diff options
Diffstat (limited to 'kernel')
270 files changed, 30197 insertions, 17706 deletions
diff --git a/kernel/Makefile b/kernel/Makefile index 12c679f769c6..4cb8e8b23c6e 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -59,15 +59,13 @@ obj-$(CONFIG_MODULES) += module.o obj-$(CONFIG_MODULE_SIG) += module_signing.o obj-$(CONFIG_KALLSYMS) += kallsyms.o obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o +obj-$(CONFIG_CRASH_CORE) += crash_core.o obj-$(CONFIG_KEXEC_CORE) += kexec_core.o obj-$(CONFIG_KEXEC) += kexec.o obj-$(CONFIG_KEXEC_FILE) += kexec_file.o obj-$(CONFIG_BACKTRACE_SELF_TEST) += backtracetest.o obj-$(CONFIG_COMPAT) += compat.o -obj-$(CONFIG_CGROUPS) += cgroup.o -obj-$(CONFIG_CGROUP_FREEZER) += cgroup_freezer.o -obj-$(CONFIG_CGROUP_PIDS) += cgroup_pids.o -obj-$(CONFIG_CPUSETS) += cpuset.o +obj-$(CONFIG_CGROUPS) += cgroup/ obj-$(CONFIG_UTS_NS) += utsname.o obj-$(CONFIG_USER_NS) += user_namespace.o obj-$(CONFIG_PID_NS) += pid_namespace.o @@ -84,7 +82,7 @@ obj-$(CONFIG_KPROBES) += kprobes.o obj-$(CONFIG_KGDB) += debug/ obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o obj-$(CONFIG_LOCKUP_DETECTOR) += watchdog.o -obj-$(CONFIG_HARDLOCKUP_DETECTOR) += watchdog_hld.o +obj-$(CONFIG_HARDLOCKUP_DETECTOR_PERF) += watchdog_hld.o obj-$(CONFIG_SECCOMP) += seccomp.o obj-$(CONFIG_RELAY) += relay.o obj-$(CONFIG_SYSCTL) += utsname_sysctl.o diff --git a/kernel/acct.c b/kernel/acct.c index 74963d192c5d..5b1284370367 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -56,6 +56,8 @@ #include <linux/syscalls.h> #include <linux/mount.h> #include <linux/uaccess.h> +#include <linux/sched/cputime.h> + #include <asm/div64.h> #include <linux/blkdev.h> /* sector_div */ #include <linux/pid_namespace.h> @@ -453,8 +455,8 @@ static void fill_ac(acct_t *ac) spin_lock_irq(¤t->sighand->siglock); tty = current->signal->tty; /* Safe as we hold the siglock */ ac->ac_tty = tty ? old_encode_dev(tty_devnum(tty)) : 0; - ac->ac_utime = encode_comp_t(jiffies_to_AHZ(cputime_to_jiffies(pacct->ac_utime))); - ac->ac_stime = encode_comp_t(jiffies_to_AHZ(cputime_to_jiffies(pacct->ac_stime))); + ac->ac_utime = encode_comp_t(nsec_to_AHZ(pacct->ac_utime)); + ac->ac_stime = encode_comp_t(nsec_to_AHZ(pacct->ac_stime)); ac->ac_flag = pacct->ac_flag; ac->ac_mem = encode_comp_t(pacct->ac_mem); ac->ac_minflt = encode_comp_t(pacct->ac_minflt); @@ -530,7 +532,7 @@ out: void acct_collect(long exitcode, int group_dead) { struct pacct_struct *pacct = ¤t->signal->pacct; - cputime_t utime, stime; + u64 utime, stime; unsigned long vsize = 0; if (group_dead && current->mm) { @@ -559,6 +561,7 @@ void acct_collect(long exitcode, int group_dead) pacct->ac_flag |= ACORE; if (current->flags & PF_SIGNALED) pacct->ac_flag |= AXSIG; + task_cputime(current, &utime, &stime); pacct->ac_utime += utime; pacct->ac_stime += stime; diff --git a/kernel/async.c b/kernel/async.c index d2edd6efec56..2cbd3dd5940d 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -114,14 +114,14 @@ static void async_run_entry_fn(struct work_struct *work) ktime_t uninitialized_var(calltime), delta, rettime; /* 1) run (and print duration) */ - if (initcall_debug && system_state == SYSTEM_BOOTING) { + if (initcall_debug && system_state < SYSTEM_RUNNING) { pr_debug("calling %lli_%pF @ %i\n", (long long)entry->cookie, entry->func, task_pid_nr(current)); calltime = ktime_get(); } entry->func(entry->data, entry->cookie); - if (initcall_debug && system_state == SYSTEM_BOOTING) { + if (initcall_debug && system_state < SYSTEM_RUNNING) { rettime = ktime_get(); delta = ktime_sub(rettime, calltime); pr_debug("initcall %lli_%pF returned 0 after %lld usecs\n", @@ -284,14 +284,14 @@ void async_synchronize_cookie_domain(async_cookie_t cookie, struct async_domain { ktime_t uninitialized_var(starttime), delta, endtime; - if (initcall_debug && system_state == SYSTEM_BOOTING) { + if (initcall_debug && system_state < SYSTEM_RUNNING) { pr_debug("async_waiting @ %i\n", task_pid_nr(current)); starttime = ktime_get(); } wait_event(async_done, lowest_in_progress(domain) >= cookie); - if (initcall_debug && system_state == SYSTEM_BOOTING) { + if (initcall_debug && system_state < SYSTEM_RUNNING) { endtime = ktime_get(); delta = ktime_sub(endtime, starttime); diff --git a/kernel/audit.c b/kernel/audit.c index 6e399bb69d7c..6dd556931739 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -54,6 +54,12 @@ #include <linux/kthread.h> #include <linux/kernel.h> #include <linux/syscalls.h> +#include <linux/spinlock.h> +#include <linux/rcupdate.h> +#include <linux/mutex.h> +#include <linux/gfp.h> +#include <linux/pid.h> +#include <linux/slab.h> #include <linux/audit.h> @@ -90,13 +96,35 @@ static u32 audit_default; /* If auditing cannot proceed, audit_failure selects what happens. */ static u32 audit_failure = AUDIT_FAIL_PRINTK; -/* - * If audit records are to be written to the netlink socket, audit_pid - * contains the pid of the auditd process and audit_nlk_portid contains - * the portid to use to send netlink messages to that process. +/* private audit network namespace index */ +static unsigned int audit_net_id; + +/** + * struct audit_net - audit private network namespace data + * @sk: communication socket */ -int audit_pid; -static __u32 audit_nlk_portid; +struct audit_net { + struct sock *sk; +}; + +/** + * struct auditd_connection - kernel/auditd connection state + * @pid: auditd PID + * @portid: netlink portid + * @net: the associated network namespace + * @rcu: RCU head + * + * Description: + * This struct is RCU protected; you must either hold the RCU lock for reading + * or the associated spinlock for writing. + */ +static struct auditd_connection { + struct pid *pid; + u32 portid; + struct net *net; + struct rcu_head rcu; +} *auditd_conn = NULL; +static DEFINE_SPINLOCK(auditd_conn_lock); /* If audit_rate_limit is non-zero, limit the rate of sending audit records * to that number per second. This prevents DoS attacks, but results in @@ -121,21 +149,12 @@ u32 audit_sig_sid = 0; 3) suppressed due to audit_rate_limit 4) suppressed due to audit_backlog_limit */ -static atomic_t audit_lost = ATOMIC_INIT(0); - -/* The netlink socket. */ -static struct sock *audit_sock; -static unsigned int audit_net_id; +static atomic_t audit_lost = ATOMIC_INIT(0); /* Hash for inode-based rules */ struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS]; -/* The audit_freelist is a list of pre-allocated audit buffers (if more - * than AUDIT_MAXFREE are in use, the audit buffer is freed instead of - * being placed on the freelist). */ -static DEFINE_SPINLOCK(audit_freelist_lock); -static int audit_freelist_count; -static LIST_HEAD(audit_freelist); +static struct kmem_cache *audit_buffer_cache; /* queue msgs to send via kauditd_task */ static struct sk_buff_head audit_queue; @@ -170,17 +189,12 @@ DEFINE_MUTEX(audit_cmd_mutex); * should be at least that large. */ #define AUDIT_BUFSIZ 1024 -/* AUDIT_MAXFREE is the number of empty audit_buffers we keep on the - * audit_freelist. Doing so eliminates many kmalloc/kfree calls. */ -#define AUDIT_MAXFREE (2*NR_CPUS) - /* The audit_buffer is used when formatting an audit record. The caller * locks briefly to get the record off the freelist or to allocate the * buffer, and locks briefly to send the buffer to the netlink layer or * to place it on a transmit queue. Multiple audit_buffers can be in * use simultaneously. */ struct audit_buffer { - struct list_head list; struct sk_buff *skb; /* formatted skb ready to send */ struct audit_context *ctx; /* NULL or associated context */ gfp_t gfp_mask; @@ -192,12 +206,65 @@ struct audit_reply { struct sk_buff *skb; }; -static void audit_set_portid(struct audit_buffer *ab, __u32 portid) +/** + * auditd_test_task - Check to see if a given task is an audit daemon + * @task: the task to check + * + * Description: + * Return 1 if the task is a registered audit daemon, 0 otherwise. + */ +int auditd_test_task(struct task_struct *task) { - if (ab) { - struct nlmsghdr *nlh = nlmsg_hdr(ab->skb); - nlh->nlmsg_pid = portid; - } + int rc; + struct auditd_connection *ac; + + rcu_read_lock(); + ac = rcu_dereference(auditd_conn); + rc = (ac && ac->pid == task_tgid(task) ? 1 : 0); + rcu_read_unlock(); + + return rc; +} + +/** + * auditd_pid_vnr - Return the auditd PID relative to the namespace + * + * Description: + * Returns the PID in relation to the namespace, 0 on failure. + */ +static pid_t auditd_pid_vnr(void) +{ + pid_t pid; + const struct auditd_connection *ac; + + rcu_read_lock(); + ac = rcu_dereference(auditd_conn); + if (!ac || !ac->pid) + pid = 0; + else + pid = pid_vnr(ac->pid); + rcu_read_unlock(); + + return pid; +} + +/** + * audit_get_sk - Return the audit socket for the given network namespace + * @net: the destination network namespace + * + * Description: + * Returns the sock pointer if valid, NULL otherwise. The caller must ensure + * that a reference is held for the network namespace while the sock is in use. + */ +static struct sock *audit_get_sk(const struct net *net) +{ + struct audit_net *aunet; + + if (!net) + return NULL; + + aunet = net_generic(net, audit_net_id); + return aunet->sk; } void audit_panic(const char *message) @@ -210,9 +277,7 @@ void audit_panic(const char *message) pr_err("%s\n", message); break; case AUDIT_FAIL_PANIC: - /* test audit_pid since printk is always losey, why bother? */ - if (audit_pid) - panic("audit: %s\n", message); + panic("audit: %s\n", message); break; } } @@ -370,21 +435,89 @@ static int audit_set_failure(u32 state) return audit_do_config_change("audit_failure", &audit_failure, state); } -/* - * For one reason or another this nlh isn't getting delivered to the userspace - * audit daemon, just send it to printk. +/** + * auditd_conn_free - RCU helper to release an auditd connection struct + * @rcu: RCU head + * + * Description: + * Drop any references inside the auditd connection tracking struct and free + * the memory. + */ + static void auditd_conn_free(struct rcu_head *rcu) + { + struct auditd_connection *ac; + + ac = container_of(rcu, struct auditd_connection, rcu); + put_pid(ac->pid); + put_net(ac->net); + kfree(ac); + } + +/** + * auditd_set - Set/Reset the auditd connection state + * @pid: auditd PID + * @portid: auditd netlink portid + * @net: auditd network namespace pointer + * + * Description: + * This function will obtain and drop network namespace references as + * necessary. Returns zero on success, negative values on failure. + */ +static int auditd_set(struct pid *pid, u32 portid, struct net *net) +{ + unsigned long flags; + struct auditd_connection *ac_old, *ac_new; + + if (!pid || !net) + return -EINVAL; + + ac_new = kzalloc(sizeof(*ac_new), GFP_KERNEL); + if (!ac_new) + return -ENOMEM; + ac_new->pid = get_pid(pid); + ac_new->portid = portid; + ac_new->net = get_net(net); + + spin_lock_irqsave(&auditd_conn_lock, flags); + ac_old = rcu_dereference_protected(auditd_conn, + lockdep_is_held(&auditd_conn_lock)); + rcu_assign_pointer(auditd_conn, ac_new); + spin_unlock_irqrestore(&auditd_conn_lock, flags); + + if (ac_old) + call_rcu(&ac_old->rcu, auditd_conn_free); + + return 0; +} + +/** + * kauditd_print_skb - Print the audit record to the ring buffer + * @skb: audit record + * + * Whatever the reason, this packet may not make it to the auditd connection + * so write it via printk so the information isn't completely lost. */ static void kauditd_printk_skb(struct sk_buff *skb) { struct nlmsghdr *nlh = nlmsg_hdr(skb); char *data = nlmsg_data(nlh); - if (nlh->nlmsg_type != AUDIT_EOE) { - if (printk_ratelimit()) - pr_notice("type=%d %s\n", nlh->nlmsg_type, data); - else - audit_log_lost("printk limit exceeded"); - } + if (nlh->nlmsg_type != AUDIT_EOE && printk_ratelimit()) + pr_notice("type=%d %s\n", nlh->nlmsg_type, data); +} + +/** + * kauditd_rehold_skb - Handle a audit record send failure in the hold queue + * @skb: audit record + * + * Description: + * This should only be used by the kauditd_thread when it fails to flush the + * hold queue. + */ +static void kauditd_rehold_skb(struct sk_buff *skb) +{ + /* put the record back in the queue at the same place */ + skb_queue_head(&audit_hold_queue, skb); } /** @@ -442,67 +575,181 @@ static void kauditd_retry_skb(struct sk_buff *skb) /** * auditd_reset - Disconnect the auditd connection + * @ac: auditd connection state * * Description: - * Break the auditd/kauditd connection and move all the records in the retry - * queue into the hold queue in case auditd reconnects. The audit_cmd_mutex - * must be held when calling this function. + * Break the auditd/kauditd connection and move all the queued records into the + * hold queue in case auditd reconnects. It is important to note that the @ac + * pointer should never be dereferenced inside this function as it may be NULL + * or invalid, you can only compare the memory address! If @ac is NULL then + * the connection will always be reset. */ -static void auditd_reset(void) +static void auditd_reset(const struct auditd_connection *ac) { + unsigned long flags; struct sk_buff *skb; - - /* break the connection */ - if (audit_sock) { - sock_put(audit_sock); - audit_sock = NULL; + struct auditd_connection *ac_old; + + /* if it isn't already broken, break the connection */ + spin_lock_irqsave(&auditd_conn_lock, flags); + ac_old = rcu_dereference_protected(auditd_conn, + lockdep_is_held(&auditd_conn_lock)); + if (ac && ac != ac_old) { + /* someone already registered a new auditd connection */ + spin_unlock_irqrestore(&auditd_conn_lock, flags); + return; } - audit_pid = 0; - audit_nlk_portid = 0; + rcu_assign_pointer(auditd_conn, NULL); + spin_unlock_irqrestore(&auditd_conn_lock, flags); - /* flush all of the retry queue to the hold queue */ + if (ac_old) + call_rcu(&ac_old->rcu, auditd_conn_free); + + /* flush the retry queue to the hold queue, but don't touch the main + * queue since we need to process that normally for multicast */ while ((skb = skb_dequeue(&audit_retry_queue))) kauditd_hold_skb(skb); } /** - * kauditd_send_unicast_skb - Send a record via unicast to auditd + * auditd_send_unicast_skb - Send a record via unicast to auditd * @skb: audit record + * + * Description: + * Send a skb to the audit daemon, returns positive/zero values on success and + * negative values on failure; in all cases the skb will be consumed by this + * function. If the send results in -ECONNREFUSED the connection with auditd + * will be reset. This function may sleep so callers should not hold any locks + * where this would cause a problem. */ -static int kauditd_send_unicast_skb(struct sk_buff *skb) +static int auditd_send_unicast_skb(struct sk_buff *skb) { int rc; + u32 portid; + struct net *net; + struct sock *sk; + struct auditd_connection *ac; + + /* NOTE: we can't call netlink_unicast while in the RCU section so + * take a reference to the network namespace and grab local + * copies of the namespace, the sock, and the portid; the + * namespace and sock aren't going to go away while we hold a + * reference and if the portid does become invalid after the RCU + * section netlink_unicast() should safely return an error */ + + rcu_read_lock(); + ac = rcu_dereference(auditd_conn); + if (!ac) { + rcu_read_unlock(); + kfree_skb(skb); + rc = -ECONNREFUSED; + goto err; + } + net = get_net(ac->net); + sk = audit_get_sk(net); + portid = ac->portid; + rcu_read_unlock(); - /* if we know nothing is connected, don't even try the netlink call */ - if (!audit_pid) - return -ECONNREFUSED; + rc = netlink_unicast(sk, skb, portid, 0); + put_net(net); + if (rc < 0) + goto err; - /* get an extra skb reference in case we fail to send */ - skb_get(skb); - rc = netlink_unicast(audit_sock, skb, audit_nlk_portid, 0); - if (rc >= 0) { - consume_skb(skb); - rc = 0; - } + return rc; +err: + if (ac && rc == -ECONNREFUSED) + auditd_reset(ac); return rc; } +/** + * kauditd_send_queue - Helper for kauditd_thread to flush skb queues + * @sk: the sending sock + * @portid: the netlink destination + * @queue: the skb queue to process + * @retry_limit: limit on number of netlink unicast failures + * @skb_hook: per-skb hook for additional processing + * @err_hook: hook called if the skb fails the netlink unicast send + * + * Description: + * Run through the given queue and attempt to send the audit records to auditd, + * returns zero on success, negative values on failure. It is up to the caller + * to ensure that the @sk is valid for the duration of this function. + * + */ +static int kauditd_send_queue(struct sock *sk, u32 portid, + struct sk_buff_head *queue, + unsigned int retry_limit, + void (*skb_hook)(struct sk_buff *skb), + void (*err_hook)(struct sk_buff *skb)) +{ + int rc = 0; + struct sk_buff *skb; + static unsigned int failed = 0; + + /* NOTE: kauditd_thread takes care of all our locking, we just use + * the netlink info passed to us (e.g. sk and portid) */ + + while ((skb = skb_dequeue(queue))) { + /* call the skb_hook for each skb we touch */ + if (skb_hook) + (*skb_hook)(skb); + + /* can we send to anyone via unicast? */ + if (!sk) { + if (err_hook) + (*err_hook)(skb); + continue; + } + + /* grab an extra skb reference in case of error */ + skb_get(skb); + rc = netlink_unicast(sk, skb, portid, 0); + if (rc < 0) { + /* fatal failure for our queue flush attempt? */ + if (++failed >= retry_limit || + rc == -ECONNREFUSED || rc == -EPERM) { + /* yes - error processing for the queue */ + sk = NULL; + if (err_hook) + (*err_hook)(skb); + if (!skb_hook) + goto out; + /* keep processing with the skb_hook */ + continue; + } else + /* no - requeue to preserve ordering */ + skb_queue_head(queue, skb); + } else { + /* it worked - drop the extra reference and continue */ + consume_skb(skb); + failed = 0; + } + } + +out: + return (rc >= 0 ? 0 : rc); +} + /* * kauditd_send_multicast_skb - Send a record to any multicast listeners * @skb: audit record * * Description: - * This function doesn't consume an skb as might be expected since it has to - * copy it anyways. + * Write a multicast message to anyone listening in the initial network + * namespace. This function doesn't consume an skb as might be expected since + * it has to copy it anyways. */ static void kauditd_send_multicast_skb(struct sk_buff *skb) { struct sk_buff *copy; - struct audit_net *aunet = net_generic(&init_net, audit_net_id); - struct sock *sock = aunet->nlsk; + struct sock *sock = audit_get_sk(&init_net); struct nlmsghdr *nlh; + /* NOTE: we are not taking an additional reference for init_net since + * we don't have to worry about it going away */ + if (!netlink_has_listeners(sock, AUDIT_NLGRP_READLOG)) return; @@ -526,149 +773,81 @@ static void kauditd_send_multicast_skb(struct sk_buff *skb) } /** - * kauditd_wake_condition - Return true when it is time to wake kauditd_thread - * - * Description: - * This function is for use by the wait_event_freezable() call in - * kauditd_thread(). + * kauditd_thread - Worker thread to send audit records to userspace + * @dummy: unused */ -static int kauditd_wake_condition(void) -{ - static int pid_last = 0; - int rc; - int pid = audit_pid; - - /* wake on new messages or a change in the connected auditd */ - rc = skb_queue_len(&audit_queue) || (pid && pid != pid_last); - if (rc) - pid_last = pid; - - return rc; -} - static int kauditd_thread(void *dummy) { int rc; - int auditd = 0; - int reschedule = 0; - struct sk_buff *skb; - struct nlmsghdr *nlh; + u32 portid = 0; + struct net *net = NULL; + struct sock *sk = NULL; + struct auditd_connection *ac; #define UNICAST_RETRIES 5 -#define AUDITD_BAD(x,y) \ - ((x) == -ECONNREFUSED || (x) == -EPERM || ++(y) >= UNICAST_RETRIES) - - /* NOTE: we do invalidate the auditd connection flag on any sending - * errors, but we only "restore" the connection flag at specific places - * in the loop in order to help ensure proper ordering of audit - * records */ set_freezable(); while (!kthread_should_stop()) { - /* NOTE: possible area for future improvement is to look at - * the hold and retry queues, since only this thread - * has access to these queues we might be able to do - * our own queuing and skip some/all of the locking */ - - /* NOTE: it might be a fun experiment to split the hold and - * retry queue handling to another thread, but the - * synchronization issues and other overhead might kill - * any performance gains */ + /* NOTE: see the lock comments in auditd_send_unicast_skb() */ + rcu_read_lock(); + ac = rcu_dereference(auditd_conn); + if (!ac) { + rcu_read_unlock(); + goto main_queue; + } + net = get_net(ac->net); + sk = audit_get_sk(net); + portid = ac->portid; + rcu_read_unlock(); /* attempt to flush the hold queue */ - while (auditd && (skb = skb_dequeue(&audit_hold_queue))) { - rc = kauditd_send_unicast_skb(skb); - if (rc) { - /* requeue to the same spot */ - skb_queue_head(&audit_hold_queue, skb); - - auditd = 0; - if (AUDITD_BAD(rc, reschedule)) { - mutex_lock(&audit_cmd_mutex); - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - reschedule = 0; - } - } else - /* we were able to send successfully */ - reschedule = 0; + rc = kauditd_send_queue(sk, portid, + &audit_hold_queue, UNICAST_RETRIES, + NULL, kauditd_rehold_skb); + if (ac && rc < 0) { + sk = NULL; + auditd_reset(ac); + goto main_queue; } /* attempt to flush the retry queue */ - while (auditd && (skb = skb_dequeue(&audit_retry_queue))) { - rc = kauditd_send_unicast_skb(skb); - if (rc) { - auditd = 0; - if (AUDITD_BAD(rc, reschedule)) { - kauditd_hold_skb(skb); - mutex_lock(&audit_cmd_mutex); - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - reschedule = 0; - } else - /* temporary problem (we hope), queue - * to the same spot and retry */ - skb_queue_head(&audit_retry_queue, skb); - } else - /* we were able to send successfully */ - reschedule = 0; + rc = kauditd_send_queue(sk, portid, + &audit_retry_queue, UNICAST_RETRIES, + NULL, kauditd_hold_skb); + if (ac && rc < 0) { + sk = NULL; + auditd_reset(ac); + goto main_queue; } - /* standard queue processing, try to be as quick as possible */ -quick_loop: - skb = skb_dequeue(&audit_queue); - if (skb) { - /* setup the netlink header, see the comments in - * kauditd_send_multicast_skb() for length quirks */ - nlh = nlmsg_hdr(skb); - nlh->nlmsg_len = skb->len - NLMSG_HDRLEN; - - /* attempt to send to any multicast listeners */ - kauditd_send_multicast_skb(skb); - - /* attempt to send to auditd, queue on failure */ - if (auditd) { - rc = kauditd_send_unicast_skb(skb); - if (rc) { - auditd = 0; - if (AUDITD_BAD(rc, reschedule)) { - mutex_lock(&audit_cmd_mutex); - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - reschedule = 0; - } - - /* move to the retry queue */ - kauditd_retry_skb(skb); - } else - /* everything is working so go fast! */ - goto quick_loop; - } else if (reschedule) - /* we are currently having problems, move to - * the retry queue */ - kauditd_retry_skb(skb); - else - /* dump the message via printk and hold it */ - kauditd_hold_skb(skb); - } else { - /* we have flushed the backlog so wake everyone */ - wake_up(&audit_backlog_wait); - - /* if everything is okay with auditd (if present), go - * to sleep until there is something new in the queue - * or we have a change in the connected auditd; - * otherwise simply reschedule to give things a chance - * to recover */ - if (reschedule) { - set_current_state(TASK_INTERRUPTIBLE); - schedule(); - } else - wait_event_freezable(kauditd_wait, - kauditd_wake_condition()); - - /* update the auditd connection status */ - auditd = (audit_pid ? 1 : 0); +main_queue: + /* process the main queue - do the multicast send and attempt + * unicast, dump failed record sends to the retry queue; if + * sk == NULL due to previous failures we will just do the + * multicast send and move the record to the hold queue */ + rc = kauditd_send_queue(sk, portid, &audit_queue, 1, + kauditd_send_multicast_skb, + (sk ? + kauditd_retry_skb : kauditd_hold_skb)); + if (ac && rc < 0) + auditd_reset(ac); + sk = NULL; + + /* drop our netns reference, no auditd sends past this line */ + if (net) { + put_net(net); + net = NULL; } + + /* we have processed all the queues so wake everyone */ + wake_up(&audit_backlog_wait); + + /* NOTE: we want to wake up if there is anything on the queue, + * regardless of if an auditd is connected, as we need to + * do the multicast send and rotate records from the + * main queue to the retry/hold queues */ + wait_event_freezable(kauditd_wait, + (skb_queue_len(&audit_queue) ? 1 : 0)); } return 0; @@ -678,23 +857,22 @@ int audit_send_list(void *_dest) { struct audit_netlink_list *dest = _dest; struct sk_buff *skb; - struct net *net = dest->net; - struct audit_net *aunet = net_generic(net, audit_net_id); + struct sock *sk = audit_get_sk(dest->net); /* wait for parent to finish and send an ACK */ mutex_lock(&audit_cmd_mutex); mutex_unlock(&audit_cmd_mutex); while ((skb = __skb_dequeue(&dest->q)) != NULL) - netlink_unicast(aunet->nlsk, skb, dest->portid, 0); + netlink_unicast(sk, skb, dest->portid, 0); - put_net(net); + put_net(dest->net); kfree(dest); return 0; } -struct sk_buff *audit_make_reply(__u32 portid, int seq, int type, int done, +struct sk_buff *audit_make_reply(int seq, int type, int done, int multi, const void *payload, int size) { struct sk_buff *skb; @@ -707,7 +885,7 @@ struct sk_buff *audit_make_reply(__u32 portid, int seq, int type, int done, if (!skb) return NULL; - nlh = nlmsg_put(skb, portid, seq, t, size, flags); + nlh = nlmsg_put(skb, 0, seq, t, size, flags); if (!nlh) goto out_kfree_skb; data = nlmsg_data(nlh); @@ -722,16 +900,15 @@ out_kfree_skb: static int audit_send_reply_thread(void *arg) { struct audit_reply *reply = (struct audit_reply *)arg; - struct net *net = reply->net; - struct audit_net *aunet = net_generic(net, audit_net_id); + struct sock *sk = audit_get_sk(reply->net); mutex_lock(&audit_cmd_mutex); mutex_unlock(&audit_cmd_mutex); /* Ignore failure. It'll only happen if the sender goes away, because our timeout is set to infinite. */ - netlink_unicast(aunet->nlsk , reply->skb, reply->portid, 0); - put_net(net); + netlink_unicast(sk, reply->skb, reply->portid, 0); + put_net(reply->net); kfree(reply); return 0; } @@ -752,7 +929,6 @@ static int audit_send_reply_thread(void *arg) static void audit_send_reply(struct sk_buff *request_skb, int seq, int type, int done, int multi, const void *payload, int size) { - u32 portid = NETLINK_CB(request_skb).portid; struct net *net = sock_net(NETLINK_CB(request_skb).sk); struct sk_buff *skb; struct task_struct *tsk; @@ -762,12 +938,12 @@ static void audit_send_reply(struct sk_buff *request_skb, int seq, int type, int if (!reply) return; - skb = audit_make_reply(portid, seq, type, done, multi, payload, size); + skb = audit_make_reply(seq, type, done, multi, payload, size); if (!skb) goto out; reply->net = get_net(net); - reply->portid = portid; + reply->portid = NETLINK_CB(request_skb).portid; reply->skb = skb; tsk = kthread_run(audit_send_reply_thread, reply, "audit_send_reply"); @@ -947,14 +1123,16 @@ static int audit_set_feature(struct sk_buff *skb) return 0; } -static int audit_replace(pid_t pid) +static int audit_replace(struct pid *pid) { - struct sk_buff *skb = audit_make_reply(0, 0, AUDIT_REPLACE, 0, 0, - &pid, sizeof(pid)); + pid_t pvnr; + struct sk_buff *skb; + pvnr = pid_vnr(pid); + skb = audit_make_reply(0, AUDIT_REPLACE, 0, 0, &pvnr, sizeof(pvnr)); if (!skb) return -ENOMEM; - return netlink_unicast(audit_sock, skb, audit_nlk_portid, 0); + return auditd_send_unicast_skb(skb); } static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) @@ -981,7 +1159,9 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) memset(&s, 0, sizeof(s)); s.enabled = audit_enabled; s.failure = audit_failure; - s.pid = audit_pid; + /* NOTE: use pid_vnr() so the PID is relative to the current + * namespace */ + s.pid = auditd_pid_vnr(); s.rate_limit = audit_rate_limit; s.backlog_limit = audit_backlog_limit; s.lost = atomic_read(&audit_lost); @@ -1007,37 +1187,61 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return err; } if (s.mask & AUDIT_STATUS_PID) { - /* NOTE: we are using task_tgid_vnr() below because - * the s.pid value is relative to the namespace - * of the caller; at present this doesn't matter - * much since you can really only run auditd - * from the initial pid namespace, but something - * to keep in mind if this changes */ - int new_pid = s.pid; - pid_t requesting_pid = task_tgid_vnr(current); - - if ((!new_pid) && (requesting_pid != audit_pid)) { - audit_log_config_change("audit_pid", new_pid, audit_pid, 0); + /* NOTE: we are using the vnr PID functions below + * because the s.pid value is relative to the + * namespace of the caller; at present this + * doesn't matter much since you can really only + * run auditd from the initial pid namespace, but + * something to keep in mind if this changes */ + pid_t new_pid = s.pid; + pid_t auditd_pid; + struct pid *req_pid = task_tgid(current); + + /* sanity check - PID values must match */ + if (new_pid != pid_vnr(req_pid)) + return -EINVAL; + + /* test the auditd connection */ + audit_replace(req_pid); + + auditd_pid = auditd_pid_vnr(); + /* only the current auditd can unregister itself */ + if ((!new_pid) && (new_pid != auditd_pid)) { + audit_log_config_change("audit_pid", new_pid, + auditd_pid, 0); return -EACCES; } - if (audit_pid && new_pid && - audit_replace(requesting_pid) != -ECONNREFUSED) { - audit_log_config_change("audit_pid", new_pid, audit_pid, 0); + /* replacing a healthy auditd is not allowed */ + if (auditd_pid && new_pid) { + audit_log_config_change("audit_pid", new_pid, + auditd_pid, 0); return -EEXIST; } - if (audit_enabled != AUDIT_OFF) - audit_log_config_change("audit_pid", new_pid, audit_pid, 1); + if (new_pid) { - if (audit_sock) - sock_put(audit_sock); - audit_pid = new_pid; - audit_nlk_portid = NETLINK_CB(skb).portid; - sock_hold(skb->sk); - audit_sock = skb->sk; + /* register a new auditd connection */ + err = auditd_set(req_pid, + NETLINK_CB(skb).portid, + sock_net(NETLINK_CB(skb).sk)); + if (audit_enabled != AUDIT_OFF) + audit_log_config_change("audit_pid", + new_pid, + auditd_pid, + err ? 0 : 1); + if (err) + return err; + + /* try to process any backlog */ + wake_up_interruptible(&kauditd_wait); } else { - auditd_reset(); + if (audit_enabled != AUDIT_OFF) + audit_log_config_change("audit_pid", + new_pid, + auditd_pid, 1); + + /* unregister the auditd connection */ + auditd_reset(NULL); } - wake_up_interruptible(&kauditd_wait); } if (s.mask & AUDIT_STATUS_RATE_LIMIT) { err = audit_set_rate_limit(s.rate_limit); @@ -1058,6 +1262,12 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (err < 0) return err; } + if (s.mask == AUDIT_STATUS_LOST) { + u32 lost = atomic_xchg(&audit_lost, 0); + + audit_log_config_change("lost", 0, lost, 1); + return lost; + } break; } case AUDIT_GET_FEATURE: @@ -1084,7 +1294,6 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (err) break; } - mutex_unlock(&audit_cmd_mutex); audit_log_common_recv_msg(&ab, msg_type); if (msg_type != AUDIT_USER_TTY) audit_log_format(ab, " msg='%.*s'", @@ -1100,9 +1309,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) size--; audit_log_n_untrustedstring(ab, data, size); } - audit_set_portid(ab, NETLINK_CB(skb).portid); audit_log_end(ab); - mutex_lock(&audit_cmd_mutex); } break; case AUDIT_ADD_RULE: @@ -1115,8 +1322,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) audit_log_end(ab); return -EPERM; } - err = audit_rule_change(msg_type, NETLINK_CB(skb).portid, - seq, data, nlmsg_len(nlh)); + err = audit_rule_change(msg_type, seq, data, nlmsg_len(nlh)); break; case AUDIT_LIST_RULES: err = audit_list_rules_send(skb, seq); @@ -1237,11 +1443,14 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return err < 0 ? err : 0; } -/* - * Get message from skb. Each message is processed by audit_receive_msg. - * Malformed skbs with wrong length are discarded silently. +/** + * audit_receive - receive messages from a netlink control socket + * @skb: the message buffer + * + * Parse the provided skb and deal with any messages that may be present, + * malformed skbs are discarded. */ -static void audit_receive_skb(struct sk_buff *skb) +static void audit_receive(struct sk_buff *skb) { struct nlmsghdr *nlh; /* @@ -1254,21 +1463,15 @@ static void audit_receive_skb(struct sk_buff *skb) nlh = nlmsg_hdr(skb); len = skb->len; + mutex_lock(&audit_cmd_mutex); while (nlmsg_ok(nlh, len)) { err = audit_receive_msg(skb, nlh); /* if err or if this message says it wants a response */ if (err || (nlh->nlmsg_flags & NLM_F_ACK)) - netlink_ack(skb, nlh, err); + netlink_ack(skb, nlh, err, NULL); nlh = nlmsg_next(nlh, &len); } -} - -/* Receive messages from netlink socket. */ -static void audit_receive(struct sk_buff *skb) -{ - mutex_lock(&audit_cmd_mutex); - audit_receive_skb(skb); mutex_unlock(&audit_cmd_mutex); } @@ -1292,26 +1495,27 @@ static int __net_init audit_net_init(struct net *net) struct audit_net *aunet = net_generic(net, audit_net_id); - aunet->nlsk = netlink_kernel_create(net, NETLINK_AUDIT, &cfg); - if (aunet->nlsk == NULL) { + aunet->sk = netlink_kernel_create(net, NETLINK_AUDIT, &cfg); + if (aunet->sk == NULL) { audit_panic("cannot initialize netlink socket in namespace"); return -ENOMEM; } - aunet->nlsk->sk_sndtimeo = MAX_SCHEDULE_TIMEOUT; + aunet->sk->sk_sndtimeo = MAX_SCHEDULE_TIMEOUT; + return 0; } static void __net_exit audit_net_exit(struct net *net) { struct audit_net *aunet = net_generic(net, audit_net_id); - struct sock *sock = aunet->nlsk; - mutex_lock(&audit_cmd_mutex); - if (sock == audit_sock) - auditd_reset(); - mutex_unlock(&audit_cmd_mutex); - netlink_kernel_release(sock); - aunet->nlsk = NULL; + /* NOTE: you would think that we would want to check the auditd + * connection and potentially reset it here if it lives in this + * namespace, but since the auditd connection tracking struct holds a + * reference to this namespace (see auditd_set()) we are only ever + * going to get here after that connection has been released */ + + netlink_kernel_release(aunet->sk); } static struct pernet_operations audit_net_ops __net_initdata = { @@ -1329,27 +1533,34 @@ static int __init audit_init(void) if (audit_initialized == AUDIT_DISABLED) return 0; - pr_info("initializing netlink subsys (%s)\n", - audit_default ? "enabled" : "disabled"); - register_pernet_subsys(&audit_net_ops); + audit_buffer_cache = kmem_cache_create("audit_buffer", + sizeof(struct audit_buffer), + 0, SLAB_PANIC, NULL); skb_queue_head_init(&audit_queue); skb_queue_head_init(&audit_retry_queue); skb_queue_head_init(&audit_hold_queue); - audit_initialized = AUDIT_INITIALIZED; - audit_enabled = audit_default; - audit_ever_enabled |= !!audit_default; for (i = 0; i < AUDIT_INODE_BUCKETS; i++) INIT_LIST_HEAD(&audit_inode_hash[i]); + pr_info("initializing netlink subsys (%s)\n", + audit_default ? "enabled" : "disabled"); + register_pernet_subsys(&audit_net_ops); + + audit_initialized = AUDIT_INITIALIZED; + audit_enabled = audit_default; + audit_ever_enabled |= !!audit_default; + kauditd_task = kthread_run(kauditd_thread, NULL, "kauditd"); if (IS_ERR(kauditd_task)) { int err = PTR_ERR(kauditd_task); panic("audit: failed to start the kauditd thread (%d)\n", err); } - audit_log(NULL, GFP_KERNEL, AUDIT_KERNEL, "initialized"); + audit_log(NULL, GFP_KERNEL, AUDIT_KERNEL, + "state=initialized audit_enabled=%u res=1", + audit_enabled); return 0; } @@ -1391,60 +1602,33 @@ __setup("audit_backlog_limit=", audit_backlog_limit_set); static void audit_buffer_free(struct audit_buffer *ab) { - unsigned long flags; - if (!ab) return; kfree_skb(ab->skb); - spin_lock_irqsave(&audit_freelist_lock, flags); - if (audit_freelist_count > AUDIT_MAXFREE) - kfree(ab); - else { - audit_freelist_count++; - list_add(&ab->list, &audit_freelist); - } - spin_unlock_irqrestore(&audit_freelist_lock, flags); + kmem_cache_free(audit_buffer_cache, ab); } -static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, - gfp_t gfp_mask, int type) +static struct audit_buffer *audit_buffer_alloc(struct audit_context *ctx, + gfp_t gfp_mask, int type) { - unsigned long flags; - struct audit_buffer *ab = NULL; - struct nlmsghdr *nlh; - - spin_lock_irqsave(&audit_freelist_lock, flags); - if (!list_empty(&audit_freelist)) { - ab = list_entry(audit_freelist.next, - struct audit_buffer, list); - list_del(&ab->list); - --audit_freelist_count; - } - spin_unlock_irqrestore(&audit_freelist_lock, flags); - - if (!ab) { - ab = kmalloc(sizeof(*ab), gfp_mask); - if (!ab) - goto err; - } + struct audit_buffer *ab; - ab->ctx = ctx; - ab->gfp_mask = gfp_mask; + ab = kmem_cache_alloc(audit_buffer_cache, gfp_mask); + if (!ab) + return NULL; ab->skb = nlmsg_new(AUDIT_BUFSIZ, gfp_mask); if (!ab->skb) goto err; + if (!nlmsg_put(ab->skb, 0, 0, type, 0, 0)) + goto err; - nlh = nlmsg_put(ab->skb, 0, 0, type, 0, 0); - if (!nlh) - goto out_kfree_skb; + ab->ctx = ctx; + ab->gfp_mask = gfp_mask; return ab; -out_kfree_skb: - kfree_skb(ab->skb); - ab->skb = NULL; err: audit_buffer_free(ab); return NULL; @@ -1475,10 +1659,10 @@ unsigned int audit_serial(void) } static inline void audit_get_stamp(struct audit_context *ctx, - struct timespec *t, unsigned int *serial) + struct timespec64 *t, unsigned int *serial) { if (!ctx || !auditsc_get_stamp(ctx, t, serial)) { - *t = CURRENT_TIME; + ktime_get_real_ts64(t); *serial = audit_serial(); } } @@ -1502,7 +1686,7 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, int type) { struct audit_buffer *ab; - struct timespec t; + struct timespec64 t; unsigned int uninitialized_var(serial); if (audit_initialized != AUDIT_INITIALIZED) @@ -1511,20 +1695,16 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, if (unlikely(!audit_filter(type, AUDIT_FILTER_TYPE))) return NULL; - /* don't ever fail/sleep on these two conditions: + /* NOTE: don't ever fail/sleep on these two conditions: * 1. auditd generated record - since we need auditd to drain the * queue; also, when we are checking for auditd, compare PIDs using * task_tgid_vnr() since auditd_pid is set in audit_receive_msg() * using a PID anchored in the caller's namespace - * 2. audit command message - record types 1000 through 1099 inclusive - * are command messages/records used to manage the kernel subsystem - * and the audit userspace, blocking on these messages could cause - * problems under load so don't do it (note: not all of these - * command types are valid as record types, but it is quicker to - * just check two ints than a series of ints in a if/switch stmt) */ - if (!((audit_pid && audit_pid == task_tgid_vnr(current)) || - (type >= 1000 && type <= 1099))) { - long sleep_time = audit_backlog_wait_time; + * 2. generator holding the audit_cmd_mutex - we don't want to block + * while holding the mutex */ + if (!(auditd_test_task(current) || + (current == __mutex_owner(&audit_cmd_mutex)))) { + long stime = audit_backlog_wait_time; while (audit_backlog_limit && (skb_queue_len(&audit_queue) > audit_backlog_limit)) { @@ -1533,14 +1713,13 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, /* sleep if we are allowed and we haven't exhausted our * backlog wait limit */ - if ((gfp_mask & __GFP_DIRECT_RECLAIM) && - (sleep_time > 0)) { + if (gfpflags_allow_blocking(gfp_mask) && (stime > 0)) { DECLARE_WAITQUEUE(wait, current); add_wait_queue_exclusive(&audit_backlog_wait, &wait); set_current_state(TASK_UNINTERRUPTIBLE); - sleep_time = schedule_timeout(sleep_time); + stime = schedule_timeout(stime); remove_wait_queue(&audit_backlog_wait, &wait); } else { if (audit_rate_check() && printk_ratelimit()) @@ -1560,8 +1739,8 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, } audit_get_stamp(ab->ctx, &t, &serial); - audit_log_format(ab, "audit(%lu.%03lu:%u): ", - t.tv_sec, t.tv_nsec/1000000, serial); + audit_log_format(ab, "audit(%llu.%03lu:%u): ", + (unsigned long long)t.tv_sec, t.tv_nsec/1000000, serial); return ab; } @@ -1830,22 +2009,10 @@ void audit_log_cap(struct audit_buffer *ab, char *prefix, kernel_cap_t *cap) static void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name) { - kernel_cap_t *perm = &name->fcap.permitted; - kernel_cap_t *inh = &name->fcap.inheritable; - int log = 0; - - if (!cap_isclear(*perm)) { - audit_log_cap(ab, "cap_fp", perm); - log = 1; - } - if (!cap_isclear(*inh)) { - audit_log_cap(ab, "cap_fi", inh); - log = 1; - } - - if (log) - audit_log_format(ab, " cap_fe=%d cap_fver=%x", - name->fcap.fE, name->fcap_ver); + audit_log_cap(ab, "cap_fp", &name->fcap.permitted); + audit_log_cap(ab, "cap_fi", &name->fcap.inheritable); + audit_log_format(ab, " cap_fe=%d cap_fver=%x", + name->fcap.fE, name->fcap_ver); } static inline int audit_copy_fcaps(struct audit_names *name, @@ -2119,15 +2286,27 @@ out: */ void audit_log_end(struct audit_buffer *ab) { + struct sk_buff *skb; + struct nlmsghdr *nlh; + if (!ab) return; - if (!audit_rate_check()) { - audit_log_lost("rate limit exceeded"); - } else { - skb_queue_tail(&audit_queue, ab->skb); - wake_up_interruptible(&kauditd_wait); + + if (audit_rate_check()) { + skb = ab->skb; ab->skb = NULL; - } + + /* setup the netlink header, see the comments in + * kauditd_send_multicast_skb() for length quirks */ + nlh = nlmsg_hdr(skb); + nlh->nlmsg_len = skb->len - NLMSG_HDRLEN; + + /* queue the netlink packet and poke the kauditd thread */ + skb_queue_tail(&audit_queue, skb); + wake_up_interruptible(&kauditd_wait); + } else + audit_log_lost("rate limit exceeded"); + audit_buffer_free(ab); } diff --git a/kernel/audit.h b/kernel/audit.h index 960d49c9db5e..b331d9b83f63 100644 --- a/kernel/audit.h +++ b/kernel/audit.h @@ -68,6 +68,7 @@ struct audit_cap_data { unsigned int fE; /* effective bit of file cap */ kernel_cap_t effective; /* effective set of process */ }; + kernel_cap_t ambient; }; /* When fs/namei.c:getname() is called, we store the pointer in name and bump @@ -112,7 +113,7 @@ struct audit_context { enum audit_state state, current_state; unsigned int serial; /* serial number for record */ int major; /* syscall number */ - struct timespec ctime; /* time of syscall entry */ + struct timespec64 ctime; /* time of syscall entry */ unsigned long argv[4]; /* syscall arguments */ long return_code;/* syscall return code */ u64 prio; @@ -199,6 +200,9 @@ struct audit_context { struct { int argc; } execve; + struct { + char *name; + } module; }; int fds[2]; struct audit_proctitle proctitle; @@ -215,7 +219,7 @@ extern void audit_log_name(struct audit_context *context, struct audit_names *n, const struct path *path, int record_num, int *call_panic); -extern int audit_pid; +extern int auditd_test_task(struct task_struct *task); #define AUDIT_INODE_BUCKETS 32 extern struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS]; @@ -234,8 +238,7 @@ extern int audit_uid_comparator(kuid_t left, u32 op, kuid_t right); extern int audit_gid_comparator(kgid_t left, u32 op, kgid_t right); extern int parent_len(const char *path); extern int audit_compare_dname_path(const char *dname, const char *path, int plen); -extern struct sk_buff *audit_make_reply(__u32 portid, int seq, int type, - int done, int multi, +extern struct sk_buff *audit_make_reply(int seq, int type, int done, int multi, const void *payload, int size); extern void audit_panic(const char *message); @@ -245,17 +248,13 @@ struct audit_netlink_list { struct sk_buff_head q; }; -int audit_send_list(void *); - -struct audit_net { - struct sock *nlsk; -}; +int audit_send_list(void *_dest); extern int selinux_audit_rule_update(void); extern struct mutex audit_filter_mutex; -extern int audit_del_rule(struct audit_entry *); -extern void audit_free_rule_rcu(struct rcu_head *); +extern int audit_del_rule(struct audit_entry *entry); +extern void audit_free_rule_rcu(struct rcu_head *head); extern struct list_head audit_filter_list[]; extern struct audit_entry *audit_dupe_rule(struct audit_krule *old); @@ -303,17 +302,17 @@ extern int audit_exe_compare(struct task_struct *tsk, struct audit_fsnotify_mark #endif /* CONFIG_AUDIT_WATCH */ #ifdef CONFIG_AUDIT_TREE -extern struct audit_chunk *audit_tree_lookup(const struct inode *); -extern void audit_put_chunk(struct audit_chunk *); -extern bool audit_tree_match(struct audit_chunk *, struct audit_tree *); -extern int audit_make_tree(struct audit_krule *, char *, u32); -extern int audit_add_tree_rule(struct audit_krule *); -extern int audit_remove_tree_rule(struct audit_krule *); +extern struct audit_chunk *audit_tree_lookup(const struct inode *inode); +extern void audit_put_chunk(struct audit_chunk *chunk); +extern bool audit_tree_match(struct audit_chunk *chunk, struct audit_tree *tree); +extern int audit_make_tree(struct audit_krule *rule, char *pathname, u32 op); +extern int audit_add_tree_rule(struct audit_krule *rule); +extern int audit_remove_tree_rule(struct audit_krule *rule); extern void audit_trim_trees(void); extern int audit_tag_tree(char *old, char *new); -extern const char *audit_tree_path(struct audit_tree *); -extern void audit_put_tree(struct audit_tree *); -extern void audit_kill_trees(struct list_head *); +extern const char *audit_tree_path(struct audit_tree *tree); +extern void audit_put_tree(struct audit_tree *tree); +extern void audit_kill_trees(struct list_head *list); #else #define audit_remove_tree_rule(rule) BUG() #define audit_add_tree_rule(rule) -EINVAL @@ -325,7 +324,7 @@ extern void audit_kill_trees(struct list_head *); #define audit_kill_trees(list) BUG() #endif -extern char *audit_unpack_string(void **, size_t *, size_t); +extern char *audit_unpack_string(void **bufp, size_t *remain, size_t len); extern pid_t audit_sig_pid; extern kuid_t audit_sig_uid; @@ -334,15 +333,8 @@ extern u32 audit_sig_sid; extern int audit_filter(int msgtype, unsigned int listtype); #ifdef CONFIG_AUDITSYSCALL -extern int __audit_signal_info(int sig, struct task_struct *t); -static inline int audit_signal_info(int sig, struct task_struct *t) -{ - if (unlikely((audit_pid && t->tgid == audit_pid) || - (audit_signals && !audit_dummy_context()))) - return __audit_signal_info(sig, t); - return 0; -} -extern void audit_filter_inodes(struct task_struct *, struct audit_context *); +extern int audit_signal_info(int sig, struct task_struct *t); +extern void audit_filter_inodes(struct task_struct *tsk, struct audit_context *ctx); extern struct list_head *audit_killed_trees(void); #else #define audit_signal_info(s,t) AUDIT_DISABLED diff --git a/kernel/audit_fsnotify.c b/kernel/audit_fsnotify.c index 7ea57e516029..52f368b6561e 100644 --- a/kernel/audit_fsnotify.c +++ b/kernel/audit_fsnotify.c @@ -103,15 +103,15 @@ struct audit_fsnotify_mark *audit_alloc_mark(struct audit_krule *krule, char *pa goto out; } - fsnotify_init_mark(&audit_mark->mark, audit_fsnotify_free_mark); + fsnotify_init_mark(&audit_mark->mark, audit_fsnotify_group); audit_mark->mark.mask = AUDIT_FS_EVENTS; audit_mark->path = pathname; audit_update_mark(audit_mark, dentry->d_inode); audit_mark->rule = krule; - ret = fsnotify_add_mark(&audit_mark->mark, audit_fsnotify_group, inode, NULL, true); + ret = fsnotify_add_mark(&audit_mark->mark, inode, NULL, true); if (ret < 0) { - audit_fsnotify_mark_free(audit_mark); + fsnotify_put_mark(&audit_mark->mark); audit_mark = ERR_PTR(ret); } out: @@ -168,7 +168,8 @@ static int audit_mark_handle_event(struct fsnotify_group *group, struct fsnotify_mark *inode_mark, struct fsnotify_mark *vfsmount_mark, u32 mask, const void *data, int data_type, - const unsigned char *dname, u32 cookie) + const unsigned char *dname, u32 cookie, + struct fsnotify_iter_info *iter_info) { struct audit_fsnotify_mark *audit_mark; const struct inode *inode = NULL; @@ -187,7 +188,7 @@ static int audit_mark_handle_event(struct fsnotify_group *group, default: BUG(); return 0; - }; + } if (mask & (FS_CREATE|FS_MOVED_TO|FS_DELETE|FS_MOVED_FROM)) { if (audit_compare_dname_path(dname, audit_mark->path, AUDIT_NAME_FULL)) @@ -201,6 +202,7 @@ static int audit_mark_handle_event(struct fsnotify_group *group, static const struct fsnotify_ops audit_mark_fsnotify_ops = { .handle_event = audit_mark_handle_event, + .free_mark = audit_fsnotify_free_mark, }; static int __init audit_fsnotify_init(void) diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 7b44195da81b..011d46e5f73f 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -3,13 +3,14 @@ #include <linux/namei.h> #include <linux/mount.h> #include <linux/kthread.h> +#include <linux/refcount.h> #include <linux/slab.h> struct audit_tree; struct audit_chunk; struct audit_tree { - atomic_t count; + refcount_t count; int goner; struct audit_chunk *root; struct list_head chunks; @@ -77,7 +78,7 @@ static struct audit_tree *alloc_tree(const char *s) tree = kmalloc(sizeof(struct audit_tree) + strlen(s) + 1, GFP_KERNEL); if (tree) { - atomic_set(&tree->count, 1); + refcount_set(&tree->count, 1); tree->goner = 0; INIT_LIST_HEAD(&tree->chunks); INIT_LIST_HEAD(&tree->rules); @@ -91,12 +92,12 @@ static struct audit_tree *alloc_tree(const char *s) static inline void get_tree(struct audit_tree *tree) { - atomic_inc(&tree->count); + refcount_inc(&tree->count); } static inline void put_tree(struct audit_tree *tree) { - if (atomic_dec_and_test(&tree->count)) + if (refcount_dec_and_test(&tree->count)) kfree_rcu(tree, head); } @@ -154,7 +155,7 @@ static struct audit_chunk *alloc_chunk(int count) INIT_LIST_HEAD(&chunk->owners[i].list); chunk->owners[i].index = i; } - fsnotify_init_mark(&chunk->mark, audit_tree_destroy_watch); + fsnotify_init_mark(&chunk->mark, audit_tree_group); chunk->mark.mask = FS_IN_IGNORED; return chunk; } @@ -163,33 +164,54 @@ enum {HASH_SIZE = 128}; static struct list_head chunk_hash_heads[HASH_SIZE]; static __cacheline_aligned_in_smp DEFINE_SPINLOCK(hash_lock); -static inline struct list_head *chunk_hash(const struct inode *inode) +/* Function to return search key in our hash from inode. */ +static unsigned long inode_to_key(const struct inode *inode) { - unsigned long n = (unsigned long)inode / L1_CACHE_BYTES; + return (unsigned long)inode; +} + +/* + * Function to return search key in our hash from chunk. Key 0 is special and + * should never be present in the hash. + */ +static unsigned long chunk_to_key(struct audit_chunk *chunk) +{ + /* + * We have a reference to the mark so it should be attached to a + * connector. + */ + if (WARN_ON_ONCE(!chunk->mark.connector)) + return 0; + return (unsigned long)chunk->mark.connector->inode; +} + +static inline struct list_head *chunk_hash(unsigned long key) +{ + unsigned long n = key / L1_CACHE_BYTES; return chunk_hash_heads + n % HASH_SIZE; } /* hash_lock & entry->lock is held by caller */ static void insert_hash(struct audit_chunk *chunk) { - struct fsnotify_mark *entry = &chunk->mark; + unsigned long key = chunk_to_key(chunk); struct list_head *list; - if (!entry->inode) + if (!(chunk->mark.flags & FSNOTIFY_MARK_FLAG_ATTACHED)) return; - list = chunk_hash(entry->inode); + list = chunk_hash(key); list_add_rcu(&chunk->hash, list); } /* called under rcu_read_lock */ struct audit_chunk *audit_tree_lookup(const struct inode *inode) { - struct list_head *list = chunk_hash(inode); + unsigned long key = inode_to_key(inode); + struct list_head *list = chunk_hash(key); struct audit_chunk *p; list_for_each_entry_rcu(p, list, hash) { - /* mark.inode may have gone NULL, but who cares? */ - if (p->mark.inode == inode) { + if (chunk_to_key(p) == key) { atomic_long_inc(&p->refs); return p; } @@ -233,11 +255,15 @@ static void untag_chunk(struct node *p) mutex_lock(&entry->group->mark_mutex); spin_lock(&entry->lock); - if (chunk->dead || !entry->inode) { + /* + * mark_mutex protects mark from getting detached and thus also from + * mark->connector->inode getting NULL. + */ + if (chunk->dead || !(entry->flags & FSNOTIFY_MARK_FLAG_ATTACHED)) { spin_unlock(&entry->lock); mutex_unlock(&entry->group->mark_mutex); if (new) - free_chunk(new); + fsnotify_put_mark(&new->mark); goto out; } @@ -261,7 +287,7 @@ static void untag_chunk(struct node *p) if (!new) goto Fallback; - if (fsnotify_add_mark_locked(&new->mark, entry->group, entry->inode, + if (fsnotify_add_mark_locked(&new->mark, entry->connector->inode, NULL, 1)) { fsnotify_put_mark(&new->mark); goto Fallback; @@ -327,7 +353,7 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) return -ENOMEM; entry = &chunk->mark; - if (fsnotify_add_mark(entry, audit_tree_group, inode, NULL, 0)) { + if (fsnotify_add_mark(entry, inode, NULL, 0)) { fsnotify_put_mark(entry); return -ENOSPC; } @@ -366,7 +392,8 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) struct node *p; int n; - old_entry = fsnotify_find_inode_mark(audit_tree_group, inode); + old_entry = fsnotify_find_mark(&inode->i_fsnotify_marks, + audit_tree_group); if (!old_entry) return create_chunk(inode, tree); @@ -393,17 +420,21 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) mutex_lock(&old_entry->group->mark_mutex); spin_lock(&old_entry->lock); - if (!old_entry->inode) { + /* + * mark_mutex protects mark from getting detached and thus also from + * mark->connector->inode getting NULL. + */ + if (!(old_entry->flags & FSNOTIFY_MARK_FLAG_ATTACHED)) { /* old_entry is being shot, lets just lie */ spin_unlock(&old_entry->lock); mutex_unlock(&old_entry->group->mark_mutex); fsnotify_put_mark(old_entry); - free_chunk(chunk); + fsnotify_put_mark(&chunk->mark); return -ENOENT; } - if (fsnotify_add_mark_locked(chunk_entry, old_entry->group, - old_entry->inode, NULL, 1)) { + if (fsnotify_add_mark_locked(chunk_entry, + old_entry->connector->inode, NULL, 1)) { spin_unlock(&old_entry->lock); mutex_unlock(&old_entry->group->mark_mutex); fsnotify_put_mark(chunk_entry); @@ -588,7 +619,8 @@ int audit_remove_tree_rule(struct audit_krule *rule) static int compare_root(struct vfsmount *mnt, void *arg) { - return d_backing_inode(mnt->mnt_root) == arg; + return inode_to_key(d_backing_inode(mnt->mnt_root)) == + (unsigned long)arg; } void audit_trim_trees(void) @@ -623,9 +655,10 @@ void audit_trim_trees(void) list_for_each_entry(node, &tree->chunks, list) { struct audit_chunk *chunk = find_chunk(node); /* this could be NULL if the watch is dying else where... */ - struct inode *inode = chunk->mark.inode; node->index |= 1U<<31; - if (iterate_mounts(compare_root, inode, root_mnt)) + if (iterate_mounts(compare_root, + (void *)chunk_to_key(chunk), + root_mnt)) node->index &= ~(1U<<31); } spin_unlock(&hash_lock); @@ -958,7 +991,8 @@ static int audit_tree_handle_event(struct fsnotify_group *group, struct fsnotify_mark *inode_mark, struct fsnotify_mark *vfsmount_mark, u32 mask, const void *data, int data_type, - const unsigned char *file_name, u32 cookie) + const unsigned char *file_name, u32 cookie, + struct fsnotify_iter_info *iter_info) { return 0; } @@ -979,6 +1013,7 @@ static void audit_tree_freeing_mark(struct fsnotify_mark *entry, struct fsnotify static const struct fsnotify_ops audit_tree_ops = { .handle_event = audit_tree_handle_event, .freeing_mark = audit_tree_freeing_mark, + .free_mark = audit_tree_destroy_watch, }; static int __init audit_tree_init(void) diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index f79e4658433d..62d686d96581 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -28,6 +28,7 @@ #include <linux/fsnotify_backend.h> #include <linux/namei.h> #include <linux/netlink.h> +#include <linux/refcount.h> #include <linux/sched.h> #include <linux/slab.h> #include <linux/security.h> @@ -46,7 +47,7 @@ */ struct audit_watch { - atomic_t count; /* reference count */ + refcount_t count; /* reference count */ dev_t dev; /* associated superblock device */ char *path; /* insertion path */ unsigned long ino; /* associated inode number */ @@ -102,7 +103,7 @@ static inline struct audit_parent *audit_find_parent(struct inode *inode) struct audit_parent *parent = NULL; struct fsnotify_mark *entry; - entry = fsnotify_find_inode_mark(audit_watch_group, inode); + entry = fsnotify_find_mark(&inode->i_fsnotify_marks, audit_watch_group); if (entry) parent = container_of(entry, struct audit_parent, mark); @@ -111,12 +112,12 @@ static inline struct audit_parent *audit_find_parent(struct inode *inode) void audit_get_watch(struct audit_watch *watch) { - atomic_inc(&watch->count); + refcount_inc(&watch->count); } void audit_put_watch(struct audit_watch *watch) { - if (atomic_dec_and_test(&watch->count)) { + if (refcount_dec_and_test(&watch->count)) { WARN_ON(watch->parent); WARN_ON(!list_empty(&watch->rules)); kfree(watch->path); @@ -157,9 +158,9 @@ static struct audit_parent *audit_init_parent(struct path *path) INIT_LIST_HEAD(&parent->watches); - fsnotify_init_mark(&parent->mark, audit_watch_free_mark); + fsnotify_init_mark(&parent->mark, audit_watch_group); parent->mark.mask = AUDIT_FS_WATCH; - ret = fsnotify_add_mark(&parent->mark, audit_watch_group, inode, NULL, 0); + ret = fsnotify_add_mark(&parent->mark, inode, NULL, 0); if (ret < 0) { audit_free_parent(parent); return ERR_PTR(ret); @@ -178,7 +179,7 @@ static struct audit_watch *audit_init_watch(char *path) return ERR_PTR(-ENOMEM); INIT_LIST_HEAD(&watch->rules); - atomic_set(&watch->count, 1); + refcount_set(&watch->count, 1); watch->path = path; watch->dev = AUDIT_DEV_UNSET; watch->ino = AUDIT_INO_UNSET; @@ -472,7 +473,8 @@ static int audit_watch_handle_event(struct fsnotify_group *group, struct fsnotify_mark *inode_mark, struct fsnotify_mark *vfsmount_mark, u32 mask, const void *data, int data_type, - const unsigned char *dname, u32 cookie) + const unsigned char *dname, u32 cookie, + struct fsnotify_iter_info *iter_info) { const struct inode *inode; struct audit_parent *parent; @@ -492,7 +494,7 @@ static int audit_watch_handle_event(struct fsnotify_group *group, BUG(); inode = NULL; break; - }; + } if (mask & (FS_CREATE|FS_MOVED_TO) && inode) audit_update_watch(parent, dname, inode->i_sb->s_dev, inode->i_ino, 0); @@ -506,6 +508,7 @@ static int audit_watch_handle_event(struct fsnotify_group *group, static const struct fsnotify_ops audit_watch_fsnotify_ops = { .handle_event = audit_watch_handle_event, + .free_mark = audit_watch_free_mark, }; static int __init audit_watch_init(void) diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index 880519d6cf2a..0b0aa5854dac 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -338,7 +338,7 @@ static int audit_field_valid(struct audit_entry *entry, struct audit_field *f) entry->rule.listnr != AUDIT_FILTER_USER) return -EINVAL; break; - }; + } switch(f->type) { default: @@ -412,7 +412,7 @@ static int audit_field_valid(struct audit_entry *entry, struct audit_field *f) if (entry->rule.listnr != AUDIT_FILTER_EXIT) return -EINVAL; break; - }; + } return 0; } @@ -1033,7 +1033,7 @@ out: } /* List rules using struct audit_rule_data. */ -static void audit_list_rules(__u32 portid, int seq, struct sk_buff_head *q) +static void audit_list_rules(int seq, struct sk_buff_head *q) { struct sk_buff *skb; struct audit_krule *r; @@ -1048,15 +1048,15 @@ static void audit_list_rules(__u32 portid, int seq, struct sk_buff_head *q) data = audit_krule_to_data(r); if (unlikely(!data)) break; - skb = audit_make_reply(portid, seq, AUDIT_LIST_RULES, - 0, 1, data, + skb = audit_make_reply(seq, AUDIT_LIST_RULES, 0, 1, + data, sizeof(*data) + data->buflen); if (skb) skb_queue_tail(q, skb); kfree(data); } } - skb = audit_make_reply(portid, seq, AUDIT_LIST_RULES, 1, 1, NULL, 0); + skb = audit_make_reply(seq, AUDIT_LIST_RULES, 1, 1, NULL, 0); if (skb) skb_queue_tail(q, skb); } @@ -1085,13 +1085,11 @@ static void audit_log_rule_change(char *action, struct audit_krule *rule, int re /** * audit_rule_change - apply all rules to the specified message type * @type: audit message type - * @portid: target port id for netlink audit messages * @seq: netlink audit message sequence (serial) number * @data: payload data * @datasz: size of payload data */ -int audit_rule_change(int type, __u32 portid, int seq, void *data, - size_t datasz) +int audit_rule_change(int type, int seq, void *data, size_t datasz) { int err = 0; struct audit_entry *entry; @@ -1150,7 +1148,7 @@ int audit_list_rules_send(struct sk_buff *request_skb, int seq) skb_queue_head_init(&dest->q); mutex_lock(&audit_filter_mutex); - audit_list_rules(portid, seq, &dest->q); + audit_list_rules(seq, &dest->q); mutex_unlock(&audit_filter_mutex); tsk = kthread_run(audit_send_list, dest, "audit_send_list"); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index cf1fa43512c1..3260ba2312a9 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -73,6 +73,7 @@ #include <linux/ctype.h> #include <linux/string.h> #include <linux/uaccess.h> +#include <linux/fsnotify_backend.h> #include <uapi/linux/limits.h> #include "audit.h" @@ -762,7 +763,7 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk, struct audit_entry *e; enum audit_state state; - if (audit_pid && tsk->tgid == audit_pid) + if (auditd_test_task(tsk)) return AUDIT_DISABLED; rcu_read_lock(); @@ -816,7 +817,7 @@ void audit_filter_inodes(struct task_struct *tsk, struct audit_context *ctx) { struct audit_names *n; - if (audit_pid && tsk->tgid == audit_pid) + if (auditd_test_task(tsk)) return; rcu_read_lock(); @@ -1221,7 +1222,7 @@ static void show_special(struct audit_context *context, int *call_panic) context->ipc.perm_mode); } break; } - case AUDIT_MQ_OPEN: { + case AUDIT_MQ_OPEN: audit_log_format(ab, "oflag=0x%x mode=%#ho mq_flags=0x%lx mq_maxmsg=%ld " "mq_msgsize=%ld mq_curmsgs=%ld", @@ -1230,8 +1231,8 @@ static void show_special(struct audit_context *context, int *call_panic) context->mq_open.attr.mq_maxmsg, context->mq_open.attr.mq_msgsize, context->mq_open.attr.mq_curmsgs); - break; } - case AUDIT_MQ_SENDRECV: { + break; + case AUDIT_MQ_SENDRECV: audit_log_format(ab, "mqdes=%d msg_len=%zd msg_prio=%u " "abs_timeout_sec=%ld abs_timeout_nsec=%ld", @@ -1240,12 +1241,12 @@ static void show_special(struct audit_context *context, int *call_panic) context->mq_sendrecv.msg_prio, context->mq_sendrecv.abs_timeout.tv_sec, context->mq_sendrecv.abs_timeout.tv_nsec); - break; } - case AUDIT_MQ_NOTIFY: { + break; + case AUDIT_MQ_NOTIFY: audit_log_format(ab, "mqdes=%d sigev_signo=%d", context->mq_notify.mqdes, context->mq_notify.sigev_signo); - break; } + break; case AUDIT_MQ_GETSETATTR: { struct mq_attr *attr = &context->mq_getsetattr.mqstat; audit_log_format(ab, @@ -1255,19 +1256,25 @@ static void show_special(struct audit_context *context, int *call_panic) attr->mq_flags, attr->mq_maxmsg, attr->mq_msgsize, attr->mq_curmsgs); break; } - case AUDIT_CAPSET: { + case AUDIT_CAPSET: audit_log_format(ab, "pid=%d", context->capset.pid); audit_log_cap(ab, "cap_pi", &context->capset.cap.inheritable); audit_log_cap(ab, "cap_pp", &context->capset.cap.permitted); audit_log_cap(ab, "cap_pe", &context->capset.cap.effective); - break; } - case AUDIT_MMAP: { + audit_log_cap(ab, "cap_pa", &context->capset.cap.ambient); + break; + case AUDIT_MMAP: audit_log_format(ab, "fd=%d flags=0x%x", context->mmap.fd, context->mmap.flags); - break; } - case AUDIT_EXECVE: { + break; + case AUDIT_EXECVE: audit_log_execve_info(context, &ab); - break; } + break; + case AUDIT_KERN_MODULE: + audit_log_format(ab, "name="); + audit_log_untrustedstring(ab, context->module.name); + kfree(context->module.name); + break; } audit_log_end(ab); } @@ -1376,9 +1383,11 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_cap(ab, "old_pp", &axs->old_pcap.permitted); audit_log_cap(ab, "old_pi", &axs->old_pcap.inheritable); audit_log_cap(ab, "old_pe", &axs->old_pcap.effective); - audit_log_cap(ab, "new_pp", &axs->new_pcap.permitted); - audit_log_cap(ab, "new_pi", &axs->new_pcap.inheritable); - audit_log_cap(ab, "new_pe", &axs->new_pcap.effective); + audit_log_cap(ab, "old_pa", &axs->old_pcap.ambient); + audit_log_cap(ab, "pp", &axs->new_pcap.permitted); + audit_log_cap(ab, "pi", &axs->new_pcap.inheritable); + audit_log_cap(ab, "pe", &axs->new_pcap.effective); + audit_log_cap(ab, "pa", &axs->new_pcap.ambient); break; } } @@ -1527,7 +1536,7 @@ void __audit_syscall_entry(int major, unsigned long a1, unsigned long a2, return; context->serial = 0; - context->ctime = CURRENT_TIME; + ktime_get_real_ts64(&context->ctime); context->in_syscall = 1; context->current_state = state; context->ppid = 0; @@ -1591,7 +1600,7 @@ static inline void handle_one(const struct inode *inode) struct audit_tree_refs *p; struct audit_chunk *chunk; int count; - if (likely(hlist_empty(&inode->i_fsnotify_marks))) + if (likely(!inode->i_fsnotify_marks)) return; context = current->audit_context; p = context->trees; @@ -1634,7 +1643,7 @@ retry: seq = read_seqbegin(&rename_lock); for(;;) { struct inode *inode = d_backing_inode(d); - if (inode && unlikely(!hlist_empty(&inode->i_fsnotify_marks))) { + if (inode && unlikely(inode->i_fsnotify_marks)) { struct audit_chunk *chunk; chunk = audit_tree_lookup(inode); if (chunk) { @@ -1936,13 +1945,13 @@ EXPORT_SYMBOL_GPL(__audit_inode_child); /** * auditsc_get_stamp - get local copies of audit_context values * @ctx: audit_context for the task - * @t: timespec to store time recorded in the audit_context + * @t: timespec64 to store time recorded in the audit_context * @serial: serial value that is recorded in the audit_context * * Also sets the context as auditable. */ int auditsc_get_stamp(struct audit_context *ctx, - struct timespec *t, unsigned int *serial) + struct timespec64 *t, unsigned int *serial) { if (!ctx->in_syscall) return 0; @@ -2244,26 +2253,27 @@ void __audit_ptrace(struct task_struct *t) * If the audit subsystem is being terminated, record the task (pid) * and uid that is doing that. */ -int __audit_signal_info(int sig, struct task_struct *t) +int audit_signal_info(int sig, struct task_struct *t) { struct audit_aux_data_pids *axp; struct task_struct *tsk = current; struct audit_context *ctx = tsk->audit_context; kuid_t uid = current_uid(), t_uid = task_uid(t); - if (audit_pid && t->tgid == audit_pid) { - if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1 || sig == SIGUSR2) { - audit_sig_pid = task_tgid_nr(tsk); - if (uid_valid(tsk->loginuid)) - audit_sig_uid = tsk->loginuid; - else - audit_sig_uid = uid; - security_task_getsecid(tsk, &audit_sig_sid); - } - if (!audit_signals || audit_dummy_context()) - return 0; + if (auditd_test_task(t) && + (sig == SIGTERM || sig == SIGHUP || + sig == SIGUSR1 || sig == SIGUSR2)) { + audit_sig_pid = task_tgid_nr(tsk); + if (uid_valid(tsk->loginuid)) + audit_sig_uid = tsk->loginuid; + else + audit_sig_uid = uid; + security_task_getsecid(tsk, &audit_sig_sid); } + if (!audit_signals || audit_dummy_context()) + return 0; + /* optimize the common case by putting first signal recipient directly * in audit_context */ if (!ctx->target_pid) { @@ -2335,10 +2345,12 @@ int __audit_log_bprm_fcaps(struct linux_binprm *bprm, ax->old_pcap.permitted = old->cap_permitted; ax->old_pcap.inheritable = old->cap_inheritable; ax->old_pcap.effective = old->cap_effective; + ax->old_pcap.ambient = old->cap_ambient; ax->new_pcap.permitted = new->cap_permitted; ax->new_pcap.inheritable = new->cap_inheritable; ax->new_pcap.effective = new->cap_effective; + ax->new_pcap.ambient = new->cap_ambient; return 0; } @@ -2357,6 +2369,7 @@ void __audit_log_capset(const struct cred *new, const struct cred *old) context->capset.cap.effective = new->cap_effective; context->capset.cap.inheritable = new->cap_effective; context->capset.cap.permitted = new->cap_permitted; + context->capset.cap.ambient = new->cap_ambient; context->type = AUDIT_CAPSET; } @@ -2368,6 +2381,15 @@ void __audit_mmap_fd(int fd, int flags) context->type = AUDIT_MMAP; } +void __audit_log_kern_module(char *name) +{ + struct audit_context *context = current->audit_context; + + context->module.name = kmalloc(strlen(name) + 1, GFP_KERNEL); + strcpy(context->module.name, name); + context->type = AUDIT_KERN_MODULE; +} + static void audit_log_task(struct audit_buffer *ab) { kuid_t auid, uid; @@ -2411,7 +2433,7 @@ void audit_core_dumps(long signr) if (unlikely(!ab)) return; audit_log_task(ab); - audit_log_format(ab, " sig=%ld", signr); + audit_log_format(ab, " sig=%ld res=1", signr); audit_log_end(ab); } diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile index 1276474ac3cd..e1e5e658f2db 100644 --- a/kernel/bpf/Makefile +++ b/kernel/bpf/Makefile @@ -1,7 +1,7 @@ obj-y := core.o obj-$(CONFIG_BPF_SYSCALL) += syscall.o verifier.o inode.o helpers.o -obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o percpu_freelist.o bpf_lru_list.o +obj-$(CONFIG_BPF_SYSCALL) += hashtab.o arraymap.o percpu_freelist.o bpf_lru_list.o lpm_trie.o map_in_map.o ifeq ($(CONFIG_PERF_EVENTS),y) obj-$(CONFIG_BPF_SYSCALL) += stackmap.o endif diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 3d55d95dcf49..d771a3872500 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -1,4 +1,5 @@ /* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com + * Copyright (c) 2016,2017 Facebook * * This program is free software; you can redistribute it and/or * modify it under the terms of version 2 of the GNU General Public @@ -16,6 +17,8 @@ #include <linux/filter.h> #include <linux/perf_event.h> +#include "map_in_map.h" + static void bpf_array_free_percpu(struct bpf_array *array) { int i; @@ -83,6 +86,7 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) array->map.key_size = attr->key_size; array->map.value_size = attr->value_size; array->map.max_entries = attr->max_entries; + array->map.map_flags = attr->map_flags; array->elem_size = elem_size; if (!percpu) @@ -113,6 +117,30 @@ static void *array_map_lookup_elem(struct bpf_map *map, void *key) return array->value + array->elem_size * index; } +/* emit BPF instructions equivalent to C code of array_map_lookup_elem() */ +static u32 array_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) +{ + struct bpf_insn *insn = insn_buf; + u32 elem_size = round_up(map->value_size, 8); + const int ret = BPF_REG_0; + const int map_ptr = BPF_REG_1; + const int index = BPF_REG_2; + + *insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value)); + *insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0); + *insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 3); + + if (is_power_of_2(elem_size)) { + *insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size)); + } else { + *insn++ = BPF_ALU64_IMM(BPF_MUL, ret, elem_size); + } + *insn++ = BPF_ALU64_REG(BPF_ADD, ret, map_ptr); + *insn++ = BPF_JMP_IMM(BPF_JA, 0, 0, 1); + *insn++ = BPF_MOV64_IMM(ret, 0); + return insn - insn_buf; +} + /* Called from eBPF program */ static void *percpu_array_map_lookup_elem(struct bpf_map *map, void *key) { @@ -155,7 +183,7 @@ int bpf_percpu_array_copy(struct bpf_map *map, void *key, void *value) static int array_map_get_next_key(struct bpf_map *map, void *key, void *next_key) { struct bpf_array *array = container_of(map, struct bpf_array, map); - u32 index = *(u32 *)key; + u32 index = key ? *(u32 *)key : U32_MAX; u32 *next = (u32 *)next_key; if (index >= array->map.max_entries) { @@ -260,21 +288,17 @@ static void array_map_free(struct bpf_map *map) bpf_map_area_free(array); } -static const struct bpf_map_ops array_ops = { +const struct bpf_map_ops array_map_ops = { .map_alloc = array_map_alloc, .map_free = array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = array_map_lookup_elem, .map_update_elem = array_map_update_elem, .map_delete_elem = array_map_delete_elem, + .map_gen_lookup = array_map_gen_lookup, }; -static struct bpf_map_type_list array_type __read_mostly = { - .ops = &array_ops, - .type = BPF_MAP_TYPE_ARRAY, -}; - -static const struct bpf_map_ops percpu_array_ops = { +const struct bpf_map_ops percpu_array_map_ops = { .map_alloc = array_map_alloc, .map_free = array_map_free, .map_get_next_key = array_map_get_next_key, @@ -283,19 +307,6 @@ static const struct bpf_map_ops percpu_array_ops = { .map_delete_elem = array_map_delete_elem, }; -static struct bpf_map_type_list percpu_array_type __read_mostly = { - .ops = &percpu_array_ops, - .type = BPF_MAP_TYPE_PERCPU_ARRAY, -}; - -static int __init register_array_map(void) -{ - bpf_register_map_type(&array_type); - bpf_register_map_type(&percpu_array_type); - return 0; -} -late_initcall(register_array_map); - static struct bpf_map *fd_array_map_alloc(union bpf_attr *attr) { /* only file descriptors can be stored in this type of map */ @@ -324,6 +335,26 @@ static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key) } /* only called from syscall */ +int bpf_fd_array_map_lookup_elem(struct bpf_map *map, void *key, u32 *value) +{ + void **elem, *ptr; + int ret = 0; + + if (!map->ops->map_fd_sys_lookup_elem) + return -ENOTSUPP; + + rcu_read_lock(); + elem = array_map_lookup_elem(map, key); + if (elem && (ptr = READ_ONCE(*elem))) + *value = map->ops->map_fd_sys_lookup_elem(ptr); + else + ret = -ENOENT; + rcu_read_unlock(); + + return ret; +} + +/* only called from syscall */ int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, void *key, void *value, u64 map_flags) { @@ -389,6 +420,11 @@ static void prog_fd_array_put_ptr(void *ptr) bpf_prog_put(ptr); } +static u32 prog_fd_array_sys_lookup_elem(void *ptr) +{ + return ((struct bpf_prog *)ptr)->aux->id; +} + /* decrement refcnt of all bpf_progs that are stored in this map */ void bpf_fd_array_map_clear(struct bpf_map *map) { @@ -399,7 +435,7 @@ void bpf_fd_array_map_clear(struct bpf_map *map) fd_array_map_delete_elem(map, &i); } -static const struct bpf_map_ops prog_array_ops = { +const struct bpf_map_ops prog_array_map_ops = { .map_alloc = fd_array_map_alloc, .map_free = fd_array_map_free, .map_get_next_key = array_map_get_next_key, @@ -407,20 +443,9 @@ static const struct bpf_map_ops prog_array_ops = { .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = prog_fd_array_get_ptr, .map_fd_put_ptr = prog_fd_array_put_ptr, + .map_fd_sys_lookup_elem = prog_fd_array_sys_lookup_elem, }; -static struct bpf_map_type_list prog_array_type __read_mostly = { - .ops = &prog_array_ops, - .type = BPF_MAP_TYPE_PROG_ARRAY, -}; - -static int __init register_prog_array_map(void) -{ - bpf_register_map_type(&prog_array_type); - return 0; -} -late_initcall(register_prog_array_map); - static struct bpf_event_entry *bpf_event_entry_gen(struct file *perf_file, struct file *map_file) { @@ -453,38 +478,24 @@ static void bpf_event_entry_free_rcu(struct bpf_event_entry *ee) static void *perf_event_fd_array_get_ptr(struct bpf_map *map, struct file *map_file, int fd) { - const struct perf_event_attr *attr; struct bpf_event_entry *ee; struct perf_event *event; struct file *perf_file; + u64 value; perf_file = perf_event_get(fd); if (IS_ERR(perf_file)) return perf_file; + ee = ERR_PTR(-EOPNOTSUPP); event = perf_file->private_data; - ee = ERR_PTR(-EINVAL); - - attr = perf_event_attrs(event); - if (IS_ERR(attr) || attr->inherit) + if (perf_event_read_local(event, &value) == -EOPNOTSUPP) goto err_out; - switch (attr->type) { - case PERF_TYPE_SOFTWARE: - if (attr->config != PERF_COUNT_SW_BPF_OUTPUT) - goto err_out; - /* fall-through */ - case PERF_TYPE_RAW: - case PERF_TYPE_HARDWARE: - ee = bpf_event_entry_gen(perf_file, map_file); - if (ee) - return ee; - ee = ERR_PTR(-ENOMEM); - /* fall-through */ - default: - break; - } - + ee = bpf_event_entry_gen(perf_file, map_file); + if (ee) + return ee; + ee = ERR_PTR(-ENOMEM); err_out: fput(perf_file); return ee; @@ -511,7 +522,7 @@ static void perf_event_fd_array_release(struct bpf_map *map, rcu_read_unlock(); } -static const struct bpf_map_ops perf_event_array_ops = { +const struct bpf_map_ops perf_event_array_map_ops = { .map_alloc = fd_array_map_alloc, .map_free = fd_array_map_free, .map_get_next_key = array_map_get_next_key, @@ -522,18 +533,6 @@ static const struct bpf_map_ops perf_event_array_ops = { .map_release = perf_event_fd_array_release, }; -static struct bpf_map_type_list perf_event_array_type __read_mostly = { - .ops = &perf_event_array_ops, - .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, -}; - -static int __init register_perf_event_array_map(void) -{ - bpf_register_map_type(&perf_event_array_type); - return 0; -} -late_initcall(register_perf_event_array_map); - #ifdef CONFIG_CGROUPS static void *cgroup_fd_array_get_ptr(struct bpf_map *map, struct file *map_file /* not used */, @@ -554,7 +553,7 @@ static void cgroup_fd_array_free(struct bpf_map *map) fd_array_map_free(map); } -static const struct bpf_map_ops cgroup_array_ops = { +const struct bpf_map_ops cgroup_array_map_ops = { .map_alloc = fd_array_map_alloc, .map_free = cgroup_fd_array_free, .map_get_next_key = array_map_get_next_key, @@ -563,16 +562,54 @@ static const struct bpf_map_ops cgroup_array_ops = { .map_fd_get_ptr = cgroup_fd_array_get_ptr, .map_fd_put_ptr = cgroup_fd_array_put_ptr, }; +#endif -static struct bpf_map_type_list cgroup_array_type __read_mostly = { - .ops = &cgroup_array_ops, - .type = BPF_MAP_TYPE_CGROUP_ARRAY, -}; +static struct bpf_map *array_of_map_alloc(union bpf_attr *attr) +{ + struct bpf_map *map, *inner_map_meta; + + inner_map_meta = bpf_map_meta_alloc(attr->inner_map_fd); + if (IS_ERR(inner_map_meta)) + return inner_map_meta; + + map = fd_array_map_alloc(attr); + if (IS_ERR(map)) { + bpf_map_meta_free(inner_map_meta); + return map; + } + + map->inner_map_meta = inner_map_meta; -static int __init register_cgroup_array_map(void) + return map; +} + +static void array_of_map_free(struct bpf_map *map) { - bpf_register_map_type(&cgroup_array_type); - return 0; + /* map->inner_map_meta is only accessed by syscall which + * is protected by fdget/fdput. + */ + bpf_map_meta_free(map->inner_map_meta); + bpf_fd_array_map_clear(map); + fd_array_map_free(map); } -late_initcall(register_cgroup_array_map); -#endif + +static void *array_of_map_lookup_elem(struct bpf_map *map, void *key) +{ + struct bpf_map **inner_map = array_map_lookup_elem(map, key); + + if (!inner_map) + return NULL; + + return READ_ONCE(*inner_map); +} + +const struct bpf_map_ops array_of_maps_map_ops = { + .map_alloc = array_of_map_alloc, + .map_free = array_of_map_free, + .map_get_next_key = array_map_get_next_key, + .map_lookup_elem = array_of_map_lookup_elem, + .map_delete_elem = fd_array_map_delete_elem, + .map_fd_get_ptr = bpf_map_fd_get_ptr, + .map_fd_put_ptr = bpf_map_fd_put_ptr, + .map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem, +}; diff --git a/kernel/bpf/bpf_lru_list.c b/kernel/bpf/bpf_lru_list.c index 89b7ef41c86b..e6ef4401a138 100644 --- a/kernel/bpf/bpf_lru_list.c +++ b/kernel/bpf/bpf_lru_list.c @@ -13,7 +13,7 @@ #define LOCAL_FREE_TARGET (128) #define LOCAL_NR_SCANS LOCAL_FREE_TARGET -#define PERCPU_FREE_TARGET (16) +#define PERCPU_FREE_TARGET (4) #define PERCPU_NR_SCANS PERCPU_FREE_TARGET /* Helpers to get the local list index */ @@ -213,11 +213,10 @@ __bpf_lru_list_shrink_inactive(struct bpf_lru *lru, enum bpf_lru_list_type tgt_free_type) { struct list_head *inactive = &l->lists[BPF_LRU_LIST_T_INACTIVE]; - struct bpf_lru_node *node, *tmp_node, *first_node; + struct bpf_lru_node *node, *tmp_node; unsigned int nshrinked = 0; unsigned int i = 0; - first_node = list_first_entry(inactive, struct bpf_lru_node, list); list_for_each_entry_safe_reverse(node, tmp_node, inactive, list) { if (bpf_lru_node_is_ref(node)) { __bpf_lru_node_move(l, node, BPF_LRU_LIST_T_ACTIVE); @@ -361,7 +360,8 @@ static void __local_list_add_pending(struct bpf_lru *lru, list_add(&node->list, local_pending_list(loc_l)); } -struct bpf_lru_node *__local_list_pop_free(struct bpf_lru_locallist *loc_l) +static struct bpf_lru_node * +__local_list_pop_free(struct bpf_lru_locallist *loc_l) { struct bpf_lru_node *node; @@ -374,8 +374,8 @@ struct bpf_lru_node *__local_list_pop_free(struct bpf_lru_locallist *loc_l) return node; } -struct bpf_lru_node *__local_list_pop_pending(struct bpf_lru *lru, - struct bpf_lru_locallist *loc_l) +static struct bpf_lru_node * +__local_list_pop_pending(struct bpf_lru *lru, struct bpf_lru_locallist *loc_l) { struct bpf_lru_node *node; bool force = false; @@ -558,8 +558,9 @@ void bpf_lru_push_free(struct bpf_lru *lru, struct bpf_lru_node *node) bpf_common_lru_push_free(lru, node); } -void bpf_common_lru_populate(struct bpf_lru *lru, void *buf, u32 node_offset, - u32 elem_size, u32 nr_elems) +static void bpf_common_lru_populate(struct bpf_lru *lru, void *buf, + u32 node_offset, u32 elem_size, + u32 nr_elems) { struct bpf_lru_list *l = &lru->common_lru.lru_list; u32 i; @@ -575,8 +576,9 @@ void bpf_common_lru_populate(struct bpf_lru *lru, void *buf, u32 node_offset, } } -void bpf_percpu_lru_populate(struct bpf_lru *lru, void *buf, u32 node_offset, - u32 elem_size, u32 nr_elems) +static void bpf_percpu_lru_populate(struct bpf_lru *lru, void *buf, + u32 node_offset, u32 elem_size, + u32 nr_elems) { u32 i, pcpu_entries; int cpu; diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c index a515f7b007c6..546113430049 100644 --- a/kernel/bpf/cgroup.c +++ b/kernel/bpf/cgroup.c @@ -52,6 +52,7 @@ void cgroup_bpf_inherit(struct cgroup *cgrp, struct cgroup *parent) e = rcu_dereference_protected(parent->bpf.effective[type], lockdep_is_held(&cgroup_mutex)); rcu_assign_pointer(cgrp->bpf.effective[type], e); + cgrp->bpf.disallow_override[type] = parent->bpf.disallow_override[type]; } } @@ -82,30 +83,63 @@ void cgroup_bpf_inherit(struct cgroup *cgrp, struct cgroup *parent) * * Must be called with cgroup_mutex held. */ -void __cgroup_bpf_update(struct cgroup *cgrp, - struct cgroup *parent, - struct bpf_prog *prog, - enum bpf_attach_type type) +int __cgroup_bpf_update(struct cgroup *cgrp, struct cgroup *parent, + struct bpf_prog *prog, enum bpf_attach_type type, + bool new_overridable) { - struct bpf_prog *old_prog, *effective; + struct bpf_prog *old_prog, *effective = NULL; struct cgroup_subsys_state *pos; + bool overridable = true; - old_prog = xchg(cgrp->bpf.prog + type, prog); + if (parent) { + overridable = !parent->bpf.disallow_override[type]; + effective = rcu_dereference_protected(parent->bpf.effective[type], + lockdep_is_held(&cgroup_mutex)); + } + + if (prog && effective && !overridable) + /* if parent has non-overridable prog attached, disallow + * attaching new programs to descendent cgroup + */ + return -EPERM; + + if (prog && effective && overridable != new_overridable) + /* if parent has overridable prog attached, only + * allow overridable programs in descendent cgroup + */ + return -EPERM; - effective = (!prog && parent) ? - rcu_dereference_protected(parent->bpf.effective[type], - lockdep_is_held(&cgroup_mutex)) : - prog; + old_prog = cgrp->bpf.prog[type]; + + if (prog) { + overridable = new_overridable; + effective = prog; + if (old_prog && + cgrp->bpf.disallow_override[type] == new_overridable) + /* disallow attaching non-overridable on top + * of existing overridable in this cgroup + * and vice versa + */ + return -EPERM; + } + + if (!prog && !old_prog) + /* report error when trying to detach and nothing is attached */ + return -ENOENT; + + cgrp->bpf.prog[type] = prog; css_for_each_descendant_pre(pos, &cgrp->self) { struct cgroup *desc = container_of(pos, struct cgroup, self); /* skip the subtree if the descendant has its own program */ - if (desc->bpf.prog[type] && desc != cgrp) + if (desc->bpf.prog[type] && desc != cgrp) { pos = css_rightmost_descendant(pos); - else + } else { rcu_assign_pointer(desc->bpf.effective[type], effective); + desc->bpf.disallow_override[type] = !overridable; + } } if (prog) @@ -115,11 +149,12 @@ void __cgroup_bpf_update(struct cgroup *cgrp, bpf_prog_put(old_prog); static_branch_dec(&cgroup_bpf_enabled_key); } + return 0; } /** * __cgroup_bpf_run_filter_skb() - Run a program for packet filtering - * @sk: The socken sending or receiving traffic + * @sk: The socket sending or receiving traffic * @skb: The skb that is being sent or received * @type: The type of program to be exectuted * @@ -154,10 +189,13 @@ int __cgroup_bpf_run_filter_skb(struct sock *sk, prog = rcu_dereference(cgrp->bpf.effective[type]); if (prog) { unsigned int offset = skb->data - skb_network_header(skb); + struct sock *save_sk = skb->sk; + skb->sk = sk; __skb_push(skb, offset); ret = bpf_prog_run_save_cb(prog, skb) == 1 ? 0 : -EPERM; __skb_pull(skb, offset); + skb->sk = save_sk; } rcu_read_unlock(); @@ -198,3 +236,40 @@ int __cgroup_bpf_run_filter_sk(struct sock *sk, return ret; } EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk); + +/** + * __cgroup_bpf_run_filter_sock_ops() - Run a program on a sock + * @sk: socket to get cgroup from + * @sock_ops: bpf_sock_ops_kern struct to pass to program. Contains + * sk with connection information (IP addresses, etc.) May not contain + * cgroup info if it is a req sock. + * @type: The type of program to be exectuted + * + * socket passed is expected to be of type INET or INET6. + * + * The program type passed in via @type must be suitable for sock_ops + * filtering. No further check is performed to assert that. + * + * This function will return %-EPERM if any if an attached program was found + * and if it returned != 1 during execution. In all other cases, 0 is returned. + */ +int __cgroup_bpf_run_filter_sock_ops(struct sock *sk, + struct bpf_sock_ops_kern *sock_ops, + enum bpf_attach_type type) +{ + struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data); + struct bpf_prog *prog; + int ret = 0; + + + rcu_read_lock(); + + prog = rcu_dereference(cgrp->bpf.effective[type]); + if (prog) + ret = BPF_PROG_RUN(prog, sock_ops) == 1 ? 0 : -EPERM; + + rcu_read_unlock(); + + return ret; +} +EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_ops); diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 503d4211988a..ad5f55922a13 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -28,6 +28,9 @@ #include <linux/moduleloader.h> #include <linux/bpf.h> #include <linux/frame.h> +#include <linux/rbtree_latch.h> +#include <linux/kallsyms.h> +#include <linux/rcupdate.h> #include <asm/unaligned.h> @@ -73,8 +76,7 @@ void *bpf_internal_load_pointer_neg_helper(const struct sk_buff *skb, int k, uns struct bpf_prog *bpf_prog_alloc(unsigned int size, gfp_t gfp_extra_flags) { - gfp_t gfp_flags = GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO | - gfp_extra_flags; + gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO | gfp_extra_flags; struct bpf_prog_aux *aux; struct bpf_prog *fp; @@ -95,6 +97,8 @@ struct bpf_prog *bpf_prog_alloc(unsigned int size, gfp_t gfp_extra_flags) fp->aux = aux; fp->aux->prog = fp; + INIT_LIST_HEAD_RCU(&fp->aux->ksym_lnode); + return fp; } EXPORT_SYMBOL_GPL(bpf_prog_alloc); @@ -102,8 +106,7 @@ EXPORT_SYMBOL_GPL(bpf_prog_alloc); struct bpf_prog *bpf_prog_realloc(struct bpf_prog *fp_old, unsigned int size, gfp_t gfp_extra_flags) { - gfp_t gfp_flags = GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO | - gfp_extra_flags; + gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO | gfp_extra_flags; struct bpf_prog *fp; u32 pages, delta; int ret; @@ -290,6 +293,202 @@ struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, } #ifdef CONFIG_BPF_JIT +static __always_inline void +bpf_get_prog_addr_region(const struct bpf_prog *prog, + unsigned long *symbol_start, + unsigned long *symbol_end) +{ + const struct bpf_binary_header *hdr = bpf_jit_binary_hdr(prog); + unsigned long addr = (unsigned long)hdr; + + WARN_ON_ONCE(!bpf_prog_ebpf_jited(prog)); + + *symbol_start = addr; + *symbol_end = addr + hdr->pages * PAGE_SIZE; +} + +static void bpf_get_prog_name(const struct bpf_prog *prog, char *sym) +{ + BUILD_BUG_ON(sizeof("bpf_prog_") + + sizeof(prog->tag) * 2 + 1 > KSYM_NAME_LEN); + + sym += snprintf(sym, KSYM_NAME_LEN, "bpf_prog_"); + sym = bin2hex(sym, prog->tag, sizeof(prog->tag)); + *sym = 0; +} + +static __always_inline unsigned long +bpf_get_prog_addr_start(struct latch_tree_node *n) +{ + unsigned long symbol_start, symbol_end; + const struct bpf_prog_aux *aux; + + aux = container_of(n, struct bpf_prog_aux, ksym_tnode); + bpf_get_prog_addr_region(aux->prog, &symbol_start, &symbol_end); + + return symbol_start; +} + +static __always_inline bool bpf_tree_less(struct latch_tree_node *a, + struct latch_tree_node *b) +{ + return bpf_get_prog_addr_start(a) < bpf_get_prog_addr_start(b); +} + +static __always_inline int bpf_tree_comp(void *key, struct latch_tree_node *n) +{ + unsigned long val = (unsigned long)key; + unsigned long symbol_start, symbol_end; + const struct bpf_prog_aux *aux; + + aux = container_of(n, struct bpf_prog_aux, ksym_tnode); + bpf_get_prog_addr_region(aux->prog, &symbol_start, &symbol_end); + + if (val < symbol_start) + return -1; + if (val >= symbol_end) + return 1; + + return 0; +} + +static const struct latch_tree_ops bpf_tree_ops = { + .less = bpf_tree_less, + .comp = bpf_tree_comp, +}; + +static DEFINE_SPINLOCK(bpf_lock); +static LIST_HEAD(bpf_kallsyms); +static struct latch_tree_root bpf_tree __cacheline_aligned; + +int bpf_jit_kallsyms __read_mostly; + +static void bpf_prog_ksym_node_add(struct bpf_prog_aux *aux) +{ + WARN_ON_ONCE(!list_empty(&aux->ksym_lnode)); + list_add_tail_rcu(&aux->ksym_lnode, &bpf_kallsyms); + latch_tree_insert(&aux->ksym_tnode, &bpf_tree, &bpf_tree_ops); +} + +static void bpf_prog_ksym_node_del(struct bpf_prog_aux *aux) +{ + if (list_empty(&aux->ksym_lnode)) + return; + + latch_tree_erase(&aux->ksym_tnode, &bpf_tree, &bpf_tree_ops); + list_del_rcu(&aux->ksym_lnode); +} + +static bool bpf_prog_kallsyms_candidate(const struct bpf_prog *fp) +{ + return fp->jited && !bpf_prog_was_classic(fp); +} + +static bool bpf_prog_kallsyms_verify_off(const struct bpf_prog *fp) +{ + return list_empty(&fp->aux->ksym_lnode) || + fp->aux->ksym_lnode.prev == LIST_POISON2; +} + +void bpf_prog_kallsyms_add(struct bpf_prog *fp) +{ + if (!bpf_prog_kallsyms_candidate(fp) || + !capable(CAP_SYS_ADMIN)) + return; + + spin_lock_bh(&bpf_lock); + bpf_prog_ksym_node_add(fp->aux); + spin_unlock_bh(&bpf_lock); +} + +void bpf_prog_kallsyms_del(struct bpf_prog *fp) +{ + if (!bpf_prog_kallsyms_candidate(fp)) + return; + + spin_lock_bh(&bpf_lock); + bpf_prog_ksym_node_del(fp->aux); + spin_unlock_bh(&bpf_lock); +} + +static struct bpf_prog *bpf_prog_kallsyms_find(unsigned long addr) +{ + struct latch_tree_node *n; + + if (!bpf_jit_kallsyms_enabled()) + return NULL; + + n = latch_tree_find((void *)addr, &bpf_tree, &bpf_tree_ops); + return n ? + container_of(n, struct bpf_prog_aux, ksym_tnode)->prog : + NULL; +} + +const char *__bpf_address_lookup(unsigned long addr, unsigned long *size, + unsigned long *off, char *sym) +{ + unsigned long symbol_start, symbol_end; + struct bpf_prog *prog; + char *ret = NULL; + + rcu_read_lock(); + prog = bpf_prog_kallsyms_find(addr); + if (prog) { + bpf_get_prog_addr_region(prog, &symbol_start, &symbol_end); + bpf_get_prog_name(prog, sym); + + ret = sym; + if (size) + *size = symbol_end - symbol_start; + if (off) + *off = addr - symbol_start; + } + rcu_read_unlock(); + + return ret; +} + +bool is_bpf_text_address(unsigned long addr) +{ + bool ret; + + rcu_read_lock(); + ret = bpf_prog_kallsyms_find(addr) != NULL; + rcu_read_unlock(); + + return ret; +} + +int bpf_get_kallsym(unsigned int symnum, unsigned long *value, char *type, + char *sym) +{ + unsigned long symbol_start, symbol_end; + struct bpf_prog_aux *aux; + unsigned int it = 0; + int ret = -ERANGE; + + if (!bpf_jit_kallsyms_enabled()) + return ret; + + rcu_read_lock(); + list_for_each_entry_rcu(aux, &bpf_kallsyms, ksym_lnode) { + if (it++ != symnum) + continue; + + bpf_get_prog_addr_region(aux->prog, &symbol_start, &symbol_end); + bpf_get_prog_name(aux->prog, sym); + + *value = symbol_start; + *type = BPF_SYM_ELF_TYPE; + + ret = 0; + break; + } + rcu_read_unlock(); + + return ret; +} + struct bpf_binary_header * bpf_jit_binary_alloc(unsigned int proglen, u8 **image_ptr, unsigned int alignment, @@ -326,6 +525,24 @@ void bpf_jit_binary_free(struct bpf_binary_header *hdr) module_memfree(hdr); } +/* This symbol is only overridden by archs that have different + * requirements than the usual eBPF JITs, f.e. when they only + * implement cBPF JIT, do not set images read-only, etc. + */ +void __weak bpf_jit_free(struct bpf_prog *fp) +{ + if (fp->jited) { + struct bpf_binary_header *hdr = bpf_jit_binary_hdr(fp); + + bpf_jit_binary_unlock_ro(hdr); + bpf_jit_binary_free(hdr); + + WARN_ON_ONCE(!bpf_prog_kallsyms_verify_off(fp)); + } + + bpf_prog_unlock_free(fp); +} + int bpf_jit_harden __read_mostly; static int bpf_jit_blind_insn(const struct bpf_insn *from, @@ -436,8 +653,7 @@ out: static struct bpf_prog *bpf_prog_clone_create(struct bpf_prog *fp_other, gfp_t gfp_extra_flags) { - gfp_t gfp_flags = GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO | - gfp_extra_flags; + gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO | gfp_extra_flags; struct bpf_prog *fp; fp = __vmalloc(fp_other->pages * PAGE_SIZE, gfp_flags, PAGE_KERNEL); @@ -547,10 +763,10 @@ EXPORT_SYMBOL_GPL(__bpf_call_base); * * Decode and execute eBPF instructions. */ -static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn) +static unsigned int ___bpf_prog_run(u64 *regs, const struct bpf_insn *insn, + u64 *stack) { - u64 stack[MAX_BPF_STACK / sizeof(u64)]; - u64 regs[MAX_BPF_REG], tmp; + u64 tmp; static const void *jumptable[256] = { [0 ... 255] = &&default_label, /* Now overwrite non-defaults ... */ @@ -608,7 +824,7 @@ static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn) [BPF_ALU64 | BPF_NEG] = &&ALU64_NEG, /* Call instruction */ [BPF_JMP | BPF_CALL] = &&JMP_CALL, - [BPF_JMP | BPF_CALL | BPF_X] = &&JMP_TAIL_CALL, + [BPF_JMP | BPF_TAIL_CALL] = &&JMP_TAIL_CALL, /* Jumps */ [BPF_JMP | BPF_JA] = &&JMP_JA, [BPF_JMP | BPF_JEQ | BPF_X] = &&JMP_JEQ_X, @@ -658,9 +874,6 @@ static unsigned int __bpf_prog_run(void *ctx, const struct bpf_insn *insn) #define CONT ({ insn++; goto select_insn; }) #define CONT_JMP ({ insn++; goto select_insn; }) - FP = (u64) (unsigned long) &stack[ARRAY_SIZE(stack)]; - ARG1 = (u64) (unsigned long) ctx; - select_insn: goto *jumptable[insn->code]; @@ -939,12 +1152,12 @@ out: LD_ABS_W: /* BPF_R0 = ntohl(*(u32 *) (skb->data + imm32)) */ off = IMM; load_word: - /* BPF_LD + BPD_ABS and BPF_LD + BPF_IND insns are - * only appearing in the programs where ctx == - * skb. All programs keep 'ctx' in regs[BPF_REG_CTX] - * == BPF_R6, bpf_convert_filter() saves it in BPF_R6, - * internal BPF verifier will check that BPF_R6 == - * ctx. + /* BPF_LD + BPD_ABS and BPF_LD + BPF_IND insns are only + * appearing in the programs where ctx == skb + * (see may_access_skb() in the verifier). All programs + * keep 'ctx' in regs[BPF_REG_CTX] == BPF_R6, + * bpf_convert_filter() saves it in BPF_R6, internal BPF + * verifier will check that BPF_R6 == ctx. * * BPF_ABS and BPF_IND are wrappers of function calls, * so they scratch BPF_R1-BPF_R5 registers, preserve @@ -1003,7 +1216,39 @@ load_byte: WARN_RATELIMIT(1, "unknown opcode %02x\n", insn->code); return 0; } -STACK_FRAME_NON_STANDARD(__bpf_prog_run); /* jump table */ +STACK_FRAME_NON_STANDARD(___bpf_prog_run); /* jump table */ + +#define PROG_NAME(stack_size) __bpf_prog_run##stack_size +#define DEFINE_BPF_PROG_RUN(stack_size) \ +static unsigned int PROG_NAME(stack_size)(const void *ctx, const struct bpf_insn *insn) \ +{ \ + u64 stack[stack_size / sizeof(u64)]; \ + u64 regs[MAX_BPF_REG]; \ +\ + FP = (u64) (unsigned long) &stack[ARRAY_SIZE(stack)]; \ + ARG1 = (u64) (unsigned long) ctx; \ + return ___bpf_prog_run(regs, insn, stack); \ +} + +#define EVAL1(FN, X) FN(X) +#define EVAL2(FN, X, Y...) FN(X) EVAL1(FN, Y) +#define EVAL3(FN, X, Y...) FN(X) EVAL2(FN, Y) +#define EVAL4(FN, X, Y...) FN(X) EVAL3(FN, Y) +#define EVAL5(FN, X, Y...) FN(X) EVAL4(FN, Y) +#define EVAL6(FN, X, Y...) FN(X) EVAL5(FN, Y) + +EVAL6(DEFINE_BPF_PROG_RUN, 32, 64, 96, 128, 160, 192); +EVAL6(DEFINE_BPF_PROG_RUN, 224, 256, 288, 320, 352, 384); +EVAL4(DEFINE_BPF_PROG_RUN, 416, 448, 480, 512); + +#define PROG_NAME_LIST(stack_size) PROG_NAME(stack_size), + +static unsigned int (*interpreters[])(const void *ctx, + const struct bpf_insn *insn) = { +EVAL6(PROG_NAME_LIST, 32, 64, 96, 128, 160, 192) +EVAL6(PROG_NAME_LIST, 224, 256, 288, 320, 352, 384) +EVAL4(PROG_NAME_LIST, 416, 448, 480, 512) +}; bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp) @@ -1052,7 +1297,9 @@ static int bpf_check_tail_call(const struct bpf_prog *fp) */ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err) { - fp->bpf_func = (void *) __bpf_prog_run; + u32 stack_depth = max_t(u32, fp->aux->stack_depth, 1); + + fp->bpf_func = interpreters[(round_up(stack_depth, 32) / 32) - 1]; /* eBPF JITs can rewrite the program in case constant * blinding is active. However, in case of error during @@ -1154,12 +1401,22 @@ const struct bpf_func_proto bpf_tail_call_proto = { .arg3_type = ARG_ANYTHING, }; -/* For classic BPF JITs that don't implement bpf_int_jit_compile(). */ +/* Stub for JITs that only support cBPF. eBPF programs are interpreted. + * It is encouraged to implement bpf_int_jit_compile() instead, so that + * eBPF and implicitly also cBPF can get JITed! + */ struct bpf_prog * __weak bpf_int_jit_compile(struct bpf_prog *prog) { return prog; } +/* Stub for JITs that support eBPF. All cBPF code gets transformed into + * eBPF by the kernel and is later compiled by bpf_int_jit_compile(). + */ +void __weak bpf_jit_compile(struct bpf_prog *prog) +{ +} + bool __weak bpf_helper_changes_pkt_data(void *func) { return false; @@ -1173,3 +1430,12 @@ int __weak skb_copy_bits(const struct sk_buff *skb, int offset, void *to, { return -EFAULT; } + +/* All definitions of tracepoints related to BPF. */ +#define CREATE_TRACE_POINTS +#include <linux/bpf_trace.h> + +EXPORT_TRACEPOINT_SYMBOL_GPL(xdp_exception); + +EXPORT_TRACEPOINT_SYMBOL_GPL(bpf_prog_get_type); +EXPORT_TRACEPOINT_SYMBOL_GPL(bpf_prog_put_rcu); diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index a753bbe7df0a..4fb463172aa8 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -13,11 +13,13 @@ #include <linux/bpf.h> #include <linux/jhash.h> #include <linux/filter.h> +#include <linux/rculist_nulls.h> #include "percpu_freelist.h" #include "bpf_lru_list.h" +#include "map_in_map.h" struct bucket { - struct hlist_head head; + struct hlist_nulls_head head; raw_spinlock_t lock; }; @@ -29,28 +31,26 @@ struct bpf_htab { struct pcpu_freelist freelist; struct bpf_lru lru; }; - void __percpu *extra_elems; + struct htab_elem *__percpu *extra_elems; atomic_t count; /* number of elements in this hashtable */ u32 n_buckets; /* number of hash buckets */ u32 elem_size; /* size of each element in bytes */ }; -enum extra_elem_state { - HTAB_NOT_AN_EXTRA_ELEM = 0, - HTAB_EXTRA_ELEM_FREE, - HTAB_EXTRA_ELEM_USED -}; - /* each htab element is struct htab_elem + key + value */ struct htab_elem { union { - struct hlist_node hash_node; - struct bpf_htab *htab; - struct pcpu_freelist_node fnode; + struct hlist_nulls_node hash_node; + struct { + void *padding; + union { + struct bpf_htab *htab; + struct pcpu_freelist_node fnode; + }; + }; }; union { struct rcu_head rcu; - enum extra_elem_state state; struct bpf_lru_node lru_node; }; u32 hash; @@ -71,6 +71,11 @@ static bool htab_is_percpu(const struct bpf_htab *htab) htab->map.map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH; } +static bool htab_is_prealloc(const struct bpf_htab *htab) +{ + return !(htab->map.map_flags & BPF_F_NO_PREALLOC); +} + static inline void htab_elem_set_ptr(struct htab_elem *l, u32 key_size, void __percpu *pptr) { @@ -82,6 +87,11 @@ static inline void __percpu *htab_elem_get_ptr(struct htab_elem *l, u32 key_size return *(void __percpu **)(l->key + key_size); } +static void *fd_htab_map_get_ptr(const struct bpf_map *map, struct htab_elem *l) +{ + return *(void **)(l->key + roundup(map->key_size, 8)); +} + static struct htab_elem *get_htab_elem(struct bpf_htab *htab, int i) { return (struct htab_elem *) (htab->elems + i * htab->elem_size); @@ -122,17 +132,20 @@ static struct htab_elem *prealloc_lru_pop(struct bpf_htab *htab, void *key, static int prealloc_init(struct bpf_htab *htab) { + u32 num_entries = htab->map.max_entries; int err = -ENOMEM, i; - htab->elems = bpf_map_area_alloc(htab->elem_size * - htab->map.max_entries); + if (!htab_is_percpu(htab) && !htab_is_lru(htab)) + num_entries += num_possible_cpus(); + + htab->elems = bpf_map_area_alloc(htab->elem_size * num_entries); if (!htab->elems) return -ENOMEM; if (!htab_is_percpu(htab)) goto skip_percpu_elems; - for (i = 0; i < htab->map.max_entries; i++) { + for (i = 0; i < num_entries; i++) { u32 size = round_up(htab->map.value_size, 8); void __percpu *pptr; @@ -160,10 +173,11 @@ skip_percpu_elems: if (htab_is_lru(htab)) bpf_lru_populate(&htab->lru, htab->elems, offsetof(struct htab_elem, lru_node), - htab->elem_size, htab->map.max_entries); + htab->elem_size, num_entries); else - pcpu_freelist_populate(&htab->freelist, htab->elems, - htab->elem_size, htab->map.max_entries); + pcpu_freelist_populate(&htab->freelist, + htab->elems + offsetof(struct htab_elem, fnode), + htab->elem_size, num_entries); return 0; @@ -184,16 +198,22 @@ static void prealloc_destroy(struct bpf_htab *htab) static int alloc_extra_elems(struct bpf_htab *htab) { - void __percpu *pptr; + struct htab_elem *__percpu *pptr, *l_new; + struct pcpu_freelist_node *l; int cpu; - pptr = __alloc_percpu_gfp(htab->elem_size, 8, GFP_USER | __GFP_NOWARN); + pptr = __alloc_percpu_gfp(sizeof(struct htab_elem *), 8, + GFP_USER | __GFP_NOWARN); if (!pptr) return -ENOMEM; for_each_possible_cpu(cpu) { - ((struct htab_elem *)per_cpu_ptr(pptr, cpu))->state = - HTAB_EXTRA_ELEM_FREE; + l = pcpu_freelist_pop(&htab->freelist); + /* pop will succeed, since prealloc_init() + * preallocated extra num_possible_cpus elements + */ + l_new = container_of(l, struct htab_elem, fnode); + *per_cpu_ptr(pptr, cpu) = l_new; } htab->extra_elems = pptr; return 0; @@ -217,6 +237,11 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) int err, i; u64 cost; + BUILD_BUG_ON(offsetof(struct htab_elem, htab) != + offsetof(struct htab_elem, hash_node.pprev)); + BUILD_BUG_ON(offsetof(struct htab_elem, fnode.next) != + offsetof(struct htab_elem, hash_node.pprev)); + if (lru && !capable(CAP_SYS_ADMIN)) /* LRU implementation is much complicated than other * maps. Hence, limit to CAP_SYS_ADMIN for now. @@ -326,29 +351,29 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) goto free_htab; for (i = 0; i < htab->n_buckets; i++) { - INIT_HLIST_HEAD(&htab->buckets[i].head); + INIT_HLIST_NULLS_HEAD(&htab->buckets[i].head, i); raw_spin_lock_init(&htab->buckets[i].lock); } - if (!percpu && !lru) { - /* lru itself can remove the least used element, so - * there is no need for an extra elem during map_update. - */ - err = alloc_extra_elems(htab); - if (err) - goto free_buckets; - } - if (prealloc) { err = prealloc_init(htab); if (err) - goto free_extra_elems; + goto free_buckets; + + if (!percpu && !lru) { + /* lru itself can remove the least used element, so + * there is no need for an extra elem during map_update. + */ + err = alloc_extra_elems(htab); + if (err) + goto free_prealloc; + } } return &htab->map; -free_extra_elems: - free_percpu(htab->extra_elems); +free_prealloc: + prealloc_destroy(htab); free_buckets: bpf_map_area_free(htab->buckets); free_htab: @@ -366,28 +391,56 @@ static inline struct bucket *__select_bucket(struct bpf_htab *htab, u32 hash) return &htab->buckets[hash & (htab->n_buckets - 1)]; } -static inline struct hlist_head *select_bucket(struct bpf_htab *htab, u32 hash) +static inline struct hlist_nulls_head *select_bucket(struct bpf_htab *htab, u32 hash) { return &__select_bucket(htab, hash)->head; } -static struct htab_elem *lookup_elem_raw(struct hlist_head *head, u32 hash, +/* this lookup function can only be called with bucket lock taken */ +static struct htab_elem *lookup_elem_raw(struct hlist_nulls_head *head, u32 hash, void *key, u32 key_size) { + struct hlist_nulls_node *n; struct htab_elem *l; - hlist_for_each_entry_rcu(l, head, hash_node) + hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) if (l->hash == hash && !memcmp(&l->key, key, key_size)) return l; return NULL; } -/* Called from syscall or from eBPF program */ +/* can be called without bucket lock. it will repeat the loop in + * the unlikely event when elements moved from one bucket into another + * while link list is being walked + */ +static struct htab_elem *lookup_nulls_elem_raw(struct hlist_nulls_head *head, + u32 hash, void *key, + u32 key_size, u32 n_buckets) +{ + struct hlist_nulls_node *n; + struct htab_elem *l; + +again: + hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) + if (l->hash == hash && !memcmp(&l->key, key, key_size)) + return l; + + if (unlikely(get_nulls_value(n) != (hash & (n_buckets - 1)))) + goto again; + + return NULL; +} + +/* Called from syscall or from eBPF program directly, so + * arguments have to match bpf_map_lookup_elem() exactly. + * The return value is adjusted by BPF instructions + * in htab_map_gen_lookup(). + */ static void *__htab_map_lookup_elem(struct bpf_map *map, void *key) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - struct hlist_head *head; + struct hlist_nulls_head *head; struct htab_elem *l; u32 hash, key_size; @@ -400,7 +453,7 @@ static void *__htab_map_lookup_elem(struct bpf_map *map, void *key) head = select_bucket(htab, hash); - l = lookup_elem_raw(head, hash, key, key_size); + l = lookup_nulls_elem_raw(head, hash, key, key_size, htab->n_buckets); return l; } @@ -415,6 +468,30 @@ static void *htab_map_lookup_elem(struct bpf_map *map, void *key) return NULL; } +/* inline bpf_map_lookup_elem() call. + * Instead of: + * bpf_prog + * bpf_map_lookup_elem + * map->ops->map_lookup_elem + * htab_map_lookup_elem + * __htab_map_lookup_elem + * do: + * bpf_prog + * __htab_map_lookup_elem + */ +static u32 htab_map_gen_lookup(struct bpf_map *map, struct bpf_insn *insn_buf) +{ + struct bpf_insn *insn = insn_buf; + const int ret = BPF_REG_0; + + *insn++ = BPF_EMIT_CALL((u64 (*)(u64, u64, u64, u64, u64))__htab_map_lookup_elem); + *insn++ = BPF_JMP_IMM(BPF_JEQ, ret, 0, 1); + *insn++ = BPF_ALU64_IMM(BPF_ADD, ret, + offsetof(struct htab_elem, key) + + round_up(map->key_size, 8)); + return insn - insn_buf; +} + static void *htab_lru_map_lookup_elem(struct bpf_map *map, void *key) { struct htab_elem *l = __htab_map_lookup_elem(map, key); @@ -433,8 +510,9 @@ static void *htab_lru_map_lookup_elem(struct bpf_map *map, void *key) static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node) { struct bpf_htab *htab = (struct bpf_htab *)arg; - struct htab_elem *l, *tgt_l; - struct hlist_head *head; + struct htab_elem *l = NULL, *tgt_l; + struct hlist_nulls_head *head; + struct hlist_nulls_node *n; unsigned long flags; struct bucket *b; @@ -444,9 +522,9 @@ static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node) raw_spin_lock_irqsave(&b->lock, flags); - hlist_for_each_entry_rcu(l, head, hash_node) + hlist_nulls_for_each_entry_rcu(l, n, head, hash_node) if (l == tgt_l) { - hlist_del_rcu(&l->hash_node); + hlist_nulls_del_rcu(&l->hash_node); break; } @@ -459,29 +537,30 @@ static bool htab_lru_map_delete_node(void *arg, struct bpf_lru_node *node) static int htab_map_get_next_key(struct bpf_map *map, void *key, void *next_key) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - struct hlist_head *head; + struct hlist_nulls_head *head; struct htab_elem *l, *next_l; u32 hash, key_size; - int i; + int i = 0; WARN_ON_ONCE(!rcu_read_lock_held()); key_size = map->key_size; + if (!key) + goto find_first_elem; + hash = htab_map_hash(key, key_size); head = select_bucket(htab, hash); /* lookup the key */ - l = lookup_elem_raw(head, hash, key, key_size); + l = lookup_nulls_elem_raw(head, hash, key, key_size, htab->n_buckets); - if (!l) { - i = 0; + if (!l) goto find_first_elem; - } /* key was found, get next key in the same bucket */ - next_l = hlist_entry_safe(rcu_dereference_raw(hlist_next_rcu(&l->hash_node)), + next_l = hlist_nulls_entry_safe(rcu_dereference_raw(hlist_nulls_next_rcu(&l->hash_node)), struct htab_elem, hash_node); if (next_l) { @@ -500,7 +579,7 @@ find_first_elem: head = select_bucket(htab, i); /* pick first element in the bucket */ - next_l = hlist_entry_safe(rcu_dereference_raw(hlist_first_rcu(head)), + next_l = hlist_nulls_entry_safe(rcu_dereference_raw(hlist_nulls_first_rcu(head)), struct htab_elem, hash_node); if (next_l) { /* if it's not empty, just return it */ @@ -538,12 +617,15 @@ static void htab_elem_free_rcu(struct rcu_head *head) static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l) { - if (l->state == HTAB_EXTRA_ELEM_USED) { - l->state = HTAB_EXTRA_ELEM_FREE; - return; + struct bpf_map *map = &htab->map; + + if (map->ops->map_fd_put_ptr) { + void *ptr = fd_htab_map_get_ptr(map, l); + + map->ops->map_fd_put_ptr(ptr); } - if (!(htab->map.map_flags & BPF_F_NO_PREALLOC)) { + if (htab_is_prealloc(htab)) { pcpu_freelist_push(&htab->freelist, &l->fnode); } else { atomic_dec(&htab->count); @@ -573,43 +655,43 @@ static void pcpu_copy_value(struct bpf_htab *htab, void __percpu *pptr, static struct htab_elem *alloc_htab_elem(struct bpf_htab *htab, void *key, void *value, u32 key_size, u32 hash, bool percpu, bool onallcpus, - bool old_elem_exists) + struct htab_elem *old_elem) { u32 size = htab->map.value_size; - bool prealloc = !(htab->map.map_flags & BPF_F_NO_PREALLOC); - struct htab_elem *l_new; + bool prealloc = htab_is_prealloc(htab); + struct htab_elem *l_new, **pl_new; void __percpu *pptr; - int err = 0; if (prealloc) { - l_new = (struct htab_elem *)pcpu_freelist_pop(&htab->freelist); - if (!l_new) - err = -E2BIG; - } else { - if (atomic_inc_return(&htab->count) > htab->map.max_entries) { - atomic_dec(&htab->count); - err = -E2BIG; + if (old_elem) { + /* if we're updating the existing element, + * use per-cpu extra elems to avoid freelist_pop/push + */ + pl_new = this_cpu_ptr(htab->extra_elems); + l_new = *pl_new; + *pl_new = old_elem; } else { - l_new = kmalloc(htab->elem_size, - GFP_ATOMIC | __GFP_NOWARN); - if (!l_new) - return ERR_PTR(-ENOMEM); - } - } + struct pcpu_freelist_node *l; - if (err) { - if (!old_elem_exists) - return ERR_PTR(err); - - /* if we're updating the existing element and the hash table - * is full, use per-cpu extra elems - */ - l_new = this_cpu_ptr(htab->extra_elems); - if (l_new->state != HTAB_EXTRA_ELEM_FREE) - return ERR_PTR(-E2BIG); - l_new->state = HTAB_EXTRA_ELEM_USED; + l = pcpu_freelist_pop(&htab->freelist); + if (!l) + return ERR_PTR(-E2BIG); + l_new = container_of(l, struct htab_elem, fnode); + } } else { - l_new->state = HTAB_NOT_AN_EXTRA_ELEM; + if (atomic_inc_return(&htab->count) > htab->map.max_entries) + if (!old_elem) { + /* when map is full and update() is replacing + * old element, it's ok to allocate, since + * old element will be freed immediately. + * Otherwise return an error + */ + atomic_dec(&htab->count); + return ERR_PTR(-E2BIG); + } + l_new = kmalloc(htab->elem_size, GFP_ATOMIC | __GFP_NOWARN); + if (!l_new) + return ERR_PTR(-ENOMEM); } memcpy(l_new->key, key, key_size); @@ -661,7 +743,7 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new = NULL, *l_old; - struct hlist_head *head; + struct hlist_nulls_head *head; unsigned long flags; struct bucket *b; u32 key_size, hash; @@ -690,7 +772,7 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, goto err; l_new = alloc_htab_elem(htab, key, value, key_size, hash, false, false, - !!l_old); + l_old); if (IS_ERR(l_new)) { /* all pre-allocated elements are in use or memory exhausted */ ret = PTR_ERR(l_new); @@ -700,10 +782,11 @@ static int htab_map_update_elem(struct bpf_map *map, void *key, void *value, /* add new element to the head of the list, so that * concurrent search will find it before old elem */ - hlist_add_head_rcu(&l_new->hash_node, head); + hlist_nulls_add_head_rcu(&l_new->hash_node, head); if (l_old) { - hlist_del_rcu(&l_old->hash_node); - free_htab_elem(htab, l_old); + hlist_nulls_del_rcu(&l_old->hash_node); + if (!htab_is_prealloc(htab)) + free_htab_elem(htab, l_old); } ret = 0; err: @@ -716,7 +799,7 @@ static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value, { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new, *l_old = NULL; - struct hlist_head *head; + struct hlist_nulls_head *head; unsigned long flags; struct bucket *b; u32 key_size, hash; @@ -757,10 +840,10 @@ static int htab_lru_map_update_elem(struct bpf_map *map, void *key, void *value, /* add new element to the head of the list, so that * concurrent search will find it before old elem */ - hlist_add_head_rcu(&l_new->hash_node, head); + hlist_nulls_add_head_rcu(&l_new->hash_node, head); if (l_old) { bpf_lru_node_set_ref(&l_new->lru_node); - hlist_del_rcu(&l_old->hash_node); + hlist_nulls_del_rcu(&l_old->hash_node); } ret = 0; @@ -781,7 +864,7 @@ static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key, { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new = NULL, *l_old; - struct hlist_head *head; + struct hlist_nulls_head *head; unsigned long flags; struct bucket *b; u32 key_size, hash; @@ -815,12 +898,12 @@ static int __htab_percpu_map_update_elem(struct bpf_map *map, void *key, value, onallcpus); } else { l_new = alloc_htab_elem(htab, key, value, key_size, - hash, true, onallcpus, false); + hash, true, onallcpus, NULL); if (IS_ERR(l_new)) { ret = PTR_ERR(l_new); goto err; } - hlist_add_head_rcu(&l_new->hash_node, head); + hlist_nulls_add_head_rcu(&l_new->hash_node, head); } ret = 0; err: @@ -834,7 +917,7 @@ static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); struct htab_elem *l_new = NULL, *l_old; - struct hlist_head *head; + struct hlist_nulls_head *head; unsigned long flags; struct bucket *b; u32 key_size, hash; @@ -882,7 +965,7 @@ static int __htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, } else { pcpu_copy_value(htab, htab_elem_get_ptr(l_new, key_size), value, onallcpus); - hlist_add_head_rcu(&l_new->hash_node, head); + hlist_nulls_add_head_rcu(&l_new->hash_node, head); l_new = NULL; } ret = 0; @@ -910,7 +993,7 @@ static int htab_lru_percpu_map_update_elem(struct bpf_map *map, void *key, static int htab_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - struct hlist_head *head; + struct hlist_nulls_head *head; struct bucket *b; struct htab_elem *l; unsigned long flags; @@ -930,7 +1013,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key) l = lookup_elem_raw(head, hash, key, key_size); if (l) { - hlist_del_rcu(&l->hash_node); + hlist_nulls_del_rcu(&l->hash_node); free_htab_elem(htab, l); ret = 0; } @@ -942,7 +1025,7 @@ static int htab_map_delete_elem(struct bpf_map *map, void *key) static int htab_lru_map_delete_elem(struct bpf_map *map, void *key) { struct bpf_htab *htab = container_of(map, struct bpf_htab, map); - struct hlist_head *head; + struct hlist_nulls_head *head; struct bucket *b; struct htab_elem *l; unsigned long flags; @@ -962,7 +1045,7 @@ static int htab_lru_map_delete_elem(struct bpf_map *map, void *key) l = lookup_elem_raw(head, hash, key, key_size); if (l) { - hlist_del_rcu(&l->hash_node); + hlist_nulls_del_rcu(&l->hash_node); ret = 0; } @@ -977,17 +1060,17 @@ static void delete_all_elements(struct bpf_htab *htab) int i; for (i = 0; i < htab->n_buckets; i++) { - struct hlist_head *head = select_bucket(htab, i); - struct hlist_node *n; + struct hlist_nulls_head *head = select_bucket(htab, i); + struct hlist_nulls_node *n; struct htab_elem *l; - hlist_for_each_entry_safe(l, n, head, hash_node) { - hlist_del_rcu(&l->hash_node); - if (l->state != HTAB_EXTRA_ELEM_USED) - htab_elem_free(htab, l); + hlist_nulls_for_each_entry_safe(l, n, head, hash_node) { + hlist_nulls_del_rcu(&l->hash_node); + htab_elem_free(htab, l); } } } + /* Called when map->refcnt goes to zero, either from workqueue or from syscall */ static void htab_map_free(struct bpf_map *map) { @@ -1004,7 +1087,7 @@ static void htab_map_free(struct bpf_map *map) * not have executed. Wait for them. */ rcu_barrier(); - if (htab->map.map_flags & BPF_F_NO_PREALLOC) + if (!htab_is_prealloc(htab)) delete_all_elements(htab); else prealloc_destroy(htab); @@ -1014,21 +1097,17 @@ static void htab_map_free(struct bpf_map *map) kfree(htab); } -static const struct bpf_map_ops htab_ops = { +const struct bpf_map_ops htab_map_ops = { .map_alloc = htab_map_alloc, .map_free = htab_map_free, .map_get_next_key = htab_map_get_next_key, .map_lookup_elem = htab_map_lookup_elem, .map_update_elem = htab_map_update_elem, .map_delete_elem = htab_map_delete_elem, + .map_gen_lookup = htab_map_gen_lookup, }; -static struct bpf_map_type_list htab_type __read_mostly = { - .ops = &htab_ops, - .type = BPF_MAP_TYPE_HASH, -}; - -static const struct bpf_map_ops htab_lru_ops = { +const struct bpf_map_ops htab_lru_map_ops = { .map_alloc = htab_map_alloc, .map_free = htab_map_free, .map_get_next_key = htab_map_get_next_key, @@ -1037,11 +1116,6 @@ static const struct bpf_map_ops htab_lru_ops = { .map_delete_elem = htab_lru_map_delete_elem, }; -static struct bpf_map_type_list htab_lru_type __read_mostly = { - .ops = &htab_lru_ops, - .type = BPF_MAP_TYPE_LRU_HASH, -}; - /* Called from eBPF program */ static void *htab_percpu_map_lookup_elem(struct bpf_map *map, void *key) { @@ -1115,7 +1189,7 @@ int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value, return ret; } -static const struct bpf_map_ops htab_percpu_ops = { +const struct bpf_map_ops htab_percpu_map_ops = { .map_alloc = htab_map_alloc, .map_free = htab_map_free, .map_get_next_key = htab_map_get_next_key, @@ -1124,12 +1198,7 @@ static const struct bpf_map_ops htab_percpu_ops = { .map_delete_elem = htab_map_delete_elem, }; -static struct bpf_map_type_list htab_percpu_type __read_mostly = { - .ops = &htab_percpu_ops, - .type = BPF_MAP_TYPE_PERCPU_HASH, -}; - -static const struct bpf_map_ops htab_lru_percpu_ops = { +const struct bpf_map_ops htab_lru_percpu_map_ops = { .map_alloc = htab_map_alloc, .map_free = htab_map_free, .map_get_next_key = htab_map_get_next_key, @@ -1138,17 +1207,123 @@ static const struct bpf_map_ops htab_lru_percpu_ops = { .map_delete_elem = htab_lru_map_delete_elem, }; -static struct bpf_map_type_list htab_lru_percpu_type __read_mostly = { - .ops = &htab_lru_percpu_ops, - .type = BPF_MAP_TYPE_LRU_PERCPU_HASH, -}; +static struct bpf_map *fd_htab_map_alloc(union bpf_attr *attr) +{ + struct bpf_map *map; + + if (attr->value_size != sizeof(u32)) + return ERR_PTR(-EINVAL); + + /* pointer is stored internally */ + attr->value_size = sizeof(void *); + map = htab_map_alloc(attr); + attr->value_size = sizeof(u32); -static int __init register_htab_map(void) + return map; +} + +static void fd_htab_map_free(struct bpf_map *map) { - bpf_register_map_type(&htab_type); - bpf_register_map_type(&htab_percpu_type); - bpf_register_map_type(&htab_lru_type); - bpf_register_map_type(&htab_lru_percpu_type); - return 0; + struct bpf_htab *htab = container_of(map, struct bpf_htab, map); + struct hlist_nulls_node *n; + struct hlist_nulls_head *head; + struct htab_elem *l; + int i; + + for (i = 0; i < htab->n_buckets; i++) { + head = select_bucket(htab, i); + + hlist_nulls_for_each_entry_safe(l, n, head, hash_node) { + void *ptr = fd_htab_map_get_ptr(map, l); + + map->ops->map_fd_put_ptr(ptr); + } + } + + htab_map_free(map); +} + +/* only called from syscall */ +int bpf_fd_htab_map_lookup_elem(struct bpf_map *map, void *key, u32 *value) +{ + void **ptr; + int ret = 0; + + if (!map->ops->map_fd_sys_lookup_elem) + return -ENOTSUPP; + + rcu_read_lock(); + ptr = htab_map_lookup_elem(map, key); + if (ptr) + *value = map->ops->map_fd_sys_lookup_elem(READ_ONCE(*ptr)); + else + ret = -ENOENT; + rcu_read_unlock(); + + return ret; +} + +/* only called from syscall */ +int bpf_fd_htab_map_update_elem(struct bpf_map *map, struct file *map_file, + void *key, void *value, u64 map_flags) +{ + void *ptr; + int ret; + u32 ufd = *(u32 *)value; + + ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); + if (IS_ERR(ptr)) + return PTR_ERR(ptr); + + ret = htab_map_update_elem(map, key, &ptr, map_flags); + if (ret) + map->ops->map_fd_put_ptr(ptr); + + return ret; } -late_initcall(register_htab_map); + +static struct bpf_map *htab_of_map_alloc(union bpf_attr *attr) +{ + struct bpf_map *map, *inner_map_meta; + + inner_map_meta = bpf_map_meta_alloc(attr->inner_map_fd); + if (IS_ERR(inner_map_meta)) + return inner_map_meta; + + map = fd_htab_map_alloc(attr); + if (IS_ERR(map)) { + bpf_map_meta_free(inner_map_meta); + return map; + } + + map->inner_map_meta = inner_map_meta; + + return map; +} + +static void *htab_of_map_lookup_elem(struct bpf_map *map, void *key) +{ + struct bpf_map **inner_map = htab_map_lookup_elem(map, key); + + if (!inner_map) + return NULL; + + return READ_ONCE(*inner_map); +} + +static void htab_of_map_free(struct bpf_map *map) +{ + bpf_map_meta_free(map->inner_map_meta); + fd_htab_map_free(map); +} + +const struct bpf_map_ops htab_of_maps_map_ops = { + .map_alloc = htab_of_map_alloc, + .map_free = htab_of_map_free, + .map_get_next_key = htab_map_get_next_key, + .map_lookup_elem = htab_of_map_lookup_elem, + .map_delete_elem = htab_map_delete_elem, + .map_fd_get_ptr = bpf_map_fd_get_ptr, + .map_fd_put_ptr = bpf_map_fd_put_ptr, + .map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem, +}; diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index 045cbe673356..3d24e238221e 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -176,6 +176,6 @@ const struct bpf_func_proto bpf_get_current_comm_proto = { .func = bpf_get_current_comm, .gpl_only = false, .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_RAW_STACK, - .arg2_type = ARG_CONST_STACK_SIZE, + .arg1_type = ARG_PTR_TO_UNINIT_MEM, + .arg2_type = ARG_CONST_SIZE, }; diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 0b030c9126d3..e833ed914358 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -21,6 +21,7 @@ #include <linux/parser.h> #include <linux/filter.h> #include <linux/bpf.h> +#include <linux/bpf_trace.h> enum bpf_type { BPF_TYPE_UNSPEC = 0, @@ -281,6 +282,13 @@ int bpf_obj_pin_user(u32 ufd, const char __user *pathname) ret = bpf_obj_do_pin(pname, raw, type); if (ret != 0) bpf_any_put(raw, type); + if ((trace_bpf_obj_pin_prog_enabled() || + trace_bpf_obj_pin_map_enabled()) && !ret) { + if (type == BPF_TYPE_PROG) + trace_bpf_obj_pin_prog(raw, ufd, pname); + if (type == BPF_TYPE_MAP) + trace_bpf_obj_pin_map(raw, ufd, pname); + } out: putname(pname); return ret; @@ -342,8 +350,15 @@ int bpf_obj_get_user(const char __user *pathname) else goto out; - if (ret < 0) + if (ret < 0) { bpf_any_put(raw, type); + } else if (trace_bpf_obj_get_prog_enabled() || + trace_bpf_obj_get_map_enabled()) { + if (type == BPF_TYPE_PROG) + trace_bpf_obj_get_prog(raw, ret, pname); + if (type == BPF_TYPE_MAP) + trace_bpf_obj_get_map(raw, ret, pname); + } out: putname(pname); return ret; @@ -362,10 +377,22 @@ static void bpf_evict_inode(struct inode *inode) bpf_any_put(inode->i_private, type); } +/* + * Display the mount options in /proc/mounts. + */ +static int bpf_show_options(struct seq_file *m, struct dentry *root) +{ + umode_t mode = d_inode(root)->i_mode & S_IALLUGO & ~S_ISVTX; + + if (mode != S_IRWXUGO) + seq_printf(m, ",mode=%o", mode); + return 0; +} + static const struct super_operations bpf_super_ops = { .statfs = simple_statfs, .drop_inode = generic_delete_inode, - .show_options = generic_show_options, + .show_options = bpf_show_options, .evict_inode = bpf_evict_inode, }; @@ -414,13 +441,11 @@ static int bpf_parse_options(char *data, struct bpf_mount_opts *opts) static int bpf_fill_super(struct super_block *sb, void *data, int silent) { - static struct tree_descr bpf_rfiles[] = { { "" } }; + static const struct tree_descr bpf_rfiles[] = { { "" } }; struct bpf_mount_opts opts; struct inode *inode; int ret; - save_mount_options(sb, data); - ret = bpf_parse_options(data, &opts); if (ret) return ret; diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c new file mode 100644 index 000000000000..b09185f0f17d --- /dev/null +++ b/kernel/bpf/lpm_trie.c @@ -0,0 +1,516 @@ +/* + * Longest prefix match list implementation + * + * Copyright (c) 2016,2017 Daniel Mack + * Copyright (c) 2016 David Herrmann + * + * This file is subject to the terms and conditions of version 2 of the GNU + * General Public License. See the file COPYING in the main directory of the + * Linux distribution for more details. + */ + +#include <linux/bpf.h> +#include <linux/err.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/vmalloc.h> +#include <net/ipv6.h> + +/* Intermediate node */ +#define LPM_TREE_NODE_FLAG_IM BIT(0) + +struct lpm_trie_node; + +struct lpm_trie_node { + struct rcu_head rcu; + struct lpm_trie_node __rcu *child[2]; + u32 prefixlen; + u32 flags; + u8 data[0]; +}; + +struct lpm_trie { + struct bpf_map map; + struct lpm_trie_node __rcu *root; + size_t n_entries; + size_t max_prefixlen; + size_t data_size; + raw_spinlock_t lock; +}; + +/* This trie implements a longest prefix match algorithm that can be used to + * match IP addresses to a stored set of ranges. + * + * Data stored in @data of struct bpf_lpm_key and struct lpm_trie_node is + * interpreted as big endian, so data[0] stores the most significant byte. + * + * Match ranges are internally stored in instances of struct lpm_trie_node + * which each contain their prefix length as well as two pointers that may + * lead to more nodes containing more specific matches. Each node also stores + * a value that is defined by and returned to userspace via the update_elem + * and lookup functions. + * + * For instance, let's start with a trie that was created with a prefix length + * of 32, so it can be used for IPv4 addresses, and one single element that + * matches 192.168.0.0/16. The data array would hence contain + * [0xc0, 0xa8, 0x00, 0x00] in big-endian notation. This documentation will + * stick to IP-address notation for readability though. + * + * As the trie is empty initially, the new node (1) will be places as root + * node, denoted as (R) in the example below. As there are no other node, both + * child pointers are %NULL. + * + * +----------------+ + * | (1) (R) | + * | 192.168.0.0/16 | + * | value: 1 | + * | [0] [1] | + * +----------------+ + * + * Next, let's add a new node (2) matching 192.168.0.0/24. As there is already + * a node with the same data and a smaller prefix (ie, a less specific one), + * node (2) will become a child of (1). In child index depends on the next bit + * that is outside of what (1) matches, and that bit is 0, so (2) will be + * child[0] of (1): + * + * +----------------+ + * | (1) (R) | + * | 192.168.0.0/16 | + * | value: 1 | + * | [0] [1] | + * +----------------+ + * | + * +----------------+ + * | (2) | + * | 192.168.0.0/24 | + * | value: 2 | + * | [0] [1] | + * +----------------+ + * + * The child[1] slot of (1) could be filled with another node which has bit #17 + * (the next bit after the ones that (1) matches on) set to 1. For instance, + * 192.168.128.0/24: + * + * +----------------+ + * | (1) (R) | + * | 192.168.0.0/16 | + * | value: 1 | + * | [0] [1] | + * +----------------+ + * | | + * +----------------+ +------------------+ + * | (2) | | (3) | + * | 192.168.0.0/24 | | 192.168.128.0/24 | + * | value: 2 | | value: 3 | + * | [0] [1] | | [0] [1] | + * +----------------+ +------------------+ + * + * Let's add another node (4) to the game for 192.168.1.0/24. In order to place + * it, node (1) is looked at first, and because (4) of the semantics laid out + * above (bit #17 is 0), it would normally be attached to (1) as child[0]. + * However, that slot is already allocated, so a new node is needed in between. + * That node does not have a value attached to it and it will never be + * returned to users as result of a lookup. It is only there to differentiate + * the traversal further. It will get a prefix as wide as necessary to + * distinguish its two children: + * + * +----------------+ + * | (1) (R) | + * | 192.168.0.0/16 | + * | value: 1 | + * | [0] [1] | + * +----------------+ + * | | + * +----------------+ +------------------+ + * | (4) (I) | | (3) | + * | 192.168.0.0/23 | | 192.168.128.0/24 | + * | value: --- | | value: 3 | + * | [0] [1] | | [0] [1] | + * +----------------+ +------------------+ + * | | + * +----------------+ +----------------+ + * | (2) | | (5) | + * | 192.168.0.0/24 | | 192.168.1.0/24 | + * | value: 2 | | value: 5 | + * | [0] [1] | | [0] [1] | + * +----------------+ +----------------+ + * + * 192.168.1.1/32 would be a child of (5) etc. + * + * An intermediate node will be turned into a 'real' node on demand. In the + * example above, (4) would be re-used if 192.168.0.0/23 is added to the trie. + * + * A fully populated trie would have a height of 32 nodes, as the trie was + * created with a prefix length of 32. + * + * The lookup starts at the root node. If the current node matches and if there + * is a child that can be used to become more specific, the trie is traversed + * downwards. The last node in the traversal that is a non-intermediate one is + * returned. + */ + +static inline int extract_bit(const u8 *data, size_t index) +{ + return !!(data[index / 8] & (1 << (7 - (index % 8)))); +} + +/** + * longest_prefix_match() - determine the longest prefix + * @trie: The trie to get internal sizes from + * @node: The node to operate on + * @key: The key to compare to @node + * + * Determine the longest prefix of @node that matches the bits in @key. + */ +static size_t longest_prefix_match(const struct lpm_trie *trie, + const struct lpm_trie_node *node, + const struct bpf_lpm_trie_key *key) +{ + size_t prefixlen = 0; + size_t i; + + for (i = 0; i < trie->data_size; i++) { + size_t b; + + b = 8 - fls(node->data[i] ^ key->data[i]); + prefixlen += b; + + if (prefixlen >= node->prefixlen || prefixlen >= key->prefixlen) + return min(node->prefixlen, key->prefixlen); + + if (b < 8) + break; + } + + return prefixlen; +} + +/* Called from syscall or from eBPF program */ +static void *trie_lookup_elem(struct bpf_map *map, void *_key) +{ + struct lpm_trie *trie = container_of(map, struct lpm_trie, map); + struct lpm_trie_node *node, *found = NULL; + struct bpf_lpm_trie_key *key = _key; + + /* Start walking the trie from the root node ... */ + + for (node = rcu_dereference(trie->root); node;) { + unsigned int next_bit; + size_t matchlen; + + /* Determine the longest prefix of @node that matches @key. + * If it's the maximum possible prefix for this trie, we have + * an exact match and can return it directly. + */ + matchlen = longest_prefix_match(trie, node, key); + if (matchlen == trie->max_prefixlen) { + found = node; + break; + } + + /* If the number of bits that match is smaller than the prefix + * length of @node, bail out and return the node we have seen + * last in the traversal (ie, the parent). + */ + if (matchlen < node->prefixlen) + break; + + /* Consider this node as return candidate unless it is an + * artificially added intermediate one. + */ + if (!(node->flags & LPM_TREE_NODE_FLAG_IM)) + found = node; + + /* If the node match is fully satisfied, let's see if we can + * become more specific. Determine the next bit in the key and + * traverse down. + */ + next_bit = extract_bit(key->data, node->prefixlen); + node = rcu_dereference(node->child[next_bit]); + } + + if (!found) + return NULL; + + return found->data + trie->data_size; +} + +static struct lpm_trie_node *lpm_trie_node_alloc(const struct lpm_trie *trie, + const void *value) +{ + struct lpm_trie_node *node; + size_t size = sizeof(struct lpm_trie_node) + trie->data_size; + + if (value) + size += trie->map.value_size; + + node = kmalloc(size, GFP_ATOMIC | __GFP_NOWARN); + if (!node) + return NULL; + + node->flags = 0; + + if (value) + memcpy(node->data + trie->data_size, value, + trie->map.value_size); + + return node; +} + +/* Called from syscall or from eBPF program */ +static int trie_update_elem(struct bpf_map *map, + void *_key, void *value, u64 flags) +{ + struct lpm_trie *trie = container_of(map, struct lpm_trie, map); + struct lpm_trie_node *node, *im_node = NULL, *new_node = NULL; + struct lpm_trie_node __rcu **slot; + struct bpf_lpm_trie_key *key = _key; + unsigned long irq_flags; + unsigned int next_bit; + size_t matchlen = 0; + int ret = 0; + + if (unlikely(flags > BPF_EXIST)) + return -EINVAL; + + if (key->prefixlen > trie->max_prefixlen) + return -EINVAL; + + raw_spin_lock_irqsave(&trie->lock, irq_flags); + + /* Allocate and fill a new node */ + + if (trie->n_entries == trie->map.max_entries) { + ret = -ENOSPC; + goto out; + } + + new_node = lpm_trie_node_alloc(trie, value); + if (!new_node) { + ret = -ENOMEM; + goto out; + } + + trie->n_entries++; + + new_node->prefixlen = key->prefixlen; + RCU_INIT_POINTER(new_node->child[0], NULL); + RCU_INIT_POINTER(new_node->child[1], NULL); + memcpy(new_node->data, key->data, trie->data_size); + + /* Now find a slot to attach the new node. To do that, walk the tree + * from the root and match as many bits as possible for each node until + * we either find an empty slot or a slot that needs to be replaced by + * an intermediate node. + */ + slot = &trie->root; + + while ((node = rcu_dereference_protected(*slot, + lockdep_is_held(&trie->lock)))) { + matchlen = longest_prefix_match(trie, node, key); + + if (node->prefixlen != matchlen || + node->prefixlen == key->prefixlen || + node->prefixlen == trie->max_prefixlen) + break; + + next_bit = extract_bit(key->data, node->prefixlen); + slot = &node->child[next_bit]; + } + + /* If the slot is empty (a free child pointer or an empty root), + * simply assign the @new_node to that slot and be done. + */ + if (!node) { + rcu_assign_pointer(*slot, new_node); + goto out; + } + + /* If the slot we picked already exists, replace it with @new_node + * which already has the correct data array set. + */ + if (node->prefixlen == matchlen) { + new_node->child[0] = node->child[0]; + new_node->child[1] = node->child[1]; + + if (!(node->flags & LPM_TREE_NODE_FLAG_IM)) + trie->n_entries--; + + rcu_assign_pointer(*slot, new_node); + kfree_rcu(node, rcu); + + goto out; + } + + /* If the new node matches the prefix completely, it must be inserted + * as an ancestor. Simply insert it between @node and *@slot. + */ + if (matchlen == key->prefixlen) { + next_bit = extract_bit(node->data, matchlen); + rcu_assign_pointer(new_node->child[next_bit], node); + rcu_assign_pointer(*slot, new_node); + goto out; + } + + im_node = lpm_trie_node_alloc(trie, NULL); + if (!im_node) { + ret = -ENOMEM; + goto out; + } + + im_node->prefixlen = matchlen; + im_node->flags |= LPM_TREE_NODE_FLAG_IM; + memcpy(im_node->data, node->data, trie->data_size); + + /* Now determine which child to install in which slot */ + if (extract_bit(key->data, matchlen)) { + rcu_assign_pointer(im_node->child[0], node); + rcu_assign_pointer(im_node->child[1], new_node); + } else { + rcu_assign_pointer(im_node->child[0], new_node); + rcu_assign_pointer(im_node->child[1], node); + } + + /* Finally, assign the intermediate node to the determined spot */ + rcu_assign_pointer(*slot, im_node); + +out: + if (ret) { + if (new_node) + trie->n_entries--; + + kfree(new_node); + kfree(im_node); + } + + raw_spin_unlock_irqrestore(&trie->lock, irq_flags); + + return ret; +} + +static int trie_delete_elem(struct bpf_map *map, void *key) +{ + /* TODO */ + return -ENOSYS; +} + +#define LPM_DATA_SIZE_MAX 256 +#define LPM_DATA_SIZE_MIN 1 + +#define LPM_VAL_SIZE_MAX (KMALLOC_MAX_SIZE - LPM_DATA_SIZE_MAX - \ + sizeof(struct lpm_trie_node)) +#define LPM_VAL_SIZE_MIN 1 + +#define LPM_KEY_SIZE(X) (sizeof(struct bpf_lpm_trie_key) + (X)) +#define LPM_KEY_SIZE_MAX LPM_KEY_SIZE(LPM_DATA_SIZE_MAX) +#define LPM_KEY_SIZE_MIN LPM_KEY_SIZE(LPM_DATA_SIZE_MIN) + +static struct bpf_map *trie_alloc(union bpf_attr *attr) +{ + struct lpm_trie *trie; + u64 cost = sizeof(*trie), cost_per_node; + int ret; + + if (!capable(CAP_SYS_ADMIN)) + return ERR_PTR(-EPERM); + + /* check sanity of attributes */ + if (attr->max_entries == 0 || + attr->map_flags != BPF_F_NO_PREALLOC || + attr->key_size < LPM_KEY_SIZE_MIN || + attr->key_size > LPM_KEY_SIZE_MAX || + attr->value_size < LPM_VAL_SIZE_MIN || + attr->value_size > LPM_VAL_SIZE_MAX) + return ERR_PTR(-EINVAL); + + trie = kzalloc(sizeof(*trie), GFP_USER | __GFP_NOWARN); + if (!trie) + return ERR_PTR(-ENOMEM); + + /* copy mandatory map attributes */ + trie->map.map_type = attr->map_type; + trie->map.key_size = attr->key_size; + trie->map.value_size = attr->value_size; + trie->map.max_entries = attr->max_entries; + trie->map.map_flags = attr->map_flags; + trie->data_size = attr->key_size - + offsetof(struct bpf_lpm_trie_key, data); + trie->max_prefixlen = trie->data_size * 8; + + cost_per_node = sizeof(struct lpm_trie_node) + + attr->value_size + trie->data_size; + cost += (u64) attr->max_entries * cost_per_node; + if (cost >= U32_MAX - PAGE_SIZE) { + ret = -E2BIG; + goto out_err; + } + + trie->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT; + + ret = bpf_map_precharge_memlock(trie->map.pages); + if (ret) + goto out_err; + + raw_spin_lock_init(&trie->lock); + + return &trie->map; +out_err: + kfree(trie); + return ERR_PTR(ret); +} + +static void trie_free(struct bpf_map *map) +{ + struct lpm_trie *trie = container_of(map, struct lpm_trie, map); + struct lpm_trie_node __rcu **slot; + struct lpm_trie_node *node; + + raw_spin_lock(&trie->lock); + + /* Always start at the root and walk down to a node that has no + * children. Then free that node, nullify its reference in the parent + * and start over. + */ + + for (;;) { + slot = &trie->root; + + for (;;) { + node = rcu_dereference_protected(*slot, + lockdep_is_held(&trie->lock)); + if (!node) + goto unlock; + + if (rcu_access_pointer(node->child[0])) { + slot = &node->child[0]; + continue; + } + + if (rcu_access_pointer(node->child[1])) { + slot = &node->child[1]; + continue; + } + + kfree(node); + RCU_INIT_POINTER(*slot, NULL); + break; + } + } + +unlock: + raw_spin_unlock(&trie->lock); +} + +static int trie_get_next_key(struct bpf_map *map, void *key, void *next_key) +{ + return -ENOTSUPP; +} + +const struct bpf_map_ops trie_map_ops = { + .map_alloc = trie_alloc, + .map_free = trie_free, + .map_get_next_key = trie_get_next_key, + .map_lookup_elem = trie_lookup_elem, + .map_update_elem = trie_update_elem, + .map_delete_elem = trie_delete_elem, +}; diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c new file mode 100644 index 000000000000..1da574612bea --- /dev/null +++ b/kernel/bpf/map_in_map.c @@ -0,0 +1,102 @@ +/* Copyright (c) 2017 Facebook + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#include <linux/slab.h> +#include <linux/bpf.h> + +#include "map_in_map.h" + +struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd) +{ + struct bpf_map *inner_map, *inner_map_meta; + struct fd f; + + f = fdget(inner_map_ufd); + inner_map = __bpf_map_get(f); + if (IS_ERR(inner_map)) + return inner_map; + + /* prog_array->owner_prog_type and owner_jited + * is a runtime binding. Doing static check alone + * in the verifier is not enough. + */ + if (inner_map->map_type == BPF_MAP_TYPE_PROG_ARRAY) { + fdput(f); + return ERR_PTR(-ENOTSUPP); + } + + /* Does not support >1 level map-in-map */ + if (inner_map->inner_map_meta) { + fdput(f); + return ERR_PTR(-EINVAL); + } + + inner_map_meta = kzalloc(sizeof(*inner_map_meta), GFP_USER); + if (!inner_map_meta) { + fdput(f); + return ERR_PTR(-ENOMEM); + } + + inner_map_meta->map_type = inner_map->map_type; + inner_map_meta->key_size = inner_map->key_size; + inner_map_meta->value_size = inner_map->value_size; + inner_map_meta->map_flags = inner_map->map_flags; + inner_map_meta->ops = inner_map->ops; + inner_map_meta->max_entries = inner_map->max_entries; + + fdput(f); + return inner_map_meta; +} + +void bpf_map_meta_free(struct bpf_map *map_meta) +{ + kfree(map_meta); +} + +bool bpf_map_meta_equal(const struct bpf_map *meta0, + const struct bpf_map *meta1) +{ + /* No need to compare ops because it is covered by map_type */ + return meta0->map_type == meta1->map_type && + meta0->key_size == meta1->key_size && + meta0->value_size == meta1->value_size && + meta0->map_flags == meta1->map_flags && + meta0->max_entries == meta1->max_entries; +} + +void *bpf_map_fd_get_ptr(struct bpf_map *map, + struct file *map_file /* not used */, + int ufd) +{ + struct bpf_map *inner_map; + struct fd f; + + f = fdget(ufd); + inner_map = __bpf_map_get(f); + if (IS_ERR(inner_map)) + return inner_map; + + if (bpf_map_meta_equal(map->inner_map_meta, inner_map)) + inner_map = bpf_map_inc(inner_map, false); + else + inner_map = ERR_PTR(-EINVAL); + + fdput(f); + return inner_map; +} + +void bpf_map_fd_put_ptr(void *ptr) +{ + /* ptr->ops->map_free() has to go through one + * rcu grace period by itself. + */ + bpf_map_put(ptr); +} + +u32 bpf_map_fd_sys_lookup_elem(void *ptr) +{ + return ((struct bpf_map *)ptr)->id; +} diff --git a/kernel/bpf/map_in_map.h b/kernel/bpf/map_in_map.h new file mode 100644 index 000000000000..6183db9ec08c --- /dev/null +++ b/kernel/bpf/map_in_map.h @@ -0,0 +1,24 @@ +/* Copyright (c) 2017 Facebook + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#ifndef __MAP_IN_MAP_H__ +#define __MAP_IN_MAP_H__ + +#include <linux/types.h> + +struct file; +struct bpf_map; + +struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd); +void bpf_map_meta_free(struct bpf_map *map_meta); +bool bpf_map_meta_equal(const struct bpf_map *meta0, + const struct bpf_map *meta1); +void *bpf_map_fd_get_ptr(struct bpf_map *map, struct file *map_file, + int ufd); +void bpf_map_fd_put_ptr(void *ptr); +u32 bpf_map_fd_sys_lookup_elem(void *ptr); + +#endif diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c index be8519148c25..31147d730abf 100644 --- a/kernel/bpf/stackmap.c +++ b/kernel/bpf/stackmap.c @@ -88,6 +88,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr) smap->map.key_size = attr->key_size; smap->map.value_size = value_size; smap->map.max_entries = attr->max_entries; + smap->map.map_flags = attr->map_flags; smap->n_buckets = n_buckets; smap->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT; @@ -264,7 +265,7 @@ static void stack_map_free(struct bpf_map *map) put_callchain_buffers(); } -static const struct bpf_map_ops stack_map_ops = { +const struct bpf_map_ops stack_map_ops = { .map_alloc = stack_map_alloc, .map_free = stack_map_free, .map_get_next_key = stack_map_get_next_key, @@ -272,15 +273,3 @@ static const struct bpf_map_ops stack_map_ops = { .map_update_elem = stack_map_update_elem, .map_delete_elem = stack_map_delete_elem, }; - -static struct bpf_map_type_list stack_map_type __read_mostly = { - .ops = &stack_map_ops, - .type = BPF_MAP_TYPE_STACK_TRACE, -}; - -static int __init register_stack_map(void) -{ - bpf_register_map_type(&stack_map_type); - return 0; -} -late_initcall(register_stack_map); diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 19b6129eab23..045646da97cc 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -10,8 +10,10 @@ * General Public License for more details. */ #include <linux/bpf.h> +#include <linux/bpf_trace.h> #include <linux/syscalls.h> #include <linux/slab.h> +#include <linux/sched/signal.h> #include <linux/vmalloc.h> #include <linux/mmzone.h> #include <linux/anon_inodes.h> @@ -20,35 +22,46 @@ #include <linux/filter.h> #include <linux/version.h> #include <linux/kernel.h> +#include <linux/idr.h> + +#define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PROG_ARRAY || \ + (map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \ + (map)->map_type == BPF_MAP_TYPE_CGROUP_ARRAY || \ + (map)->map_type == BPF_MAP_TYPE_ARRAY_OF_MAPS) +#define IS_FD_HASH(map) ((map)->map_type == BPF_MAP_TYPE_HASH_OF_MAPS) +#define IS_FD_MAP(map) (IS_FD_ARRAY(map) || IS_FD_HASH(map)) DEFINE_PER_CPU(int, bpf_prog_active); +static DEFINE_IDR(prog_idr); +static DEFINE_SPINLOCK(prog_idr_lock); +static DEFINE_IDR(map_idr); +static DEFINE_SPINLOCK(map_idr_lock); int sysctl_unprivileged_bpf_disabled __read_mostly; -static LIST_HEAD(bpf_map_types); +static const struct bpf_map_ops * const bpf_map_types[] = { +#define BPF_PROG_TYPE(_id, _ops) +#define BPF_MAP_TYPE(_id, _ops) \ + [_id] = &_ops, +#include <linux/bpf_types.h> +#undef BPF_PROG_TYPE +#undef BPF_MAP_TYPE +}; static struct bpf_map *find_and_alloc_map(union bpf_attr *attr) { - struct bpf_map_type_list *tl; struct bpf_map *map; - list_for_each_entry(tl, &bpf_map_types, list_node) { - if (tl->type == attr->map_type) { - map = tl->ops->map_alloc(attr); - if (IS_ERR(map)) - return map; - map->ops = tl->ops; - map->map_type = attr->map_type; - return map; - } - } - return ERR_PTR(-EINVAL); -} + if (attr->map_type >= ARRAY_SIZE(bpf_map_types) || + !bpf_map_types[attr->map_type]) + return ERR_PTR(-EINVAL); -/* boot time registration of different map implementations */ -void bpf_register_map_type(struct bpf_map_type_list *tl) -{ - list_add(&tl->list_node, &bpf_map_types); + map = bpf_map_types[attr->map_type]->map_alloc(attr); + if (IS_ERR(map)) + return map; + map->ops = bpf_map_types[attr->map_type]; + map->map_type = attr->map_type; + return map; } void *bpf_map_area_alloc(size_t size) @@ -66,8 +79,7 @@ void *bpf_map_area_alloc(size_t size) return area; } - return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | flags, - PAGE_KERNEL); + return __vmalloc(size, GFP_KERNEL | flags, PAGE_KERNEL); } void bpf_map_area_free(void *area) @@ -114,6 +126,37 @@ static void bpf_map_uncharge_memlock(struct bpf_map *map) free_uid(user); } +static int bpf_map_alloc_id(struct bpf_map *map) +{ + int id; + + spin_lock_bh(&map_idr_lock); + id = idr_alloc_cyclic(&map_idr, map, 1, INT_MAX, GFP_ATOMIC); + if (id > 0) + map->id = id; + spin_unlock_bh(&map_idr_lock); + + if (WARN_ON_ONCE(!id)) + return -ENOSPC; + + return id > 0 ? 0 : id; +} + +static void bpf_map_free_id(struct bpf_map *map, bool do_idr_lock) +{ + if (do_idr_lock) + spin_lock_bh(&map_idr_lock); + else + __acquire(&map_idr_lock); + + idr_remove(&map_idr, map->id); + + if (do_idr_lock) + spin_unlock_bh(&map_idr_lock); + else + __release(&map_idr_lock); +} + /* called from workqueue */ static void bpf_map_free_deferred(struct work_struct *work) { @@ -135,14 +178,21 @@ static void bpf_map_put_uref(struct bpf_map *map) /* decrement map refcnt and schedule it for freeing via workqueue * (unrelying map implementation ops->map_free() might sleep) */ -void bpf_map_put(struct bpf_map *map) +static void __bpf_map_put(struct bpf_map *map, bool do_idr_lock) { if (atomic_dec_and_test(&map->refcnt)) { + /* bpf_map_free_id() must be called first */ + bpf_map_free_id(map, do_idr_lock); INIT_WORK(&map->work, bpf_map_free_deferred); schedule_work(&map->work); } } +void bpf_map_put(struct bpf_map *map) +{ + __bpf_map_put(map, true); +} + void bpf_map_put_with_uref(struct bpf_map *map) { bpf_map_put_uref(map); @@ -166,10 +216,12 @@ static void bpf_map_show_fdinfo(struct seq_file *m, struct file *filp) const struct bpf_map *map = filp->private_data; const struct bpf_array *array; u32 owner_prog_type = 0; + u32 owner_jited = 0; if (map->map_type == BPF_MAP_TYPE_PROG_ARRAY) { array = container_of(map, struct bpf_array, map); owner_prog_type = array->owner_prog_type; + owner_jited = array->owner_jited; } seq_printf(m, @@ -186,9 +238,12 @@ static void bpf_map_show_fdinfo(struct seq_file *m, struct file *filp) map->map_flags, map->pages * 1ULL << PAGE_SHIFT); - if (owner_prog_type) + if (owner_prog_type) { seq_printf(m, "owner_prog_type:\t%u\n", owner_prog_type); + seq_printf(m, "owner_jited:\t%u\n", + owner_jited); + } } #endif @@ -213,7 +268,7 @@ int bpf_map_new_fd(struct bpf_map *map) offsetof(union bpf_attr, CMD##_LAST_FIELD) - \ sizeof(attr->CMD##_LAST_FIELD)) != NULL -#define BPF_MAP_CREATE_LAST_FIELD map_flags +#define BPF_MAP_CREATE_LAST_FIELD inner_map_fd /* called via syscall */ static int map_create(union bpf_attr *attr) { @@ -236,11 +291,23 @@ static int map_create(union bpf_attr *attr) if (err) goto free_map_nouncharge; - err = bpf_map_new_fd(map); - if (err < 0) - /* failed to allocate fd */ + err = bpf_map_alloc_id(map); + if (err) goto free_map; + err = bpf_map_new_fd(map); + if (err < 0) { + /* failed to allocate fd. + * bpf_map_put() is needed because the above + * bpf_map_alloc_id() has published the map + * to the userspace and the userspace may + * have refcnt-ed it through BPF_MAP_GET_FD_BY_ID. + */ + bpf_map_put(map); + return err; + } + + trace_bpf_map_create(map, err); return err; free_map: @@ -294,6 +361,28 @@ struct bpf_map *bpf_map_get_with_uref(u32 ufd) return map; } +/* map_idr_lock should have been held */ +static struct bpf_map *bpf_map_inc_not_zero(struct bpf_map *map, + bool uref) +{ + int refold; + + refold = __atomic_add_unless(&map->refcnt, 1, 0); + + if (refold >= BPF_MAX_REFCNT) { + __bpf_map_put(map, false); + return ERR_PTR(-EBUSY); + } + + if (!refold) + return ERR_PTR(-ENOENT); + + if (uref) + atomic_inc(&map->usercnt); + + return map; +} + int __weak bpf_stackmap_copy(struct bpf_map *map, void *key, void *value) { return -ENOTSUPP; @@ -321,19 +410,18 @@ static int map_lookup_elem(union bpf_attr *attr) if (IS_ERR(map)) return PTR_ERR(map); - err = -ENOMEM; - key = kmalloc(map->key_size, GFP_USER); - if (!key) + key = memdup_user(ukey, map->key_size); + if (IS_ERR(key)) { + err = PTR_ERR(key); goto err_put; - - err = -EFAULT; - if (copy_from_user(key, ukey, map->key_size) != 0) - goto free_key; + } if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH || map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH || map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) value_size = round_up(map->value_size, 8) * num_possible_cpus(); + else if (IS_FD_MAP(map)) + value_size = sizeof(u32); else value_size = map->value_size; @@ -349,6 +437,10 @@ static int map_lookup_elem(union bpf_attr *attr) err = bpf_percpu_array_copy(map, key, value); } else if (map->map_type == BPF_MAP_TYPE_STACK_TRACE) { err = bpf_stackmap_copy(map, key, value); + } else if (IS_FD_ARRAY(map)) { + err = bpf_fd_array_map_lookup_elem(map, key, value); + } else if (IS_FD_HASH(map)) { + err = bpf_fd_htab_map_lookup_elem(map, key, value); } else { rcu_read_lock(); ptr = map->ops->map_lookup_elem(map, key); @@ -365,6 +457,7 @@ static int map_lookup_elem(union bpf_attr *attr) if (copy_to_user(uvalue, value, value_size) != 0) goto free_value; + trace_bpf_map_lookup_elem(map, ufd, key, value); err = 0; free_value: @@ -397,14 +490,11 @@ static int map_update_elem(union bpf_attr *attr) if (IS_ERR(map)) return PTR_ERR(map); - err = -ENOMEM; - key = kmalloc(map->key_size, GFP_USER); - if (!key) + key = memdup_user(ukey, map->key_size); + if (IS_ERR(key)) { + err = PTR_ERR(key); goto err_put; - - err = -EFAULT; - if (copy_from_user(key, ukey, map->key_size) != 0) - goto free_key; + } if (map->map_type == BPF_MAP_TYPE_PERCPU_HASH || map->map_type == BPF_MAP_TYPE_LRU_PERCPU_HASH || @@ -434,11 +524,17 @@ static int map_update_elem(union bpf_attr *attr) err = bpf_percpu_array_update(map, key, value, attr->flags); } else if (map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || map->map_type == BPF_MAP_TYPE_PROG_ARRAY || - map->map_type == BPF_MAP_TYPE_CGROUP_ARRAY) { + map->map_type == BPF_MAP_TYPE_CGROUP_ARRAY || + map->map_type == BPF_MAP_TYPE_ARRAY_OF_MAPS) { rcu_read_lock(); err = bpf_fd_array_map_update_elem(map, f.file, key, value, attr->flags); rcu_read_unlock(); + } else if (map->map_type == BPF_MAP_TYPE_HASH_OF_MAPS) { + rcu_read_lock(); + err = bpf_fd_htab_map_update_elem(map, f.file, key, value, + attr->flags); + rcu_read_unlock(); } else { rcu_read_lock(); err = map->ops->map_update_elem(map, key, value, attr->flags); @@ -447,6 +543,8 @@ static int map_update_elem(union bpf_attr *attr) __this_cpu_dec(bpf_prog_active); preempt_enable(); + if (!err) + trace_bpf_map_update_elem(map, ufd, key, value); free_value: kfree(value); free_key: @@ -475,14 +573,11 @@ static int map_delete_elem(union bpf_attr *attr) if (IS_ERR(map)) return PTR_ERR(map); - err = -ENOMEM; - key = kmalloc(map->key_size, GFP_USER); - if (!key) + key = memdup_user(ukey, map->key_size); + if (IS_ERR(key)) { + err = PTR_ERR(key); goto err_put; - - err = -EFAULT; - if (copy_from_user(key, ukey, map->key_size) != 0) - goto free_key; + } preempt_disable(); __this_cpu_inc(bpf_prog_active); @@ -492,7 +587,8 @@ static int map_delete_elem(union bpf_attr *attr) __this_cpu_dec(bpf_prog_active); preempt_enable(); -free_key: + if (!err) + trace_bpf_map_delete_elem(map, ufd, key); kfree(key); err_put: fdput(f); @@ -520,14 +616,15 @@ static int map_get_next_key(union bpf_attr *attr) if (IS_ERR(map)) return PTR_ERR(map); - err = -ENOMEM; - key = kmalloc(map->key_size, GFP_USER); - if (!key) - goto err_put; - - err = -EFAULT; - if (copy_from_user(key, ukey, map->key_size) != 0) - goto free_key; + if (ukey) { + key = memdup_user(ukey, map->key_size); + if (IS_ERR(key)) { + err = PTR_ERR(key); + goto err_put; + } + } else { + key = NULL; + } err = -ENOMEM; next_key = kmalloc(map->key_size, GFP_USER); @@ -544,6 +641,7 @@ static int map_get_next_key(union bpf_attr *attr) if (copy_to_user(unext_key, next_key, map->key_size) != 0) goto free_next_key; + trace_bpf_map_next_key(map, ufd, key, next_key); err = 0; free_next_key: @@ -555,79 +653,23 @@ err_put: return err; } -static LIST_HEAD(bpf_prog_types); +static const struct bpf_verifier_ops * const bpf_prog_types[] = { +#define BPF_PROG_TYPE(_id, _ops) \ + [_id] = &_ops, +#define BPF_MAP_TYPE(_id, _ops) +#include <linux/bpf_types.h> +#undef BPF_PROG_TYPE +#undef BPF_MAP_TYPE +}; static int find_prog_type(enum bpf_prog_type type, struct bpf_prog *prog) { - struct bpf_prog_type_list *tl; - - list_for_each_entry(tl, &bpf_prog_types, list_node) { - if (tl->type == type) { - prog->aux->ops = tl->ops; - prog->type = type; - return 0; - } - } - - return -EINVAL; -} - -void bpf_register_prog_type(struct bpf_prog_type_list *tl) -{ - list_add(&tl->list_node, &bpf_prog_types); -} - -/* fixup insn->imm field of bpf_call instructions: - * if (insn->imm == BPF_FUNC_map_lookup_elem) - * insn->imm = bpf_map_lookup_elem - __bpf_call_base; - * else if (insn->imm == BPF_FUNC_map_update_elem) - * insn->imm = bpf_map_update_elem - __bpf_call_base; - * else ... - * - * this function is called after eBPF program passed verification - */ -static void fixup_bpf_calls(struct bpf_prog *prog) -{ - const struct bpf_func_proto *fn; - int i; + if (type >= ARRAY_SIZE(bpf_prog_types) || !bpf_prog_types[type]) + return -EINVAL; - for (i = 0; i < prog->len; i++) { - struct bpf_insn *insn = &prog->insnsi[i]; - - if (insn->code == (BPF_JMP | BPF_CALL)) { - /* we reach here when program has bpf_call instructions - * and it passed bpf_check(), means that - * ops->get_func_proto must have been supplied, check it - */ - BUG_ON(!prog->aux->ops->get_func_proto); - - if (insn->imm == BPF_FUNC_get_route_realm) - prog->dst_needed = 1; - if (insn->imm == BPF_FUNC_get_prandom_u32) - bpf_user_rnd_init_once(); - if (insn->imm == BPF_FUNC_xdp_adjust_head) - prog->xdp_adjust_head = 1; - if (insn->imm == BPF_FUNC_tail_call) { - /* mark bpf_tail_call as different opcode - * to avoid conditional branch in - * interpeter for every normal call - * and to prevent accidental JITing by - * JIT compiler that doesn't support - * bpf_tail_call yet - */ - insn->imm = 0; - insn->code |= BPF_X; - continue; - } - - fn = prog->aux->ops->get_func_proto(insn->imm); - /* all functions that have prototype and verifier allowed - * programs to call them, must be real in-kernel functions - */ - BUG_ON(!fn->func); - insn->imm = fn->func - __bpf_call_base; - } - } + prog->aux->ops = bpf_prog_types[type]; + prog->type = type; + return 0; } /* drop refcnt on maps used by eBPF program and free auxilary data */ @@ -686,6 +728,42 @@ static void bpf_prog_uncharge_memlock(struct bpf_prog *prog) free_uid(user); } +static int bpf_prog_alloc_id(struct bpf_prog *prog) +{ + int id; + + spin_lock_bh(&prog_idr_lock); + id = idr_alloc_cyclic(&prog_idr, prog, 1, INT_MAX, GFP_ATOMIC); + if (id > 0) + prog->aux->id = id; + spin_unlock_bh(&prog_idr_lock); + + /* id is in [1, INT_MAX) */ + if (WARN_ON_ONCE(!id)) + return -ENOSPC; + + return id > 0 ? 0 : id; +} + +static void bpf_prog_free_id(struct bpf_prog *prog, bool do_idr_lock) +{ + /* cBPF to eBPF migrations are currently not in the idr store. */ + if (!prog->aux->id) + return; + + if (do_idr_lock) + spin_lock_bh(&prog_idr_lock); + else + __acquire(&prog_idr_lock); + + idr_remove(&prog_idr, prog->aux->id); + + if (do_idr_lock) + spin_unlock_bh(&prog_idr_lock); + else + __release(&prog_idr_lock); +} + static void __bpf_prog_put_rcu(struct rcu_head *rcu) { struct bpf_prog_aux *aux = container_of(rcu, struct bpf_prog_aux, rcu); @@ -695,10 +773,20 @@ static void __bpf_prog_put_rcu(struct rcu_head *rcu) bpf_prog_free(aux->prog); } -void bpf_prog_put(struct bpf_prog *prog) +static void __bpf_prog_put(struct bpf_prog *prog, bool do_idr_lock) { - if (atomic_dec_and_test(&prog->aux->refcnt)) + if (atomic_dec_and_test(&prog->aux->refcnt)) { + trace_bpf_prog_put_rcu(prog); + /* bpf_prog_free_id() must be called first */ + bpf_prog_free_id(prog, do_idr_lock); + bpf_prog_kallsyms_del(prog); call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu); + } +} + +void bpf_prog_put(struct bpf_prog *prog) +{ + __bpf_prog_put(prog, true); } EXPORT_SYMBOL_GPL(bpf_prog_put); @@ -781,6 +869,24 @@ struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) } EXPORT_SYMBOL_GPL(bpf_prog_inc); +/* prog_idr_lock should have been held */ +static struct bpf_prog *bpf_prog_inc_not_zero(struct bpf_prog *prog) +{ + int refold; + + refold = __atomic_add_unless(&prog->aux->refcnt, 1, 0); + + if (refold >= BPF_MAX_REFCNT) { + __bpf_prog_put(prog, false); + return ERR_PTR(-EBUSY); + } + + if (!refold) + return ERR_PTR(-ENOENT); + + return prog; +} + static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *type) { struct fd f = fdget(ufd); @@ -807,12 +913,16 @@ struct bpf_prog *bpf_prog_get(u32 ufd) struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type) { - return __bpf_prog_get(ufd, &type); + struct bpf_prog *prog = __bpf_prog_get(ufd, &type); + + if (!IS_ERR(prog)) + trace_bpf_prog_get_type(prog); + return prog; } EXPORT_SYMBOL_GPL(bpf_prog_get_type); /* last field in 'union bpf_attr' used by this command */ -#define BPF_PROG_LOAD_LAST_FIELD kern_version +#define BPF_PROG_LOAD_LAST_FIELD prog_flags static int bpf_prog_load(union bpf_attr *attr) { @@ -825,6 +935,9 @@ static int bpf_prog_load(union bpf_attr *attr) if (CHECK_ATTR(BPF_PROG_LOAD)) return -EINVAL; + if (attr->prog_flags & ~BPF_F_STRICT_ALIGNMENT) + return -EINVAL; + /* copy eBPF program license from user space */ if (strncpy_from_user(license, u64_to_user_ptr(attr->license), sizeof(license) - 1) < 0) @@ -841,7 +954,9 @@ static int bpf_prog_load(union bpf_attr *attr) attr->kern_version != LINUX_VERSION_CODE) return -EINVAL; - if (type != BPF_PROG_TYPE_SOCKET_FILTER && !capable(CAP_SYS_ADMIN)) + if (type != BPF_PROG_TYPE_SOCKET_FILTER && + type != BPF_PROG_TYPE_CGROUP_SKB && + !capable(CAP_SYS_ADMIN)) return -EPERM; /* plain bpf_prog allocation */ @@ -876,19 +991,29 @@ static int bpf_prog_load(union bpf_attr *attr) if (err < 0) goto free_used_maps; - /* fixup BPF_CALL->imm field */ - fixup_bpf_calls(prog); - /* eBPF program is ready to be JITed */ prog = bpf_prog_select_runtime(prog, &err); if (err < 0) goto free_used_maps; - err = bpf_prog_new_fd(prog); - if (err < 0) - /* failed to allocate fd */ + err = bpf_prog_alloc_id(prog); + if (err) goto free_used_maps; + err = bpf_prog_new_fd(prog); + if (err < 0) { + /* failed to allocate fd. + * bpf_prog_put() is needed because the above + * bpf_prog_alloc_id() has published the prog + * to the userspace and the userspace may + * have refcnt-ed it through BPF_PROG_GET_FD_BY_ID. + */ + bpf_prog_put(prog); + return err; + } + + bpf_prog_kallsyms_add(prog); + trace_bpf_prog_load(prog, err); return err; free_used_maps: @@ -920,13 +1045,14 @@ static int bpf_obj_get(const union bpf_attr *attr) #ifdef CONFIG_CGROUP_BPF -#define BPF_PROG_ATTACH_LAST_FIELD attach_type +#define BPF_PROG_ATTACH_LAST_FIELD attach_flags static int bpf_prog_attach(const union bpf_attr *attr) { + enum bpf_prog_type ptype; struct bpf_prog *prog; struct cgroup *cgrp; - enum bpf_prog_type ptype; + int ret; if (!capable(CAP_NET_ADMIN)) return -EPERM; @@ -934,6 +1060,9 @@ static int bpf_prog_attach(const union bpf_attr *attr) if (CHECK_ATTR(BPF_PROG_ATTACH)) return -EINVAL; + if (attr->attach_flags & ~BPF_F_ALLOW_OVERRIDE) + return -EINVAL; + switch (attr->attach_type) { case BPF_CGROUP_INET_INGRESS: case BPF_CGROUP_INET_EGRESS: @@ -942,6 +1071,9 @@ static int bpf_prog_attach(const union bpf_attr *attr) case BPF_CGROUP_INET_SOCK_CREATE: ptype = BPF_PROG_TYPE_CGROUP_SOCK; break; + case BPF_CGROUP_SOCK_OPS: + ptype = BPF_PROG_TYPE_SOCK_OPS; + break; default: return -EINVAL; } @@ -956,10 +1088,13 @@ static int bpf_prog_attach(const union bpf_attr *attr) return PTR_ERR(cgrp); } - cgroup_bpf_update(cgrp, prog, attr->attach_type); + ret = cgroup_bpf_update(cgrp, prog, attr->attach_type, + attr->attach_flags & BPF_F_ALLOW_OVERRIDE); + if (ret) + bpf_prog_put(prog); cgroup_put(cgrp); - return 0; + return ret; } #define BPF_PROG_DETACH_LAST_FIELD attach_type @@ -967,6 +1102,7 @@ static int bpf_prog_attach(const union bpf_attr *attr) static int bpf_prog_detach(const union bpf_attr *attr) { struct cgroup *cgrp; + int ret; if (!capable(CAP_NET_ADMIN)) return -EPERM; @@ -978,11 +1114,12 @@ static int bpf_prog_detach(const union bpf_attr *attr) case BPF_CGROUP_INET_INGRESS: case BPF_CGROUP_INET_EGRESS: case BPF_CGROUP_INET_SOCK_CREATE: + case BPF_CGROUP_SOCK_OPS: cgrp = cgroup_get_from_fd(attr->target_fd); if (IS_ERR(cgrp)) return PTR_ERR(cgrp); - cgroup_bpf_update(cgrp, NULL, attr->attach_type); + ret = cgroup_bpf_update(cgrp, NULL, attr->attach_type, false); cgroup_put(cgrp); break; @@ -990,10 +1127,264 @@ static int bpf_prog_detach(const union bpf_attr *attr) return -EINVAL; } - return 0; + return ret; } + #endif /* CONFIG_CGROUP_BPF */ +#define BPF_PROG_TEST_RUN_LAST_FIELD test.duration + +static int bpf_prog_test_run(const union bpf_attr *attr, + union bpf_attr __user *uattr) +{ + struct bpf_prog *prog; + int ret = -ENOTSUPP; + + if (CHECK_ATTR(BPF_PROG_TEST_RUN)) + return -EINVAL; + + prog = bpf_prog_get(attr->test.prog_fd); + if (IS_ERR(prog)) + return PTR_ERR(prog); + + if (prog->aux->ops->test_run) + ret = prog->aux->ops->test_run(prog, attr, uattr); + + bpf_prog_put(prog); + return ret; +} + +#define BPF_OBJ_GET_NEXT_ID_LAST_FIELD next_id + +static int bpf_obj_get_next_id(const union bpf_attr *attr, + union bpf_attr __user *uattr, + struct idr *idr, + spinlock_t *lock) +{ + u32 next_id = attr->start_id; + int err = 0; + + if (CHECK_ATTR(BPF_OBJ_GET_NEXT_ID) || next_id >= INT_MAX) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + next_id++; + spin_lock_bh(lock); + if (!idr_get_next(idr, &next_id)) + err = -ENOENT; + spin_unlock_bh(lock); + + if (!err) + err = put_user(next_id, &uattr->next_id); + + return err; +} + +#define BPF_PROG_GET_FD_BY_ID_LAST_FIELD prog_id + +static int bpf_prog_get_fd_by_id(const union bpf_attr *attr) +{ + struct bpf_prog *prog; + u32 id = attr->prog_id; + int fd; + + if (CHECK_ATTR(BPF_PROG_GET_FD_BY_ID)) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + spin_lock_bh(&prog_idr_lock); + prog = idr_find(&prog_idr, id); + if (prog) + prog = bpf_prog_inc_not_zero(prog); + else + prog = ERR_PTR(-ENOENT); + spin_unlock_bh(&prog_idr_lock); + + if (IS_ERR(prog)) + return PTR_ERR(prog); + + fd = bpf_prog_new_fd(prog); + if (fd < 0) + bpf_prog_put(prog); + + return fd; +} + +#define BPF_MAP_GET_FD_BY_ID_LAST_FIELD map_id + +static int bpf_map_get_fd_by_id(const union bpf_attr *attr) +{ + struct bpf_map *map; + u32 id = attr->map_id; + int fd; + + if (CHECK_ATTR(BPF_MAP_GET_FD_BY_ID)) + return -EINVAL; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + spin_lock_bh(&map_idr_lock); + map = idr_find(&map_idr, id); + if (map) + map = bpf_map_inc_not_zero(map, true); + else + map = ERR_PTR(-ENOENT); + spin_unlock_bh(&map_idr_lock); + + if (IS_ERR(map)) + return PTR_ERR(map); + + fd = bpf_map_new_fd(map); + if (fd < 0) + bpf_map_put(map); + + return fd; +} + +static int check_uarg_tail_zero(void __user *uaddr, + size_t expected_size, + size_t actual_size) +{ + unsigned char __user *addr; + unsigned char __user *end; + unsigned char val; + int err; + + if (actual_size <= expected_size) + return 0; + + addr = uaddr + expected_size; + end = uaddr + actual_size; + + for (; addr < end; addr++) { + err = get_user(val, addr); + if (err) + return err; + if (val) + return -E2BIG; + } + + return 0; +} + +static int bpf_prog_get_info_by_fd(struct bpf_prog *prog, + const union bpf_attr *attr, + union bpf_attr __user *uattr) +{ + struct bpf_prog_info __user *uinfo = u64_to_user_ptr(attr->info.info); + struct bpf_prog_info info = {}; + u32 info_len = attr->info.info_len; + char __user *uinsns; + u32 ulen; + int err; + + err = check_uarg_tail_zero(uinfo, sizeof(info), info_len); + if (err) + return err; + info_len = min_t(u32, sizeof(info), info_len); + + if (copy_from_user(&info, uinfo, info_len)) + return err; + + info.type = prog->type; + info.id = prog->aux->id; + + memcpy(info.tag, prog->tag, sizeof(prog->tag)); + + if (!capable(CAP_SYS_ADMIN)) { + info.jited_prog_len = 0; + info.xlated_prog_len = 0; + goto done; + } + + ulen = info.jited_prog_len; + info.jited_prog_len = prog->jited_len; + if (info.jited_prog_len && ulen) { + uinsns = u64_to_user_ptr(info.jited_prog_insns); + ulen = min_t(u32, info.jited_prog_len, ulen); + if (copy_to_user(uinsns, prog->bpf_func, ulen)) + return -EFAULT; + } + + ulen = info.xlated_prog_len; + info.xlated_prog_len = bpf_prog_size(prog->len); + if (info.xlated_prog_len && ulen) { + uinsns = u64_to_user_ptr(info.xlated_prog_insns); + ulen = min_t(u32, info.xlated_prog_len, ulen); + if (copy_to_user(uinsns, prog->insnsi, ulen)) + return -EFAULT; + } + +done: + if (copy_to_user(uinfo, &info, info_len) || + put_user(info_len, &uattr->info.info_len)) + return -EFAULT; + + return 0; +} + +static int bpf_map_get_info_by_fd(struct bpf_map *map, + const union bpf_attr *attr, + union bpf_attr __user *uattr) +{ + struct bpf_map_info __user *uinfo = u64_to_user_ptr(attr->info.info); + struct bpf_map_info info = {}; + u32 info_len = attr->info.info_len; + int err; + + err = check_uarg_tail_zero(uinfo, sizeof(info), info_len); + if (err) + return err; + info_len = min_t(u32, sizeof(info), info_len); + + info.type = map->map_type; + info.id = map->id; + info.key_size = map->key_size; + info.value_size = map->value_size; + info.max_entries = map->max_entries; + info.map_flags = map->map_flags; + + if (copy_to_user(uinfo, &info, info_len) || + put_user(info_len, &uattr->info.info_len)) + return -EFAULT; + + return 0; +} + +#define BPF_OBJ_GET_INFO_BY_FD_LAST_FIELD info.info + +static int bpf_obj_get_info_by_fd(const union bpf_attr *attr, + union bpf_attr __user *uattr) +{ + int ufd = attr->info.bpf_fd; + struct fd f; + int err; + + if (CHECK_ATTR(BPF_OBJ_GET_INFO_BY_FD)) + return -EINVAL; + + f = fdget(ufd); + if (!f.file) + return -EBADFD; + + if (f.file->f_op == &bpf_prog_fops) + err = bpf_prog_get_info_by_fd(f.file->private_data, attr, + uattr); + else if (f.file->f_op == &bpf_map_fops) + err = bpf_map_get_info_by_fd(f.file->private_data, attr, + uattr); + else + err = -EINVAL; + + fdput(f); + return err; +} + SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, size) { union bpf_attr attr = {}; @@ -1013,23 +1404,10 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz * user-space does not rely on any kernel feature * extensions we dont know about yet. */ - if (size > sizeof(attr)) { - unsigned char __user *addr; - unsigned char __user *end; - unsigned char val; - - addr = (void __user *)uattr + sizeof(attr); - end = (void __user *)uattr + size; - - for (; addr < end; addr++) { - err = get_user(val, addr); - if (err) - return err; - if (val) - return -E2BIG; - } - size = sizeof(attr); - } + err = check_uarg_tail_zero(uattr, sizeof(attr), size); + if (err) + return err; + size = min_t(u32, size, sizeof(attr)); /* copy attributes from user space, may be less than sizeof(bpf_attr) */ if (copy_from_user(&attr, uattr, size) != 0) @@ -1060,7 +1438,6 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz case BPF_OBJ_GET: err = bpf_obj_get(&attr); break; - #ifdef CONFIG_CGROUP_BPF case BPF_PROG_ATTACH: err = bpf_prog_attach(&attr); @@ -1069,7 +1446,26 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz err = bpf_prog_detach(&attr); break; #endif - + case BPF_PROG_TEST_RUN: + err = bpf_prog_test_run(&attr, uattr); + break; + case BPF_PROG_GET_NEXT_ID: + err = bpf_obj_get_next_id(&attr, uattr, + &prog_idr, &prog_idr_lock); + break; + case BPF_MAP_GET_NEXT_ID: + err = bpf_obj_get_next_id(&attr, uattr, + &map_idr, &map_idr_lock); + break; + case BPF_PROG_GET_FD_BY_ID: + err = bpf_prog_get_fd_by_id(&attr); + break; + case BPF_MAP_GET_FD_BY_ID: + err = bpf_map_get_fd_by_id(&attr); + break; + case BPF_OBJ_GET_INFO_BY_FD: + err = bpf_obj_get_info_by_fd(&attr, uattr); + break; default: err = -EINVAL; break; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index cdc43b899f28..af9e84a4944e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -33,7 +33,7 @@ * - out of bounds or malformed jumps * The second pass is all possible path descent from the 1st insn. * Since it's analyzing all pathes through the program, the length of the - * analysis is limited to 32k insn, which may be hit even if total number of + * analysis is limited to 64k insn, which may be hit even if total number of * insn is less then 4K, but there are too many branches that change stack/regs. * Number of 'branches to be analyzed' is limited to 1k * @@ -140,9 +140,11 @@ struct bpf_verifier_stack_elem { struct bpf_verifier_stack_elem *next; }; -#define BPF_COMPLEXITY_LIMIT_INSNS 65536 +#define BPF_COMPLEXITY_LIMIT_INSNS 98304 #define BPF_COMPLEXITY_LIMIT_STACK 1024 +#define BPF_MAP_PTR_POISON ((void *)0xeB9F + POISON_POINTER_DELTA) + struct bpf_call_arg_meta { struct bpf_map *map_ptr; bool raw_mode; @@ -239,6 +241,12 @@ static void print_verifier_state(struct bpf_verifier_state *state) if (reg->max_value != BPF_REGISTER_MAX_RANGE) verbose(",max_value=%llu", (unsigned long long)reg->max_value); + if (reg->min_align) + verbose(",min_align=%u", reg->min_align); + if (reg->aux_off) + verbose(",aux_off=%u", reg->aux_off); + if (reg->aux_off_align) + verbose(",aux_off_align=%u", reg->aux_off_align); } for (i = 0; i < MAX_BPF_STACK; i += BPF_REG_SIZE) { if (state->stack_slot_type[i] == STACK_SPILL) @@ -296,7 +304,8 @@ static const char *const bpf_jmp_string[16] = { [BPF_EXIT >> 4] = "exit", }; -static void print_bpf_insn(struct bpf_insn *insn) +static void print_bpf_insn(const struct bpf_verifier_env *env, + const struct bpf_insn *insn) { u8 class = BPF_CLASS(insn->code); @@ -360,9 +369,19 @@ static void print_bpf_insn(struct bpf_insn *insn) insn->code, bpf_ldst_string[BPF_SIZE(insn->code) >> 3], insn->src_reg, insn->imm); - } else if (BPF_MODE(insn->code) == BPF_IMM) { - verbose("(%02x) r%d = 0x%x\n", - insn->code, insn->dst_reg, insn->imm); + } else if (BPF_MODE(insn->code) == BPF_IMM && + BPF_SIZE(insn->code) == BPF_DW) { + /* At this point, we already made sure that the second + * part of the ldimm64 insn is accessible. + */ + u64 imm = ((u64)(insn + 1)->imm << 32) | (u32)insn->imm; + bool map_ptr = insn->src_reg == BPF_PSEUDO_MAP_FD; + + if (map_ptr && !env->allow_ptr_leaks) + imm = 0; + + verbose("(%02x) r%d = 0x%llx\n", insn->code, + insn->dst_reg, (unsigned long long)imm); } else { verbose("BUG_ld_%02x\n", insn->code); return; @@ -444,16 +463,22 @@ static const int caller_saved[CALLER_SAVED_REGS] = { BPF_REG_0, BPF_REG_1, BPF_REG_2, BPF_REG_3, BPF_REG_4, BPF_REG_5 }; +static void mark_reg_not_init(struct bpf_reg_state *regs, u32 regno) +{ + BUG_ON(regno >= MAX_BPF_REG); + + memset(®s[regno], 0, sizeof(regs[regno])); + regs[regno].type = NOT_INIT; + regs[regno].min_value = BPF_REGISTER_MIN_RANGE; + regs[regno].max_value = BPF_REGISTER_MAX_RANGE; +} + static void init_reg_state(struct bpf_reg_state *regs) { int i; - for (i = 0; i < MAX_BPF_REG; i++) { - regs[i].type = NOT_INIT; - regs[i].imm = 0; - regs[i].min_value = BPF_REGISTER_MIN_RANGE; - regs[i].max_value = BPF_REGISTER_MAX_RANGE; - } + for (i = 0; i < MAX_BPF_REG; i++) + mark_reg_not_init(regs, i); /* frame pointer */ regs[BPF_REG_FP].type = FRAME_PTR; @@ -479,6 +504,15 @@ static void reset_reg_range_values(struct bpf_reg_state *regs, u32 regno) { regs[regno].min_value = BPF_REGISTER_MIN_RANGE; regs[regno].max_value = BPF_REGISTER_MAX_RANGE; + regs[regno].value_from_signed = false; + regs[regno].min_align = 0; +} + +static void mark_reg_unknown_value_and_range(struct bpf_reg_state *regs, + u32 regno) +{ + mark_reg_unknown_value(regs, regno); + reset_reg_range_values(regs, regno); } enum reg_arg_type { @@ -513,25 +547,12 @@ static int check_reg_arg(struct bpf_reg_state *regs, u32 regno, return 0; } -static int bpf_size_to_bytes(int bpf_size) -{ - if (bpf_size == BPF_W) - return 4; - else if (bpf_size == BPF_H) - return 2; - else if (bpf_size == BPF_B) - return 1; - else if (bpf_size == BPF_DW) - return 8; - else - return -EINVAL; -} - static bool is_spillable_regtype(enum bpf_reg_type type) { switch (type) { case PTR_TO_MAP_VALUE: case PTR_TO_MAP_VALUE_OR_NULL: + case PTR_TO_MAP_VALUE_ADJ: case PTR_TO_STACK: case PTR_TO_CTX: case PTR_TO_PACKET: @@ -616,7 +637,8 @@ static int check_stack_read(struct bpf_verifier_state *state, int off, int size, } if (value_regno >= 0) /* have read misc data from the stack */ - mark_reg_unknown_value(state->regs, value_regno); + mark_reg_unknown_value_and_range(state->regs, + value_regno); return 0; } } @@ -627,7 +649,7 @@ static int check_map_access(struct bpf_verifier_env *env, u32 regno, int off, { struct bpf_map *map = env->cur_state.regs[regno].map_ptr; - if (off < 0 || off + size > map->value_size) { + if (off < 0 || size <= 0 || off + size > map->value_size) { verbose("invalid access to map value, value_size=%d off=%d size=%d\n", map->value_size, off, size); return -EACCES; @@ -635,6 +657,51 @@ static int check_map_access(struct bpf_verifier_env *env, u32 regno, int off, return 0; } +/* check read/write into an adjusted map element */ +static int check_map_access_adj(struct bpf_verifier_env *env, u32 regno, + int off, int size) +{ + struct bpf_verifier_state *state = &env->cur_state; + struct bpf_reg_state *reg = &state->regs[regno]; + int err; + + /* We adjusted the register to this map value, so we + * need to change off and size to min_value and max_value + * respectively to make sure our theoretical access will be + * safe. + */ + if (log_level) + print_verifier_state(state); + env->varlen_map_value_access = true; + /* The minimum value is only important with signed + * comparisons where we can't assume the floor of a + * value is 0. If we are using signed variables for our + * index'es we need to make sure that whatever we use + * will have a set floor within our range. + */ + if (reg->min_value < 0) { + verbose("R%d min value is negative, either use unsigned index or do a if (index >=0) check.\n", + regno); + return -EACCES; + } + err = check_map_access(env, regno, reg->min_value + off, size); + if (err) { + verbose("R%d min value is outside of the array range\n", + regno); + return err; + } + + /* If we haven't set a max value then we need to bail + * since we can't be sure we won't do bad things. + */ + if (reg->max_value == BPF_REGISTER_MAX_RANGE) { + verbose("R%d unbounded memory access, make sure to bounds check any array access into a map\n", + regno); + return -EACCES; + } + return check_map_access(env, regno, reg->max_value + off, size); +} + #define MAX_PACKET_OFF 0xffff static bool may_access_direct_pkt_data(struct bpf_verifier_env *env, @@ -647,6 +714,7 @@ static bool may_access_direct_pkt_data(struct bpf_verifier_env *env, /* dst_input() and dst_output() can't write for now */ if (t == BPF_WRITE) return false; + /* fallthrough */ case BPF_PROG_TYPE_SCHED_CLS: case BPF_PROG_TYPE_SCHED_ACT: case BPF_PROG_TYPE_XDP: @@ -677,15 +745,29 @@ static int check_packet_access(struct bpf_verifier_env *env, u32 regno, int off, } /* check access to 'struct bpf_context' fields */ -static int check_ctx_access(struct bpf_verifier_env *env, int off, int size, +static int check_ctx_access(struct bpf_verifier_env *env, int insn_idx, int off, int size, enum bpf_access_type t, enum bpf_reg_type *reg_type) { + struct bpf_insn_access_aux info = { + .reg_type = *reg_type, + }; + /* for analyzer ctx accesses are already validated and converted */ if (env->analyzer_ops) return 0; if (env->prog->aux->ops->is_valid_access && - env->prog->aux->ops->is_valid_access(off, size, t, reg_type)) { + env->prog->aux->ops->is_valid_access(off, size, t, &info)) { + /* A non zero info.ctx_field_size indicates that this field is a + * candidate for later verifier transformation to load the whole + * field and then apply a mask when accessed with a narrower + * access than actual ctx access size. A zero info.ctx_field_size + * will only allow for whole field access and rejects any other + * type of narrower access. + */ + env->insn_aux_data[insn_idx].ctx_field_size = info.ctx_field_size; + *reg_type = info.reg_type; + /* remember the offset of last byte accessed in ctx */ if (env->prog->aux->max_ctx_offset < off + size) env->prog->aux->max_ctx_offset = off + size; @@ -696,12 +778,13 @@ static int check_ctx_access(struct bpf_verifier_env *env, int off, int size, return -EACCES; } -static bool is_pointer_value(struct bpf_verifier_env *env, int regno) +static bool __is_pointer_value(bool allow_ptr_leaks, + const struct bpf_reg_state *reg) { - if (env->allow_ptr_leaks) + if (allow_ptr_leaks) return false; - switch (env->cur_state.regs[regno].type) { + switch (reg->type) { case UNKNOWN_VALUE: case CONST_IMM: return false; @@ -710,45 +793,89 @@ static bool is_pointer_value(struct bpf_verifier_env *env, int regno) } } -static int check_ptr_alignment(struct bpf_verifier_env *env, - struct bpf_reg_state *reg, int off, int size) +static bool is_pointer_value(struct bpf_verifier_env *env, int regno) { - if (reg->type != PTR_TO_PACKET && reg->type != PTR_TO_MAP_VALUE_ADJ) { - if (off % size != 0) { - verbose("misaligned access off %d size %d\n", - off, size); + return __is_pointer_value(env->allow_ptr_leaks, &env->cur_state.regs[regno]); +} + +static int check_pkt_ptr_alignment(const struct bpf_reg_state *reg, + int off, int size, bool strict) +{ + int ip_align; + int reg_off; + + /* Byte size accesses are always allowed. */ + if (!strict || size == 1) + return 0; + + reg_off = reg->off; + if (reg->id) { + if (reg->aux_off_align % size) { + verbose("Packet access is only %u byte aligned, %d byte access not allowed\n", + reg->aux_off_align, size); return -EACCES; - } else { - return 0; } + reg_off += reg->aux_off; } - if (IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) - /* misaligned access to packet is ok on x86,arm,arm64 */ - return 0; - - if (reg->id && size != 1) { - verbose("Unknown packet alignment. Only byte-sized access allowed\n"); + /* For platforms that do not have a Kconfig enabling + * CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS the value of + * NET_IP_ALIGN is universally set to '2'. And on platforms + * that do set CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS, we get + * to this code only in strict mode where we want to emulate + * the NET_IP_ALIGN==2 checking. Therefore use an + * unconditional IP align value of '2'. + */ + ip_align = 2; + if ((ip_align + reg_off + off) % size != 0) { + verbose("misaligned packet access off %d+%d+%d size %d\n", + ip_align, reg_off, off, size); return -EACCES; } - /* skb->data is NET_IP_ALIGN-ed */ - if (reg->type == PTR_TO_PACKET && - (NET_IP_ALIGN + reg->off + off) % size != 0) { - verbose("misaligned packet access off %d+%d+%d size %d\n", - NET_IP_ALIGN, reg->off, off, size); + return 0; +} + +static int check_val_ptr_alignment(const struct bpf_reg_state *reg, + int size, bool strict) +{ + if (strict && size != 1) { + verbose("Unknown alignment. Only byte-sized access allowed in value access.\n"); return -EACCES; } + return 0; } +static int check_ptr_alignment(struct bpf_verifier_env *env, + const struct bpf_reg_state *reg, + int off, int size) +{ + bool strict = env->strict_alignment; + + switch (reg->type) { + case PTR_TO_PACKET: + return check_pkt_ptr_alignment(reg, off, size, strict); + case PTR_TO_MAP_VALUE_ADJ: + return check_val_ptr_alignment(reg, size, strict); + default: + if (off % size != 0) { + verbose("misaligned access off %d size %d\n", + off, size); + return -EACCES; + } + + return 0; + } +} + /* check whether memory at (regno + off) is accessible for t = (read | write) * if t==write, value_regno is a register which value is stored into memory * if t==read, value_regno is a register which will receive the value from memory * if t==write && value_regno==-1, some unknown value is stored into memory * if t==read && value_regno==-1, don't care what we read from memory */ -static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, +static int check_mem_access(struct bpf_verifier_env *env, int insn_idx, u32 regno, int off, int bpf_size, enum bpf_access_type t, int value_regno) { @@ -775,47 +902,13 @@ static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, return -EACCES; } - /* If we adjusted the register to this map value at all then we - * need to change off and size to min_value and max_value - * respectively to make sure our theoretical access will be - * safe. - */ - if (reg->type == PTR_TO_MAP_VALUE_ADJ) { - if (log_level) - print_verifier_state(state); - env->varlen_map_value_access = true; - /* The minimum value is only important with signed - * comparisons where we can't assume the floor of a - * value is 0. If we are using signed variables for our - * index'es we need to make sure that whatever we use - * will have a set floor within our range. - */ - if (reg->min_value < 0) { - verbose("R%d min value is negative, either use unsigned index or do a if (index >=0) check.\n", - regno); - return -EACCES; - } - err = check_map_access(env, regno, reg->min_value + off, - size); - if (err) { - verbose("R%d min value is outside of the array range\n", - regno); - return err; - } - - /* If we haven't set a max value then we need to bail - * since we can't be sure we won't do bad things. - */ - if (reg->max_value == BPF_REGISTER_MAX_RANGE) { - verbose("R%d unbounded memory access, make sure to bounds check any array access into a map\n", - regno); - return -EACCES; - } - off += reg->max_value; - } - err = check_map_access(env, regno, off, size); + if (reg->type == PTR_TO_MAP_VALUE_ADJ) + err = check_map_access_adj(env, regno, off, size); + else + err = check_map_access(env, regno, off, size); if (!err && t == BPF_READ && value_regno >= 0) - mark_reg_unknown_value(state->regs, value_regno); + mark_reg_unknown_value_and_range(state->regs, + value_regno); } else if (reg->type == PTR_TO_CTX) { enum bpf_reg_type reg_type = UNKNOWN_VALUE; @@ -825,11 +918,14 @@ static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, verbose("R%d leaks addr into ctx\n", value_regno); return -EACCES; } - err = check_ctx_access(env, off, size, t, ®_type); + err = check_ctx_access(env, insn_idx, off, size, t, ®_type); if (!err && t == BPF_READ && value_regno >= 0) { - mark_reg_unknown_value(state->regs, value_regno); + mark_reg_unknown_value_and_range(state->regs, + value_regno); /* note that reg.[id|off|range] == 0 */ state->regs[value_regno].type = reg_type; + state->regs[value_regno].aux_off = 0; + state->regs[value_regno].aux_off_align = 0; } } else if (reg->type == FRAME_PTR || reg->type == PTR_TO_STACK) { @@ -837,6 +933,10 @@ static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, verbose("invalid stack off=%d size=%d\n", off, size); return -EACCES; } + + if (env->prog->aux->stack_depth < -off) + env->prog->aux->stack_depth = -off; + if (t == BPF_WRITE) { if (!env->allow_ptr_leaks && state->stack_slot_type[MAX_BPF_STACK + off] == STACK_SPILL && @@ -860,7 +960,8 @@ static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, } err = check_packet_access(env, regno, off, size); if (!err && t == BPF_READ && value_regno >= 0) - mark_reg_unknown_value(state->regs, value_regno); + mark_reg_unknown_value_and_range(state->regs, + value_regno); } else { verbose("R%d invalid mem access '%s'\n", regno, reg_type_str[reg->type]); @@ -878,7 +979,7 @@ static int check_mem_access(struct bpf_verifier_env *env, u32 regno, int off, return err; } -static int check_xadd(struct bpf_verifier_env *env, struct bpf_insn *insn) +static int check_xadd(struct bpf_verifier_env *env, int insn_idx, struct bpf_insn *insn) { struct bpf_reg_state *regs = env->cur_state.regs; int err; @@ -899,14 +1000,19 @@ static int check_xadd(struct bpf_verifier_env *env, struct bpf_insn *insn) if (err) return err; + if (is_pointer_value(env, insn->src_reg)) { + verbose("R%d leaks addr into mem\n", insn->src_reg); + return -EACCES; + } + /* check whether atomic_add can read the memory */ - err = check_mem_access(env, insn->dst_reg, insn->off, + err = check_mem_access(env, insn_idx, insn->dst_reg, insn->off, BPF_SIZE(insn->code), BPF_READ, -1); if (err) return err; /* check whether atomic_add can write into the same memory */ - return check_mem_access(env, insn->dst_reg, insn->off, + return check_mem_access(env, insn_idx, insn->dst_reg, insn->off, BPF_SIZE(insn->code), BPF_WRITE, -1); } @@ -942,6 +1048,9 @@ static int check_stack_boundary(struct bpf_verifier_env *env, int regno, return -EACCES; } + if (env->prog->aux->stack_depth < -off) + env->prog->aux->stack_depth = -off; + if (meta && meta->raw_mode) { meta->access_size = access_size; meta->regno = regno; @@ -958,6 +1067,25 @@ static int check_stack_boundary(struct bpf_verifier_env *env, int regno, return 0; } +static int check_helper_mem_access(struct bpf_verifier_env *env, int regno, + int access_size, bool zero_size_allowed, + struct bpf_call_arg_meta *meta) +{ + struct bpf_reg_state *regs = env->cur_state.regs; + + switch (regs[regno].type) { + case PTR_TO_PACKET: + return check_packet_access(env, regno, 0, access_size); + case PTR_TO_MAP_VALUE: + return check_map_access(env, regno, 0, access_size); + case PTR_TO_MAP_VALUE_ADJ: + return check_map_access_adj(env, regno, 0, access_size); + default: /* const_imm|ptr_to_stack or invalid ptr */ + return check_stack_boundary(env, regno, access_size, + zero_size_allowed, meta); + } +} + static int check_func_arg(struct bpf_verifier_env *env, u32 regno, enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta) @@ -993,10 +1121,13 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno, expected_type = PTR_TO_STACK; if (type != PTR_TO_PACKET && type != expected_type) goto err_type; - } else if (arg_type == ARG_CONST_STACK_SIZE || - arg_type == ARG_CONST_STACK_SIZE_OR_ZERO) { + } else if (arg_type == ARG_CONST_SIZE || + arg_type == ARG_CONST_SIZE_OR_ZERO) { expected_type = CONST_IMM; - if (type != expected_type) + /* One exception. Allow UNKNOWN_VALUE registers when the + * boundaries are known and don't cause unsafe memory accesses + */ + if (type != UNKNOWN_VALUE && type != expected_type) goto err_type; } else if (arg_type == ARG_CONST_MAP_PTR) { expected_type = CONST_PTR_TO_MAP; @@ -1006,8 +1137,8 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno, expected_type = PTR_TO_CTX; if (type != expected_type) goto err_type; - } else if (arg_type == ARG_PTR_TO_STACK || - arg_type == ARG_PTR_TO_RAW_STACK) { + } else if (arg_type == ARG_PTR_TO_MEM || + arg_type == ARG_PTR_TO_UNINIT_MEM) { expected_type = PTR_TO_STACK; /* One exception here. In case function allows for NULL to be * passed in as argument, it's a CONST_IMM type. Final test @@ -1015,9 +1146,10 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno, */ if (type == CONST_IMM && reg->imm == 0) /* final test in check_stack_boundary() */; - else if (type != PTR_TO_PACKET && type != expected_type) + else if (type != PTR_TO_PACKET && type != PTR_TO_MAP_VALUE && + type != PTR_TO_MAP_VALUE_ADJ && type != expected_type) goto err_type; - meta->raw_mode = arg_type == ARG_PTR_TO_RAW_STACK; + meta->raw_mode = arg_type == ARG_PTR_TO_UNINIT_MEM; } else { verbose("unsupported arg_type %d\n", arg_type); return -EFAULT; @@ -1063,9 +1195,9 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno, err = check_stack_boundary(env, regno, meta->map_ptr->value_size, false, NULL); - } else if (arg_type == ARG_CONST_STACK_SIZE || - arg_type == ARG_CONST_STACK_SIZE_OR_ZERO) { - bool zero_size_allowed = (arg_type == ARG_CONST_STACK_SIZE_OR_ZERO); + } else if (arg_type == ARG_CONST_SIZE || + arg_type == ARG_CONST_SIZE_OR_ZERO) { + bool zero_size_allowed = (arg_type == ARG_CONST_SIZE_OR_ZERO); /* bpf_xxx(..., buf, len) call will access 'len' bytes * from stack pointer 'buf'. Check it @@ -1073,14 +1205,50 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno, */ if (regno == 0) { /* kernel subsystem misconfigured verifier */ - verbose("ARG_CONST_STACK_SIZE cannot be first argument\n"); + verbose("ARG_CONST_SIZE cannot be first argument\n"); return -EACCES; } - if (regs[regno - 1].type == PTR_TO_PACKET) - err = check_packet_access(env, regno - 1, 0, reg->imm); - else - err = check_stack_boundary(env, regno - 1, reg->imm, - zero_size_allowed, meta); + + /* If the register is UNKNOWN_VALUE, the access check happens + * using its boundaries. Otherwise, just use its imm + */ + if (type == UNKNOWN_VALUE) { + /* For unprivileged variable accesses, disable raw + * mode so that the program is required to + * initialize all the memory that the helper could + * just partially fill up. + */ + meta = NULL; + + if (reg->min_value < 0) { + verbose("R%d min value is negative, either use unsigned or 'var &= const'\n", + regno); + return -EACCES; + } + + if (reg->min_value == 0) { + err = check_helper_mem_access(env, regno - 1, 0, + zero_size_allowed, + meta); + if (err) + return err; + } + + if (reg->max_value == BPF_REGISTER_MAX_RANGE) { + verbose("R%d unbounded memory access, use 'var &= const' or 'if (var < const)'\n", + regno); + return -EACCES; + } + err = check_helper_mem_access(env, regno - 1, + reg->max_value, + zero_size_allowed, meta); + if (err) + return err; + } else { + /* register is CONST_IMM */ + err = check_helper_mem_access(env, regno - 1, reg->imm, + zero_size_allowed, meta); + } } return err; @@ -1115,6 +1283,10 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) func_id != BPF_FUNC_current_task_under_cgroup) goto error; break; + case BPF_MAP_TYPE_ARRAY_OF_MAPS: + case BPF_MAP_TYPE_HASH_OF_MAPS: + if (func_id != BPF_FUNC_map_lookup_elem) + goto error; default: break; } @@ -1154,15 +1326,15 @@ static int check_raw_mode(const struct bpf_func_proto *fn) { int count = 0; - if (fn->arg1_type == ARG_PTR_TO_RAW_STACK) + if (fn->arg1_type == ARG_PTR_TO_UNINIT_MEM) count++; - if (fn->arg2_type == ARG_PTR_TO_RAW_STACK) + if (fn->arg2_type == ARG_PTR_TO_UNINIT_MEM) count++; - if (fn->arg3_type == ARG_PTR_TO_RAW_STACK) + if (fn->arg3_type == ARG_PTR_TO_UNINIT_MEM) count++; - if (fn->arg4_type == ARG_PTR_TO_RAW_STACK) + if (fn->arg4_type == ARG_PTR_TO_UNINIT_MEM) count++; - if (fn->arg5_type == ARG_PTR_TO_RAW_STACK) + if (fn->arg5_type == ARG_PTR_TO_UNINIT_MEM) count++; return count > 1 ? -EINVAL : 0; @@ -1186,17 +1358,16 @@ static void clear_all_pkt_pointers(struct bpf_verifier_env *env) if (reg->type != PTR_TO_PACKET && reg->type != PTR_TO_PACKET_END) continue; - reg->type = UNKNOWN_VALUE; - reg->imm = 0; + __mark_reg_unknown_value(state->spilled_regs, + i / BPF_REG_SIZE); } } -static int check_call(struct bpf_verifier_env *env, int func_id) +static int check_call(struct bpf_verifier_env *env, int func_id, int insn_idx) { struct bpf_verifier_state *state = &env->cur_state; const struct bpf_func_proto *fn = NULL; struct bpf_reg_state *regs = state->regs; - struct bpf_reg_state *reg; struct bpf_call_arg_meta meta; bool changes_data; int i, err; @@ -1257,17 +1428,14 @@ static int check_call(struct bpf_verifier_env *env, int func_id) * is inferred from register state. */ for (i = 0; i < meta.access_size; i++) { - err = check_mem_access(env, meta.regno, i, BPF_B, BPF_WRITE, -1); + err = check_mem_access(env, insn_idx, meta.regno, i, BPF_B, BPF_WRITE, -1); if (err) return err; } /* reset caller saved regs */ - for (i = 0; i < CALLER_SAVED_REGS; i++) { - reg = regs + caller_saved[i]; - reg->type = NOT_INIT; - reg->imm = 0; - } + for (i = 0; i < CALLER_SAVED_REGS; i++) + mark_reg_not_init(regs, caller_saved[i]); /* update return register */ if (fn->ret_type == RET_INTEGER) { @@ -1275,6 +1443,8 @@ static int check_call(struct bpf_verifier_env *env, int func_id) } else if (fn->ret_type == RET_VOID) { regs[BPF_REG_0].type = NOT_INIT; } else if (fn->ret_type == RET_PTR_TO_MAP_VALUE_OR_NULL) { + struct bpf_insn_aux_data *insn_aux; + regs[BPF_REG_0].type = PTR_TO_MAP_VALUE_OR_NULL; regs[BPF_REG_0].max_value = regs[BPF_REG_0].min_value = 0; /* remember map_ptr, so that check_map_access() @@ -1287,6 +1457,11 @@ static int check_call(struct bpf_verifier_env *env, int func_id) } regs[BPF_REG_0].map_ptr = meta.map_ptr; regs[BPF_REG_0].id = ++env->id_gen; + insn_aux = &env->insn_aux_data[insn_idx]; + if (!insn_aux->map_ptr) + insn_aux->map_ptr = meta.map_ptr; + else if (insn_aux->map_ptr != meta.map_ptr) + insn_aux->map_ptr = BPF_MAP_PTR_POISON; } else { verbose("unknown return type %d of func %s#%d\n", fn->ret_type, func_id_name(func_id), func_id); @@ -1316,7 +1491,7 @@ static int check_packet_ptr_add(struct bpf_verifier_env *env, imm = insn->imm; add_imm: - if (imm <= 0) { + if (imm < 0) { verbose("addition of negative constant to packet pointer is not allowed\n"); return -EACCES; } @@ -1331,6 +1506,8 @@ add_imm: */ dst_reg->off += imm; } else { + bool had_id; + if (src_reg->type == PTR_TO_PACKET) { /* R6=pkt(id=0,off=0,r=62) R7=imm22; r7 += r6 */ tmp_reg = *dst_reg; /* save r7 state */ @@ -1364,14 +1541,23 @@ add_imm: src_reg->imm); return -EACCES; } + + had_id = (dst_reg->id != 0); + /* dst_reg stays as pkt_ptr type and since some positive * integer value was added to the pointer, increment its 'id' */ dst_reg->id = ++env->id_gen; - /* something was added to pkt_ptr, set range and off to zero */ + /* something was added to pkt_ptr, set range to zero */ + dst_reg->aux_off += dst_reg->off; dst_reg->off = 0; dst_reg->range = 0; + if (had_id) + dst_reg->aux_off_align = min(dst_reg->aux_off_align, + src_reg->min_align); + else + dst_reg->aux_off_align = src_reg->min_align; } return 0; } @@ -1478,6 +1664,65 @@ static int evaluate_reg_alu(struct bpf_verifier_env *env, struct bpf_insn *insn) return 0; } +static int evaluate_reg_imm_alu_unknown(struct bpf_verifier_env *env, + struct bpf_insn *insn) +{ + struct bpf_reg_state *regs = env->cur_state.regs; + struct bpf_reg_state *dst_reg = ®s[insn->dst_reg]; + struct bpf_reg_state *src_reg = ®s[insn->src_reg]; + u8 opcode = BPF_OP(insn->code); + s64 imm_log2 = __ilog2_u64((long long)dst_reg->imm); + + /* BPF_X code with src_reg->type UNKNOWN_VALUE here. */ + if (src_reg->imm > 0 && dst_reg->imm) { + switch (opcode) { + case BPF_ADD: + /* dreg += sreg + * where both have zero upper bits. Adding them + * can only result making one more bit non-zero + * in the larger value. + * Ex. 0xffff (imm=48) + 1 (imm=63) = 0x10000 (imm=47) + * 0xffff (imm=48) + 0xffff = 0x1fffe (imm=47) + */ + dst_reg->imm = min(src_reg->imm, 63 - imm_log2); + dst_reg->imm--; + break; + case BPF_AND: + /* dreg &= sreg + * AND can not extend zero bits only shrink + * Ex. 0x00..00ffffff + * & 0x0f..ffffffff + * ---------------- + * 0x00..00ffffff + */ + dst_reg->imm = max(src_reg->imm, 63 - imm_log2); + break; + case BPF_OR: + /* dreg |= sreg + * OR can only extend zero bits + * Ex. 0x00..00ffffff + * | 0x0f..ffffffff + * ---------------- + * 0x0f..00ffffff + */ + dst_reg->imm = min(src_reg->imm, 63 - imm_log2); + break; + case BPF_SUB: + case BPF_MUL: + case BPF_RSH: + case BPF_LSH: + /* These may be flushed out later */ + default: + mark_reg_unknown_value(regs, insn->dst_reg); + } + } else { + mark_reg_unknown_value(regs, insn->dst_reg); + } + + dst_reg->type = UNKNOWN_VALUE; + return 0; +} + static int evaluate_reg_imm_alu(struct bpf_verifier_env *env, struct bpf_insn *insn) { @@ -1485,22 +1730,57 @@ static int evaluate_reg_imm_alu(struct bpf_verifier_env *env, struct bpf_reg_state *dst_reg = ®s[insn->dst_reg]; struct bpf_reg_state *src_reg = ®s[insn->src_reg]; u8 opcode = BPF_OP(insn->code); + u64 dst_imm = dst_reg->imm; - /* dst_reg->type == CONST_IMM here, simulate execution of 'add'/'or' - * insn. Don't care about overflow or negative values, just add them + if (BPF_SRC(insn->code) == BPF_X && src_reg->type == UNKNOWN_VALUE) + return evaluate_reg_imm_alu_unknown(env, insn); + + /* dst_reg->type == CONST_IMM here. Simulate execution of insns + * containing ALU ops. Don't care about overflow or negative + * values, just add/sub/... them; registers are in u64. */ - if (opcode == BPF_ADD && BPF_SRC(insn->code) == BPF_K) - dst_reg->imm += insn->imm; - else if (opcode == BPF_ADD && BPF_SRC(insn->code) == BPF_X && - src_reg->type == CONST_IMM) - dst_reg->imm += src_reg->imm; - else if (opcode == BPF_OR && BPF_SRC(insn->code) == BPF_K) - dst_reg->imm |= insn->imm; - else if (opcode == BPF_OR && BPF_SRC(insn->code) == BPF_X && - src_reg->type == CONST_IMM) - dst_reg->imm |= src_reg->imm; - else + if (opcode == BPF_ADD && BPF_SRC(insn->code) == BPF_K) { + dst_imm += insn->imm; + } else if (opcode == BPF_ADD && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm += src_reg->imm; + } else if (opcode == BPF_SUB && BPF_SRC(insn->code) == BPF_K) { + dst_imm -= insn->imm; + } else if (opcode == BPF_SUB && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm -= src_reg->imm; + } else if (opcode == BPF_MUL && BPF_SRC(insn->code) == BPF_K) { + dst_imm *= insn->imm; + } else if (opcode == BPF_MUL && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm *= src_reg->imm; + } else if (opcode == BPF_OR && BPF_SRC(insn->code) == BPF_K) { + dst_imm |= insn->imm; + } else if (opcode == BPF_OR && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm |= src_reg->imm; + } else if (opcode == BPF_AND && BPF_SRC(insn->code) == BPF_K) { + dst_imm &= insn->imm; + } else if (opcode == BPF_AND && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm &= src_reg->imm; + } else if (opcode == BPF_RSH && BPF_SRC(insn->code) == BPF_K) { + dst_imm >>= insn->imm; + } else if (opcode == BPF_RSH && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm >>= src_reg->imm; + } else if (opcode == BPF_LSH && BPF_SRC(insn->code) == BPF_K) { + dst_imm <<= insn->imm; + } else if (opcode == BPF_LSH && BPF_SRC(insn->code) == BPF_X && + src_reg->type == CONST_IMM) { + dst_imm <<= src_reg->imm; + } else { mark_reg_unknown_value(regs, insn->dst_reg); + goto out; + } + + dst_reg->imm = dst_imm; +out: return 0; } @@ -1513,6 +1793,13 @@ static void check_reg_overflow(struct bpf_reg_state *reg) reg->min_value = BPF_REGISTER_MIN_RANGE; } +static u32 calc_align(u32 imm) +{ + if (!imm) + return 1U << 31; + return imm - ((imm - 1) & imm); +} + static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, struct bpf_insn *insn) { @@ -1520,8 +1807,10 @@ static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, s64 min_val = BPF_REGISTER_MIN_RANGE; u64 max_val = BPF_REGISTER_MAX_RANGE; u8 opcode = BPF_OP(insn->code); + u32 dst_align, src_align; dst_reg = ®s[insn->dst_reg]; + src_align = 0; if (BPF_SRC(insn->code) == BPF_X) { check_reg_overflow(®s[insn->src_reg]); min_val = regs[insn->src_reg].min_value; @@ -1537,17 +1826,37 @@ static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, regs[insn->src_reg].type != UNKNOWN_VALUE) { min_val = BPF_REGISTER_MIN_RANGE; max_val = BPF_REGISTER_MAX_RANGE; + src_align = 0; + } else { + src_align = regs[insn->src_reg].min_align; } } else if (insn->imm < BPF_REGISTER_MAX_RANGE && (s64)insn->imm > BPF_REGISTER_MIN_RANGE) { min_val = max_val = insn->imm; + src_align = calc_align(insn->imm); } + dst_align = dst_reg->min_align; + /* We don't know anything about what was done to this register, mark it - * as unknown. + * as unknown. Also, if both derived bounds came from signed/unsigned + * mixed compares and one side is unbounded, we cannot really do anything + * with them as boundaries cannot be trusted. Thus, arithmetic of two + * regs of such kind will get invalidated bounds on the dst side. */ - if (min_val == BPF_REGISTER_MIN_RANGE && - max_val == BPF_REGISTER_MAX_RANGE) { + if ((min_val == BPF_REGISTER_MIN_RANGE && + max_val == BPF_REGISTER_MAX_RANGE) || + (BPF_SRC(insn->code) == BPF_X && + ((min_val != BPF_REGISTER_MIN_RANGE && + max_val == BPF_REGISTER_MAX_RANGE) || + (min_val == BPF_REGISTER_MIN_RANGE && + max_val != BPF_REGISTER_MAX_RANGE) || + (dst_reg->min_value != BPF_REGISTER_MIN_RANGE && + dst_reg->max_value == BPF_REGISTER_MAX_RANGE) || + (dst_reg->min_value == BPF_REGISTER_MIN_RANGE && + dst_reg->max_value != BPF_REGISTER_MAX_RANGE)) && + regs[insn->dst_reg].value_from_signed != + regs[insn->src_reg].value_from_signed)) { reset_reg_range_values(regs, insn->dst_reg); return; } @@ -1567,18 +1876,21 @@ static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, dst_reg->min_value += min_val; if (dst_reg->max_value != BPF_REGISTER_MAX_RANGE) dst_reg->max_value += max_val; + dst_reg->min_align = min(src_align, dst_align); break; case BPF_SUB: if (dst_reg->min_value != BPF_REGISTER_MIN_RANGE) dst_reg->min_value -= min_val; if (dst_reg->max_value != BPF_REGISTER_MAX_RANGE) dst_reg->max_value -= max_val; + dst_reg->min_align = min(src_align, dst_align); break; case BPF_MUL: if (dst_reg->min_value != BPF_REGISTER_MIN_RANGE) dst_reg->min_value *= min_val; if (dst_reg->max_value != BPF_REGISTER_MAX_RANGE) dst_reg->max_value *= max_val; + dst_reg->min_align = max(src_align, dst_align); break; case BPF_AND: /* Disallow AND'ing of negative numbers, ain't nobody got time @@ -1590,17 +1902,23 @@ static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, else dst_reg->min_value = 0; dst_reg->max_value = max_val; + dst_reg->min_align = max(src_align, dst_align); break; case BPF_LSH: /* Gotta have special overflow logic here, if we're shifting * more than MAX_RANGE then just assume we have an invalid * range. */ - if (min_val > ilog2(BPF_REGISTER_MAX_RANGE)) + if (min_val > ilog2(BPF_REGISTER_MAX_RANGE)) { dst_reg->min_value = BPF_REGISTER_MIN_RANGE; - else if (dst_reg->min_value != BPF_REGISTER_MIN_RANGE) - dst_reg->min_value <<= min_val; - + dst_reg->min_align = 1; + } else { + if (dst_reg->min_value != BPF_REGISTER_MIN_RANGE) + dst_reg->min_value <<= min_val; + if (!dst_reg->min_align) + dst_reg->min_align = 1; + dst_reg->min_align <<= min_val; + } if (max_val > ilog2(BPF_REGISTER_MAX_RANGE)) dst_reg->max_value = BPF_REGISTER_MAX_RANGE; else if (dst_reg->max_value != BPF_REGISTER_MAX_RANGE) @@ -1610,11 +1928,19 @@ static void adjust_reg_min_max_vals(struct bpf_verifier_env *env, /* RSH by a negative number is undefined, and the BPF_RSH is an * unsigned shift, so make the appropriate casts. */ - if (min_val < 0 || dst_reg->min_value < 0) + if (min_val < 0 || dst_reg->min_value < 0) { dst_reg->min_value = BPF_REGISTER_MIN_RANGE; - else + } else { dst_reg->min_value = (u64)(dst_reg->min_value) >> min_val; + } + if (min_val < 0) { + dst_reg->min_align = 1; + } else { + dst_reg->min_align >>= (u64) min_val; + if (!dst_reg->min_align) + dst_reg->min_align = 1; + } if (dst_reg->max_value != BPF_REGISTER_MAX_RANGE) dst_reg->max_value >>= max_val; break; @@ -1714,8 +2040,11 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn) */ regs[insn->dst_reg].type = CONST_IMM; regs[insn->dst_reg].imm = insn->imm; + regs[insn->dst_reg].id = 0; regs[insn->dst_reg].max_value = insn->imm; regs[insn->dst_reg].min_value = insn->imm; + regs[insn->dst_reg].min_align = calc_align(insn->imm); + regs[insn->dst_reg].value_from_signed = false; } } else if (opcode > BPF_END) { @@ -1779,6 +2108,17 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn) return 0; } else if (opcode == BPF_ADD && BPF_CLASS(insn->code) == BPF_ALU64 && + dst_reg->type == PTR_TO_STACK && + ((BPF_SRC(insn->code) == BPF_X && + regs[insn->src_reg].type == CONST_IMM) || + BPF_SRC(insn->code) == BPF_K)) { + if (BPF_SRC(insn->code) == BPF_X) + dst_reg->imm += regs[insn->src_reg].imm; + else + dst_reg->imm += insn->imm; + return 0; + } else if (opcode == BPF_ADD && + BPF_CLASS(insn->code) == BPF_ALU64 && (dst_reg->type == PTR_TO_PACKET || (BPF_SRC(insn->code) == BPF_X && regs[insn->src_reg].type == PTR_TO_PACKET))) { @@ -1811,6 +2151,7 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn) * register as unknown. */ if (env->allow_ptr_leaks && + BPF_CLASS(insn->code) == BPF_ALU64 && opcode == BPF_ADD && (dst_reg->type == PTR_TO_MAP_VALUE || dst_reg->type == PTR_TO_MAP_VALUE_ADJ)) dst_reg->type = PTR_TO_MAP_VALUE_ADJ; @@ -1859,14 +2200,15 @@ static void find_good_pkt_pointers(struct bpf_verifier_state *state, for (i = 0; i < MAX_BPF_REG; i++) if (regs[i].type == PTR_TO_PACKET && regs[i].id == dst_reg->id) - regs[i].range = dst_reg->off; + /* keep the maximum range already checked */ + regs[i].range = max(regs[i].range, dst_reg->off); for (i = 0; i < MAX_BPF_STACK; i += BPF_REG_SIZE) { if (state->stack_slot_type[i] != STACK_SPILL) continue; reg = &state->spilled_regs[i / BPF_REG_SIZE]; if (reg->type == PTR_TO_PACKET && reg->id == dst_reg->id) - reg->range = dst_reg->off; + reg->range = max(reg->range, dst_reg->off); } } @@ -1878,38 +2220,63 @@ static void reg_set_min_max(struct bpf_reg_state *true_reg, struct bpf_reg_state *false_reg, u64 val, u8 opcode) { + bool value_from_signed = true; + bool is_range = true; + switch (opcode) { case BPF_JEQ: /* If this is false then we know nothing Jon Snow, but if it is * true then we know for sure. */ true_reg->max_value = true_reg->min_value = val; + is_range = false; break; case BPF_JNE: /* If this is true we know nothing Jon Snow, but if it is false * we know the value for sure; */ false_reg->max_value = false_reg->min_value = val; + is_range = false; break; case BPF_JGT: - /* Unsigned comparison, the minimum value is 0. */ - false_reg->min_value = 0; + value_from_signed = false; + /* fallthrough */ case BPF_JSGT: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGT) { + /* Unsigned comparison, the minimum value is 0. */ + false_reg->min_value = 0; + } /* If this is false then we know the maximum val is val, * otherwise we know the min val is val+1. */ false_reg->max_value = val; + false_reg->value_from_signed = value_from_signed; true_reg->min_value = val + 1; + true_reg->value_from_signed = value_from_signed; break; case BPF_JGE: - /* Unsigned comparison, the minimum value is 0. */ - false_reg->min_value = 0; + value_from_signed = false; + /* fallthrough */ case BPF_JSGE: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGE) { + /* Unsigned comparison, the minimum value is 0. */ + false_reg->min_value = 0; + } /* If this is false then we know the maximum value is val - 1, * otherwise we know the mimimum value is val. */ false_reg->max_value = val - 1; + false_reg->value_from_signed = value_from_signed; true_reg->min_value = val; + true_reg->value_from_signed = value_from_signed; break; default: break; @@ -1917,6 +2284,12 @@ static void reg_set_min_max(struct bpf_reg_state *true_reg, check_reg_overflow(false_reg); check_reg_overflow(true_reg); + if (is_range) { + if (__is_pointer_value(false, false_reg)) + reset_reg_range_values(false_reg, 0); + if (__is_pointer_value(false, true_reg)) + reset_reg_range_values(true_reg, 0); + } } /* Same as above, but for the case that dst_reg is a CONST_IMM reg and src_reg @@ -1926,39 +2299,64 @@ static void reg_set_min_max_inv(struct bpf_reg_state *true_reg, struct bpf_reg_state *false_reg, u64 val, u8 opcode) { + bool value_from_signed = true; + bool is_range = true; + switch (opcode) { case BPF_JEQ: /* If this is false then we know nothing Jon Snow, but if it is * true then we know for sure. */ true_reg->max_value = true_reg->min_value = val; + is_range = false; break; case BPF_JNE: /* If this is true we know nothing Jon Snow, but if it is false * we know the value for sure; */ false_reg->max_value = false_reg->min_value = val; + is_range = false; break; case BPF_JGT: - /* Unsigned comparison, the minimum value is 0. */ - true_reg->min_value = 0; + value_from_signed = false; + /* fallthrough */ case BPF_JSGT: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGT) { + /* Unsigned comparison, the minimum value is 0. */ + true_reg->min_value = 0; + } /* * If this is false, then the val is <= the register, if it is * true the register <= to the val. */ false_reg->min_value = val; + false_reg->value_from_signed = value_from_signed; true_reg->max_value = val - 1; + true_reg->value_from_signed = value_from_signed; break; case BPF_JGE: - /* Unsigned comparison, the minimum value is 0. */ - true_reg->min_value = 0; + value_from_signed = false; + /* fallthrough */ case BPF_JSGE: + if (true_reg->value_from_signed != value_from_signed) + reset_reg_range_values(true_reg, 0); + if (false_reg->value_from_signed != value_from_signed) + reset_reg_range_values(false_reg, 0); + if (opcode == BPF_JGE) { + /* Unsigned comparison, the minimum value is 0. */ + true_reg->min_value = 0; + } /* If this is false then constant < register, if it is true then * the register < constant. */ false_reg->min_value = val + 1; + false_reg->value_from_signed = value_from_signed; true_reg->max_value = val; + true_reg->value_from_signed = value_from_signed; break; default: break; @@ -1966,6 +2364,12 @@ static void reg_set_min_max_inv(struct bpf_reg_state *true_reg, check_reg_overflow(false_reg); check_reg_overflow(true_reg); + if (is_range) { + if (__is_pointer_value(false, false_reg)) + reset_reg_range_values(false_reg, 0); + if (__is_pointer_value(false, true_reg)) + reset_reg_range_values(true_reg, 0); + } } static void mark_map_reg(struct bpf_reg_state *regs, u32 regno, u32 id, @@ -1974,14 +2378,19 @@ static void mark_map_reg(struct bpf_reg_state *regs, u32 regno, u32 id, struct bpf_reg_state *reg = ®s[regno]; if (reg->type == PTR_TO_MAP_VALUE_OR_NULL && reg->id == id) { - reg->type = type; + if (type == UNKNOWN_VALUE) { + __mark_reg_unknown_value(regs, regno); + } else if (reg->map_ptr->inner_map_meta) { + reg->type = CONST_PTR_TO_MAP; + reg->map_ptr = reg->map_ptr->inner_map_meta; + } else { + reg->type = type; + } /* We don't need id from this point onwards anymore, thus we * should better reset it, so that state pruning has chances * to take effect. */ reg->id = 0; - if (type == UNKNOWN_VALUE) - __mark_reg_unknown_value(regs, regno); } } @@ -2144,16 +2553,11 @@ static int check_ld_imm(struct bpf_verifier_env *env, struct bpf_insn *insn) return err; if (insn->src_reg == 0) { - /* generic move 64-bit immediate into a register, - * only analyzer needs to collect the ld_imm value. - */ u64 imm = ((u64)(insn + 1)->imm << 32) | (u32)insn->imm; - if (!env->analyzer_ops) - return 0; - regs[insn->dst_reg].type = CONST_IMM; regs[insn->dst_reg].imm = imm; + regs[insn->dst_reg].id = 0; return 0; } @@ -2196,7 +2600,6 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn) { struct bpf_reg_state *regs = env->cur_state.regs; u8 mode = BPF_MODE(insn->code); - struct bpf_reg_state *reg; int i, err; if (!may_access_skb(env->prog->type)) { @@ -2229,11 +2632,8 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn) } /* reset caller saved regs to unreadable */ - for (i = 0; i < CALLER_SAVED_REGS; i++) { - reg = regs + caller_saved[i]; - reg->type = NOT_INIT; - reg->imm = 0; - } + for (i = 0; i < CALLER_SAVED_REGS; i++) + mark_reg_not_init(regs, caller_saved[i]); /* mark destination R0 register as readable, since it contains * the value fetched from the packet @@ -2392,6 +2792,7 @@ peek_stack: env->explored_states[t + 1] = STATE_LIST_MARK; } else { /* conditional jump with two edges */ + env->explored_states[t] = STATE_LIST_MARK; ret = push_insn(t, t + 1, FALLTHROUGH, env); if (ret == 1) goto peek_stack; @@ -2443,7 +2844,8 @@ err_free: /* the following conditions reduce the number of explored insns * from ~140k to ~80k for ultra large programs that use a lot of ptr_to_packet */ -static bool compare_ptrs_to_packet(struct bpf_reg_state *old, +static bool compare_ptrs_to_packet(struct bpf_verifier_env *env, + struct bpf_reg_state *old, struct bpf_reg_state *cur) { if (old->id != cur->id) @@ -2486,7 +2888,7 @@ static bool compare_ptrs_to_packet(struct bpf_reg_state *old, * 'if (R4 > data_end)' and all further insn were already good with r=20, * so they will be good with r=30 and we can prune the search. */ - if (old->off <= cur->off && + if (!env->strict_alignment && old->off <= cur->off && old->off >= old->range && cur->off >= cur->range) return true; @@ -2550,8 +2952,14 @@ static bool states_equal(struct bpf_verifier_env *env, rcur->type != NOT_INIT)) continue; + /* Don't care about the reg->id in this case. */ + if (rold->type == PTR_TO_MAP_VALUE_OR_NULL && + rcur->type == PTR_TO_MAP_VALUE_OR_NULL && + rold->map_ptr == rcur->map_ptr) + continue; + if (rold->type == PTR_TO_PACKET && rcur->type == PTR_TO_PACKET && - compare_ptrs_to_packet(rold, rcur)) + compare_ptrs_to_packet(env, rold, rcur)) continue; return false; @@ -2569,6 +2977,8 @@ static bool states_equal(struct bpf_verifier_env *env, return false; if (i % BPF_REG_SIZE) continue; + if (old->stack_slot_type[i] != STACK_SPILL) + continue; if (memcmp(&old->spilled_regs[i / BPF_REG_SIZE], &cur->spilled_regs[i / BPF_REG_SIZE], sizeof(old->spilled_regs[0]))) @@ -2664,7 +3074,7 @@ static int do_check(struct bpf_verifier_env *env) class = BPF_CLASS(insn->code); if (++insn_processed > BPF_COMPLEXITY_LIMIT_INSNS) { - verbose("BPF program is too large. Proccessed %d insn\n", + verbose("BPF program is too large. Processed %d insn\n", insn_processed); return -E2BIG; } @@ -2684,15 +3094,22 @@ static int do_check(struct bpf_verifier_env *env) goto process_bpf_exit; } - if (log_level && do_print_state) { - verbose("\nfrom %d to %d:", prev_insn_idx, insn_idx); + if (need_resched()) + cond_resched(); + + if (log_level > 1 || (log_level && do_print_state)) { + if (log_level > 1) + verbose("%d:", insn_idx); + else + verbose("\nfrom %d to %d:", + prev_insn_idx, insn_idx); print_verifier_state(&env->cur_state); do_print_state = false; } if (log_level) { verbose("%d: ", insn_idx); - print_bpf_insn(insn); + print_bpf_insn(env, insn); } err = ext_analyzer_insn_hook(env, insn_idx, prev_insn_idx); @@ -2723,19 +3140,12 @@ static int do_check(struct bpf_verifier_env *env) /* check that memory (src_reg + off) is readable, * the state of dst_reg will be updated by this func */ - err = check_mem_access(env, insn->src_reg, insn->off, + err = check_mem_access(env, insn_idx, insn->src_reg, insn->off, BPF_SIZE(insn->code), BPF_READ, insn->dst_reg); if (err) return err; - reset_reg_range_values(regs, insn->dst_reg); - if (BPF_SIZE(insn->code) != BPF_W && - BPF_SIZE(insn->code) != BPF_DW) { - insn_idx++; - continue; - } - prev_src_type = &env->insn_aux_data[insn_idx].ptr_type; if (*prev_src_type == NOT_INIT) { @@ -2763,7 +3173,7 @@ static int do_check(struct bpf_verifier_env *env) enum bpf_reg_type *prev_dst_type, dst_reg_type; if (BPF_MODE(insn->code) == BPF_XADD) { - err = check_xadd(env, insn); + err = check_xadd(env, insn_idx, insn); if (err) return err; insn_idx++; @@ -2782,7 +3192,7 @@ static int do_check(struct bpf_verifier_env *env) dst_reg_type = regs[insn->dst_reg].type; /* check that memory (dst_reg + off) is writeable */ - err = check_mem_access(env, insn->dst_reg, insn->off, + err = check_mem_access(env, insn_idx, insn->dst_reg, insn->off, BPF_SIZE(insn->code), BPF_WRITE, insn->src_reg); if (err) @@ -2811,7 +3221,7 @@ static int do_check(struct bpf_verifier_env *env) return err; /* check that memory (dst_reg + off) is writeable */ - err = check_mem_access(env, insn->dst_reg, insn->off, + err = check_mem_access(env, insn_idx, insn->dst_reg, insn->off, BPF_SIZE(insn->code), BPF_WRITE, -1); if (err) @@ -2829,7 +3239,7 @@ static int do_check(struct bpf_verifier_env *env) return -EINVAL; } - err = check_call(env, insn->imm); + err = check_call(env, insn->imm, insn_idx); if (err) return err; @@ -2909,20 +3319,38 @@ process_bpf_exit: insn_idx++; } - verbose("processed %d insns\n", insn_processed); + verbose("processed %d insns, stack depth %d\n", + insn_processed, env->prog->aux->stack_depth); return 0; } +static int check_map_prealloc(struct bpf_map *map) +{ + return (map->map_type != BPF_MAP_TYPE_HASH && + map->map_type != BPF_MAP_TYPE_PERCPU_HASH && + map->map_type != BPF_MAP_TYPE_HASH_OF_MAPS) || + !(map->map_flags & BPF_F_NO_PREALLOC); +} + static int check_map_prog_compatibility(struct bpf_map *map, struct bpf_prog *prog) { - if (prog->type == BPF_PROG_TYPE_PERF_EVENT && - (map->map_type == BPF_MAP_TYPE_HASH || - map->map_type == BPF_MAP_TYPE_PERCPU_HASH) && - (map->map_flags & BPF_F_NO_PREALLOC)) { - verbose("perf_event programs can only use preallocated hash map\n"); - return -EINVAL; + /* Make sure that BPF_PROG_TYPE_PERF_EVENT programs only use + * preallocated hash maps, since doing memory allocation + * in overflow_handler can crash depending on where nmi got + * triggered. + */ + if (prog->type == BPF_PROG_TYPE_PERF_EVENT) { + if (!check_map_prealloc(map)) { + verbose("perf_event programs can only use preallocated hash map\n"); + return -EINVAL; + } + if (map->inner_map_meta && + !check_map_prealloc(map->inner_map_meta)) { + verbose("perf_event programs can only use preallocated inner hash map\n"); + return -EINVAL; + } } return 0; } @@ -3051,17 +3479,54 @@ static void convert_pseudo_ld_imm64(struct bpf_verifier_env *env) insn->src_reg = 0; } +/* single env->prog->insni[off] instruction was replaced with the range + * insni[off, off + cnt). Adjust corresponding insn_aux_data by copying + * [0, off) and [off, end) to new locations, so the patched range stays zero + */ +static int adjust_insn_aux_data(struct bpf_verifier_env *env, u32 prog_len, + u32 off, u32 cnt) +{ + struct bpf_insn_aux_data *new_data, *old_data = env->insn_aux_data; + + if (cnt == 1) + return 0; + new_data = vzalloc(sizeof(struct bpf_insn_aux_data) * prog_len); + if (!new_data) + return -ENOMEM; + memcpy(new_data, old_data, sizeof(struct bpf_insn_aux_data) * off); + memcpy(new_data + off + cnt - 1, old_data + off, + sizeof(struct bpf_insn_aux_data) * (prog_len - off - cnt + 1)); + env->insn_aux_data = new_data; + vfree(old_data); + return 0; +} + +static struct bpf_prog *bpf_patch_insn_data(struct bpf_verifier_env *env, u32 off, + const struct bpf_insn *patch, u32 len) +{ + struct bpf_prog *new_prog; + + new_prog = bpf_patch_insn_single(env->prog, off, patch, len); + if (!new_prog) + return NULL; + if (adjust_insn_aux_data(env, new_prog->len, off, len)) + return NULL; + return new_prog; +} + /* convert load instructions that access fields of 'struct __sk_buff' * into sequence of instructions that access fields of 'struct sk_buff' */ static int convert_ctx_accesses(struct bpf_verifier_env *env) { const struct bpf_verifier_ops *ops = env->prog->aux->ops; + int i, cnt, size, ctx_field_size, delta = 0; const int insn_cnt = env->prog->len; struct bpf_insn insn_buf[16], *insn; struct bpf_prog *new_prog; enum bpf_access_type type; - int i, cnt, delta = 0; + bool is_narrower_load; + u32 target_size; if (ops->gen_prologue) { cnt = ops->gen_prologue(insn_buf, env->seen_direct_write, @@ -3070,10 +3535,10 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) verbose("bpf verifier is misconfigured\n"); return -EINVAL; } else if (cnt) { - new_prog = bpf_patch_insn_single(env->prog, 0, - insn_buf, cnt); + new_prog = bpf_patch_insn_data(env, 0, insn_buf, cnt); if (!new_prog) return -ENOMEM; + env->prog = new_prog; delta += cnt - 1; } @@ -3085,27 +3550,69 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) insn = env->prog->insnsi + delta; for (i = 0; i < insn_cnt; i++, insn++) { - if (insn->code == (BPF_LDX | BPF_MEM | BPF_W) || + if (insn->code == (BPF_LDX | BPF_MEM | BPF_B) || + insn->code == (BPF_LDX | BPF_MEM | BPF_H) || + insn->code == (BPF_LDX | BPF_MEM | BPF_W) || insn->code == (BPF_LDX | BPF_MEM | BPF_DW)) type = BPF_READ; - else if (insn->code == (BPF_STX | BPF_MEM | BPF_W) || + else if (insn->code == (BPF_STX | BPF_MEM | BPF_B) || + insn->code == (BPF_STX | BPF_MEM | BPF_H) || + insn->code == (BPF_STX | BPF_MEM | BPF_W) || insn->code == (BPF_STX | BPF_MEM | BPF_DW)) type = BPF_WRITE; else continue; - if (env->insn_aux_data[i].ptr_type != PTR_TO_CTX) + if (env->insn_aux_data[i + delta].ptr_type != PTR_TO_CTX) continue; - cnt = ops->convert_ctx_access(type, insn->dst_reg, insn->src_reg, - insn->off, insn_buf, env->prog); - if (cnt == 0 || cnt >= ARRAY_SIZE(insn_buf)) { + ctx_field_size = env->insn_aux_data[i + delta].ctx_field_size; + size = BPF_LDST_BYTES(insn); + + /* If the read access is a narrower load of the field, + * convert to a 4/8-byte load, to minimum program type specific + * convert_ctx_access changes. If conversion is successful, + * we will apply proper mask to the result. + */ + is_narrower_load = size < ctx_field_size; + if (is_narrower_load) { + u32 off = insn->off; + u8 size_code; + + if (type == BPF_WRITE) { + verbose("bpf verifier narrow ctx access misconfigured\n"); + return -EINVAL; + } + + size_code = BPF_H; + if (ctx_field_size == 4) + size_code = BPF_W; + else if (ctx_field_size == 8) + size_code = BPF_DW; + + insn->off = off & ~(ctx_field_size - 1); + insn->code = BPF_LDX | BPF_MEM | size_code; + } + + target_size = 0; + cnt = ops->convert_ctx_access(type, insn, insn_buf, env->prog, + &target_size); + if (cnt == 0 || cnt >= ARRAY_SIZE(insn_buf) || + (ctx_field_size && !target_size)) { verbose("bpf verifier is misconfigured\n"); return -EINVAL; } - new_prog = bpf_patch_insn_single(env->prog, i + delta, insn_buf, - cnt); + if (is_narrower_load && size < target_size) { + if (ctx_field_size <= 4) + insn_buf[cnt++] = BPF_ALU32_IMM(BPF_AND, insn->dst_reg, + (1 << size * 8) - 1); + else + insn_buf[cnt++] = BPF_ALU64_IMM(BPF_AND, insn->dst_reg, + (1 << size * 8) - 1); + } + + new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, cnt); if (!new_prog) return -ENOMEM; @@ -3119,6 +3626,90 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) return 0; } +/* fixup insn->imm field of bpf_call instructions + * and inline eligible helpers as explicit sequence of BPF instructions + * + * this function is called after eBPF program passed verification + */ +static int fixup_bpf_calls(struct bpf_verifier_env *env) +{ + struct bpf_prog *prog = env->prog; + struct bpf_insn *insn = prog->insnsi; + const struct bpf_func_proto *fn; + const int insn_cnt = prog->len; + struct bpf_insn insn_buf[16]; + struct bpf_prog *new_prog; + struct bpf_map *map_ptr; + int i, cnt, delta = 0; + + for (i = 0; i < insn_cnt; i++, insn++) { + if (insn->code != (BPF_JMP | BPF_CALL)) + continue; + + if (insn->imm == BPF_FUNC_get_route_realm) + prog->dst_needed = 1; + if (insn->imm == BPF_FUNC_get_prandom_u32) + bpf_user_rnd_init_once(); + if (insn->imm == BPF_FUNC_tail_call) { + /* If we tail call into other programs, we + * cannot make any assumptions since they can + * be replaced dynamically during runtime in + * the program array. + */ + prog->cb_access = 1; + env->prog->aux->stack_depth = MAX_BPF_STACK; + + /* mark bpf_tail_call as different opcode to avoid + * conditional branch in the interpeter for every normal + * call and to prevent accidental JITing by JIT compiler + * that doesn't support bpf_tail_call yet + */ + insn->imm = 0; + insn->code = BPF_JMP | BPF_TAIL_CALL; + continue; + } + + if (ebpf_jit_enabled() && insn->imm == BPF_FUNC_map_lookup_elem) { + map_ptr = env->insn_aux_data[i + delta].map_ptr; + if (map_ptr == BPF_MAP_PTR_POISON || + !map_ptr->ops->map_gen_lookup) + goto patch_call_imm; + + cnt = map_ptr->ops->map_gen_lookup(map_ptr, insn_buf); + if (cnt == 0 || cnt >= ARRAY_SIZE(insn_buf)) { + verbose("bpf verifier is misconfigured\n"); + return -EINVAL; + } + + new_prog = bpf_patch_insn_data(env, i + delta, insn_buf, + cnt); + if (!new_prog) + return -ENOMEM; + + delta += cnt - 1; + + /* keep walking new program and skip insns we just inserted */ + env->prog = prog = new_prog; + insn = new_prog->insnsi + i + delta; + continue; + } + +patch_call_imm: + fn = prog->aux->ops->get_func_proto(insn->imm); + /* all functions that have prototype and verifier allowed + * programs to call them, must be real in-kernel functions + */ + if (!fn->func) { + verbose("kernel subsystem misconfigured func %s#%d\n", + func_id_name(insn->imm), insn->imm); + return -EFAULT; + } + insn->imm = fn->func - __bpf_call_base; + } + + return 0; +} + static void free_states(struct bpf_verifier_env *env) { struct bpf_verifier_state_list *sl, *sln; @@ -3187,6 +3778,10 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr) log_level = 0; } + env->strict_alignment = !!(attr->prog_flags & BPF_F_STRICT_ALIGNMENT); + if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) + env->strict_alignment = true; + ret = replace_map_fd_with_map_ptr(env); if (ret < 0) goto skip_full_check; @@ -3214,6 +3809,9 @@ skip_full_check: /* program is valid, convert *(u32*)(ctx + off) accesses */ ret = convert_ctx_accesses(env); + if (ret == 0) + ret = fixup_bpf_calls(env); + if (log_level && log_len >= log_size - 1) { BUG_ON(log_len >= log_size); /* verifier log exceeded user supplied buffer */ @@ -3289,6 +3887,10 @@ int bpf_analyzer(struct bpf_prog *prog, const struct bpf_ext_analyzer_ops *ops, log_level = 0; + env->strict_alignment = false; + if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS)) + env->strict_alignment = true; + env->explored_states = kcalloc(env->prog->len, sizeof(struct bpf_verifier_state_list *), GFP_KERNEL); diff --git a/kernel/cgroup/Makefile b/kernel/cgroup/Makefile new file mode 100644 index 000000000000..ce693ccb8c58 --- /dev/null +++ b/kernel/cgroup/Makefile @@ -0,0 +1,7 @@ +obj-y := cgroup.o namespace.o cgroup-v1.o + +obj-$(CONFIG_CGROUP_FREEZER) += freezer.o +obj-$(CONFIG_CGROUP_PIDS) += pids.o +obj-$(CONFIG_CGROUP_RDMA) += rdma.o +obj-$(CONFIG_CPUSETS) += cpuset.o +obj-$(CONFIG_CGROUP_DEBUG) += debug.o diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h new file mode 100644 index 000000000000..793565c05742 --- /dev/null +++ b/kernel/cgroup/cgroup-internal.h @@ -0,0 +1,217 @@ +#ifndef __CGROUP_INTERNAL_H +#define __CGROUP_INTERNAL_H + +#include <linux/cgroup.h> +#include <linux/kernfs.h> +#include <linux/workqueue.h> +#include <linux/list.h> +#include <linux/refcount.h> + +/* + * A cgroup can be associated with multiple css_sets as different tasks may + * belong to different cgroups on different hierarchies. In the other + * direction, a css_set is naturally associated with multiple cgroups. + * This M:N relationship is represented by the following link structure + * which exists for each association and allows traversing the associations + * from both sides. + */ +struct cgrp_cset_link { + /* the cgroup and css_set this link associates */ + struct cgroup *cgrp; + struct css_set *cset; + + /* list of cgrp_cset_links anchored at cgrp->cset_links */ + struct list_head cset_link; + + /* list of cgrp_cset_links anchored at css_set->cgrp_links */ + struct list_head cgrp_link; +}; + +/* used to track tasks and csets during migration */ +struct cgroup_taskset { + /* the src and dst cset list running through cset->mg_node */ + struct list_head src_csets; + struct list_head dst_csets; + + /* the subsys currently being processed */ + int ssid; + + /* + * Fields for cgroup_taskset_*() iteration. + * + * Before migration is committed, the target migration tasks are on + * ->mg_tasks of the csets on ->src_csets. After, on ->mg_tasks of + * the csets on ->dst_csets. ->csets point to either ->src_csets + * or ->dst_csets depending on whether migration is committed. + * + * ->cur_csets and ->cur_task point to the current task position + * during iteration. + */ + struct list_head *csets; + struct css_set *cur_cset; + struct task_struct *cur_task; +}; + +/* migration context also tracks preloading */ +struct cgroup_mgctx { + /* + * Preloaded source and destination csets. Used to guarantee + * atomic success or failure on actual migration. + */ + struct list_head preloaded_src_csets; + struct list_head preloaded_dst_csets; + + /* tasks and csets to migrate */ + struct cgroup_taskset tset; + + /* subsystems affected by migration */ + u16 ss_mask; +}; + +#define CGROUP_TASKSET_INIT(tset) \ +{ \ + .src_csets = LIST_HEAD_INIT(tset.src_csets), \ + .dst_csets = LIST_HEAD_INIT(tset.dst_csets), \ + .csets = &tset.src_csets, \ +} + +#define CGROUP_MGCTX_INIT(name) \ +{ \ + LIST_HEAD_INIT(name.preloaded_src_csets), \ + LIST_HEAD_INIT(name.preloaded_dst_csets), \ + CGROUP_TASKSET_INIT(name.tset), \ +} + +#define DEFINE_CGROUP_MGCTX(name) \ + struct cgroup_mgctx name = CGROUP_MGCTX_INIT(name) + +struct cgroup_sb_opts { + u16 subsys_mask; + unsigned int flags; + char *release_agent; + bool cpuset_clone_children; + char *name; + /* User explicitly requested empty subsystem */ + bool none; +}; + +extern struct mutex cgroup_mutex; +extern spinlock_t css_set_lock; +extern struct cgroup_subsys *cgroup_subsys[]; +extern struct list_head cgroup_roots; +extern struct file_system_type cgroup_fs_type; + +/* iterate across the hierarchies */ +#define for_each_root(root) \ + list_for_each_entry((root), &cgroup_roots, root_list) + +/** + * for_each_subsys - iterate all enabled cgroup subsystems + * @ss: the iteration cursor + * @ssid: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end + */ +#define for_each_subsys(ss, ssid) \ + for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT && \ + (((ss) = cgroup_subsys[ssid]) || true); (ssid)++) + +static inline bool cgroup_is_dead(const struct cgroup *cgrp) +{ + return !(cgrp->self.flags & CSS_ONLINE); +} + +static inline bool notify_on_release(const struct cgroup *cgrp) +{ + return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); +} + +void put_css_set_locked(struct css_set *cset); + +static inline void put_css_set(struct css_set *cset) +{ + unsigned long flags; + + /* + * Ensure that the refcount doesn't hit zero while any readers + * can see it. Similar to atomic_dec_and_lock(), but for an + * rwlock + */ + if (refcount_dec_not_one(&cset->refcount)) + return; + + spin_lock_irqsave(&css_set_lock, flags); + put_css_set_locked(cset); + spin_unlock_irqrestore(&css_set_lock, flags); +} + +/* + * refcounted get/put for css_set objects + */ +static inline void get_css_set(struct css_set *cset) +{ + refcount_inc(&cset->refcount); +} + +bool cgroup_ssid_enabled(int ssid); +bool cgroup_on_dfl(const struct cgroup *cgrp); + +struct cgroup_root *cgroup_root_from_kf(struct kernfs_root *kf_root); +struct cgroup *task_cgroup_from_root(struct task_struct *task, + struct cgroup_root *root); +struct cgroup *cgroup_kn_lock_live(struct kernfs_node *kn, bool drain_offline); +void cgroup_kn_unlock(struct kernfs_node *kn); +int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen, + struct cgroup_namespace *ns); + +void cgroup_free_root(struct cgroup_root *root); +void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts); +int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags); +int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask); +struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags, + struct cgroup_root *root, unsigned long magic, + struct cgroup_namespace *ns); + +bool cgroup_may_migrate_to(struct cgroup *dst_cgrp); +void cgroup_migrate_finish(struct cgroup_mgctx *mgctx); +void cgroup_migrate_add_src(struct css_set *src_cset, struct cgroup *dst_cgrp, + struct cgroup_mgctx *mgctx); +int cgroup_migrate_prepare_dst(struct cgroup_mgctx *mgctx); +int cgroup_migrate(struct task_struct *leader, bool threadgroup, + struct cgroup_mgctx *mgctx); + +int cgroup_attach_task(struct cgroup *dst_cgrp, struct task_struct *leader, + bool threadgroup); +ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, + size_t nbytes, loff_t off, bool threadgroup); +ssize_t cgroup_procs_write(struct kernfs_open_file *of, char *buf, size_t nbytes, + loff_t off); + +void cgroup_lock_and_drain_offline(struct cgroup *cgrp); + +int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, umode_t mode); +int cgroup_rmdir(struct kernfs_node *kn); +int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node, + struct kernfs_root *kf_root); + +int cgroup_task_count(const struct cgroup *cgrp); + +/* + * namespace.c + */ +extern const struct proc_ns_operations cgroupns_operations; + +/* + * cgroup-v1.c + */ +extern struct cftype cgroup1_base_files[]; +extern const struct file_operations proc_cgroupstats_operations; +extern struct kernfs_syscall_ops cgroup1_kf_syscall_ops; + +bool cgroup1_ssid_disabled(int ssid); +void cgroup1_pidlist_destroy_all(struct cgroup *cgrp); +void cgroup1_release_agent(struct work_struct *work); +void cgroup1_check_for_release(struct cgroup *cgrp); +struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags, + void *data, unsigned long magic, + struct cgroup_namespace *ns); + +#endif /* __CGROUP_INTERNAL_H */ diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c new file mode 100644 index 000000000000..7bf4b1533f34 --- /dev/null +++ b/kernel/cgroup/cgroup-v1.c @@ -0,0 +1,1261 @@ +#include "cgroup-internal.h" + +#include <linux/ctype.h> +#include <linux/kmod.h> +#include <linux/sort.h> +#include <linux/delay.h> +#include <linux/mm.h> +#include <linux/sched/signal.h> +#include <linux/sched/task.h> +#include <linux/magic.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <linux/delayacct.h> +#include <linux/pid_namespace.h> +#include <linux/cgroupstats.h> + +#include <trace/events/cgroup.h> + +/* + * pidlists linger the following amount before being destroyed. The goal + * is avoiding frequent destruction in the middle of consecutive read calls + * Expiring in the middle is a performance problem not a correctness one. + * 1 sec should be enough. + */ +#define CGROUP_PIDLIST_DESTROY_DELAY HZ + +/* Controllers blocked by the commandline in v1 */ +static u16 cgroup_no_v1_mask; + +/* + * pidlist destructions need to be flushed on cgroup destruction. Use a + * separate workqueue as flush domain. + */ +static struct workqueue_struct *cgroup_pidlist_destroy_wq; + +/* + * Protects cgroup_subsys->release_agent_path. Modifying it also requires + * cgroup_mutex. Reading requires either cgroup_mutex or this spinlock. + */ +static DEFINE_SPINLOCK(release_agent_path_lock); + +bool cgroup1_ssid_disabled(int ssid) +{ + return cgroup_no_v1_mask & (1 << ssid); +} + +/** + * cgroup_attach_task_all - attach task 'tsk' to all cgroups of task 'from' + * @from: attach to all cgroups of a given task + * @tsk: the task to be attached + */ +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) +{ + struct cgroup_root *root; + int retval = 0; + + mutex_lock(&cgroup_mutex); + percpu_down_write(&cgroup_threadgroup_rwsem); + for_each_root(root) { + struct cgroup *from_cgrp; + + if (root == &cgrp_dfl_root) + continue; + + spin_lock_irq(&css_set_lock); + from_cgrp = task_cgroup_from_root(from, root); + spin_unlock_irq(&css_set_lock); + + retval = cgroup_attach_task(from_cgrp, tsk, false); + if (retval) + break; + } + percpu_up_write(&cgroup_threadgroup_rwsem); + mutex_unlock(&cgroup_mutex); + + return retval; +} +EXPORT_SYMBOL_GPL(cgroup_attach_task_all); + +/** + * cgroup_trasnsfer_tasks - move tasks from one cgroup to another + * @to: cgroup to which the tasks will be moved + * @from: cgroup in which the tasks currently reside + * + * Locking rules between cgroup_post_fork() and the migration path + * guarantee that, if a task is forking while being migrated, the new child + * is guaranteed to be either visible in the source cgroup after the + * parent's migration is complete or put into the target cgroup. No task + * can slip out of migration through forking. + */ +int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from) +{ + DEFINE_CGROUP_MGCTX(mgctx); + struct cgrp_cset_link *link; + struct css_task_iter it; + struct task_struct *task; + int ret; + + if (cgroup_on_dfl(to)) + return -EINVAL; + + if (!cgroup_may_migrate_to(to)) + return -EBUSY; + + mutex_lock(&cgroup_mutex); + + percpu_down_write(&cgroup_threadgroup_rwsem); + + /* all tasks in @from are being moved, all csets are source */ + spin_lock_irq(&css_set_lock); + list_for_each_entry(link, &from->cset_links, cset_link) + cgroup_migrate_add_src(link->cset, to, &mgctx); + spin_unlock_irq(&css_set_lock); + + ret = cgroup_migrate_prepare_dst(&mgctx); + if (ret) + goto out_err; + + /* + * Migrate tasks one-by-one until @from is empty. This fails iff + * ->can_attach() fails. + */ + do { + css_task_iter_start(&from->self, &it); + task = css_task_iter_next(&it); + if (task) + get_task_struct(task); + css_task_iter_end(&it); + + if (task) { + ret = cgroup_migrate(task, false, &mgctx); + if (!ret) + trace_cgroup_transfer_tasks(to, task, false); + put_task_struct(task); + } + } while (task && !ret); +out_err: + cgroup_migrate_finish(&mgctx); + percpu_up_write(&cgroup_threadgroup_rwsem); + mutex_unlock(&cgroup_mutex); + return ret; +} + +/* + * Stuff for reading the 'tasks'/'procs' files. + * + * Reading this file can return large amounts of data if a cgroup has + * *lots* of attached tasks. So it may need several calls to read(), + * but we cannot guarantee that the information we produce is correct + * unless we produce it entirely atomically. + * + */ + +/* which pidlist file are we talking about? */ +enum cgroup_filetype { + CGROUP_FILE_PROCS, + CGROUP_FILE_TASKS, +}; + +/* + * A pidlist is a list of pids that virtually represents the contents of one + * of the cgroup files ("procs" or "tasks"). We keep a list of such pidlists, + * a pair (one each for procs, tasks) for each pid namespace that's relevant + * to the cgroup. + */ +struct cgroup_pidlist { + /* + * used to find which pidlist is wanted. doesn't change as long as + * this particular list stays in the list. + */ + struct { enum cgroup_filetype type; struct pid_namespace *ns; } key; + /* array of xids */ + pid_t *list; + /* how many elements the above list has */ + int length; + /* each of these stored in a list by its cgroup */ + struct list_head links; + /* pointer to the cgroup we belong to, for list removal purposes */ + struct cgroup *owner; + /* for delayed destruction */ + struct delayed_work destroy_dwork; +}; + +/* + * The following two functions "fix" the issue where there are more pids + * than kmalloc will give memory for; in such cases, we use vmalloc/vfree. + * TODO: replace with a kernel-wide solution to this problem + */ +#define PIDLIST_TOO_LARGE(c) ((c) * sizeof(pid_t) > (PAGE_SIZE * 2)) +static void *pidlist_allocate(int count) +{ + if (PIDLIST_TOO_LARGE(count)) + return vmalloc(count * sizeof(pid_t)); + else + return kmalloc(count * sizeof(pid_t), GFP_KERNEL); +} + +static void pidlist_free(void *p) +{ + kvfree(p); +} + +/* + * Used to destroy all pidlists lingering waiting for destroy timer. None + * should be left afterwards. + */ +void cgroup1_pidlist_destroy_all(struct cgroup *cgrp) +{ + struct cgroup_pidlist *l, *tmp_l; + + mutex_lock(&cgrp->pidlist_mutex); + list_for_each_entry_safe(l, tmp_l, &cgrp->pidlists, links) + mod_delayed_work(cgroup_pidlist_destroy_wq, &l->destroy_dwork, 0); + mutex_unlock(&cgrp->pidlist_mutex); + + flush_workqueue(cgroup_pidlist_destroy_wq); + BUG_ON(!list_empty(&cgrp->pidlists)); +} + +static void cgroup_pidlist_destroy_work_fn(struct work_struct *work) +{ + struct delayed_work *dwork = to_delayed_work(work); + struct cgroup_pidlist *l = container_of(dwork, struct cgroup_pidlist, + destroy_dwork); + struct cgroup_pidlist *tofree = NULL; + + mutex_lock(&l->owner->pidlist_mutex); + + /* + * Destroy iff we didn't get queued again. The state won't change + * as destroy_dwork can only be queued while locked. + */ + if (!delayed_work_pending(dwork)) { + list_del(&l->links); + pidlist_free(l->list); + put_pid_ns(l->key.ns); + tofree = l; + } + + mutex_unlock(&l->owner->pidlist_mutex); + kfree(tofree); +} + +/* + * pidlist_uniq - given a kmalloc()ed list, strip out all duplicate entries + * Returns the number of unique elements. + */ +static int pidlist_uniq(pid_t *list, int length) +{ + int src, dest = 1; + + /* + * we presume the 0th element is unique, so i starts at 1. trivial + * edge cases first; no work needs to be done for either + */ + if (length == 0 || length == 1) + return length; + /* src and dest walk down the list; dest counts unique elements */ + for (src = 1; src < length; src++) { + /* find next unique element */ + while (list[src] == list[src-1]) { + src++; + if (src == length) + goto after; + } + /* dest always points to where the next unique element goes */ + list[dest] = list[src]; + dest++; + } +after: + return dest; +} + +/* + * The two pid files - task and cgroup.procs - guaranteed that the result + * is sorted, which forced this whole pidlist fiasco. As pid order is + * different per namespace, each namespace needs differently sorted list, + * making it impossible to use, for example, single rbtree of member tasks + * sorted by task pointer. As pidlists can be fairly large, allocating one + * per open file is dangerous, so cgroup had to implement shared pool of + * pidlists keyed by cgroup and namespace. + */ +static int cmppid(const void *a, const void *b) +{ + return *(pid_t *)a - *(pid_t *)b; +} + +static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp, + enum cgroup_filetype type) +{ + struct cgroup_pidlist *l; + /* don't need task_nsproxy() if we're looking at ourself */ + struct pid_namespace *ns = task_active_pid_ns(current); + + lockdep_assert_held(&cgrp->pidlist_mutex); + + list_for_each_entry(l, &cgrp->pidlists, links) + if (l->key.type == type && l->key.ns == ns) + return l; + return NULL; +} + +/* + * find the appropriate pidlist for our purpose (given procs vs tasks) + * returns with the lock on that pidlist already held, and takes care + * of the use count, or returns NULL with no locks held if we're out of + * memory. + */ +static struct cgroup_pidlist *cgroup_pidlist_find_create(struct cgroup *cgrp, + enum cgroup_filetype type) +{ + struct cgroup_pidlist *l; + + lockdep_assert_held(&cgrp->pidlist_mutex); + + l = cgroup_pidlist_find(cgrp, type); + if (l) + return l; + + /* entry not found; create a new one */ + l = kzalloc(sizeof(struct cgroup_pidlist), GFP_KERNEL); + if (!l) + return l; + + INIT_DELAYED_WORK(&l->destroy_dwork, cgroup_pidlist_destroy_work_fn); + l->key.type = type; + /* don't need task_nsproxy() if we're looking at ourself */ + l->key.ns = get_pid_ns(task_active_pid_ns(current)); + l->owner = cgrp; + list_add(&l->links, &cgrp->pidlists); + return l; +} + +/** + * cgroup_task_count - count the number of tasks in a cgroup. + * @cgrp: the cgroup in question + */ +int cgroup_task_count(const struct cgroup *cgrp) +{ + int count = 0; + struct cgrp_cset_link *link; + + spin_lock_irq(&css_set_lock); + list_for_each_entry(link, &cgrp->cset_links, cset_link) + count += link->cset->nr_tasks; + spin_unlock_irq(&css_set_lock); + return count; +} + +/* + * Load a cgroup's pidarray with either procs' tgids or tasks' pids + */ +static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, + struct cgroup_pidlist **lp) +{ + pid_t *array; + int length; + int pid, n = 0; /* used for populating the array */ + struct css_task_iter it; + struct task_struct *tsk; + struct cgroup_pidlist *l; + + lockdep_assert_held(&cgrp->pidlist_mutex); + + /* + * If cgroup gets more users after we read count, we won't have + * enough space - tough. This race is indistinguishable to the + * caller from the case that the additional cgroup users didn't + * show up until sometime later on. + */ + length = cgroup_task_count(cgrp); + array = pidlist_allocate(length); + if (!array) + return -ENOMEM; + /* now, populate the array */ + css_task_iter_start(&cgrp->self, &it); + while ((tsk = css_task_iter_next(&it))) { + if (unlikely(n == length)) + break; + /* get tgid or pid for procs or tasks file respectively */ + if (type == CGROUP_FILE_PROCS) + pid = task_tgid_vnr(tsk); + else + pid = task_pid_vnr(tsk); + if (pid > 0) /* make sure to only use valid results */ + array[n++] = pid; + } + css_task_iter_end(&it); + length = n; + /* now sort & (if procs) strip out duplicates */ + sort(array, length, sizeof(pid_t), cmppid, NULL); + if (type == CGROUP_FILE_PROCS) + length = pidlist_uniq(array, length); + + l = cgroup_pidlist_find_create(cgrp, type); + if (!l) { + pidlist_free(array); + return -ENOMEM; + } + + /* store array, freeing old if necessary */ + pidlist_free(l->list); + l->list = array; + l->length = length; + *lp = l; + return 0; +} + +/* + * seq_file methods for the tasks/procs files. The seq_file position is the + * next pid to display; the seq_file iterator is a pointer to the pid + * in the cgroup->l->list array. + */ + +static void *cgroup_pidlist_start(struct seq_file *s, loff_t *pos) +{ + /* + * Initially we receive a position value that corresponds to + * one more than the last pid shown (or 0 on the first call or + * after a seek to the start). Use a binary-search to find the + * next pid to display, if any + */ + struct kernfs_open_file *of = s->private; + struct cgroup *cgrp = seq_css(s)->cgroup; + struct cgroup_pidlist *l; + enum cgroup_filetype type = seq_cft(s)->private; + int index = 0, pid = *pos; + int *iter, ret; + + mutex_lock(&cgrp->pidlist_mutex); + + /* + * !NULL @of->priv indicates that this isn't the first start() + * after open. If the matching pidlist is around, we can use that. + * Look for it. Note that @of->priv can't be used directly. It + * could already have been destroyed. + */ + if (of->priv) + of->priv = cgroup_pidlist_find(cgrp, type); + + /* + * Either this is the first start() after open or the matching + * pidlist has been destroyed inbetween. Create a new one. + */ + if (!of->priv) { + ret = pidlist_array_load(cgrp, type, + (struct cgroup_pidlist **)&of->priv); + if (ret) + return ERR_PTR(ret); + } + l = of->priv; + + if (pid) { + int end = l->length; + + while (index < end) { + int mid = (index + end) / 2; + if (l->list[mid] == pid) { + index = mid; + break; + } else if (l->list[mid] <= pid) + index = mid + 1; + else + end = mid; + } + } + /* If we're off the end of the array, we're done */ + if (index >= l->length) + return NULL; + /* Update the abstract position to be the actual pid that we found */ + iter = l->list + index; + *pos = *iter; + return iter; +} + +static void cgroup_pidlist_stop(struct seq_file *s, void *v) +{ + struct kernfs_open_file *of = s->private; + struct cgroup_pidlist *l = of->priv; + + if (l) + mod_delayed_work(cgroup_pidlist_destroy_wq, &l->destroy_dwork, + CGROUP_PIDLIST_DESTROY_DELAY); + mutex_unlock(&seq_css(s)->cgroup->pidlist_mutex); +} + +static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct kernfs_open_file *of = s->private; + struct cgroup_pidlist *l = of->priv; + pid_t *p = v; + pid_t *end = l->list + l->length; + /* + * Advance to the next pid in the array. If this goes off the + * end, we're done + */ + p++; + if (p >= end) { + return NULL; + } else { + *pos = *p; + return p; + } +} + +static int cgroup_pidlist_show(struct seq_file *s, void *v) +{ + seq_printf(s, "%d\n", *(int *)v); + + return 0; +} + +static ssize_t cgroup_tasks_write(struct kernfs_open_file *of, + char *buf, size_t nbytes, loff_t off) +{ + return __cgroup_procs_write(of, buf, nbytes, off, false); +} + +static ssize_t cgroup_release_agent_write(struct kernfs_open_file *of, + char *buf, size_t nbytes, loff_t off) +{ + struct cgroup *cgrp; + + BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX); + + cgrp = cgroup_kn_lock_live(of->kn, false); + if (!cgrp) + return -ENODEV; + spin_lock(&release_agent_path_lock); + strlcpy(cgrp->root->release_agent_path, strstrip(buf), + sizeof(cgrp->root->release_agent_path)); + spin_unlock(&release_agent_path_lock); + cgroup_kn_unlock(of->kn); + return nbytes; +} + +static int cgroup_release_agent_show(struct seq_file *seq, void *v) +{ + struct cgroup *cgrp = seq_css(seq)->cgroup; + + spin_lock(&release_agent_path_lock); + seq_puts(seq, cgrp->root->release_agent_path); + spin_unlock(&release_agent_path_lock); + seq_putc(seq, '\n'); + return 0; +} + +static int cgroup_sane_behavior_show(struct seq_file *seq, void *v) +{ + seq_puts(seq, "0\n"); + return 0; +} + +static u64 cgroup_read_notify_on_release(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + return notify_on_release(css->cgroup); +} + +static int cgroup_write_notify_on_release(struct cgroup_subsys_state *css, + struct cftype *cft, u64 val) +{ + if (val) + set_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags); + else + clear_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags); + return 0; +} + +static u64 cgroup_clone_children_read(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + return test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); +} + +static int cgroup_clone_children_write(struct cgroup_subsys_state *css, + struct cftype *cft, u64 val) +{ + if (val) + set_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); + else + clear_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); + return 0; +} + +/* cgroup core interface files for the legacy hierarchies */ +struct cftype cgroup1_base_files[] = { + { + .name = "cgroup.procs", + .seq_start = cgroup_pidlist_start, + .seq_next = cgroup_pidlist_next, + .seq_stop = cgroup_pidlist_stop, + .seq_show = cgroup_pidlist_show, + .private = CGROUP_FILE_PROCS, + .write = cgroup_procs_write, + }, + { + .name = "cgroup.clone_children", + .read_u64 = cgroup_clone_children_read, + .write_u64 = cgroup_clone_children_write, + }, + { + .name = "cgroup.sane_behavior", + .flags = CFTYPE_ONLY_ON_ROOT, + .seq_show = cgroup_sane_behavior_show, + }, + { + .name = "tasks", + .seq_start = cgroup_pidlist_start, + .seq_next = cgroup_pidlist_next, + .seq_stop = cgroup_pidlist_stop, + .seq_show = cgroup_pidlist_show, + .private = CGROUP_FILE_TASKS, + .write = cgroup_tasks_write, + }, + { + .name = "notify_on_release", + .read_u64 = cgroup_read_notify_on_release, + .write_u64 = cgroup_write_notify_on_release, + }, + { + .name = "release_agent", + .flags = CFTYPE_ONLY_ON_ROOT, + .seq_show = cgroup_release_agent_show, + .write = cgroup_release_agent_write, + .max_write_len = PATH_MAX - 1, + }, + { } /* terminate */ +}; + +/* Display information about each subsystem and each hierarchy */ +static int proc_cgroupstats_show(struct seq_file *m, void *v) +{ + struct cgroup_subsys *ss; + int i; + + seq_puts(m, "#subsys_name\thierarchy\tnum_cgroups\tenabled\n"); + /* + * ideally we don't want subsystems moving around while we do this. + * cgroup_mutex is also necessary to guarantee an atomic snapshot of + * subsys/hierarchy state. + */ + mutex_lock(&cgroup_mutex); + + for_each_subsys(ss, i) + seq_printf(m, "%s\t%d\t%d\t%d\n", + ss->legacy_name, ss->root->hierarchy_id, + atomic_read(&ss->root->nr_cgrps), + cgroup_ssid_enabled(i)); + + mutex_unlock(&cgroup_mutex); + return 0; +} + +static int cgroupstats_open(struct inode *inode, struct file *file) +{ + return single_open(file, proc_cgroupstats_show, NULL); +} + +const struct file_operations proc_cgroupstats_operations = { + .open = cgroupstats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/** + * cgroupstats_build - build and fill cgroupstats + * @stats: cgroupstats to fill information into + * @dentry: A dentry entry belonging to the cgroup for which stats have + * been requested. + * + * Build and fill cgroupstats so that taskstats can export it to user + * space. + */ +int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) +{ + struct kernfs_node *kn = kernfs_node_from_dentry(dentry); + struct cgroup *cgrp; + struct css_task_iter it; + struct task_struct *tsk; + + /* it should be kernfs_node belonging to cgroupfs and is a directory */ + if (dentry->d_sb->s_type != &cgroup_fs_type || !kn || + kernfs_type(kn) != KERNFS_DIR) + return -EINVAL; + + mutex_lock(&cgroup_mutex); + + /* + * We aren't being called from kernfs and there's no guarantee on + * @kn->priv's validity. For this and css_tryget_online_from_dir(), + * @kn->priv is RCU safe. Let's do the RCU dancing. + */ + rcu_read_lock(); + cgrp = rcu_dereference(*(void __rcu __force **)&kn->priv); + if (!cgrp || cgroup_is_dead(cgrp)) { + rcu_read_unlock(); + mutex_unlock(&cgroup_mutex); + return -ENOENT; + } + rcu_read_unlock(); + + css_task_iter_start(&cgrp->self, &it); + while ((tsk = css_task_iter_next(&it))) { + switch (tsk->state) { + case TASK_RUNNING: + stats->nr_running++; + break; + case TASK_INTERRUPTIBLE: + stats->nr_sleeping++; + break; + case TASK_UNINTERRUPTIBLE: + stats->nr_uninterruptible++; + break; + case TASK_STOPPED: + stats->nr_stopped++; + break; + default: + if (delayacct_is_task_waiting_on_io(tsk)) + stats->nr_io_wait++; + break; + } + } + css_task_iter_end(&it); + + mutex_unlock(&cgroup_mutex); + return 0; +} + +void cgroup1_check_for_release(struct cgroup *cgrp) +{ + if (notify_on_release(cgrp) && !cgroup_is_populated(cgrp) && + !css_has_online_children(&cgrp->self) && !cgroup_is_dead(cgrp)) + schedule_work(&cgrp->release_agent_work); +} + +/* + * Notify userspace when a cgroup is released, by running the + * configured release agent with the name of the cgroup (path + * relative to the root of cgroup file system) as the argument. + * + * Most likely, this user command will try to rmdir this cgroup. + * + * This races with the possibility that some other task will be + * attached to this cgroup before it is removed, or that some other + * user task will 'mkdir' a child cgroup of this cgroup. That's ok. + * The presumed 'rmdir' will fail quietly if this cgroup is no longer + * unused, and this cgroup will be reprieved from its death sentence, + * to continue to serve a useful existence. Next time it's released, + * we will get notified again, if it still has 'notify_on_release' set. + * + * The final arg to call_usermodehelper() is UMH_WAIT_EXEC, which + * means only wait until the task is successfully execve()'d. The + * separate release agent task is forked by call_usermodehelper(), + * then control in this thread returns here, without waiting for the + * release agent task. We don't bother to wait because the caller of + * this routine has no use for the exit status of the release agent + * task, so no sense holding our caller up for that. + */ +void cgroup1_release_agent(struct work_struct *work) +{ + struct cgroup *cgrp = + container_of(work, struct cgroup, release_agent_work); + char *pathbuf = NULL, *agentbuf = NULL; + char *argv[3], *envp[3]; + int ret; + + mutex_lock(&cgroup_mutex); + + pathbuf = kmalloc(PATH_MAX, GFP_KERNEL); + agentbuf = kstrdup(cgrp->root->release_agent_path, GFP_KERNEL); + if (!pathbuf || !agentbuf) + goto out; + + spin_lock_irq(&css_set_lock); + ret = cgroup_path_ns_locked(cgrp, pathbuf, PATH_MAX, &init_cgroup_ns); + spin_unlock_irq(&css_set_lock); + if (ret < 0 || ret >= PATH_MAX) + goto out; + + argv[0] = agentbuf; + argv[1] = pathbuf; + argv[2] = NULL; + + /* minimal command environment */ + envp[0] = "HOME=/"; + envp[1] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin"; + envp[2] = NULL; + + mutex_unlock(&cgroup_mutex); + call_usermodehelper(argv[0], argv, envp, UMH_WAIT_EXEC); + goto out_free; +out: + mutex_unlock(&cgroup_mutex); +out_free: + kfree(agentbuf); + kfree(pathbuf); +} + +/* + * cgroup_rename - Only allow simple rename of directories in place. + */ +static int cgroup1_rename(struct kernfs_node *kn, struct kernfs_node *new_parent, + const char *new_name_str) +{ + struct cgroup *cgrp = kn->priv; + int ret; + + if (kernfs_type(kn) != KERNFS_DIR) + return -ENOTDIR; + if (kn->parent != new_parent) + return -EIO; + + /* + * We're gonna grab cgroup_mutex which nests outside kernfs + * active_ref. kernfs_rename() doesn't require active_ref + * protection. Break them before grabbing cgroup_mutex. + */ + kernfs_break_active_protection(new_parent); + kernfs_break_active_protection(kn); + + mutex_lock(&cgroup_mutex); + + ret = kernfs_rename(kn, new_parent, new_name_str); + if (!ret) + trace_cgroup_rename(cgrp); + + mutex_unlock(&cgroup_mutex); + + kernfs_unbreak_active_protection(kn); + kernfs_unbreak_active_protection(new_parent); + return ret; +} + +static int cgroup1_show_options(struct seq_file *seq, struct kernfs_root *kf_root) +{ + struct cgroup_root *root = cgroup_root_from_kf(kf_root); + struct cgroup_subsys *ss; + int ssid; + + for_each_subsys(ss, ssid) + if (root->subsys_mask & (1 << ssid)) + seq_show_option(seq, ss->legacy_name, NULL); + if (root->flags & CGRP_ROOT_NOPREFIX) + seq_puts(seq, ",noprefix"); + if (root->flags & CGRP_ROOT_XATTR) + seq_puts(seq, ",xattr"); + + spin_lock(&release_agent_path_lock); + if (strlen(root->release_agent_path)) + seq_show_option(seq, "release_agent", + root->release_agent_path); + spin_unlock(&release_agent_path_lock); + + if (test_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags)) + seq_puts(seq, ",clone_children"); + if (strlen(root->name)) + seq_show_option(seq, "name", root->name); + return 0; +} + +static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) +{ + char *token, *o = data; + bool all_ss = false, one_ss = false; + u16 mask = U16_MAX; + struct cgroup_subsys *ss; + int nr_opts = 0; + int i; + +#ifdef CONFIG_CPUSETS + mask = ~((u16)1 << cpuset_cgrp_id); +#endif + + memset(opts, 0, sizeof(*opts)); + + while ((token = strsep(&o, ",")) != NULL) { + nr_opts++; + + if (!*token) + return -EINVAL; + if (!strcmp(token, "none")) { + /* Explicitly have no subsystems */ + opts->none = true; + continue; + } + if (!strcmp(token, "all")) { + /* Mutually exclusive option 'all' + subsystem name */ + if (one_ss) + return -EINVAL; + all_ss = true; + continue; + } + if (!strcmp(token, "noprefix")) { + opts->flags |= CGRP_ROOT_NOPREFIX; + continue; + } + if (!strcmp(token, "clone_children")) { + opts->cpuset_clone_children = true; + continue; + } + if (!strcmp(token, "xattr")) { + opts->flags |= CGRP_ROOT_XATTR; + continue; + } + if (!strncmp(token, "release_agent=", 14)) { + /* Specifying two release agents is forbidden */ + if (opts->release_agent) + return -EINVAL; + opts->release_agent = + kstrndup(token + 14, PATH_MAX - 1, GFP_KERNEL); + if (!opts->release_agent) + return -ENOMEM; + continue; + } + if (!strncmp(token, "name=", 5)) { + const char *name = token + 5; + /* Can't specify an empty name */ + if (!strlen(name)) + return -EINVAL; + /* Must match [\w.-]+ */ + for (i = 0; i < strlen(name); i++) { + char c = name[i]; + if (isalnum(c)) + continue; + if ((c == '.') || (c == '-') || (c == '_')) + continue; + return -EINVAL; + } + /* Specifying two names is forbidden */ + if (opts->name) + return -EINVAL; + opts->name = kstrndup(name, + MAX_CGROUP_ROOT_NAMELEN - 1, + GFP_KERNEL); + if (!opts->name) + return -ENOMEM; + + continue; + } + + for_each_subsys(ss, i) { + if (strcmp(token, ss->legacy_name)) + continue; + if (!cgroup_ssid_enabled(i)) + continue; + if (cgroup1_ssid_disabled(i)) + continue; + + /* Mutually exclusive option 'all' + subsystem name */ + if (all_ss) + return -EINVAL; + opts->subsys_mask |= (1 << i); + one_ss = true; + + break; + } + if (i == CGROUP_SUBSYS_COUNT) + return -ENOENT; + } + + /* + * If the 'all' option was specified select all the subsystems, + * otherwise if 'none', 'name=' and a subsystem name options were + * not specified, let's default to 'all' + */ + if (all_ss || (!one_ss && !opts->none && !opts->name)) + for_each_subsys(ss, i) + if (cgroup_ssid_enabled(i) && !cgroup1_ssid_disabled(i)) + opts->subsys_mask |= (1 << i); + + /* + * We either have to specify by name or by subsystems. (So all + * empty hierarchies must have a name). + */ + if (!opts->subsys_mask && !opts->name) + return -EINVAL; + + /* + * Option noprefix was introduced just for backward compatibility + * with the old cpuset, so we allow noprefix only if mounting just + * the cpuset subsystem. + */ + if ((opts->flags & CGRP_ROOT_NOPREFIX) && (opts->subsys_mask & mask)) + return -EINVAL; + + /* Can't specify "none" and some subsystems */ + if (opts->subsys_mask && opts->none) + return -EINVAL; + + return 0; +} + +static int cgroup1_remount(struct kernfs_root *kf_root, int *flags, char *data) +{ + int ret = 0; + struct cgroup_root *root = cgroup_root_from_kf(kf_root); + struct cgroup_sb_opts opts; + u16 added_mask, removed_mask; + + cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp); + + /* See what subsystems are wanted */ + ret = parse_cgroupfs_options(data, &opts); + if (ret) + goto out_unlock; + + if (opts.subsys_mask != root->subsys_mask || opts.release_agent) + pr_warn("option changes via remount are deprecated (pid=%d comm=%s)\n", + task_tgid_nr(current), current->comm); + + added_mask = opts.subsys_mask & ~root->subsys_mask; + removed_mask = root->subsys_mask & ~opts.subsys_mask; + + /* Don't allow flags or name to change at remount */ + if ((opts.flags ^ root->flags) || + (opts.name && strcmp(opts.name, root->name))) { + pr_err("option or name mismatch, new: 0x%x \"%s\", old: 0x%x \"%s\"\n", + opts.flags, opts.name ?: "", root->flags, root->name); + ret = -EINVAL; + goto out_unlock; + } + + /* remounting is not allowed for populated hierarchies */ + if (!list_empty(&root->cgrp.self.children)) { + ret = -EBUSY; + goto out_unlock; + } + + ret = rebind_subsystems(root, added_mask); + if (ret) + goto out_unlock; + + WARN_ON(rebind_subsystems(&cgrp_dfl_root, removed_mask)); + + if (opts.release_agent) { + spin_lock(&release_agent_path_lock); + strcpy(root->release_agent_path, opts.release_agent); + spin_unlock(&release_agent_path_lock); + } + + trace_cgroup_remount(root); + + out_unlock: + kfree(opts.release_agent); + kfree(opts.name); + mutex_unlock(&cgroup_mutex); + return ret; +} + +struct kernfs_syscall_ops cgroup1_kf_syscall_ops = { + .rename = cgroup1_rename, + .show_options = cgroup1_show_options, + .remount_fs = cgroup1_remount, + .mkdir = cgroup_mkdir, + .rmdir = cgroup_rmdir, + .show_path = cgroup_show_path, +}; + +struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags, + void *data, unsigned long magic, + struct cgroup_namespace *ns) +{ + struct super_block *pinned_sb = NULL; + struct cgroup_sb_opts opts; + struct cgroup_root *root; + struct cgroup_subsys *ss; + struct dentry *dentry; + int i, ret; + bool new_root = false; + + cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp); + + /* First find the desired set of subsystems */ + ret = parse_cgroupfs_options(data, &opts); + if (ret) + goto out_unlock; + + /* + * Destruction of cgroup root is asynchronous, so subsystems may + * still be dying after the previous unmount. Let's drain the + * dying subsystems. We just need to ensure that the ones + * unmounted previously finish dying and don't care about new ones + * starting. Testing ref liveliness is good enough. + */ + for_each_subsys(ss, i) { + if (!(opts.subsys_mask & (1 << i)) || + ss->root == &cgrp_dfl_root) + continue; + + if (!percpu_ref_tryget_live(&ss->root->cgrp.self.refcnt)) { + mutex_unlock(&cgroup_mutex); + msleep(10); + ret = restart_syscall(); + goto out_free; + } + cgroup_put(&ss->root->cgrp); + } + + for_each_root(root) { + bool name_match = false; + + if (root == &cgrp_dfl_root) + continue; + + /* + * If we asked for a name then it must match. Also, if + * name matches but sybsys_mask doesn't, we should fail. + * Remember whether name matched. + */ + if (opts.name) { + if (strcmp(opts.name, root->name)) + continue; + name_match = true; + } + + /* + * If we asked for subsystems (or explicitly for no + * subsystems) then they must match. + */ + if ((opts.subsys_mask || opts.none) && + (opts.subsys_mask != root->subsys_mask)) { + if (!name_match) + continue; + ret = -EBUSY; + goto out_unlock; + } + + if (root->flags ^ opts.flags) + pr_warn("new mount options do not match the existing superblock, will be ignored\n"); + + /* + * We want to reuse @root whose lifetime is governed by its + * ->cgrp. Let's check whether @root is alive and keep it + * that way. As cgroup_kill_sb() can happen anytime, we + * want to block it by pinning the sb so that @root doesn't + * get killed before mount is complete. + * + * With the sb pinned, tryget_live can reliably indicate + * whether @root can be reused. If it's being killed, + * drain it. We can use wait_queue for the wait but this + * path is super cold. Let's just sleep a bit and retry. + */ + pinned_sb = kernfs_pin_sb(root->kf_root, NULL); + if (IS_ERR(pinned_sb) || + !percpu_ref_tryget_live(&root->cgrp.self.refcnt)) { + mutex_unlock(&cgroup_mutex); + if (!IS_ERR_OR_NULL(pinned_sb)) + deactivate_super(pinned_sb); + msleep(10); + ret = restart_syscall(); + goto out_free; + } + + ret = 0; + goto out_unlock; + } + + /* + * No such thing, create a new one. name= matching without subsys + * specification is allowed for already existing hierarchies but we + * can't create new one without subsys specification. + */ + if (!opts.subsys_mask && !opts.none) { + ret = -EINVAL; + goto out_unlock; + } + + /* Hierarchies may only be created in the initial cgroup namespace. */ + if (ns != &init_cgroup_ns) { + ret = -EPERM; + goto out_unlock; + } + + root = kzalloc(sizeof(*root), GFP_KERNEL); + if (!root) { + ret = -ENOMEM; + goto out_unlock; + } + new_root = true; + + init_cgroup_root(root, &opts); + + ret = cgroup_setup_root(root, opts.subsys_mask, PERCPU_REF_INIT_DEAD); + if (ret) + cgroup_free_root(root); + +out_unlock: + mutex_unlock(&cgroup_mutex); +out_free: + kfree(opts.release_agent); + kfree(opts.name); + + if (ret) + return ERR_PTR(ret); + + dentry = cgroup_do_mount(&cgroup_fs_type, flags, root, + CGROUP_SUPER_MAGIC, ns); + + /* + * There's a race window after we release cgroup_mutex and before + * allocating a superblock. Make sure a concurrent process won't + * be able to re-use the root during this window by delaying the + * initialization of root refcnt. + */ + if (new_root) { + mutex_lock(&cgroup_mutex); + percpu_ref_reinit(&root->cgrp.self.refcnt); + mutex_unlock(&cgroup_mutex); + } + + /* + * If @pinned_sb, we're reusing an existing root and holding an + * extra ref on its sb. Mount is complete. Put the extra ref. + */ + if (pinned_sb) + deactivate_super(pinned_sb); + + return dentry; +} + +static int __init cgroup1_wq_init(void) +{ + /* + * Used to destroy pidlists and separate to serve as flush domain. + * Cap @max_active to 1 too. + */ + cgroup_pidlist_destroy_wq = alloc_workqueue("cgroup_pidlist_destroy", + 0, 1); + BUG_ON(!cgroup_pidlist_destroy_wq); + return 0; +} +core_initcall(cgroup1_wq_init); + +static int __init cgroup_no_v1(char *str) +{ + struct cgroup_subsys *ss; + char *token; + int i; + + while ((token = strsep(&str, ",")) != NULL) { + if (!*token) + continue; + + if (!strcmp(token, "all")) { + cgroup_no_v1_mask = U16_MAX; + break; + } + + for_each_subsys(ss, i) { + if (strcmp(token, ss->name) && + strcmp(token, ss->legacy_name)) + continue; + + cgroup_no_v1_mask |= 1 << i; + } + } + return 1; +} +__setup("cgroup_no_v1=", cgroup_no_v1); diff --git a/kernel/cgroup.c b/kernel/cgroup/cgroup.c index 688dd02af985..620794a20a33 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -28,35 +28,27 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt -#include <linux/cgroup.h> +#include "cgroup-internal.h" + #include <linux/cred.h> -#include <linux/ctype.h> #include <linux/errno.h> #include <linux/init_task.h> #include <linux/kernel.h> -#include <linux/list.h> #include <linux/magic.h> -#include <linux/mm.h> #include <linux/mutex.h> #include <linux/mount.h> #include <linux/pagemap.h> #include <linux/proc_fs.h> #include <linux/rcupdate.h> #include <linux/sched.h> +#include <linux/sched/task.h> #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/percpu-rwsem.h> #include <linux/string.h> -#include <linux/sort.h> -#include <linux/kmod.h> -#include <linux/delayacct.h> -#include <linux/cgroupstats.h> #include <linux/hashtable.h> -#include <linux/pid_namespace.h> #include <linux/idr.h> -#include <linux/vmalloc.h> /* TODO: replace with more sophisticated array */ #include <linux/kthread.h> -#include <linux/delay.h> #include <linux/atomic.h> #include <linux/cpuset.h> #include <linux/proc_ns.h> @@ -67,14 +59,6 @@ #define CREATE_TRACE_POINTS #include <trace/events/cgroup.h> -/* - * pidlists linger the following amount before being destroyed. The goal - * is avoiding frequent destruction in the middle of consecutive read calls - * Expiring in the middle is a performance problem not a correctness one. - * 1 sec should be enough. - */ -#define CGROUP_PIDLIST_DESTROY_DELAY HZ - #define CGROUP_FILE_NAME_MAX (MAX_CGROUP_TYPE_NAMELEN + \ MAX_CFTYPE_NAME + 2) @@ -88,14 +72,12 @@ * These locks are exported if CONFIG_PROVE_RCU so that accessors in * cgroup.h can use them for lockdep annotations. */ -#ifdef CONFIG_PROVE_RCU DEFINE_MUTEX(cgroup_mutex); DEFINE_SPINLOCK(css_set_lock); + +#ifdef CONFIG_PROVE_RCU EXPORT_SYMBOL_GPL(cgroup_mutex); EXPORT_SYMBOL_GPL(css_set_lock); -#else -static DEFINE_MUTEX(cgroup_mutex); -static DEFINE_SPINLOCK(css_set_lock); #endif /* @@ -110,12 +92,6 @@ static DEFINE_SPINLOCK(cgroup_idr_lock); */ static DEFINE_SPINLOCK(cgroup_file_kn_lock); -/* - * Protects cgroup_subsys->release_agent_path. Modifying it also requires - * cgroup_mutex. Reading requires either cgroup_mutex or this spinlock. - */ -static DEFINE_SPINLOCK(release_agent_path_lock); - struct percpu_rw_semaphore cgroup_threadgroup_rwsem; #define cgroup_assert_mutex_or_rcu_locked() \ @@ -131,15 +107,9 @@ struct percpu_rw_semaphore cgroup_threadgroup_rwsem; */ static struct workqueue_struct *cgroup_destroy_wq; -/* - * pidlist destructions need to be flushed on cgroup destruction. Use a - * separate workqueue as flush domain. - */ -static struct workqueue_struct *cgroup_pidlist_destroy_wq; - /* generate an array of cgroup subsystem pointers */ #define SUBSYS(_x) [_x ## _cgrp_id] = &_x ## _cgrp_subsys, -static struct cgroup_subsys *cgroup_subsys[] = { +struct cgroup_subsys *cgroup_subsys[] = { #include <linux/cgroup_subsys.h> }; #undef SUBSYS @@ -186,18 +156,14 @@ EXPORT_SYMBOL_GPL(cgrp_dfl_root); */ static bool cgrp_dfl_visible; -/* Controllers blocked by the commandline in v1 */ -static u16 cgroup_no_v1_mask; - /* some controllers are not supported in the default hierarchy */ static u16 cgrp_dfl_inhibit_ss_mask; /* some controllers are implicitly enabled on the default hierarchy */ -static unsigned long cgrp_dfl_implicit_ss_mask; +static u16 cgrp_dfl_implicit_ss_mask; /* The list of hierarchy roots */ - -static LIST_HEAD(cgroup_roots); +LIST_HEAD(cgroup_roots); static int cgroup_root_count; /* hierarchy ID allocation and mapping, protected by cgroup_mutex */ @@ -213,32 +179,26 @@ static DEFINE_IDR(cgroup_hierarchy_idr); static u64 css_serial_nr_next = 1; /* - * These bitmask flags indicate whether tasks in the fork and exit paths have - * fork/exit handlers to call. This avoids us having to do extra work in the - * fork/exit path to check which subsystems have fork/exit callbacks. + * These bitmasks identify subsystems with specific features to avoid + * having to do iterative checks repeatedly. */ static u16 have_fork_callback __read_mostly; static u16 have_exit_callback __read_mostly; static u16 have_free_callback __read_mostly; +static u16 have_canfork_callback __read_mostly; /* cgroup namespace for init task */ struct cgroup_namespace init_cgroup_ns = { - .count = { .counter = 2, }, + .count = REFCOUNT_INIT(2), .user_ns = &init_user_ns, .ns.ops = &cgroupns_operations, .ns.inum = PROC_CGROUP_INIT_INO, .root_cset = &init_css_set, }; -/* Ditto for the can_fork callback. */ -static u16 have_canfork_callback __read_mostly; - static struct file_system_type cgroup2_fs_type; -static struct cftype cgroup_dfl_base_files[]; -static struct cftype cgroup_legacy_base_files[]; +static struct cftype cgroup_base_files[]; -static int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask); -static void cgroup_lock_and_drain_offline(struct cgroup *cgrp); static int cgroup_apply_control(struct cgroup *cgrp); static void cgroup_finalize_control(struct cgroup *cgrp, int ret); static void css_task_iter_advance(struct css_task_iter *it); @@ -259,7 +219,7 @@ static int cgroup_addrm_files(struct cgroup_subsys_state *css, * is fine for individual subsystems but unsuitable for cgroup core. This * is slower static_key_enabled() based test indexed by @ssid. */ -static bool cgroup_ssid_enabled(int ssid) +bool cgroup_ssid_enabled(int ssid) { if (CGROUP_SUBSYS_COUNT == 0) return false; @@ -267,11 +227,6 @@ static bool cgroup_ssid_enabled(int ssid) return static_key_enabled(cgroup_subsys_enabled_key[ssid]); } -static bool cgroup_ssid_no_v1(int ssid) -{ - return cgroup_no_v1_mask & (1 << ssid); -} - /** * cgroup_on_dfl - test whether a cgroup is on the default hierarchy * @cgrp: the cgroup of interest @@ -325,7 +280,7 @@ static bool cgroup_ssid_no_v1(int ssid) * * - debug: disallowed on the default hierarchy. */ -static bool cgroup_on_dfl(const struct cgroup *cgrp) +bool cgroup_on_dfl(const struct cgroup *cgrp) { return cgrp->root == &cgrp_dfl_root; } @@ -481,13 +436,12 @@ out_unlock: return css; } -/* convenient tests for these bits */ -static inline bool cgroup_is_dead(const struct cgroup *cgrp) +static void __maybe_unused cgroup_get(struct cgroup *cgrp) { - return !(cgrp->self.flags & CSS_ONLINE); + css_get(&cgrp->self); } -static void cgroup_get(struct cgroup *cgrp) +static void cgroup_get_live(struct cgroup *cgrp) { WARN_ON_ONCE(cgroup_is_dead(cgrp)); css_get(&cgrp->self); @@ -518,11 +472,6 @@ struct cgroup_subsys_state *of_css(struct kernfs_open_file *of) } EXPORT_SYMBOL_GPL(of_css); -static int notify_on_release(const struct cgroup *cgrp) -{ - return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); -} - /** * for_each_css - iterate all css's of a cgroup * @css: the iteration cursor @@ -553,15 +502,6 @@ static int notify_on_release(const struct cgroup *cgrp) else /** - * for_each_subsys - iterate all enabled cgroup subsystems - * @ss: the iteration cursor - * @ssid: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end - */ -#define for_each_subsys(ss, ssid) \ - for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT && \ - (((ss) = cgroup_subsys[ssid]) || true); (ssid)++) - -/** * do_each_subsys_mask - filter for_each_subsys with a bitmask * @ss: the iteration cursor * @ssid: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end @@ -585,10 +525,6 @@ static int notify_on_release(const struct cgroup *cgrp) } \ } while (false) -/* iterate across the hierarchies */ -#define for_each_root(root) \ - list_for_each_entry((root), &cgroup_roots, root_list) - /* iterate over child cgrps, lock should be held throughout iteration */ #define cgroup_for_each_live_child(child, cgrp) \ list_for_each_entry((child), &(cgrp)->self.children, self.sibling) \ @@ -615,29 +551,6 @@ static int notify_on_release(const struct cgroup *cgrp) ; \ else -static void cgroup_release_agent(struct work_struct *work); -static void check_for_release(struct cgroup *cgrp); - -/* - * A cgroup can be associated with multiple css_sets as different tasks may - * belong to different cgroups on different hierarchies. In the other - * direction, a css_set is naturally associated with multiple cgroups. - * This M:N relationship is represented by the following link structure - * which exists for each association and allows traversing the associations - * from both sides. - */ -struct cgrp_cset_link { - /* the cgroup and css_set this link associates */ - struct cgroup *cgrp; - struct css_set *cset; - - /* list of cgrp_cset_links anchored at cgrp->cset_links */ - struct list_head cset_link; - - /* list of cgrp_cset_links anchored at css_set->cgrp_links */ - struct list_head cgrp_link; -}; - /* * The default css_set - used by init and its children prior to any * hierarchies being mounted. It contains a pointer to the root state @@ -646,13 +559,13 @@ struct cgrp_cset_link { * haven't been created. */ struct css_set init_css_set = { - .refcount = ATOMIC_INIT(1), - .cgrp_links = LIST_HEAD_INIT(init_css_set.cgrp_links), + .refcount = REFCOUNT_INIT(1), .tasks = LIST_HEAD_INIT(init_css_set.tasks), .mg_tasks = LIST_HEAD_INIT(init_css_set.mg_tasks), + .task_iters = LIST_HEAD_INIT(init_css_set.task_iters), + .cgrp_links = LIST_HEAD_INIT(init_css_set.cgrp_links), .mg_preload_node = LIST_HEAD_INIT(init_css_set.mg_preload_node), .mg_node = LIST_HEAD_INIT(init_css_set.mg_node), - .task_iters = LIST_HEAD_INIT(init_css_set.task_iters), }; static int css_set_count = 1; /* 1 for init_css_set */ @@ -660,6 +573,11 @@ static int css_set_count = 1; /* 1 for init_css_set */ /** * css_set_populated - does a css_set contain any tasks? * @cset: target css_set + * + * css_set_populated() should be the same as !!cset->nr_tasks at steady + * state. However, css_set_populated() can be called while a task is being + * added to or removed from the linked list before the nr_tasks is + * properly updated. Hence, we can't just look at ->nr_tasks here. */ static bool css_set_populated(struct css_set *cset) { @@ -699,7 +617,7 @@ static void cgroup_update_populated(struct cgroup *cgrp, bool populated) if (!trigger) break; - check_for_release(cgrp); + cgroup1_check_for_release(cgrp); cgroup_file_notify(&cgrp->events_file); cgrp = cgroup_parent(cgrp); @@ -808,7 +726,7 @@ static unsigned long css_set_hash(struct cgroup_subsys_state *css[]) return key; } -static void put_css_set_locked(struct css_set *cset) +void put_css_set_locked(struct css_set *cset) { struct cgrp_cset_link *link, *tmp_link; struct cgroup_subsys *ss; @@ -816,7 +734,7 @@ static void put_css_set_locked(struct css_set *cset) lockdep_assert_held(&css_set_lock); - if (!atomic_dec_and_test(&cset->refcount)) + if (!refcount_dec_and_test(&cset->refcount)) return; /* This css_set is dead. unlink it and release cgroup and css refs */ @@ -838,31 +756,6 @@ static void put_css_set_locked(struct css_set *cset) kfree_rcu(cset, rcu_head); } -static void put_css_set(struct css_set *cset) -{ - unsigned long flags; - - /* - * Ensure that the refcount doesn't hit zero while any readers - * can see it. Similar to atomic_dec_and_lock(), but for an - * rwlock - */ - if (atomic_add_unless(&cset->refcount, -1, 1)) - return; - - spin_lock_irqsave(&css_set_lock, flags); - put_css_set_locked(cset); - spin_unlock_irqrestore(&css_set_lock, flags); -} - -/* - * refcounted get/put for css_set objects - */ -static inline void get_css_set(struct css_set *cset) -{ - atomic_inc(&cset->refcount); -} - /** * compare_css_sets - helper function for find_existing_css_set(). * @cset: candidate css_set being tested @@ -1049,7 +942,7 @@ static void link_css_set(struct list_head *tmp_links, struct css_set *cset, list_add_tail(&link->cgrp_link, &cset->cgrp_links); if (cgroup_parent(cgrp)) - cgroup_get(cgrp); + cgroup_get_live(cgrp); } /** @@ -1094,14 +987,14 @@ static struct css_set *find_css_set(struct css_set *old_cset, return NULL; } - atomic_set(&cset->refcount, 1); - INIT_LIST_HEAD(&cset->cgrp_links); + refcount_set(&cset->refcount, 1); INIT_LIST_HEAD(&cset->tasks); INIT_LIST_HEAD(&cset->mg_tasks); - INIT_LIST_HEAD(&cset->mg_preload_node); - INIT_LIST_HEAD(&cset->mg_node); INIT_LIST_HEAD(&cset->task_iters); INIT_HLIST_NODE(&cset->hlist); + INIT_LIST_HEAD(&cset->cgrp_links); + INIT_LIST_HEAD(&cset->mg_preload_node); + INIT_LIST_HEAD(&cset->mg_node); /* Copy the set of subsystem state objects generated in * find_existing_css_set() */ @@ -1138,7 +1031,7 @@ static struct css_set *find_css_set(struct css_set *old_cset, return cset; } -static struct cgroup_root *cgroup_root_from_kf(struct kernfs_root *kf_root) +struct cgroup_root *cgroup_root_from_kf(struct kernfs_root *kf_root) { struct cgroup *root_cgrp = kf_root->kn->priv; @@ -1166,7 +1059,7 @@ static void cgroup_exit_root_id(struct cgroup_root *root) idr_remove(&cgroup_hierarchy_idr, root->hierarchy_id); } -static void cgroup_free_root(struct cgroup_root *root) +void cgroup_free_root(struct cgroup_root *root) { if (root) { idr_destroy(&root->cgroup_idr); @@ -1283,8 +1176,8 @@ static struct cgroup *cset_cgroup_from_root(struct css_set *cset, * Return the cgroup for "task" from the given hierarchy. Must be * called with cgroup_mutex and css_set_lock held. */ -static struct cgroup *task_cgroup_from_root(struct task_struct *task, - struct cgroup_root *root) +struct cgroup *task_cgroup_from_root(struct task_struct *task, + struct cgroup_root *root) { /* * No need to lock the task - since we hold cgroup_mutex the @@ -1321,7 +1214,6 @@ static struct cgroup *task_cgroup_from_root(struct task_struct *task, */ static struct kernfs_syscall_ops cgroup_kf_syscall_ops; -static const struct file_operations proc_cgroupstats_operations; static char *cgroup_file_name(struct cgroup *cgrp, const struct cftype *cft, char *buf) @@ -1415,7 +1307,7 @@ static u16 cgroup_calc_subtree_ss_mask(u16 subtree_control, u16 this_ss_mask) * inaccessible any time. If the caller intends to continue to access the * cgroup, it should pin it before invoking this function. */ -static void cgroup_kn_unlock(struct kernfs_node *kn) +void cgroup_kn_unlock(struct kernfs_node *kn) { struct cgroup *cgrp; @@ -1447,8 +1339,7 @@ static void cgroup_kn_unlock(struct kernfs_node *kn) * locking under kernfs active protection and allows all kernfs operations * including self-removal. */ -static struct cgroup *cgroup_kn_lock_live(struct kernfs_node *kn, - bool drain_offline) +struct cgroup *cgroup_kn_lock_live(struct kernfs_node *kn, bool drain_offline) { struct cgroup *cgrp; @@ -1532,9 +1423,9 @@ static int css_populate_dir(struct cgroup_subsys_state *css) if (!css->ss) { if (cgroup_on_dfl(cgrp)) - cfts = cgroup_dfl_base_files; + cfts = cgroup_base_files; else - cfts = cgroup_legacy_base_files; + cfts = cgroup1_base_files; return cgroup_addrm_files(&cgrp->self, cgrp, cfts, true); } @@ -1559,7 +1450,7 @@ err: return ret; } -static int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask) +int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask) { struct cgroup *dcgrp = &dst_root->cgrp; struct cgroup_subsys *ss; @@ -1629,8 +1520,8 @@ static int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask) return 0; } -static int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node, - struct kernfs_root *kf_root) +int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node, + struct kernfs_root *kf_root) { int len = 0; char *buf = NULL; @@ -1656,237 +1547,56 @@ static int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node, return len; } -static int cgroup_show_options(struct seq_file *seq, - struct kernfs_root *kf_root) -{ - struct cgroup_root *root = cgroup_root_from_kf(kf_root); - struct cgroup_subsys *ss; - int ssid; - - if (root != &cgrp_dfl_root) - for_each_subsys(ss, ssid) - if (root->subsys_mask & (1 << ssid)) - seq_show_option(seq, ss->legacy_name, NULL); - if (root->flags & CGRP_ROOT_NOPREFIX) - seq_puts(seq, ",noprefix"); - if (root->flags & CGRP_ROOT_XATTR) - seq_puts(seq, ",xattr"); - - spin_lock(&release_agent_path_lock); - if (strlen(root->release_agent_path)) - seq_show_option(seq, "release_agent", - root->release_agent_path); - spin_unlock(&release_agent_path_lock); - - if (test_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags)) - seq_puts(seq, ",clone_children"); - if (strlen(root->name)) - seq_show_option(seq, "name", root->name); - return 0; -} - -struct cgroup_sb_opts { - u16 subsys_mask; - unsigned int flags; - char *release_agent; - bool cpuset_clone_children; - char *name; - /* User explicitly requested empty subsystem */ - bool none; -}; - -static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) +static int parse_cgroup_root_flags(char *data, unsigned int *root_flags) { - char *token, *o = data; - bool all_ss = false, one_ss = false; - u16 mask = U16_MAX; - struct cgroup_subsys *ss; - int nr_opts = 0; - int i; - -#ifdef CONFIG_CPUSETS - mask = ~((u16)1 << cpuset_cgrp_id); -#endif - - memset(opts, 0, sizeof(*opts)); + char *token; - while ((token = strsep(&o, ",")) != NULL) { - nr_opts++; + *root_flags = 0; - if (!*token) - return -EINVAL; - if (!strcmp(token, "none")) { - /* Explicitly have no subsystems */ - opts->none = true; - continue; - } - if (!strcmp(token, "all")) { - /* Mutually exclusive option 'all' + subsystem name */ - if (one_ss) - return -EINVAL; - all_ss = true; - continue; - } - if (!strcmp(token, "noprefix")) { - opts->flags |= CGRP_ROOT_NOPREFIX; - continue; - } - if (!strcmp(token, "clone_children")) { - opts->cpuset_clone_children = true; - continue; - } - if (!strcmp(token, "xattr")) { - opts->flags |= CGRP_ROOT_XATTR; - continue; - } - if (!strncmp(token, "release_agent=", 14)) { - /* Specifying two release agents is forbidden */ - if (opts->release_agent) - return -EINVAL; - opts->release_agent = - kstrndup(token + 14, PATH_MAX - 1, GFP_KERNEL); - if (!opts->release_agent) - return -ENOMEM; - continue; - } - if (!strncmp(token, "name=", 5)) { - const char *name = token + 5; - /* Can't specify an empty name */ - if (!strlen(name)) - return -EINVAL; - /* Must match [\w.-]+ */ - for (i = 0; i < strlen(name); i++) { - char c = name[i]; - if (isalnum(c)) - continue; - if ((c == '.') || (c == '-') || (c == '_')) - continue; - return -EINVAL; - } - /* Specifying two names is forbidden */ - if (opts->name) - return -EINVAL; - opts->name = kstrndup(name, - MAX_CGROUP_ROOT_NAMELEN - 1, - GFP_KERNEL); - if (!opts->name) - return -ENOMEM; + if (!data) + return 0; + while ((token = strsep(&data, ",")) != NULL) { + if (!strcmp(token, "nsdelegate")) { + *root_flags |= CGRP_ROOT_NS_DELEGATE; continue; } - for_each_subsys(ss, i) { - if (strcmp(token, ss->legacy_name)) - continue; - if (!cgroup_ssid_enabled(i)) - continue; - if (cgroup_ssid_no_v1(i)) - continue; - - /* Mutually exclusive option 'all' + subsystem name */ - if (all_ss) - return -EINVAL; - opts->subsys_mask |= (1 << i); - one_ss = true; - - break; - } - if (i == CGROUP_SUBSYS_COUNT) - return -ENOENT; - } - - /* - * If the 'all' option was specified select all the subsystems, - * otherwise if 'none', 'name=' and a subsystem name options were - * not specified, let's default to 'all' - */ - if (all_ss || (!one_ss && !opts->none && !opts->name)) - for_each_subsys(ss, i) - if (cgroup_ssid_enabled(i) && !cgroup_ssid_no_v1(i)) - opts->subsys_mask |= (1 << i); - - /* - * We either have to specify by name or by subsystems. (So all - * empty hierarchies must have a name). - */ - if (!opts->subsys_mask && !opts->name) - return -EINVAL; - - /* - * Option noprefix was introduced just for backward compatibility - * with the old cpuset, so we allow noprefix only if mounting just - * the cpuset subsystem. - */ - if ((opts->flags & CGRP_ROOT_NOPREFIX) && (opts->subsys_mask & mask)) - return -EINVAL; - - /* Can't specify "none" and some subsystems */ - if (opts->subsys_mask && opts->none) + pr_err("cgroup2: unknown option \"%s\"\n", token); return -EINVAL; + } return 0; } -static int cgroup_remount(struct kernfs_root *kf_root, int *flags, char *data) +static void apply_cgroup_root_flags(unsigned int root_flags) { - int ret = 0; - struct cgroup_root *root = cgroup_root_from_kf(kf_root); - struct cgroup_sb_opts opts; - u16 added_mask, removed_mask; - - if (root == &cgrp_dfl_root) { - pr_err("remount is not allowed\n"); - return -EINVAL; + if (current->nsproxy->cgroup_ns == &init_cgroup_ns) { + if (root_flags & CGRP_ROOT_NS_DELEGATE) + cgrp_dfl_root.flags |= CGRP_ROOT_NS_DELEGATE; + else + cgrp_dfl_root.flags &= ~CGRP_ROOT_NS_DELEGATE; } +} - cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp); - - /* See what subsystems are wanted */ - ret = parse_cgroupfs_options(data, &opts); - if (ret) - goto out_unlock; - - if (opts.subsys_mask != root->subsys_mask || opts.release_agent) - pr_warn("option changes via remount are deprecated (pid=%d comm=%s)\n", - task_tgid_nr(current), current->comm); - - added_mask = opts.subsys_mask & ~root->subsys_mask; - removed_mask = root->subsys_mask & ~opts.subsys_mask; - - /* Don't allow flags or name to change at remount */ - if ((opts.flags ^ root->flags) || - (opts.name && strcmp(opts.name, root->name))) { - pr_err("option or name mismatch, new: 0x%x \"%s\", old: 0x%x \"%s\"\n", - opts.flags, opts.name ?: "", root->flags, root->name); - ret = -EINVAL; - goto out_unlock; - } +static int cgroup_show_options(struct seq_file *seq, struct kernfs_root *kf_root) +{ + if (cgrp_dfl_root.flags & CGRP_ROOT_NS_DELEGATE) + seq_puts(seq, ",nsdelegate"); + return 0; +} - /* remounting is not allowed for populated hierarchies */ - if (!list_empty(&root->cgrp.self.children)) { - ret = -EBUSY; - goto out_unlock; - } +static int cgroup_remount(struct kernfs_root *kf_root, int *flags, char *data) +{ + unsigned int root_flags; + int ret; - ret = rebind_subsystems(root, added_mask); + ret = parse_cgroup_root_flags(data, &root_flags); if (ret) - goto out_unlock; - - WARN_ON(rebind_subsystems(&cgrp_dfl_root, removed_mask)); - - if (opts.release_agent) { - spin_lock(&release_agent_path_lock); - strcpy(root->release_agent_path, opts.release_agent); - spin_unlock(&release_agent_path_lock); - } - - trace_cgroup_remount(root); + return ret; - out_unlock: - kfree(opts.release_agent); - kfree(opts.name); - mutex_unlock(&cgroup_mutex); - return ret; + apply_cgroup_root_flags(root_flags); + return 0; } /* @@ -1939,6 +1649,7 @@ static void cgroup_enable_task_cg_lists(void) css_set_update_populated(cset, true); list_add_tail(&p->cg_list, &cset->tasks); get_css_set(cset); + cset->nr_tasks++; } spin_unlock(&p->sighand->siglock); } while_each_thread(g, p); @@ -1964,11 +1675,10 @@ static void init_cgroup_housekeeping(struct cgroup *cgrp) INIT_LIST_HEAD(&cgrp->e_csets[ssid]); init_waitqueue_head(&cgrp->offline_waitq); - INIT_WORK(&cgrp->release_agent_work, cgroup_release_agent); + INIT_WORK(&cgrp->release_agent_work, cgroup1_release_agent); } -static void init_cgroup_root(struct cgroup_root *root, - struct cgroup_sb_opts *opts) +void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts) { struct cgroup *cgrp = &root->cgrp; @@ -1987,10 +1697,11 @@ static void init_cgroup_root(struct cgroup_root *root, set_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags); } -static int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask) +int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags) { LIST_HEAD(tmp_links); struct cgroup *root_cgrp = &root->cgrp; + struct kernfs_syscall_ops *kf_sops; struct css_set *cset; int i, ret; @@ -2002,8 +1713,8 @@ static int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask) root_cgrp->id = ret; root_cgrp->ancestor_ids[0] = ret; - ret = percpu_ref_init(&root_cgrp->self.refcnt, css_release, 0, - GFP_KERNEL); + ret = percpu_ref_init(&root_cgrp->self.refcnt, css_release, + ref_flags, GFP_KERNEL); if (ret) goto out; @@ -2022,7 +1733,10 @@ static int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask) if (ret) goto cancel_ref; - root->kf_root = kernfs_create_root(&cgroup_kf_syscall_ops, + kf_sops = root == &cgrp_dfl_root ? + &cgroup_kf_syscall_ops : &cgroup1_kf_syscall_ops; + + root->kf_root = kernfs_create_root(kf_sops, KERNFS_ROOT_CREATE_DEACTIVATED, root_cgrp); if (IS_ERR(root->kf_root)) { @@ -2080,182 +1794,18 @@ out: return ret; } -static struct dentry *cgroup_mount(struct file_system_type *fs_type, - int flags, const char *unused_dev_name, - void *data) +struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags, + struct cgroup_root *root, unsigned long magic, + struct cgroup_namespace *ns) { - bool is_v2 = fs_type == &cgroup2_fs_type; - struct super_block *pinned_sb = NULL; - struct cgroup_namespace *ns = current->nsproxy->cgroup_ns; - struct cgroup_subsys *ss; - struct cgroup_root *root; - struct cgroup_sb_opts opts; struct dentry *dentry; - int ret; - int i; bool new_sb; - get_cgroup_ns(ns); - - /* Check if the caller has permission to mount. */ - if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN)) { - put_cgroup_ns(ns); - return ERR_PTR(-EPERM); - } + dentry = kernfs_mount(fs_type, flags, root->kf_root, magic, &new_sb); /* - * The first time anyone tries to mount a cgroup, enable the list - * linking each css_set to its tasks and fix up all existing tasks. - */ - if (!use_task_css_set_links) - cgroup_enable_task_cg_lists(); - - if (is_v2) { - if (data) { - pr_err("cgroup2: unknown option \"%s\"\n", (char *)data); - put_cgroup_ns(ns); - return ERR_PTR(-EINVAL); - } - cgrp_dfl_visible = true; - root = &cgrp_dfl_root; - cgroup_get(&root->cgrp); - goto out_mount; - } - - cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp); - - /* First find the desired set of subsystems */ - ret = parse_cgroupfs_options(data, &opts); - if (ret) - goto out_unlock; - - /* - * Destruction of cgroup root is asynchronous, so subsystems may - * still be dying after the previous unmount. Let's drain the - * dying subsystems. We just need to ensure that the ones - * unmounted previously finish dying and don't care about new ones - * starting. Testing ref liveliness is good enough. - */ - for_each_subsys(ss, i) { - if (!(opts.subsys_mask & (1 << i)) || - ss->root == &cgrp_dfl_root) - continue; - - if (!percpu_ref_tryget_live(&ss->root->cgrp.self.refcnt)) { - mutex_unlock(&cgroup_mutex); - msleep(10); - ret = restart_syscall(); - goto out_free; - } - cgroup_put(&ss->root->cgrp); - } - - for_each_root(root) { - bool name_match = false; - - if (root == &cgrp_dfl_root) - continue; - - /* - * If we asked for a name then it must match. Also, if - * name matches but sybsys_mask doesn't, we should fail. - * Remember whether name matched. - */ - if (opts.name) { - if (strcmp(opts.name, root->name)) - continue; - name_match = true; - } - - /* - * If we asked for subsystems (or explicitly for no - * subsystems) then they must match. - */ - if ((opts.subsys_mask || opts.none) && - (opts.subsys_mask != root->subsys_mask)) { - if (!name_match) - continue; - ret = -EBUSY; - goto out_unlock; - } - - if (root->flags ^ opts.flags) - pr_warn("new mount options do not match the existing superblock, will be ignored\n"); - - /* - * We want to reuse @root whose lifetime is governed by its - * ->cgrp. Let's check whether @root is alive and keep it - * that way. As cgroup_kill_sb() can happen anytime, we - * want to block it by pinning the sb so that @root doesn't - * get killed before mount is complete. - * - * With the sb pinned, tryget_live can reliably indicate - * whether @root can be reused. If it's being killed, - * drain it. We can use wait_queue for the wait but this - * path is super cold. Let's just sleep a bit and retry. - */ - pinned_sb = kernfs_pin_sb(root->kf_root, NULL); - if (IS_ERR(pinned_sb) || - !percpu_ref_tryget_live(&root->cgrp.self.refcnt)) { - mutex_unlock(&cgroup_mutex); - if (!IS_ERR_OR_NULL(pinned_sb)) - deactivate_super(pinned_sb); - msleep(10); - ret = restart_syscall(); - goto out_free; - } - - ret = 0; - goto out_unlock; - } - - /* - * No such thing, create a new one. name= matching without subsys - * specification is allowed for already existing hierarchies but we - * can't create new one without subsys specification. - */ - if (!opts.subsys_mask && !opts.none) { - ret = -EINVAL; - goto out_unlock; - } - - /* Hierarchies may only be created in the initial cgroup namespace. */ - if (ns != &init_cgroup_ns) { - ret = -EPERM; - goto out_unlock; - } - - root = kzalloc(sizeof(*root), GFP_KERNEL); - if (!root) { - ret = -ENOMEM; - goto out_unlock; - } - - init_cgroup_root(root, &opts); - - ret = cgroup_setup_root(root, opts.subsys_mask); - if (ret) - cgroup_free_root(root); - -out_unlock: - mutex_unlock(&cgroup_mutex); -out_free: - kfree(opts.release_agent); - kfree(opts.name); - - if (ret) { - put_cgroup_ns(ns); - return ERR_PTR(ret); - } -out_mount: - dentry = kernfs_mount(fs_type, flags, root->kf_root, - is_v2 ? CGROUP2_SUPER_MAGIC : CGROUP_SUPER_MAGIC, - &new_sb); - - /* - * In non-init cgroup namespace, instead of root cgroup's - * dentry, we return the dentry corresponding to the - * cgroupns->root_cgrp. + * In non-init cgroup namespace, instead of root cgroup's dentry, + * we return the dentry corresponding to the cgroupns->root_cgrp. */ if (!IS_ERR(dentry) && ns != &init_cgroup_ns) { struct dentry *nsdentry; @@ -2277,13 +1827,51 @@ out_mount: if (IS_ERR(dentry) || !new_sb) cgroup_put(&root->cgrp); + return dentry; +} + +static struct dentry *cgroup_mount(struct file_system_type *fs_type, + int flags, const char *unused_dev_name, + void *data) +{ + struct cgroup_namespace *ns = current->nsproxy->cgroup_ns; + struct dentry *dentry; + int ret; + + get_cgroup_ns(ns); + + /* Check if the caller has permission to mount. */ + if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN)) { + put_cgroup_ns(ns); + return ERR_PTR(-EPERM); + } + /* - * If @pinned_sb, we're reusing an existing root and holding an - * extra ref on its sb. Mount is complete. Put the extra ref. + * The first time anyone tries to mount a cgroup, enable the list + * linking each css_set to its tasks and fix up all existing tasks. */ - if (pinned_sb) { - WARN_ON(new_sb); - deactivate_super(pinned_sb); + if (!use_task_css_set_links) + cgroup_enable_task_cg_lists(); + + if (fs_type == &cgroup2_fs_type) { + unsigned int root_flags; + + ret = parse_cgroup_root_flags(data, &root_flags); + if (ret) { + put_cgroup_ns(ns); + return ERR_PTR(ret); + } + + cgrp_dfl_visible = true; + cgroup_get_live(&cgrp_dfl_root.cgrp); + + dentry = cgroup_do_mount(&cgroup2_fs_type, flags, &cgrp_dfl_root, + CGROUP2_SUPER_MAGIC, ns); + if (!IS_ERR(dentry)) + apply_cgroup_root_flags(root_flags); + } else { + dentry = cgroup1_mount(&cgroup_fs_type, flags, data, + CGROUP_SUPER_MAGIC, ns); } put_cgroup_ns(ns); @@ -2311,7 +1899,7 @@ static void cgroup_kill_sb(struct super_block *sb) kernfs_kill_sb(sb); } -static struct file_system_type cgroup_fs_type = { +struct file_system_type cgroup_fs_type = { .name = "cgroup", .mount = cgroup_mount, .kill_sb = cgroup_kill_sb, @@ -2325,8 +1913,8 @@ static struct file_system_type cgroup2_fs_type = { .fs_flags = FS_USERNS_MOUNT, }; -static int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen, - struct cgroup_namespace *ns) +int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen, + struct cgroup_namespace *ns) { struct cgroup *root = cset_cgroup_from_root(ns->root_cset, cgrp->root); @@ -2389,49 +1977,18 @@ int task_cgroup_path(struct task_struct *task, char *buf, size_t buflen) } EXPORT_SYMBOL_GPL(task_cgroup_path); -/* used to track tasks and other necessary states during migration */ -struct cgroup_taskset { - /* the src and dst cset list running through cset->mg_node */ - struct list_head src_csets; - struct list_head dst_csets; - - /* the subsys currently being processed */ - int ssid; - - /* - * Fields for cgroup_taskset_*() iteration. - * - * Before migration is committed, the target migration tasks are on - * ->mg_tasks of the csets on ->src_csets. After, on ->mg_tasks of - * the csets on ->dst_csets. ->csets point to either ->src_csets - * or ->dst_csets depending on whether migration is committed. - * - * ->cur_csets and ->cur_task point to the current task position - * during iteration. - */ - struct list_head *csets; - struct css_set *cur_cset; - struct task_struct *cur_task; -}; - -#define CGROUP_TASKSET_INIT(tset) (struct cgroup_taskset){ \ - .src_csets = LIST_HEAD_INIT(tset.src_csets), \ - .dst_csets = LIST_HEAD_INIT(tset.dst_csets), \ - .csets = &tset.src_csets, \ -} - /** - * cgroup_taskset_add - try to add a migration target task to a taskset + * cgroup_migrate_add_task - add a migration target task to a migration context * @task: target task - * @tset: target taskset + * @mgctx: target migration context * - * Add @task, which is a migration target, to @tset. This function becomes - * noop if @task doesn't need to be migrated. @task's css_set should have - * been added as a migration source and @task->cg_list will be moved from - * the css_set's tasks list to mg_tasks one. + * Add @task, which is a migration target, to @mgctx->tset. This function + * becomes noop if @task doesn't need to be migrated. @task's css_set + * should have been added as a migration source and @task->cg_list will be + * moved from the css_set's tasks list to mg_tasks one. */ -static void cgroup_taskset_add(struct task_struct *task, - struct cgroup_taskset *tset) +static void cgroup_migrate_add_task(struct task_struct *task, + struct cgroup_mgctx *mgctx) { struct css_set *cset; @@ -2451,10 +2008,11 @@ static void cgroup_taskset_add(struct task_struct *task, list_move_tail(&task->cg_list, &cset->mg_tasks); if (list_empty(&cset->mg_node)) - list_add_tail(&cset->mg_node, &tset->src_csets); + list_add_tail(&cset->mg_node, + &mgctx->tset.src_csets); if (list_empty(&cset->mg_dst_cset->mg_node)) - list_move_tail(&cset->mg_dst_cset->mg_node, - &tset->dst_csets); + list_add_tail(&cset->mg_dst_cset->mg_node, + &mgctx->tset.dst_csets); } /** @@ -2521,17 +2079,16 @@ struct task_struct *cgroup_taskset_next(struct cgroup_taskset *tset, /** * cgroup_taskset_migrate - migrate a taskset - * @tset: taget taskset - * @root: cgroup root the migration is taking place on + * @mgctx: migration context * - * Migrate tasks in @tset as setup by migration preparation functions. + * Migrate tasks in @mgctx as setup by migration preparation functions. * This function fails iff one of the ->can_attach callbacks fails and - * guarantees that either all or none of the tasks in @tset are migrated. - * @tset is consumed regardless of success. + * guarantees that either all or none of the tasks in @mgctx are migrated. + * @mgctx is consumed regardless of success. */ -static int cgroup_taskset_migrate(struct cgroup_taskset *tset, - struct cgroup_root *root) +static int cgroup_migrate_execute(struct cgroup_mgctx *mgctx) { + struct cgroup_taskset *tset = &mgctx->tset; struct cgroup_subsys *ss; struct task_struct *task, *tmp_task; struct css_set *cset, *tmp_cset; @@ -2542,7 +2099,7 @@ static int cgroup_taskset_migrate(struct cgroup_taskset *tset, return 0; /* check that we can legitimately attach to the cgroup */ - do_each_subsys_mask(ss, ssid, root->subsys_mask) { + do_each_subsys_mask(ss, ssid, mgctx->ss_mask) { if (ss->can_attach) { tset->ssid = ssid; ret = ss->can_attach(tset); @@ -2565,8 +2122,10 @@ static int cgroup_taskset_migrate(struct cgroup_taskset *tset, struct css_set *to_cset = cset->mg_dst_cset; get_css_set(to_cset); + to_cset->nr_tasks++; css_set_move_task(task, from_cset, to_cset, true); put_css_set_locked(from_cset); + from_cset->nr_tasks--; } } spin_unlock_irq(&css_set_lock); @@ -2578,7 +2137,7 @@ static int cgroup_taskset_migrate(struct cgroup_taskset *tset, */ tset->csets = &tset->dst_csets; - do_each_subsys_mask(ss, ssid, root->subsys_mask) { + do_each_subsys_mask(ss, ssid, mgctx->ss_mask) { if (ss->attach) { tset->ssid = ssid; ss->attach(tset); @@ -2589,7 +2148,7 @@ static int cgroup_taskset_migrate(struct cgroup_taskset *tset, goto out_release_tset; out_cancel_attach: - do_each_subsys_mask(ss, ssid, root->subsys_mask) { + do_each_subsys_mask(ss, ssid, mgctx->ss_mask) { if (ssid == failed_ssid) break; if (ss->cancel_attach) { @@ -2616,7 +2175,7 @@ out_release_tset: * zero for migration destination cgroups with tasks so that child cgroups * don't compete against tasks. */ -static bool cgroup_may_migrate_to(struct cgroup *dst_cgrp) +bool cgroup_may_migrate_to(struct cgroup *dst_cgrp) { return !cgroup_on_dfl(dst_cgrp) || !cgroup_parent(dst_cgrp) || !dst_cgrp->subtree_control; @@ -2624,25 +2183,31 @@ static bool cgroup_may_migrate_to(struct cgroup *dst_cgrp) /** * cgroup_migrate_finish - cleanup after attach - * @preloaded_csets: list of preloaded css_sets + * @mgctx: migration context * * Undo cgroup_migrate_add_src() and cgroup_migrate_prepare_dst(). See * those functions for details. */ -static void cgroup_migrate_finish(struct list_head *preloaded_csets) +void cgroup_migrate_finish(struct cgroup_mgctx *mgctx) { + LIST_HEAD(preloaded); struct css_set *cset, *tmp_cset; lockdep_assert_held(&cgroup_mutex); spin_lock_irq(&css_set_lock); - list_for_each_entry_safe(cset, tmp_cset, preloaded_csets, mg_preload_node) { + + list_splice_tail_init(&mgctx->preloaded_src_csets, &preloaded); + list_splice_tail_init(&mgctx->preloaded_dst_csets, &preloaded); + + list_for_each_entry_safe(cset, tmp_cset, &preloaded, mg_preload_node) { cset->mg_src_cgrp = NULL; cset->mg_dst_cgrp = NULL; cset->mg_dst_cset = NULL; list_del_init(&cset->mg_preload_node); put_css_set_locked(cset); } + spin_unlock_irq(&css_set_lock); } @@ -2650,10 +2215,10 @@ static void cgroup_migrate_finish(struct list_head *preloaded_csets) * cgroup_migrate_add_src - add a migration source css_set * @src_cset: the source css_set to add * @dst_cgrp: the destination cgroup - * @preloaded_csets: list of preloaded css_sets + * @mgctx: migration context * * Tasks belonging to @src_cset are about to be migrated to @dst_cgrp. Pin - * @src_cset and add it to @preloaded_csets, which should later be cleaned + * @src_cset and add it to @mgctx->src_csets, which should later be cleaned * up by cgroup_migrate_finish(). * * This function may be called without holding cgroup_threadgroup_rwsem @@ -2662,9 +2227,9 @@ static void cgroup_migrate_finish(struct list_head *preloaded_csets) * into play and the preloaded css_sets are guaranteed to cover all * migrations. */ -static void cgroup_migrate_add_src(struct css_set *src_cset, - struct cgroup *dst_cgrp, - struct list_head *preloaded_csets) +void cgroup_migrate_add_src(struct css_set *src_cset, + struct cgroup *dst_cgrp, + struct cgroup_mgctx *mgctx) { struct cgroup *src_cgrp; @@ -2692,33 +2257,35 @@ static void cgroup_migrate_add_src(struct css_set *src_cset, src_cset->mg_src_cgrp = src_cgrp; src_cset->mg_dst_cgrp = dst_cgrp; get_css_set(src_cset); - list_add(&src_cset->mg_preload_node, preloaded_csets); + list_add_tail(&src_cset->mg_preload_node, &mgctx->preloaded_src_csets); } /** * cgroup_migrate_prepare_dst - prepare destination css_sets for migration - * @preloaded_csets: list of preloaded source css_sets + * @mgctx: migration context * * Tasks are about to be moved and all the source css_sets have been - * preloaded to @preloaded_csets. This function looks up and pins all - * destination css_sets, links each to its source, and append them to - * @preloaded_csets. + * preloaded to @mgctx->preloaded_src_csets. This function looks up and + * pins all destination css_sets, links each to its source, and append them + * to @mgctx->preloaded_dst_csets. * * This function must be called after cgroup_migrate_add_src() has been * called on each migration source css_set. After migration is performed * using cgroup_migrate(), cgroup_migrate_finish() must be called on - * @preloaded_csets. + * @mgctx. */ -static int cgroup_migrate_prepare_dst(struct list_head *preloaded_csets) +int cgroup_migrate_prepare_dst(struct cgroup_mgctx *mgctx) { - LIST_HEAD(csets); struct css_set *src_cset, *tmp_cset; lockdep_assert_held(&cgroup_mutex); /* look up the dst cset for each src cset and link it to src */ - list_for_each_entry_safe(src_cset, tmp_cset, preloaded_csets, mg_preload_node) { + list_for_each_entry_safe(src_cset, tmp_cset, &mgctx->preloaded_src_csets, + mg_preload_node) { struct css_set *dst_cset; + struct cgroup_subsys *ss; + int ssid; dst_cset = find_css_set(src_cset, src_cset->mg_dst_cgrp); if (!dst_cset) @@ -2743,15 +2310,19 @@ static int cgroup_migrate_prepare_dst(struct list_head *preloaded_csets) src_cset->mg_dst_cset = dst_cset; if (list_empty(&dst_cset->mg_preload_node)) - list_add(&dst_cset->mg_preload_node, &csets); + list_add_tail(&dst_cset->mg_preload_node, + &mgctx->preloaded_dst_csets); else put_css_set(dst_cset); + + for_each_subsys(ss, ssid) + if (src_cset->subsys[ssid] != dst_cset->subsys[ssid]) + mgctx->ss_mask |= 1 << ssid; } - list_splice_tail(&csets, preloaded_csets); return 0; err: - cgroup_migrate_finish(&csets); + cgroup_migrate_finish(mgctx); return -ENOMEM; } @@ -2759,7 +2330,7 @@ err: * cgroup_migrate - migrate a process or task to a cgroup * @leader: the leader of the process or the task to migrate * @threadgroup: whether @leader points to the whole process or a single task - * @root: cgroup root migration is taking place on + * @mgctx: migration context * * Migrate a process or task denoted by @leader. If migrating a process, * the caller must be holding cgroup_threadgroup_rwsem. The caller is also @@ -2773,10 +2344,9 @@ err: * decided for all targets by invoking group_migrate_prepare_dst() before * actually starting migrating. */ -static int cgroup_migrate(struct task_struct *leader, bool threadgroup, - struct cgroup_root *root) +int cgroup_migrate(struct task_struct *leader, bool threadgroup, + struct cgroup_mgctx *mgctx) { - struct cgroup_taskset tset = CGROUP_TASKSET_INIT(tset); struct task_struct *task; /* @@ -2788,14 +2358,14 @@ static int cgroup_migrate(struct task_struct *leader, bool threadgroup, rcu_read_lock(); task = leader; do { - cgroup_taskset_add(task, &tset); + cgroup_migrate_add_task(task, mgctx); if (!threadgroup) break; } while_each_thread(leader, task); rcu_read_unlock(); spin_unlock_irq(&css_set_lock); - return cgroup_taskset_migrate(&tset, root); + return cgroup_migrate_execute(mgctx); } /** @@ -2806,10 +2376,10 @@ static int cgroup_migrate(struct task_struct *leader, bool threadgroup, * * Call holding cgroup_mutex and cgroup_threadgroup_rwsem. */ -static int cgroup_attach_task(struct cgroup *dst_cgrp, - struct task_struct *leader, bool threadgroup) +int cgroup_attach_task(struct cgroup *dst_cgrp, struct task_struct *leader, + bool threadgroup) { - LIST_HEAD(preloaded_csets); + DEFINE_CGROUP_MGCTX(mgctx); struct task_struct *task; int ret; @@ -2821,8 +2391,7 @@ static int cgroup_attach_task(struct cgroup *dst_cgrp, rcu_read_lock(); task = leader; do { - cgroup_migrate_add_src(task_css_set(task), dst_cgrp, - &preloaded_csets); + cgroup_migrate_add_src(task_css_set(task), dst_cgrp, &mgctx); if (!threadgroup) break; } while_each_thread(leader, task); @@ -2830,11 +2399,11 @@ static int cgroup_attach_task(struct cgroup *dst_cgrp, spin_unlock_irq(&css_set_lock); /* prepare dst csets and commit */ - ret = cgroup_migrate_prepare_dst(&preloaded_csets); + ret = cgroup_migrate_prepare_dst(&mgctx); if (!ret) - ret = cgroup_migrate(leader, threadgroup, dst_cgrp->root); + ret = cgroup_migrate(leader, threadgroup, &mgctx); - cgroup_migrate_finish(&preloaded_csets); + cgroup_migrate_finish(&mgctx); if (!ret) trace_cgroup_attach_task(dst_cgrp, leader, threadgroup); @@ -2846,41 +2415,62 @@ static int cgroup_procs_write_permission(struct task_struct *task, struct cgroup *dst_cgrp, struct kernfs_open_file *of) { - const struct cred *cred = current_cred(); - const struct cred *tcred = get_task_cred(task); - int ret = 0; + struct super_block *sb = of->file->f_path.dentry->d_sb; + struct cgroup_namespace *ns = current->nsproxy->cgroup_ns; + struct cgroup *root_cgrp = ns->root_cset->dfl_cgrp; + struct cgroup *src_cgrp, *com_cgrp; + struct inode *inode; + int ret; - /* - * even if we're attaching all tasks in the thread group, we only - * need to check permissions on one of them. - */ - if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) && - !uid_eq(cred->euid, tcred->uid) && - !uid_eq(cred->euid, tcred->suid)) - ret = -EACCES; + if (!cgroup_on_dfl(dst_cgrp)) { + const struct cred *cred = current_cred(); + const struct cred *tcred = get_task_cred(task); - if (!ret && cgroup_on_dfl(dst_cgrp)) { - struct super_block *sb = of->file->f_path.dentry->d_sb; - struct cgroup *cgrp; - struct inode *inode; + /* + * even if we're attaching all tasks in the thread group, + * we only need to check permissions on one of them. + */ + if (uid_eq(cred->euid, GLOBAL_ROOT_UID) || + uid_eq(cred->euid, tcred->uid) || + uid_eq(cred->euid, tcred->suid)) + ret = 0; + else + ret = -EACCES; - spin_lock_irq(&css_set_lock); - cgrp = task_cgroup_from_root(task, &cgrp_dfl_root); - spin_unlock_irq(&css_set_lock); + put_cred(tcred); + return ret; + } - while (!cgroup_is_descendant(dst_cgrp, cgrp)) - cgrp = cgroup_parent(cgrp); + /* find the source cgroup */ + spin_lock_irq(&css_set_lock); + src_cgrp = task_cgroup_from_root(task, &cgrp_dfl_root); + spin_unlock_irq(&css_set_lock); - ret = -ENOMEM; - inode = kernfs_get_inode(sb, cgrp->procs_file.kn); - if (inode) { - ret = inode_permission(inode, MAY_WRITE); - iput(inode); - } - } + /* and the common ancestor */ + com_cgrp = src_cgrp; + while (!cgroup_is_descendant(dst_cgrp, com_cgrp)) + com_cgrp = cgroup_parent(com_cgrp); - put_cred(tcred); - return ret; + /* %current should be authorized to migrate to the common ancestor */ + inode = kernfs_get_inode(sb, com_cgrp->procs_file.kn); + if (!inode) + return -ENOMEM; + + ret = inode_permission(inode, MAY_WRITE); + iput(inode); + if (ret) + return ret; + + /* + * If namespaces are delegation boundaries, %current must be able + * to see both source and destination cgroups from its namespace. + */ + if ((cgrp_dfl_root.flags & CGRP_ROOT_NS_DELEGATE) && + (!cgroup_is_descendant(src_cgrp, root_cgrp) || + !cgroup_is_descendant(dst_cgrp, root_cgrp))) + return -ENOENT; + + return 0; } /* @@ -2888,8 +2478,8 @@ static int cgroup_procs_write_permission(struct task_struct *task, * function to attach either it or all tasks in its threadgroup. Will lock * cgroup_mutex and threadgroup. */ -static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, - size_t nbytes, loff_t off, bool threadgroup) +ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, + size_t nbytes, loff_t off, bool threadgroup) { struct task_struct *tsk; struct cgroup_subsys *ss; @@ -2920,11 +2510,12 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, tsk = tsk->group_leader; /* - * Workqueue threads may acquire PF_NO_SETAFFINITY and become - * trapped in a cpuset, or RT worker may be born in a cgroup - * with no rt_runtime allocated. Just say no. + * kthreads may acquire PF_NO_SETAFFINITY during initialization. + * If userland migrates such a kthread to a non-root cgroup, it can + * become trapped in a cpuset, or RT kthread may be born in a + * cgroup with no rt_runtime allocated. Just say no. */ - if (tsk == kthreadd_task || (tsk->flags & PF_NO_SETAFFINITY)) { + if (tsk->no_cgroup_migration || (tsk->flags & PF_NO_SETAFFINITY)) { ret = -EINVAL; goto out_unlock_rcu; } @@ -2950,86 +2541,12 @@ out_unlock_threadgroup: return ret ?: nbytes; } -/** - * cgroup_attach_task_all - attach task 'tsk' to all cgroups of task 'from' - * @from: attach to all cgroups of a given task - * @tsk: the task to be attached - */ -int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) -{ - struct cgroup_root *root; - int retval = 0; - - mutex_lock(&cgroup_mutex); - percpu_down_write(&cgroup_threadgroup_rwsem); - for_each_root(root) { - struct cgroup *from_cgrp; - - if (root == &cgrp_dfl_root) - continue; - - spin_lock_irq(&css_set_lock); - from_cgrp = task_cgroup_from_root(from, root); - spin_unlock_irq(&css_set_lock); - - retval = cgroup_attach_task(from_cgrp, tsk, false); - if (retval) - break; - } - percpu_up_write(&cgroup_threadgroup_rwsem); - mutex_unlock(&cgroup_mutex); - - return retval; -} -EXPORT_SYMBOL_GPL(cgroup_attach_task_all); - -static ssize_t cgroup_tasks_write(struct kernfs_open_file *of, - char *buf, size_t nbytes, loff_t off) -{ - return __cgroup_procs_write(of, buf, nbytes, off, false); -} - -static ssize_t cgroup_procs_write(struct kernfs_open_file *of, - char *buf, size_t nbytes, loff_t off) +ssize_t cgroup_procs_write(struct kernfs_open_file *of, char *buf, size_t nbytes, + loff_t off) { return __cgroup_procs_write(of, buf, nbytes, off, true); } -static ssize_t cgroup_release_agent_write(struct kernfs_open_file *of, - char *buf, size_t nbytes, loff_t off) -{ - struct cgroup *cgrp; - - BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX); - - cgrp = cgroup_kn_lock_live(of->kn, false); - if (!cgrp) - return -ENODEV; - spin_lock(&release_agent_path_lock); - strlcpy(cgrp->root->release_agent_path, strstrip(buf), - sizeof(cgrp->root->release_agent_path)); - spin_unlock(&release_agent_path_lock); - cgroup_kn_unlock(of->kn); - return nbytes; -} - -static int cgroup_release_agent_show(struct seq_file *seq, void *v) -{ - struct cgroup *cgrp = seq_css(seq)->cgroup; - - spin_lock(&release_agent_path_lock); - seq_puts(seq, cgrp->root->release_agent_path); - spin_unlock(&release_agent_path_lock); - seq_putc(seq, '\n'); - return 0; -} - -static int cgroup_sane_behavior_show(struct seq_file *seq, void *v) -{ - seq_puts(seq, "0\n"); - return 0; -} - static void cgroup_print_ss_mask(struct seq_file *seq, u16 ss_mask) { struct cgroup_subsys *ss; @@ -3075,8 +2592,7 @@ static int cgroup_subtree_control_show(struct seq_file *seq, void *v) */ static int cgroup_update_dfl_csses(struct cgroup *cgrp) { - LIST_HEAD(preloaded_csets); - struct cgroup_taskset tset = CGROUP_TASKSET_INIT(tset); + DEFINE_CGROUP_MGCTX(mgctx); struct cgroup_subsys_state *d_css; struct cgroup *dsct; struct css_set *src_cset; @@ -3092,33 +2608,28 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) struct cgrp_cset_link *link; list_for_each_entry(link, &dsct->cset_links, cset_link) - cgroup_migrate_add_src(link->cset, dsct, - &preloaded_csets); + cgroup_migrate_add_src(link->cset, dsct, &mgctx); } spin_unlock_irq(&css_set_lock); /* NULL dst indicates self on default hierarchy */ - ret = cgroup_migrate_prepare_dst(&preloaded_csets); + ret = cgroup_migrate_prepare_dst(&mgctx); if (ret) goto out_finish; spin_lock_irq(&css_set_lock); - list_for_each_entry(src_cset, &preloaded_csets, mg_preload_node) { + list_for_each_entry(src_cset, &mgctx.preloaded_src_csets, mg_preload_node) { struct task_struct *task, *ntask; - /* src_csets precede dst_csets, break on the first dst_cset */ - if (!src_cset->mg_src_cgrp) - break; - /* all tasks in src_csets need to be migrated */ list_for_each_entry_safe(task, ntask, &src_cset->tasks, cg_list) - cgroup_taskset_add(task, &tset); + cgroup_migrate_add_task(task, &mgctx); } spin_unlock_irq(&css_set_lock); - ret = cgroup_taskset_migrate(&tset, cgrp->root); + ret = cgroup_migrate_execute(&mgctx); out_finish: - cgroup_migrate_finish(&preloaded_csets); + cgroup_migrate_finish(&mgctx); percpu_up_write(&cgroup_threadgroup_rwsem); return ret; } @@ -3131,7 +2642,7 @@ out_finish: * controller while the previous css is still around. This function grabs * cgroup_mutex and drains the previous css instances of @cgrp's subtree. */ -static void cgroup_lock_and_drain_offline(struct cgroup *cgrp) +void cgroup_lock_and_drain_offline(struct cgroup *cgrp) __acquires(&cgroup_mutex) { struct cgroup *dsct; @@ -3150,7 +2661,7 @@ restart: if (!css || !percpu_ref_is_dying(&css->refcnt)) continue; - cgroup_get(dsct); + cgroup_get_live(dsct); prepare_to_wait(&dsct->offline_waitq, &wait, TASK_UNINTERRUPTIBLE); @@ -3244,7 +2755,7 @@ static bool css_visible(struct cgroup_subsys_state *css) * * Returns 0 on success, -errno on failure. On failure, csses which have * been processed already aren't cleaned up. The caller is responsible for - * cleaning up with cgroup_apply_control_disble(). + * cleaning up with cgroup_apply_control_disable(). */ static int cgroup_apply_control_enable(struct cgroup *cgrp) { @@ -3503,14 +3014,43 @@ static int cgroup_events_show(struct seq_file *seq, void *v) return 0; } +static int cgroup_file_open(struct kernfs_open_file *of) +{ + struct cftype *cft = of->kn->priv; + + if (cft->open) + return cft->open(of); + return 0; +} + +static void cgroup_file_release(struct kernfs_open_file *of) +{ + struct cftype *cft = of->kn->priv; + + if (cft->release) + cft->release(of); +} + static ssize_t cgroup_file_write(struct kernfs_open_file *of, char *buf, size_t nbytes, loff_t off) { + struct cgroup_namespace *ns = current->nsproxy->cgroup_ns; struct cgroup *cgrp = of->kn->parent->priv; struct cftype *cft = of->kn->priv; struct cgroup_subsys_state *css; int ret; + /* + * If namespaces are delegation boundaries, disallow writes to + * files in an non-init namespace root from inside the namespace + * except for the files explicitly marked delegatable - + * cgroup.procs and cgroup.subtree_control. + */ + if ((cgrp->root->flags & CGRP_ROOT_NS_DELEGATE) && + !(cft->flags & CFTYPE_NS_DELEGATABLE) && + ns != &init_cgroup_ns && ns->root_cset->dfl_cgrp == cgrp) + return -EPERM; + if (cft->write) return cft->write(of, buf, nbytes, off); @@ -3553,7 +3093,8 @@ static void *cgroup_seqfile_next(struct seq_file *seq, void *v, loff_t *ppos) static void cgroup_seqfile_stop(struct seq_file *seq, void *v) { - seq_cft(seq)->seq_stop(seq, v); + if (seq_cft(seq)->seq_stop) + seq_cft(seq)->seq_stop(seq, v); } static int cgroup_seqfile_show(struct seq_file *m, void *arg) @@ -3575,12 +3116,16 @@ static int cgroup_seqfile_show(struct seq_file *m, void *arg) static struct kernfs_ops cgroup_kf_single_ops = { .atomic_write_len = PAGE_SIZE, + .open = cgroup_file_open, + .release = cgroup_file_release, .write = cgroup_file_write, .seq_show = cgroup_seqfile_show, }; static struct kernfs_ops cgroup_kf_ops = { .atomic_write_len = PAGE_SIZE, + .open = cgroup_file_open, + .release = cgroup_file_release, .write = cgroup_file_write, .seq_start = cgroup_seqfile_start, .seq_next = cgroup_seqfile_next, @@ -3588,48 +3133,6 @@ static struct kernfs_ops cgroup_kf_ops = { .seq_show = cgroup_seqfile_show, }; -/* - * cgroup_rename - Only allow simple rename of directories in place. - */ -static int cgroup_rename(struct kernfs_node *kn, struct kernfs_node *new_parent, - const char *new_name_str) -{ - struct cgroup *cgrp = kn->priv; - int ret; - - if (kernfs_type(kn) != KERNFS_DIR) - return -ENOTDIR; - if (kn->parent != new_parent) - return -EIO; - - /* - * This isn't a proper migration and its usefulness is very - * limited. Disallow on the default hierarchy. - */ - if (cgroup_on_dfl(cgrp)) - return -EPERM; - - /* - * We're gonna grab cgroup_mutex which nests outside kernfs - * active_ref. kernfs_rename() doesn't require active_ref - * protection. Break them before grabbing cgroup_mutex. - */ - kernfs_break_active_protection(new_parent); - kernfs_break_active_protection(kn); - - mutex_lock(&cgroup_mutex); - - ret = kernfs_rename(kn, new_parent, new_name_str); - if (!ret) - trace_cgroup_rename(cgrp); - - mutex_unlock(&cgroup_mutex); - - kernfs_unbreak_active_protection(kn); - kernfs_unbreak_active_protection(new_parent); - return ret; -} - /* set uid and gid of cgroup dirs and files to that of the creator */ static int cgroup_kn_set_ugid(struct kernfs_node *kn) { @@ -3926,26 +3429,6 @@ void cgroup_file_notify(struct cgroup_file *cfile) } /** - * cgroup_task_count - count the number of tasks in a cgroup. - * @cgrp: the cgroup in question - * - * Return the number of tasks in the cgroup. The returned number can be - * higher than the actual number of tasks due to css_set references from - * namespace roots and temporary usages. - */ -static int cgroup_task_count(const struct cgroup *cgrp) -{ - int count = 0; - struct cgrp_cset_link *link; - - spin_lock_irq(&css_set_lock); - list_for_each_entry(link, &cgrp->cset_links, cset_link) - count += atomic_read(&link->cset->refcount); - spin_unlock_irq(&css_set_lock); - return count; -} - -/** * css_next_child - find the next child of a given css * @pos: the current position (%NULL to initiate traversal) * @parent: css whose children to walk @@ -4343,560 +3826,70 @@ void css_task_iter_end(struct css_task_iter *it) put_task_struct(it->cur_task); } -/** - * cgroup_trasnsfer_tasks - move tasks from one cgroup to another - * @to: cgroup to which the tasks will be moved - * @from: cgroup in which the tasks currently reside - * - * Locking rules between cgroup_post_fork() and the migration path - * guarantee that, if a task is forking while being migrated, the new child - * is guaranteed to be either visible in the source cgroup after the - * parent's migration is complete or put into the target cgroup. No task - * can slip out of migration through forking. - */ -int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from) +static void cgroup_procs_release(struct kernfs_open_file *of) { - LIST_HEAD(preloaded_csets); - struct cgrp_cset_link *link; - struct css_task_iter it; - struct task_struct *task; - int ret; - - if (!cgroup_may_migrate_to(to)) - return -EBUSY; - - mutex_lock(&cgroup_mutex); - - percpu_down_write(&cgroup_threadgroup_rwsem); - - /* all tasks in @from are being moved, all csets are source */ - spin_lock_irq(&css_set_lock); - list_for_each_entry(link, &from->cset_links, cset_link) - cgroup_migrate_add_src(link->cset, to, &preloaded_csets); - spin_unlock_irq(&css_set_lock); - - ret = cgroup_migrate_prepare_dst(&preloaded_csets); - if (ret) - goto out_err; - - /* - * Migrate tasks one-by-one until @from is empty. This fails iff - * ->can_attach() fails. - */ - do { - css_task_iter_start(&from->self, &it); - task = css_task_iter_next(&it); - if (task) - get_task_struct(task); - css_task_iter_end(&it); - - if (task) { - ret = cgroup_migrate(task, false, to->root); - if (!ret) - trace_cgroup_transfer_tasks(to, task, false); - put_task_struct(task); - } - } while (task && !ret); -out_err: - cgroup_migrate_finish(&preloaded_csets); - percpu_up_write(&cgroup_threadgroup_rwsem); - mutex_unlock(&cgroup_mutex); - return ret; -} - -/* - * Stuff for reading the 'tasks'/'procs' files. - * - * Reading this file can return large amounts of data if a cgroup has - * *lots* of attached tasks. So it may need several calls to read(), - * but we cannot guarantee that the information we produce is correct - * unless we produce it entirely atomically. - * - */ - -/* which pidlist file are we talking about? */ -enum cgroup_filetype { - CGROUP_FILE_PROCS, - CGROUP_FILE_TASKS, -}; - -/* - * A pidlist is a list of pids that virtually represents the contents of one - * of the cgroup files ("procs" or "tasks"). We keep a list of such pidlists, - * a pair (one each for procs, tasks) for each pid namespace that's relevant - * to the cgroup. - */ -struct cgroup_pidlist { - /* - * used to find which pidlist is wanted. doesn't change as long as - * this particular list stays in the list. - */ - struct { enum cgroup_filetype type; struct pid_namespace *ns; } key; - /* array of xids */ - pid_t *list; - /* how many elements the above list has */ - int length; - /* each of these stored in a list by its cgroup */ - struct list_head links; - /* pointer to the cgroup we belong to, for list removal purposes */ - struct cgroup *owner; - /* for delayed destruction */ - struct delayed_work destroy_dwork; -}; - -/* - * The following two functions "fix" the issue where there are more pids - * than kmalloc will give memory for; in such cases, we use vmalloc/vfree. - * TODO: replace with a kernel-wide solution to this problem - */ -#define PIDLIST_TOO_LARGE(c) ((c) * sizeof(pid_t) > (PAGE_SIZE * 2)) -static void *pidlist_allocate(int count) -{ - if (PIDLIST_TOO_LARGE(count)) - return vmalloc(count * sizeof(pid_t)); - else - return kmalloc(count * sizeof(pid_t), GFP_KERNEL); -} - -static void pidlist_free(void *p) -{ - kvfree(p); -} - -/* - * Used to destroy all pidlists lingering waiting for destroy timer. None - * should be left afterwards. - */ -static void cgroup_pidlist_destroy_all(struct cgroup *cgrp) -{ - struct cgroup_pidlist *l, *tmp_l; - - mutex_lock(&cgrp->pidlist_mutex); - list_for_each_entry_safe(l, tmp_l, &cgrp->pidlists, links) - mod_delayed_work(cgroup_pidlist_destroy_wq, &l->destroy_dwork, 0); - mutex_unlock(&cgrp->pidlist_mutex); - - flush_workqueue(cgroup_pidlist_destroy_wq); - BUG_ON(!list_empty(&cgrp->pidlists)); -} - -static void cgroup_pidlist_destroy_work_fn(struct work_struct *work) -{ - struct delayed_work *dwork = to_delayed_work(work); - struct cgroup_pidlist *l = container_of(dwork, struct cgroup_pidlist, - destroy_dwork); - struct cgroup_pidlist *tofree = NULL; - - mutex_lock(&l->owner->pidlist_mutex); - - /* - * Destroy iff we didn't get queued again. The state won't change - * as destroy_dwork can only be queued while locked. - */ - if (!delayed_work_pending(dwork)) { - list_del(&l->links); - pidlist_free(l->list); - put_pid_ns(l->key.ns); - tofree = l; - } - - mutex_unlock(&l->owner->pidlist_mutex); - kfree(tofree); -} - -/* - * pidlist_uniq - given a kmalloc()ed list, strip out all duplicate entries - * Returns the number of unique elements. - */ -static int pidlist_uniq(pid_t *list, int length) -{ - int src, dest = 1; - - /* - * we presume the 0th element is unique, so i starts at 1. trivial - * edge cases first; no work needs to be done for either - */ - if (length == 0 || length == 1) - return length; - /* src and dest walk down the list; dest counts unique elements */ - for (src = 1; src < length; src++) { - /* find next unique element */ - while (list[src] == list[src-1]) { - src++; - if (src == length) - goto after; - } - /* dest always points to where the next unique element goes */ - list[dest] = list[src]; - dest++; - } -after: - return dest; -} - -/* - * The two pid files - task and cgroup.procs - guaranteed that the result - * is sorted, which forced this whole pidlist fiasco. As pid order is - * different per namespace, each namespace needs differently sorted list, - * making it impossible to use, for example, single rbtree of member tasks - * sorted by task pointer. As pidlists can be fairly large, allocating one - * per open file is dangerous, so cgroup had to implement shared pool of - * pidlists keyed by cgroup and namespace. - * - * All this extra complexity was caused by the original implementation - * committing to an entirely unnecessary property. In the long term, we - * want to do away with it. Explicitly scramble sort order if on the - * default hierarchy so that no such expectation exists in the new - * interface. - * - * Scrambling is done by swapping every two consecutive bits, which is - * non-identity one-to-one mapping which disturbs sort order sufficiently. - */ -static pid_t pid_fry(pid_t pid) -{ - unsigned a = pid & 0x55555555; - unsigned b = pid & 0xAAAAAAAA; - - return (a << 1) | (b >> 1); -} - -static pid_t cgroup_pid_fry(struct cgroup *cgrp, pid_t pid) -{ - if (cgroup_on_dfl(cgrp)) - return pid_fry(pid); - else - return pid; -} - -static int cmppid(const void *a, const void *b) -{ - return *(pid_t *)a - *(pid_t *)b; -} - -static int fried_cmppid(const void *a, const void *b) -{ - return pid_fry(*(pid_t *)a) - pid_fry(*(pid_t *)b); -} - -static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp, - enum cgroup_filetype type) -{ - struct cgroup_pidlist *l; - /* don't need task_nsproxy() if we're looking at ourself */ - struct pid_namespace *ns = task_active_pid_ns(current); - - lockdep_assert_held(&cgrp->pidlist_mutex); - - list_for_each_entry(l, &cgrp->pidlists, links) - if (l->key.type == type && l->key.ns == ns) - return l; - return NULL; -} - -/* - * find the appropriate pidlist for our purpose (given procs vs tasks) - * returns with the lock on that pidlist already held, and takes care - * of the use count, or returns NULL with no locks held if we're out of - * memory. - */ -static struct cgroup_pidlist *cgroup_pidlist_find_create(struct cgroup *cgrp, - enum cgroup_filetype type) -{ - struct cgroup_pidlist *l; - - lockdep_assert_held(&cgrp->pidlist_mutex); - - l = cgroup_pidlist_find(cgrp, type); - if (l) - return l; - - /* entry not found; create a new one */ - l = kzalloc(sizeof(struct cgroup_pidlist), GFP_KERNEL); - if (!l) - return l; - - INIT_DELAYED_WORK(&l->destroy_dwork, cgroup_pidlist_destroy_work_fn); - l->key.type = type; - /* don't need task_nsproxy() if we're looking at ourself */ - l->key.ns = get_pid_ns(task_active_pid_ns(current)); - l->owner = cgrp; - list_add(&l->links, &cgrp->pidlists); - return l; -} - -/* - * Load a cgroup's pidarray with either procs' tgids or tasks' pids - */ -static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, - struct cgroup_pidlist **lp) -{ - pid_t *array; - int length; - int pid, n = 0; /* used for populating the array */ - struct css_task_iter it; - struct task_struct *tsk; - struct cgroup_pidlist *l; - - lockdep_assert_held(&cgrp->pidlist_mutex); - - /* - * If cgroup gets more users after we read count, we won't have - * enough space - tough. This race is indistinguishable to the - * caller from the case that the additional cgroup users didn't - * show up until sometime later on. - */ - length = cgroup_task_count(cgrp); - array = pidlist_allocate(length); - if (!array) - return -ENOMEM; - /* now, populate the array */ - css_task_iter_start(&cgrp->self, &it); - while ((tsk = css_task_iter_next(&it))) { - if (unlikely(n == length)) - break; - /* get tgid or pid for procs or tasks file respectively */ - if (type == CGROUP_FILE_PROCS) - pid = task_tgid_vnr(tsk); - else - pid = task_pid_vnr(tsk); - if (pid > 0) /* make sure to only use valid results */ - array[n++] = pid; - } - css_task_iter_end(&it); - length = n; - /* now sort & (if procs) strip out duplicates */ - if (cgroup_on_dfl(cgrp)) - sort(array, length, sizeof(pid_t), fried_cmppid, NULL); - else - sort(array, length, sizeof(pid_t), cmppid, NULL); - if (type == CGROUP_FILE_PROCS) - length = pidlist_uniq(array, length); - - l = cgroup_pidlist_find_create(cgrp, type); - if (!l) { - pidlist_free(array); - return -ENOMEM; + if (of->priv) { + css_task_iter_end(of->priv); + kfree(of->priv); } - - /* store array, freeing old if necessary */ - pidlist_free(l->list); - l->list = array; - l->length = length; - *lp = l; - return 0; } -/** - * cgroupstats_build - build and fill cgroupstats - * @stats: cgroupstats to fill information into - * @dentry: A dentry entry belonging to the cgroup for which stats have - * been requested. - * - * Build and fill cgroupstats so that taskstats can export it to user - * space. - */ -int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) +static void *cgroup_procs_next(struct seq_file *s, void *v, loff_t *pos) { - struct kernfs_node *kn = kernfs_node_from_dentry(dentry); - struct cgroup *cgrp; - struct css_task_iter it; - struct task_struct *tsk; - - /* it should be kernfs_node belonging to cgroupfs and is a directory */ - if (dentry->d_sb->s_type != &cgroup_fs_type || !kn || - kernfs_type(kn) != KERNFS_DIR) - return -EINVAL; - - mutex_lock(&cgroup_mutex); - - /* - * We aren't being called from kernfs and there's no guarantee on - * @kn->priv's validity. For this and css_tryget_online_from_dir(), - * @kn->priv is RCU safe. Let's do the RCU dancing. - */ - rcu_read_lock(); - cgrp = rcu_dereference(kn->priv); - if (!cgrp || cgroup_is_dead(cgrp)) { - rcu_read_unlock(); - mutex_unlock(&cgroup_mutex); - return -ENOENT; - } - rcu_read_unlock(); + struct kernfs_open_file *of = s->private; + struct css_task_iter *it = of->priv; + struct task_struct *task; - css_task_iter_start(&cgrp->self, &it); - while ((tsk = css_task_iter_next(&it))) { - switch (tsk->state) { - case TASK_RUNNING: - stats->nr_running++; - break; - case TASK_INTERRUPTIBLE: - stats->nr_sleeping++; - break; - case TASK_UNINTERRUPTIBLE: - stats->nr_uninterruptible++; - break; - case TASK_STOPPED: - stats->nr_stopped++; - break; - default: - if (delayacct_is_task_waiting_on_io(tsk)) - stats->nr_io_wait++; - break; - } - } - css_task_iter_end(&it); + do { + task = css_task_iter_next(it); + } while (task && !thread_group_leader(task)); - mutex_unlock(&cgroup_mutex); - return 0; + return task; } - -/* - * seq_file methods for the tasks/procs files. The seq_file position is the - * next pid to display; the seq_file iterator is a pointer to the pid - * in the cgroup->l->list array. - */ - -static void *cgroup_pidlist_start(struct seq_file *s, loff_t *pos) +static void *cgroup_procs_start(struct seq_file *s, loff_t *pos) { - /* - * Initially we receive a position value that corresponds to - * one more than the last pid shown (or 0 on the first call or - * after a seek to the start). Use a binary-search to find the - * next pid to display, if any - */ struct kernfs_open_file *of = s->private; struct cgroup *cgrp = seq_css(s)->cgroup; - struct cgroup_pidlist *l; - enum cgroup_filetype type = seq_cft(s)->private; - int index = 0, pid = *pos; - int *iter, ret; - - mutex_lock(&cgrp->pidlist_mutex); - - /* - * !NULL @of->priv indicates that this isn't the first start() - * after open. If the matching pidlist is around, we can use that. - * Look for it. Note that @of->priv can't be used directly. It - * could already have been destroyed. - */ - if (of->priv) - of->priv = cgroup_pidlist_find(cgrp, type); + struct css_task_iter *it = of->priv; /* - * Either this is the first start() after open or the matching - * pidlist has been destroyed inbetween. Create a new one. + * When a seq_file is seeked, it's always traversed sequentially + * from position 0, so we can simply keep iterating on !0 *pos. */ - if (!of->priv) { - ret = pidlist_array_load(cgrp, type, - (struct cgroup_pidlist **)&of->priv); - if (ret) - return ERR_PTR(ret); - } - l = of->priv; - - if (pid) { - int end = l->length; - - while (index < end) { - int mid = (index + end) / 2; - if (cgroup_pid_fry(cgrp, l->list[mid]) == pid) { - index = mid; - break; - } else if (cgroup_pid_fry(cgrp, l->list[mid]) <= pid) - index = mid + 1; - else - end = mid; - } - } - /* If we're off the end of the array, we're done */ - if (index >= l->length) - return NULL; - /* Update the abstract position to be the actual pid that we found */ - iter = l->list + index; - *pos = cgroup_pid_fry(cgrp, *iter); - return iter; -} - -static void cgroup_pidlist_stop(struct seq_file *s, void *v) -{ - struct kernfs_open_file *of = s->private; - struct cgroup_pidlist *l = of->priv; - - if (l) - mod_delayed_work(cgroup_pidlist_destroy_wq, &l->destroy_dwork, - CGROUP_PIDLIST_DESTROY_DELAY); - mutex_unlock(&seq_css(s)->cgroup->pidlist_mutex); -} + if (!it) { + if (WARN_ON_ONCE((*pos)++)) + return ERR_PTR(-EINVAL); -static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos) -{ - struct kernfs_open_file *of = s->private; - struct cgroup_pidlist *l = of->priv; - pid_t *p = v; - pid_t *end = l->list + l->length; - /* - * Advance to the next pid in the array. If this goes off the - * end, we're done - */ - p++; - if (p >= end) { - return NULL; - } else { - *pos = cgroup_pid_fry(seq_css(s)->cgroup, *p); - return p; + it = kzalloc(sizeof(*it), GFP_KERNEL); + if (!it) + return ERR_PTR(-ENOMEM); + of->priv = it; + css_task_iter_start(&cgrp->self, it); + } else if (!(*pos)++) { + css_task_iter_end(it); + css_task_iter_start(&cgrp->self, it); } -} - -static int cgroup_pidlist_show(struct seq_file *s, void *v) -{ - seq_printf(s, "%d\n", *(int *)v); - - return 0; -} -static u64 cgroup_read_notify_on_release(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return notify_on_release(css->cgroup); + return cgroup_procs_next(s, NULL, NULL); } -static int cgroup_write_notify_on_release(struct cgroup_subsys_state *css, - struct cftype *cft, u64 val) +static int cgroup_procs_show(struct seq_file *s, void *v) { - if (val) - set_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags); - else - clear_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags); - return 0; -} - -static u64 cgroup_clone_children_read(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); -} - -static int cgroup_clone_children_write(struct cgroup_subsys_state *css, - struct cftype *cft, u64 val) -{ - if (val) - set_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); - else - clear_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); + seq_printf(s, "%d\n", task_tgid_vnr(v)); return 0; } /* cgroup core interface files for the default hierarchy */ -static struct cftype cgroup_dfl_base_files[] = { +static struct cftype cgroup_base_files[] = { { .name = "cgroup.procs", + .flags = CFTYPE_NS_DELEGATABLE, .file_offset = offsetof(struct cgroup, procs_file), - .seq_start = cgroup_pidlist_start, - .seq_next = cgroup_pidlist_next, - .seq_stop = cgroup_pidlist_stop, - .seq_show = cgroup_pidlist_show, - .private = CGROUP_FILE_PROCS, + .release = cgroup_procs_release, + .seq_start = cgroup_procs_start, + .seq_next = cgroup_procs_next, + .seq_show = cgroup_procs_show, .write = cgroup_procs_write, }, { @@ -4905,6 +3898,7 @@ static struct cftype cgroup_dfl_base_files[] = { }, { .name = "cgroup.subtree_control", + .flags = CFTYPE_NS_DELEGATABLE, .seq_show = cgroup_subtree_control_show, .write = cgroup_subtree_control_write, }, @@ -4917,51 +3911,6 @@ static struct cftype cgroup_dfl_base_files[] = { { } /* terminate */ }; -/* cgroup core interface files for the legacy hierarchies */ -static struct cftype cgroup_legacy_base_files[] = { - { - .name = "cgroup.procs", - .seq_start = cgroup_pidlist_start, - .seq_next = cgroup_pidlist_next, - .seq_stop = cgroup_pidlist_stop, - .seq_show = cgroup_pidlist_show, - .private = CGROUP_FILE_PROCS, - .write = cgroup_procs_write, - }, - { - .name = "cgroup.clone_children", - .read_u64 = cgroup_clone_children_read, - .write_u64 = cgroup_clone_children_write, - }, - { - .name = "cgroup.sane_behavior", - .flags = CFTYPE_ONLY_ON_ROOT, - .seq_show = cgroup_sane_behavior_show, - }, - { - .name = "tasks", - .seq_start = cgroup_pidlist_start, - .seq_next = cgroup_pidlist_next, - .seq_stop = cgroup_pidlist_stop, - .seq_show = cgroup_pidlist_show, - .private = CGROUP_FILE_TASKS, - .write = cgroup_tasks_write, - }, - { - .name = "notify_on_release", - .read_u64 = cgroup_read_notify_on_release, - .write_u64 = cgroup_write_notify_on_release, - }, - { - .name = "release_agent", - .flags = CFTYPE_ONLY_ON_ROOT, - .seq_show = cgroup_release_agent_show, - .write = cgroup_release_agent_write, - .max_write_len = PATH_MAX - 1, - }, - { } /* terminate */ -}; - /* * css destruction is four-stage process. * @@ -5007,7 +3956,7 @@ static void css_free_work_fn(struct work_struct *work) } else { /* cgroup free path */ atomic_dec(&cgrp->root->nr_cgrps); - cgroup_pidlist_destroy_all(cgrp); + cgroup1_pidlist_destroy_all(cgrp); cancel_work_sync(&cgrp->release_agent_work); if (cgroup_parent(cgrp)) { @@ -5097,7 +4046,7 @@ static void init_and_link_css(struct cgroup_subsys_state *css, { lockdep_assert_held(&cgroup_mutex); - cgroup_get(cgrp); + cgroup_get_live(cgrp); memset(css, 0, sizeof(*css)); css->cgroup = cgrp; @@ -5273,7 +4222,7 @@ static struct cgroup *cgroup_create(struct cgroup *parent) /* allocation complete, commit to creation */ list_add_tail_rcu(&cgrp->self.sibling, &cgroup_parent(cgrp)->self.children); atomic_inc(&root->nr_cgrps); - cgroup_get(parent); + cgroup_get_live(parent); /* * @cgrp is now fully operational. If something fails after this @@ -5302,8 +4251,7 @@ out_free_cgrp: return ERR_PTR(ret); } -static int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, - umode_t mode) +int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name, umode_t mode) { struct cgroup *parent, *cgrp; struct kernfs_node *kn; @@ -5411,6 +4359,11 @@ static void kill_css(struct cgroup_subsys_state *css) { lockdep_assert_held(&cgroup_mutex); + if (css->flags & CSS_DYING) + return; + + css->flags |= CSS_DYING; + /* * This must happen before css is disassociated with its cgroup. * See seq_css() for details. @@ -5507,7 +4460,7 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) */ kernfs_remove(cgrp->kn); - check_for_release(cgroup_parent(cgrp)); + cgroup1_check_for_release(cgroup_parent(cgrp)); /* put the base reference */ percpu_ref_kill(&cgrp->self.refcnt); @@ -5515,7 +4468,7 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) return 0; }; -static int cgroup_rmdir(struct kernfs_node *kn) +int cgroup_rmdir(struct kernfs_node *kn) { struct cgroup *cgrp; int ret = 0; @@ -5534,11 +4487,10 @@ static int cgroup_rmdir(struct kernfs_node *kn) } static struct kernfs_syscall_ops cgroup_kf_syscall_ops = { - .remount_fs = cgroup_remount, .show_options = cgroup_show_options, + .remount_fs = cgroup_remount, .mkdir = cgroup_mkdir, .rmdir = cgroup_rmdir, - .rename = cgroup_rename, .show_path = cgroup_show_path, }; @@ -5646,8 +4598,8 @@ int __init cgroup_init(void) BUILD_BUG_ON(CGROUP_SUBSYS_COUNT > 16); BUG_ON(percpu_init_rwsem(&cgroup_threadgroup_rwsem)); - BUG_ON(cgroup_init_cftypes(NULL, cgroup_dfl_base_files)); - BUG_ON(cgroup_init_cftypes(NULL, cgroup_legacy_base_files)); + BUG_ON(cgroup_init_cftypes(NULL, cgroup_base_files)); + BUG_ON(cgroup_init_cftypes(NULL, cgroup1_base_files)); /* * The latency of the synchronize_sched() is too high for cgroups, @@ -5666,7 +4618,7 @@ int __init cgroup_init(void) hash_add(css_set_table, &init_css_set.hlist, css_set_hash(init_css_set.subsys)); - BUG_ON(cgroup_setup_root(&cgrp_dfl_root, 0)); + BUG_ON(cgroup_setup_root(&cgrp_dfl_root, 0, 0)); mutex_unlock(&cgroup_mutex); @@ -5697,7 +4649,7 @@ int __init cgroup_init(void) continue; } - if (cgroup_ssid_no_v1(ssid)) + if (cgroup1_ssid_disabled(ssid)) printk(KERN_INFO "Disabling %s control group subsystem in v1 mounts\n", ss->name); @@ -5744,15 +4696,6 @@ static int __init cgroup_wq_init(void) */ cgroup_destroy_wq = alloc_workqueue("cgroup_destroy", 0, 1); BUG_ON(!cgroup_destroy_wq); - - /* - * Used to destroy pidlists and separate to serve as flush domain. - * Cap @max_active to 1 too. - */ - cgroup_pidlist_destroy_wq = alloc_workqueue("cgroup_pidlist_destroy", - 0, 1); - BUG_ON(!cgroup_pidlist_destroy_wq); - return 0; } core_initcall(cgroup_wq_init); @@ -5835,42 +4778,6 @@ out: return retval; } -/* Display information about each subsystem and each hierarchy */ -static int proc_cgroupstats_show(struct seq_file *m, void *v) -{ - struct cgroup_subsys *ss; - int i; - - seq_puts(m, "#subsys_name\thierarchy\tnum_cgroups\tenabled\n"); - /* - * ideally we don't want subsystems moving around while we do this. - * cgroup_mutex is also necessary to guarantee an atomic snapshot of - * subsys/hierarchy state. - */ - mutex_lock(&cgroup_mutex); - - for_each_subsys(ss, i) - seq_printf(m, "%s\t%d\t%d\t%d\n", - ss->legacy_name, ss->root->hierarchy_id, - atomic_read(&ss->root->nr_cgrps), - cgroup_ssid_enabled(i)); - - mutex_unlock(&cgroup_mutex); - return 0; -} - -static int cgroupstats_open(struct inode *inode, struct file *file) -{ - return single_open(file, proc_cgroupstats_show, NULL); -} - -static const struct file_operations proc_cgroupstats_operations = { - .open = cgroupstats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - /** * cgroup_fork - initialize cgroup related fields during copy_process() * @child: pointer to task_struct of forking parent process. @@ -5977,6 +4884,7 @@ void cgroup_post_fork(struct task_struct *child) cset = task_css_set(current); if (list_empty(&child->cg_list)) { get_css_set(cset); + cset->nr_tasks++; css_set_move_task(child, NULL, cset, false); } spin_unlock_irq(&css_set_lock); @@ -6026,6 +4934,7 @@ void cgroup_exit(struct task_struct *tsk) if (!list_empty(&tsk->cg_list)) { spin_lock_irq(&css_set_lock); css_set_move_task(tsk, cset, NULL, false); + cset->nr_tasks--; spin_unlock_irq(&css_set_lock); } else { get_css_set(cset); @@ -6050,76 +4959,6 @@ void cgroup_free(struct task_struct *task) put_css_set(cset); } -static void check_for_release(struct cgroup *cgrp) -{ - if (notify_on_release(cgrp) && !cgroup_is_populated(cgrp) && - !css_has_online_children(&cgrp->self) && !cgroup_is_dead(cgrp)) - schedule_work(&cgrp->release_agent_work); -} - -/* - * Notify userspace when a cgroup is released, by running the - * configured release agent with the name of the cgroup (path - * relative to the root of cgroup file system) as the argument. - * - * Most likely, this user command will try to rmdir this cgroup. - * - * This races with the possibility that some other task will be - * attached to this cgroup before it is removed, or that some other - * user task will 'mkdir' a child cgroup of this cgroup. That's ok. - * The presumed 'rmdir' will fail quietly if this cgroup is no longer - * unused, and this cgroup will be reprieved from its death sentence, - * to continue to serve a useful existence. Next time it's released, - * we will get notified again, if it still has 'notify_on_release' set. - * - * The final arg to call_usermodehelper() is UMH_WAIT_EXEC, which - * means only wait until the task is successfully execve()'d. The - * separate release agent task is forked by call_usermodehelper(), - * then control in this thread returns here, without waiting for the - * release agent task. We don't bother to wait because the caller of - * this routine has no use for the exit status of the release agent - * task, so no sense holding our caller up for that. - */ -static void cgroup_release_agent(struct work_struct *work) -{ - struct cgroup *cgrp = - container_of(work, struct cgroup, release_agent_work); - char *pathbuf = NULL, *agentbuf = NULL; - char *argv[3], *envp[3]; - int ret; - - mutex_lock(&cgroup_mutex); - - pathbuf = kmalloc(PATH_MAX, GFP_KERNEL); - agentbuf = kstrdup(cgrp->root->release_agent_path, GFP_KERNEL); - if (!pathbuf || !agentbuf) - goto out; - - spin_lock_irq(&css_set_lock); - ret = cgroup_path_ns_locked(cgrp, pathbuf, PATH_MAX, &init_cgroup_ns); - spin_unlock_irq(&css_set_lock); - if (ret < 0 || ret >= PATH_MAX) - goto out; - - argv[0] = agentbuf; - argv[1] = pathbuf; - argv[2] = NULL; - - /* minimal command environment */ - envp[0] = "HOME=/"; - envp[1] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin"; - envp[2] = NULL; - - mutex_unlock(&cgroup_mutex); - call_usermodehelper(argv[0], argv, envp, UMH_WAIT_EXEC); - goto out_free; -out: - mutex_unlock(&cgroup_mutex); -out_free: - kfree(agentbuf); - kfree(pathbuf); -} - static int __init cgroup_disable(char *str) { struct cgroup_subsys *ss; @@ -6141,33 +4980,6 @@ static int __init cgroup_disable(char *str) } __setup("cgroup_disable=", cgroup_disable); -static int __init cgroup_no_v1(char *str) -{ - struct cgroup_subsys *ss; - char *token; - int i; - - while ((token = strsep(&str, ",")) != NULL) { - if (!*token) - continue; - - if (!strcmp(token, "all")) { - cgroup_no_v1_mask = U16_MAX; - break; - } - - for_each_subsys(ss, i) { - if (strcmp(token, ss->name) && - strcmp(token, ss->legacy_name)) - continue; - - cgroup_no_v1_mask |= 1 << i; - } - } - return 1; -} -__setup("cgroup_no_v1=", cgroup_no_v1); - /** * css_tryget_online_from_dir - get corresponding css from a cgroup dentry * @dentry: directory dentry of interest @@ -6197,7 +5009,7 @@ struct cgroup_subsys_state *css_tryget_online_from_dir(struct dentry *dentry, * have been or be removed at any point. @kn->priv is RCU * protected for this access. See css_release_work_fn() for details. */ - cgrp = rcu_dereference(kn->priv); + cgrp = rcu_dereference(*(void __rcu __force **)&kn->priv); if (cgrp) css = cgroup_css(cgrp, ss); @@ -6242,7 +5054,7 @@ struct cgroup *cgroup_get_from_path(const char *path) if (kn) { if (kernfs_type(kn) == KERNFS_DIR) { cgrp = kn->priv; - cgroup_get(cgrp); + cgroup_get_live(cgrp); } else { cgrp = ERR_PTR(-ENOTDIR); } @@ -6322,6 +5134,11 @@ void cgroup_sk_alloc(struct sock_cgroup_data *skcd) /* Socket clone path */ if (skcd->val) { + /* + * We might be cloning a socket which is left in an empty + * cgroup and the cgroup might have already been rmdir'd. + * Don't use cgroup_get_live(). + */ cgroup_get(sock_cgroup_ptr(skcd)); return; } @@ -6349,309 +5166,16 @@ void cgroup_sk_free(struct sock_cgroup_data *skcd) #endif /* CONFIG_SOCK_CGROUP_DATA */ -/* cgroup namespaces */ - -static struct ucounts *inc_cgroup_namespaces(struct user_namespace *ns) -{ - return inc_ucount(ns, current_euid(), UCOUNT_CGROUP_NAMESPACES); -} - -static void dec_cgroup_namespaces(struct ucounts *ucounts) -{ - dec_ucount(ucounts, UCOUNT_CGROUP_NAMESPACES); -} - -static struct cgroup_namespace *alloc_cgroup_ns(void) -{ - struct cgroup_namespace *new_ns; - int ret; - - new_ns = kzalloc(sizeof(struct cgroup_namespace), GFP_KERNEL); - if (!new_ns) - return ERR_PTR(-ENOMEM); - ret = ns_alloc_inum(&new_ns->ns); - if (ret) { - kfree(new_ns); - return ERR_PTR(ret); - } - atomic_set(&new_ns->count, 1); - new_ns->ns.ops = &cgroupns_operations; - return new_ns; -} - -void free_cgroup_ns(struct cgroup_namespace *ns) -{ - put_css_set(ns->root_cset); - dec_cgroup_namespaces(ns->ucounts); - put_user_ns(ns->user_ns); - ns_free_inum(&ns->ns); - kfree(ns); -} -EXPORT_SYMBOL(free_cgroup_ns); - -struct cgroup_namespace *copy_cgroup_ns(unsigned long flags, - struct user_namespace *user_ns, - struct cgroup_namespace *old_ns) -{ - struct cgroup_namespace *new_ns; - struct ucounts *ucounts; - struct css_set *cset; - - BUG_ON(!old_ns); - - if (!(flags & CLONE_NEWCGROUP)) { - get_cgroup_ns(old_ns); - return old_ns; - } - - /* Allow only sysadmin to create cgroup namespace. */ - if (!ns_capable(user_ns, CAP_SYS_ADMIN)) - return ERR_PTR(-EPERM); - - ucounts = inc_cgroup_namespaces(user_ns); - if (!ucounts) - return ERR_PTR(-ENOSPC); - - /* It is not safe to take cgroup_mutex here */ - spin_lock_irq(&css_set_lock); - cset = task_css_set(current); - get_css_set(cset); - spin_unlock_irq(&css_set_lock); - - new_ns = alloc_cgroup_ns(); - if (IS_ERR(new_ns)) { - put_css_set(cset); - dec_cgroup_namespaces(ucounts); - return new_ns; - } - - new_ns->user_ns = get_user_ns(user_ns); - new_ns->ucounts = ucounts; - new_ns->root_cset = cset; - - return new_ns; -} - -static inline struct cgroup_namespace *to_cg_ns(struct ns_common *ns) -{ - return container_of(ns, struct cgroup_namespace, ns); -} - -static int cgroupns_install(struct nsproxy *nsproxy, struct ns_common *ns) -{ - struct cgroup_namespace *cgroup_ns = to_cg_ns(ns); - - if (!ns_capable(current_user_ns(), CAP_SYS_ADMIN) || - !ns_capable(cgroup_ns->user_ns, CAP_SYS_ADMIN)) - return -EPERM; - - /* Don't need to do anything if we are attaching to our own cgroupns. */ - if (cgroup_ns == nsproxy->cgroup_ns) - return 0; - - get_cgroup_ns(cgroup_ns); - put_cgroup_ns(nsproxy->cgroup_ns); - nsproxy->cgroup_ns = cgroup_ns; - - return 0; -} - -static struct ns_common *cgroupns_get(struct task_struct *task) -{ - struct cgroup_namespace *ns = NULL; - struct nsproxy *nsproxy; - - task_lock(task); - nsproxy = task->nsproxy; - if (nsproxy) { - ns = nsproxy->cgroup_ns; - get_cgroup_ns(ns); - } - task_unlock(task); - - return ns ? &ns->ns : NULL; -} - -static void cgroupns_put(struct ns_common *ns) -{ - put_cgroup_ns(to_cg_ns(ns)); -} - -static struct user_namespace *cgroupns_owner(struct ns_common *ns) -{ - return to_cg_ns(ns)->user_ns; -} - -const struct proc_ns_operations cgroupns_operations = { - .name = "cgroup", - .type = CLONE_NEWCGROUP, - .get = cgroupns_get, - .put = cgroupns_put, - .install = cgroupns_install, - .owner = cgroupns_owner, -}; - -static __init int cgroup_namespaces_init(void) -{ - return 0; -} -subsys_initcall(cgroup_namespaces_init); - #ifdef CONFIG_CGROUP_BPF -void cgroup_bpf_update(struct cgroup *cgrp, - struct bpf_prog *prog, - enum bpf_attach_type type) +int cgroup_bpf_update(struct cgroup *cgrp, struct bpf_prog *prog, + enum bpf_attach_type type, bool overridable) { struct cgroup *parent = cgroup_parent(cgrp); + int ret; mutex_lock(&cgroup_mutex); - __cgroup_bpf_update(cgrp, parent, prog, type); + ret = __cgroup_bpf_update(cgrp, parent, prog, type, overridable); mutex_unlock(&cgroup_mutex); + return ret; } #endif /* CONFIG_CGROUP_BPF */ - -#ifdef CONFIG_CGROUP_DEBUG -static struct cgroup_subsys_state * -debug_css_alloc(struct cgroup_subsys_state *parent_css) -{ - struct cgroup_subsys_state *css = kzalloc(sizeof(*css), GFP_KERNEL); - - if (!css) - return ERR_PTR(-ENOMEM); - - return css; -} - -static void debug_css_free(struct cgroup_subsys_state *css) -{ - kfree(css); -} - -static u64 debug_taskcount_read(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return cgroup_task_count(css->cgroup); -} - -static u64 current_css_set_read(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return (u64)(unsigned long)current->cgroups; -} - -static u64 current_css_set_refcount_read(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - u64 count; - - rcu_read_lock(); - count = atomic_read(&task_css_set(current)->refcount); - rcu_read_unlock(); - return count; -} - -static int current_css_set_cg_links_read(struct seq_file *seq, void *v) -{ - struct cgrp_cset_link *link; - struct css_set *cset; - char *name_buf; - - name_buf = kmalloc(NAME_MAX + 1, GFP_KERNEL); - if (!name_buf) - return -ENOMEM; - - spin_lock_irq(&css_set_lock); - rcu_read_lock(); - cset = rcu_dereference(current->cgroups); - list_for_each_entry(link, &cset->cgrp_links, cgrp_link) { - struct cgroup *c = link->cgrp; - - cgroup_name(c, name_buf, NAME_MAX + 1); - seq_printf(seq, "Root %d group %s\n", - c->root->hierarchy_id, name_buf); - } - rcu_read_unlock(); - spin_unlock_irq(&css_set_lock); - kfree(name_buf); - return 0; -} - -#define MAX_TASKS_SHOWN_PER_CSS 25 -static int cgroup_css_links_read(struct seq_file *seq, void *v) -{ - struct cgroup_subsys_state *css = seq_css(seq); - struct cgrp_cset_link *link; - - spin_lock_irq(&css_set_lock); - list_for_each_entry(link, &css->cgroup->cset_links, cset_link) { - struct css_set *cset = link->cset; - struct task_struct *task; - int count = 0; - - seq_printf(seq, "css_set %p\n", cset); - - list_for_each_entry(task, &cset->tasks, cg_list) { - if (count++ > MAX_TASKS_SHOWN_PER_CSS) - goto overflow; - seq_printf(seq, " task %d\n", task_pid_vnr(task)); - } - - list_for_each_entry(task, &cset->mg_tasks, cg_list) { - if (count++ > MAX_TASKS_SHOWN_PER_CSS) - goto overflow; - seq_printf(seq, " task %d\n", task_pid_vnr(task)); - } - continue; - overflow: - seq_puts(seq, " ...\n"); - } - spin_unlock_irq(&css_set_lock); - return 0; -} - -static u64 releasable_read(struct cgroup_subsys_state *css, struct cftype *cft) -{ - return (!cgroup_is_populated(css->cgroup) && - !css_has_online_children(&css->cgroup->self)); -} - -static struct cftype debug_files[] = { - { - .name = "taskcount", - .read_u64 = debug_taskcount_read, - }, - - { - .name = "current_css_set", - .read_u64 = current_css_set_read, - }, - - { - .name = "current_css_set_refcount", - .read_u64 = current_css_set_refcount_read, - }, - - { - .name = "current_css_set_cg_links", - .seq_show = current_css_set_cg_links_read, - }, - - { - .name = "cgroup_css_links", - .seq_show = cgroup_css_links_read, - }, - - { - .name = "releasable", - .read_u64 = releasable_read, - }, - - { } /* terminate */ -}; - -struct cgroup_subsys debug_cgrp_subsys = { - .css_alloc = debug_css_alloc, - .css_free = debug_css_free, - .legacy_cftypes = debug_files, -}; -#endif /* CONFIG_CGROUP_DEBUG */ diff --git a/kernel/cpuset.c b/kernel/cgroup/cpuset.c index b3088886cd37..ca8376e5008c 100644 --- a/kernel/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -44,6 +44,8 @@ #include <linux/proc_fs.h> #include <linux/rcupdate.h> #include <linux/sched.h> +#include <linux/sched/mm.h> +#include <linux/sched/task.h> #include <linux/seq_file.h> #include <linux/security.h> #include <linux/slab.h> @@ -174,9 +176,9 @@ typedef enum { } cpuset_flagbits_t; /* convenient tests for these bits */ -static inline bool is_cpuset_online(const struct cpuset *cs) +static inline bool is_cpuset_online(struct cpuset *cs) { - return test_bit(CS_ONLINE, &cs->flags); + return test_bit(CS_ONLINE, &cs->flags) && !css_is_dying(&cs->css); } static inline int is_cpu_exclusive(const struct cpuset *cs) @@ -1036,40 +1038,25 @@ static void cpuset_post_attach(void) * @tsk: the task to change * @newmems: new nodes that the task will be set * - * In order to avoid seeing no nodes if the old and new nodes are disjoint, - * we structure updates as setting all new allowed nodes, then clearing newly - * disallowed ones. + * We use the mems_allowed_seq seqlock to safely update both tsk->mems_allowed + * and rebind an eventual tasks' mempolicy. If the task is allocating in + * parallel, it might temporarily see an empty intersection, which results in + * a seqlock check and retry before OOM or allocation failure. */ static void cpuset_change_task_nodemask(struct task_struct *tsk, nodemask_t *newmems) { - bool need_loop; - task_lock(tsk); - /* - * Determine if a loop is necessary if another thread is doing - * read_mems_allowed_begin(). If at least one node remains unchanged and - * tsk does not have a mempolicy, then an empty nodemask will not be - * possible when mems_allowed is larger than a word. - */ - need_loop = task_has_mempolicy(tsk) || - !nodes_intersects(*newmems, tsk->mems_allowed); - if (need_loop) { - local_irq_disable(); - write_seqcount_begin(&tsk->mems_allowed_seq); - } + local_irq_disable(); + write_seqcount_begin(&tsk->mems_allowed_seq); nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); - mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); - - mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP2); + mpol_rebind_task(tsk, newmems); tsk->mems_allowed = *newmems; - if (need_loop) { - write_seqcount_end(&tsk->mems_allowed_seq); - local_irq_enable(); - } + write_seqcount_end(&tsk->mems_allowed_seq); + local_irq_enable(); task_unlock(tsk); } @@ -2119,10 +2106,8 @@ int __init cpuset_init(void) { int err = 0; - if (!alloc_cpumask_var(&top_cpuset.cpus_allowed, GFP_KERNEL)) - BUG(); - if (!alloc_cpumask_var(&top_cpuset.effective_cpus, GFP_KERNEL)) - BUG(); + BUG_ON(!alloc_cpumask_var(&top_cpuset.cpus_allowed, GFP_KERNEL)); + BUG_ON(!alloc_cpumask_var(&top_cpuset.effective_cpus, GFP_KERNEL)); cpumask_setall(top_cpuset.cpus_allowed); nodes_setall(top_cpuset.mems_allowed); @@ -2137,8 +2122,7 @@ int __init cpuset_init(void) if (err < 0) return err; - if (!alloc_cpumask_var(&cpus_attach, GFP_KERNEL)) - BUG(); + BUG_ON(!alloc_cpumask_var(&cpus_attach, GFP_KERNEL)); return 0; } @@ -2352,7 +2336,7 @@ static void cpuset_hotplug_workfn(struct work_struct *work) rebuild_sched_domains(); } -void cpuset_update_active_cpus(bool cpu_online) +void cpuset_update_active_cpus(void) { /* * We're inside cpu hotplug critical region which usually nests diff --git a/kernel/cgroup/debug.c b/kernel/cgroup/debug.c new file mode 100644 index 000000000000..dac46af22782 --- /dev/null +++ b/kernel/cgroup/debug.c @@ -0,0 +1,357 @@ +/* + * Debug controller + * + * WARNING: This controller is for cgroup core debugging only. + * Its interfaces are unstable and subject to changes at any time. + */ +#include <linux/ctype.h> +#include <linux/mm.h> +#include <linux/slab.h> + +#include "cgroup-internal.h" + +static struct cgroup_subsys_state * +debug_css_alloc(struct cgroup_subsys_state *parent_css) +{ + struct cgroup_subsys_state *css = kzalloc(sizeof(*css), GFP_KERNEL); + + if (!css) + return ERR_PTR(-ENOMEM); + + return css; +} + +static void debug_css_free(struct cgroup_subsys_state *css) +{ + kfree(css); +} + +/* + * debug_taskcount_read - return the number of tasks in a cgroup. + * @cgrp: the cgroup in question + */ +static u64 debug_taskcount_read(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + return cgroup_task_count(css->cgroup); +} + +static int current_css_set_read(struct seq_file *seq, void *v) +{ + struct kernfs_open_file *of = seq->private; + struct css_set *cset; + struct cgroup_subsys *ss; + struct cgroup_subsys_state *css; + int i, refcnt; + + if (!cgroup_kn_lock_live(of->kn, false)) + return -ENODEV; + + spin_lock_irq(&css_set_lock); + rcu_read_lock(); + cset = rcu_dereference(current->cgroups); + refcnt = refcount_read(&cset->refcount); + seq_printf(seq, "css_set %pK %d", cset, refcnt); + if (refcnt > cset->nr_tasks) + seq_printf(seq, " +%d", refcnt - cset->nr_tasks); + seq_puts(seq, "\n"); + + /* + * Print the css'es stored in the current css_set. + */ + for_each_subsys(ss, i) { + css = cset->subsys[ss->id]; + if (!css) + continue; + seq_printf(seq, "%2d: %-4s\t- %lx[%d]\n", ss->id, ss->name, + (unsigned long)css, css->id); + } + rcu_read_unlock(); + spin_unlock_irq(&css_set_lock); + cgroup_kn_unlock(of->kn); + return 0; +} + +static u64 current_css_set_refcount_read(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + u64 count; + + rcu_read_lock(); + count = refcount_read(&task_css_set(current)->refcount); + rcu_read_unlock(); + return count; +} + +static int current_css_set_cg_links_read(struct seq_file *seq, void *v) +{ + struct cgrp_cset_link *link; + struct css_set *cset; + char *name_buf; + + name_buf = kmalloc(NAME_MAX + 1, GFP_KERNEL); + if (!name_buf) + return -ENOMEM; + + spin_lock_irq(&css_set_lock); + rcu_read_lock(); + cset = rcu_dereference(current->cgroups); + list_for_each_entry(link, &cset->cgrp_links, cgrp_link) { + struct cgroup *c = link->cgrp; + + cgroup_name(c, name_buf, NAME_MAX + 1); + seq_printf(seq, "Root %d group %s\n", + c->root->hierarchy_id, name_buf); + } + rcu_read_unlock(); + spin_unlock_irq(&css_set_lock); + kfree(name_buf); + return 0; +} + +#define MAX_TASKS_SHOWN_PER_CSS 25 +static int cgroup_css_links_read(struct seq_file *seq, void *v) +{ + struct cgroup_subsys_state *css = seq_css(seq); + struct cgrp_cset_link *link; + int dead_cnt = 0, extra_refs = 0; + + spin_lock_irq(&css_set_lock); + list_for_each_entry(link, &css->cgroup->cset_links, cset_link) { + struct css_set *cset = link->cset; + struct task_struct *task; + int count = 0; + int refcnt = refcount_read(&cset->refcount); + + seq_printf(seq, " %d", refcnt); + if (refcnt - cset->nr_tasks > 0) { + int extra = refcnt - cset->nr_tasks; + + seq_printf(seq, " +%d", extra); + /* + * Take out the one additional reference in + * init_css_set. + */ + if (cset == &init_css_set) + extra--; + extra_refs += extra; + } + seq_puts(seq, "\n"); + + list_for_each_entry(task, &cset->tasks, cg_list) { + if (count++ <= MAX_TASKS_SHOWN_PER_CSS) + seq_printf(seq, " task %d\n", + task_pid_vnr(task)); + } + + list_for_each_entry(task, &cset->mg_tasks, cg_list) { + if (count++ <= MAX_TASKS_SHOWN_PER_CSS) + seq_printf(seq, " task %d\n", + task_pid_vnr(task)); + } + /* show # of overflowed tasks */ + if (count > MAX_TASKS_SHOWN_PER_CSS) + seq_printf(seq, " ... (%d)\n", + count - MAX_TASKS_SHOWN_PER_CSS); + + if (cset->dead) { + seq_puts(seq, " [dead]\n"); + dead_cnt++; + } + + WARN_ON(count != cset->nr_tasks); + } + spin_unlock_irq(&css_set_lock); + + if (!dead_cnt && !extra_refs) + return 0; + + seq_puts(seq, "\n"); + if (extra_refs) + seq_printf(seq, "extra references = %d\n", extra_refs); + if (dead_cnt) + seq_printf(seq, "dead css_sets = %d\n", dead_cnt); + + return 0; +} + +static int cgroup_subsys_states_read(struct seq_file *seq, void *v) +{ + struct kernfs_open_file *of = seq->private; + struct cgroup *cgrp; + struct cgroup_subsys *ss; + struct cgroup_subsys_state *css; + char pbuf[16]; + int i; + + cgrp = cgroup_kn_lock_live(of->kn, false); + if (!cgrp) + return -ENODEV; + + for_each_subsys(ss, i) { + css = rcu_dereference_check(cgrp->subsys[ss->id], true); + if (!css) + continue; + + pbuf[0] = '\0'; + + /* Show the parent CSS if applicable*/ + if (css->parent) + snprintf(pbuf, sizeof(pbuf) - 1, " P=%d", + css->parent->id); + seq_printf(seq, "%2d: %-4s\t- %lx[%d] %d%s\n", ss->id, ss->name, + (unsigned long)css, css->id, + atomic_read(&css->online_cnt), pbuf); + } + + cgroup_kn_unlock(of->kn); + return 0; +} + +static void cgroup_masks_read_one(struct seq_file *seq, const char *name, + u16 mask) +{ + struct cgroup_subsys *ss; + int ssid; + bool first = true; + + seq_printf(seq, "%-17s: ", name); + for_each_subsys(ss, ssid) { + if (!(mask & (1 << ssid))) + continue; + if (!first) + seq_puts(seq, ", "); + seq_puts(seq, ss->name); + first = false; + } + seq_putc(seq, '\n'); +} + +static int cgroup_masks_read(struct seq_file *seq, void *v) +{ + struct kernfs_open_file *of = seq->private; + struct cgroup *cgrp; + + cgrp = cgroup_kn_lock_live(of->kn, false); + if (!cgrp) + return -ENODEV; + + cgroup_masks_read_one(seq, "subtree_control", cgrp->subtree_control); + cgroup_masks_read_one(seq, "subtree_ss_mask", cgrp->subtree_ss_mask); + + cgroup_kn_unlock(of->kn); + return 0; +} + +static u64 releasable_read(struct cgroup_subsys_state *css, struct cftype *cft) +{ + return (!cgroup_is_populated(css->cgroup) && + !css_has_online_children(&css->cgroup->self)); +} + +static struct cftype debug_legacy_files[] = { + { + .name = "taskcount", + .read_u64 = debug_taskcount_read, + }, + + { + .name = "current_css_set", + .seq_show = current_css_set_read, + .flags = CFTYPE_ONLY_ON_ROOT, + }, + + { + .name = "current_css_set_refcount", + .read_u64 = current_css_set_refcount_read, + .flags = CFTYPE_ONLY_ON_ROOT, + }, + + { + .name = "current_css_set_cg_links", + .seq_show = current_css_set_cg_links_read, + .flags = CFTYPE_ONLY_ON_ROOT, + }, + + { + .name = "cgroup_css_links", + .seq_show = cgroup_css_links_read, + }, + + { + .name = "cgroup_subsys_states", + .seq_show = cgroup_subsys_states_read, + }, + + { + .name = "cgroup_masks", + .seq_show = cgroup_masks_read, + }, + + { + .name = "releasable", + .read_u64 = releasable_read, + }, + + { } /* terminate */ +}; + +static struct cftype debug_files[] = { + { + .name = "taskcount", + .read_u64 = debug_taskcount_read, + }, + + { + .name = "current_css_set", + .seq_show = current_css_set_read, + .flags = CFTYPE_ONLY_ON_ROOT, + }, + + { + .name = "current_css_set_refcount", + .read_u64 = current_css_set_refcount_read, + .flags = CFTYPE_ONLY_ON_ROOT, + }, + + { + .name = "current_css_set_cg_links", + .seq_show = current_css_set_cg_links_read, + .flags = CFTYPE_ONLY_ON_ROOT, + }, + + { + .name = "css_links", + .seq_show = cgroup_css_links_read, + }, + + { + .name = "csses", + .seq_show = cgroup_subsys_states_read, + }, + + { + .name = "masks", + .seq_show = cgroup_masks_read, + }, + + { } /* terminate */ +}; + +struct cgroup_subsys debug_cgrp_subsys = { + .css_alloc = debug_css_alloc, + .css_free = debug_css_free, + .legacy_cftypes = debug_legacy_files, +}; + +/* + * On v2, debug is an implicit controller enabled by "cgroup_debug" boot + * parameter. + */ +static int __init enable_cgroup_debug(char *str) +{ + debug_cgrp_subsys.dfl_cftypes = debug_files; + debug_cgrp_subsys.implicit_on_dfl = true; + return 1; +} +__setup("cgroup_debug", enable_cgroup_debug); diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup/freezer.c index 1b72d56edce5..1b72d56edce5 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup/freezer.c diff --git a/kernel/cgroup/namespace.c b/kernel/cgroup/namespace.c new file mode 100644 index 000000000000..66129eb4371d --- /dev/null +++ b/kernel/cgroup/namespace.c @@ -0,0 +1,155 @@ +#include "cgroup-internal.h" + +#include <linux/sched/task.h> +#include <linux/slab.h> +#include <linux/nsproxy.h> +#include <linux/proc_ns.h> + + +/* cgroup namespaces */ + +static struct ucounts *inc_cgroup_namespaces(struct user_namespace *ns) +{ + return inc_ucount(ns, current_euid(), UCOUNT_CGROUP_NAMESPACES); +} + +static void dec_cgroup_namespaces(struct ucounts *ucounts) +{ + dec_ucount(ucounts, UCOUNT_CGROUP_NAMESPACES); +} + +static struct cgroup_namespace *alloc_cgroup_ns(void) +{ + struct cgroup_namespace *new_ns; + int ret; + + new_ns = kzalloc(sizeof(struct cgroup_namespace), GFP_KERNEL); + if (!new_ns) + return ERR_PTR(-ENOMEM); + ret = ns_alloc_inum(&new_ns->ns); + if (ret) { + kfree(new_ns); + return ERR_PTR(ret); + } + refcount_set(&new_ns->count, 1); + new_ns->ns.ops = &cgroupns_operations; + return new_ns; +} + +void free_cgroup_ns(struct cgroup_namespace *ns) +{ + put_css_set(ns->root_cset); + dec_cgroup_namespaces(ns->ucounts); + put_user_ns(ns->user_ns); + ns_free_inum(&ns->ns); + kfree(ns); +} +EXPORT_SYMBOL(free_cgroup_ns); + +struct cgroup_namespace *copy_cgroup_ns(unsigned long flags, + struct user_namespace *user_ns, + struct cgroup_namespace *old_ns) +{ + struct cgroup_namespace *new_ns; + struct ucounts *ucounts; + struct css_set *cset; + + BUG_ON(!old_ns); + + if (!(flags & CLONE_NEWCGROUP)) { + get_cgroup_ns(old_ns); + return old_ns; + } + + /* Allow only sysadmin to create cgroup namespace. */ + if (!ns_capable(user_ns, CAP_SYS_ADMIN)) + return ERR_PTR(-EPERM); + + ucounts = inc_cgroup_namespaces(user_ns); + if (!ucounts) + return ERR_PTR(-ENOSPC); + + /* It is not safe to take cgroup_mutex here */ + spin_lock_irq(&css_set_lock); + cset = task_css_set(current); + get_css_set(cset); + spin_unlock_irq(&css_set_lock); + + new_ns = alloc_cgroup_ns(); + if (IS_ERR(new_ns)) { + put_css_set(cset); + dec_cgroup_namespaces(ucounts); + return new_ns; + } + + new_ns->user_ns = get_user_ns(user_ns); + new_ns->ucounts = ucounts; + new_ns->root_cset = cset; + + return new_ns; +} + +static inline struct cgroup_namespace *to_cg_ns(struct ns_common *ns) +{ + return container_of(ns, struct cgroup_namespace, ns); +} + +static int cgroupns_install(struct nsproxy *nsproxy, struct ns_common *ns) +{ + struct cgroup_namespace *cgroup_ns = to_cg_ns(ns); + + if (!ns_capable(current_user_ns(), CAP_SYS_ADMIN) || + !ns_capable(cgroup_ns->user_ns, CAP_SYS_ADMIN)) + return -EPERM; + + /* Don't need to do anything if we are attaching to our own cgroupns. */ + if (cgroup_ns == nsproxy->cgroup_ns) + return 0; + + get_cgroup_ns(cgroup_ns); + put_cgroup_ns(nsproxy->cgroup_ns); + nsproxy->cgroup_ns = cgroup_ns; + + return 0; +} + +static struct ns_common *cgroupns_get(struct task_struct *task) +{ + struct cgroup_namespace *ns = NULL; + struct nsproxy *nsproxy; + + task_lock(task); + nsproxy = task->nsproxy; + if (nsproxy) { + ns = nsproxy->cgroup_ns; + get_cgroup_ns(ns); + } + task_unlock(task); + + return ns ? &ns->ns : NULL; +} + +static void cgroupns_put(struct ns_common *ns) +{ + put_cgroup_ns(to_cg_ns(ns)); +} + +static struct user_namespace *cgroupns_owner(struct ns_common *ns) +{ + return to_cg_ns(ns)->user_ns; +} + +const struct proc_ns_operations cgroupns_operations = { + .name = "cgroup", + .type = CLONE_NEWCGROUP, + .get = cgroupns_get, + .put = cgroupns_put, + .install = cgroupns_install, + .owner = cgroupns_owner, +}; + +static __init int cgroup_namespaces_init(void) +{ + return 0; +} +subsys_initcall(cgroup_namespaces_init); diff --git a/kernel/cgroup_pids.c b/kernel/cgroup/pids.c index 2bd673783f1a..2237201d66d5 100644 --- a/kernel/cgroup_pids.c +++ b/kernel/cgroup/pids.c @@ -214,7 +214,7 @@ static void pids_cancel_attach(struct cgroup_taskset *tset) /* * task_css_check(true) in pids_can_fork() and pids_cancel_fork() relies - * on threadgroup_change_begin() held by the copy_process(). + * on cgroup_threadgroup_change_begin() held by the copy_process(). */ static int pids_can_fork(struct task_struct *task) { @@ -229,7 +229,7 @@ static int pids_can_fork(struct task_struct *task) /* Only log the first time events_limit is incremented. */ if (atomic64_inc_return(&pids->events_limit) == 1) { pr_info("cgroup: fork rejected by pids controller in "); - pr_cont_cgroup_path(task_cgroup(current, pids_cgrp_id)); + pr_cont_cgroup_path(css->cgroup); pr_cont("\n"); } cgroup_file_notify(&pids->events_file); diff --git a/kernel/cgroup/rdma.c b/kernel/cgroup/rdma.c new file mode 100644 index 000000000000..defad3c5e7dc --- /dev/null +++ b/kernel/cgroup/rdma.c @@ -0,0 +1,619 @@ +/* + * RDMA resource limiting controller for cgroups. + * + * Used to allow a cgroup hierarchy to stop processes from consuming + * additional RDMA resources after a certain limit is reached. + * + * Copyright (C) 2016 Parav Pandit <[email protected]> + * + * This file is subject to the terms and conditions of version 2 of the GNU + * General Public License. See the file COPYING in the main directory of the + * Linux distribution for more details. + */ + +#include <linux/bitops.h> +#include <linux/slab.h> +#include <linux/seq_file.h> +#include <linux/cgroup.h> +#include <linux/parser.h> +#include <linux/cgroup_rdma.h> + +#define RDMACG_MAX_STR "max" + +/* + * Protects list of resource pools maintained on per cgroup basis + * and rdma device list. + */ +static DEFINE_MUTEX(rdmacg_mutex); +static LIST_HEAD(rdmacg_devices); + +enum rdmacg_file_type { + RDMACG_RESOURCE_TYPE_MAX, + RDMACG_RESOURCE_TYPE_STAT, +}; + +/* + * resource table definition as to be seen by the user. + * Need to add entries to it when more resources are + * added/defined at IB verb/core layer. + */ +static char const *rdmacg_resource_names[] = { + [RDMACG_RESOURCE_HCA_HANDLE] = "hca_handle", + [RDMACG_RESOURCE_HCA_OBJECT] = "hca_object", +}; + +/* resource tracker for each resource of rdma cgroup */ +struct rdmacg_resource { + int max; + int usage; +}; + +/* + * resource pool object which represents per cgroup, per device + * resources. There are multiple instances of this object per cgroup, + * therefore it cannot be embedded within rdma_cgroup structure. It + * is maintained as list. + */ +struct rdmacg_resource_pool { + struct rdmacg_device *device; + struct rdmacg_resource resources[RDMACG_RESOURCE_MAX]; + + struct list_head cg_node; + struct list_head dev_node; + + /* count active user tasks of this pool */ + u64 usage_sum; + /* total number counts which are set to max */ + int num_max_cnt; +}; + +static struct rdma_cgroup *css_rdmacg(struct cgroup_subsys_state *css) +{ + return container_of(css, struct rdma_cgroup, css); +} + +static struct rdma_cgroup *parent_rdmacg(struct rdma_cgroup *cg) +{ + return css_rdmacg(cg->css.parent); +} + +static inline struct rdma_cgroup *get_current_rdmacg(void) +{ + return css_rdmacg(task_get_css(current, rdma_cgrp_id)); +} + +static void set_resource_limit(struct rdmacg_resource_pool *rpool, + int index, int new_max) +{ + if (new_max == S32_MAX) { + if (rpool->resources[index].max != S32_MAX) + rpool->num_max_cnt++; + } else { + if (rpool->resources[index].max == S32_MAX) + rpool->num_max_cnt--; + } + rpool->resources[index].max = new_max; +} + +static void set_all_resource_max_limit(struct rdmacg_resource_pool *rpool) +{ + int i; + + for (i = 0; i < RDMACG_RESOURCE_MAX; i++) + set_resource_limit(rpool, i, S32_MAX); +} + +static void free_cg_rpool_locked(struct rdmacg_resource_pool *rpool) +{ + lockdep_assert_held(&rdmacg_mutex); + + list_del(&rpool->cg_node); + list_del(&rpool->dev_node); + kfree(rpool); +} + +static struct rdmacg_resource_pool * +find_cg_rpool_locked(struct rdma_cgroup *cg, + struct rdmacg_device *device) + +{ + struct rdmacg_resource_pool *pool; + + lockdep_assert_held(&rdmacg_mutex); + + list_for_each_entry(pool, &cg->rpools, cg_node) + if (pool->device == device) + return pool; + + return NULL; +} + +static struct rdmacg_resource_pool * +get_cg_rpool_locked(struct rdma_cgroup *cg, struct rdmacg_device *device) +{ + struct rdmacg_resource_pool *rpool; + + rpool = find_cg_rpool_locked(cg, device); + if (rpool) + return rpool; + + rpool = kzalloc(sizeof(*rpool), GFP_KERNEL); + if (!rpool) + return ERR_PTR(-ENOMEM); + + rpool->device = device; + set_all_resource_max_limit(rpool); + + INIT_LIST_HEAD(&rpool->cg_node); + INIT_LIST_HEAD(&rpool->dev_node); + list_add_tail(&rpool->cg_node, &cg->rpools); + list_add_tail(&rpool->dev_node, &device->rpools); + return rpool; +} + +/** + * uncharge_cg_locked - uncharge resource for rdma cgroup + * @cg: pointer to cg to uncharge and all parents in hierarchy + * @device: pointer to rdmacg device + * @index: index of the resource to uncharge in cg (resource pool) + * + * It also frees the resource pool which was created as part of + * charging operation when there are no resources attached to + * resource pool. + */ +static void +uncharge_cg_locked(struct rdma_cgroup *cg, + struct rdmacg_device *device, + enum rdmacg_resource_type index) +{ + struct rdmacg_resource_pool *rpool; + + rpool = find_cg_rpool_locked(cg, device); + + /* + * rpool cannot be null at this stage. Let kernel operate in case + * if there a bug in IB stack or rdma controller, instead of crashing + * the system. + */ + if (unlikely(!rpool)) { + pr_warn("Invalid device %p or rdma cgroup %p\n", cg, device); + return; + } + + rpool->resources[index].usage--; + + /* + * A negative count (or overflow) is invalid, + * it indicates a bug in the rdma controller. + */ + WARN_ON_ONCE(rpool->resources[index].usage < 0); + rpool->usage_sum--; + if (rpool->usage_sum == 0 && + rpool->num_max_cnt == RDMACG_RESOURCE_MAX) { + /* + * No user of the rpool and all entries are set to max, so + * safe to delete this rpool. + */ + free_cg_rpool_locked(rpool); + } +} + +/** + * rdmacg_uncharge_hierarchy - hierarchically uncharge rdma resource count + * @device: pointer to rdmacg device + * @stop_cg: while traversing hirerchy, when meet with stop_cg cgroup + * stop uncharging + * @index: index of the resource to uncharge in cg in given resource pool + */ +static void rdmacg_uncharge_hierarchy(struct rdma_cgroup *cg, + struct rdmacg_device *device, + struct rdma_cgroup *stop_cg, + enum rdmacg_resource_type index) +{ + struct rdma_cgroup *p; + + mutex_lock(&rdmacg_mutex); + + for (p = cg; p != stop_cg; p = parent_rdmacg(p)) + uncharge_cg_locked(p, device, index); + + mutex_unlock(&rdmacg_mutex); + + css_put(&cg->css); +} + +/** + * rdmacg_uncharge - hierarchically uncharge rdma resource count + * @device: pointer to rdmacg device + * @index: index of the resource to uncharge in cgroup in given resource pool + */ +void rdmacg_uncharge(struct rdma_cgroup *cg, + struct rdmacg_device *device, + enum rdmacg_resource_type index) +{ + if (index >= RDMACG_RESOURCE_MAX) + return; + + rdmacg_uncharge_hierarchy(cg, device, NULL, index); +} +EXPORT_SYMBOL(rdmacg_uncharge); + +/** + * rdmacg_try_charge - hierarchically try to charge the rdma resource + * @rdmacg: pointer to rdma cgroup which will own this resource + * @device: pointer to rdmacg device + * @index: index of the resource to charge in cgroup (resource pool) + * + * This function follows charging resource in hierarchical way. + * It will fail if the charge would cause the new value to exceed the + * hierarchical limit. + * Returns 0 if the charge succeded, otherwise -EAGAIN, -ENOMEM or -EINVAL. + * Returns pointer to rdmacg for this resource when charging is successful. + * + * Charger needs to account resources on two criteria. + * (a) per cgroup & (b) per device resource usage. + * Per cgroup resource usage ensures that tasks of cgroup doesn't cross + * the configured limits. Per device provides granular configuration + * in multi device usage. It allocates resource pool in the hierarchy + * for each parent it come across for first resource. Later on resource + * pool will be available. Therefore it will be much faster thereon + * to charge/uncharge. + */ +int rdmacg_try_charge(struct rdma_cgroup **rdmacg, + struct rdmacg_device *device, + enum rdmacg_resource_type index) +{ + struct rdma_cgroup *cg, *p; + struct rdmacg_resource_pool *rpool; + s64 new; + int ret = 0; + + if (index >= RDMACG_RESOURCE_MAX) + return -EINVAL; + + /* + * hold on to css, as cgroup can be removed but resource + * accounting happens on css. + */ + cg = get_current_rdmacg(); + + mutex_lock(&rdmacg_mutex); + for (p = cg; p; p = parent_rdmacg(p)) { + rpool = get_cg_rpool_locked(p, device); + if (IS_ERR(rpool)) { + ret = PTR_ERR(rpool); + goto err; + } else { + new = rpool->resources[index].usage + 1; + if (new > rpool->resources[index].max) { + ret = -EAGAIN; + goto err; + } else { + rpool->resources[index].usage = new; + rpool->usage_sum++; + } + } + } + mutex_unlock(&rdmacg_mutex); + + *rdmacg = cg; + return 0; + +err: + mutex_unlock(&rdmacg_mutex); + rdmacg_uncharge_hierarchy(cg, device, p, index); + return ret; +} +EXPORT_SYMBOL(rdmacg_try_charge); + +/** + * rdmacg_register_device - register rdmacg device to rdma controller. + * @device: pointer to rdmacg device whose resources need to be accounted. + * + * If IB stack wish a device to participate in rdma cgroup resource + * tracking, it must invoke this API to register with rdma cgroup before + * any user space application can start using the RDMA resources. + * Returns 0 on success or EINVAL when table length given is beyond + * supported size. + */ +int rdmacg_register_device(struct rdmacg_device *device) +{ + INIT_LIST_HEAD(&device->dev_node); + INIT_LIST_HEAD(&device->rpools); + + mutex_lock(&rdmacg_mutex); + list_add_tail(&device->dev_node, &rdmacg_devices); + mutex_unlock(&rdmacg_mutex); + return 0; +} +EXPORT_SYMBOL(rdmacg_register_device); + +/** + * rdmacg_unregister_device - unregister rdmacg device from rdma controller. + * @device: pointer to rdmacg device which was previously registered with rdma + * controller using rdmacg_register_device(). + * + * IB stack must invoke this after all the resources of the IB device + * are destroyed and after ensuring that no more resources will be created + * when this API is invoked. + */ +void rdmacg_unregister_device(struct rdmacg_device *device) +{ + struct rdmacg_resource_pool *rpool, *tmp; + + /* + * Synchronize with any active resource settings, + * usage query happening via configfs. + */ + mutex_lock(&rdmacg_mutex); + list_del_init(&device->dev_node); + + /* + * Now that this device is off the cgroup list, its safe to free + * all the rpool resources. + */ + list_for_each_entry_safe(rpool, tmp, &device->rpools, dev_node) + free_cg_rpool_locked(rpool); + + mutex_unlock(&rdmacg_mutex); +} +EXPORT_SYMBOL(rdmacg_unregister_device); + +static int parse_resource(char *c, int *intval) +{ + substring_t argstr; + const char **table = &rdmacg_resource_names[0]; + char *name, *value = c; + size_t len; + int ret, i = 0; + + name = strsep(&value, "="); + if (!name || !value) + return -EINVAL; + + len = strlen(value); + + for (i = 0; i < RDMACG_RESOURCE_MAX; i++) { + if (strcmp(table[i], name)) + continue; + + argstr.from = value; + argstr.to = value + len; + + ret = match_int(&argstr, intval); + if (ret >= 0) { + if (*intval < 0) + break; + return i; + } + if (strncmp(value, RDMACG_MAX_STR, len) == 0) { + *intval = S32_MAX; + return i; + } + break; + } + return -EINVAL; +} + +static int rdmacg_parse_limits(char *options, + int *new_limits, unsigned long *enables) +{ + char *c; + int err = -EINVAL; + + /* parse resource options */ + while ((c = strsep(&options, " ")) != NULL) { + int index, intval; + + index = parse_resource(c, &intval); + if (index < 0) + goto err; + + new_limits[index] = intval; + *enables |= BIT(index); + } + return 0; + +err: + return err; +} + +static struct rdmacg_device *rdmacg_get_device_locked(const char *name) +{ + struct rdmacg_device *device; + + lockdep_assert_held(&rdmacg_mutex); + + list_for_each_entry(device, &rdmacg_devices, dev_node) + if (!strcmp(name, device->name)) + return device; + + return NULL; +} + +static ssize_t rdmacg_resource_set_max(struct kernfs_open_file *of, + char *buf, size_t nbytes, loff_t off) +{ + struct rdma_cgroup *cg = css_rdmacg(of_css(of)); + const char *dev_name; + struct rdmacg_resource_pool *rpool; + struct rdmacg_device *device; + char *options = strstrip(buf); + int *new_limits; + unsigned long enables = 0; + int i = 0, ret = 0; + + /* extract the device name first */ + dev_name = strsep(&options, " "); + if (!dev_name) { + ret = -EINVAL; + goto err; + } + + new_limits = kcalloc(RDMACG_RESOURCE_MAX, sizeof(int), GFP_KERNEL); + if (!new_limits) { + ret = -ENOMEM; + goto err; + } + + ret = rdmacg_parse_limits(options, new_limits, &enables); + if (ret) + goto parse_err; + + /* acquire lock to synchronize with hot plug devices */ + mutex_lock(&rdmacg_mutex); + + device = rdmacg_get_device_locked(dev_name); + if (!device) { + ret = -ENODEV; + goto dev_err; + } + + rpool = get_cg_rpool_locked(cg, device); + if (IS_ERR(rpool)) { + ret = PTR_ERR(rpool); + goto dev_err; + } + + /* now set the new limits of the rpool */ + for_each_set_bit(i, &enables, RDMACG_RESOURCE_MAX) + set_resource_limit(rpool, i, new_limits[i]); + + if (rpool->usage_sum == 0 && + rpool->num_max_cnt == RDMACG_RESOURCE_MAX) { + /* + * No user of the rpool and all entries are set to max, so + * safe to delete this rpool. + */ + free_cg_rpool_locked(rpool); + } + +dev_err: + mutex_unlock(&rdmacg_mutex); + +parse_err: + kfree(new_limits); + +err: + return ret ?: nbytes; +} + +static void print_rpool_values(struct seq_file *sf, + struct rdmacg_resource_pool *rpool) +{ + enum rdmacg_file_type sf_type; + int i; + u32 value; + + sf_type = seq_cft(sf)->private; + + for (i = 0; i < RDMACG_RESOURCE_MAX; i++) { + seq_puts(sf, rdmacg_resource_names[i]); + seq_putc(sf, '='); + if (sf_type == RDMACG_RESOURCE_TYPE_MAX) { + if (rpool) + value = rpool->resources[i].max; + else + value = S32_MAX; + } else { + if (rpool) + value = rpool->resources[i].usage; + else + value = 0; + } + + if (value == S32_MAX) + seq_puts(sf, RDMACG_MAX_STR); + else + seq_printf(sf, "%d", value); + seq_putc(sf, ' '); + } +} + +static int rdmacg_resource_read(struct seq_file *sf, void *v) +{ + struct rdmacg_device *device; + struct rdmacg_resource_pool *rpool; + struct rdma_cgroup *cg = css_rdmacg(seq_css(sf)); + + mutex_lock(&rdmacg_mutex); + + list_for_each_entry(device, &rdmacg_devices, dev_node) { + seq_printf(sf, "%s ", device->name); + + rpool = find_cg_rpool_locked(cg, device); + print_rpool_values(sf, rpool); + + seq_putc(sf, '\n'); + } + + mutex_unlock(&rdmacg_mutex); + return 0; +} + +static struct cftype rdmacg_files[] = { + { + .name = "max", + .write = rdmacg_resource_set_max, + .seq_show = rdmacg_resource_read, + .private = RDMACG_RESOURCE_TYPE_MAX, + .flags = CFTYPE_NOT_ON_ROOT, + }, + { + .name = "current", + .seq_show = rdmacg_resource_read, + .private = RDMACG_RESOURCE_TYPE_STAT, + .flags = CFTYPE_NOT_ON_ROOT, + }, + { } /* terminate */ +}; + +static struct cgroup_subsys_state * +rdmacg_css_alloc(struct cgroup_subsys_state *parent) +{ + struct rdma_cgroup *cg; + + cg = kzalloc(sizeof(*cg), GFP_KERNEL); + if (!cg) + return ERR_PTR(-ENOMEM); + + INIT_LIST_HEAD(&cg->rpools); + return &cg->css; +} + +static void rdmacg_css_free(struct cgroup_subsys_state *css) +{ + struct rdma_cgroup *cg = css_rdmacg(css); + + kfree(cg); +} + +/** + * rdmacg_css_offline - cgroup css_offline callback + * @css: css of interest + * + * This function is called when @css is about to go away and responsible + * for shooting down all rdmacg associated with @css. As part of that it + * marks all the resource pool entries to max value, so that when resources are + * uncharged, associated resource pool can be freed as well. + */ +static void rdmacg_css_offline(struct cgroup_subsys_state *css) +{ + struct rdma_cgroup *cg = css_rdmacg(css); + struct rdmacg_resource_pool *rpool; + + mutex_lock(&rdmacg_mutex); + + list_for_each_entry(rpool, &cg->rpools, cg_node) + set_all_resource_max_limit(rpool); + + mutex_unlock(&rdmacg_mutex); +} + +struct cgroup_subsys rdma_cgrp_subsys = { + .css_alloc = rdmacg_css_alloc, + .css_free = rdmacg_css_free, + .css_offline = rdmacg_css_offline, + .legacy_cftypes = rdmacg_files, + .dfl_cftypes = rdmacg_files, +}; diff --git a/kernel/compat.c b/kernel/compat.c index 19aec5d98108..6f0a0e723a06 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -30,100 +30,66 @@ #include <linux/uaccess.h> -static int compat_get_timex(struct timex *txc, struct compat_timex __user *utp) +int compat_get_timex(struct timex *txc, const struct compat_timex __user *utp) { - memset(txc, 0, sizeof(struct timex)); - - if (!access_ok(VERIFY_READ, utp, sizeof(struct compat_timex)) || - __get_user(txc->modes, &utp->modes) || - __get_user(txc->offset, &utp->offset) || - __get_user(txc->freq, &utp->freq) || - __get_user(txc->maxerror, &utp->maxerror) || - __get_user(txc->esterror, &utp->esterror) || - __get_user(txc->status, &utp->status) || - __get_user(txc->constant, &utp->constant) || - __get_user(txc->precision, &utp->precision) || - __get_user(txc->tolerance, &utp->tolerance) || - __get_user(txc->time.tv_sec, &utp->time.tv_sec) || - __get_user(txc->time.tv_usec, &utp->time.tv_usec) || - __get_user(txc->tick, &utp->tick) || - __get_user(txc->ppsfreq, &utp->ppsfreq) || - __get_user(txc->jitter, &utp->jitter) || - __get_user(txc->shift, &utp->shift) || - __get_user(txc->stabil, &utp->stabil) || - __get_user(txc->jitcnt, &utp->jitcnt) || - __get_user(txc->calcnt, &utp->calcnt) || - __get_user(txc->errcnt, &utp->errcnt) || - __get_user(txc->stbcnt, &utp->stbcnt)) - return -EFAULT; + struct compat_timex tx32; - return 0; -} - -static int compat_put_timex(struct compat_timex __user *utp, struct timex *txc) -{ - if (!access_ok(VERIFY_WRITE, utp, sizeof(struct compat_timex)) || - __put_user(txc->modes, &utp->modes) || - __put_user(txc->offset, &utp->offset) || - __put_user(txc->freq, &utp->freq) || - __put_user(txc->maxerror, &utp->maxerror) || - __put_user(txc->esterror, &utp->esterror) || - __put_user(txc->status, &utp->status) || - __put_user(txc->constant, &utp->constant) || - __put_user(txc->precision, &utp->precision) || - __put_user(txc->tolerance, &utp->tolerance) || - __put_user(txc->time.tv_sec, &utp->time.tv_sec) || - __put_user(txc->time.tv_usec, &utp->time.tv_usec) || - __put_user(txc->tick, &utp->tick) || - __put_user(txc->ppsfreq, &utp->ppsfreq) || - __put_user(txc->jitter, &utp->jitter) || - __put_user(txc->shift, &utp->shift) || - __put_user(txc->stabil, &utp->stabil) || - __put_user(txc->jitcnt, &utp->jitcnt) || - __put_user(txc->calcnt, &utp->calcnt) || - __put_user(txc->errcnt, &utp->errcnt) || - __put_user(txc->stbcnt, &utp->stbcnt) || - __put_user(txc->tai, &utp->tai)) + if (copy_from_user(&tx32, utp, sizeof(struct compat_timex))) return -EFAULT; - return 0; -} -COMPAT_SYSCALL_DEFINE2(gettimeofday, struct compat_timeval __user *, tv, - struct timezone __user *, tz) -{ - if (tv) { - struct timeval ktv; - do_gettimeofday(&ktv); - if (compat_put_timeval(&ktv, tv)) - return -EFAULT; - } - if (tz) { - if (copy_to_user(tz, &sys_tz, sizeof(sys_tz))) - return -EFAULT; - } + txc->modes = tx32.modes; + txc->offset = tx32.offset; + txc->freq = tx32.freq; + txc->maxerror = tx32.maxerror; + txc->esterror = tx32.esterror; + txc->status = tx32.status; + txc->constant = tx32.constant; + txc->precision = tx32.precision; + txc->tolerance = tx32.tolerance; + txc->time.tv_sec = tx32.time.tv_sec; + txc->time.tv_usec = tx32.time.tv_usec; + txc->tick = tx32.tick; + txc->ppsfreq = tx32.ppsfreq; + txc->jitter = tx32.jitter; + txc->shift = tx32.shift; + txc->stabil = tx32.stabil; + txc->jitcnt = tx32.jitcnt; + txc->calcnt = tx32.calcnt; + txc->errcnt = tx32.errcnt; + txc->stbcnt = tx32.stbcnt; return 0; } -COMPAT_SYSCALL_DEFINE2(settimeofday, struct compat_timeval __user *, tv, - struct timezone __user *, tz) -{ - struct timeval user_tv; - struct timespec new_ts; - struct timezone new_tz; - - if (tv) { - if (compat_get_timeval(&user_tv, tv)) - return -EFAULT; - new_ts.tv_sec = user_tv.tv_sec; - new_ts.tv_nsec = user_tv.tv_usec * NSEC_PER_USEC; - } - if (tz) { - if (copy_from_user(&new_tz, tz, sizeof(*tz))) - return -EFAULT; - } - - return do_sys_settimeofday(tv ? &new_ts : NULL, tz ? &new_tz : NULL); +int compat_put_timex(struct compat_timex __user *utp, const struct timex *txc) +{ + struct compat_timex tx32; + + memset(&tx32, 0, sizeof(struct compat_timex)); + tx32.modes = txc->modes; + tx32.offset = txc->offset; + tx32.freq = txc->freq; + tx32.maxerror = txc->maxerror; + tx32.esterror = txc->esterror; + tx32.status = txc->status; + tx32.constant = txc->constant; + tx32.precision = txc->precision; + tx32.tolerance = txc->tolerance; + tx32.time.tv_sec = txc->time.tv_sec; + tx32.time.tv_usec = txc->time.tv_usec; + tx32.tick = txc->tick; + tx32.ppsfreq = txc->ppsfreq; + tx32.jitter = txc->jitter; + tx32.shift = txc->shift; + tx32.stabil = txc->stabil; + tx32.jitcnt = txc->jitcnt; + tx32.calcnt = txc->calcnt; + tx32.errcnt = txc->errcnt; + tx32.stbcnt = txc->stbcnt; + tx32.tai = txc->tai; + if (copy_to_user(utp, &tx32, sizeof(struct compat_timex))) + return -EFAULT; + return 0; } static int __compat_get_timeval(struct timeval *tv, const struct compat_timeval __user *ctv) @@ -154,6 +120,50 @@ static int __compat_put_timespec(const struct timespec *ts, struct compat_timesp __put_user(ts->tv_nsec, &cts->tv_nsec)) ? -EFAULT : 0; } +static int __compat_get_timespec64(struct timespec64 *ts64, + const struct compat_timespec __user *cts) +{ + struct compat_timespec ts; + int ret; + + ret = copy_from_user(&ts, cts, sizeof(ts)); + if (ret) + return -EFAULT; + + ts64->tv_sec = ts.tv_sec; + ts64->tv_nsec = ts.tv_nsec; + + return 0; +} + +static int __compat_put_timespec64(const struct timespec64 *ts64, + struct compat_timespec __user *cts) +{ + struct compat_timespec ts = { + .tv_sec = ts64->tv_sec, + .tv_nsec = ts64->tv_nsec + }; + return copy_to_user(cts, &ts, sizeof(ts)) ? -EFAULT : 0; +} + +int compat_get_timespec64(struct timespec64 *ts, const void __user *uts) +{ + if (COMPAT_USE_64BIT_TIME) + return copy_from_user(ts, uts, sizeof(*ts)) ? -EFAULT : 0; + else + return __compat_get_timespec64(ts, uts); +} +EXPORT_SYMBOL_GPL(compat_get_timespec64); + +int compat_put_timespec64(const struct timespec64 *ts, void __user *uts) +{ + if (COMPAT_USE_64BIT_TIME) + return copy_to_user(uts, ts, sizeof(*ts)) ? -EFAULT : 0; + else + return __compat_put_timespec64(ts, uts); +} +EXPORT_SYMBOL_GPL(compat_put_timespec64); + int compat_get_timeval(struct timeval *tv, const void __user *utv) { if (COMPAT_USE_64BIT_TIME) @@ -213,188 +223,30 @@ int compat_convert_timespec(struct timespec __user **kts, return 0; } -static long compat_nanosleep_restart(struct restart_block *restart) -{ - struct compat_timespec __user *rmtp; - struct timespec rmt; - mm_segment_t oldfs; - long ret; - - restart->nanosleep.rmtp = (struct timespec __user *) &rmt; - oldfs = get_fs(); - set_fs(KERNEL_DS); - ret = hrtimer_nanosleep_restart(restart); - set_fs(oldfs); - - if (ret == -ERESTART_RESTARTBLOCK) { - rmtp = restart->nanosleep.compat_rmtp; - - if (rmtp && compat_put_timespec(&rmt, rmtp)) - return -EFAULT; - } - - return ret; -} - -COMPAT_SYSCALL_DEFINE2(nanosleep, struct compat_timespec __user *, rqtp, - struct compat_timespec __user *, rmtp) -{ - struct timespec tu, rmt; - mm_segment_t oldfs; - long ret; - - if (compat_get_timespec(&tu, rqtp)) - return -EFAULT; - - if (!timespec_valid(&tu)) - return -EINVAL; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - ret = hrtimer_nanosleep(&tu, - rmtp ? (struct timespec __user *)&rmt : NULL, - HRTIMER_MODE_REL, CLOCK_MONOTONIC); - set_fs(oldfs); - - /* - * hrtimer_nanosleep() can only return 0 or - * -ERESTART_RESTARTBLOCK here because: - * - * - we call it with HRTIMER_MODE_REL and therefor exclude the - * -ERESTARTNOHAND return path. - * - * - we supply the rmtp argument from the task stack (due to - * the necessary compat conversion. So the update cannot - * fail, which excludes the -EFAULT return path as well. If - * it fails nevertheless we have a bigger problem and wont - * reach this place anymore. - * - * - if the return value is 0, we do not have to update rmtp - * because there is no remaining time. - * - * We check for -ERESTART_RESTARTBLOCK nevertheless if the - * core implementation decides to return random nonsense. - */ - if (ret == -ERESTART_RESTARTBLOCK) { - struct restart_block *restart = ¤t->restart_block; - - restart->fn = compat_nanosleep_restart; - restart->nanosleep.compat_rmtp = rmtp; - - if (rmtp && compat_put_timespec(&rmt, rmtp)) - return -EFAULT; - } - return ret; -} - -static inline long get_compat_itimerval(struct itimerval *o, - struct compat_itimerval __user *i) +int get_compat_itimerval(struct itimerval *o, const struct compat_itimerval __user *i) { - return (!access_ok(VERIFY_READ, i, sizeof(*i)) || - (__get_user(o->it_interval.tv_sec, &i->it_interval.tv_sec) | - __get_user(o->it_interval.tv_usec, &i->it_interval.tv_usec) | - __get_user(o->it_value.tv_sec, &i->it_value.tv_sec) | - __get_user(o->it_value.tv_usec, &i->it_value.tv_usec))); -} - -static inline long put_compat_itimerval(struct compat_itimerval __user *o, - struct itimerval *i) -{ - return (!access_ok(VERIFY_WRITE, o, sizeof(*o)) || - (__put_user(i->it_interval.tv_sec, &o->it_interval.tv_sec) | - __put_user(i->it_interval.tv_usec, &o->it_interval.tv_usec) | - __put_user(i->it_value.tv_sec, &o->it_value.tv_sec) | - __put_user(i->it_value.tv_usec, &o->it_value.tv_usec))); -} - -asmlinkage long sys_ni_posix_timers(void); + struct compat_itimerval v32; -COMPAT_SYSCALL_DEFINE2(getitimer, int, which, - struct compat_itimerval __user *, it) -{ - struct itimerval kit; - int error; - - if (!IS_ENABLED(CONFIG_POSIX_TIMERS)) - return sys_ni_posix_timers(); - - error = do_getitimer(which, &kit); - if (!error && put_compat_itimerval(it, &kit)) - error = -EFAULT; - return error; -} - -COMPAT_SYSCALL_DEFINE3(setitimer, int, which, - struct compat_itimerval __user *, in, - struct compat_itimerval __user *, out) -{ - struct itimerval kin, kout; - int error; - - if (!IS_ENABLED(CONFIG_POSIX_TIMERS)) - return sys_ni_posix_timers(); - - if (in) { - if (get_compat_itimerval(&kin, in)) - return -EFAULT; - } else - memset(&kin, 0, sizeof(kin)); - - error = do_setitimer(which, &kin, out ? &kout : NULL); - if (error || !out) - return error; - if (put_compat_itimerval(out, &kout)) + if (copy_from_user(&v32, i, sizeof(struct compat_itimerval))) return -EFAULT; + o->it_interval.tv_sec = v32.it_interval.tv_sec; + o->it_interval.tv_usec = v32.it_interval.tv_usec; + o->it_value.tv_sec = v32.it_value.tv_sec; + o->it_value.tv_usec = v32.it_value.tv_usec; return 0; } -static compat_clock_t clock_t_to_compat_clock_t(clock_t x) -{ - return compat_jiffies_to_clock_t(clock_t_to_jiffies(x)); -} - -COMPAT_SYSCALL_DEFINE1(times, struct compat_tms __user *, tbuf) +int put_compat_itimerval(struct compat_itimerval __user *o, const struct itimerval *i) { - if (tbuf) { - struct tms tms; - struct compat_tms tmp; - - do_sys_times(&tms); - /* Convert our struct tms to the compat version. */ - tmp.tms_utime = clock_t_to_compat_clock_t(tms.tms_utime); - tmp.tms_stime = clock_t_to_compat_clock_t(tms.tms_stime); - tmp.tms_cutime = clock_t_to_compat_clock_t(tms.tms_cutime); - tmp.tms_cstime = clock_t_to_compat_clock_t(tms.tms_cstime); - if (copy_to_user(tbuf, &tmp, sizeof(tmp))) - return -EFAULT; - } - force_successful_syscall_return(); - return compat_jiffies_to_clock_t(jiffies); -} - -#ifdef __ARCH_WANT_SYS_SIGPENDING + struct compat_itimerval v32; -/* - * Assumption: old_sigset_t and compat_old_sigset_t are both - * types that can be passed to put_user()/get_user(). - */ - -COMPAT_SYSCALL_DEFINE1(sigpending, compat_old_sigset_t __user *, set) -{ - old_sigset_t s; - long ret; - mm_segment_t old_fs = get_fs(); - - set_fs(KERNEL_DS); - ret = sys_sigpending((old_sigset_t __user *) &s); - set_fs(old_fs); - if (ret == 0) - ret = put_user(s, set); - return ret; + v32.it_interval.tv_sec = i->it_interval.tv_sec; + v32.it_interval.tv_usec = i->it_interval.tv_usec; + v32.it_value.tv_sec = i->it_value.tv_sec; + v32.it_value.tv_usec = i->it_value.tv_usec; + return copy_to_user(o, &v32, sizeof(struct compat_itimerval)) ? -EFAULT : 0; } -#endif - #ifdef __ARCH_WANT_SYS_SIGPROCMASK /* @@ -449,164 +301,33 @@ COMPAT_SYSCALL_DEFINE3(sigprocmask, int, how, #endif -COMPAT_SYSCALL_DEFINE2(setrlimit, unsigned int, resource, - struct compat_rlimit __user *, rlim) -{ - struct rlimit r; - - if (!access_ok(VERIFY_READ, rlim, sizeof(*rlim)) || - __get_user(r.rlim_cur, &rlim->rlim_cur) || - __get_user(r.rlim_max, &rlim->rlim_max)) - return -EFAULT; - - if (r.rlim_cur == COMPAT_RLIM_INFINITY) - r.rlim_cur = RLIM_INFINITY; - if (r.rlim_max == COMPAT_RLIM_INFINITY) - r.rlim_max = RLIM_INFINITY; - return do_prlimit(current, resource, &r, NULL); -} - -#ifdef COMPAT_RLIM_OLD_INFINITY - -COMPAT_SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, - struct compat_rlimit __user *, rlim) -{ - struct rlimit r; - int ret; - mm_segment_t old_fs = get_fs(); - - set_fs(KERNEL_DS); - ret = sys_old_getrlimit(resource, (struct rlimit __user *)&r); - set_fs(old_fs); - - if (!ret) { - if (r.rlim_cur > COMPAT_RLIM_OLD_INFINITY) - r.rlim_cur = COMPAT_RLIM_INFINITY; - if (r.rlim_max > COMPAT_RLIM_OLD_INFINITY) - r.rlim_max = COMPAT_RLIM_INFINITY; - - if (!access_ok(VERIFY_WRITE, rlim, sizeof(*rlim)) || - __put_user(r.rlim_cur, &rlim->rlim_cur) || - __put_user(r.rlim_max, &rlim->rlim_max)) - return -EFAULT; - } - return ret; -} - -#endif - -COMPAT_SYSCALL_DEFINE2(getrlimit, unsigned int, resource, - struct compat_rlimit __user *, rlim) -{ - struct rlimit r; - int ret; - - ret = do_prlimit(current, resource, NULL, &r); - if (!ret) { - if (r.rlim_cur > COMPAT_RLIM_INFINITY) - r.rlim_cur = COMPAT_RLIM_INFINITY; - if (r.rlim_max > COMPAT_RLIM_INFINITY) - r.rlim_max = COMPAT_RLIM_INFINITY; - - if (!access_ok(VERIFY_WRITE, rlim, sizeof(*rlim)) || - __put_user(r.rlim_cur, &rlim->rlim_cur) || - __put_user(r.rlim_max, &rlim->rlim_max)) - return -EFAULT; - } - return ret; -} - int put_compat_rusage(const struct rusage *r, struct compat_rusage __user *ru) { - if (!access_ok(VERIFY_WRITE, ru, sizeof(*ru)) || - __put_user(r->ru_utime.tv_sec, &ru->ru_utime.tv_sec) || - __put_user(r->ru_utime.tv_usec, &ru->ru_utime.tv_usec) || - __put_user(r->ru_stime.tv_sec, &ru->ru_stime.tv_sec) || - __put_user(r->ru_stime.tv_usec, &ru->ru_stime.tv_usec) || - __put_user(r->ru_maxrss, &ru->ru_maxrss) || - __put_user(r->ru_ixrss, &ru->ru_ixrss) || - __put_user(r->ru_idrss, &ru->ru_idrss) || - __put_user(r->ru_isrss, &ru->ru_isrss) || - __put_user(r->ru_minflt, &ru->ru_minflt) || - __put_user(r->ru_majflt, &ru->ru_majflt) || - __put_user(r->ru_nswap, &ru->ru_nswap) || - __put_user(r->ru_inblock, &ru->ru_inblock) || - __put_user(r->ru_oublock, &ru->ru_oublock) || - __put_user(r->ru_msgsnd, &ru->ru_msgsnd) || - __put_user(r->ru_msgrcv, &ru->ru_msgrcv) || - __put_user(r->ru_nsignals, &ru->ru_nsignals) || - __put_user(r->ru_nvcsw, &ru->ru_nvcsw) || - __put_user(r->ru_nivcsw, &ru->ru_nivcsw)) + struct compat_rusage r32; + memset(&r32, 0, sizeof(r32)); + r32.ru_utime.tv_sec = r->ru_utime.tv_sec; + r32.ru_utime.tv_usec = r->ru_utime.tv_usec; + r32.ru_stime.tv_sec = r->ru_stime.tv_sec; + r32.ru_stime.tv_usec = r->ru_stime.tv_usec; + r32.ru_maxrss = r->ru_maxrss; + r32.ru_ixrss = r->ru_ixrss; + r32.ru_idrss = r->ru_idrss; + r32.ru_isrss = r->ru_isrss; + r32.ru_minflt = r->ru_minflt; + r32.ru_majflt = r->ru_majflt; + r32.ru_nswap = r->ru_nswap; + r32.ru_inblock = r->ru_inblock; + r32.ru_oublock = r->ru_oublock; + r32.ru_msgsnd = r->ru_msgsnd; + r32.ru_msgrcv = r->ru_msgrcv; + r32.ru_nsignals = r->ru_nsignals; + r32.ru_nvcsw = r->ru_nvcsw; + r32.ru_nivcsw = r->ru_nivcsw; + if (copy_to_user(ru, &r32, sizeof(r32))) return -EFAULT; return 0; } -COMPAT_SYSCALL_DEFINE4(wait4, - compat_pid_t, pid, - compat_uint_t __user *, stat_addr, - int, options, - struct compat_rusage __user *, ru) -{ - if (!ru) { - return sys_wait4(pid, stat_addr, options, NULL); - } else { - struct rusage r; - int ret; - unsigned int status; - mm_segment_t old_fs = get_fs(); - - set_fs (KERNEL_DS); - ret = sys_wait4(pid, - (stat_addr ? - (unsigned int __user *) &status : NULL), - options, (struct rusage __user *) &r); - set_fs (old_fs); - - if (ret > 0) { - if (put_compat_rusage(&r, ru)) - return -EFAULT; - if (stat_addr && put_user(status, stat_addr)) - return -EFAULT; - } - return ret; - } -} - -COMPAT_SYSCALL_DEFINE5(waitid, - int, which, compat_pid_t, pid, - struct compat_siginfo __user *, uinfo, int, options, - struct compat_rusage __user *, uru) -{ - siginfo_t info; - struct rusage ru; - long ret; - mm_segment_t old_fs = get_fs(); - - memset(&info, 0, sizeof(info)); - - set_fs(KERNEL_DS); - ret = sys_waitid(which, pid, (siginfo_t __user *)&info, options, - uru ? (struct rusage __user *)&ru : NULL); - set_fs(old_fs); - - if ((ret < 0) || (info.si_signo == 0)) - return ret; - - if (uru) { - /* sys_waitid() overwrites everything in ru */ - if (COMPAT_USE_64BIT_TIME) - ret = copy_to_user(uru, &ru, sizeof(ru)); - else - ret = put_compat_rusage(&ru, uru); - if (ret) - return -EFAULT; - } - - BUG_ON(info.si_code & __SI_MASK); - info.si_code |= __SI_CHLD; - return copy_siginfo_to_user32(uinfo, &info); -} - static int compat_get_user_cpu_mask(compat_ulong_t __user *user_mask_ptr, unsigned len, struct cpumask *new_mask) { @@ -687,192 +408,26 @@ int put_compat_itimerspec(struct compat_itimerspec __user *dst, return 0; } -COMPAT_SYSCALL_DEFINE3(timer_create, clockid_t, which_clock, - struct compat_sigevent __user *, timer_event_spec, - timer_t __user *, created_timer_id) -{ - struct sigevent __user *event = NULL; - - if (timer_event_spec) { - struct sigevent kevent; - - event = compat_alloc_user_space(sizeof(*event)); - if (get_compat_sigevent(&kevent, timer_event_spec) || - copy_to_user(event, &kevent, sizeof(*event))) - return -EFAULT; - } - - return sys_timer_create(which_clock, event, created_timer_id); -} - -COMPAT_SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, - struct compat_itimerspec __user *, new, - struct compat_itimerspec __user *, old) -{ - long err; - mm_segment_t oldfs; - struct itimerspec newts, oldts; - - if (!new) - return -EINVAL; - if (get_compat_itimerspec(&newts, new)) - return -EFAULT; - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = sys_timer_settime(timer_id, flags, - (struct itimerspec __user *) &newts, - (struct itimerspec __user *) &oldts); - set_fs(oldfs); - if (!err && old && put_compat_itimerspec(old, &oldts)) - return -EFAULT; - return err; -} - -COMPAT_SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, - struct compat_itimerspec __user *, setting) -{ - long err; - mm_segment_t oldfs; - struct itimerspec ts; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = sys_timer_gettime(timer_id, - (struct itimerspec __user *) &ts); - set_fs(oldfs); - if (!err && put_compat_itimerspec(setting, &ts)) - return -EFAULT; - return err; -} - -COMPAT_SYSCALL_DEFINE2(clock_settime, clockid_t, which_clock, - struct compat_timespec __user *, tp) -{ - long err; - mm_segment_t oldfs; - struct timespec ts; - - if (compat_get_timespec(&ts, tp)) - return -EFAULT; - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = sys_clock_settime(which_clock, - (struct timespec __user *) &ts); - set_fs(oldfs); - return err; -} - -COMPAT_SYSCALL_DEFINE2(clock_gettime, clockid_t, which_clock, - struct compat_timespec __user *, tp) -{ - long err; - mm_segment_t oldfs; - struct timespec ts; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = sys_clock_gettime(which_clock, - (struct timespec __user *) &ts); - set_fs(oldfs); - if (!err && compat_put_timespec(&ts, tp)) - return -EFAULT; - return err; -} - -COMPAT_SYSCALL_DEFINE2(clock_adjtime, clockid_t, which_clock, - struct compat_timex __user *, utp) +int get_compat_itimerspec64(struct itimerspec64 *its, + const struct compat_itimerspec __user *uits) { - struct timex txc; - mm_segment_t oldfs; - int err, ret; - err = compat_get_timex(&txc, utp); - if (err) - return err; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - ret = sys_clock_adjtime(which_clock, (struct timex __user *) &txc); - set_fs(oldfs); - - err = compat_put_timex(utp, &txc); - if (err) - return err; - - return ret; -} - -COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, - struct compat_timespec __user *, tp) -{ - long err; - mm_segment_t oldfs; - struct timespec ts; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = sys_clock_getres(which_clock, - (struct timespec __user *) &ts); - set_fs(oldfs); - if (!err && tp && compat_put_timespec(&ts, tp)) + if (__compat_get_timespec64(&its->it_interval, &uits->it_interval) || + __compat_get_timespec64(&its->it_value, &uits->it_value)) return -EFAULT; - return err; -} - -static long compat_clock_nanosleep_restart(struct restart_block *restart) -{ - long err; - mm_segment_t oldfs; - struct timespec tu; - struct compat_timespec __user *rmtp = restart->nanosleep.compat_rmtp; - - restart->nanosleep.rmtp = (struct timespec __user *) &tu; - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = clock_nanosleep_restart(restart); - set_fs(oldfs); - - if ((err == -ERESTART_RESTARTBLOCK) && rmtp && - compat_put_timespec(&tu, rmtp)) - return -EFAULT; - - if (err == -ERESTART_RESTARTBLOCK) { - restart->fn = compat_clock_nanosleep_restart; - restart->nanosleep.compat_rmtp = rmtp; - } - return err; + return 0; } +EXPORT_SYMBOL_GPL(get_compat_itimerspec64); -COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, - struct compat_timespec __user *, rqtp, - struct compat_timespec __user *, rmtp) +int put_compat_itimerspec64(const struct itimerspec64 *its, + struct compat_itimerspec __user *uits) { - long err; - mm_segment_t oldfs; - struct timespec in, out; - struct restart_block *restart; - - if (compat_get_timespec(&in, rqtp)) - return -EFAULT; - - oldfs = get_fs(); - set_fs(KERNEL_DS); - err = sys_clock_nanosleep(which_clock, flags, - (struct timespec __user *) &in, - (struct timespec __user *) &out); - set_fs(oldfs); - - if ((err == -ERESTART_RESTARTBLOCK) && rmtp && - compat_put_timespec(&out, rmtp)) + if (__compat_put_timespec64(&its->it_interval, &uits->it_interval) || + __compat_put_timespec64(&its->it_value, &uits->it_value)) return -EFAULT; - - if (err == -ERESTART_RESTARTBLOCK) { - restart = ¤t->restart_block; - restart->fn = compat_clock_nanosleep_restart; - restart->nanosleep.compat_rmtp = rmtp; - } - return err; + return 0; } +EXPORT_SYMBOL_GPL(put_compat_itimerspec64); /* * We currently only need the following fields from the sigevent @@ -898,84 +453,59 @@ int get_compat_sigevent(struct sigevent *event, long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, unsigned long bitmap_size) { - int i, j; - unsigned long m; - compat_ulong_t um; unsigned long nr_compat_longs; /* align bitmap up to nearest compat_long_t boundary */ bitmap_size = ALIGN(bitmap_size, BITS_PER_COMPAT_LONG); + nr_compat_longs = BITS_TO_COMPAT_LONGS(bitmap_size); if (!access_ok(VERIFY_READ, umask, bitmap_size / 8)) return -EFAULT; - nr_compat_longs = BITS_TO_COMPAT_LONGS(bitmap_size); - - for (i = 0; i < BITS_TO_LONGS(bitmap_size); i++) { - m = 0; - - for (j = 0; j < sizeof(m)/sizeof(um); j++) { - /* - * We dont want to read past the end of the userspace - * bitmap. We must however ensure the end of the - * kernel bitmap is zeroed. - */ - if (nr_compat_longs) { - nr_compat_longs--; - if (__get_user(um, umask)) - return -EFAULT; - } else { - um = 0; - } - - umask++; - m |= (long)um << (j * BITS_PER_COMPAT_LONG); - } - *mask++ = m; + user_access_begin(); + while (nr_compat_longs > 1) { + compat_ulong_t l1, l2; + unsafe_get_user(l1, umask++, Efault); + unsafe_get_user(l2, umask++, Efault); + *mask++ = ((unsigned long)l2 << BITS_PER_COMPAT_LONG) | l1; + nr_compat_longs -= 2; } - + if (nr_compat_longs) + unsafe_get_user(*mask, umask++, Efault); + user_access_end(); return 0; + +Efault: + user_access_end(); + return -EFAULT; } long compat_put_bitmap(compat_ulong_t __user *umask, unsigned long *mask, unsigned long bitmap_size) { - int i, j; - unsigned long m; - compat_ulong_t um; unsigned long nr_compat_longs; /* align bitmap up to nearest compat_long_t boundary */ bitmap_size = ALIGN(bitmap_size, BITS_PER_COMPAT_LONG); + nr_compat_longs = BITS_TO_COMPAT_LONGS(bitmap_size); if (!access_ok(VERIFY_WRITE, umask, bitmap_size / 8)) return -EFAULT; - nr_compat_longs = BITS_TO_COMPAT_LONGS(bitmap_size); - - for (i = 0; i < BITS_TO_LONGS(bitmap_size); i++) { - m = *mask++; - - for (j = 0; j < sizeof(m)/sizeof(um); j++) { - um = m; - - /* - * We dont want to write past the end of the userspace - * bitmap. - */ - if (nr_compat_longs) { - nr_compat_longs--; - if (__put_user(um, umask)) - return -EFAULT; - } - - umask++; - m >>= 4*sizeof(um); - m >>= 4*sizeof(um); - } + user_access_begin(); + while (nr_compat_longs > 1) { + unsigned long m = *mask++; + unsafe_put_user((compat_ulong_t)m, umask++, Efault); + unsafe_put_user(m >> BITS_PER_COMPAT_LONG, umask++, Efault); + nr_compat_longs -= 2; } - + if (nr_compat_longs) + unsafe_put_user((compat_ulong_t)*mask, umask++, Efault); + user_access_end(); return 0; +Efault: + user_access_end(); + return -EFAULT; } void @@ -1001,96 +531,6 @@ sigset_to_compat(compat_sigset_t *compat, const sigset_t *set) } } -COMPAT_SYSCALL_DEFINE4(rt_sigtimedwait, compat_sigset_t __user *, uthese, - struct compat_siginfo __user *, uinfo, - struct compat_timespec __user *, uts, compat_size_t, sigsetsize) -{ - compat_sigset_t s32; - sigset_t s; - struct timespec t; - siginfo_t info; - long ret; - - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&s32, uthese, sizeof(compat_sigset_t))) - return -EFAULT; - sigset_from_compat(&s, &s32); - - if (uts) { - if (compat_get_timespec(&t, uts)) - return -EFAULT; - } - - ret = do_sigtimedwait(&s, &info, uts ? &t : NULL); - - if (ret > 0 && uinfo) { - if (copy_siginfo_to_user32(uinfo, &info)) - ret = -EFAULT; - } - - return ret; -} - -#ifdef __ARCH_WANT_COMPAT_SYS_TIME - -/* compat_time_t is a 32 bit "long" and needs to get converted. */ - -COMPAT_SYSCALL_DEFINE1(time, compat_time_t __user *, tloc) -{ - compat_time_t i; - struct timeval tv; - - do_gettimeofday(&tv); - i = tv.tv_sec; - - if (tloc) { - if (put_user(i,tloc)) - return -EFAULT; - } - force_successful_syscall_return(); - return i; -} - -COMPAT_SYSCALL_DEFINE1(stime, compat_time_t __user *, tptr) -{ - struct timespec tv; - int err; - - if (get_user(tv.tv_sec, tptr)) - return -EFAULT; - - tv.tv_nsec = 0; - - err = security_settime(&tv, NULL); - if (err) - return err; - - do_settimeofday(&tv); - return 0; -} - -#endif /* __ARCH_WANT_COMPAT_SYS_TIME */ - -COMPAT_SYSCALL_DEFINE1(adjtimex, struct compat_timex __user *, utp) -{ - struct timex txc; - int err, ret; - - err = compat_get_timex(&txc, utp); - if (err) - return err; - - ret = do_adjtimex(&txc); - - err = compat_put_timex(utp, &txc); - if (err) - return err; - - return ret; -} - #ifdef CONFIG_NUMA COMPAT_SYSCALL_DEFINE6(move_pages, pid_t, pid, compat_ulong_t, nr_pages, compat_uptr_t __user *, pages32, diff --git a/kernel/configs/android-base.config b/kernel/configs/android-base.config index 1a8f34f63601..d70829033bb7 100644 --- a/kernel/configs/android-base.config +++ b/kernel/configs/android-base.config @@ -1,10 +1,13 @@ # KEEP ALPHABETICALLY SORTED # CONFIG_DEVKMEM is not set # CONFIG_DEVMEM is not set +# CONFIG_FHANDLE is not set # CONFIG_INET_LRO is not set -# CONFIG_MODULES is not set +# CONFIG_NFSD is not set +# CONFIG_NFS_FS is not set # CONFIG_OABI_COMPAT is not set # CONFIG_SYSVIPC is not set +# CONFIG_USELIB is not set CONFIG_ANDROID=y CONFIG_ANDROID_BINDER_IPC=y CONFIG_ANDROID_LOW_MEMORY_KILLER=y @@ -13,6 +16,7 @@ CONFIG_ASHMEM=y CONFIG_AUDIT=y CONFIG_BLK_DEV_INITRD=y CONFIG_CGROUPS=y +CONFIG_CGROUP_BPF=y CONFIG_CGROUP_CPUACCT=y CONFIG_CGROUP_DEBUG=y CONFIG_CGROUP_FREEZER=y @@ -21,7 +25,10 @@ CONFIG_CP15_BARRIER_EMULATION=y CONFIG_DEFAULT_SECURITY_SELINUX=y CONFIG_EMBEDDED=y CONFIG_FB=y +CONFIG_HARDENED_USERCOPY=y CONFIG_HIGH_RES_TIMERS=y +CONFIG_IKCONFIG=y +CONFIG_IKCONFIG_PROC=y CONFIG_INET6_AH=y CONFIG_INET6_ESP=y CONFIG_INET6_IPCOMP=y @@ -59,6 +66,9 @@ CONFIG_IP_NF_TARGET_MASQUERADE=y CONFIG_IP_NF_TARGET_NETMAP=y CONFIG_IP_NF_TARGET_REDIRECT=y CONFIG_IP_NF_TARGET_REJECT=y +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +CONFIG_MODVERSIONS=y CONFIG_NET=y CONFIG_NETDEVICES=y CONFIG_NETFILTER=y @@ -129,6 +139,7 @@ CONFIG_PPP_DEFLATE=y CONFIG_PPP_MPPE=y CONFIG_PREEMPT=y CONFIG_QUOTA=y +CONFIG_RANDOMIZE_BASE=y CONFIG_RTC_CLASS=y CONFIG_RT_GROUP_SCHED=y CONFIG_SECCOMP=y diff --git a/kernel/configs/android-recommended.config b/kernel/configs/android-recommended.config index 297756be369c..946fb92418f7 100644 --- a/kernel/configs/android-recommended.config +++ b/kernel/configs/android-recommended.config @@ -1,17 +1,20 @@ # KEEP ALPHABETICALLY SORTED +# CONFIG_AIO is not set # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set # CONFIG_INPUT_MOUSE is not set # CONFIG_LEGACY_PTYS is not set # CONFIG_NF_CONNTRACK_SIP is not set # CONFIG_PM_WAKELOCKS_GC is not set # CONFIG_VT is not set +CONFIG_ARM64_SW_TTBR0_PAN=y CONFIG_BACKLIGHT_LCD_SUPPORT=y CONFIG_BLK_DEV_DM=y CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_SIZE=8192 +CONFIG_CC_STACKPROTECTOR_STRONG=y CONFIG_COMPACTION=y -CONFIG_DEBUG_RODATA=y +CONFIG_CPU_SW_DOMAIN_PAN=y CONFIG_DM_CRYPT=y CONFIG_DM_UEVENT=y CONFIG_DM_VERITY=y @@ -104,6 +107,7 @@ CONFIG_SCHEDSTATS=y CONFIG_SMARTJOYPLUS_FF=y CONFIG_SND=y CONFIG_SOUND=y +CONFIG_STRICT_KERNEL_RWX=y CONFIG_SUSPEND_TIME=y CONFIG_TABLET_USB_ACECAD=y CONFIG_TABLET_USB_AIPTEK=y diff --git a/kernel/cpu.c b/kernel/cpu.c index 0a5f630f5c54..eee033134262 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -7,7 +7,9 @@ #include <linux/smp.h> #include <linux/init.h> #include <linux/notifier.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/hotplug.h> +#include <linux/sched/task.h> #include <linux/unistd.h> #include <linux/cpu.h> #include <linux/oom.h> @@ -25,6 +27,7 @@ #include <linux/smpboot.h> #include <linux/relay.h> #include <linux/slab.h> +#include <linux/percpu-rwsem.h> #include <trace/events/power.h> #define CREATE_TRACE_POINTS @@ -63,6 +66,12 @@ struct cpuhp_cpu_state { static DEFINE_PER_CPU(struct cpuhp_cpu_state, cpuhp_state); +#if defined(CONFIG_LOCKDEP) && defined(CONFIG_SMP) +static struct lock_class_key cpuhp_state_key; +static struct lockdep_map cpuhp_state_lock_map = + STATIC_LOCKDEP_MAP_INIT("cpuhp_state", &cpuhp_state_key); +#endif + /** * cpuhp_step - Hotplug state machine step * @name: Name of the step @@ -194,121 +203,41 @@ void cpu_maps_update_done(void) mutex_unlock(&cpu_add_remove_lock); } -/* If set, cpu_up and cpu_down will return -EBUSY and do nothing. +/* + * If set, cpu_up and cpu_down will return -EBUSY and do nothing. * Should always be manipulated under cpu_add_remove_lock */ static int cpu_hotplug_disabled; #ifdef CONFIG_HOTPLUG_CPU -static struct { - struct task_struct *active_writer; - /* wait queue to wake up the active_writer */ - wait_queue_head_t wq; - /* verifies that no writer will get active while readers are active */ - struct mutex lock; - /* - * Also blocks the new readers during - * an ongoing cpu hotplug operation. - */ - atomic_t refcount; +DEFINE_STATIC_PERCPU_RWSEM(cpu_hotplug_lock); -#ifdef CONFIG_DEBUG_LOCK_ALLOC - struct lockdep_map dep_map; -#endif -} cpu_hotplug = { - .active_writer = NULL, - .wq = __WAIT_QUEUE_HEAD_INITIALIZER(cpu_hotplug.wq), - .lock = __MUTEX_INITIALIZER(cpu_hotplug.lock), -#ifdef CONFIG_DEBUG_LOCK_ALLOC - .dep_map = STATIC_LOCKDEP_MAP_INIT("cpu_hotplug.dep_map", &cpu_hotplug.dep_map), -#endif -}; - -/* Lockdep annotations for get/put_online_cpus() and cpu_hotplug_begin/end() */ -#define cpuhp_lock_acquire_read() lock_map_acquire_read(&cpu_hotplug.dep_map) -#define cpuhp_lock_acquire_tryread() \ - lock_map_acquire_tryread(&cpu_hotplug.dep_map) -#define cpuhp_lock_acquire() lock_map_acquire(&cpu_hotplug.dep_map) -#define cpuhp_lock_release() lock_map_release(&cpu_hotplug.dep_map) - - -void get_online_cpus(void) +void cpus_read_lock(void) { - might_sleep(); - if (cpu_hotplug.active_writer == current) - return; - cpuhp_lock_acquire_read(); - mutex_lock(&cpu_hotplug.lock); - atomic_inc(&cpu_hotplug.refcount); - mutex_unlock(&cpu_hotplug.lock); + percpu_down_read(&cpu_hotplug_lock); } -EXPORT_SYMBOL_GPL(get_online_cpus); +EXPORT_SYMBOL_GPL(cpus_read_lock); -void put_online_cpus(void) +void cpus_read_unlock(void) { - int refcount; - - if (cpu_hotplug.active_writer == current) - return; - - refcount = atomic_dec_return(&cpu_hotplug.refcount); - if (WARN_ON(refcount < 0)) /* try to fix things up */ - atomic_inc(&cpu_hotplug.refcount); - - if (refcount <= 0 && waitqueue_active(&cpu_hotplug.wq)) - wake_up(&cpu_hotplug.wq); - - cpuhp_lock_release(); - + percpu_up_read(&cpu_hotplug_lock); } -EXPORT_SYMBOL_GPL(put_online_cpus); +EXPORT_SYMBOL_GPL(cpus_read_unlock); -/* - * This ensures that the hotplug operation can begin only when the - * refcount goes to zero. - * - * Note that during a cpu-hotplug operation, the new readers, if any, - * will be blocked by the cpu_hotplug.lock - * - * Since cpu_hotplug_begin() is always called after invoking - * cpu_maps_update_begin(), we can be sure that only one writer is active. - * - * Note that theoretically, there is a possibility of a livelock: - * - Refcount goes to zero, last reader wakes up the sleeping - * writer. - * - Last reader unlocks the cpu_hotplug.lock. - * - A new reader arrives at this moment, bumps up the refcount. - * - The writer acquires the cpu_hotplug.lock finds the refcount - * non zero and goes to sleep again. - * - * However, this is very difficult to achieve in practice since - * get_online_cpus() not an api which is called all that often. - * - */ -void cpu_hotplug_begin(void) +void cpus_write_lock(void) { - DEFINE_WAIT(wait); - - cpu_hotplug.active_writer = current; - cpuhp_lock_acquire(); + percpu_down_write(&cpu_hotplug_lock); +} - for (;;) { - mutex_lock(&cpu_hotplug.lock); - prepare_to_wait(&cpu_hotplug.wq, &wait, TASK_UNINTERRUPTIBLE); - if (likely(!atomic_read(&cpu_hotplug.refcount))) - break; - mutex_unlock(&cpu_hotplug.lock); - schedule(); - } - finish_wait(&cpu_hotplug.wq, &wait); +void cpus_write_unlock(void) +{ + percpu_up_write(&cpu_hotplug_lock); } -void cpu_hotplug_done(void) +void lockdep_assert_cpus_held(void) { - cpu_hotplug.active_writer = NULL; - mutex_unlock(&cpu_hotplug.lock); - cpuhp_lock_release(); + percpu_rwsem_assert_held(&cpu_hotplug_lock); } /* @@ -342,13 +271,26 @@ void cpu_hotplug_enable(void) EXPORT_SYMBOL_GPL(cpu_hotplug_enable); #endif /* CONFIG_HOTPLUG_CPU */ -/* Notifier wrappers for transitioning to state machine */ +static void __cpuhp_kick_ap_work(struct cpuhp_cpu_state *st); static int bringup_wait_for_ap(unsigned int cpu) { struct cpuhp_cpu_state *st = per_cpu_ptr(&cpuhp_state, cpu); + /* Wait for the CPU to reach CPUHP_AP_ONLINE_IDLE */ wait_for_completion(&st->done); + if (WARN_ON_ONCE((!cpu_online(cpu)))) + return -ECANCELED; + + /* Unpark the stopper thread and the hotplug thread of the target cpu */ + stop_machine_unpark(cpu); + kthread_unpark(st->thread); + + /* Should we go further up ? */ + if (st->target > CPUHP_AP_ONLINE_IDLE) { + __cpuhp_kick_ap_work(st); + wait_for_completion(&st->done); + } return st->result; } @@ -369,9 +311,7 @@ static int bringup_cpu(unsigned int cpu) irq_unlock_sparse(); if (ret) return ret; - ret = bringup_wait_for_ap(cpu); - BUG_ON(!cpu_online(cpu)); - return ret; + return bringup_wait_for_ap(cpu); } /* @@ -482,6 +422,7 @@ static void cpuhp_thread_fun(unsigned int cpu) st->should_run = false; + lock_map_acquire(&cpuhp_state_lock_map); /* Single callback invocation for [un]install ? */ if (st->single) { if (st->cb_state < CPUHP_AP_ONLINE) { @@ -508,6 +449,7 @@ static void cpuhp_thread_fun(unsigned int cpu) else if (st->state > st->target) ret = cpuhp_ap_offline(cpu, st); } + lock_map_release(&cpuhp_state_lock_map); st->result = ret; complete(&st->done); } @@ -522,6 +464,9 @@ cpuhp_invoke_ap_callback(int cpu, enum cpuhp_state state, bool bringup, if (!cpu_online(cpu)) return 0; + lock_map_acquire(&cpuhp_state_lock_map); + lock_map_release(&cpuhp_state_lock_map); + /* * If we are up and running, use the hotplug thread. For early calls * we invoke the thread function directly. @@ -565,6 +510,8 @@ static int cpuhp_kick_ap_work(unsigned int cpu) enum cpuhp_state state = st->state; trace_cpuhp_enter(cpu, st->target, state, cpuhp_kick_ap_work); + lock_map_acquire(&cpuhp_state_lock_map); + lock_map_release(&cpuhp_state_lock_map); __cpuhp_kick_ap_work(st); wait_for_completion(&st->done); trace_cpuhp_exit(cpu, st->state, state, st->result); @@ -628,30 +575,6 @@ void clear_tasks_mm_cpumask(int cpu) rcu_read_unlock(); } -static inline void check_for_tasks(int dead_cpu) -{ - struct task_struct *g, *p; - - read_lock(&tasklist_lock); - for_each_process_thread(g, p) { - if (!p->on_rq) - continue; - /* - * We do the check with unlocked task_rq(p)->lock. - * Order the reading to do not warn about a task, - * which was running on this cpu in the past, and - * it's just been woken on another cpu. - */ - rmb(); - if (task_cpu(p) != dead_cpu) - continue; - - pr_warn("Task %s (pid=%d) is on cpu %d (state=%ld, flags=%x)\n", - p->comm, task_pid_nr(p), dead_cpu, p->state, p->flags); - } - read_unlock(&tasklist_lock); -} - /* Take this CPU down. */ static int take_cpu_down(void *_param) { @@ -699,7 +622,7 @@ static int takedown_cpu(unsigned int cpu) /* * So now all preempt/rcu users must observe !cpu_active(). */ - err = stop_machine(take_cpu_down, NULL, cpumask_of(cpu)); + err = stop_machine_cpuslocked(take_cpu_down, NULL, cpumask_of(cpu)); if (err) { /* CPU refused to die */ irq_unlock_sparse(); @@ -771,7 +694,7 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen, if (!cpu_present(cpu)) return -EINVAL; - cpu_hotplug_begin(); + cpus_write_lock(); cpuhp_tasks_frozen = tasks_frozen; @@ -809,7 +732,7 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen, } out: - cpu_hotplug_done(); + cpus_write_unlock(); return ret; } @@ -857,31 +780,20 @@ void notify_cpu_starting(unsigned int cpu) } /* - * Called from the idle task. We need to set active here, so we can kick off - * the stopper thread and unpark the smpboot threads. If the target state is - * beyond CPUHP_AP_ONLINE_IDLE we kick cpuhp thread and let it bring up the - * cpu further. + * Called from the idle task. Wake up the controlling task which brings the + * stopper and the hotplug thread of the upcoming CPU up and then delegates + * the rest of the online bringup to the hotplug thread. */ void cpuhp_online_idle(enum cpuhp_state state) { struct cpuhp_cpu_state *st = this_cpu_ptr(&cpuhp_state); - unsigned int cpu = smp_processor_id(); /* Happens for the boot cpu */ if (state != CPUHP_AP_ONLINE_IDLE) return; st->state = CPUHP_AP_ONLINE_IDLE; - - /* Unpark the stopper thread and the hotplug thread of this cpu */ - stop_machine_unpark(cpu); - kthread_unpark(st->thread); - - /* Should we go further up ? */ - if (st->target > CPUHP_AP_ONLINE_IDLE) - __cpuhp_kick_ap_work(st); - else - complete(&st->done); + complete(&st->done); } /* Requires cpu_add_remove_lock to be held */ @@ -891,7 +803,7 @@ static int _cpu_up(unsigned int cpu, int tasks_frozen, enum cpuhp_state target) struct task_struct *idle; int ret = 0; - cpu_hotplug_begin(); + cpus_write_lock(); if (!cpu_present(cpu)) { ret = -EINVAL; @@ -939,7 +851,7 @@ static int _cpu_up(unsigned int cpu, int tasks_frozen, enum cpuhp_state target) target = min((int)target, CPUHP_BRINGUP_CPU); ret = cpuhp_up_callbacks(cpu, st, target); out: - cpu_hotplug_done(); + cpus_write_unlock(); return ret; } @@ -1123,6 +1035,8 @@ core_initcall(cpu_hotplug_pm_sync_init); #endif /* CONFIG_PM_SLEEP_SMP */ +int __boot_cpu_id; + #endif /* CONFIG_SMP */ /* Boot processor state steps */ @@ -1248,6 +1162,11 @@ static struct cpuhp_step cpuhp_ap_states[] = { .startup.single = smpboot_unpark_threads, .teardown.single = NULL, }, + [CPUHP_AP_IRQ_AFFINITY_ONLINE] = { + .name = "irq/affinity:online", + .startup.single = irq_affinity_online_cpu, + .teardown.single = NULL, + }, [CPUHP_AP_PERF_ONLINE] = { .name = "perf:online", .startup.single = perf_event_init_cpu, @@ -1333,26 +1252,21 @@ static int cpuhp_store_callbacks(enum cpuhp_state state, const char *name, struct cpuhp_step *sp; int ret = 0; - mutex_lock(&cpuhp_state_mutex); - if (state == CPUHP_AP_ONLINE_DYN || state == CPUHP_BP_PREPARE_DYN) { ret = cpuhp_reserve_state(state); if (ret < 0) - goto out; + return ret; state = ret; } sp = cpuhp_get_step(state); - if (name && sp->name) { - ret = -EBUSY; - goto out; - } + if (name && sp->name) + return -EBUSY; + sp->startup.single = startup; sp->teardown.single = teardown; sp->name = name; sp->multi_instance = multi_instance; INIT_HLIST_HEAD(&sp->list); -out: - mutex_unlock(&cpuhp_state_mutex); return ret; } @@ -1414,18 +1328,21 @@ static void cpuhp_rollback_install(int failedcpu, enum cpuhp_state state, } } -int __cpuhp_state_add_instance(enum cpuhp_state state, struct hlist_node *node, - bool invoke) +int __cpuhp_state_add_instance_cpuslocked(enum cpuhp_state state, + struct hlist_node *node, + bool invoke) { struct cpuhp_step *sp; int cpu; int ret; + lockdep_assert_cpus_held(); + sp = cpuhp_get_step(state); if (sp->multi_instance == false) return -EINVAL; - get_online_cpus(); + mutex_lock(&cpuhp_state_mutex); if (!invoke || !sp->startup.multi) goto add_node; @@ -1445,23 +1362,31 @@ int __cpuhp_state_add_instance(enum cpuhp_state state, struct hlist_node *node, if (ret) { if (sp->teardown.multi) cpuhp_rollback_install(cpu, state, node); - goto err; + goto unlock; } } add_node: ret = 0; - mutex_lock(&cpuhp_state_mutex); hlist_add_head(node, &sp->list); +unlock: mutex_unlock(&cpuhp_state_mutex); + return ret; +} -err: - put_online_cpus(); +int __cpuhp_state_add_instance(enum cpuhp_state state, struct hlist_node *node, + bool invoke) +{ + int ret; + + cpus_read_lock(); + ret = __cpuhp_state_add_instance_cpuslocked(state, node, invoke); + cpus_read_unlock(); return ret; } EXPORT_SYMBOL_GPL(__cpuhp_state_add_instance); /** - * __cpuhp_setup_state - Setup the callbacks for an hotplug machine state + * __cpuhp_setup_state_cpuslocked - Setup the callbacks for an hotplug machine state * @state: The state to setup * @invoke: If true, the startup function is invoked for cpus where * cpu state >= @state @@ -1470,25 +1395,28 @@ EXPORT_SYMBOL_GPL(__cpuhp_state_add_instance); * @multi_instance: State is set up for multiple instances which get * added afterwards. * + * The caller needs to hold cpus read locked while calling this function. * Returns: * On success: * Positive state number if @state is CPUHP_AP_ONLINE_DYN * 0 for all other states * On failure: proper (negative) error code */ -int __cpuhp_setup_state(enum cpuhp_state state, - const char *name, bool invoke, - int (*startup)(unsigned int cpu), - int (*teardown)(unsigned int cpu), - bool multi_instance) +int __cpuhp_setup_state_cpuslocked(enum cpuhp_state state, + const char *name, bool invoke, + int (*startup)(unsigned int cpu), + int (*teardown)(unsigned int cpu), + bool multi_instance) { int cpu, ret = 0; bool dynstate; + lockdep_assert_cpus_held(); + if (cpuhp_cb_check(state) || !name) return -EINVAL; - get_online_cpus(); + mutex_lock(&cpuhp_state_mutex); ret = cpuhp_store_callbacks(state, name, startup, teardown, multi_instance); @@ -1522,7 +1450,7 @@ int __cpuhp_setup_state(enum cpuhp_state state, } } out: - put_online_cpus(); + mutex_unlock(&cpuhp_state_mutex); /* * If the requested state is CPUHP_AP_ONLINE_DYN, return the * dynamically allocated state in case of success. @@ -1531,6 +1459,22 @@ out: return state; return ret; } +EXPORT_SYMBOL(__cpuhp_setup_state_cpuslocked); + +int __cpuhp_setup_state(enum cpuhp_state state, + const char *name, bool invoke, + int (*startup)(unsigned int cpu), + int (*teardown)(unsigned int cpu), + bool multi_instance) +{ + int ret; + + cpus_read_lock(); + ret = __cpuhp_setup_state_cpuslocked(state, name, invoke, startup, + teardown, multi_instance); + cpus_read_unlock(); + return ret; +} EXPORT_SYMBOL(__cpuhp_setup_state); int __cpuhp_state_remove_instance(enum cpuhp_state state, @@ -1544,7 +1488,9 @@ int __cpuhp_state_remove_instance(enum cpuhp_state state, if (!sp->multi_instance) return -EINVAL; - get_online_cpus(); + cpus_read_lock(); + mutex_lock(&cpuhp_state_mutex); + if (!invoke || !cpuhp_get_teardown_cb(state)) goto remove; /* @@ -1561,32 +1507,34 @@ int __cpuhp_state_remove_instance(enum cpuhp_state state, } remove: - mutex_lock(&cpuhp_state_mutex); hlist_del(node); mutex_unlock(&cpuhp_state_mutex); - put_online_cpus(); + cpus_read_unlock(); return 0; } EXPORT_SYMBOL_GPL(__cpuhp_state_remove_instance); + /** - * __cpuhp_remove_state - Remove the callbacks for an hotplug machine state + * __cpuhp_remove_state_cpuslocked - Remove the callbacks for an hotplug machine state * @state: The state to remove * @invoke: If true, the teardown function is invoked for cpus where * cpu state >= @state * + * The caller needs to hold cpus read locked while calling this function. * The teardown callback is currently not allowed to fail. Think * about module removal! */ -void __cpuhp_remove_state(enum cpuhp_state state, bool invoke) +void __cpuhp_remove_state_cpuslocked(enum cpuhp_state state, bool invoke) { struct cpuhp_step *sp = cpuhp_get_step(state); int cpu; BUG_ON(cpuhp_cb_check(state)); - get_online_cpus(); + lockdep_assert_cpus_held(); + mutex_lock(&cpuhp_state_mutex); if (sp->multi_instance) { WARN(!hlist_empty(&sp->list), "Error: Removing state %d which has instances left.\n", @@ -1611,7 +1559,15 @@ void __cpuhp_remove_state(enum cpuhp_state state, bool invoke) } remove: cpuhp_store_callbacks(state, NULL, NULL, NULL, false); - put_online_cpus(); + mutex_unlock(&cpuhp_state_mutex); +} +EXPORT_SYMBOL(__cpuhp_remove_state_cpuslocked); + +void __cpuhp_remove_state(enum cpuhp_state state, bool invoke) +{ + cpus_read_lock(); + __cpuhp_remove_state_cpuslocked(state, invoke); + cpus_read_unlock(); } EXPORT_SYMBOL(__cpuhp_remove_state); @@ -1654,13 +1610,13 @@ static ssize_t write_cpuhp_target(struct device *dev, ret = !sp->name || sp->cant_stop ? -EINVAL : 0; mutex_unlock(&cpuhp_state_mutex); if (ret) - return ret; + goto out; if (st->state < target) ret = do_cpu_up(dev->id, target); else ret = do_cpu_down(dev->id, target); - +out: unlock_device_hotplug(); return ret ? ret : count; } @@ -1680,7 +1636,7 @@ static struct attribute *cpuhp_cpu_attrs[] = { NULL }; -static struct attribute_group cpuhp_cpu_attr_group = { +static const struct attribute_group cpuhp_cpu_attr_group = { .attrs = cpuhp_cpu_attrs, .name = "hotplug", NULL @@ -1712,7 +1668,7 @@ static struct attribute *cpuhp_cpu_root_attrs[] = { NULL }; -static struct attribute_group cpuhp_cpu_root_attr_group = { +static const struct attribute_group cpuhp_cpu_root_attr_group = { .attrs = cpuhp_cpu_root_attrs, .name = "hotplug", NULL @@ -1813,6 +1769,10 @@ void __init boot_cpu_init(void) set_cpu_active(cpu, true); set_cpu_present(cpu, true); set_cpu_possible(cpu, true); + +#ifdef CONFIG_SMP + __boot_cpu_id = cpu; +#endif } /* diff --git a/kernel/crash_core.c b/kernel/crash_core.c new file mode 100644 index 000000000000..6db80fc0810b --- /dev/null +++ b/kernel/crash_core.c @@ -0,0 +1,471 @@ +/* + * crash.c - kernel crash support code. + * Copyright (C) 2002-2004 Eric Biederman <[email protected]> + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + */ + +#include <linux/crash_core.h> +#include <linux/utsname.h> +#include <linux/vmalloc.h> + +#include <asm/page.h> +#include <asm/sections.h> + +/* vmcoreinfo stuff */ +static unsigned char *vmcoreinfo_data; +static size_t vmcoreinfo_size; +u32 *vmcoreinfo_note; + +/* trusted vmcoreinfo, e.g. we can make a copy in the crash memory */ +static unsigned char *vmcoreinfo_data_safecopy; + +/* + * parsing the "crashkernel" commandline + * + * this code is intended to be called from architecture specific code + */ + + +/* + * This function parses command lines in the format + * + * crashkernel=ramsize-range:size[,...][@offset] + * + * The function returns 0 on success and -EINVAL on failure. + */ +static int __init parse_crashkernel_mem(char *cmdline, + unsigned long long system_ram, + unsigned long long *crash_size, + unsigned long long *crash_base) +{ + char *cur = cmdline, *tmp; + + /* for each entry of the comma-separated list */ + do { + unsigned long long start, end = ULLONG_MAX, size; + + /* get the start of the range */ + start = memparse(cur, &tmp); + if (cur == tmp) { + pr_warn("crashkernel: Memory value expected\n"); + return -EINVAL; + } + cur = tmp; + if (*cur != '-') { + pr_warn("crashkernel: '-' expected\n"); + return -EINVAL; + } + cur++; + + /* if no ':' is here, than we read the end */ + if (*cur != ':') { + end = memparse(cur, &tmp); + if (cur == tmp) { + pr_warn("crashkernel: Memory value expected\n"); + return -EINVAL; + } + cur = tmp; + if (end <= start) { + pr_warn("crashkernel: end <= start\n"); + return -EINVAL; + } + } + + if (*cur != ':') { + pr_warn("crashkernel: ':' expected\n"); + return -EINVAL; + } + cur++; + + size = memparse(cur, &tmp); + if (cur == tmp) { + pr_warn("Memory value expected\n"); + return -EINVAL; + } + cur = tmp; + if (size >= system_ram) { + pr_warn("crashkernel: invalid size\n"); + return -EINVAL; + } + + /* match ? */ + if (system_ram >= start && system_ram < end) { + *crash_size = size; + break; + } + } while (*cur++ == ','); + + if (*crash_size > 0) { + while (*cur && *cur != ' ' && *cur != '@') + cur++; + if (*cur == '@') { + cur++; + *crash_base = memparse(cur, &tmp); + if (cur == tmp) { + pr_warn("Memory value expected after '@'\n"); + return -EINVAL; + } + } + } + + return 0; +} + +/* + * That function parses "simple" (old) crashkernel command lines like + * + * crashkernel=size[@offset] + * + * It returns 0 on success and -EINVAL on failure. + */ +static int __init parse_crashkernel_simple(char *cmdline, + unsigned long long *crash_size, + unsigned long long *crash_base) +{ + char *cur = cmdline; + + *crash_size = memparse(cmdline, &cur); + if (cmdline == cur) { + pr_warn("crashkernel: memory value expected\n"); + return -EINVAL; + } + + if (*cur == '@') + *crash_base = memparse(cur+1, &cur); + else if (*cur != ' ' && *cur != '\0') { + pr_warn("crashkernel: unrecognized char: %c\n", *cur); + return -EINVAL; + } + + return 0; +} + +#define SUFFIX_HIGH 0 +#define SUFFIX_LOW 1 +#define SUFFIX_NULL 2 +static __initdata char *suffix_tbl[] = { + [SUFFIX_HIGH] = ",high", + [SUFFIX_LOW] = ",low", + [SUFFIX_NULL] = NULL, +}; + +/* + * That function parses "suffix" crashkernel command lines like + * + * crashkernel=size,[high|low] + * + * It returns 0 on success and -EINVAL on failure. + */ +static int __init parse_crashkernel_suffix(char *cmdline, + unsigned long long *crash_size, + const char *suffix) +{ + char *cur = cmdline; + + *crash_size = memparse(cmdline, &cur); + if (cmdline == cur) { + pr_warn("crashkernel: memory value expected\n"); + return -EINVAL; + } + + /* check with suffix */ + if (strncmp(cur, suffix, strlen(suffix))) { + pr_warn("crashkernel: unrecognized char: %c\n", *cur); + return -EINVAL; + } + cur += strlen(suffix); + if (*cur != ' ' && *cur != '\0') { + pr_warn("crashkernel: unrecognized char: %c\n", *cur); + return -EINVAL; + } + + return 0; +} + +static __init char *get_last_crashkernel(char *cmdline, + const char *name, + const char *suffix) +{ + char *p = cmdline, *ck_cmdline = NULL; + + /* find crashkernel and use the last one if there are more */ + p = strstr(p, name); + while (p) { + char *end_p = strchr(p, ' '); + char *q; + + if (!end_p) + end_p = p + strlen(p); + + if (!suffix) { + int i; + + /* skip the one with any known suffix */ + for (i = 0; suffix_tbl[i]; i++) { + q = end_p - strlen(suffix_tbl[i]); + if (!strncmp(q, suffix_tbl[i], + strlen(suffix_tbl[i]))) + goto next; + } + ck_cmdline = p; + } else { + q = end_p - strlen(suffix); + if (!strncmp(q, suffix, strlen(suffix))) + ck_cmdline = p; + } +next: + p = strstr(p+1, name); + } + + if (!ck_cmdline) + return NULL; + + return ck_cmdline; +} + +static int __init __parse_crashkernel(char *cmdline, + unsigned long long system_ram, + unsigned long long *crash_size, + unsigned long long *crash_base, + const char *name, + const char *suffix) +{ + char *first_colon, *first_space; + char *ck_cmdline; + + BUG_ON(!crash_size || !crash_base); + *crash_size = 0; + *crash_base = 0; + + ck_cmdline = get_last_crashkernel(cmdline, name, suffix); + + if (!ck_cmdline) + return -EINVAL; + + ck_cmdline += strlen(name); + + if (suffix) + return parse_crashkernel_suffix(ck_cmdline, crash_size, + suffix); + /* + * if the commandline contains a ':', then that's the extended + * syntax -- if not, it must be the classic syntax + */ + first_colon = strchr(ck_cmdline, ':'); + first_space = strchr(ck_cmdline, ' '); + if (first_colon && (!first_space || first_colon < first_space)) + return parse_crashkernel_mem(ck_cmdline, system_ram, + crash_size, crash_base); + + return parse_crashkernel_simple(ck_cmdline, crash_size, crash_base); +} + +/* + * That function is the entry point for command line parsing and should be + * called from the arch-specific code. + */ +int __init parse_crashkernel(char *cmdline, + unsigned long long system_ram, + unsigned long long *crash_size, + unsigned long long *crash_base) +{ + return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, + "crashkernel=", NULL); +} + +int __init parse_crashkernel_high(char *cmdline, + unsigned long long system_ram, + unsigned long long *crash_size, + unsigned long long *crash_base) +{ + return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, + "crashkernel=", suffix_tbl[SUFFIX_HIGH]); +} + +int __init parse_crashkernel_low(char *cmdline, + unsigned long long system_ram, + unsigned long long *crash_size, + unsigned long long *crash_base) +{ + return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, + "crashkernel=", suffix_tbl[SUFFIX_LOW]); +} + +Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type, + void *data, size_t data_len) +{ + struct elf_note *note = (struct elf_note *)buf; + + note->n_namesz = strlen(name) + 1; + note->n_descsz = data_len; + note->n_type = type; + buf += DIV_ROUND_UP(sizeof(*note), sizeof(Elf_Word)); + memcpy(buf, name, note->n_namesz); + buf += DIV_ROUND_UP(note->n_namesz, sizeof(Elf_Word)); + memcpy(buf, data, data_len); + buf += DIV_ROUND_UP(data_len, sizeof(Elf_Word)); + + return buf; +} + +void final_note(Elf_Word *buf) +{ + memset(buf, 0, sizeof(struct elf_note)); +} + +static void update_vmcoreinfo_note(void) +{ + u32 *buf = vmcoreinfo_note; + + if (!vmcoreinfo_size) + return; + buf = append_elf_note(buf, VMCOREINFO_NOTE_NAME, 0, vmcoreinfo_data, + vmcoreinfo_size); + final_note(buf); +} + +void crash_update_vmcoreinfo_safecopy(void *ptr) +{ + if (ptr) + memcpy(ptr, vmcoreinfo_data, vmcoreinfo_size); + + vmcoreinfo_data_safecopy = ptr; +} + +void crash_save_vmcoreinfo(void) +{ + if (!vmcoreinfo_note) + return; + + /* Use the safe copy to generate vmcoreinfo note if have */ + if (vmcoreinfo_data_safecopy) + vmcoreinfo_data = vmcoreinfo_data_safecopy; + + vmcoreinfo_append_str("CRASHTIME=%ld\n", get_seconds()); + update_vmcoreinfo_note(); +} + +void vmcoreinfo_append_str(const char *fmt, ...) +{ + va_list args; + char buf[0x50]; + size_t r; + + va_start(args, fmt); + r = vscnprintf(buf, sizeof(buf), fmt, args); + va_end(args); + + r = min(r, (size_t)VMCOREINFO_BYTES - vmcoreinfo_size); + + memcpy(&vmcoreinfo_data[vmcoreinfo_size], buf, r); + + vmcoreinfo_size += r; +} + +/* + * provide an empty default implementation here -- architecture + * code may override this + */ +void __weak arch_crash_save_vmcoreinfo(void) +{} + +phys_addr_t __weak paddr_vmcoreinfo_note(void) +{ + return __pa(vmcoreinfo_note); +} + +static int __init crash_save_vmcoreinfo_init(void) +{ + vmcoreinfo_data = (unsigned char *)get_zeroed_page(GFP_KERNEL); + if (!vmcoreinfo_data) { + pr_warn("Memory allocation for vmcoreinfo_data failed\n"); + return -ENOMEM; + } + + vmcoreinfo_note = alloc_pages_exact(VMCOREINFO_NOTE_SIZE, + GFP_KERNEL | __GFP_ZERO); + if (!vmcoreinfo_note) { + free_page((unsigned long)vmcoreinfo_data); + vmcoreinfo_data = NULL; + pr_warn("Memory allocation for vmcoreinfo_note failed\n"); + return -ENOMEM; + } + + VMCOREINFO_OSRELEASE(init_uts_ns.name.release); + VMCOREINFO_PAGESIZE(PAGE_SIZE); + + VMCOREINFO_SYMBOL(init_uts_ns); + VMCOREINFO_SYMBOL(node_online_map); +#ifdef CONFIG_MMU + VMCOREINFO_SYMBOL(swapper_pg_dir); +#endif + VMCOREINFO_SYMBOL(_stext); + VMCOREINFO_SYMBOL(vmap_area_list); + +#ifndef CONFIG_NEED_MULTIPLE_NODES + VMCOREINFO_SYMBOL(mem_map); + VMCOREINFO_SYMBOL(contig_page_data); +#endif +#ifdef CONFIG_SPARSEMEM + VMCOREINFO_SYMBOL(mem_section); + VMCOREINFO_LENGTH(mem_section, NR_SECTION_ROOTS); + VMCOREINFO_STRUCT_SIZE(mem_section); + VMCOREINFO_OFFSET(mem_section, section_mem_map); +#endif + VMCOREINFO_STRUCT_SIZE(page); + VMCOREINFO_STRUCT_SIZE(pglist_data); + VMCOREINFO_STRUCT_SIZE(zone); + VMCOREINFO_STRUCT_SIZE(free_area); + VMCOREINFO_STRUCT_SIZE(list_head); + VMCOREINFO_SIZE(nodemask_t); + VMCOREINFO_OFFSET(page, flags); + VMCOREINFO_OFFSET(page, _refcount); + VMCOREINFO_OFFSET(page, mapping); + VMCOREINFO_OFFSET(page, lru); + VMCOREINFO_OFFSET(page, _mapcount); + VMCOREINFO_OFFSET(page, private); + VMCOREINFO_OFFSET(page, compound_dtor); + VMCOREINFO_OFFSET(page, compound_order); + VMCOREINFO_OFFSET(page, compound_head); + VMCOREINFO_OFFSET(pglist_data, node_zones); + VMCOREINFO_OFFSET(pglist_data, nr_zones); +#ifdef CONFIG_FLAT_NODE_MEM_MAP + VMCOREINFO_OFFSET(pglist_data, node_mem_map); +#endif + VMCOREINFO_OFFSET(pglist_data, node_start_pfn); + VMCOREINFO_OFFSET(pglist_data, node_spanned_pages); + VMCOREINFO_OFFSET(pglist_data, node_id); + VMCOREINFO_OFFSET(zone, free_area); + VMCOREINFO_OFFSET(zone, vm_stat); + VMCOREINFO_OFFSET(zone, spanned_pages); + VMCOREINFO_OFFSET(free_area, free_list); + VMCOREINFO_OFFSET(list_head, next); + VMCOREINFO_OFFSET(list_head, prev); + VMCOREINFO_OFFSET(vmap_area, va_start); + VMCOREINFO_OFFSET(vmap_area, list); + VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER); + log_buf_vmcoreinfo_setup(); + VMCOREINFO_LENGTH(free_area.free_list, MIGRATE_TYPES); + VMCOREINFO_NUMBER(NR_FREE_PAGES); + VMCOREINFO_NUMBER(PG_lru); + VMCOREINFO_NUMBER(PG_private); + VMCOREINFO_NUMBER(PG_swapcache); + VMCOREINFO_NUMBER(PG_slab); +#ifdef CONFIG_MEMORY_FAILURE + VMCOREINFO_NUMBER(PG_hwpoison); +#endif + VMCOREINFO_NUMBER(PG_head_mask); + VMCOREINFO_NUMBER(PAGE_BUDDY_MAPCOUNT_VALUE); +#ifdef CONFIG_HUGETLB_PAGE + VMCOREINFO_NUMBER(HUGETLB_PAGE_DTOR); +#endif + + arch_crash_save_vmcoreinfo(); + update_vmcoreinfo_note(); + + return 0; +} + +subsys_initcall(crash_save_vmcoreinfo_init); diff --git a/kernel/cred.c b/kernel/cred.c index 5f264fb5737d..ecf03657e71c 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -1,4 +1,4 @@ -/* Task credentials management - see Documentation/security/credentials.txt +/* Task credentials management - see Documentation/security/credentials.rst * * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. * Written by David Howells ([email protected]) @@ -12,6 +12,7 @@ #include <linux/cred.h> #include <linux/slab.h> #include <linux/sched.h> +#include <linux/sched/coredump.h> #include <linux/key.h> #include <linux/keyctl.h> #include <linux/init_task.h> diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c index 79517e5549f1..65c0f1363788 100644 --- a/kernel/debug/debug_core.c +++ b/kernel/debug/debug_core.c @@ -49,6 +49,7 @@ #include <linux/init.h> #include <linux/kgdb.h> #include <linux/kdb.h> +#include <linux/nmi.h> #include <linux/pid.h> #include <linux/smp.h> #include <linux/mm.h> @@ -232,9 +233,9 @@ static void kgdb_flush_swbreak_addr(unsigned long addr) int i; for (i = 0; i < VMACACHE_SIZE; i++) { - if (!current->vmacache[i]) + if (!current->vmacache.vmas[i]) continue; - flush_cache_range(current->vmacache[i], + flush_cache_range(current->vmacache.vmas[i], addr, addr + BREAK_INSTR_SIZE); } } diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c index 19d9a578c753..7510dc687c0d 100644 --- a/kernel/debug/gdbstub.c +++ b/kernel/debug/gdbstub.c @@ -29,6 +29,7 @@ */ #include <linux/kernel.h> +#include <linux/sched/signal.h> #include <linux/kgdb.h> #include <linux/kdb.h> #include <linux/serial_core.h> diff --git a/kernel/debug/kdb/kdb_bt.c b/kernel/debug/kdb/kdb_bt.c index fe15fff5df53..6ad4a9fcbd6f 100644 --- a/kernel/debug/kdb/kdb_bt.c +++ b/kernel/debug/kdb/kdb_bt.c @@ -12,7 +12,8 @@ #include <linux/ctype.h> #include <linux/string.h> #include <linux/kernel.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> #include <linux/kdb.h> #include <linux/nmi.h> #include "kdb_private.h" diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index ca183919d302..c8146d53ca67 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -18,6 +18,9 @@ #include <linux/kmsg_dump.h> #include <linux/reboot.h> #include <linux/sched.h> +#include <linux/sched/loadavg.h> +#include <linux/sched/stat.h> +#include <linux/sched/debug.h> #include <linux/sysrq.h> #include <linux/smp.h> #include <linux/utsname.h> diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 435c14a45118..4a1c33416b6a 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -14,6 +14,8 @@ */ #include <linux/sched.h> +#include <linux/sched/task.h> +#include <linux/sched/cputime.h> #include <linux/slab.h> #include <linux/taskstats.h> #include <linux/time.h> @@ -82,19 +84,19 @@ void __delayacct_blkio_end(void) int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) { - cputime_t utime, stime, stimescaled, utimescaled; + u64 utime, stime, stimescaled, utimescaled; unsigned long long t2, t3; unsigned long flags, t1; s64 tmp; task_cputime(tsk, &utime, &stime); tmp = (s64)d->cpu_run_real_total; - tmp += cputime_to_nsecs(utime + stime); + tmp += utime + stime; d->cpu_run_real_total = (tmp < (s64)d->cpu_run_real_total) ? 0 : tmp; task_cputime_scaled(tsk, &utimescaled, &stimescaled); tmp = (s64)d->cpu_scaled_run_real_total; - tmp += cputime_to_nsecs(utimescaled + stimescaled); + tmp += utimescaled + stimescaled; d->cpu_scaled_run_real_total = (tmp < (s64)d->cpu_scaled_run_real_total) ? 0 : tmp; diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c index e9fdb5203de5..1b2be63c8528 100644 --- a/kernel/events/callchain.c +++ b/kernel/events/callchain.c @@ -11,6 +11,8 @@ #include <linux/perf_event.h> #include <linux/slab.h> +#include <linux/sched/task_stack.h> + #include "internal.h" struct callchain_cpus_entries { @@ -227,12 +229,18 @@ get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user, } if (regs) { + mm_segment_t fs; + if (crosstask) goto exit_put; if (add_mark) perf_callchain_store_context(&ctx, PERF_CONTEXT_USER); + + fs = get_fs(); + set_fs(USER_DS); perf_callchain_user(&ctx, regs); + set_fs(fs); } } diff --git a/kernel/events/core.c b/kernel/events/core.c index e5aaa806702d..426c2ffba16d 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -46,6 +46,10 @@ #include <linux/filter.h> #include <linux/namei.h> #include <linux/parser.h> +#include <linux/sched/clock.h> +#include <linux/sched/mm.h> +#include <linux/proc_ns.h> +#include <linux/mount.h> #include "internal.h" @@ -355,6 +359,8 @@ enum event_type_t { EVENT_FLEXIBLE = 0x1, EVENT_PINNED = 0x2, EVENT_TIME = 0x4, + /* see ctx_resched() for details */ + EVENT_CPU = 0x8, EVENT_ALL = EVENT_FLEXIBLE | EVENT_PINNED, }; @@ -375,6 +381,7 @@ static DEFINE_PER_CPU(struct pmu_event_list, pmu_sb_events); static atomic_t nr_mmap_events __read_mostly; static atomic_t nr_comm_events __read_mostly; +static atomic_t nr_namespaces_events __read_mostly; static atomic_t nr_task_events __read_mostly; static atomic_t nr_freq_events __read_mostly; static atomic_t nr_switch_events __read_mostly; @@ -382,6 +389,7 @@ static atomic_t nr_switch_events __read_mostly; static LIST_HEAD(pmus); static DEFINE_MUTEX(pmus_lock); static struct srcu_struct pmus_srcu; +static cpumask_var_t perf_online_mask; /* * perf event paranoia level: @@ -453,7 +461,7 @@ int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - int ret = proc_dointvec(table, write, buffer, lenp, ppos); + int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (ret || !write) return ret; @@ -678,6 +686,8 @@ perf_cgroup_set_timestamp(struct task_struct *task, info->timestamp = ctx->timestamp; } +static DEFINE_PER_CPU(struct list_head, cgrp_cpuctx_list); + #define PERF_CGROUP_SWOUT 0x1 /* cgroup switch out every event */ #define PERF_CGROUP_SWIN 0x2 /* cgroup switch in events based on task */ @@ -690,61 +700,46 @@ perf_cgroup_set_timestamp(struct task_struct *task, static void perf_cgroup_switch(struct task_struct *task, int mode) { struct perf_cpu_context *cpuctx; - struct pmu *pmu; + struct list_head *list; unsigned long flags; /* - * disable interrupts to avoid geting nr_cgroup - * changes via __perf_event_disable(). Also - * avoids preemption. + * Disable interrupts and preemption to avoid this CPU's + * cgrp_cpuctx_entry to change under us. */ local_irq_save(flags); - /* - * we reschedule only in the presence of cgroup - * constrained events. - */ - - list_for_each_entry_rcu(pmu, &pmus, entry) { - cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); - if (cpuctx->unique_pmu != pmu) - continue; /* ensure we process each cpuctx once */ + list = this_cpu_ptr(&cgrp_cpuctx_list); + list_for_each_entry(cpuctx, list, cgrp_cpuctx_entry) { + WARN_ON_ONCE(cpuctx->ctx.nr_cgroups == 0); - /* - * perf_cgroup_events says at least one - * context on this CPU has cgroup events. - * - * ctx->nr_cgroups reports the number of cgroup - * events for a context. - */ - if (cpuctx->ctx.nr_cgroups > 0) { - perf_ctx_lock(cpuctx, cpuctx->task_ctx); - perf_pmu_disable(cpuctx->ctx.pmu); + perf_ctx_lock(cpuctx, cpuctx->task_ctx); + perf_pmu_disable(cpuctx->ctx.pmu); - if (mode & PERF_CGROUP_SWOUT) { - cpu_ctx_sched_out(cpuctx, EVENT_ALL); - /* - * must not be done before ctxswout due - * to event_filter_match() in event_sched_out() - */ - cpuctx->cgrp = NULL; - } + if (mode & PERF_CGROUP_SWOUT) { + cpu_ctx_sched_out(cpuctx, EVENT_ALL); + /* + * must not be done before ctxswout due + * to event_filter_match() in event_sched_out() + */ + cpuctx->cgrp = NULL; + } - if (mode & PERF_CGROUP_SWIN) { - WARN_ON_ONCE(cpuctx->cgrp); - /* - * set cgrp before ctxsw in to allow - * event_filter_match() to not have to pass - * task around - * we pass the cpuctx->ctx to perf_cgroup_from_task() - * because cgorup events are only per-cpu - */ - cpuctx->cgrp = perf_cgroup_from_task(task, &cpuctx->ctx); - cpu_ctx_sched_in(cpuctx, EVENT_ALL, task); - } - perf_pmu_enable(cpuctx->ctx.pmu); - perf_ctx_unlock(cpuctx, cpuctx->task_ctx); + if (mode & PERF_CGROUP_SWIN) { + WARN_ON_ONCE(cpuctx->cgrp); + /* + * set cgrp before ctxsw in to allow + * event_filter_match() to not have to pass + * task around + * we pass the cpuctx->ctx to perf_cgroup_from_task() + * because cgorup events are only per-cpu + */ + cpuctx->cgrp = perf_cgroup_from_task(task, + &cpuctx->ctx); + cpu_ctx_sched_in(cpuctx, EVENT_ALL, task); } + perf_pmu_enable(cpuctx->ctx.pmu); + perf_ctx_unlock(cpuctx, cpuctx->task_ctx); } local_irq_restore(flags); @@ -889,6 +884,7 @@ list_update_cgroup_event(struct perf_event *event, struct perf_event_context *ctx, bool add) { struct perf_cpu_context *cpuctx; + struct list_head *cpuctx_entry; if (!is_cgroup_event(event)) return; @@ -902,15 +898,16 @@ list_update_cgroup_event(struct perf_event *event, * this will always be called from the right CPU. */ cpuctx = __get_cpu_context(ctx); - - /* - * cpuctx->cgrp is NULL until a cgroup event is sched in or - * ctx->nr_cgroup == 0 . - */ - if (add && perf_cgroup_from_task(current, ctx) == event->cgrp) - cpuctx->cgrp = event->cgrp; - else if (!add) + cpuctx_entry = &cpuctx->cgrp_cpuctx_entry; + /* cpuctx->cgrp is NULL unless a cgroup event is active in this CPU .*/ + if (add) { + list_add(cpuctx_entry, this_cpu_ptr(&cgrp_cpuctx_list)); + if (perf_cgroup_from_task(current, ctx) == event->cgrp) + cpuctx->cgrp = event->cgrp; + } else { + list_del(cpuctx_entry); cpuctx->cgrp = NULL; + } } #else /* !CONFIG_CGROUP_PERF */ @@ -929,11 +926,6 @@ static inline int is_cgroup_event(struct perf_event *event) return 0; } -static inline u64 perf_cgroup_event_cgrp_time(struct perf_event *event) -{ - return 0; -} - static inline void update_cgrp_time_from_event(struct perf_event *event) { } @@ -1005,7 +997,7 @@ list_update_cgroup_event(struct perf_event *event, */ #define PERF_CPU_HRTIMER (1000 / HZ) /* - * function must be called with interrupts disbled + * function must be called with interrupts disabled */ static enum hrtimer_restart perf_mux_hrtimer_handler(struct hrtimer *hr) { @@ -1453,6 +1445,27 @@ static void update_group_times(struct perf_event *leader) update_event_times(event); } +static enum event_type_t get_event_type(struct perf_event *event) +{ + struct perf_event_context *ctx = event->ctx; + enum event_type_t event_type; + + lockdep_assert_held(&ctx->lock); + + /* + * It's 'group type', really, because if our group leader is + * pinned, so are we. + */ + if (event->group_leader != event) + event = event->group_leader; + + event_type = event->attr.pinned ? EVENT_PINNED : EVENT_FLEXIBLE; + if (!ctx->task) + event_type |= EVENT_CPU; + + return event_type; +} + static struct list_head * ctx_group_list(struct perf_event *event, struct perf_event_context *ctx) { @@ -2226,7 +2239,8 @@ ctx_sched_in(struct perf_event_context *ctx, struct task_struct *task); static void task_ctx_sched_out(struct perf_cpu_context *cpuctx, - struct perf_event_context *ctx) + struct perf_event_context *ctx, + enum event_type_t event_type) { if (!cpuctx->task_ctx) return; @@ -2234,7 +2248,7 @@ static void task_ctx_sched_out(struct perf_cpu_context *cpuctx, if (WARN_ON_ONCE(ctx != cpuctx->task_ctx)) return; - ctx_sched_out(ctx, cpuctx, EVENT_ALL); + ctx_sched_out(ctx, cpuctx, event_type); } static void perf_event_sched_in(struct perf_cpu_context *cpuctx, @@ -2249,13 +2263,51 @@ static void perf_event_sched_in(struct perf_cpu_context *cpuctx, ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE, task); } +/* + * We want to maintain the following priority of scheduling: + * - CPU pinned (EVENT_CPU | EVENT_PINNED) + * - task pinned (EVENT_PINNED) + * - CPU flexible (EVENT_CPU | EVENT_FLEXIBLE) + * - task flexible (EVENT_FLEXIBLE). + * + * In order to avoid unscheduling and scheduling back in everything every + * time an event is added, only do it for the groups of equal priority and + * below. + * + * This can be called after a batch operation on task events, in which case + * event_type is a bit mask of the types of events involved. For CPU events, + * event_type is only either EVENT_PINNED or EVENT_FLEXIBLE. + */ static void ctx_resched(struct perf_cpu_context *cpuctx, - struct perf_event_context *task_ctx) + struct perf_event_context *task_ctx, + enum event_type_t event_type) { + enum event_type_t ctx_event_type = event_type & EVENT_ALL; + bool cpu_event = !!(event_type & EVENT_CPU); + + /* + * If pinned groups are involved, flexible groups also need to be + * scheduled out. + */ + if (event_type & EVENT_PINNED) + event_type |= EVENT_FLEXIBLE; + perf_pmu_disable(cpuctx->ctx.pmu); if (task_ctx) - task_ctx_sched_out(cpuctx, task_ctx); - cpu_ctx_sched_out(cpuctx, EVENT_ALL); + task_ctx_sched_out(cpuctx, task_ctx, event_type); + + /* + * Decide which cpu ctx groups to schedule out based on the types + * of events that caused rescheduling: + * - EVENT_CPU: schedule out corresponding groups; + * - EVENT_PINNED task events: schedule out EVENT_FLEXIBLE groups; + * - otherwise, do nothing more. + */ + if (cpu_event) + cpu_ctx_sched_out(cpuctx, ctx_event_type); + else if (ctx_event_type & EVENT_PINNED) + cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE); + perf_event_sched_in(cpuctx, task_ctx, current); perf_pmu_enable(cpuctx->ctx.pmu); } @@ -2302,7 +2354,7 @@ static int __perf_install_in_context(void *info) if (reprogram) { ctx_sched_out(ctx, cpuctx, EVENT_TIME); add_event_to_ctx(event, ctx); - ctx_resched(cpuctx, task_ctx); + ctx_resched(cpuctx, task_ctx, get_event_type(event)); } else { add_event_to_ctx(event, ctx); } @@ -2469,7 +2521,7 @@ static void __perf_event_enable(struct perf_event *event, if (ctx->task) WARN_ON_ONCE(task_ctx != ctx); - ctx_resched(cpuctx, task_ctx); + ctx_resched(cpuctx, task_ctx, get_event_type(event)); } /* @@ -2896,7 +2948,7 @@ unlock: if (do_switch) { raw_spin_lock(&ctx->lock); - task_ctx_sched_out(cpuctx, ctx); + task_ctx_sched_out(cpuctx, ctx, EVENT_ALL); raw_spin_unlock(&ctx->lock); } } @@ -2943,7 +2995,7 @@ static void perf_pmu_sched_task(struct task_struct *prev, return; list_for_each_entry(cpuctx, this_cpu_ptr(&sched_cb_list), sched_cb_entry) { - pmu = cpuctx->unique_pmu; /* software PMUs will not have sched_task */ + pmu = cpuctx->ctx.pmu; /* software PMUs will not have sched_task */ if (WARN_ON_ONCE(!pmu->sched_task)) continue; @@ -3133,8 +3185,12 @@ static void perf_event_context_sched_in(struct perf_event_context *ctx, * We want to keep the following priority order: * cpu pinned (that don't need to move), task pinned, * cpu flexible, task flexible. + * + * However, if task's ctx is not carrying any pinned + * events, no need to flip the cpuctx's events around. */ - cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE); + if (!list_empty(&ctx->pinned_groups)) + cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE); perf_event_sched_in(cpuctx, ctx, task); perf_pmu_enable(ctx->pmu); perf_ctx_unlock(cpuctx, ctx); @@ -3449,6 +3505,7 @@ static int event_enable_on_exec(struct perf_event *event, static void perf_event_enable_on_exec(int ctxn) { struct perf_event_context *ctx, *clone_ctx = NULL; + enum event_type_t event_type = 0; struct perf_cpu_context *cpuctx; struct perf_event *event; unsigned long flags; @@ -3462,15 +3519,19 @@ static void perf_event_enable_on_exec(int ctxn) cpuctx = __get_cpu_context(ctx); perf_ctx_lock(cpuctx, ctx); ctx_sched_out(ctx, cpuctx, EVENT_TIME); - list_for_each_entry(event, &ctx->event_list, event_entry) + list_for_each_entry(event, &ctx->event_list, event_entry) { enabled |= event_enable_on_exec(event, ctx); + event_type |= get_event_type(event); + } /* * Unclone and reschedule this context if we enabled any event. */ if (enabled) { clone_ctx = unclone_ctx(ctx); - ctx_resched(cpuctx, ctx); + ctx_resched(cpuctx, ctx, event_type); + } else { + ctx_sched_in(ctx, cpuctx, EVENT_TIME, current); } perf_ctx_unlock(cpuctx, ctx); @@ -3487,14 +3548,15 @@ struct perf_read_data { int ret; }; -static int find_cpu_to_read(struct perf_event *event, int local_cpu) +static int __perf_event_read_cpu(struct perf_event *event, int event_cpu) { - int event_cpu = event->oncpu; u16 local_pkg, event_pkg; if (event->group_caps & PERF_EV_CAP_READ_ACTIVE_PKG) { - event_pkg = topology_physical_package_id(event_cpu); - local_pkg = topology_physical_package_id(local_cpu); + int local_cpu = smp_processor_id(); + + event_pkg = topology_physical_package_id(event_cpu); + local_pkg = topology_physical_package_id(local_cpu); if (event_pkg == local_pkg) return local_cpu; @@ -3577,10 +3639,10 @@ static inline u64 perf_event_count(struct perf_event *event) * will not be local and we cannot read them atomically * - must not have a pmu::count method */ -u64 perf_event_read_local(struct perf_event *event) +int perf_event_read_local(struct perf_event *event, u64 *value) { unsigned long flags; - u64 val; + int ret = 0; /* * Disabling interrupts avoids all counter scheduling (context @@ -3588,25 +3650,37 @@ u64 perf_event_read_local(struct perf_event *event) */ local_irq_save(flags); - /* If this is a per-task event, it must be for current */ - WARN_ON_ONCE((event->attach_state & PERF_ATTACH_TASK) && - event->hw.target != current); - - /* If this is a per-CPU event, it must be for this CPU */ - WARN_ON_ONCE(!(event->attach_state & PERF_ATTACH_TASK) && - event->cpu != smp_processor_id()); - /* * It must not be an event with inherit set, we cannot read * all child counters from atomic context. */ - WARN_ON_ONCE(event->attr.inherit); + if (event->attr.inherit) { + ret = -EOPNOTSUPP; + goto out; + } /* * It must not have a pmu::count method, those are not * NMI safe. */ - WARN_ON_ONCE(event->pmu->count); + if (event->pmu->count) { + ret = -EOPNOTSUPP; + goto out; + } + + /* If this is a per-task event, it must be for current */ + if ((event->attach_state & PERF_ATTACH_TASK) && + event->hw.target != current) { + ret = -EINVAL; + goto out; + } + + /* If this is a per-CPU event, it must be for this CPU */ + if (!(event->attach_state & PERF_ATTACH_TASK) && + event->cpu != smp_processor_id()) { + ret = -EINVAL; + goto out; + } /* * If the event is currently on this CPU, its either a per-task event, @@ -3616,15 +3690,16 @@ u64 perf_event_read_local(struct perf_event *event) if (event->oncpu == smp_processor_id()) event->pmu->read(event); - val = local64_read(&event->count); + *value = local64_read(&event->count); +out: local_irq_restore(flags); - return val; + return ret; } static int perf_event_read(struct perf_event *event, bool group) { - int ret = 0, cpu_to_read, local_cpu; + int event_cpu, ret = 0; /* * If event is enabled and currently active on a CPU, update the @@ -3637,21 +3712,25 @@ static int perf_event_read(struct perf_event *event, bool group) .ret = 0, }; - local_cpu = get_cpu(); - cpu_to_read = find_cpu_to_read(event, local_cpu); - put_cpu(); + event_cpu = READ_ONCE(event->oncpu); + if ((unsigned)event_cpu >= nr_cpu_ids) + return 0; + + preempt_disable(); + event_cpu = __perf_event_read_cpu(event, event_cpu); /* * Purposely ignore the smp_call_function_single() return * value. * - * If event->oncpu isn't a valid CPU it means the event got + * If event_cpu isn't a valid CPU it means the event got * scheduled out and that will have updated the event count. * * Therefore, either way, we'll have an up-to-date event count * after this. */ - (void)smp_call_function_single(cpu_to_read, __perf_event_read, &data, 1); + (void)smp_call_function_single(event_cpu, __perf_event_read, &data, 1); + preempt_enable(); ret = data.ret; } else if (event->state == PERF_EVENT_STATE_INACTIVE) { struct perf_event_context *ctx = event->ctx; @@ -3749,14 +3828,6 @@ find_get_context(struct pmu *pmu, struct task_struct *task, if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) return ERR_PTR(-EACCES); - /* - * We could be clever and allow to attach a event to an - * offline CPU and activate it when the CPU comes up, but - * that's for later. - */ - if (!cpu_online(cpu)) - return ERR_PTR(-ENODEV); - cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); ctx = &cpuctx->ctx; get_ctx(ctx); @@ -3931,6 +4002,8 @@ static void unaccount_event(struct perf_event *event) atomic_dec(&nr_mmap_events); if (event->attr.comm) atomic_dec(&nr_comm_events); + if (event->attr.namespaces) + atomic_dec(&nr_namespaces_events); if (event->attr.task) atomic_dec(&nr_task_events); if (event->attr.freq) @@ -4196,7 +4269,7 @@ int perf_event_release_kernel(struct perf_event *event) raw_spin_lock_irq(&ctx->lock); /* - * Mark this even as STATE_DEAD, there is no external reference to it + * Mark this event as STATE_DEAD, there is no external reference to it * anymore. * * Anybody acquiring event->child_mutex after the below loop _must_ @@ -4312,7 +4385,9 @@ EXPORT_SYMBOL_GPL(perf_event_read_value); static int __perf_read_group_add(struct perf_event *leader, u64 read_format, u64 *values) { + struct perf_event_context *ctx = leader->ctx; struct perf_event *sub; + unsigned long flags; int n = 1; /* skip @nr */ int ret; @@ -4342,12 +4417,15 @@ static int __perf_read_group_add(struct perf_event *leader, if (read_format & PERF_FORMAT_ID) values[n++] = primary_event_id(leader); + raw_spin_lock_irqsave(&ctx->lock, flags); + list_for_each_entry(sub, &leader->sibling_list, group_entry) { values[n++] += perf_event_count(sub); if (read_format & PERF_FORMAT_ID) values[n++] = primary_event_id(sub); } + raw_spin_unlock_irqrestore(&ctx->lock, flags); return 0; } @@ -4869,9 +4947,9 @@ unlock: rcu_read_unlock(); } -static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +static int perf_mmap_fault(struct vm_fault *vmf) { - struct perf_event *event = vma->vm_file->private_data; + struct perf_event *event = vmf->vma->vm_file->private_data; struct ring_buffer *rb; int ret = VM_FAULT_SIGBUS; @@ -4894,7 +4972,7 @@ static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) goto unlock; get_page(vmf->page); - vmf->page->mapping = vma->vm_file->f_mapping; + vmf->page->mapping = vmf->vma->vm_file->f_mapping; vmf->page->index = vmf->pgoff; ret = 0; @@ -5664,9 +5742,6 @@ static void perf_output_read_one(struct perf_output_handle *handle, __output_copy(handle, values, n * sizeof(u64)); } -/* - * XXX PERF_FORMAT_GROUP vs inherited events seems difficult. - */ static void perf_output_read_group(struct perf_output_handle *handle, struct perf_event *event, u64 enabled, u64 running) @@ -5711,6 +5786,13 @@ static void perf_output_read_group(struct perf_output_handle *handle, #define PERF_FORMAT_TOTAL_TIMES (PERF_FORMAT_TOTAL_TIME_ENABLED|\ PERF_FORMAT_TOTAL_TIME_RUNNING) +/* + * XXX PERF_SAMPLE_READ vs inherited events seems difficult. + * + * The problem is that its both hard and excessively expensive to iterate the + * child list, not to mention that its impossible to IPI the children running + * on another CPU, from interrupt/NMI context. + */ static void perf_output_read(struct perf_output_handle *handle, struct perf_event *event) { @@ -6431,6 +6513,7 @@ static void perf_event_task(struct task_struct *task, void perf_event_fork(struct task_struct *task) { perf_event_task(task, NULL, 1); + perf_event_namespaces(task); } /* @@ -6533,6 +6616,132 @@ void perf_event_comm(struct task_struct *task, bool exec) } /* + * namespaces tracking + */ + +struct perf_namespaces_event { + struct task_struct *task; + + struct { + struct perf_event_header header; + + u32 pid; + u32 tid; + u64 nr_namespaces; + struct perf_ns_link_info link_info[NR_NAMESPACES]; + } event_id; +}; + +static int perf_event_namespaces_match(struct perf_event *event) +{ + return event->attr.namespaces; +} + +static void perf_event_namespaces_output(struct perf_event *event, + void *data) +{ + struct perf_namespaces_event *namespaces_event = data; + struct perf_output_handle handle; + struct perf_sample_data sample; + int ret; + + if (!perf_event_namespaces_match(event)) + return; + + perf_event_header__init_id(&namespaces_event->event_id.header, + &sample, event); + ret = perf_output_begin(&handle, event, + namespaces_event->event_id.header.size); + if (ret) + return; + + namespaces_event->event_id.pid = perf_event_pid(event, + namespaces_event->task); + namespaces_event->event_id.tid = perf_event_tid(event, + namespaces_event->task); + + perf_output_put(&handle, namespaces_event->event_id); + + perf_event__output_id_sample(event, &handle, &sample); + + perf_output_end(&handle); +} + +static void perf_fill_ns_link_info(struct perf_ns_link_info *ns_link_info, + struct task_struct *task, + const struct proc_ns_operations *ns_ops) +{ + struct path ns_path; + struct inode *ns_inode; + void *error; + + error = ns_get_path(&ns_path, task, ns_ops); + if (!error) { + ns_inode = ns_path.dentry->d_inode; + ns_link_info->dev = new_encode_dev(ns_inode->i_sb->s_dev); + ns_link_info->ino = ns_inode->i_ino; + } +} + +void perf_event_namespaces(struct task_struct *task) +{ + struct perf_namespaces_event namespaces_event; + struct perf_ns_link_info *ns_link_info; + + if (!atomic_read(&nr_namespaces_events)) + return; + + namespaces_event = (struct perf_namespaces_event){ + .task = task, + .event_id = { + .header = { + .type = PERF_RECORD_NAMESPACES, + .misc = 0, + .size = sizeof(namespaces_event.event_id), + }, + /* .pid */ + /* .tid */ + .nr_namespaces = NR_NAMESPACES, + /* .link_info[NR_NAMESPACES] */ + }, + }; + + ns_link_info = namespaces_event.event_id.link_info; + + perf_fill_ns_link_info(&ns_link_info[MNT_NS_INDEX], + task, &mntns_operations); + +#ifdef CONFIG_USER_NS + perf_fill_ns_link_info(&ns_link_info[USER_NS_INDEX], + task, &userns_operations); +#endif +#ifdef CONFIG_NET_NS + perf_fill_ns_link_info(&ns_link_info[NET_NS_INDEX], + task, &netns_operations); +#endif +#ifdef CONFIG_UTS_NS + perf_fill_ns_link_info(&ns_link_info[UTS_NS_INDEX], + task, &utsns_operations); +#endif +#ifdef CONFIG_IPC_NS + perf_fill_ns_link_info(&ns_link_info[IPC_NS_INDEX], + task, &ipcns_operations); +#endif +#ifdef CONFIG_PID_NS + perf_fill_ns_link_info(&ns_link_info[PID_NS_INDEX], + task, &pidns_operations); +#endif +#ifdef CONFIG_CGROUPS + perf_fill_ns_link_info(&ns_link_info[CGROUP_NS_INDEX], + task, &cgroupns_operations); +#endif + + perf_iterate_sb(perf_event_namespaces_output, + &namespaces_event, + NULL); +} + +/* * mmap tracking */ @@ -7511,7 +7720,8 @@ static int swevent_hlist_get_cpu(int cpu) int err = 0; mutex_lock(&swhash->hlist_mutex); - if (!swevent_hlist_deref(swhash) && cpu_online(cpu)) { + if (!swevent_hlist_deref(swhash) && + cpumask_test_cpu(cpu, perf_online_mask)) { struct swevent_hlist *hlist; hlist = kzalloc(sizeof(*hlist), GFP_KERNEL); @@ -7532,7 +7742,7 @@ static int swevent_hlist_get(void) { int err, cpu, failed_cpu; - get_online_cpus(); + mutex_lock(&pmus_lock); for_each_possible_cpu(cpu) { err = swevent_hlist_get_cpu(cpu); if (err) { @@ -7540,8 +7750,7 @@ static int swevent_hlist_get(void) goto fail; } } - put_online_cpus(); - + mutex_unlock(&pmus_lock); return 0; fail: for_each_possible_cpu(cpu) { @@ -7549,8 +7758,7 @@ fail: break; swevent_hlist_put_cpu(cpu); } - - put_online_cpus(); + mutex_unlock(&pmus_lock); return err; } @@ -7845,12 +8053,8 @@ static int perf_event_set_bpf_prog(struct perf_event *event, u32 prog_fd) bool is_kprobe, is_tracepoint; struct bpf_prog *prog; - if (event->attr.type == PERF_TYPE_HARDWARE || - event->attr.type == PERF_TYPE_SOFTWARE) - return perf_event_set_bpf_handler(event, prog_fd); - if (event->attr.type != PERF_TYPE_TRACEPOINT) - return -EINVAL; + return perf_event_set_bpf_handler(event, prog_fd); if (event->tp_event->prog) return -EEXIST; @@ -8039,6 +8243,9 @@ static void perf_event_addr_filters_apply(struct perf_event *event) if (task == TASK_TOMBSTONE) return; + if (!ifh->nr_file_filters) + return; + mm = get_task_mm(event->ctx->task); if (!mm) goto restart; @@ -8209,6 +8416,7 @@ perf_event_parse_addr_filter(struct perf_event *event, char *fstr, * attribute. */ if (state == IF_STATE_END) { + ret = -EINVAL; if (kernel && event->attr.exclude_kernel) goto fail; @@ -8216,6 +8424,18 @@ perf_event_parse_addr_filter(struct perf_event *event, char *fstr, if (!filename) goto fail; + /* + * For now, we only support file-based filters + * in per-task events; doing so for CPU-wide + * events requires additional context switching + * trickery, since same object code will be + * mapped at different virtual addresses in + * different processes. + */ + ret = -EOPNOTSUPP; + if (!event->ctx->task) + goto fail_free_name; + /* look up the path and grab its inode */ ret = kern_path(filename, LOOKUP_FOLLOW, &path); if (ret) @@ -8231,6 +8451,8 @@ perf_event_parse_addr_filter(struct perf_event *event, char *fstr, !S_ISREG(filter->inode->i_mode)) /* free_filters_list() will iput() */ goto fail; + + event->addr_filters.nr_file_filters++; } /* ready to consume more filters */ @@ -8270,24 +8492,13 @@ perf_event_set_addr_filter(struct perf_event *event, char *filter_str) if (WARN_ON_ONCE(event->parent)) return -EINVAL; - /* - * For now, we only support filtering in per-task events; doing so - * for CPU-wide events requires additional context switching trickery, - * since same object code will be mapped at different virtual - * addresses in different processes. - */ - if (!event->ctx->task) - return -EOPNOTSUPP; - ret = perf_event_parse_addr_filter(event, filter_str, &filters); if (ret) - return ret; + goto fail_clear_files; ret = event->pmu->addr_filters_validate(&filters); - if (ret) { - free_filters_list(&filters); - return ret; - } + if (ret) + goto fail_free_filters; /* remove existing filters, if any */ perf_addr_filters_splice(event, &filters); @@ -8296,6 +8507,14 @@ perf_event_set_addr_filter(struct perf_event *event, char *filter_str) perf_event_for_each_child(event, perf_event_addr_filters_apply); return ret; + +fail_free_filters: + free_filters_list(&filters); + +fail_clear_files: + event->addr_filters.nr_file_filters = 0; + + return ret; } static int perf_event_set_filter(struct perf_event *event, void __user *arg) @@ -8647,37 +8866,10 @@ static struct perf_cpu_context __percpu *find_pmu_context(int ctxn) return NULL; } -static void update_pmu_context(struct pmu *pmu, struct pmu *old_pmu) -{ - int cpu; - - for_each_possible_cpu(cpu) { - struct perf_cpu_context *cpuctx; - - cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); - - if (cpuctx->unique_pmu == old_pmu) - cpuctx->unique_pmu = pmu; - } -} - static void free_pmu_context(struct pmu *pmu) { - struct pmu *i; - mutex_lock(&pmus_lock); - /* - * Like a real lame refcount. - */ - list_for_each_entry(i, &pmus, entry) { - if (i->pmu_cpu_context == pmu->pmu_cpu_context) { - update_pmu_context(i, pmu); - goto out; - } - } - free_percpu(pmu->pmu_cpu_context); -out: mutex_unlock(&pmus_lock); } @@ -8740,7 +8932,7 @@ perf_event_mux_interval_ms_store(struct device *dev, pmu->hrtimer_interval_ms = timer; /* update all cpuctx for this PMU */ - get_online_cpus(); + cpus_read_lock(); for_each_online_cpu(cpu) { struct perf_cpu_context *cpuctx; cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); @@ -8749,7 +8941,7 @@ perf_event_mux_interval_ms_store(struct device *dev, cpu_function_call(cpu, (remote_function_f)perf_mux_hrtimer_restart, cpuctx); } - put_online_cpus(); + cpus_read_unlock(); mutex_unlock(&mux_interval_mutex); return count; @@ -8879,10 +9071,9 @@ skip_type: lockdep_set_class(&cpuctx->ctx.mutex, &cpuctx_mutex); lockdep_set_class(&cpuctx->ctx.lock, &cpuctx_lock); cpuctx->ctx.pmu = pmu; + cpuctx->online = cpumask_test_cpu(cpu, perf_online_mask); __perf_mux_hrtimer_init(cpuctx, cpu); - - cpuctx->unique_pmu = pmu; } got_cpu_context: @@ -8994,12 +9185,20 @@ static int perf_try_init_event(struct pmu *pmu, struct perf_event *event) static struct pmu *perf_init_event(struct perf_event *event) { - struct pmu *pmu = NULL; + struct pmu *pmu; int idx; int ret; idx = srcu_read_lock(&pmus_srcu); + /* Try parent's PMU first: */ + if (event->parent && event->parent->pmu) { + pmu = event->parent->pmu; + ret = perf_try_init_event(pmu, event); + if (!ret) + goto unlock; + } + rcu_read_lock(); pmu = idr_find(&pmu_idr, event->attr.type); rcu_read_unlock(); @@ -9092,6 +9291,8 @@ static void account_event(struct perf_event *event) atomic_inc(&nr_mmap_events); if (event->attr.comm) atomic_inc(&nr_comm_events); + if (event->attr.namespaces) + atomic_inc(&nr_namespaces_events); if (event->attr.task) atomic_inc(&nr_task_events); if (event->attr.freq) @@ -9253,9 +9454,10 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, local64_set(&hwc->period_left, hwc->sample_period); /* - * we currently do not support PERF_FORMAT_GROUP on inherited events + * We currently do not support PERF_SAMPLE_READ on inherited events. + * See perf_output_read(). */ - if (attr->inherit && (attr->read_format & PERF_FORMAT_GROUP)) + if (attr->inherit && (attr->sample_type & PERF_SAMPLE_READ)) goto err_ns; if (!has_branch_stack(event)) @@ -9268,9 +9470,7 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, } pmu = perf_init_event(event); - if (!pmu) - goto err_ns; - else if (IS_ERR(pmu)) { + if (IS_ERR(pmu)) { err = PTR_ERR(pmu); goto err_ns; } @@ -9283,8 +9483,10 @@ perf_event_alloc(struct perf_event_attr *attr, int cpu, event->addr_filters_offs = kcalloc(pmu->nr_addr_filters, sizeof(unsigned long), GFP_KERNEL); - if (!event->addr_filters_offs) + if (!event->addr_filters_offs) { + err = -ENOMEM; goto err_per_task; + } /* force hw sync on the address filters */ event->addr_filters_gen = 1; @@ -9637,6 +9839,11 @@ SYSCALL_DEFINE5(perf_event_open, return -EACCES; } + if (attr.namespaces) { + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + } + if (attr.freq) { if (attr.sample_freq > sysctl_perf_event_sample_rate) return -EINVAL; @@ -9689,12 +9896,10 @@ SYSCALL_DEFINE5(perf_event_open, goto err_task; } - get_online_cpus(); - if (task) { err = mutex_lock_interruptible(&task->signal->cred_guard_mutex); if (err) - goto err_cpus; + goto err_task; /* * Reuse ptrace permission checks for now. @@ -9880,6 +10085,23 @@ SYSCALL_DEFINE5(perf_event_open, goto err_locked; } + if (!task) { + /* + * Check if the @cpu we're creating an event for is online. + * + * We use the perf_cpu_context::ctx::mutex to serialize against + * the hotplug notifiers. See perf_event_{init,exit}_cpu(). + */ + struct perf_cpu_context *cpuctx = + container_of(ctx, struct perf_cpu_context, ctx); + + if (!cpuctx->online) { + err = -ENODEV; + goto err_locked; + } + } + + /* * Must be under the same ctx::mutex as perf_install_in_context(), * because we need to serialize with concurrent event creation. @@ -9905,6 +10127,7 @@ SYSCALL_DEFINE5(perf_event_open, * of swizzling perf_event::ctx. */ perf_remove_from_context(group_leader, 0); + put_ctx(gctx); list_for_each_entry(sibling, &group_leader->sibling_list, group_entry) { @@ -9943,13 +10166,6 @@ SYSCALL_DEFINE5(perf_event_open, perf_event__state_init(group_leader); perf_install_in_context(ctx, group_leader, group_leader->cpu); get_ctx(ctx); - - /* - * Now that all events are installed in @ctx, nothing - * references @gctx anymore, so drop the last reference we have - * on it. - */ - put_ctx(gctx); } /* @@ -9975,8 +10191,6 @@ SYSCALL_DEFINE5(perf_event_open, put_task_struct(task); } - put_online_cpus(); - mutex_lock(¤t->perf_event_mutex); list_add_tail(&event->owner_entry, ¤t->perf_event_list); mutex_unlock(¤t->perf_event_mutex); @@ -10010,8 +10224,6 @@ err_alloc: err_cred: if (task) mutex_unlock(&task->signal->cred_guard_mutex); -err_cpus: - put_online_cpus(); err_task: if (task) put_task_struct(task); @@ -10066,6 +10278,21 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu, goto err_unlock; } + if (!task) { + /* + * Check if the @cpu we're creating an event for is online. + * + * We use the perf_cpu_context::ctx::mutex to serialize against + * the hotplug notifiers. See perf_event_{init,exit}_cpu(). + */ + struct perf_cpu_context *cpuctx = + container_of(ctx, struct perf_cpu_context, ctx); + if (!cpuctx->online) { + err = -ENODEV; + goto err_unlock; + } + } + if (!exclusive_event_installable(event, ctx)) { err = -EBUSY; goto err_unlock; @@ -10260,7 +10487,7 @@ static void perf_event_exit_task_context(struct task_struct *child, int ctxn) * in. */ raw_spin_lock_irq(&child_ctx->lock); - task_ctx_sched_out(__get_cpu_context(child_ctx), child_ctx); + task_ctx_sched_out(__get_cpu_context(child_ctx), child_ctx, EVENT_ALL); /* * Now that the context is inactive, destroy the task <-> ctx relation @@ -10369,21 +10596,22 @@ void perf_event_free_task(struct task_struct *task) continue; mutex_lock(&ctx->mutex); -again: - list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, - group_entry) - perf_free_event(event, ctx); + raw_spin_lock_irq(&ctx->lock); + /* + * Destroy the task <-> ctx relation and mark the context dead. + * + * This is important because even though the task hasn't been + * exposed yet the context has been (through child_list). + */ + RCU_INIT_POINTER(task->perf_event_ctxp[ctxn], NULL); + WRITE_ONCE(ctx->task, TASK_TOMBSTONE); + put_task_struct(task); /* cannot be last */ + raw_spin_unlock_irq(&ctx->lock); - list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, - group_entry) + list_for_each_entry_safe(event, tmp, &ctx->event_list, event_entry) perf_free_event(event, ctx); - if (!list_empty(&ctx->pinned_groups) || - !list_empty(&ctx->flexible_groups)) - goto again; - mutex_unlock(&ctx->mutex); - put_ctx(ctx); } } @@ -10421,7 +10649,12 @@ const struct perf_event_attr *perf_event_attrs(struct perf_event *event) } /* - * inherit a event from parent task to child task: + * Inherit a event from parent task to child task. + * + * Returns: + * - valid pointer on success + * - NULL for orphaned events + * - IS_ERR() on error */ static struct perf_event * inherit_event(struct perf_event *parent_event, @@ -10515,6 +10748,16 @@ inherit_event(struct perf_event *parent_event, return child_event; } +/* + * Inherits an event group. + * + * This will quietly suppress orphaned events; !inherit_event() is not an error. + * This matches with perf_event_release_kernel() removing all child events. + * + * Returns: + * - 0 on success + * - <0 on error + */ static int inherit_group(struct perf_event *parent_event, struct task_struct *parent, struct perf_event_context *parent_ctx, @@ -10529,6 +10772,11 @@ static int inherit_group(struct perf_event *parent_event, child, NULL, child_ctx); if (IS_ERR(leader)) return PTR_ERR(leader); + /* + * @leader can be NULL here because of is_orphaned_event(). In this + * case inherit_event() will create individual events, similar to what + * perf_group_detach() would do anyway. + */ list_for_each_entry(sub, &parent_event->sibling_list, group_entry) { child_ctr = inherit_event(sub, parent, parent_ctx, child, leader, child_ctx); @@ -10538,6 +10786,17 @@ static int inherit_group(struct perf_event *parent_event, return 0; } +/* + * Creates the child task context and tries to inherit the event-group. + * + * Clears @inherited_all on !attr.inherited or error. Note that we'll leave + * inherited_all set when we 'fail' to inherit an orphaned event; this is + * consistent with perf_event_release_kernel() removing all child events. + * + * Returns: + * - 0 on success + * - <0 on error + */ static int inherit_task_group(struct perf_event *event, struct task_struct *parent, struct perf_event_context *parent_ctx, @@ -10560,7 +10819,6 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent, * First allocate and initialize a context for the * child. */ - child_ctx = alloc_perf_context(parent_ctx->pmu, child); if (!child_ctx) return -ENOMEM; @@ -10622,7 +10880,7 @@ static int perf_event_init_context(struct task_struct *child, int ctxn) ret = inherit_task_group(event, parent, parent_ctx, child, ctxn, &inherited_all); if (ret) - break; + goto out_unlock; } /* @@ -10638,7 +10896,7 @@ static int perf_event_init_context(struct task_struct *child, int ctxn) ret = inherit_task_group(event, parent, parent_ctx, child, ctxn, &inherited_all); if (ret) - break; + goto out_unlock; } raw_spin_lock_irqsave(&parent_ctx->lock, flags); @@ -10666,6 +10924,7 @@ static int perf_event_init_context(struct task_struct *child, int ctxn) } raw_spin_unlock_irqrestore(&parent_ctx->lock, flags); +out_unlock: mutex_unlock(&parent_ctx->mutex); perf_unpin_context(parent_ctx); @@ -10701,6 +10960,8 @@ static void __init perf_event_init_all_cpus(void) struct swevent_htable *swhash; int cpu; + zalloc_cpumask_var(&perf_online_mask, GFP_KERNEL); + for_each_possible_cpu(cpu) { swhash = &per_cpu(swevent_htable, cpu); mutex_init(&swhash->hlist_mutex); @@ -10709,11 +10970,14 @@ static void __init perf_event_init_all_cpus(void) INIT_LIST_HEAD(&per_cpu(pmu_sb_events.list, cpu)); raw_spin_lock_init(&per_cpu(pmu_sb_events.lock, cpu)); +#ifdef CONFIG_CGROUP_PERF + INIT_LIST_HEAD(&per_cpu(cgrp_cpuctx_list, cpu)); +#endif INIT_LIST_HEAD(&per_cpu(sched_cb_list, cpu)); } } -int perf_event_init_cpu(unsigned int cpu) +void perf_swevent_init_cpu(unsigned int cpu) { struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); @@ -10726,7 +10990,6 @@ int perf_event_init_cpu(unsigned int cpu) rcu_assign_pointer(swhash->swevent_hlist, hlist); } mutex_unlock(&swhash->hlist_mutex); - return 0; } #if defined CONFIG_HOTPLUG_CPU || defined CONFIG_KEXEC_CORE @@ -10744,19 +11007,22 @@ static void __perf_event_exit_context(void *__info) static void perf_event_exit_cpu_context(int cpu) { + struct perf_cpu_context *cpuctx; struct perf_event_context *ctx; struct pmu *pmu; - int idx; - idx = srcu_read_lock(&pmus_srcu); - list_for_each_entry_rcu(pmu, &pmus, entry) { - ctx = &per_cpu_ptr(pmu->pmu_cpu_context, cpu)->ctx; + mutex_lock(&pmus_lock); + list_for_each_entry(pmu, &pmus, entry) { + cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); + ctx = &cpuctx->ctx; mutex_lock(&ctx->mutex); smp_call_function_single(cpu, __perf_event_exit_context, ctx, 1); + cpuctx->online = 0; mutex_unlock(&ctx->mutex); } - srcu_read_unlock(&pmus_srcu, idx); + cpumask_clear_cpu(cpu, perf_online_mask); + mutex_unlock(&pmus_lock); } #else @@ -10764,6 +11030,29 @@ static void perf_event_exit_cpu_context(int cpu) { } #endif +int perf_event_init_cpu(unsigned int cpu) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + struct pmu *pmu; + + perf_swevent_init_cpu(cpu); + + mutex_lock(&pmus_lock); + cpumask_set_cpu(cpu, perf_online_mask); + list_for_each_entry(pmu, &pmus, entry) { + cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu); + ctx = &cpuctx->ctx; + + mutex_lock(&ctx->mutex); + cpuctx->online = 1; + mutex_unlock(&ctx->mutex); + } + mutex_unlock(&pmus_lock); + + return 0; +} + int perf_event_exit_cpu(unsigned int cpu) { perf_event_exit_cpu_context(cpu); @@ -10906,5 +11195,11 @@ struct cgroup_subsys perf_event_cgrp_subsys = { .css_alloc = perf_cgroup_css_alloc, .css_free = perf_cgroup_css_free, .attach = perf_cgroup_attach, + /* + * Implicitly enable on dfl hierarchy so that perf events can + * always be filtered by cgroup2 path as long as perf_event + * controller is not mounted on a legacy hierarchy. + */ + .implicit_on_dfl = true, }; #endif /* CONFIG_CGROUP_PERF */ diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 257fa460b846..ee97196bb151 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -297,6 +297,19 @@ ring_buffer_init(struct ring_buffer *rb, long watermark, int flags) rb->paused = 1; } +void perf_aux_output_flag(struct perf_output_handle *handle, u64 flags) +{ + /* + * OVERWRITE is determined by perf_aux_output_end() and can't + * be passed in directly. + */ + if (WARN_ON_ONCE(flags & PERF_AUX_FLAG_OVERWRITE)) + return; + + handle->aux_flags |= flags; +} +EXPORT_SYMBOL_GPL(perf_aux_output_flag); + /* * This is called before hardware starts writing to the AUX area to * obtain an output handle and make sure there's room in the buffer. @@ -360,6 +373,7 @@ void *perf_aux_output_begin(struct perf_output_handle *handle, handle->event = event; handle->head = aux_head; handle->size = 0; + handle->aux_flags = 0; /* * In overwrite mode, AUX data stores do not depend on aux_tail, @@ -408,34 +422,32 @@ err: * of the AUX buffer management code is that after pmu::stop(), the AUX * transaction must be stopped and therefore drop the AUX reference count. */ -void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size, - bool truncated) +void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size) { + bool wakeup = !!(handle->aux_flags & PERF_AUX_FLAG_TRUNCATED); struct ring_buffer *rb = handle->rb; - bool wakeup = truncated; unsigned long aux_head; - u64 flags = 0; - - if (truncated) - flags |= PERF_AUX_FLAG_TRUNCATED; /* in overwrite mode, driver provides aux_head via handle */ if (rb->aux_overwrite) { - flags |= PERF_AUX_FLAG_OVERWRITE; + handle->aux_flags |= PERF_AUX_FLAG_OVERWRITE; aux_head = handle->head; local_set(&rb->aux_head, aux_head); } else { + handle->aux_flags &= ~PERF_AUX_FLAG_OVERWRITE; + aux_head = local_read(&rb->aux_head); local_add(size, &rb->aux_head); } - if (size || flags) { + if (size || handle->aux_flags) { /* * Only send RECORD_AUX if we have something useful to communicate */ - perf_event_aux_event(handle->event, aux_head, size, flags); + perf_event_aux_event(handle->event, aux_head, size, + handle->aux_flags); } aux_head = rb->user_page->aux_head = local_read(&rb->aux_head); @@ -446,7 +458,7 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size, } if (wakeup) { - if (truncated) + if (handle->aux_flags & PERF_AUX_FLAG_TRUNCATED) handle->event->pending_disable = 1; perf_output_wakeup(handle); } @@ -568,7 +580,7 @@ int rb_alloc_aux(struct ring_buffer *rb, struct perf_event *event, int ret = -ENOMEM, max_order = 0; if (!has_aux(event)) - return -ENOTSUPP; + return -EOPNOTSUPP; if (event->pmu->capabilities & PERF_PMU_CAP_AUX_NO_SG) { /* diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index d416f3baf392..0e137f98a50c 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -27,6 +27,8 @@ #include <linux/pagemap.h> /* read_mapping_page */ #include <linux/slab.h> #include <linux/sched.h> +#include <linux/sched/mm.h> +#include <linux/sched/coredump.h> #include <linux/export.h> #include <linux/rmap.h> /* anon_vma_prepare */ #include <linux/mmu_notifier.h> /* set_pte_at_notify */ @@ -153,14 +155,19 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, struct page *old_page, struct page *new_page) { struct mm_struct *mm = vma->vm_mm; - spinlock_t *ptl; - pte_t *ptep; + struct page_vma_mapped_walk pvmw = { + .page = old_page, + .vma = vma, + .address = addr, + }; int err; /* For mmu_notifiers */ const unsigned long mmun_start = addr; const unsigned long mmun_end = addr + PAGE_SIZE; struct mem_cgroup *memcg; + VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page); + err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg, false); if (err) @@ -171,11 +178,11 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); err = -EAGAIN; - ptep = page_check_address(old_page, mm, addr, &ptl, 0); - if (!ptep) { + if (!page_vma_mapped_walk(&pvmw)) { mem_cgroup_cancel_charge(new_page, memcg, false); goto unlock; } + VM_BUG_ON_PAGE(addr != pvmw.address, old_page); get_page(new_page); page_add_new_anon_rmap(new_page, vma, addr, false); @@ -187,14 +194,15 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, inc_mm_counter(mm, MM_ANONPAGES); } - flush_cache_page(vma, addr, pte_pfn(*ptep)); - ptep_clear_flush_notify(vma, addr, ptep); - set_pte_at_notify(mm, addr, ptep, mk_pte(new_page, vma->vm_page_prot)); + flush_cache_page(vma, addr, pte_pfn(*pvmw.pte)); + ptep_clear_flush_notify(vma, addr, pvmw.pte); + set_pte_at_notify(mm, addr, pvmw.pte, + mk_pte(new_page, vma->vm_page_prot)); page_remove_rmap(old_page, false); if (!page_mapped(old_page)) try_to_free_swap(old_page); - pte_unmap_unlock(ptep, ptl); + page_vma_mapped_walk_done(&pvmw); if (vma->vm_flags & VM_LOCKED) munlock_vma_page(old_page); @@ -300,8 +308,8 @@ int uprobe_write_opcode(struct mm_struct *mm, unsigned long vaddr, retry: /* Read the page with vaddr into memory */ - ret = get_user_pages_remote(NULL, mm, vaddr, 1, FOLL_FORCE, &old_page, - &vma, NULL); + ret = get_user_pages_remote(NULL, mm, vaddr, 1, + FOLL_FORCE | FOLL_SPLIT, &old_page, &vma, NULL); if (ret <= 0) return ret; @@ -741,7 +749,7 @@ build_map_info(struct address_space *mapping, loff_t offset, bool is_register) continue; } - if (!atomic_inc_not_zero(&vma->vm_mm->mm_users)) + if (!mmget_not_zero(vma->vm_mm)) continue; info = prev; diff --git a/kernel/exit.c b/kernel/exit.c index 8f14b866f9f6..c5548faa9f37 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -6,6 +6,12 @@ #include <linux/mm.h> #include <linux/slab.h> +#include <linux/sched/autogroup.h> +#include <linux/sched/mm.h> +#include <linux/sched/stat.h> +#include <linux/sched/task.h> +#include <linux/sched/task_stack.h> +#include <linux/sched/cputime.h> #include <linux/interrupt.h> #include <linux/module.h> #include <linux/capability.h> @@ -14,7 +20,6 @@ #include <linux/tty.h> #include <linux/iocontext.h> #include <linux/key.h> -#include <linux/security.h> #include <linux/cpu.h> #include <linux/acct.h> #include <linux/tsacct_kern.h> @@ -55,6 +60,8 @@ #include <linux/shm.h> #include <linux/kcov.h> #include <linux/random.h> +#include <linux/rcuwait.h> +#include <linux/compat.h> #include <linux/uaccess.h> #include <asm/unistd.h> @@ -86,7 +93,7 @@ static void __exit_signal(struct task_struct *tsk) bool group_dead = thread_group_leader(tsk); struct sighand_struct *sighand; struct tty_struct *uninitialized_var(tty); - cputime_t utime, stime; + u64 utime, stime; sighand = rcu_dereference_check(tsk->sighand, lockdep_tasklist_lock_is_held()); @@ -282,17 +289,33 @@ retry: return task; } -struct task_struct *try_get_task_struct(struct task_struct **ptask) +void rcuwait_wake_up(struct rcuwait *w) { struct task_struct *task; rcu_read_lock(); - task = task_rcu_dereference(ptask); + + /* + * Order condition vs @task, such that everything prior to the load + * of @task is visible. This is the condition as to why the user called + * rcuwait_trywake() in the first place. Pairs with set_current_state() + * barrier (A) in rcuwait_wait_event(). + * + * WAIT WAKE + * [S] tsk = current [S] cond = true + * MB (A) MB (B) + * [L] cond [L] tsk + */ + smp_rmb(); /* (B) */ + + /* + * Avoid using task_rcu_dereference() magic as long as we are careful, + * see comment in rcuwait_wait_event() regarding ->exit_state. + */ + task = rcu_dereference(w->task); if (task) - get_task_struct(task); + wake_up_process(task); rcu_read_unlock(); - - return task; } /* @@ -468,12 +491,12 @@ assign_new_owner: * Turn us into a lazy TLB process if we * aren't already.. */ -static void exit_mm(struct task_struct *tsk) +static void exit_mm(void) { - struct mm_struct *mm = tsk->mm; + struct mm_struct *mm = current->mm; struct core_state *core_state; - mm_release(tsk, mm); + mm_release(current, mm); if (!mm) return; sync_mm_rss(mm); @@ -491,7 +514,7 @@ static void exit_mm(struct task_struct *tsk) up_read(&mm->mmap_sem); - self.task = tsk; + self.task = current; self.next = xchg(&core_state->dumper.next, &self); /* * Implies mb(), the result of xchg() must be visible @@ -501,22 +524,22 @@ static void exit_mm(struct task_struct *tsk) complete(&core_state->startup); for (;;) { - set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE); if (!self.task) /* see coredump_finish() */ break; freezable_schedule(); } - __set_task_state(tsk, TASK_RUNNING); + __set_current_state(TASK_RUNNING); down_read(&mm->mmap_sem); } - atomic_inc(&mm->mm_count); - BUG_ON(mm != tsk->active_mm); + mmgrab(mm); + BUG_ON(mm != current->active_mm); /* more a memory barrier than a real lock */ - task_lock(tsk); - tsk->mm = NULL; + task_lock(current); + current->mm = NULL; up_read(&mm->mmap_sem); enter_lazy_tlb(mm, current); - task_unlock(tsk); + task_unlock(current); mm_update_next_owner(mm); mmput(mm); if (test_thread_flag(TIF_MEMDIE)) @@ -578,15 +601,18 @@ static struct task_struct *find_new_reaper(struct task_struct *father, return thread; if (father->signal->has_child_subreaper) { + unsigned int ns_level = task_pid(father)->level; /* * Find the first ->is_child_subreaper ancestor in our pid_ns. - * We start from father to ensure we can not look into another - * namespace, this is safe because all its threads are dead. + * We can't check reaper != child_reaper to ensure we do not + * cross the namespaces, the exiting parent could be injected + * by setns() + fork(). + * We check pid->level, this is slightly more efficient than + * task_active_pid_ns(reaper) != task_active_pid_ns(father). */ - for (reaper = father; - !same_thread_group(reaper, child_reaper); + for (reaper = father->real_parent; + task_pid(reaper)->level == ns_level; reaper = reaper->real_parent) { - /* call_usermodehelper() descendants need this check */ if (reaper == &init_task) break; if (!reaper->signal->is_child_subreaper) @@ -823,7 +849,7 @@ void __noreturn do_exit(long code) tsk->exit_code = code; taskstats_exit(tsk, group_dead); - exit_mm(tsk); + exit_mm(); if (group_dead) acct_process(); @@ -956,16 +982,23 @@ SYSCALL_DEFINE1(exit_group, int, error_code) return 0; } +struct waitid_info { + pid_t pid; + uid_t uid; + int status; + int cause; +}; + struct wait_opts { enum pid_type wo_type; int wo_flags; struct pid *wo_pid; - struct siginfo __user *wo_info; - int __user *wo_stat; - struct rusage __user *wo_rusage; + struct waitid_info *wo_info; + int wo_stat; + struct rusage *wo_rusage; - wait_queue_t child_wait; + wait_queue_entry_t child_wait; int notask_error; }; @@ -1010,34 +1043,6 @@ eligible_child(struct wait_opts *wo, bool ptrace, struct task_struct *p) return 1; } -static int wait_noreap_copyout(struct wait_opts *wo, struct task_struct *p, - pid_t pid, uid_t uid, int why, int status) -{ - struct siginfo __user *infop; - int retval = wo->wo_rusage - ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; - - put_task_struct(p); - infop = wo->wo_info; - if (infop) { - if (!retval) - retval = put_user(SIGCHLD, &infop->si_signo); - if (!retval) - retval = put_user(0, &infop->si_errno); - if (!retval) - retval = put_user((short)why, &infop->si_code); - if (!retval) - retval = put_user(pid, &infop->si_pid); - if (!retval) - retval = put_user(uid, &infop->si_uid); - if (!retval) - retval = put_user(status, &infop->si_status); - } - if (!retval) - retval = pid; - return retval; -} - /* * Handle sys_wait4 work for one task in state EXIT_ZOMBIE. We hold * read_lock(&tasklist_lock) on entry. If we return zero, we still hold @@ -1046,30 +1051,23 @@ static int wait_noreap_copyout(struct wait_opts *wo, struct task_struct *p, */ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) { - int state, retval, status; + int state, status; pid_t pid = task_pid_vnr(p); uid_t uid = from_kuid_munged(current_user_ns(), task_uid(p)); - struct siginfo __user *infop; + struct waitid_info *infop; if (!likely(wo->wo_flags & WEXITED)) return 0; if (unlikely(wo->wo_flags & WNOWAIT)) { - int exit_code = p->exit_code; - int why; - + status = p->exit_code; get_task_struct(p); read_unlock(&tasklist_lock); sched_annotate_sleep(); - - if ((exit_code & 0x7f) == 0) { - why = CLD_EXITED; - status = exit_code >> 8; - } else { - why = (exit_code & 0x80) ? CLD_DUMPED : CLD_KILLED; - status = exit_code & 0x7f; - } - return wait_noreap_copyout(wo, p, pid, uid, why, status); + if (wo->wo_rusage) + getrusage(p, RUSAGE_BOTH, wo->wo_rusage); + put_task_struct(p); + goto out_info; } /* * Move the task's state to DEAD/TRACE, only one thread can do this. @@ -1091,7 +1089,7 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) struct signal_struct *sig = p->signal; struct signal_struct *psig = current->signal; unsigned long maxrss; - cputime_t tgutime, tgstime; + u64 tgutime, tgstime; /* * The resource counters for the group leader are in its @@ -1142,38 +1140,11 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) spin_unlock_irq(¤t->sighand->siglock); } - retval = wo->wo_rusage - ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; + if (wo->wo_rusage) + getrusage(p, RUSAGE_BOTH, wo->wo_rusage); status = (p->signal->flags & SIGNAL_GROUP_EXIT) ? p->signal->group_exit_code : p->exit_code; - if (!retval && wo->wo_stat) - retval = put_user(status, wo->wo_stat); - - infop = wo->wo_info; - if (!retval && infop) - retval = put_user(SIGCHLD, &infop->si_signo); - if (!retval && infop) - retval = put_user(0, &infop->si_errno); - if (!retval && infop) { - int why; - - if ((status & 0x7f) == 0) { - why = CLD_EXITED; - status >>= 8; - } else { - why = (status & 0x80) ? CLD_DUMPED : CLD_KILLED; - status &= 0x7f; - } - retval = put_user((short)why, &infop->si_code); - if (!retval) - retval = put_user(status, &infop->si_status); - } - if (!retval && infop) - retval = put_user(pid, &infop->si_pid); - if (!retval && infop) - retval = put_user(uid, &infop->si_uid); - if (!retval) - retval = pid; + wo->wo_stat = status; if (state == EXIT_TRACE) { write_lock_irq(&tasklist_lock); @@ -1190,7 +1161,21 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) if (state == EXIT_DEAD) release_task(p); - return retval; +out_info: + infop = wo->wo_info; + if (infop) { + if ((status & 0x7f) == 0) { + infop->cause = CLD_EXITED; + infop->status = status >> 8; + } else { + infop->cause = (status & 0x80) ? CLD_DUMPED : CLD_KILLED; + infop->status = status & 0x7f; + } + infop->pid = pid; + infop->uid = uid; + } + + return pid; } static int *task_stopped_code(struct task_struct *p, bool ptrace) @@ -1226,8 +1211,8 @@ static int *task_stopped_code(struct task_struct *p, bool ptrace) static int wait_task_stopped(struct wait_opts *wo, int ptrace, struct task_struct *p) { - struct siginfo __user *infop; - int retval, exit_code, *p_code, why; + struct waitid_info *infop; + int exit_code, *p_code, why; uid_t uid = 0; /* unneeded, required by compiler */ pid_t pid; @@ -1272,34 +1257,21 @@ unlock_sig: why = ptrace ? CLD_TRAPPED : CLD_STOPPED; read_unlock(&tasklist_lock); sched_annotate_sleep(); + if (wo->wo_rusage) + getrusage(p, RUSAGE_BOTH, wo->wo_rusage); + put_task_struct(p); - if (unlikely(wo->wo_flags & WNOWAIT)) - return wait_noreap_copyout(wo, p, pid, uid, why, exit_code); - - retval = wo->wo_rusage - ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; - if (!retval && wo->wo_stat) - retval = put_user((exit_code << 8) | 0x7f, wo->wo_stat); + if (likely(!(wo->wo_flags & WNOWAIT))) + wo->wo_stat = (exit_code << 8) | 0x7f; infop = wo->wo_info; - if (!retval && infop) - retval = put_user(SIGCHLD, &infop->si_signo); - if (!retval && infop) - retval = put_user(0, &infop->si_errno); - if (!retval && infop) - retval = put_user((short)why, &infop->si_code); - if (!retval && infop) - retval = put_user(exit_code, &infop->si_status); - if (!retval && infop) - retval = put_user(pid, &infop->si_pid); - if (!retval && infop) - retval = put_user(uid, &infop->si_uid); - if (!retval) - retval = pid; - put_task_struct(p); - - BUG_ON(!retval); - return retval; + if (infop) { + infop->cause = why; + infop->status = exit_code; + infop->pid = pid; + infop->uid = uid; + } + return pid; } /* @@ -1310,7 +1282,7 @@ unlock_sig: */ static int wait_task_continued(struct wait_opts *wo, struct task_struct *p) { - int retval; + struct waitid_info *infop; pid_t pid; uid_t uid; @@ -1335,22 +1307,20 @@ static int wait_task_continued(struct wait_opts *wo, struct task_struct *p) get_task_struct(p); read_unlock(&tasklist_lock); sched_annotate_sleep(); + if (wo->wo_rusage) + getrusage(p, RUSAGE_BOTH, wo->wo_rusage); + put_task_struct(p); - if (!wo->wo_info) { - retval = wo->wo_rusage - ? getrusage(p, RUSAGE_BOTH, wo->wo_rusage) : 0; - put_task_struct(p); - if (!retval && wo->wo_stat) - retval = put_user(0xffff, wo->wo_stat); - if (!retval) - retval = pid; + infop = wo->wo_info; + if (!infop) { + wo->wo_stat = 0xffff; } else { - retval = wait_noreap_copyout(wo, p, pid, uid, - CLD_CONTINUED, SIGCONT); - BUG_ON(retval == 0); + infop->cause = CLD_CONTINUED; + infop->pid = pid; + infop->uid = uid; + infop->status = SIGCONT; } - - return retval; + return pid; } /* @@ -1360,7 +1330,7 @@ static int wait_task_continued(struct wait_opts *wo, struct task_struct *p) * Returns nonzero for a final return, when we have unlocked tasklist_lock. * Returns zero if the search for a child should continue; * then ->notask_error is 0 if @p is an eligible child, - * or another error from security_task_wait(), or still -ECHILD. + * or still -ECHILD. */ static int wait_consider_task(struct wait_opts *wo, int ptrace, struct task_struct *p) @@ -1380,20 +1350,6 @@ static int wait_consider_task(struct wait_opts *wo, int ptrace, if (!ret) return ret; - ret = security_task_wait(p); - if (unlikely(ret < 0)) { - /* - * If we have not yet seen any eligible child, - * then let this error code replace -ECHILD. - * A permission error will give the user a clue - * to look for security policy problems, rather - * than for mysterious wait bugs. - */ - if (wo->notask_error) - wo->notask_error = ret; - return 0; - } - if (unlikely(exit_state == EXIT_TRACE)) { /* * ptrace == 0 means we are the natural parent. In this case @@ -1486,7 +1442,7 @@ static int wait_consider_task(struct wait_opts *wo, int ptrace, * Returns nonzero for a final return, when we have unlocked tasklist_lock. * Returns zero if the search for a child should continue; then * ->notask_error is 0 if there were any eligible children, - * or another error from security_task_wait(), or still -ECHILD. + * or still -ECHILD. */ static int do_wait_thread(struct wait_opts *wo, struct task_struct *tsk) { @@ -1516,7 +1472,7 @@ static int ptrace_do_wait(struct wait_opts *wo, struct task_struct *tsk) return 0; } -static int child_wait_callback(wait_queue_t *wait, unsigned mode, +static int child_wait_callback(wait_queue_entry_t *wait, unsigned mode, int sync, void *key) { struct wait_opts *wo = container_of(wait, struct wait_opts, @@ -1592,8 +1548,8 @@ end: return retval; } -SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, - infop, int, options, struct rusage __user *, ru) +static long kernel_waitid(int which, pid_t upid, struct waitid_info *infop, + int options, struct rusage *ru) { struct wait_opts wo; struct pid *pid = NULL; @@ -1631,38 +1587,48 @@ SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, wo.wo_pid = pid; wo.wo_flags = options; wo.wo_info = infop; - wo.wo_stat = NULL; wo.wo_rusage = ru; ret = do_wait(&wo); - if (ret > 0) { - ret = 0; - } else if (infop) { - /* - * For a WNOHANG return, clear out all the fields - * we would set so the user can easily tell the - * difference. - */ - if (!ret) - ret = put_user(0, &infop->si_signo); - if (!ret) - ret = put_user(0, &infop->si_errno); - if (!ret) - ret = put_user(0, &infop->si_code); - if (!ret) - ret = put_user(0, &infop->si_pid); - if (!ret) - ret = put_user(0, &infop->si_uid); - if (!ret) - ret = put_user(0, &infop->si_status); - } - put_pid(pid); return ret; } -SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, - int, options, struct rusage __user *, ru) +SYSCALL_DEFINE5(waitid, int, which, pid_t, upid, struct siginfo __user *, + infop, int, options, struct rusage __user *, ru) +{ + struct rusage r; + struct waitid_info info = {.status = 0}; + long err = kernel_waitid(which, upid, &info, options, ru ? &r : NULL); + int signo = 0; + if (err > 0) { + signo = SIGCHLD; + err = 0; + } + + if (!err) { + if (ru && copy_to_user(ru, &r, sizeof(struct rusage))) + return -EFAULT; + } + if (!infop) + return err; + + user_access_begin(); + unsafe_put_user(signo, &infop->si_signo, Efault); + unsafe_put_user(0, &infop->si_errno, Efault); + unsafe_put_user((short)info.cause, &infop->si_code, Efault); + unsafe_put_user(info.pid, &infop->si_pid, Efault); + unsafe_put_user(info.uid, &infop->si_uid, Efault); + unsafe_put_user(info.status, &infop->si_status, Efault); + user_access_end(); + return err; +Efault: + user_access_end(); + return -EFAULT; +} + +long kernel_wait4(pid_t upid, int __user *stat_addr, int options, + struct rusage *ru) { struct wait_opts wo; struct pid *pid = NULL; @@ -1673,6 +1639,10 @@ SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, __WNOTHREAD|__WCLONE|__WALL)) return -EINVAL; + /* -INT_MIN is not defined */ + if (upid == INT_MIN) + return -ESRCH; + if (upid == -1) type = PIDTYPE_MAX; else if (upid < 0) { @@ -1690,14 +1660,29 @@ SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, wo.wo_pid = pid; wo.wo_flags = options | WEXITED; wo.wo_info = NULL; - wo.wo_stat = stat_addr; + wo.wo_stat = 0; wo.wo_rusage = ru; ret = do_wait(&wo); put_pid(pid); + if (ret > 0 && stat_addr && put_user(wo.wo_stat, stat_addr)) + ret = -EFAULT; return ret; } +SYSCALL_DEFINE4(wait4, pid_t, upid, int __user *, stat_addr, + int, options, struct rusage __user *, ru) +{ + struct rusage r; + long err = kernel_wait4(upid, stat_addr, options, ru ? &r : NULL); + + if (err > 0) { + if (ru && copy_to_user(ru, &r, sizeof(struct rusage))) + return -EFAULT; + } + return err; +} + #ifdef __ARCH_WANT_SYS_WAITPID /* @@ -1710,3 +1695,61 @@ SYSCALL_DEFINE3(waitpid, pid_t, pid, int __user *, stat_addr, int, options) } #endif + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(wait4, + compat_pid_t, pid, + compat_uint_t __user *, stat_addr, + int, options, + struct compat_rusage __user *, ru) +{ + struct rusage r; + long err = kernel_wait4(pid, stat_addr, options, ru ? &r : NULL); + if (err > 0) { + if (ru && put_compat_rusage(&r, ru)) + return -EFAULT; + } + return err; +} + +COMPAT_SYSCALL_DEFINE5(waitid, + int, which, compat_pid_t, pid, + struct compat_siginfo __user *, infop, int, options, + struct compat_rusage __user *, uru) +{ + struct rusage ru; + struct waitid_info info = {.status = 0}; + long err = kernel_waitid(which, pid, &info, options, uru ? &ru : NULL); + int signo = 0; + if (err > 0) { + signo = SIGCHLD; + err = 0; + } + + if (!err && uru) { + /* kernel_waitid() overwrites everything in ru */ + if (COMPAT_USE_64BIT_TIME) + err = copy_to_user(uru, &ru, sizeof(ru)); + else + err = put_compat_rusage(&ru, uru); + if (err) + return -EFAULT; + } + + if (!infop) + return err; + + user_access_begin(); + unsafe_put_user(signo, &infop->si_signo, Efault); + unsafe_put_user(0, &infop->si_errno, Efault); + unsafe_put_user((short)info.cause, &infop->si_code, Efault); + unsafe_put_user(info.pid, &infop->si_pid, Efault); + unsafe_put_user(info.uid, &infop->si_uid, Efault); + unsafe_put_user(info.status, &infop->si_status, Efault); + user_access_end(); + return err; +Efault: + user_access_end(); + return -EFAULT; +} +#endif diff --git a/kernel/extable.c b/kernel/extable.c index e3beec4a2339..38c2412401a1 100644 --- a/kernel/extable.c +++ b/kernel/extable.c @@ -17,9 +17,12 @@ */ #include <linux/ftrace.h> #include <linux/memory.h> +#include <linux/extable.h> #include <linux/module.h> #include <linux/mutex.h> #include <linux/init.h> +#include <linux/kprobes.h> +#include <linux/filter.h> #include <asm/sections.h> #include <linux/uaccess.h> @@ -52,7 +55,8 @@ const struct exception_table_entry *search_exception_tables(unsigned long addr) { const struct exception_table_entry *e; - e = search_extable(__start___ex_table, __stop___ex_table-1, addr); + e = search_extable(__start___ex_table, + __stop___ex_table - __start___ex_table, addr); if (!e) e = search_module_extables(addr); return e; @@ -66,13 +70,13 @@ static inline int init_kernel_text(unsigned long addr) return 0; } -int core_kernel_text(unsigned long addr) +int notrace core_kernel_text(unsigned long addr) { if (addr >= (unsigned long)_stext && addr < (unsigned long)_etext) return 1; - if (system_state == SYSTEM_BOOTING && + if (system_state < SYSTEM_RUNNING && init_kernel_text(addr)) return 1; return 0; @@ -104,6 +108,10 @@ int __kernel_text_address(unsigned long addr) return 1; if (is_ftrace_trampoline(addr)) return 1; + if (is_kprobe_optinsn_slot(addr) || is_kprobe_insn_slot(addr)) + return 1; + if (is_bpf_text_address(addr)) + return 1; /* * There might be init symbols in saved stacktraces. * Give those symbols a chance to be printed in @@ -123,7 +131,13 @@ int kernel_text_address(unsigned long addr) return 1; if (is_module_text_address(addr)) return 1; - return is_ftrace_trampoline(addr); + if (is_ftrace_trampoline(addr)) + return 1; + if (is_kprobe_optinsn_slot(addr) || is_kprobe_insn_slot(addr)) + return 1; + if (is_bpf_text_address(addr)) + return 1; + return 0; } /* diff --git a/kernel/fork.c b/kernel/fork.c index 11c5c8ab827c..17921b0390b4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -12,6 +12,16 @@ */ #include <linux/slab.h> +#include <linux/sched/autogroup.h> +#include <linux/sched/mm.h> +#include <linux/sched/coredump.h> +#include <linux/sched/user.h> +#include <linux/sched/numa_balancing.h> +#include <linux/sched/stat.h> +#include <linux/sched/task.h> +#include <linux/sched/task_stack.h> +#include <linux/sched/cputime.h> +#include <linux/rtmutex.h> #include <linux/init.h> #include <linux/unistd.h> #include <linux/module.h> @@ -55,6 +65,7 @@ #include <linux/rmap.h> #include <linux/ksm.h> #include <linux/acct.h> +#include <linux/userfaultfd_k.h> #include <linux/tsacct_kern.h> #include <linux/cn_proc.h> #include <linux/freezer.h> @@ -76,6 +87,7 @@ #include <linux/compiler.h> #include <linux/sysctl.h> #include <linux/kcov.h> +#include <linux/livepatch.h> #include <asm/pgtable.h> #include <asm/pgalloc.h> @@ -167,6 +179,24 @@ void __weak arch_release_thread_stack(unsigned long *stack) */ #define NR_CACHED_STACKS 2 static DEFINE_PER_CPU(struct vm_struct *, cached_stacks[NR_CACHED_STACKS]); + +static int free_vm_stack_cache(unsigned int cpu) +{ + struct vm_struct **cached_vm_stacks = per_cpu_ptr(cached_stacks, cpu); + int i; + + for (i = 0; i < NR_CACHED_STACKS; i++) { + struct vm_struct *vm_stack = cached_vm_stacks[i]; + + if (!vm_stack) + continue; + + vfree(vm_stack->addr); + cached_vm_stacks[i] = NULL; + } + + return 0; +} #endif static unsigned long *alloc_thread_stack_node(struct task_struct *tsk, int node) @@ -175,23 +205,21 @@ static unsigned long *alloc_thread_stack_node(struct task_struct *tsk, int node) void *stack; int i; - local_irq_disable(); for (i = 0; i < NR_CACHED_STACKS; i++) { - struct vm_struct *s = this_cpu_read(cached_stacks[i]); + struct vm_struct *s; + + s = this_cpu_xchg(cached_stacks[i], NULL); if (!s) continue; - this_cpu_write(cached_stacks[i], NULL); tsk->stack_vm_area = s; - local_irq_enable(); return s->addr; } - local_irq_enable(); stack = __vmalloc_node_range(THREAD_SIZE, THREAD_SIZE, VMALLOC_START, VMALLOC_END, - THREADINFO_GFP | __GFP_HIGHMEM, + THREADINFO_GFP, PAGE_KERNEL, 0, node, __builtin_return_address(0)); @@ -215,19 +243,15 @@ static inline void free_thread_stack(struct task_struct *tsk) { #ifdef CONFIG_VMAP_STACK if (task_stack_vm_area(tsk)) { - unsigned long flags; int i; - local_irq_save(flags); for (i = 0; i < NR_CACHED_STACKS; i++) { - if (this_cpu_read(cached_stacks[i])) + if (this_cpu_cmpxchg(cached_stacks[i], + NULL, tsk->stack_vm_area) != NULL) continue; - this_cpu_write(cached_stacks[i], tsk->stack_vm_area); - local_irq_restore(flags); return; } - local_irq_restore(flags); vfree_atomic(tsk->stack); return; @@ -296,8 +320,8 @@ static void account_kernel_stack(struct task_struct *tsk, int account) } /* All stack pages belong to the same memcg. */ - memcg_kmem_update_page_stat(vm->pages[0], MEMCG_KERNEL_STACK_KB, - account * (THREAD_SIZE / 1024)); + mod_memcg_page_state(vm->pages[0], MEMCG_KERNEL_STACK_KB, + account * (THREAD_SIZE / 1024)); } else { /* * All stack pages are in the same zone and belong to the @@ -308,8 +332,8 @@ static void account_kernel_stack(struct task_struct *tsk, int account) mod_zone_page_state(page_zone(first_page), NR_KERNEL_STACK_KB, THREAD_SIZE / 1024 * account); - memcg_kmem_update_page_stat(first_page, MEMCG_KERNEL_STACK_KB, - account * (THREAD_SIZE / 1024)); + mod_memcg_page_state(first_page, MEMCG_KERNEL_STACK_KB, + account * (THREAD_SIZE / 1024)); } } @@ -432,11 +456,13 @@ void __init fork_init(void) int i; #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR #ifndef ARCH_MIN_TASKALIGN -#define ARCH_MIN_TASKALIGN L1_CACHE_BYTES +#define ARCH_MIN_TASKALIGN 0 #endif + int align = max_t(int, L1_CACHE_BYTES, ARCH_MIN_TASKALIGN); + /* create a slab on which task_structs can be allocated */ task_struct_cachep = kmem_cache_create("task_struct", - arch_task_struct_size, ARCH_MIN_TASKALIGN, + arch_task_struct_size, align, SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, NULL); #endif @@ -453,6 +479,11 @@ void __init fork_init(void) for (i = 0; i < UCOUNT_COUNTS; i++) { init_user_ns.ucount_max[i] = max_threads/2; } + +#ifdef CONFIG_VMAP_STACK + cpuhp_setup_state(CPUHP_BP_PREPARE_DYN, "fork:vm_stack_cache", + NULL, free_vm_stack_cache); +#endif } int __weak arch_dup_task_struct(struct task_struct *dst, @@ -523,7 +554,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node) set_task_stack_end_magic(tsk); #ifdef CONFIG_CC_STACKPROTECTOR - tsk->stack_canary = get_random_int(); + tsk->stack_canary = get_random_canary(); #endif /* @@ -542,6 +573,10 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node) kcov_task_init(tsk); +#ifdef CONFIG_FAULT_INJECTION + tsk->fail_nth = 0; +#endif + return tsk; free_stack: @@ -559,6 +594,7 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm, struct rb_node **rb_link, *rb_parent; int retval; unsigned long charge; + LIST_HEAD(uf); uprobe_start_dup_mmap(); if (down_write_killable(&oldmm->mmap_sem)) { @@ -615,12 +651,13 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm, if (retval) goto fail_nomem_policy; tmp->vm_mm = mm; + retval = dup_userfaultfd(tmp, &uf); + if (retval) + goto fail_nomem_anon_vma_fork; if (anon_vma_fork(tmp, mpnt)) goto fail_nomem_anon_vma_fork; - tmp->vm_flags &= - ~(VM_LOCKED|VM_LOCKONFAULT|VM_UFFD_MISSING|VM_UFFD_WP); + tmp->vm_flags &= ~(VM_LOCKED | VM_LOCKONFAULT); tmp->vm_next = tmp->vm_prev = NULL; - tmp->vm_userfaultfd_ctx = NULL_VM_UFFD_CTX; file = tmp->vm_file; if (file) { struct inode *inode = file_inode(file); @@ -676,6 +713,7 @@ out: up_write(&mm->mmap_sem); flush_tlb_mm(oldmm); up_write(&oldmm->mmap_sem); + dup_userfaultfd_complete(&uf); fail_uprobe_end: uprobe_end_dup_mmap(); return retval; @@ -994,7 +1032,7 @@ struct mm_struct *get_task_mm(struct task_struct *task) if (task->flags & PF_KTHREAD) mm = NULL; else - atomic_inc(&mm->mm_users); + mmget(mm); } task_unlock(task); return mm; @@ -1182,7 +1220,7 @@ static int copy_mm(unsigned long clone_flags, struct task_struct *tsk) vmacache_flush(tsk); if (clone_flags & CLONE_VM) { - atomic_inc(&oldmm->mm_users); + mmget(oldmm); mm = oldmm; goto good_mm; } @@ -1297,13 +1335,14 @@ void __cleanup_sighand(struct sighand_struct *sighand) if (atomic_dec_and_test(&sighand->count)) { signalfd_cleanup(sighand); /* - * sighand_cachep is SLAB_DESTROY_BY_RCU so we can free it + * sighand_cachep is SLAB_TYPESAFE_BY_RCU so we can free it * without an RCU grace period, see __lock_task_sighand(). */ kmem_cache_free(sighand_cachep, sighand); } } +#ifdef CONFIG_POSIX_TIMERS /* * Initialize POSIX timer handling for a thread group. */ @@ -1313,7 +1352,7 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) cpu_limit = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); if (cpu_limit != RLIM_INFINITY) { - sig->cputime_expires.prof_exp = secs_to_cputime(cpu_limit); + sig->cputime_expires.prof_exp = cpu_limit * NSEC_PER_SEC; sig->cputimer.running = true; } @@ -1322,6 +1361,9 @@ static void posix_cpu_timers_init_group(struct signal_struct *sig) INIT_LIST_HEAD(&sig->cpu_timers[1]); INIT_LIST_HEAD(&sig->cpu_timers[2]); } +#else +static inline void posix_cpu_timers_init_group(struct signal_struct *sig) { } +#endif static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) { @@ -1346,11 +1388,11 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) init_waitqueue_head(&sig->wait_chldexit); sig->curr_target = tsk; init_sigpending(&sig->shared_pending); - INIT_LIST_HEAD(&sig->posix_timers); seqlock_init(&sig->stats_lock); prev_cputime_init(&sig->prev_cputime); #ifdef CONFIG_POSIX_TIMERS + INIT_LIST_HEAD(&sig->posix_timers); hrtimer_init(&sig->real_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); sig->real_timer.function = it_real_fn; #endif @@ -1367,9 +1409,6 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->oom_score_adj = current->signal->oom_score_adj; sig->oom_score_adj_min = current->signal->oom_score_adj_min; - sig->has_child_subreaper = current->signal->has_child_subreaper || - current->signal->is_child_subreaper; - mutex_init(&sig->cred_guard_mutex); return 0; @@ -1421,10 +1460,12 @@ static void rt_mutex_init_task(struct task_struct *p) #ifdef CONFIG_RT_MUTEXES p->pi_waiters = RB_ROOT; p->pi_waiters_leftmost = NULL; + p->pi_top_task = NULL; p->pi_blocked_on = NULL; #endif } +#ifdef CONFIG_POSIX_TIMERS /* * Initialize POSIX timer handling for a single task. */ @@ -1437,6 +1478,9 @@ static void posix_cpu_timers_init(struct task_struct *tsk) INIT_LIST_HEAD(&tsk->cpu_timers[1]); INIT_LIST_HEAD(&tsk->cpu_timers[2]); } +#else +static inline void posix_cpu_timers_init(struct task_struct *tsk) { } +#endif static inline void init_task_pid(struct task_struct *task, enum pid_type type, struct pid *pid) @@ -1444,6 +1488,21 @@ init_task_pid(struct task_struct *task, enum pid_type type, struct pid *pid) task->pids[type].pid = pid; } +static inline void rcu_copy_process(struct task_struct *p) +{ +#ifdef CONFIG_PREEMPT_RCU + p->rcu_read_lock_nesting = 0; + p->rcu_read_unlock_special.s = 0; + p->rcu_blocked_node = NULL; + INIT_LIST_HEAD(&p->rcu_node_entry); +#endif /* #ifdef CONFIG_PREEMPT_RCU */ +#ifdef CONFIG_TASKS_RCU + p->rcu_tasks_holdout = false; + INIT_LIST_HEAD(&p->rcu_tasks_holdout_list); + p->rcu_tasks_idle_cpu = -1; +#endif /* #ifdef CONFIG_TASKS_RCU */ +} + /* * This creates a new process as a copy of the old one, * but does not actually start it yet. @@ -1516,6 +1575,18 @@ static __latent_entropy struct task_struct *copy_process( if (!p) goto fork_out; + /* + * This _must_ happen before we call free_task(), i.e. before we jump + * to any of the bad_fork_* labels. This is to avoid freeing + * p->set_child_tid which is (ab)used as a kthread's data pointer for + * kernel threads (PF_KTHREAD). + */ + p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL; + /* + * Clear TID on mm_release()? + */ + p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? child_tidptr : NULL; + ftrace_graph_init_task(p); rt_mutex_init_task(p); @@ -1564,9 +1635,9 @@ static __latent_entropy struct task_struct *copy_process( prev_cputime_init(&p->prev_cputime); #ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN - seqcount_init(&p->vtime_seqcount); - p->vtime_snap = 0; - p->vtime_snap_whence = VTIME_INACTIVE; + seqcount_init(&p->vtime.seqcount); + p->vtime.starttime = 0; + p->vtime.state = VTIME_INACTIVE; #endif #if defined(SPLIT_RSS_COUNTING) @@ -1643,9 +1714,12 @@ static __latent_entropy struct task_struct *copy_process( goto bad_fork_cleanup_perf; /* copy all the process information */ shm_init_task(p); - retval = copy_semundo(clone_flags, p); + retval = security_task_alloc(p, clone_flags); if (retval) goto bad_fork_cleanup_audit; + retval = copy_semundo(clone_flags, p); + if (retval) + goto bad_fork_cleanup_security; retval = copy_files(clone_flags, p); if (retval) goto bad_fork_cleanup_semundo; @@ -1679,11 +1753,6 @@ static __latent_entropy struct task_struct *copy_process( } } - p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL; - /* - * Clear TID on mm_release()? - */ - p->clear_child_tid = (clone_flags & CLONE_CHILD_CLEARTID) ? child_tidptr : NULL; #ifdef CONFIG_BLOCK p->plug = NULL; #endif @@ -1735,7 +1804,7 @@ static __latent_entropy struct task_struct *copy_process( INIT_LIST_HEAD(&p->thread_group); p->task_works = NULL; - threadgroup_change_begin(current); + cgroup_threadgroup_change_begin(current); /* * Ensure that the cgroup subsystem policies allow the new process to be * forked. It should be noted the the new process's css_set can be changed @@ -1761,6 +1830,8 @@ static __latent_entropy struct task_struct *copy_process( p->parent_exec_id = current->self_exec_id; } + klp_copy_process(p); + spin_lock(¤t->sighand->siglock); /* @@ -1779,11 +1850,13 @@ static __latent_entropy struct task_struct *copy_process( */ recalc_sigpending(); if (signal_pending(current)) { - spin_unlock(¤t->sighand->siglock); - write_unlock_irq(&tasklist_lock); retval = -ERESTARTNOINTR; goto bad_fork_cancel_cgroup; } + if (unlikely(!(ns_of_pid(pid)->nr_hashed & PIDNS_HASH_ADDING))) { + retval = -ENOMEM; + goto bad_fork_cancel_cgroup; + } if (likely(p->pid)) { ptrace_init_task(p, (clone_flags & CLONE_PTRACE) || trace); @@ -1800,6 +1873,13 @@ static __latent_entropy struct task_struct *copy_process( p->signal->leader_pid = pid; p->signal->tty = tty_kref_get(current->signal->tty); + /* + * Inherit has_child_subreaper flag under the same + * tasklist_lock with adding child to the process tree + * for propagate_has_child_subreaper optimization. + */ + p->signal->has_child_subreaper = p->real_parent->signal->has_child_subreaper || + p->real_parent->signal->is_child_subreaper; list_add_tail(&p->sibling, &p->real_parent->children); list_add_tail_rcu(&p->tasks, &init_task.tasks); attach_pid(p, PIDTYPE_PGID); @@ -1825,7 +1905,7 @@ static __latent_entropy struct task_struct *copy_process( proc_fork_connector(p); cgroup_post_fork(p); - threadgroup_change_end(current); + cgroup_threadgroup_change_end(current); perf_event_fork(p); trace_task_newtask(p, clone_flags); @@ -1834,9 +1914,11 @@ static __latent_entropy struct task_struct *copy_process( return p; bad_fork_cancel_cgroup: + spin_unlock(¤t->sighand->siglock); + write_unlock_irq(&tasklist_lock); cgroup_cancel_fork(p); bad_fork_free_pid: - threadgroup_change_end(current); + cgroup_threadgroup_change_end(current); if (pid != &init_struct_pid) free_pid(pid); bad_fork_cleanup_thread: @@ -1860,6 +1942,8 @@ bad_fork_cleanup_files: exit_files(p); /* blocking */ bad_fork_cleanup_semundo: exit_sem(p); +bad_fork_cleanup_security: + security_task_free(p); bad_fork_cleanup_audit: audit_free(p); bad_fork_cleanup_perf: @@ -2053,6 +2137,38 @@ SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp, } #endif +void walk_process_tree(struct task_struct *top, proc_visitor visitor, void *data) +{ + struct task_struct *leader, *parent, *child; + int res; + + read_lock(&tasklist_lock); + leader = top = top->group_leader; +down: + for_each_thread(leader, parent) { + list_for_each_entry(child, &parent->children, sibling) { + res = visitor(child, data); + if (res) { + if (res < 0) + goto out; + leader = child; + goto down; + } +up: + ; + } + } + + if (leader != top) { + child = leader; + parent = child->real_parent; + leader = parent->group_leader; + goto up; + } +out: + read_unlock(&tasklist_lock); +} + #ifndef ARCH_MIN_MMSTRUCT_ALIGN #define ARCH_MIN_MMSTRUCT_ALIGN 0 #endif @@ -2069,7 +2185,7 @@ void __init proc_caches_init(void) { sighand_cachep = kmem_cache_create("sighand_cache", sizeof(struct sighand_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_DESTROY_BY_RCU| + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_TYPESAFE_BY_RCU| SLAB_NOTRACK|SLAB_ACCOUNT, sighand_ctor); signal_cachep = kmem_cache_create("signal_cache", sizeof(struct signal_struct), 0, @@ -2277,6 +2393,8 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags) } } + perf_event_namespaces(current); + bad_unshare_cleanup_cred: if (new_cred) put_cred(new_cred); diff --git a/kernel/futex.c b/kernel/futex.c index 0842c8ca534b..16dbe4c93895 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -61,6 +61,8 @@ #include <linux/nsproxy.h> #include <linux/ptrace.h> #include <linux/sched/rt.h> +#include <linux/sched/wake_q.h> +#include <linux/sched/mm.h> #include <linux/hugetlb.h> #include <linux/freezer.h> #include <linux/bootmem.h> @@ -210,7 +212,7 @@ struct futex_pi_state { atomic_t refcount; union futex_key key; -}; +} __randomize_layout; /** * struct futex_q - The hashed futex queue entry, one per waiting task @@ -223,7 +225,7 @@ struct futex_pi_state { * @requeue_pi_key: the requeue_pi target futex key * @bitset: bitset for the optional bitmasked wakeup * - * We use this hashed waitqueue, instead of a normal wait_queue_t, so + * We use this hashed waitqueue, instead of a normal wait_queue_entry_t, so * we can wake only the relevant ones (hashed queues may be shared). * * A futex_q has a woken state, just like tasks have TASK_RUNNING. @@ -244,7 +246,7 @@ struct futex_q { struct rt_mutex_waiter *rt_waiter; union futex_key *requeue_pi_key; u32 bitset; -}; +} __randomize_layout; static const struct futex_q futex_q_init = { /* list gets initialized in queue_me()*/ @@ -338,7 +340,7 @@ static inline bool should_fail_futex(bool fshared) static inline void futex_get_mm(union futex_key *key) { - atomic_inc(&key->private.mm->mm_count); + mmgrab(key->private.mm); /* * Ensure futex_get_mm() implies a full barrier such that * get_futex_key() implies a full barrier. This is relied upon @@ -486,7 +488,7 @@ static void drop_futex_key_refs(union futex_key *key) * * Return: a negative error code or 0 * - * The key words are stored in *key on success. + * The key words are stored in @key on success. * * For shared mappings, it's (page->index, file_inode(vma->vm_file), * offset_within_page). For private mappings, it's (uaddr, current->mm). @@ -800,7 +802,7 @@ static int refill_pi_state_cache(void) return 0; } -static struct futex_pi_state * alloc_pi_state(void) +static struct futex_pi_state *alloc_pi_state(void) { struct futex_pi_state *pi_state = current->pi_state_cache; @@ -810,6 +812,11 @@ static struct futex_pi_state * alloc_pi_state(void) return pi_state; } +static void get_pi_state(struct futex_pi_state *pi_state) +{ + WARN_ON_ONCE(!atomic_inc_not_zero(&pi_state->refcount)); +} + /* * Drops a reference to the pi_state object and frees or caches it * when the last reference is gone. @@ -854,7 +861,7 @@ static void put_pi_state(struct futex_pi_state *pi_state) * Look up the task based on what TID userspace gave us. * We dont trust it. */ -static struct task_struct * futex_find_get_task(pid_t pid) +static struct task_struct *futex_find_get_task(pid_t pid) { struct task_struct *p; @@ -914,10 +921,12 @@ void exit_pi_state_list(struct task_struct *curr) pi_state->owner = NULL; raw_spin_unlock_irq(&curr->pi_lock); - rt_mutex_unlock(&pi_state->pi_mutex); - + get_pi_state(pi_state); spin_unlock(&hb->lock); + rt_mutex_futex_unlock(&pi_state->pi_mutex); + put_pi_state(pi_state); + raw_spin_lock_irq(&curr->pi_lock); } raw_spin_unlock_irq(&curr->pi_lock); @@ -971,6 +980,39 @@ void exit_pi_state_list(struct task_struct *curr) * * [10] There is no transient state which leaves owner and user space * TID out of sync. + * + * + * Serialization and lifetime rules: + * + * hb->lock: + * + * hb -> futex_q, relation + * futex_q -> pi_state, relation + * + * (cannot be raw because hb can contain arbitrary amount + * of futex_q's) + * + * pi_mutex->wait_lock: + * + * {uval, pi_state} + * + * (and pi_mutex 'obviously') + * + * p->pi_lock: + * + * p->pi_state_list -> pi_state->list, relation + * + * pi_state->refcount: + * + * pi_state lifetime + * + * + * Lock order: + * + * hb->lock + * pi_mutex->wait_lock + * p->pi_lock + * */ /* @@ -978,10 +1020,13 @@ void exit_pi_state_list(struct task_struct *curr) * the pi_state against the user space value. If correct, attach to * it. */ -static int attach_to_pi_state(u32 uval, struct futex_pi_state *pi_state, +static int attach_to_pi_state(u32 __user *uaddr, u32 uval, + struct futex_pi_state *pi_state, struct futex_pi_state **ps) { pid_t pid = uval & FUTEX_TID_MASK; + u32 uval2; + int ret; /* * Userspace might have messed up non-PI and PI futexes [3] @@ -989,9 +1034,39 @@ static int attach_to_pi_state(u32 uval, struct futex_pi_state *pi_state, if (unlikely(!pi_state)) return -EINVAL; + /* + * We get here with hb->lock held, and having found a + * futex_top_waiter(). This means that futex_lock_pi() of said futex_q + * has dropped the hb->lock in between queue_me() and unqueue_me_pi(), + * which in turn means that futex_lock_pi() still has a reference on + * our pi_state. + * + * The waiter holding a reference on @pi_state also protects against + * the unlocked put_pi_state() in futex_unlock_pi(), futex_lock_pi() + * and futex_wait_requeue_pi() as it cannot go to 0 and consequently + * free pi_state before we can take a reference ourselves. + */ WARN_ON(!atomic_read(&pi_state->refcount)); /* + * Now that we have a pi_state, we can acquire wait_lock + * and do the state validation. + */ + raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock); + + /* + * Since {uval, pi_state} is serialized by wait_lock, and our current + * uval was read without holding it, it can have changed. Verify it + * still is what we expect it to be, otherwise retry the entire + * operation. + */ + if (get_futex_value_locked(&uval2, uaddr)) + goto out_efault; + + if (uval != uval2) + goto out_eagain; + + /* * Handle the owner died case: */ if (uval & FUTEX_OWNER_DIED) { @@ -1006,11 +1081,11 @@ static int attach_to_pi_state(u32 uval, struct futex_pi_state *pi_state, * is not 0. Inconsistent state. [5] */ if (pid) - return -EINVAL; + goto out_einval; /* * Take a ref on the state and return success. [4] */ - goto out_state; + goto out_attach; } /* @@ -1022,14 +1097,14 @@ static int attach_to_pi_state(u32 uval, struct futex_pi_state *pi_state, * Take a ref on the state and return success. [6] */ if (!pid) - goto out_state; + goto out_attach; } else { /* * If the owner died bit is not set, then the pi_state * must have an owner. [7] */ if (!pi_state->owner) - return -EINVAL; + goto out_einval; } /* @@ -1038,11 +1113,29 @@ static int attach_to_pi_state(u32 uval, struct futex_pi_state *pi_state, * user space TID. [9/10] */ if (pid != task_pid_vnr(pi_state->owner)) - return -EINVAL; -out_state: - atomic_inc(&pi_state->refcount); + goto out_einval; + +out_attach: + get_pi_state(pi_state); + raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); *ps = pi_state; return 0; + +out_einval: + ret = -EINVAL; + goto out_error; + +out_eagain: + ret = -EAGAIN; + goto out_error; + +out_efault: + ret = -EFAULT; + goto out_error; + +out_error: + raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); + return ret; } /* @@ -1093,6 +1186,9 @@ static int attach_to_pi_owner(u32 uval, union futex_key *key, /* * No existing pi state. First waiter. [2] + * + * This creates pi_state, we have hb->lock held, this means nothing can + * observe this state, wait_lock is irrelevant. */ pi_state = alloc_pi_state(); @@ -1117,17 +1213,18 @@ static int attach_to_pi_owner(u32 uval, union futex_key *key, return 0; } -static int lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, +static int lookup_pi_state(u32 __user *uaddr, u32 uval, + struct futex_hash_bucket *hb, union futex_key *key, struct futex_pi_state **ps) { - struct futex_q *match = futex_top_waiter(hb, key); + struct futex_q *top_waiter = futex_top_waiter(hb, key); /* * If there is a waiter on that futex, validate it and * attach to the pi_state when the validation succeeds. */ - if (match) - return attach_to_pi_state(uval, match->pi_state, ps); + if (top_waiter) + return attach_to_pi_state(uaddr, uval, top_waiter->pi_state, ps); /* * We are the first waiter - try to look up the owner based on @@ -1146,7 +1243,7 @@ static int lock_pi_update_atomic(u32 __user *uaddr, u32 uval, u32 newval) if (unlikely(cmpxchg_futex_value_locked(&curval, uaddr, uval, newval))) return -EFAULT; - /*If user space value changed, let the caller retry */ + /* If user space value changed, let the caller retry */ return curval != uval ? -EAGAIN : 0; } @@ -1162,9 +1259,9 @@ static int lock_pi_update_atomic(u32 __user *uaddr, u32 uval, u32 newval) * @set_waiters: force setting the FUTEX_WAITERS bit (1) or not (0) * * Return: - * 0 - ready to wait; - * 1 - acquired the lock; - * <0 - error + * - 0 - ready to wait; + * - 1 - acquired the lock; + * - <0 - error * * The hb->lock and futex_key refs shall be held by the caller. */ @@ -1174,7 +1271,7 @@ static int futex_lock_pi_atomic(u32 __user *uaddr, struct futex_hash_bucket *hb, struct task_struct *task, int set_waiters) { u32 uval, newval, vpid = task_pid_vnr(task); - struct futex_q *match; + struct futex_q *top_waiter; int ret; /* @@ -1200,9 +1297,9 @@ static int futex_lock_pi_atomic(u32 __user *uaddr, struct futex_hash_bucket *hb, * Lookup existing state first. If it exists, try to attach to * its pi_state. */ - match = futex_top_waiter(hb, key); - if (match) - return attach_to_pi_state(uval, match->pi_state, ps); + top_waiter = futex_top_waiter(hb, key); + if (top_waiter) + return attach_to_pi_state(uaddr, uval, top_waiter->pi_state, ps); /* * No waiter and user TID is 0. We are here because the @@ -1283,50 +1380,44 @@ static void mark_wake_futex(struct wake_q_head *wake_q, struct futex_q *q) wake_q_add(wake_q, p); __unqueue_futex(q); /* - * The waiting task can free the futex_q as soon as - * q->lock_ptr = NULL is written, without taking any locks. A - * memory barrier is required here to prevent the following - * store to lock_ptr from getting ahead of the plist_del. + * The waiting task can free the futex_q as soon as q->lock_ptr = NULL + * is written, without taking any locks. This is possible in the event + * of a spurious wakeup, for example. A memory barrier is required here + * to prevent the following store to lock_ptr from getting ahead of the + * plist_del in __unqueue_futex(). */ - smp_wmb(); - q->lock_ptr = NULL; + smp_store_release(&q->lock_ptr, NULL); } -static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this, - struct futex_hash_bucket *hb) +/* + * Caller must hold a reference on @pi_state. + */ +static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_state) { - struct task_struct *new_owner; - struct futex_pi_state *pi_state = this->pi_state; u32 uninitialized_var(curval), newval; + struct task_struct *new_owner; + bool postunlock = false; DEFINE_WAKE_Q(wake_q); - bool deboost; int ret = 0; - if (!pi_state) - return -EINVAL; - - /* - * If current does not own the pi_state then the futex is - * inconsistent and user space fiddled with the futex value. - */ - if (pi_state->owner != current) - return -EINVAL; - - raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock); new_owner = rt_mutex_next_owner(&pi_state->pi_mutex); + if (WARN_ON_ONCE(!new_owner)) { + /* + * As per the comment in futex_unlock_pi() this should not happen. + * + * When this happens, give up our locks and try again, giving + * the futex_lock_pi() instance time to complete, either by + * waiting on the rtmutex or removing itself from the futex + * queue. + */ + ret = -EAGAIN; + goto out_unlock; + } /* - * It is possible that the next waiter (the one that brought - * this owner to the kernel) timed out and is no longer - * waiting on the lock. - */ - if (!new_owner) - new_owner = this->task; - - /* - * We pass it to the next owner. The WAITERS bit is always - * kept enabled while there is PI state around. We cleanup the - * owner died bit, because we are the owner. + * We pass it to the next owner. The WAITERS bit is always kept + * enabled while there is PI state around. We cleanup the owner + * died bit, because we are the owner. */ newval = FUTEX_WAITERS | task_pid_vnr(new_owner); @@ -1335,6 +1426,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this, if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) { ret = -EFAULT; + } else if (curval != uval) { /* * If a unconditional UNLOCK_PI operation (user space did not @@ -1347,10 +1439,14 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this, else ret = -EINVAL; } - if (ret) { - raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); - return ret; - } + + if (ret) + goto out_unlock; + + /* + * This is a point of no return; once we modify the uval there is no + * going back and subsequent operations must not fail. + */ raw_spin_lock(&pi_state->owner->pi_lock); WARN_ON(list_empty(&pi_state->list)); @@ -1363,22 +1459,15 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this, pi_state->owner = new_owner; raw_spin_unlock(&new_owner->pi_lock); - raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); + postunlock = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q); - deboost = rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q); +out_unlock: + raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); - /* - * First unlock HB so the waiter does not spin on it once he got woken - * up. Second wake up the waiter before the priority is adjusted. If we - * deboost first (and lose our higher priority), then the task might get - * scheduled away before the wake up can take place. - */ - spin_unlock(&hb->lock); - wake_up_q(&wake_q); - if (deboost) - rt_mutex_adjust_prio(current); + if (postunlock) + rt_mutex_postunlock(&wake_q); - return 0; + return ret; } /* @@ -1628,9 +1717,9 @@ void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key, * hb1 and hb2 must be held by the caller. * * Return: - * 0 - failed to acquire the lock atomically; - * >0 - acquired the lock, return value is vpid of the top_waiter - * <0 - error + * - 0 - failed to acquire the lock atomically; + * - >0 - acquired the lock, return value is vpid of the top_waiter + * - <0 - error */ static int futex_proxy_trylock_atomic(u32 __user *pifutex, struct futex_hash_bucket *hb1, @@ -1696,8 +1785,8 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, * uaddr2 atomically on behalf of the top waiter. * * Return: - * >=0 - on success, the number of tasks requeued or woken; - * <0 - on error + * - >=0 - on success, the number of tasks requeued or woken; + * - <0 - on error */ static int futex_requeue(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2, int nr_wake, int nr_requeue, @@ -1824,7 +1913,7 @@ retry_private: * If that call succeeds then we have pi_state and an * initial refcount on it. */ - ret = lookup_pi_state(ret, hb2, &key2, &pi_state); + ret = lookup_pi_state(uaddr2, ret, hb2, &key2, &pi_state); } switch (ret) { @@ -1907,7 +1996,7 @@ retry_private: * refcount on the pi_state and store the pointer in * the futex_q object of the waiter. */ - atomic_inc(&pi_state->refcount); + get_pi_state(pi_state); this->pi_state = pi_state; ret = rt_mutex_start_proxy_lock(&pi_state->pi_mutex, this->rt_waiter, @@ -2007,20 +2096,7 @@ queue_unlock(struct futex_hash_bucket *hb) hb_waiters_dec(hb); } -/** - * queue_me() - Enqueue the futex_q on the futex_hash_bucket - * @q: The futex_q to enqueue - * @hb: The destination hash bucket - * - * The hb->lock must be held by the caller, and is released here. A call to - * queue_me() is typically paired with exactly one call to unqueue_me(). The - * exceptions involve the PI related operations, which may use unqueue_me_pi() - * or nothing if the unqueue is done as part of the wake process and the unqueue - * state is implicit in the state of woken task (see futex_wait_requeue_pi() for - * an example). - */ -static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) - __releases(&hb->lock) +static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *hb) { int prio; @@ -2037,6 +2113,24 @@ static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) plist_node_init(&q->list, prio); plist_add(&q->list, &hb->chain); q->task = current; +} + +/** + * queue_me() - Enqueue the futex_q on the futex_hash_bucket + * @q: The futex_q to enqueue + * @hb: The destination hash bucket + * + * The hb->lock must be held by the caller, and is released here. A call to + * queue_me() is typically paired with exactly one call to unqueue_me(). The + * exceptions involve the PI related operations, which may use unqueue_me_pi() + * or nothing if the unqueue is done as part of the wake process and the unqueue + * state is implicit in the state of woken task (see futex_wait_requeue_pi() for + * an example). + */ +static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) + __releases(&hb->lock) +{ + __queue_me(q, hb); spin_unlock(&hb->lock); } @@ -2048,8 +2142,8 @@ static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) * be paired with exactly one earlier call to queue_me(). * * Return: - * 1 - if the futex_q was still queued (and we removed unqueued it); - * 0 - if the futex_q was already removed by the waking thread + * - 1 - if the futex_q was still queued (and we removed unqueued it); + * - 0 - if the futex_q was already removed by the waking thread */ static int unqueue_me(struct futex_q *q) { @@ -2123,10 +2217,13 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, { u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS; struct futex_pi_state *pi_state = q->pi_state; - struct task_struct *oldowner = pi_state->owner; u32 uval, uninitialized_var(curval), newval; + struct task_struct *oldowner; int ret; + raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock); + + oldowner = pi_state->owner; /* Owner died? */ if (!pi_state->owner) newtid |= FUTEX_OWNER_DIED; @@ -2134,7 +2231,8 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, /* * We are here either because we stole the rtmutex from the * previous highest priority waiter or we are the highest priority - * waiter but failed to get the rtmutex the first time. + * waiter but have failed to get the rtmutex the first time. + * * We have to replace the newowner TID in the user space variable. * This must be atomic as we have to preserve the owner died bit here. * @@ -2142,17 +2240,16 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, * because we can fault here. Imagine swapped out pages or a fork * that marked all the anonymous memory readonly for cow. * - * Modifying pi_state _before_ the user space value would - * leave the pi_state in an inconsistent state when we fault - * here, because we need to drop the hash bucket lock to - * handle the fault. This might be observed in the PID check - * in lookup_pi_state. + * Modifying pi_state _before_ the user space value would leave the + * pi_state in an inconsistent state when we fault here, because we + * need to drop the locks to handle the fault. This might be observed + * in the PID check in lookup_pi_state. */ retry: if (get_futex_value_locked(&uval, uaddr)) goto handle_fault; - while (1) { + for (;;) { newval = (uval & FUTEX_OWNER_DIED) | newtid; if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) @@ -2167,47 +2264,60 @@ retry: * itself. */ if (pi_state->owner != NULL) { - raw_spin_lock_irq(&pi_state->owner->pi_lock); + raw_spin_lock(&pi_state->owner->pi_lock); WARN_ON(list_empty(&pi_state->list)); list_del_init(&pi_state->list); - raw_spin_unlock_irq(&pi_state->owner->pi_lock); + raw_spin_unlock(&pi_state->owner->pi_lock); } pi_state->owner = newowner; - raw_spin_lock_irq(&newowner->pi_lock); + raw_spin_lock(&newowner->pi_lock); WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &newowner->pi_state_list); - raw_spin_unlock_irq(&newowner->pi_lock); + raw_spin_unlock(&newowner->pi_lock); + raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); + return 0; /* - * To handle the page fault we need to drop the hash bucket - * lock here. That gives the other task (either the highest priority - * waiter itself or the task which stole the rtmutex) the - * chance to try the fixup of the pi_state. So once we are - * back from handling the fault we need to check the pi_state - * after reacquiring the hash bucket lock and before trying to - * do another fixup. When the fixup has been done already we - * simply return. + * To handle the page fault we need to drop the locks here. That gives + * the other task (either the highest priority waiter itself or the + * task which stole the rtmutex) the chance to try the fixup of the + * pi_state. So once we are back from handling the fault we need to + * check the pi_state after reacquiring the locks and before trying to + * do another fixup. When the fixup has been done already we simply + * return. + * + * Note: we hold both hb->lock and pi_mutex->wait_lock. We can safely + * drop hb->lock since the caller owns the hb -> futex_q relation. + * Dropping the pi_mutex->wait_lock requires the state revalidate. */ handle_fault: + raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); spin_unlock(q->lock_ptr); ret = fault_in_user_writeable(uaddr); spin_lock(q->lock_ptr); + raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock); /* * Check if someone else fixed it for us: */ - if (pi_state->owner != oldowner) - return 0; + if (pi_state->owner != oldowner) { + ret = 0; + goto out_unlock; + } if (ret) - return ret; + goto out_unlock; goto retry; + +out_unlock: + raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); + return ret; } static long futex_wait_restart(struct restart_block *restart); @@ -2223,19 +2333,22 @@ static long futex_wait_restart(struct restart_block *restart); * acquire the lock. Must be called with the hb lock held. * * Return: - * 1 - success, lock taken; - * 0 - success, lock not taken; - * <0 - on error (-EFAULT) + * - 1 - success, lock taken; + * - 0 - success, lock not taken; + * - <0 - on error (-EFAULT) */ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) { - struct task_struct *owner; int ret = 0; if (locked) { /* * Got the lock. We might not be the anticipated owner if we * did a lock-steal - fix up the PI-state in that case: + * + * We can safely read pi_state->owner without holding wait_lock + * because we now own the rt_mutex, only the owner will attempt + * to change it. */ if (q->pi_state->owner != current) ret = fixup_pi_state_owner(uaddr, q, current); @@ -2243,43 +2356,15 @@ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) } /* - * Catch the rare case, where the lock was released when we were on the - * way back before we locked the hash bucket. - */ - if (q->pi_state->owner == current) { - /* - * Try to get the rt_mutex now. This might fail as some other - * task acquired the rt_mutex after we removed ourself from the - * rt_mutex waiters list. - */ - if (rt_mutex_trylock(&q->pi_state->pi_mutex)) { - locked = 1; - goto out; - } - - /* - * pi_state is incorrect, some other task did a lock steal and - * we returned due to timeout or signal without taking the - * rt_mutex. Too late. - */ - raw_spin_lock_irq(&q->pi_state->pi_mutex.wait_lock); - owner = rt_mutex_owner(&q->pi_state->pi_mutex); - if (!owner) - owner = rt_mutex_next_owner(&q->pi_state->pi_mutex); - raw_spin_unlock_irq(&q->pi_state->pi_mutex.wait_lock); - ret = fixup_pi_state_owner(uaddr, q, owner); - goto out; - } - - /* * Paranoia check. If we did not take the lock, then we should not be * the owner of the rt_mutex. */ - if (rt_mutex_owner(&q->pi_state->pi_mutex) == current) + if (rt_mutex_owner(&q->pi_state->pi_mutex) == current) { printk(KERN_ERR "fixup_owner: ret = %d pi-mutex: %p " "pi-state %p\n", ret, q->pi_state->pi_mutex.owner, q->pi_state->owner); + } out: return ret ? ret : locked; @@ -2337,8 +2422,8 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * with no q.key reference on failure. * * Return: - * 0 - uaddr contains val and hb has been locked; - * <1 - -EFAULT or -EWOULDBLOCK (uaddr does not contain val) and hb is unlocked + * - 0 - uaddr contains val and hb has been locked; + * - <1 - -EFAULT or -EWOULDBLOCK (uaddr does not contain val) and hb is unlocked */ static int futex_wait_setup(u32 __user *uaddr, u32 val, unsigned int flags, struct futex_q *q, struct futex_hash_bucket **hb) @@ -2503,6 +2588,8 @@ static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, ktime_t *time, int trylock) { struct hrtimer_sleeper timeout, *to = NULL; + struct futex_pi_state *pi_state = NULL; + struct rt_mutex_waiter rt_waiter; struct futex_hash_bucket *hb; struct futex_q q = futex_q_init; int res, ret; @@ -2555,25 +2642,68 @@ retry_private: } } + WARN_ON(!q.pi_state); + /* * Only actually queue now that the atomic ops are done: */ - queue_me(&q, hb); + __queue_me(&q, hb); - WARN_ON(!q.pi_state); - /* - * Block on the PI mutex: - */ - if (!trylock) { - ret = rt_mutex_timed_futex_lock(&q.pi_state->pi_mutex, to); - } else { - ret = rt_mutex_trylock(&q.pi_state->pi_mutex); + if (trylock) { + ret = rt_mutex_futex_trylock(&q.pi_state->pi_mutex); /* Fixup the trylock return value: */ ret = ret ? 0 : -EWOULDBLOCK; + goto no_block; + } + + rt_mutex_init_waiter(&rt_waiter); + + /* + * On PREEMPT_RT_FULL, when hb->lock becomes an rt_mutex, we must not + * hold it while doing rt_mutex_start_proxy(), because then it will + * include hb->lock in the blocking chain, even through we'll not in + * fact hold it while blocking. This will lead it to report -EDEADLK + * and BUG when futex_unlock_pi() interleaves with this. + * + * Therefore acquire wait_lock while holding hb->lock, but drop the + * latter before calling rt_mutex_start_proxy_lock(). This still fully + * serializes against futex_unlock_pi() as that does the exact same + * lock handoff sequence. + */ + raw_spin_lock_irq(&q.pi_state->pi_mutex.wait_lock); + spin_unlock(q.lock_ptr); + ret = __rt_mutex_start_proxy_lock(&q.pi_state->pi_mutex, &rt_waiter, current); + raw_spin_unlock_irq(&q.pi_state->pi_mutex.wait_lock); + + if (ret) { + if (ret == 1) + ret = 0; + + spin_lock(q.lock_ptr); + goto no_block; } + + if (unlikely(to)) + hrtimer_start_expires(&to->timer, HRTIMER_MODE_ABS); + + ret = rt_mutex_wait_proxy_lock(&q.pi_state->pi_mutex, to, &rt_waiter); + spin_lock(q.lock_ptr); /* + * If we failed to acquire the lock (signal/timeout), we must + * first acquire the hb->lock before removing the lock from the + * rt_mutex waitqueue, such that we can keep the hb and rt_mutex + * wait lists consistent. + * + * In particular; it is important that futex_unlock_pi() can not + * observe this inconsistency. + */ + if (ret && !rt_mutex_cleanup_proxy_lock(&q.pi_state->pi_mutex, &rt_waiter)) + ret = 0; + +no_block: + /* * Fixup the pi_state owner and possibly acquire the lock if we * haven't already. */ @@ -2589,12 +2719,19 @@ retry_private: * If fixup_owner() faulted and was unable to handle the fault, unlock * it and return the fault to userspace. */ - if (ret && (rt_mutex_owner(&q.pi_state->pi_mutex) == current)) - rt_mutex_unlock(&q.pi_state->pi_mutex); + if (ret && (rt_mutex_owner(&q.pi_state->pi_mutex) == current)) { + pi_state = q.pi_state; + get_pi_state(pi_state); + } /* Unqueue and drop the lock */ unqueue_me_pi(&q); + if (pi_state) { + rt_mutex_futex_unlock(&pi_state->pi_mutex); + put_pi_state(pi_state); + } + goto out_put_key; out_unlock_put_key: @@ -2603,8 +2740,10 @@ out_unlock_put_key: out_put_key: put_futex_key(&q.key); out: - if (to) + if (to) { + hrtimer_cancel(&to->timer); destroy_hrtimer_on_stack(&to->timer); + } return ret != -EINTR ? ret : -ERESTARTNOINTR; uaddr_faulted: @@ -2631,7 +2770,7 @@ static int futex_unlock_pi(u32 __user *uaddr, unsigned int flags) u32 uninitialized_var(curval), uval, vpid = task_pid_vnr(current); union futex_key key = FUTEX_KEY_INIT; struct futex_hash_bucket *hb; - struct futex_q *match; + struct futex_q *top_waiter; int ret; retry: @@ -2655,12 +2794,37 @@ retry: * all and we at least want to know if user space fiddled * with the futex value instead of blindly unlocking. */ - match = futex_top_waiter(hb, &key); - if (match) { - ret = wake_futex_pi(uaddr, uval, match, hb); + top_waiter = futex_top_waiter(hb, &key); + if (top_waiter) { + struct futex_pi_state *pi_state = top_waiter->pi_state; + + ret = -EINVAL; + if (!pi_state) + goto out_unlock; + + /* + * If current does not own the pi_state then the futex is + * inconsistent and user space fiddled with the futex value. + */ + if (pi_state->owner != current) + goto out_unlock; + + get_pi_state(pi_state); /* - * In case of success wake_futex_pi dropped the hash - * bucket lock. + * By taking wait_lock while still holding hb->lock, we ensure + * there is no point where we hold neither; and therefore + * wake_futex_pi() must observe a state consistent with what we + * observed. + */ + raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock); + spin_unlock(&hb->lock); + + ret = wake_futex_pi(uaddr, uval, pi_state); + + put_pi_state(pi_state); + + /* + * Success, we're done! No tricky corner cases. */ if (!ret) goto out_putkey; @@ -2675,7 +2839,6 @@ retry: * setting the FUTEX_WAITERS bit. Try again. */ if (ret == -EAGAIN) { - spin_unlock(&hb->lock); put_futex_key(&key); goto retry; } @@ -2683,7 +2846,7 @@ retry: * wake_futex_pi has detected invalid state. Tell user * space. */ - goto out_unlock; + goto out_putkey; } /* @@ -2693,8 +2856,10 @@ retry: * preserve the WAITERS bit not the OWNER_DIED one. We are the * owner. */ - if (cmpxchg_futex_value_locked(&curval, uaddr, uval, 0)) + if (cmpxchg_futex_value_locked(&curval, uaddr, uval, 0)) { + spin_unlock(&hb->lock); goto pi_faulted; + } /* * If uval has changed, let user space handle it. @@ -2708,7 +2873,6 @@ out_putkey: return ret; pi_faulted: - spin_unlock(&hb->lock); put_futex_key(&key); ret = fault_in_user_writeable(uaddr); @@ -2731,8 +2895,8 @@ pi_faulted: * called with the hb lock held. * * Return: - * 0 = no early wakeup detected; - * <0 = -ETIMEDOUT or -ERESTARTNOINTR + * - 0 = no early wakeup detected; + * - <0 = -ETIMEDOUT or -ERESTARTNOINTR */ static inline int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, @@ -2804,16 +2968,16 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * If 4 or 7, we cleanup and return with -ETIMEDOUT. * * Return: - * 0 - On success; - * <0 - On error + * - 0 - On success; + * - <0 - On error */ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, u32 val, ktime_t *abs_time, u32 bitset, u32 __user *uaddr2) { struct hrtimer_sleeper timeout, *to = NULL; + struct futex_pi_state *pi_state = NULL; struct rt_mutex_waiter rt_waiter; - struct rt_mutex *pi_mutex = NULL; struct futex_hash_bucket *hb; union futex_key key2 = FUTEX_KEY_INIT; struct futex_q q = futex_q_init; @@ -2839,10 +3003,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, * The waiter is allocated on our stack, manipulated by the requeue * code while we sleep on uaddr. */ - debug_rt_mutex_init_waiter(&rt_waiter); - RB_CLEAR_NODE(&rt_waiter.pi_tree_entry); - RB_CLEAR_NODE(&rt_waiter.tree_entry); - rt_waiter.task = NULL; + rt_mutex_init_waiter(&rt_waiter); ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2, VERIFY_WRITE); if (unlikely(ret != 0)) @@ -2897,6 +3058,10 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (q.pi_state && (q.pi_state->owner != current)) { spin_lock(q.lock_ptr); ret = fixup_pi_state_owner(uaddr2, &q, current); + if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) { + pi_state = q.pi_state; + get_pi_state(pi_state); + } /* * Drop the reference to the pi state which * the requeue_pi() code acquired for us. @@ -2905,6 +3070,8 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, spin_unlock(q.lock_ptr); } } else { + struct rt_mutex *pi_mutex; + /* * We have been woken up by futex_unlock_pi(), a timeout, or a * signal. futex_unlock_pi() will not destroy the lock_ptr nor @@ -2912,10 +3079,13 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, */ WARN_ON(!q.pi_state); pi_mutex = &q.pi_state->pi_mutex; - ret = rt_mutex_finish_proxy_lock(pi_mutex, to, &rt_waiter); - debug_rt_mutex_free_waiter(&rt_waiter); + ret = rt_mutex_wait_proxy_lock(pi_mutex, to, &rt_waiter); spin_lock(q.lock_ptr); + if (ret && !rt_mutex_cleanup_proxy_lock(pi_mutex, &rt_waiter)) + ret = 0; + + debug_rt_mutex_free_waiter(&rt_waiter); /* * Fixup the pi_state owner and possibly acquire the lock if we * haven't already. @@ -2928,18 +3098,26 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (res) ret = (res < 0) ? res : 0; + /* + * If fixup_pi_state_owner() faulted and was unable to handle + * the fault, unlock the rt_mutex and return the fault to + * userspace. + */ + if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) { + pi_state = q.pi_state; + get_pi_state(pi_state); + } + /* Unqueue and drop the lock. */ unqueue_me_pi(&q); } - /* - * If fixup_pi_state_owner() faulted and was unable to handle the - * fault, unlock the rt_mutex and return the fault to userspace. - */ - if (ret == -EFAULT) { - if (pi_mutex && rt_mutex_owner(pi_mutex) == current) - rt_mutex_unlock(pi_mutex); - } else if (ret == -EINTR) { + if (pi_state) { + rt_mutex_futex_unlock(&pi_state->pi_mutex); + put_pi_state(pi_state); + } + + if (ret == -EINTR) { /* * We've already been requeued, but cannot restart by calling * futex_lock_pi() directly. We could restart this syscall, but @@ -3323,4 +3501,4 @@ static int __init futex_init(void) return 0; } -__initcall(futex_init); +core_initcall(futex_init); diff --git a/kernel/gcov/base.c b/kernel/gcov/base.c index 2f9df37940a0..c51a49c9be70 100644 --- a/kernel/gcov/base.c +++ b/kernel/gcov/base.c @@ -98,6 +98,12 @@ void __gcov_merge_icall_topn(gcov_type *counters, unsigned int n_counters) } EXPORT_SYMBOL(__gcov_merge_icall_topn); +void __gcov_exit(void) +{ + /* Unused. */ +} +EXPORT_SYMBOL(__gcov_exit); + /** * gcov_enable_events - enable event reporting through gcov_event() * diff --git a/kernel/gcov/gcc_4_7.c b/kernel/gcov/gcc_4_7.c index 6a5c239c7669..46a18e72bce6 100644 --- a/kernel/gcov/gcc_4_7.c +++ b/kernel/gcov/gcc_4_7.c @@ -18,7 +18,9 @@ #include <linux/vmalloc.h> #include "gcov.h" -#if (__GNUC__ > 5) || (__GNUC__ == 5 && __GNUC_MINOR__ >= 1) +#if (__GNUC__ >= 7) +#define GCOV_COUNTERS 9 +#elif (__GNUC__ > 5) || (__GNUC__ == 5 && __GNUC_MINOR__ >= 1) #define GCOV_COUNTERS 10 #elif __GNUC__ == 4 && __GNUC_MINOR__ >= 9 #define GCOV_COUNTERS 9 diff --git a/kernel/groups.c b/kernel/groups.c index 8dd7a61b7115..434f6665f187 100644 --- a/kernel/groups.c +++ b/kernel/groups.c @@ -5,6 +5,7 @@ #include <linux/export.h> #include <linux/slab.h> #include <linux/security.h> +#include <linux/sort.h> #include <linux/syscalls.h> #include <linux/user_namespace.h> #include <linux/vmalloc.h> @@ -18,7 +19,7 @@ struct group_info *groups_alloc(int gidsetsize) len = sizeof(struct group_info) + sizeof(kgid_t) * gidsetsize; gi = kmalloc(len, GFP_KERNEL_ACCOUNT|__GFP_NOWARN|__GFP_NORETRY); if (!gi) - gi = __vmalloc(len, GFP_KERNEL_ACCOUNT|__GFP_HIGHMEM, PAGE_KERNEL); + gi = __vmalloc(len, GFP_KERNEL_ACCOUNT, PAGE_KERNEL); if (!gi) return NULL; @@ -76,32 +77,18 @@ static int groups_from_user(struct group_info *group_info, return 0; } -/* a simple Shell sort */ +static int gid_cmp(const void *_a, const void *_b) +{ + kgid_t a = *(kgid_t *)_a; + kgid_t b = *(kgid_t *)_b; + + return gid_gt(a, b) - gid_lt(a, b); +} + static void groups_sort(struct group_info *group_info) { - int base, max, stride; - int gidsetsize = group_info->ngroups; - - for (stride = 1; stride < gidsetsize; stride = 3 * stride + 1) - ; /* nothing */ - stride /= 3; - - while (stride) { - max = gidsetsize - stride; - for (base = 0; base < max; base++) { - int left = base; - int right = left + stride; - kgid_t tmp = group_info->gid[right]; - - while (left >= 0 && gid_gt(group_info->gid[left], tmp)) { - group_info->gid[right] = group_info->gid[left]; - right = left; - left -= stride; - } - group_info->gid[right] = tmp; - } - stride /= 3; - } + sort(group_info->gid, group_info->ngroups, sizeof(*group_info->gid), + gid_cmp, NULL); } /* a simple bsearch */ diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 40c07e4fa116..751593ed7c0b 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -16,6 +16,9 @@ #include <linux/export.h> #include <linux/sysctl.h> #include <linux/utsname.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> + #include <trace/events/sched.h> /* @@ -40,6 +43,7 @@ unsigned long __read_mostly sysctl_hung_task_timeout_secs = CONFIG_DEFAULT_HUNG_ int __read_mostly sysctl_hung_task_warnings = 10; static int __read_mostly did_panic; +static bool hung_task_show_lock; static struct task_struct *watchdog_task; @@ -117,12 +121,14 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout) pr_err("\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\"" " disables this message.\n"); sched_show_task(t); - debug_show_all_locks(); + hung_task_show_lock = true; } touch_nmi_watchdog(); if (sysctl_hung_task_panic) { + if (hung_task_show_lock) + debug_show_all_locks(); trigger_all_cpu_backtrace(); panic("hung_task: blocked tasks"); } @@ -169,6 +175,7 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) if (test_taint(TAINT_DIE) || did_panic) return; + hung_task_show_lock = false; rcu_read_lock(); for_each_process_thread(g, t) { if (!max_count--) @@ -184,6 +191,8 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) } unlock: rcu_read_unlock(); + if (hung_task_show_lock) + debug_show_all_locks(); } static long hung_timeout_jiffies(unsigned long last_checked, diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig index 3bbfd6a9c475..27c4e774071c 100644 --- a/kernel/irq/Kconfig +++ b/kernel/irq/Kconfig @@ -21,6 +21,10 @@ config GENERIC_IRQ_SHOW config GENERIC_IRQ_SHOW_LEVEL bool +# Supports effective affinity mask +config GENERIC_IRQ_EFFECTIVE_AFF_MASK + bool + # Facility to allocate a hardware interrupt. This is legacy support # and should not be used in new code. Use irq domains instead. config GENERIC_IRQ_LEGACY_ALLOC_HWIRQ @@ -81,6 +85,9 @@ config GENERIC_MSI_IRQ_DOMAIN config HANDLE_DOMAIN_IRQ bool +config IRQ_TIMINGS + bool + config IRQ_DOMAIN_DEBUG bool "Expose hardware/virtual IRQ mapping via debugfs" depends on IRQ_DOMAIN && DEBUG_FS @@ -108,4 +115,15 @@ config SPARSE_IRQ If you don't know what to do here, say N. +config GENERIC_IRQ_DEBUGFS + bool "Expose irq internals in debugfs" + depends on DEBUG_FS + default n + ---help--- + + Exposes internal state information through debugfs. Mostly for + developers and debugging of hard to diagnose interrupt problems. + + If you don't know what to do here, say N. + endmenu diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile index 1d3ee3169202..e4aef7351f2b 100644 --- a/kernel/irq/Makefile +++ b/kernel/irq/Makefile @@ -1,5 +1,6 @@ obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devres.o +obj-$(CONFIG_IRQ_TIMINGS) += timings.o obj-$(CONFIG_GENERIC_IRQ_CHIP) += generic-chip.o obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o obj-$(CONFIG_IRQ_DOMAIN) += irqdomain.o @@ -10,3 +11,4 @@ obj-$(CONFIG_PM_SLEEP) += pm.o obj-$(CONFIG_GENERIC_MSI_IRQ) += msi.o obj-$(CONFIG_GENERIC_IRQ_IPI) += ipi.o obj-$(CONFIG_SMP) += affinity.o +obj-$(CONFIG_GENERIC_IRQ_DEBUGFS) += debugfs.o diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c index 4544b115f5eb..d69bd77252a7 100644 --- a/kernel/irq/affinity.c +++ b/kernel/irq/affinity.c @@ -1,4 +1,7 @@ - +/* + * Copyright (C) 2016 Thomas Gleixner. + * Copyright (C) 2016-2017 Christoph Hellwig. + */ #include <linux/interrupt.h> #include <linux/kernel.h> #include <linux/slab.h> @@ -35,13 +38,54 @@ static void irq_spread_init_one(struct cpumask *irqmsk, struct cpumask *nmsk, } } -static int get_nodes_in_cpumask(const struct cpumask *mask, nodemask_t *nodemsk) +static cpumask_var_t *alloc_node_to_present_cpumask(void) +{ + cpumask_var_t *masks; + int node; + + masks = kcalloc(nr_node_ids, sizeof(cpumask_var_t), GFP_KERNEL); + if (!masks) + return NULL; + + for (node = 0; node < nr_node_ids; node++) { + if (!zalloc_cpumask_var(&masks[node], GFP_KERNEL)) + goto out_unwind; + } + + return masks; + +out_unwind: + while (--node >= 0) + free_cpumask_var(masks[node]); + kfree(masks); + return NULL; +} + +static void free_node_to_present_cpumask(cpumask_var_t *masks) +{ + int node; + + for (node = 0; node < nr_node_ids; node++) + free_cpumask_var(masks[node]); + kfree(masks); +} + +static void build_node_to_present_cpumask(cpumask_var_t *masks) +{ + int cpu; + + for_each_present_cpu(cpu) + cpumask_set_cpu(cpu, masks[cpu_to_node(cpu)]); +} + +static int get_nodes_in_cpumask(cpumask_var_t *node_to_present_cpumask, + const struct cpumask *mask, nodemask_t *nodemsk) { int n, nodes = 0; /* Calculate the number of nodes in the supplied affinity mask */ - for_each_online_node(n) { - if (cpumask_intersects(mask, cpumask_of_node(n))) { + for_each_node(n) { + if (cpumask_intersects(mask, node_to_present_cpumask[n])) { node_set(n, *nodemsk); nodes++; } @@ -59,12 +103,19 @@ static int get_nodes_in_cpumask(const struct cpumask *mask, nodemask_t *nodemsk) struct cpumask * irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) { - int n, nodes, vecs_per_node, cpus_per_vec, extra_vecs, curvec; + int n, nodes, cpus_per_vec, extra_vecs, curvec; int affv = nvecs - affd->pre_vectors - affd->post_vectors; int last_affv = affv + affd->pre_vectors; nodemask_t nodemsk = NODE_MASK_NONE; struct cpumask *masks; - cpumask_var_t nmsk; + cpumask_var_t nmsk, *node_to_present_cpumask; + + /* + * If there aren't any vectors left after applying the pre/post + * vectors don't bother with assigning affinity. + */ + if (!affv) + return NULL; if (!zalloc_cpumask_var(&nmsk, GFP_KERNEL)) return NULL; @@ -73,13 +124,19 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) if (!masks) goto out; + node_to_present_cpumask = alloc_node_to_present_cpumask(); + if (!node_to_present_cpumask) + goto out; + /* Fill out vectors at the beginning that don't need affinity */ for (curvec = 0; curvec < affd->pre_vectors; curvec++) cpumask_copy(masks + curvec, irq_default_affinity); /* Stabilize the cpumasks */ get_online_cpus(); - nodes = get_nodes_in_cpumask(cpu_online_mask, &nodemsk); + build_node_to_present_cpumask(node_to_present_cpumask); + nodes = get_nodes_in_cpumask(node_to_present_cpumask, cpu_present_mask, + &nodemsk); /* * If the number of nodes in the mask is greater than or equal the @@ -87,26 +144,29 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) */ if (affv <= nodes) { for_each_node_mask(n, nodemsk) { - cpumask_copy(masks + curvec, cpumask_of_node(n)); + cpumask_copy(masks + curvec, + node_to_present_cpumask[n]); if (++curvec == last_affv) break; } goto done; } - /* Spread the vectors per node */ - vecs_per_node = affv / nodes; - /* Account for rounding errors */ - extra_vecs = affv - (nodes * vecs_per_node); - for_each_node_mask(n, nodemsk) { - int ncpus, v, vecs_to_assign = vecs_per_node; + int ncpus, v, vecs_to_assign, vecs_per_node; + + /* Spread the vectors per node */ + vecs_per_node = (affv - (curvec - affd->pre_vectors)) / nodes; /* Get the cpus on this node which are in the mask */ - cpumask_and(nmsk, cpu_online_mask, cpumask_of_node(n)); + cpumask_and(nmsk, cpu_present_mask, node_to_present_cpumask[n]); /* Calculate the number of cpus per vector */ ncpus = cpumask_weight(nmsk); + vecs_to_assign = min(vecs_per_node, ncpus); + + /* Account for rounding errors */ + extra_vecs = ncpus - vecs_to_assign * (ncpus / vecs_to_assign); for (v = 0; curvec < last_affv && v < vecs_to_assign; curvec++, v++) { @@ -115,14 +175,14 @@ irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd) /* Account for extra vectors to compensate rounding errors */ if (extra_vecs) { cpus_per_vec++; - if (!--extra_vecs) - vecs_per_node++; + --extra_vecs; } irq_spread_init_one(masks + curvec, nmsk, cpus_per_vec); } if (curvec >= last_affv) break; + --nodes; } done: @@ -131,6 +191,7 @@ done: /* Fill out vectors at the end that don't need affinity */ for (; curvec < nvecs; curvec++) cpumask_copy(masks + curvec, irq_default_affinity); + free_node_to_present_cpumask(node_to_present_cpumask); out: free_cpumask_var(nmsk); return masks; @@ -138,19 +199,21 @@ out: /** * irq_calc_affinity_vectors - Calculate the optimal number of vectors + * @minvec: The minimum number of vectors available * @maxvec: The maximum number of vectors available * @affd: Description of the affinity requirements */ -int irq_calc_affinity_vectors(int maxvec, const struct irq_affinity *affd) +int irq_calc_affinity_vectors(int minvec, int maxvec, const struct irq_affinity *affd) { int resv = affd->pre_vectors + affd->post_vectors; int vecs = maxvec - resv; - int cpus; + int ret; + + if (resv > minvec) + return 0; - /* Stabilize the cpumasks */ get_online_cpus(); - cpus = cpumask_weight(cpu_online_mask); + ret = min_t(int, cpumask_weight(cpu_present_mask), vecs) + resv; put_online_cpus(); - - return min(cpus, vecs) + resv; + return ret; } diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c index 0119b9d467ae..d30a0dd5cc02 100644 --- a/kernel/irq/autoprobe.c +++ b/kernel/irq/autoprobe.c @@ -53,7 +53,7 @@ unsigned long probe_irq_on(void) if (desc->irq_data.chip->irq_set_type) desc->irq_data.chip->irq_set_type(&desc->irq_data, IRQ_TYPE_PROBE); - irq_startup(desc, false); + irq_startup(desc, IRQ_NORESEND, IRQ_START_FORCE); } raw_spin_unlock_irq(&desc->lock); } @@ -70,7 +70,7 @@ unsigned long probe_irq_on(void) raw_spin_lock_irq(&desc->lock); if (!desc->action && irq_settings_can_probe(desc)) { desc->istate |= IRQS_AUTODETECT | IRQS_WAITING; - if (irq_startup(desc, false)) + if (irq_startup(desc, IRQ_NORESEND, IRQ_START_FORCE)) desc->istate |= IRQS_PENDING; } raw_spin_unlock_irq(&desc->lock); diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index be3c34e4f2ac..a3cc37c0c85e 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -7,7 +7,7 @@ * This file contains the core interrupt handling code, for irq-chip * based architectures. * - * Detailed information is available in Documentation/DocBook/genericirq + * Detailed information is available in Documentation/core-api/genericirq.rst */ #include <linux/irq.h> @@ -170,62 +170,167 @@ static void irq_state_clr_disabled(struct irq_desc *desc) irqd_clear(&desc->irq_data, IRQD_IRQ_DISABLED); } -static void irq_state_set_disabled(struct irq_desc *desc) +static void irq_state_clr_masked(struct irq_desc *desc) { - irqd_set(&desc->irq_data, IRQD_IRQ_DISABLED); + irqd_clear(&desc->irq_data, IRQD_IRQ_MASKED); } -static void irq_state_clr_masked(struct irq_desc *desc) +static void irq_state_clr_started(struct irq_desc *desc) { - irqd_clear(&desc->irq_data, IRQD_IRQ_MASKED); + irqd_clear(&desc->irq_data, IRQD_IRQ_STARTED); } -static void irq_state_set_masked(struct irq_desc *desc) +static void irq_state_set_started(struct irq_desc *desc) { - irqd_set(&desc->irq_data, IRQD_IRQ_MASKED); + irqd_set(&desc->irq_data, IRQD_IRQ_STARTED); } -int irq_startup(struct irq_desc *desc, bool resend) +enum { + IRQ_STARTUP_NORMAL, + IRQ_STARTUP_MANAGED, + IRQ_STARTUP_ABORT, +}; + +#ifdef CONFIG_SMP +static int +__irq_startup_managed(struct irq_desc *desc, struct cpumask *aff, bool force) { - int ret = 0; + struct irq_data *d = irq_desc_get_irq_data(desc); - irq_state_clr_disabled(desc); - desc->depth = 0; + if (!irqd_affinity_is_managed(d)) + return IRQ_STARTUP_NORMAL; + + irqd_clr_managed_shutdown(d); + + if (cpumask_any_and(aff, cpu_online_mask) > nr_cpu_ids) { + /* + * Catch code which fiddles with enable_irq() on a managed + * and potentially shutdown IRQ. Chained interrupt + * installment or irq auto probing should not happen on + * managed irqs either. Emit a warning, break the affinity + * and start it up as a normal interrupt. + */ + if (WARN_ON_ONCE(force)) + return IRQ_STARTUP_NORMAL; + /* + * The interrupt was requested, but there is no online CPU + * in it's affinity mask. Put it into managed shutdown + * state and let the cpu hotplug mechanism start it up once + * a CPU in the mask becomes available. + */ + irqd_set_managed_shutdown(d); + return IRQ_STARTUP_ABORT; + } + return IRQ_STARTUP_MANAGED; +} +#else +static __always_inline int +__irq_startup_managed(struct irq_desc *desc, struct cpumask *aff, bool force) +{ + return IRQ_STARTUP_NORMAL; +} +#endif + +static int __irq_startup(struct irq_desc *desc) +{ + struct irq_data *d = irq_desc_get_irq_data(desc); + int ret = 0; - irq_domain_activate_irq(&desc->irq_data); - if (desc->irq_data.chip->irq_startup) { - ret = desc->irq_data.chip->irq_startup(&desc->irq_data); + irq_domain_activate_irq(d); + if (d->chip->irq_startup) { + ret = d->chip->irq_startup(d); + irq_state_clr_disabled(desc); irq_state_clr_masked(desc); } else { irq_enable(desc); } + irq_state_set_started(desc); + return ret; +} + +int irq_startup(struct irq_desc *desc, bool resend, bool force) +{ + struct irq_data *d = irq_desc_get_irq_data(desc); + struct cpumask *aff = irq_data_get_affinity_mask(d); + int ret = 0; + + desc->depth = 0; + + if (irqd_is_started(d)) { + irq_enable(desc); + } else { + switch (__irq_startup_managed(desc, aff, force)) { + case IRQ_STARTUP_NORMAL: + ret = __irq_startup(desc); + irq_setup_affinity(desc); + break; + case IRQ_STARTUP_MANAGED: + ret = __irq_startup(desc); + irq_set_affinity_locked(d, aff, false); + break; + case IRQ_STARTUP_ABORT: + return 0; + } + } if (resend) check_irq_resend(desc); + return ret; } +static void __irq_disable(struct irq_desc *desc, bool mask); + void irq_shutdown(struct irq_desc *desc) { - irq_state_set_disabled(desc); - desc->depth = 1; - if (desc->irq_data.chip->irq_shutdown) - desc->irq_data.chip->irq_shutdown(&desc->irq_data); - else if (desc->irq_data.chip->irq_disable) - desc->irq_data.chip->irq_disable(&desc->irq_data); - else - desc->irq_data.chip->irq_mask(&desc->irq_data); + if (irqd_is_started(&desc->irq_data)) { + desc->depth = 1; + if (desc->irq_data.chip->irq_shutdown) { + desc->irq_data.chip->irq_shutdown(&desc->irq_data); + irq_state_set_disabled(desc); + irq_state_set_masked(desc); + } else { + __irq_disable(desc, true); + } + irq_state_clr_started(desc); + } + /* + * This must be called even if the interrupt was never started up, + * because the activation can happen before the interrupt is + * available for request/startup. It has it's own state tracking so + * it's safe to call it unconditionally. + */ irq_domain_deactivate_irq(&desc->irq_data); - irq_state_set_masked(desc); } void irq_enable(struct irq_desc *desc) { - irq_state_clr_disabled(desc); - if (desc->irq_data.chip->irq_enable) - desc->irq_data.chip->irq_enable(&desc->irq_data); - else - desc->irq_data.chip->irq_unmask(&desc->irq_data); - irq_state_clr_masked(desc); + if (!irqd_irq_disabled(&desc->irq_data)) { + unmask_irq(desc); + } else { + irq_state_clr_disabled(desc); + if (desc->irq_data.chip->irq_enable) { + desc->irq_data.chip->irq_enable(&desc->irq_data); + irq_state_clr_masked(desc); + } else { + unmask_irq(desc); + } + } +} + +static void __irq_disable(struct irq_desc *desc, bool mask) +{ + if (irqd_irq_disabled(&desc->irq_data)) { + if (mask) + mask_irq(desc); + } else { + irq_state_set_disabled(desc); + if (desc->irq_data.chip->irq_disable) { + desc->irq_data.chip->irq_disable(&desc->irq_data); + irq_state_set_masked(desc); + } else if (mask) { + mask_irq(desc); + } + } } /** @@ -250,13 +355,7 @@ void irq_enable(struct irq_desc *desc) */ void irq_disable(struct irq_desc *desc) { - irq_state_set_disabled(desc); - if (desc->irq_data.chip->irq_disable) { - desc->irq_data.chip->irq_disable(&desc->irq_data); - irq_state_set_masked(desc); - } else if (irq_settings_disable_unlazy(desc)) { - mask_irq(desc); - } + __irq_disable(desc, irq_settings_disable_unlazy(desc)); } void irq_percpu_enable(struct irq_desc *desc, unsigned int cpu) @@ -279,18 +378,21 @@ void irq_percpu_disable(struct irq_desc *desc, unsigned int cpu) static inline void mask_ack_irq(struct irq_desc *desc) { - if (desc->irq_data.chip->irq_mask_ack) + if (desc->irq_data.chip->irq_mask_ack) { desc->irq_data.chip->irq_mask_ack(&desc->irq_data); - else { - desc->irq_data.chip->irq_mask(&desc->irq_data); + irq_state_set_masked(desc); + } else { + mask_irq(desc); if (desc->irq_data.chip->irq_ack) desc->irq_data.chip->irq_ack(&desc->irq_data); } - irq_state_set_masked(desc); } void mask_irq(struct irq_desc *desc) { + if (irqd_irq_masked(&desc->irq_data)) + return; + if (desc->irq_data.chip->irq_mask) { desc->irq_data.chip->irq_mask(&desc->irq_data); irq_state_set_masked(desc); @@ -299,6 +401,9 @@ void mask_irq(struct irq_desc *desc) void unmask_irq(struct irq_desc *desc) { + if (!irqd_irq_masked(&desc->irq_data)) + return; + if (desc->irq_data.chip->irq_unmask) { desc->irq_data.chip->irq_unmask(&desc->irq_data); irq_state_clr_masked(desc); @@ -312,10 +417,7 @@ void unmask_threaded_irq(struct irq_desc *desc) if (chip->flags & IRQCHIP_EOI_THREADED) chip->irq_eoi(&desc->irq_data); - if (chip->irq_unmask) { - chip->irq_unmask(&desc->irq_data); - irq_state_clr_masked(desc); - } + unmask_irq(desc); } /* @@ -348,7 +450,10 @@ void handle_nested_irq(unsigned int irq) irqd_set(&desc->irq_data, IRQD_IRQ_INPROGRESS); raw_spin_unlock_irq(&desc->lock); - action_ret = action->thread_fn(action->irq, action->dev_id); + action_ret = IRQ_NONE; + for_each_action_of_desc(desc, action) + action_ret |= action->thread_fn(action->irq, action->dev_id); + if (!noirqdebug) note_interrupt(desc, action_ret); @@ -848,7 +953,7 @@ __irq_do_set_handler(struct irq_desc *desc, irq_flow_handler_t handle, irq_settings_set_norequest(desc); irq_settings_set_nothread(desc); desc->action = &chained_action; - irq_startup(desc, true); + irq_startup(desc, IRQ_RESEND, IRQ_START_FORCE); } } @@ -877,8 +982,8 @@ irq_set_chained_handler_and_data(unsigned int irq, irq_flow_handler_t handle, if (!desc) return; - __irq_do_set_handler(desc, handle, 1, NULL); desc->irq_common_data.handler_data = data; + __irq_do_set_handler(desc, handle, 1, NULL); irq_put_desc_busunlock(desc, flags); } @@ -900,6 +1005,13 @@ void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set) if (!desc) return; + + /* + * Warn when a driver sets the no autoenable flag on an already + * active interrupt. + */ + WARN_ON_ONCE(!desc->depth && (set & _IRQ_NOAUTOEN)); + irq_settings_clr_and_set(desc, clr, set); irqd_clear(&desc->irq_data, IRQD_NO_BALANCING | IRQD_PER_CPU | diff --git a/kernel/irq/cpuhotplug.c b/kernel/irq/cpuhotplug.c index 011f8c4c63da..aee8f7ec40af 100644 --- a/kernel/irq/cpuhotplug.c +++ b/kernel/irq/cpuhotplug.c @@ -14,37 +14,99 @@ #include "internals.h" +/* For !GENERIC_IRQ_EFFECTIVE_AFF_MASK this looks at general affinity mask */ +static inline bool irq_needs_fixup(struct irq_data *d) +{ + const struct cpumask *m = irq_data_get_effective_affinity_mask(d); + + return cpumask_test_cpu(smp_processor_id(), m); +} + static bool migrate_one_irq(struct irq_desc *desc) { struct irq_data *d = irq_desc_get_irq_data(desc); - const struct cpumask *affinity = d->common->affinity; - struct irq_chip *c; - bool ret = false; + struct irq_chip *chip = irq_data_get_irq_chip(d); + bool maskchip = !irq_can_move_pcntxt(d) && !irqd_irq_masked(d); + const struct cpumask *affinity; + bool brokeaff = false; + int err; /* - * If this is a per-CPU interrupt, or the affinity does not - * include this CPU, then we have nothing to do. + * IRQ chip might be already torn down, but the irq descriptor is + * still in the radix tree. Also if the chip has no affinity setter, + * nothing can be done here. */ - if (irqd_is_per_cpu(d) || - !cpumask_test_cpu(smp_processor_id(), affinity)) + if (!chip || !chip->irq_set_affinity) { + pr_debug("IRQ %u: Unable to migrate away\n", d->irq); return false; + } + + /* + * No move required, if: + * - Interrupt is per cpu + * - Interrupt is not started + * - Affinity mask does not include this CPU. + * + * Note: Do not check desc->action as this might be a chained + * interrupt. + */ + if (irqd_is_per_cpu(d) || !irqd_is_started(d) || !irq_needs_fixup(d)) { + /* + * If an irq move is pending, abort it if the dying CPU is + * the sole target. + */ + irq_fixup_move_pending(desc, false); + return false; + } + + /* + * Complete an eventually pending irq move cleanup. If this + * interrupt was moved in hard irq context, then the vectors need + * to be cleaned up. It can't wait until this interrupt actually + * happens and this CPU was involved. + */ + irq_force_complete_move(desc); + + /* + * If there is a setaffinity pending, then try to reuse the pending + * mask, so the last change of the affinity does not get lost. If + * there is no move pending or the pending mask does not contain + * any online CPU, use the current affinity mask. + */ + if (irq_fixup_move_pending(desc, true)) + affinity = irq_desc_get_pending_mask(desc); + else + affinity = irq_data_get_affinity_mask(d); + + /* Mask the chip for interrupts which cannot move in process context */ + if (maskchip && chip->irq_mask) + chip->irq_mask(d); if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) { + /* + * If the interrupt is managed, then shut it down and leave + * the affinity untouched. + */ + if (irqd_affinity_is_managed(d)) { + irqd_set_managed_shutdown(d); + irq_shutdown(desc); + return false; + } affinity = cpu_online_mask; - ret = true; + brokeaff = true; } - c = irq_data_get_irq_chip(d); - if (!c->irq_set_affinity) { - pr_debug("IRQ%u: unable to set affinity\n", d->irq); - } else { - int r = irq_do_set_affinity(d, affinity, false); - if (r) - pr_warn_ratelimited("IRQ%u: set affinity failed(%d).\n", - d->irq, r); + err = irq_do_set_affinity(d, affinity, true); + if (err) { + pr_warn_ratelimited("IRQ%u: set affinity failed(%d).\n", + d->irq, err); + brokeaff = false; } - return ret; + if (maskchip && chip->irq_unmask) + chip->irq_unmask(d); + + return brokeaff; } /** @@ -59,11 +121,8 @@ static bool migrate_one_irq(struct irq_desc *desc) */ void irq_migrate_all_off_this_cpu(void) { - unsigned int irq; struct irq_desc *desc; - unsigned long flags; - - local_irq_save(flags); + unsigned int irq; for_each_active_irq(irq) { bool affinity_broken; @@ -73,10 +132,53 @@ void irq_migrate_all_off_this_cpu(void) affinity_broken = migrate_one_irq(desc); raw_spin_unlock(&desc->lock); - if (affinity_broken) - pr_warn_ratelimited("IRQ%u no longer affine to CPU%u\n", + if (affinity_broken) { + pr_warn_ratelimited("IRQ %u: no longer affine to CPU%u\n", irq, smp_processor_id()); + } + } +} + +static void irq_restore_affinity_of_irq(struct irq_desc *desc, unsigned int cpu) +{ + struct irq_data *data = irq_desc_get_irq_data(desc); + const struct cpumask *affinity = irq_data_get_affinity_mask(data); + + if (!irqd_affinity_is_managed(data) || !desc->action || + !irq_data_get_irq_chip(data) || !cpumask_test_cpu(cpu, affinity)) + return; + + if (irqd_is_managed_and_shutdown(data)) { + irq_startup(desc, IRQ_RESEND, IRQ_START_COND); + return; + } + + /* + * If the interrupt can only be directed to a single target + * CPU then it is already assigned to a CPU in the affinity + * mask. No point in trying to move it around. + */ + if (!irqd_is_single_target(data)) + irq_set_affinity_locked(data, affinity, false); +} + +/** + * irq_affinity_online_cpu - Restore affinity for managed interrupts + * @cpu: Upcoming CPU for which interrupts should be restored + */ +int irq_affinity_online_cpu(unsigned int cpu) +{ + struct irq_desc *desc; + unsigned int irq; + + irq_lock_sparse(); + for_each_active_irq(irq) { + desc = irq_to_desc(irq); + raw_spin_lock_irq(&desc->lock); + irq_restore_affinity_of_irq(desc, cpu); + raw_spin_unlock_irq(&desc->lock); } + irq_unlock_sparse(); - local_irq_restore(flags); + return 0; } diff --git a/kernel/irq/debugfs.c b/kernel/irq/debugfs.c new file mode 100644 index 000000000000..4d384edc0c64 --- /dev/null +++ b/kernel/irq/debugfs.c @@ -0,0 +1,213 @@ +/* + * Copyright 2017 Thomas Gleixner <[email protected]> + * + * This file is licensed under the GPL V2. + */ +#include <linux/irqdomain.h> +#include <linux/irq.h> + +#include "internals.h" + +static struct dentry *irq_dir; + +struct irq_bit_descr { + unsigned int mask; + char *name; +}; +#define BIT_MASK_DESCR(m) { .mask = m, .name = #m } + +static void irq_debug_show_bits(struct seq_file *m, int ind, unsigned int state, + const struct irq_bit_descr *sd, int size) +{ + int i; + + for (i = 0; i < size; i++, sd++) { + if (state & sd->mask) + seq_printf(m, "%*s%s\n", ind + 12, "", sd->name); + } +} + +#ifdef CONFIG_SMP +static void irq_debug_show_masks(struct seq_file *m, struct irq_desc *desc) +{ + struct irq_data *data = irq_desc_get_irq_data(desc); + struct cpumask *msk; + + msk = irq_data_get_affinity_mask(data); + seq_printf(m, "affinity: %*pbl\n", cpumask_pr_args(msk)); +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + msk = irq_data_get_effective_affinity_mask(data); + seq_printf(m, "effectiv: %*pbl\n", cpumask_pr_args(msk)); +#endif +#ifdef CONFIG_GENERIC_PENDING_IRQ + msk = desc->pending_mask; + seq_printf(m, "pending: %*pbl\n", cpumask_pr_args(msk)); +#endif +} +#else +static void irq_debug_show_masks(struct seq_file *m, struct irq_desc *desc) { } +#endif + +static const struct irq_bit_descr irqchip_flags[] = { + BIT_MASK_DESCR(IRQCHIP_SET_TYPE_MASKED), + BIT_MASK_DESCR(IRQCHIP_EOI_IF_HANDLED), + BIT_MASK_DESCR(IRQCHIP_MASK_ON_SUSPEND), + BIT_MASK_DESCR(IRQCHIP_ONOFFLINE_ENABLED), + BIT_MASK_DESCR(IRQCHIP_SKIP_SET_WAKE), + BIT_MASK_DESCR(IRQCHIP_ONESHOT_SAFE), + BIT_MASK_DESCR(IRQCHIP_EOI_THREADED), +}; + +static void +irq_debug_show_chip(struct seq_file *m, struct irq_data *data, int ind) +{ + struct irq_chip *chip = data->chip; + + if (!chip) { + seq_printf(m, "chip: None\n"); + return; + } + seq_printf(m, "%*schip: %s\n", ind, "", chip->name); + seq_printf(m, "%*sflags: 0x%lx\n", ind + 1, "", chip->flags); + irq_debug_show_bits(m, ind, chip->flags, irqchip_flags, + ARRAY_SIZE(irqchip_flags)); +} + +static void +irq_debug_show_data(struct seq_file *m, struct irq_data *data, int ind) +{ + seq_printf(m, "%*sdomain: %s\n", ind, "", + data->domain ? data->domain->name : ""); + seq_printf(m, "%*shwirq: 0x%lx\n", ind + 1, "", data->hwirq); + irq_debug_show_chip(m, data, ind + 1); +#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY + if (!data->parent_data) + return; + seq_printf(m, "%*sparent:\n", ind + 1, ""); + irq_debug_show_data(m, data->parent_data, ind + 4); +#endif +} + +static const struct irq_bit_descr irqdata_states[] = { + BIT_MASK_DESCR(IRQ_TYPE_EDGE_RISING), + BIT_MASK_DESCR(IRQ_TYPE_EDGE_FALLING), + BIT_MASK_DESCR(IRQ_TYPE_LEVEL_HIGH), + BIT_MASK_DESCR(IRQ_TYPE_LEVEL_LOW), + BIT_MASK_DESCR(IRQD_LEVEL), + + BIT_MASK_DESCR(IRQD_ACTIVATED), + BIT_MASK_DESCR(IRQD_IRQ_STARTED), + BIT_MASK_DESCR(IRQD_IRQ_DISABLED), + BIT_MASK_DESCR(IRQD_IRQ_MASKED), + BIT_MASK_DESCR(IRQD_IRQ_INPROGRESS), + + BIT_MASK_DESCR(IRQD_PER_CPU), + BIT_MASK_DESCR(IRQD_NO_BALANCING), + + BIT_MASK_DESCR(IRQD_SINGLE_TARGET), + BIT_MASK_DESCR(IRQD_MOVE_PCNTXT), + BIT_MASK_DESCR(IRQD_AFFINITY_SET), + BIT_MASK_DESCR(IRQD_SETAFFINITY_PENDING), + BIT_MASK_DESCR(IRQD_AFFINITY_MANAGED), + BIT_MASK_DESCR(IRQD_MANAGED_SHUTDOWN), + + BIT_MASK_DESCR(IRQD_FORWARDED_TO_VCPU), + + BIT_MASK_DESCR(IRQD_WAKEUP_STATE), + BIT_MASK_DESCR(IRQD_WAKEUP_ARMED), +}; + +static const struct irq_bit_descr irqdesc_states[] = { + BIT_MASK_DESCR(_IRQ_NOPROBE), + BIT_MASK_DESCR(_IRQ_NOREQUEST), + BIT_MASK_DESCR(_IRQ_NOTHREAD), + BIT_MASK_DESCR(_IRQ_NOAUTOEN), + BIT_MASK_DESCR(_IRQ_NESTED_THREAD), + BIT_MASK_DESCR(_IRQ_PER_CPU_DEVID), + BIT_MASK_DESCR(_IRQ_IS_POLLED), + BIT_MASK_DESCR(_IRQ_DISABLE_UNLAZY), +}; + +static const struct irq_bit_descr irqdesc_istates[] = { + BIT_MASK_DESCR(IRQS_AUTODETECT), + BIT_MASK_DESCR(IRQS_SPURIOUS_DISABLED), + BIT_MASK_DESCR(IRQS_POLL_INPROGRESS), + BIT_MASK_DESCR(IRQS_ONESHOT), + BIT_MASK_DESCR(IRQS_REPLAY), + BIT_MASK_DESCR(IRQS_WAITING), + BIT_MASK_DESCR(IRQS_PENDING), + BIT_MASK_DESCR(IRQS_SUSPENDED), +}; + + +static int irq_debug_show(struct seq_file *m, void *p) +{ + struct irq_desc *desc = m->private; + struct irq_data *data; + + raw_spin_lock_irq(&desc->lock); + data = irq_desc_get_irq_data(desc); + seq_printf(m, "handler: %pf\n", desc->handle_irq); + seq_printf(m, "status: 0x%08x\n", desc->status_use_accessors); + irq_debug_show_bits(m, 0, desc->status_use_accessors, irqdesc_states, + ARRAY_SIZE(irqdesc_states)); + seq_printf(m, "istate: 0x%08x\n", desc->istate); + irq_debug_show_bits(m, 0, desc->istate, irqdesc_istates, + ARRAY_SIZE(irqdesc_istates)); + seq_printf(m, "ddepth: %u\n", desc->depth); + seq_printf(m, "wdepth: %u\n", desc->wake_depth); + seq_printf(m, "dstate: 0x%08x\n", irqd_get(data)); + irq_debug_show_bits(m, 0, irqd_get(data), irqdata_states, + ARRAY_SIZE(irqdata_states)); + seq_printf(m, "node: %d\n", irq_data_get_node(data)); + irq_debug_show_masks(m, desc); + irq_debug_show_data(m, data, 0); + raw_spin_unlock_irq(&desc->lock); + return 0; +} + +static int irq_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, irq_debug_show, inode->i_private); +} + +static const struct file_operations dfs_irq_ops = { + .open = irq_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +void irq_add_debugfs_entry(unsigned int irq, struct irq_desc *desc) +{ + char name [10]; + + if (!irq_dir || !desc || desc->debugfs_file) + return; + + sprintf(name, "%d", irq); + desc->debugfs_file = debugfs_create_file(name, 0444, irq_dir, desc, + &dfs_irq_ops); +} + +static int __init irq_debugfs_init(void) +{ + struct dentry *root_dir; + int irq; + + root_dir = debugfs_create_dir("irq", NULL); + if (!root_dir) + return -ENOMEM; + + irq_domain_debugfs_init(root_dir); + + irq_dir = debugfs_create_dir("irqs", root_dir); + + irq_lock_sparse(); + for_each_active_irq(irq) + irq_add_debugfs_entry(irq, irq_to_desc(irq)); + irq_unlock_sparse(); + + return 0; +} +__initcall(irq_debugfs_init); diff --git a/kernel/irq/devres.c b/kernel/irq/devres.c index 74d90a754268..194c506d9d20 100644 --- a/kernel/irq/devres.c +++ b/kernel/irq/devres.c @@ -2,6 +2,9 @@ #include <linux/interrupt.h> #include <linux/device.h> #include <linux/gfp.h> +#include <linux/irq.h> + +#include "internals.h" /* * Device resource management aware IRQ request/free implementation. @@ -33,7 +36,7 @@ static int devm_irq_match(struct device *dev, void *res, void *data) * @thread_fn: function to be called in a threaded interrupt context. NULL * for devices which handle everything in @handler * @irqflags: Interrupt type flags - * @devname: An ascii name for the claiming device + * @devname: An ascii name for the claiming device, dev_name(dev) if NULL * @dev_id: A cookie passed back to the handler function * * Except for the extra @dev argument, this function takes the @@ -57,6 +60,9 @@ int devm_request_threaded_irq(struct device *dev, unsigned int irq, if (!dr) return -ENOMEM; + if (!devname) + devname = dev_name(dev); + rc = request_threaded_irq(irq, handler, thread_fn, irqflags, devname, dev_id); if (rc) { @@ -80,7 +86,7 @@ EXPORT_SYMBOL(devm_request_threaded_irq); * @thread_fn: function to be called in a threaded interrupt context. NULL * for devices which handle everything in @handler * @irqflags: Interrupt type flags - * @devname: An ascii name for the claiming device + * @devname: An ascii name for the claiming device, dev_name(dev) if NULL * @dev_id: A cookie passed back to the handler function * * Except for the extra @dev argument, this function takes the @@ -103,6 +109,9 @@ int devm_request_any_context_irq(struct device *dev, unsigned int irq, if (!dr) return -ENOMEM; + if (!devname) + devname = dev_name(dev); + rc = request_any_context_irq(irq, handler, irqflags, devname, dev_id); if (rc < 0) { devres_free(dr); @@ -137,3 +146,141 @@ void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id) free_irq(irq, dev_id); } EXPORT_SYMBOL(devm_free_irq); + +struct irq_desc_devres { + unsigned int from; + unsigned int cnt; +}; + +static void devm_irq_desc_release(struct device *dev, void *res) +{ + struct irq_desc_devres *this = res; + + irq_free_descs(this->from, this->cnt); +} + +/** + * __devm_irq_alloc_descs - Allocate and initialize a range of irq descriptors + * for a managed device + * @dev: Device to allocate the descriptors for + * @irq: Allocate for specific irq number if irq >= 0 + * @from: Start the search from this irq number + * @cnt: Number of consecutive irqs to allocate + * @node: Preferred node on which the irq descriptor should be allocated + * @owner: Owning module (can be NULL) + * @affinity: Optional pointer to an affinity mask array of size @cnt + * which hints where the irq descriptors should be allocated + * and which default affinities to use + * + * Returns the first irq number or error code. + * + * Note: Use the provided wrappers (devm_irq_alloc_desc*) for simplicity. + */ +int __devm_irq_alloc_descs(struct device *dev, int irq, unsigned int from, + unsigned int cnt, int node, struct module *owner, + const struct cpumask *affinity) +{ + struct irq_desc_devres *dr; + int base; + + dr = devres_alloc(devm_irq_desc_release, sizeof(*dr), GFP_KERNEL); + if (!dr) + return -ENOMEM; + + base = __irq_alloc_descs(irq, from, cnt, node, owner, affinity); + if (base < 0) { + devres_free(dr); + return base; + } + + dr->from = base; + dr->cnt = cnt; + devres_add(dev, dr); + + return base; +} +EXPORT_SYMBOL_GPL(__devm_irq_alloc_descs); + +#ifdef CONFIG_GENERIC_IRQ_CHIP +/** + * devm_irq_alloc_generic_chip - Allocate and initialize a generic chip + * for a managed device + * @dev: Device to allocate the generic chip for + * @name: Name of the irq chip + * @num_ct: Number of irq_chip_type instances associated with this + * @irq_base: Interrupt base nr for this chip + * @reg_base: Register base address (virtual) + * @handler: Default flow handler associated with this chip + * + * Returns an initialized irq_chip_generic structure. The chip defaults + * to the primary (index 0) irq_chip_type and @handler + */ +struct irq_chip_generic * +devm_irq_alloc_generic_chip(struct device *dev, const char *name, int num_ct, + unsigned int irq_base, void __iomem *reg_base, + irq_flow_handler_t handler) +{ + struct irq_chip_generic *gc; + unsigned long sz = sizeof(*gc) + num_ct * sizeof(struct irq_chip_type); + + gc = devm_kzalloc(dev, sz, GFP_KERNEL); + if (gc) + irq_init_generic_chip(gc, name, num_ct, + irq_base, reg_base, handler); + + return gc; +} +EXPORT_SYMBOL_GPL(devm_irq_alloc_generic_chip); + +struct irq_generic_chip_devres { + struct irq_chip_generic *gc; + u32 msk; + unsigned int clr; + unsigned int set; +}; + +static void devm_irq_remove_generic_chip(struct device *dev, void *res) +{ + struct irq_generic_chip_devres *this = res; + + irq_remove_generic_chip(this->gc, this->msk, this->clr, this->set); +} + +/** + * devm_irq_setup_generic_chip - Setup a range of interrupts with a generic + * chip for a managed device + * + * @dev: Device to setup the generic chip for + * @gc: Generic irq chip holding all data + * @msk: Bitmask holding the irqs to initialize relative to gc->irq_base + * @flags: Flags for initialization + * @clr: IRQ_* bits to clear + * @set: IRQ_* bits to set + * + * Set up max. 32 interrupts starting from gc->irq_base. Note, this + * initializes all interrupts to the primary irq_chip_type and its + * associated handler. + */ +int devm_irq_setup_generic_chip(struct device *dev, struct irq_chip_generic *gc, + u32 msk, enum irq_gc_flags flags, + unsigned int clr, unsigned int set) +{ + struct irq_generic_chip_devres *dr; + + dr = devres_alloc(devm_irq_remove_generic_chip, + sizeof(*dr), GFP_KERNEL); + if (!dr) + return -ENOMEM; + + irq_setup_generic_chip(gc, msk, flags, clr, set); + + dr->gc = gc; + dr->msk = msk; + dr->clr = clr; + dr->set = set; + devres_add(dev, dr); + + return 0; +} +EXPORT_SYMBOL_GPL(devm_irq_setup_generic_chip); +#endif /* CONFIG_GENERIC_IRQ_CHIP */ diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c index ee32870079c9..f7086b78ad6e 100644 --- a/kernel/irq/generic-chip.c +++ b/kernel/irq/generic-chip.c @@ -201,10 +201,9 @@ static void irq_writel_be(u32 val, void __iomem *addr) iowrite32be(val, addr); } -static void -irq_init_generic_chip(struct irq_chip_generic *gc, const char *name, - int num_ct, unsigned int irq_base, - void __iomem *reg_base, irq_flow_handler_t handler) +void irq_init_generic_chip(struct irq_chip_generic *gc, const char *name, + int num_ct, unsigned int irq_base, + void __iomem *reg_base, irq_flow_handler_t handler) { raw_spin_lock_init(&gc->lock); gc->num_ct = num_ct; diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index d3f24905852c..79f987b942b8 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -6,7 +6,7 @@ * * This file contains the core interrupt handling code. * - * Detailed information is available in Documentation/DocBook/genericirq + * Detailed information is available in Documentation/core-api/genericirq.rst * */ @@ -138,6 +138,8 @@ irqreturn_t __handle_irq_event_percpu(struct irq_desc *desc, unsigned int *flags unsigned int irq = desc->irq_data.irq; struct irqaction *action; + record_irq_time(desc); + for_each_action_of_desc(desc, action) { irqreturn_t res; diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index bc226e783bd2..a2c48058354c 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -8,6 +8,7 @@ #include <linux/irqdesc.h> #include <linux/kernel_stat.h> #include <linux/pm_runtime.h> +#include <linux/sched/clock.h> #ifdef CONFIG_SPARSE_IRQ # define IRQ_BITMAP_BITS (NR_IRQS + 8196) @@ -57,6 +58,7 @@ enum { IRQS_WAITING = 0x00000080, IRQS_PENDING = 0x00000200, IRQS_SUSPENDED = 0x00000800, + IRQS_TIMINGS = 0x00001000, }; #include "debug.h" @@ -66,7 +68,14 @@ extern int __irq_set_trigger(struct irq_desc *desc, unsigned long flags); extern void __disable_irq(struct irq_desc *desc); extern void __enable_irq(struct irq_desc *desc); -extern int irq_startup(struct irq_desc *desc, bool resend); +#define IRQ_RESEND true +#define IRQ_NORESEND false + +#define IRQ_START_FORCE true +#define IRQ_START_COND false + +extern int irq_startup(struct irq_desc *desc, bool resend, bool force); + extern void irq_shutdown(struct irq_desc *desc); extern void irq_enable(struct irq_desc *desc); extern void irq_disable(struct irq_desc *desc); @@ -109,13 +118,19 @@ static inline void unregister_handler_proc(unsigned int irq, extern bool irq_can_set_affinity_usr(unsigned int irq); -extern int irq_select_affinity_usr(unsigned int irq, struct cpumask *mask); +extern int irq_select_affinity_usr(unsigned int irq); extern void irq_set_thread_affinity(struct irq_desc *desc); extern int irq_do_set_affinity(struct irq_data *data, const struct cpumask *dest, bool force); +#ifdef CONFIG_SMP +extern int irq_setup_affinity(struct irq_desc *desc); +#else +static inline int irq_setup_affinity(struct irq_desc *desc) { return 0; } +#endif + /* Inline functions for support of irq chips on slow busses */ static inline void chip_bus_lock(struct irq_desc *desc) { @@ -169,6 +184,11 @@ irq_put_desc_unlock(struct irq_desc *desc, unsigned long flags) #define __irqd_to_state(d) ACCESS_PRIVATE((d)->common, state_use_accessors) +static inline unsigned int irqd_get(struct irq_data *d) +{ + return __irqd_to_state(d); +} + /* * Manipulation functions for irq_data.state */ @@ -182,6 +202,16 @@ static inline void irqd_clr_move_pending(struct irq_data *d) __irqd_to_state(d) &= ~IRQD_SETAFFINITY_PENDING; } +static inline void irqd_set_managed_shutdown(struct irq_data *d) +{ + __irqd_to_state(d) |= IRQD_MANAGED_SHUTDOWN; +} + +static inline void irqd_clr_managed_shutdown(struct irq_data *d) +{ + __irqd_to_state(d) &= ~IRQD_MANAGED_SHUTDOWN; +} + static inline void irqd_clear(struct irq_data *d, unsigned int mask) { __irqd_to_state(d) &= ~mask; @@ -197,6 +227,16 @@ static inline bool irqd_has_set(struct irq_data *d, unsigned int mask) return __irqd_to_state(d) & mask; } +static inline void irq_state_set_disabled(struct irq_desc *desc) +{ + irqd_set(&desc->irq_data, IRQD_IRQ_DISABLED); +} + +static inline void irq_state_set_masked(struct irq_desc *desc) +{ + irqd_set(&desc->irq_data, IRQD_IRQ_MASKED); +} + #undef __irqd_to_state static inline void kstat_incr_irqs_this_cpu(struct irq_desc *desc) @@ -226,3 +266,196 @@ irq_pm_install_action(struct irq_desc *desc, struct irqaction *action) { } static inline void irq_pm_remove_action(struct irq_desc *desc, struct irqaction *action) { } #endif + +#ifdef CONFIG_IRQ_TIMINGS + +#define IRQ_TIMINGS_SHIFT 5 +#define IRQ_TIMINGS_SIZE (1 << IRQ_TIMINGS_SHIFT) +#define IRQ_TIMINGS_MASK (IRQ_TIMINGS_SIZE - 1) + +/** + * struct irq_timings - irq timings storing structure + * @values: a circular buffer of u64 encoded <timestamp,irq> values + * @count: the number of elements in the array + */ +struct irq_timings { + u64 values[IRQ_TIMINGS_SIZE]; + int count; +}; + +DECLARE_PER_CPU(struct irq_timings, irq_timings); + +extern void irq_timings_free(int irq); +extern int irq_timings_alloc(int irq); + +static inline void irq_remove_timings(struct irq_desc *desc) +{ + desc->istate &= ~IRQS_TIMINGS; + + irq_timings_free(irq_desc_get_irq(desc)); +} + +static inline void irq_setup_timings(struct irq_desc *desc, struct irqaction *act) +{ + int irq = irq_desc_get_irq(desc); + int ret; + + /* + * We don't need the measurement because the idle code already + * knows the next expiry event. + */ + if (act->flags & __IRQF_TIMER) + return; + + /* + * In case the timing allocation fails, we just want to warn, + * not fail, so letting the system boot anyway. + */ + ret = irq_timings_alloc(irq); + if (ret) { + pr_warn("Failed to allocate irq timing stats for irq%d (%d)", + irq, ret); + return; + } + + desc->istate |= IRQS_TIMINGS; +} + +extern void irq_timings_enable(void); +extern void irq_timings_disable(void); + +DECLARE_STATIC_KEY_FALSE(irq_timing_enabled); + +/* + * The interrupt number and the timestamp are encoded into a single + * u64 variable to optimize the size. + * 48 bit time stamp and 16 bit IRQ number is way sufficient. + * Who cares an IRQ after 78 hours of idle time? + */ +static inline u64 irq_timing_encode(u64 timestamp, int irq) +{ + return (timestamp << 16) | irq; +} + +static inline int irq_timing_decode(u64 value, u64 *timestamp) +{ + *timestamp = value >> 16; + return value & U16_MAX; +} + +/* + * The function record_irq_time is only called in one place in the + * interrupts handler. We want this function always inline so the code + * inside is embedded in the function and the static key branching + * code can act at the higher level. Without the explicit + * __always_inline we can end up with a function call and a small + * overhead in the hotpath for nothing. + */ +static __always_inline void record_irq_time(struct irq_desc *desc) +{ + if (!static_branch_likely(&irq_timing_enabled)) + return; + + if (desc->istate & IRQS_TIMINGS) { + struct irq_timings *timings = this_cpu_ptr(&irq_timings); + + timings->values[timings->count & IRQ_TIMINGS_MASK] = + irq_timing_encode(local_clock(), + irq_desc_get_irq(desc)); + + timings->count++; + } +} +#else +static inline void irq_remove_timings(struct irq_desc *desc) {} +static inline void irq_setup_timings(struct irq_desc *desc, + struct irqaction *act) {}; +static inline void record_irq_time(struct irq_desc *desc) {} +#endif /* CONFIG_IRQ_TIMINGS */ + + +#ifdef CONFIG_GENERIC_IRQ_CHIP +void irq_init_generic_chip(struct irq_chip_generic *gc, const char *name, + int num_ct, unsigned int irq_base, + void __iomem *reg_base, irq_flow_handler_t handler); +#else +static inline void +irq_init_generic_chip(struct irq_chip_generic *gc, const char *name, + int num_ct, unsigned int irq_base, + void __iomem *reg_base, irq_flow_handler_t handler) { } +#endif /* CONFIG_GENERIC_IRQ_CHIP */ + +#ifdef CONFIG_GENERIC_PENDING_IRQ +static inline bool irq_can_move_pcntxt(struct irq_data *data) +{ + return irqd_can_move_in_process_context(data); +} +static inline bool irq_move_pending(struct irq_data *data) +{ + return irqd_is_setaffinity_pending(data); +} +static inline void +irq_copy_pending(struct irq_desc *desc, const struct cpumask *mask) +{ + cpumask_copy(desc->pending_mask, mask); +} +static inline void +irq_get_pending(struct cpumask *mask, struct irq_desc *desc) +{ + cpumask_copy(mask, desc->pending_mask); +} +static inline struct cpumask *irq_desc_get_pending_mask(struct irq_desc *desc) +{ + return desc->pending_mask; +} +bool irq_fixup_move_pending(struct irq_desc *desc, bool force_clear); +#else /* CONFIG_GENERIC_PENDING_IRQ */ +static inline bool irq_can_move_pcntxt(struct irq_data *data) +{ + return true; +} +static inline bool irq_move_pending(struct irq_data *data) +{ + return false; +} +static inline void +irq_copy_pending(struct irq_desc *desc, const struct cpumask *mask) +{ +} +static inline void +irq_get_pending(struct cpumask *mask, struct irq_desc *desc) +{ +} +static inline struct cpumask *irq_desc_get_pending_mask(struct irq_desc *desc) +{ + return NULL; +} +static inline bool irq_fixup_move_pending(struct irq_desc *desc, bool fclear) +{ + return false; +} +#endif /* !CONFIG_GENERIC_PENDING_IRQ */ + +#ifdef CONFIG_GENERIC_IRQ_DEBUGFS +#include <linux/debugfs.h> + +void irq_add_debugfs_entry(unsigned int irq, struct irq_desc *desc); +static inline void irq_remove_debugfs_entry(struct irq_desc *desc) +{ + debugfs_remove(desc->debugfs_file); +} +# ifdef CONFIG_IRQ_DOMAIN +void irq_domain_debugfs_init(struct dentry *root); +# else +static inline void irq_domain_debugfs_init(struct dentry *root) +{ +} +# endif +#else /* CONFIG_GENERIC_IRQ_DEBUGFS */ +static inline void irq_add_debugfs_entry(unsigned int irq, struct irq_desc *d) +{ +} +static inline void irq_remove_debugfs_entry(struct irq_desc *d) +{ +} +#endif /* CONFIG_GENERIC_IRQ_DEBUGFS */ diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 00bb0aeea1d0..73be2b3909bd 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -4,7 +4,7 @@ * * This file contains the interrupt descriptor management code * - * Detailed information is available in Documentation/DocBook/genericirq + * Detailed information is available in Documentation/core-api/genericirq.rst * */ #include <linux/irq.h> @@ -54,14 +54,25 @@ static void __init init_irq_default_affinity(void) #endif #ifdef CONFIG_SMP -static int alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) +static int alloc_masks(struct irq_desc *desc, int node) { if (!zalloc_cpumask_var_node(&desc->irq_common_data.affinity, - gfp, node)) + GFP_KERNEL, node)) return -ENOMEM; +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + if (!zalloc_cpumask_var_node(&desc->irq_common_data.effective_affinity, + GFP_KERNEL, node)) { + free_cpumask_var(desc->irq_common_data.affinity); + return -ENOMEM; + } +#endif + #ifdef CONFIG_GENERIC_PENDING_IRQ - if (!zalloc_cpumask_var_node(&desc->pending_mask, gfp, node)) { + if (!zalloc_cpumask_var_node(&desc->pending_mask, GFP_KERNEL, node)) { +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + free_cpumask_var(desc->irq_common_data.effective_affinity); +#endif free_cpumask_var(desc->irq_common_data.affinity); return -ENOMEM; } @@ -86,7 +97,7 @@ static void desc_smp_init(struct irq_desc *desc, int node, #else static inline int -alloc_masks(struct irq_desc *desc, gfp_t gfp, int node) { return 0; } +alloc_masks(struct irq_desc *desc, int node) { return 0; } static inline void desc_smp_init(struct irq_desc *desc, int node, const struct cpumask *affinity) { } #endif @@ -105,6 +116,7 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node, desc->irq_data.chip_data = NULL; irq_settings_clr_and_set(desc, ~0, _IRQ_DEFAULT_INIT_FLAGS); irqd_set(&desc->irq_data, IRQD_IRQ_DISABLED); + irqd_set(&desc->irq_data, IRQD_IRQ_MASKED); desc->handle_irq = handle_bad_irq; desc->depth = 1; desc->irq_count = 0; @@ -324,6 +336,9 @@ static void free_masks(struct irq_desc *desc) free_cpumask_var(desc->pending_mask); #endif free_cpumask_var(desc->irq_common_data.affinity); +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + free_cpumask_var(desc->irq_common_data.effective_affinity); +#endif } #else static inline void free_masks(struct irq_desc *desc) { } @@ -344,9 +359,8 @@ static struct irq_desc *alloc_desc(int irq, int node, unsigned int flags, struct module *owner) { struct irq_desc *desc; - gfp_t gfp = GFP_KERNEL; - desc = kzalloc_node(sizeof(*desc), gfp, node); + desc = kzalloc_node(sizeof(*desc), GFP_KERNEL, node); if (!desc) return NULL; /* allocate based on nr_cpu_ids */ @@ -354,11 +368,12 @@ static struct irq_desc *alloc_desc(int irq, int node, unsigned int flags, if (!desc->kstat_irqs) goto err_desc; - if (alloc_masks(desc, gfp, node)) + if (alloc_masks(desc, node)) goto err_kstat; raw_spin_lock_init(&desc->lock); lockdep_set_class(&desc->lock, &irq_desc_lock_class); + mutex_init(&desc->request_mutex); init_rcu_head(&desc->rcu); desc_set_defaults(irq, desc, node, affinity, owner); @@ -394,6 +409,7 @@ static void free_desc(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); + irq_remove_debugfs_entry(desc); unregister_irq_proc(irq, desc); /* @@ -480,7 +496,8 @@ int __init early_irq_init(void) /* Let arch update nr_irqs and return the nr of preallocated irqs */ initcnt = arch_probe_nr_irqs(); - printk(KERN_INFO "NR_IRQS:%d nr_irqs:%d %d\n", NR_IRQS, nr_irqs, initcnt); + printk(KERN_INFO "NR_IRQS: %d, nr_irqs: %d, preallocated irqs: %d\n", + NR_IRQS, nr_irqs, initcnt); if (WARN_ON(nr_irqs > IRQ_BITMAP_BITS)) nr_irqs = IRQ_BITMAP_BITS; @@ -516,14 +533,14 @@ int __init early_irq_init(void) init_irq_default_affinity(); - printk(KERN_INFO "NR_IRQS:%d\n", NR_IRQS); + printk(KERN_INFO "NR_IRQS: %d\n", NR_IRQS); desc = irq_desc; count = ARRAY_SIZE(irq_desc); for (i = 0; i < count; i++) { desc[i].kstat_irqs = alloc_percpu(unsigned int); - alloc_masks(&desc[i], GFP_KERNEL, node); + alloc_masks(&desc[i], node); raw_spin_lock_init(&desc[i].lock); lockdep_set_class(&desc[i].lock, &irq_desc_lock_class); desc_set_defaults(i, &desc[i], node, NULL, NULL); diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index b59e6768c5e9..f1f251479aa6 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -1,5 +1,6 @@ #define pr_fmt(fmt) "irq: " fmt +#include <linux/acpi.h> #include <linux/debugfs.h> #include <linux/hardirq.h> #include <linux/interrupt.h> @@ -26,39 +27,69 @@ static struct irq_domain *irq_default_domain; static void irq_domain_check_hierarchy(struct irq_domain *domain); struct irqchip_fwid { - struct fwnode_handle fwnode; - char *name; + struct fwnode_handle fwnode; + unsigned int type; + char *name; void *data; }; +#ifdef CONFIG_GENERIC_IRQ_DEBUGFS +static void debugfs_add_domain_dir(struct irq_domain *d); +static void debugfs_remove_domain_dir(struct irq_domain *d); +#else +static inline void debugfs_add_domain_dir(struct irq_domain *d) { } +static inline void debugfs_remove_domain_dir(struct irq_domain *d) { } +#endif + /** * irq_domain_alloc_fwnode - Allocate a fwnode_handle suitable for * identifying an irq domain - * @data: optional user-provided data + * @type: Type of irqchip_fwnode. See linux/irqdomain.h + * @name: Optional user provided domain name + * @id: Optional user provided id if name != NULL + * @data: Optional user-provided data * - * Allocate a struct device_node, and return a poiner to the embedded + * Allocate a struct irqchip_fwid, and return a poiner to the embedded * fwnode_handle (or NULL on failure). + * + * Note: The types IRQCHIP_FWNODE_NAMED and IRQCHIP_FWNODE_NAMED_ID are + * solely to transport name information to irqdomain creation code. The + * node is not stored. For other types the pointer is kept in the irq + * domain struct. */ -struct fwnode_handle *irq_domain_alloc_fwnode(void *data) +struct fwnode_handle *__irq_domain_alloc_fwnode(unsigned int type, int id, + const char *name, void *data) { struct irqchip_fwid *fwid; - char *name; + char *n; fwid = kzalloc(sizeof(*fwid), GFP_KERNEL); - name = kasprintf(GFP_KERNEL, "irqchip@%p", data); - if (!fwid || !name) { + switch (type) { + case IRQCHIP_FWNODE_NAMED: + n = kasprintf(GFP_KERNEL, "%s", name); + break; + case IRQCHIP_FWNODE_NAMED_ID: + n = kasprintf(GFP_KERNEL, "%s-%d", name, id); + break; + default: + n = kasprintf(GFP_KERNEL, "irqchip@%p", data); + break; + } + + if (!fwid || !n) { kfree(fwid); - kfree(name); + kfree(n); return NULL; } - fwid->name = name; + fwid->type = type; + fwid->name = n; fwid->data = data; fwid->fwnode.type = FWNODE_IRQCHIP; return &fwid->fwnode; } -EXPORT_SYMBOL_GPL(irq_domain_alloc_fwnode); +EXPORT_SYMBOL_GPL(__irq_domain_alloc_fwnode); /** * irq_domain_free_fwnode - Free a non-OF-backed fwnode_handle @@ -97,26 +128,97 @@ struct irq_domain *__irq_domain_add(struct fwnode_handle *fwnode, int size, void *host_data) { struct device_node *of_node = to_of_node(fwnode); + struct irqchip_fwid *fwid; struct irq_domain *domain; + static atomic_t unknown_domains; + domain = kzalloc_node(sizeof(*domain) + (sizeof(unsigned int) * size), GFP_KERNEL, of_node_to_nid(of_node)); if (WARN_ON(!domain)) return NULL; + if (fwnode && is_fwnode_irqchip(fwnode)) { + fwid = container_of(fwnode, struct irqchip_fwid, fwnode); + + switch (fwid->type) { + case IRQCHIP_FWNODE_NAMED: + case IRQCHIP_FWNODE_NAMED_ID: + domain->name = kstrdup(fwid->name, GFP_KERNEL); + if (!domain->name) { + kfree(domain); + return NULL; + } + domain->flags |= IRQ_DOMAIN_NAME_ALLOCATED; + break; + default: + domain->fwnode = fwnode; + domain->name = fwid->name; + break; + } +#ifdef CONFIG_ACPI + } else if (is_acpi_device_node(fwnode)) { + struct acpi_buffer buf = { + .length = ACPI_ALLOCATE_BUFFER, + }; + acpi_handle handle; + + handle = acpi_device_handle(to_acpi_device_node(fwnode)); + if (acpi_get_name(handle, ACPI_FULL_PATHNAME, &buf) == AE_OK) { + domain->name = buf.pointer; + domain->flags |= IRQ_DOMAIN_NAME_ALLOCATED; + } + + domain->fwnode = fwnode; +#endif + } else if (of_node) { + char *name; + + /* + * DT paths contain '/', which debugfs is legitimately + * unhappy about. Replace them with ':', which does + * the trick and is not as offensive as '\'... + */ + name = kstrdup(of_node_full_name(of_node), GFP_KERNEL); + if (!name) { + kfree(domain); + return NULL; + } + + strreplace(name, '/', ':'); + + domain->name = name; + domain->fwnode = fwnode; + domain->flags |= IRQ_DOMAIN_NAME_ALLOCATED; + } + + if (!domain->name) { + if (fwnode) { + pr_err("Invalid fwnode type (%d) for irqdomain\n", + fwnode->type); + } + domain->name = kasprintf(GFP_KERNEL, "unknown-%d", + atomic_inc_return(&unknown_domains)); + if (!domain->name) { + kfree(domain); + return NULL; + } + domain->flags |= IRQ_DOMAIN_NAME_ALLOCATED; + } + of_node_get(of_node); /* Fill structure */ INIT_RADIX_TREE(&domain->revmap_tree, GFP_KERNEL); domain->ops = ops; domain->host_data = host_data; - domain->fwnode = fwnode; domain->hwirq_max = hwirq_max; domain->revmap_size = size; domain->revmap_direct_max_irq = direct_max; irq_domain_check_hierarchy(domain); mutex_lock(&irq_domain_mutex); + debugfs_add_domain_dir(domain); list_add(&domain->link, &irq_domain_list); mutex_unlock(&irq_domain_mutex); @@ -136,6 +238,7 @@ EXPORT_SYMBOL_GPL(__irq_domain_add); void irq_domain_remove(struct irq_domain *domain) { mutex_lock(&irq_domain_mutex); + debugfs_remove_domain_dir(domain); WARN_ON(!radix_tree_empty(&domain->revmap_tree)); @@ -152,10 +255,43 @@ void irq_domain_remove(struct irq_domain *domain) pr_debug("Removed domain %s\n", domain->name); of_node_put(irq_domain_get_of_node(domain)); + if (domain->flags & IRQ_DOMAIN_NAME_ALLOCATED) + kfree(domain->name); kfree(domain); } EXPORT_SYMBOL_GPL(irq_domain_remove); +void irq_domain_update_bus_token(struct irq_domain *domain, + enum irq_domain_bus_token bus_token) +{ + char *name; + + if (domain->bus_token == bus_token) + return; + + mutex_lock(&irq_domain_mutex); + + domain->bus_token = bus_token; + + name = kasprintf(GFP_KERNEL, "%s-%d", domain->name, bus_token); + if (!name) { + mutex_unlock(&irq_domain_mutex); + return; + } + + debugfs_remove_domain_dir(domain); + + if (domain->flags & IRQ_DOMAIN_NAME_ALLOCATED) + kfree(domain->name); + else + domain->flags |= IRQ_DOMAIN_NAME_ALLOCATED; + + domain->name = name; + debugfs_add_domain_dir(domain); + + mutex_unlock(&irq_domain_mutex); +} + /** * irq_domain_add_simple() - Register an irq_domain and optionally map a range of irqs * @of_node: pointer to interrupt controller's device tree node. @@ -278,6 +414,31 @@ struct irq_domain *irq_find_matching_fwspec(struct irq_fwspec *fwspec, EXPORT_SYMBOL_GPL(irq_find_matching_fwspec); /** + * irq_domain_check_msi_remap - Check whether all MSI irq domains implement + * IRQ remapping + * + * Return: false if any MSI irq domain does not support IRQ remapping, + * true otherwise (including if there is no MSI irq domain) + */ +bool irq_domain_check_msi_remap(void) +{ + struct irq_domain *h; + bool ret = true; + + mutex_lock(&irq_domain_mutex); + list_for_each_entry(h, &irq_domain_list, link) { + if (irq_domain_is_msi(h) && + !irq_domain_hierarchical_is_msi_remap(h)) { + ret = false; + break; + } + } + mutex_unlock(&irq_domain_mutex); + return ret; +} +EXPORT_SYMBOL_GPL(irq_domain_check_msi_remap); + +/** * irq_set_default_host() - Set a "default" irq domain * @domain: default domain pointer * @@ -319,6 +480,7 @@ void irq_domain_disassociate(struct irq_domain *domain, unsigned int irq) irq_data->domain = NULL; irq_data->hwirq = 0; + domain->mapcount--; /* Clear reverse map for this hwirq */ if (hwirq < domain->revmap_size) { @@ -370,6 +532,7 @@ int irq_domain_associate(struct irq_domain *domain, unsigned int virq, domain->name = irq_data->chip->name; } + domain->mapcount++; if (hwirq < domain->revmap_size) { domain->linear_revmap[hwirq] = virq; } else { @@ -721,13 +884,54 @@ unsigned int irq_find_mapping(struct irq_domain *domain, EXPORT_SYMBOL_GPL(irq_find_mapping); #ifdef CONFIG_IRQ_DOMAIN_DEBUG +static void virq_debug_show_one(struct seq_file *m, struct irq_desc *desc) +{ + struct irq_domain *domain; + struct irq_data *data; + + domain = desc->irq_data.domain; + data = &desc->irq_data; + + while (domain) { + unsigned int irq = data->irq; + unsigned long hwirq = data->hwirq; + struct irq_chip *chip; + bool direct; + + if (data == &desc->irq_data) + seq_printf(m, "%5d ", irq); + else + seq_printf(m, "%5d+ ", irq); + seq_printf(m, "0x%05lx ", hwirq); + + chip = irq_data_get_irq_chip(data); + seq_printf(m, "%-15s ", (chip && chip->name) ? chip->name : "none"); + + seq_printf(m, data ? "0x%p " : " %p ", + irq_data_get_irq_chip_data(data)); + + seq_printf(m, " %c ", (desc->action && desc->action->handler) ? '*' : ' '); + direct = (irq == hwirq) && (irq < domain->revmap_direct_max_irq); + seq_printf(m, "%6s%-8s ", + (hwirq < domain->revmap_size) ? "LINEAR" : "RADIX", + direct ? "(DIRECT)" : ""); + seq_printf(m, "%s\n", domain->name); +#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY + domain = domain->parent; + data = data->parent_data; +#else + domain = NULL; +#endif + } +} + static int virq_debug_show(struct seq_file *m, void *private) { unsigned long flags; struct irq_desc *desc; struct irq_domain *domain; struct radix_tree_iter iter; - void *data, **slot; + void **slot; int i; seq_printf(m, " %-16s %-6s %-10s %-10s %s\n", @@ -735,15 +939,26 @@ static int virq_debug_show(struct seq_file *m, void *private) mutex_lock(&irq_domain_mutex); list_for_each_entry(domain, &irq_domain_list, link) { struct device_node *of_node; + const char *name; + int count = 0; + of_node = irq_domain_get_of_node(domain); + if (of_node) + name = of_node_full_name(of_node); + else if (is_fwnode_irqchip(domain->fwnode)) + name = container_of(domain->fwnode, struct irqchip_fwid, + fwnode)->name; + else + name = ""; + radix_tree_for_each_slot(slot, &domain->revmap_tree, &iter, 0) count++; seq_printf(m, "%c%-16s %6u %10u %10u %s\n", domain == irq_default_domain ? '*' : ' ', domain->name, domain->revmap_size + count, domain->revmap_size, domain->revmap_direct_max_irq, - of_node ? of_node_full_name(of_node) : ""); + name); } mutex_unlock(&irq_domain_mutex); @@ -757,30 +972,7 @@ static int virq_debug_show(struct seq_file *m, void *private) continue; raw_spin_lock_irqsave(&desc->lock, flags); - domain = desc->irq_data.domain; - - if (domain) { - struct irq_chip *chip; - int hwirq = desc->irq_data.hwirq; - bool direct; - - seq_printf(m, "%5d ", i); - seq_printf(m, "0x%05x ", hwirq); - - chip = irq_desc_get_chip(desc); - seq_printf(m, "%-15s ", (chip && chip->name) ? chip->name : "none"); - - data = irq_desc_get_chip_data(desc); - seq_printf(m, data ? "0x%p " : " %p ", data); - - seq_printf(m, " %c ", (desc->action && desc->action->handler) ? '*' : ' '); - direct = (i == hwirq) && (i < domain->revmap_direct_max_irq); - seq_printf(m, "%6s%-8s ", - (hwirq < domain->revmap_size) ? "LINEAR" : "RADIX", - direct ? "(DIRECT)" : ""); - seq_printf(m, "%s\n", desc->irq_data.domain->name); - } - + virq_debug_show_one(m, desc); raw_spin_unlock_irqrestore(&desc->lock, flags); } @@ -948,6 +1140,7 @@ static void irq_domain_insert_irq(int virq) struct irq_domain *domain = data->domain; irq_hw_number_t hwirq = data->hwirq; + domain->mapcount++; if (hwirq < domain->revmap_size) { domain->linear_revmap[hwirq] = virq; } else { @@ -977,6 +1170,7 @@ static void irq_domain_remove_irq(int virq) struct irq_domain *domain = data->domain; irq_hw_number_t hwirq = data->hwirq; + domain->mapcount--; if (hwirq < domain->revmap_size) { domain->linear_revmap[hwirq] = 0; } else { @@ -1164,43 +1358,18 @@ void irq_domain_free_irqs_top(struct irq_domain *domain, unsigned int virq, irq_domain_free_irqs_common(domain, virq, nr_irqs); } -static bool irq_domain_is_auto_recursive(struct irq_domain *domain) -{ - return domain->flags & IRQ_DOMAIN_FLAG_AUTO_RECURSIVE; -} - -static void irq_domain_free_irqs_recursive(struct irq_domain *domain, +static void irq_domain_free_irqs_hierarchy(struct irq_domain *domain, unsigned int irq_base, unsigned int nr_irqs) { domain->ops->free(domain, irq_base, nr_irqs); - if (irq_domain_is_auto_recursive(domain)) { - BUG_ON(!domain->parent); - irq_domain_free_irqs_recursive(domain->parent, irq_base, - nr_irqs); - } } -int irq_domain_alloc_irqs_recursive(struct irq_domain *domain, +int irq_domain_alloc_irqs_hierarchy(struct irq_domain *domain, unsigned int irq_base, unsigned int nr_irqs, void *arg) { - int ret = 0; - struct irq_domain *parent = domain->parent; - bool recursive = irq_domain_is_auto_recursive(domain); - - BUG_ON(recursive && !parent); - if (recursive) - ret = irq_domain_alloc_irqs_recursive(parent, irq_base, - nr_irqs, arg); - if (ret < 0) - return ret; - - ret = domain->ops->alloc(domain, irq_base, nr_irqs, arg); - if (ret < 0 && recursive) - irq_domain_free_irqs_recursive(parent, irq_base, nr_irqs); - - return ret; + return domain->ops->alloc(domain, irq_base, nr_irqs, arg); } /** @@ -1261,7 +1430,7 @@ int __irq_domain_alloc_irqs(struct irq_domain *domain, int irq_base, } mutex_lock(&irq_domain_mutex); - ret = irq_domain_alloc_irqs_recursive(domain, virq, nr_irqs, arg); + ret = irq_domain_alloc_irqs_hierarchy(domain, virq, nr_irqs, arg); if (ret < 0) { mutex_unlock(&irq_domain_mutex); goto out_free_irq_data; @@ -1296,7 +1465,7 @@ void irq_domain_free_irqs(unsigned int virq, unsigned int nr_irqs) mutex_lock(&irq_domain_mutex); for (i = 0; i < nr_irqs; i++) irq_domain_remove_irq(virq + i); - irq_domain_free_irqs_recursive(data->domain, virq, nr_irqs); + irq_domain_free_irqs_hierarchy(data->domain, virq, nr_irqs); mutex_unlock(&irq_domain_mutex); irq_domain_free_irq_data(virq, nr_irqs); @@ -1316,15 +1485,11 @@ int irq_domain_alloc_irqs_parent(struct irq_domain *domain, unsigned int irq_base, unsigned int nr_irqs, void *arg) { - /* irq_domain_alloc_irqs_recursive() has called parent's alloc() */ - if (irq_domain_is_auto_recursive(domain)) - return 0; + if (!domain->parent) + return -ENOSYS; - domain = domain->parent; - if (domain) - return irq_domain_alloc_irqs_recursive(domain, irq_base, - nr_irqs, arg); - return -ENOSYS; + return irq_domain_alloc_irqs_hierarchy(domain->parent, irq_base, + nr_irqs, arg); } EXPORT_SYMBOL_GPL(irq_domain_alloc_irqs_parent); @@ -1339,10 +1504,10 @@ EXPORT_SYMBOL_GPL(irq_domain_alloc_irqs_parent); void irq_domain_free_irqs_parent(struct irq_domain *domain, unsigned int irq_base, unsigned int nr_irqs) { - /* irq_domain_free_irqs_recursive() will call parent's free */ - if (!irq_domain_is_auto_recursive(domain) && domain->parent) - irq_domain_free_irqs_recursive(domain->parent, irq_base, - nr_irqs); + if (!domain->parent) + return; + + irq_domain_free_irqs_hierarchy(domain->parent, irq_base, nr_irqs); } EXPORT_SYMBOL_GPL(irq_domain_free_irqs_parent); @@ -1408,6 +1573,20 @@ static void irq_domain_check_hierarchy(struct irq_domain *domain) if (domain->ops->alloc) domain->flags |= IRQ_DOMAIN_FLAG_HIERARCHY; } + +/** + * irq_domain_hierarchical_is_msi_remap - Check if the domain or any + * parent has MSI remapping support + * @domain: domain pointer + */ +bool irq_domain_hierarchical_is_msi_remap(struct irq_domain *domain) +{ + for (; domain; domain = domain->parent) { + if (irq_domain_is_msi_remap(domain)) + return true; + } + return false; +} #else /* CONFIG_IRQ_DOMAIN_HIERARCHY */ /** * irq_domain_get_irq_data - Get irq_data associated with @virq and @domain @@ -1448,3 +1627,77 @@ static void irq_domain_check_hierarchy(struct irq_domain *domain) { } #endif /* CONFIG_IRQ_DOMAIN_HIERARCHY */ + +#ifdef CONFIG_GENERIC_IRQ_DEBUGFS +static struct dentry *domain_dir; + +static void +irq_domain_debug_show_one(struct seq_file *m, struct irq_domain *d, int ind) +{ + seq_printf(m, "%*sname: %s\n", ind, "", d->name); + seq_printf(m, "%*ssize: %u\n", ind + 1, "", + d->revmap_size + d->revmap_direct_max_irq); + seq_printf(m, "%*smapped: %u\n", ind + 1, "", d->mapcount); + seq_printf(m, "%*sflags: 0x%08x\n", ind +1 , "", d->flags); +#ifdef CONFIG_IRQ_DOMAIN_HIERARCHY + if (!d->parent) + return; + seq_printf(m, "%*sparent: %s\n", ind + 1, "", d->parent->name); + irq_domain_debug_show_one(m, d->parent, ind + 4); +#endif +} + +static int irq_domain_debug_show(struct seq_file *m, void *p) +{ + struct irq_domain *d = m->private; + + /* Default domain? Might be NULL */ + if (!d) { + if (!irq_default_domain) + return 0; + d = irq_default_domain; + } + irq_domain_debug_show_one(m, d, 0); + return 0; +} + +static int irq_domain_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, irq_domain_debug_show, inode->i_private); +} + +static const struct file_operations dfs_domain_ops = { + .open = irq_domain_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static void debugfs_add_domain_dir(struct irq_domain *d) +{ + if (!d->name || !domain_dir || d->debugfs_file) + return; + d->debugfs_file = debugfs_create_file(d->name, 0444, domain_dir, d, + &dfs_domain_ops); +} + +static void debugfs_remove_domain_dir(struct irq_domain *d) +{ + debugfs_remove(d->debugfs_file); +} + +void __init irq_domain_debugfs_init(struct dentry *root) +{ + struct irq_domain *d; + + domain_dir = debugfs_create_dir("domains", root); + if (!domain_dir) + return; + + debugfs_create_file("default", 0444, domain_dir, NULL, &dfs_domain_ops); + mutex_lock(&irq_domain_mutex); + list_for_each_entry(d, &irq_domain_list, link) + debugfs_add_domain_dir(d); + mutex_unlock(&irq_domain_mutex); +} +#endif diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 6b669593e7eb..1d1a5b945ab4 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -17,6 +17,8 @@ #include <linux/slab.h> #include <linux/sched.h> #include <linux/sched/rt.h> +#include <linux/sched/task.h> +#include <uapi/linux/sched/types.h> #include <linux/task_work.h> #include "internals.h" @@ -166,34 +168,6 @@ void irq_set_thread_affinity(struct irq_desc *desc) set_bit(IRQTF_AFFINITY, &action->thread_flags); } -#ifdef CONFIG_GENERIC_PENDING_IRQ -static inline bool irq_can_move_pcntxt(struct irq_data *data) -{ - return irqd_can_move_in_process_context(data); -} -static inline bool irq_move_pending(struct irq_data *data) -{ - return irqd_is_setaffinity_pending(data); -} -static inline void -irq_copy_pending(struct irq_desc *desc, const struct cpumask *mask) -{ - cpumask_copy(desc->pending_mask, mask); -} -static inline void -irq_get_pending(struct cpumask *mask, struct irq_desc *desc) -{ - cpumask_copy(mask, desc->pending_mask); -} -#else -static inline bool irq_can_move_pcntxt(struct irq_data *data) { return true; } -static inline bool irq_move_pending(struct irq_data *data) { return false; } -static inline void -irq_copy_pending(struct irq_desc *desc, const struct cpumask *mask) { } -static inline void -irq_get_pending(struct cpumask *mask, struct irq_desc *desc) { } -#endif - int irq_do_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force) { @@ -343,17 +317,20 @@ EXPORT_SYMBOL_GPL(irq_set_affinity_notifier); /* * Generic version of the affinity autoselector. */ -static int setup_affinity(struct irq_desc *desc, struct cpumask *mask) +int irq_setup_affinity(struct irq_desc *desc) { struct cpumask *set = irq_default_affinity; - int node = irq_desc_get_node(desc); + int ret, node = irq_desc_get_node(desc); + static DEFINE_RAW_SPINLOCK(mask_lock); + static struct cpumask mask; /* Excludes PER_CPU and NO_BALANCE interrupts */ if (!__irq_can_set_affinity(desc)) return 0; + raw_spin_lock(&mask_lock); /* - * Preserve the managed affinity setting and an userspace affinity + * Preserve the managed affinity setting and a userspace affinity * setup, but make sure that one of the targets is online. */ if (irqd_affinity_is_managed(&desc->irq_data) || @@ -365,46 +342,40 @@ static int setup_affinity(struct irq_desc *desc, struct cpumask *mask) irqd_clear(&desc->irq_data, IRQD_AFFINITY_SET); } - cpumask_and(mask, cpu_online_mask, set); + cpumask_and(&mask, cpu_online_mask, set); if (node != NUMA_NO_NODE) { const struct cpumask *nodemask = cpumask_of_node(node); /* make sure at least one of the cpus in nodemask is online */ - if (cpumask_intersects(mask, nodemask)) - cpumask_and(mask, mask, nodemask); + if (cpumask_intersects(&mask, nodemask)) + cpumask_and(&mask, &mask, nodemask); } - irq_do_set_affinity(&desc->irq_data, mask, false); - return 0; + ret = irq_do_set_affinity(&desc->irq_data, &mask, false); + raw_spin_unlock(&mask_lock); + return ret; } #else /* Wrapper for ALPHA specific affinity selector magic */ -static inline int setup_affinity(struct irq_desc *d, struct cpumask *mask) +int irq_setup_affinity(struct irq_desc *desc) { - return irq_select_affinity(irq_desc_get_irq(d)); + return irq_select_affinity(irq_desc_get_irq(desc)); } #endif /* - * Called when affinity is set via /proc/irq + * Called when a bogus affinity is set via /proc/irq */ -int irq_select_affinity_usr(unsigned int irq, struct cpumask *mask) +int irq_select_affinity_usr(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); unsigned long flags; int ret; raw_spin_lock_irqsave(&desc->lock, flags); - ret = setup_affinity(desc, mask); + ret = irq_setup_affinity(desc); raw_spin_unlock_irqrestore(&desc->lock, flags); return ret; } - -#else -static inline int -setup_affinity(struct irq_desc *desc, struct cpumask *mask) -{ - return 0; -} #endif /** @@ -531,9 +502,15 @@ void __enable_irq(struct irq_desc *desc) goto err_out; /* Prevent probing on this irq: */ irq_settings_set_noprobe(desc); - irq_enable(desc); - check_irq_resend(desc); - /* fall-through */ + /* + * Call irq_startup() not irq_enable() here because the + * interrupt might be marked NOAUTOEN. So irq_startup() + * needs to be invoked when it gets enabled the first + * time. If it was already started up, then irq_startup() + * will invoke irq_enable() under the hood. + */ + irq_startup(desc, IRQ_RESEND, IRQ_START_COND); + break; } default: desc->depth--; @@ -850,7 +827,7 @@ irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action) * This code is triggered unconditionally. Check the affinity * mask pointer. For CPU_MASK_OFFSTACK=n this is optimized out. */ - if (desc->irq_common_data.affinity) + if (cpumask_available(desc->irq_common_data.affinity)) cpumask_copy(mask, desc->irq_common_data.affinity); else valid = false; @@ -1113,6 +1090,16 @@ setup_irq_thread(struct irqaction *new, unsigned int irq, bool secondary) /* * Internal function to register an irqaction - typically used to * allocate special interrupts that are part of the architecture. + * + * Locking rules: + * + * desc->request_mutex Provides serialization against a concurrent free_irq() + * chip_bus_lock Provides serialization for slow bus operations + * desc->lock Provides serialization against hard interrupts + * + * chip_bus_lock and desc->lock are sufficient for all other management and + * interrupt related functions. desc->request_mutex solely serializes + * request/free_irq(). */ static int __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) @@ -1120,7 +1107,6 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) struct irqaction *old, **old_ptr; unsigned long flags, thread_mask = 0; int ret, nested, shared = 0; - cpumask_var_t mask; if (!desc) return -EINVAL; @@ -1179,11 +1165,6 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) } } - if (!alloc_cpumask_var(&mask, GFP_KERNEL)) { - ret = -ENOMEM; - goto out_thread; - } - /* * Drivers are often written to work w/o knowledge about the * underlying irq chip implementation, so a request for a @@ -1197,7 +1178,34 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) new->flags &= ~IRQF_ONESHOT; /* + * Protects against a concurrent __free_irq() call which might wait + * for synchronize_irq() to complete without holding the optional + * chip bus lock and desc->lock. + */ + mutex_lock(&desc->request_mutex); + + /* + * Acquire bus lock as the irq_request_resources() callback below + * might rely on the serialization or the magic power management + * functions which are abusing the irq_bus_lock() callback, + */ + chip_bus_lock(desc); + + /* First installed action requests resources. */ + if (!desc->action) { + ret = irq_request_resources(desc); + if (ret) { + pr_err("Failed to request resources for %s (irq %d) on irqchip %s\n", + new->name, irq, desc->irq_data.chip->name); + goto out_bus_unlock; + } + } + + /* * The following block of code has to be executed atomically + * protected against a concurrent interrupt and any of the other + * management calls which are not serialized via + * desc->request_mutex or the optional bus lock. */ raw_spin_lock_irqsave(&desc->lock, flags); old_ptr = &desc->action; @@ -1210,8 +1218,10 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) * set the trigger type must match. Also all must * agree on ONESHOT. */ + unsigned int oldtype = irqd_get_trigger_type(&desc->irq_data); + if (!((old->flags & new->flags) & IRQF_SHARED) || - ((old->flags ^ new->flags) & IRQF_TRIGGER_MASK) || + (oldtype != (new->flags & IRQF_TRIGGER_MASK)) || ((old->flags ^ new->flags) & IRQF_ONESHOT)) goto mismatch; @@ -1246,7 +1256,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) */ if (thread_mask == ~0UL) { ret = -EBUSY; - goto out_mask; + goto out_unlock; } /* * The thread_mask for the action is or'ed to @@ -1290,17 +1300,10 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) pr_err("Threaded irq requested with handler=NULL and !ONESHOT for irq %d\n", irq); ret = -EINVAL; - goto out_mask; + goto out_unlock; } if (!shared) { - ret = irq_request_resources(desc); - if (ret) { - pr_err("Failed to request resources for %s (irq %d) on irqchip %s\n", - new->name, irq, desc->irq_data.chip->name); - goto out_mask; - } - init_waitqueue_head(&desc->wait_for_threads); /* Setup the type (level, edge polarity) if configured: */ @@ -1309,7 +1312,7 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) new->flags & IRQF_TRIGGER_MASK); if (ret) - goto out_mask; + goto out_unlock; } desc->istate &= ~(IRQS_AUTODETECT | IRQS_SPURIOUS_DISABLED | \ @@ -1324,20 +1327,25 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (new->flags & IRQF_ONESHOT) desc->istate |= IRQS_ONESHOT; - if (irq_settings_can_autoenable(desc)) - irq_startup(desc, true); - else - /* Undo nested disables: */ - desc->depth = 1; - /* Exclude IRQ from balancing if requested */ if (new->flags & IRQF_NOBALANCING) { irq_settings_set_no_balancing(desc); irqd_set(&desc->irq_data, IRQD_NO_BALANCING); } - /* Set default affinity mask once everything is setup */ - setup_affinity(desc, mask); + if (irq_settings_can_autoenable(desc)) { + irq_startup(desc, IRQ_RESEND, IRQ_START_COND); + } else { + /* + * Shared interrupts do not go well with disabling + * auto enable. The sharing interrupt might request + * it while it's still disabled and then wait for + * interrupts forever. + */ + WARN_ON_ONCE(new->flags & IRQF_SHARED); + /* Undo nested disables: */ + desc->depth = 1; + } } else if (new->flags & IRQF_TRIGGER_MASK) { unsigned int nmsk = new->flags & IRQF_TRIGGER_MASK; @@ -1367,6 +1375,10 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) } raw_spin_unlock_irqrestore(&desc->lock, flags); + chip_bus_sync_unlock(desc); + mutex_unlock(&desc->request_mutex); + + irq_setup_timings(desc, new); /* * Strictly no need to wake it up, but hung_task complains @@ -1378,10 +1390,9 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) wake_up_process(new->secondary->thread); register_irq_proc(irq, desc); + irq_add_debugfs_entry(irq, desc); new->dir = NULL; register_handler_proc(irq, new); - free_cpumask_var(mask); - return 0; mismatch: @@ -1394,9 +1405,14 @@ mismatch: } ret = -EBUSY; -out_mask: +out_unlock: raw_spin_unlock_irqrestore(&desc->lock, flags); - free_cpumask_var(mask); + + if (!desc->action) + irq_release_resources(desc); +out_bus_unlock: + chip_bus_sync_unlock(desc); + mutex_unlock(&desc->request_mutex); out_thread: if (new->thread) { @@ -1437,9 +1453,7 @@ int setup_irq(unsigned int irq, struct irqaction *act) if (retval < 0) return retval; - chip_bus_lock(desc); retval = __setup_irq(irq, desc, act); - chip_bus_sync_unlock(desc); if (retval) irq_chip_pm_put(&desc->irq_data); @@ -1463,6 +1477,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) if (!desc) return NULL; + mutex_lock(&desc->request_mutex); chip_bus_lock(desc); raw_spin_lock_irqsave(&desc->lock, flags); @@ -1478,6 +1493,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) WARN(1, "Trying to free already-free IRQ %d\n", irq); raw_spin_unlock_irqrestore(&desc->lock, flags); chip_bus_sync_unlock(desc); + mutex_unlock(&desc->request_mutex); return NULL; } @@ -1495,7 +1511,6 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) if (!desc->action) { irq_settings_clr_disable_unlazy(desc); irq_shutdown(desc); - irq_release_resources(desc); } #ifdef CONFIG_SMP @@ -1505,6 +1520,20 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) #endif raw_spin_unlock_irqrestore(&desc->lock, flags); + /* + * Drop bus_lock here so the changes which were done in the chip + * callbacks above are synced out to the irq chips which hang + * behind a slow bus (I2C, SPI) before calling synchronize_irq(). + * + * Aside of that the bus_lock can also be taken from the threaded + * handler in irq_finalize_oneshot() which results in a deadlock + * because synchronize_irq() would wait forever for the thread to + * complete, which is blocked on the bus lock. + * + * The still held desc->request_mutex() protects against a + * concurrent request_irq() of this irq so the release of resources + * and timing data is properly serialized. + */ chip_bus_sync_unlock(desc); unregister_handler_proc(irq, action); @@ -1537,6 +1566,20 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) } } + /* Last action releases resources */ + if (!desc->action) { + /* + * Reaquire bus lock as irq_release_resources() might + * require it to deallocate resources over the slow bus. + */ + chip_bus_lock(desc); + irq_release_resources(desc); + chip_bus_sync_unlock(desc); + irq_remove_timings(desc); + } + + mutex_unlock(&desc->request_mutex); + irq_chip_pm_put(&desc->irq_data); module_put(desc->owner); kfree(action->secondary); @@ -1555,7 +1598,7 @@ void remove_irq(unsigned int irq, struct irqaction *act) struct irq_desc *desc = irq_to_desc(irq); if (desc && !WARN_ON(irq_settings_is_per_cpu_devid(desc))) - __free_irq(irq, act->dev_id); + __free_irq(irq, act->dev_id); } EXPORT_SYMBOL_GPL(remove_irq); @@ -1572,20 +1615,27 @@ EXPORT_SYMBOL_GPL(remove_irq); * have completed. * * This function must not be called from interrupt context. + * + * Returns the devname argument passed to request_irq. */ -void free_irq(unsigned int irq, void *dev_id) +const void *free_irq(unsigned int irq, void *dev_id) { struct irq_desc *desc = irq_to_desc(irq); + struct irqaction *action; + const char *devname; if (!desc || WARN_ON(irq_settings_is_per_cpu_devid(desc))) - return; + return NULL; #ifdef CONFIG_SMP if (WARN_ON(desc->affinity_notify)) desc->affinity_notify = NULL; #endif - kfree(__free_irq(irq, dev_id)); + action = __free_irq(irq, dev_id); + devname = action->name; + kfree(action); + return devname; } EXPORT_SYMBOL(free_irq); @@ -1686,9 +1736,7 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, return retval; } - chip_bus_lock(desc); retval = __setup_irq(irq, desc, action); - chip_bus_sync_unlock(desc); if (retval) { irq_chip_pm_put(&desc->irq_data); @@ -1936,9 +1984,7 @@ int setup_percpu_irq(unsigned int irq, struct irqaction *act) if (retval < 0) return retval; - chip_bus_lock(desc); retval = __setup_irq(irq, desc, act); - chip_bus_sync_unlock(desc); if (retval) irq_chip_pm_put(&desc->irq_data); @@ -1947,9 +1993,10 @@ int setup_percpu_irq(unsigned int irq, struct irqaction *act) } /** - * request_percpu_irq - allocate a percpu interrupt line + * __request_percpu_irq - allocate a percpu interrupt line * @irq: Interrupt line to allocate * @handler: Function to be called when the IRQ occurs. + * @flags: Interrupt type flags (IRQF_TIMER only) * @devname: An ascii name for the claiming device * @dev_id: A percpu cookie passed back to the handler function * @@ -1962,8 +2009,9 @@ int setup_percpu_irq(unsigned int irq, struct irqaction *act) * the handler gets called with the interrupted CPU's instance of * that variable. */ -int request_percpu_irq(unsigned int irq, irq_handler_t handler, - const char *devname, void __percpu *dev_id) +int __request_percpu_irq(unsigned int irq, irq_handler_t handler, + unsigned long flags, const char *devname, + void __percpu *dev_id) { struct irqaction *action; struct irq_desc *desc; @@ -1977,12 +2025,15 @@ int request_percpu_irq(unsigned int irq, irq_handler_t handler, !irq_settings_is_per_cpu_devid(desc)) return -EINVAL; + if (flags && flags != IRQF_TIMER) + return -EINVAL; + action = kzalloc(sizeof(struct irqaction), GFP_KERNEL); if (!action) return -ENOMEM; action->handler = handler; - action->flags = IRQF_PERCPU | IRQF_NO_SUSPEND; + action->flags = flags | IRQF_PERCPU | IRQF_NO_SUSPEND; action->name = devname; action->percpu_dev_id = dev_id; @@ -1992,9 +2043,7 @@ int request_percpu_irq(unsigned int irq, irq_handler_t handler, return retval; } - chip_bus_lock(desc); retval = __setup_irq(irq, desc, action); - chip_bus_sync_unlock(desc); if (retval) { irq_chip_pm_put(&desc->irq_data); @@ -2003,7 +2052,7 @@ int request_percpu_irq(unsigned int irq, irq_handler_t handler, return retval; } -EXPORT_SYMBOL_GPL(request_percpu_irq); +EXPORT_SYMBOL_GPL(__request_percpu_irq); /** * irq_get_irqchip_state - returns the irqchip state of a interrupt. diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index 37ddb7bda651..6ca054a3f91d 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -4,6 +4,36 @@ #include "internals.h" +/** + * irq_fixup_move_pending - Cleanup irq move pending from a dying CPU + * @desc: Interrupt descpriptor to clean up + * @force_clear: If set clear the move pending bit unconditionally. + * If not set, clear it only when the dying CPU is the + * last one in the pending mask. + * + * Returns true if the pending bit was set and the pending mask contains an + * online CPU other than the dying CPU. + */ +bool irq_fixup_move_pending(struct irq_desc *desc, bool force_clear) +{ + struct irq_data *data = irq_desc_get_irq_data(desc); + + if (!irqd_is_setaffinity_pending(data)) + return false; + + /* + * The outgoing CPU might be the last online target in a pending + * interrupt move. If that's the case clear the pending move bit. + */ + if (cpumask_any_and(desc->pending_mask, cpu_online_mask) >= nr_cpu_ids) { + irqd_clr_move_pending(data); + return false; + } + if (force_clear) + irqd_clr_move_pending(data); + return true; +} + void irq_move_masked_irq(struct irq_data *idata) { struct irq_desc *desc = irq_data_to_desc(idata); diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c index ee230063f033..48eadf416c24 100644 --- a/kernel/irq/msi.c +++ b/kernel/irq/msi.c @@ -265,13 +265,20 @@ struct irq_domain *msi_create_irq_domain(struct fwnode_handle *fwnode, struct msi_domain_info *info, struct irq_domain *parent) { + struct irq_domain *domain; + if (info->flags & MSI_FLAG_USE_DEF_DOM_OPS) msi_domain_update_dom_ops(info); if (info->flags & MSI_FLAG_USE_DEF_CHIP_OPS) msi_domain_update_chip_ops(info); - return irq_domain_create_hierarchy(parent, 0, 0, fwnode, - &msi_domain_ops, info); + domain = irq_domain_create_hierarchy(parent, IRQ_DOMAIN_FLAG_MSI, 0, + fwnode, &msi_domain_ops, info); + + if (domain && !domain->name && info->chip) + domain->name = info->chip->name; + + return domain; } int msi_domain_prepare_irqs(struct irq_domain *domain, struct device *dev, @@ -308,7 +315,7 @@ int msi_domain_populate_irqs(struct irq_domain *domain, struct device *dev, ops->set_desc(arg, desc); /* Assumes the domain mutex is held! */ - ret = irq_domain_alloc_irqs_recursive(domain, virq, 1, arg); + ret = irq_domain_alloc_irqs_hierarchy(domain, virq, 1, arg); if (ret) break; diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index cea1de0161f1..6bd9b58429cc 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -149,6 +149,8 @@ static void resume_irq(struct irq_desc *desc) /* Pretend that it got disabled ! */ desc->depth++; + irq_state_set_disabled(desc); + irq_state_set_masked(desc); resume: desc->istate &= ~IRQS_SUSPENDED; __enable_irq(desc); diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index feaa813b84a9..7f9642a1e267 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -37,19 +37,47 @@ static struct proc_dir_entry *root_irq_dir; #ifdef CONFIG_SMP -static int show_irq_affinity(int type, struct seq_file *m, void *v) +enum { + AFFINITY, + AFFINITY_LIST, + EFFECTIVE, + EFFECTIVE_LIST, +}; + +static int show_irq_affinity(int type, struct seq_file *m) { struct irq_desc *desc = irq_to_desc((long)m->private); - const struct cpumask *mask = desc->irq_common_data.affinity; + const struct cpumask *mask; + switch (type) { + case AFFINITY: + case AFFINITY_LIST: + mask = desc->irq_common_data.affinity; #ifdef CONFIG_GENERIC_PENDING_IRQ - if (irqd_is_setaffinity_pending(&desc->irq_data)) - mask = desc->pending_mask; + if (irqd_is_setaffinity_pending(&desc->irq_data)) + mask = desc->pending_mask; #endif - if (type) + break; + case EFFECTIVE: + case EFFECTIVE_LIST: +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + mask = desc->irq_common_data.effective_affinity; + break; +#else + return -EINVAL; +#endif + }; + + switch (type) { + case AFFINITY_LIST: + case EFFECTIVE_LIST: seq_printf(m, "%*pbl\n", cpumask_pr_args(mask)); - else + break; + case AFFINITY: + case EFFECTIVE: seq_printf(m, "%*pb\n", cpumask_pr_args(mask)); + break; + } return 0; } @@ -80,12 +108,12 @@ static int irq_affinity_hint_proc_show(struct seq_file *m, void *v) int no_irq_affinity; static int irq_affinity_proc_show(struct seq_file *m, void *v) { - return show_irq_affinity(0, m, v); + return show_irq_affinity(AFFINITY, m); } static int irq_affinity_list_proc_show(struct seq_file *m, void *v) { - return show_irq_affinity(1, m, v); + return show_irq_affinity(AFFINITY_LIST, m); } @@ -120,9 +148,11 @@ static ssize_t write_irq_affinity(int type, struct file *file, * one online CPU still has to be targeted. */ if (!cpumask_intersects(new_value, cpu_online_mask)) { - /* Special case for empty set - allow the architecture - code to set default SMP affinity. */ - err = irq_select_affinity_usr(irq, new_value) ? -EINVAL : count; + /* + * Special case for empty set - allow the architecture code + * to set default SMP affinity. + */ + err = irq_select_affinity_usr(irq) ? -EINVAL : count; } else { irq_set_affinity(irq, new_value); err = count; @@ -183,6 +213,44 @@ static const struct file_operations irq_affinity_list_proc_fops = { .write = irq_affinity_list_proc_write, }; +#ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK +static int irq_effective_aff_proc_show(struct seq_file *m, void *v) +{ + return show_irq_affinity(EFFECTIVE, m); +} + +static int irq_effective_aff_list_proc_show(struct seq_file *m, void *v) +{ + return show_irq_affinity(EFFECTIVE_LIST, m); +} + +static int irq_effective_aff_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, irq_effective_aff_proc_show, PDE_DATA(inode)); +} + +static int irq_effective_aff_list_proc_open(struct inode *inode, + struct file *file) +{ + return single_open(file, irq_effective_aff_list_proc_show, + PDE_DATA(inode)); +} + +static const struct file_operations irq_effective_aff_proc_fops = { + .open = irq_effective_aff_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static const struct file_operations irq_effective_aff_list_proc_fops = { + .open = irq_effective_aff_list_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + static int default_affinity_show(struct seq_file *m, void *v) { seq_printf(m, "%*pb\n", cpumask_pr_args(irq_default_affinity)); @@ -324,6 +392,7 @@ void register_handler_proc(unsigned int irq, struct irqaction *action) void register_irq_proc(unsigned int irq, struct irq_desc *desc) { static DEFINE_MUTEX(register_lock); + void __maybe_unused *irqp = (void *)(unsigned long) irq; char name [MAX_NAMELEN]; if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip)) @@ -349,20 +418,25 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) #ifdef CONFIG_SMP /* create /proc/irq/<irq>/smp_affinity */ proc_create_data("smp_affinity", 0644, desc->dir, - &irq_affinity_proc_fops, (void *)(long)irq); + &irq_affinity_proc_fops, irqp); /* create /proc/irq/<irq>/affinity_hint */ proc_create_data("affinity_hint", 0444, desc->dir, - &irq_affinity_hint_proc_fops, (void *)(long)irq); + &irq_affinity_hint_proc_fops, irqp); /* create /proc/irq/<irq>/smp_affinity_list */ proc_create_data("smp_affinity_list", 0644, desc->dir, - &irq_affinity_list_proc_fops, (void *)(long)irq); + &irq_affinity_list_proc_fops, irqp); proc_create_data("node", 0444, desc->dir, - &irq_node_proc_fops, (void *)(long)irq); + &irq_node_proc_fops, irqp); +# ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + proc_create_data("effective_affinity", 0444, desc->dir, + &irq_effective_aff_proc_fops, irqp); + proc_create_data("effective_affinity_list", 0444, desc->dir, + &irq_effective_aff_list_proc_fops, irqp); +# endif #endif - proc_create_data("spurious", 0444, desc->dir, &irq_spurious_proc_fops, (void *)(long)irq); @@ -381,6 +455,10 @@ void unregister_irq_proc(unsigned int irq, struct irq_desc *desc) remove_proc_entry("affinity_hint", desc->dir); remove_proc_entry("smp_affinity_list", desc->dir); remove_proc_entry("node", desc->dir); +# ifdef CONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK + remove_proc_entry("effective_affinity", desc->dir); + remove_proc_entry("effective_affinity_list", desc->dir); +# endif #endif remove_proc_entry("spurious", desc->dir); @@ -487,6 +565,8 @@ int show_interrupts(struct seq_file *p, void *v) } if (desc->irq_data.domain) seq_printf(p, " %*d", prec, (int) desc->irq_data.hwirq); + else + seq_printf(p, " %*s", prec, ""); #ifdef CONFIG_GENERIC_IRQ_SHOW_LEVEL seq_printf(p, " %-8s", irqd_is_level_type(&desc->irq_data) ? "Level" : "Edge"); #endif diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 5707f97a3e6a..061ba7eed4ed 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -175,7 +175,9 @@ out: static inline int bad_action_ret(irqreturn_t action_ret) { - if (likely(action_ret <= (IRQ_HANDLED | IRQ_WAKE_THREAD))) + unsigned int r = action_ret; + + if (likely(r <= (IRQ_HANDLED | IRQ_WAKE_THREAD))) return 0; return 1; } diff --git a/kernel/irq/timings.c b/kernel/irq/timings.c new file mode 100644 index 000000000000..c8c1d073fbf1 --- /dev/null +++ b/kernel/irq/timings.c @@ -0,0 +1,369 @@ +/* + * linux/kernel/irq/timings.c + * + * Copyright (C) 2016, Linaro Ltd - Daniel Lezcano <[email protected]> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include <linux/kernel.h> +#include <linux/percpu.h> +#include <linux/slab.h> +#include <linux/static_key.h> +#include <linux/interrupt.h> +#include <linux/idr.h> +#include <linux/irq.h> +#include <linux/math64.h> + +#include <trace/events/irq.h> + +#include "internals.h" + +DEFINE_STATIC_KEY_FALSE(irq_timing_enabled); + +DEFINE_PER_CPU(struct irq_timings, irq_timings); + +struct irqt_stat { + u64 next_evt; + u64 last_ts; + u64 variance; + u32 avg; + u32 nr_samples; + int anomalies; + int valid; +}; + +static DEFINE_IDR(irqt_stats); + +void irq_timings_enable(void) +{ + static_branch_enable(&irq_timing_enabled); +} + +void irq_timings_disable(void) +{ + static_branch_disable(&irq_timing_enabled); +} + +/** + * irqs_update - update the irq timing statistics with a new timestamp + * + * @irqs: an irqt_stat struct pointer + * @ts: the new timestamp + * + * The statistics are computed online, in other words, the code is + * designed to compute the statistics on a stream of values rather + * than doing multiple passes on the values to compute the average, + * then the variance. The integer division introduces a loss of + * precision but with an acceptable error margin regarding the results + * we would have with the double floating precision: we are dealing + * with nanosec, so big numbers, consequently the mantisse is + * negligeable, especially when converting the time in usec + * afterwards. + * + * The computation happens at idle time. When the CPU is not idle, the + * interrupts' timestamps are stored in the circular buffer, when the + * CPU goes idle and this routine is called, all the buffer's values + * are injected in the statistical model continuying to extend the + * statistics from the previous busy-idle cycle. + * + * The observations showed a device will trigger a burst of periodic + * interrupts followed by one or two peaks of longer time, for + * instance when a SD card device flushes its cache, then the periodic + * intervals occur again. A one second inactivity period resets the + * stats, that gives us the certitude the statistical values won't + * exceed 1x10^9, thus the computation won't overflow. + * + * Basically, the purpose of the algorithm is to watch the periodic + * interrupts and eliminate the peaks. + * + * An interrupt is considered periodically stable if the interval of + * its occurences follow the normal distribution, thus the values + * comply with: + * + * avg - 3 x stddev < value < avg + 3 x stddev + * + * Which can be simplified to: + * + * -3 x stddev < value - avg < 3 x stddev + * + * abs(value - avg) < 3 x stddev + * + * In order to save a costly square root computation, we use the + * variance. For the record, stddev = sqrt(variance). The equation + * above becomes: + * + * abs(value - avg) < 3 x sqrt(variance) + * + * And finally we square it: + * + * (value - avg) ^ 2 < (3 x sqrt(variance)) ^ 2 + * + * (value - avg) x (value - avg) < 9 x variance + * + * Statistically speaking, any values out of this interval is + * considered as an anomaly and is discarded. However, a normal + * distribution appears when the number of samples is 30 (it is the + * rule of thumb in statistics, cf. "30 samples" on Internet). When + * there are three consecutive anomalies, the statistics are resetted. + * + */ +static void irqs_update(struct irqt_stat *irqs, u64 ts) +{ + u64 old_ts = irqs->last_ts; + u64 variance = 0; + u64 interval; + s64 diff; + + /* + * The timestamps are absolute time values, we need to compute + * the timing interval between two interrupts. + */ + irqs->last_ts = ts; + + /* + * The interval type is u64 in order to deal with the same + * type in our computation, that prevent mindfuck issues with + * overflow, sign and division. + */ + interval = ts - old_ts; + + /* + * The interrupt triggered more than one second apart, that + * ends the sequence as predictible for our purpose. In this + * case, assume we have the beginning of a sequence and the + * timestamp is the first value. As it is impossible to + * predict anything at this point, return. + * + * Note the first timestamp of the sequence will always fall + * in this test because the old_ts is zero. That is what we + * want as we need another timestamp to compute an interval. + */ + if (interval >= NSEC_PER_SEC) { + memset(irqs, 0, sizeof(*irqs)); + irqs->last_ts = ts; + return; + } + + /* + * Pre-compute the delta with the average as the result is + * used several times in this function. + */ + diff = interval - irqs->avg; + + /* + * Increment the number of samples. + */ + irqs->nr_samples++; + + /* + * Online variance divided by the number of elements if there + * is more than one sample. Normally the formula is division + * by nr_samples - 1 but we assume the number of element will be + * more than 32 and dividing by 32 instead of 31 is enough + * precise. + */ + if (likely(irqs->nr_samples > 1)) + variance = irqs->variance >> IRQ_TIMINGS_SHIFT; + + /* + * The rule of thumb in statistics for the normal distribution + * is having at least 30 samples in order to have the model to + * apply. Values outside the interval are considered as an + * anomaly. + */ + if ((irqs->nr_samples >= 30) && ((diff * diff) > (9 * variance))) { + /* + * After three consecutive anomalies, we reset the + * stats as it is no longer stable enough. + */ + if (irqs->anomalies++ >= 3) { + memset(irqs, 0, sizeof(*irqs)); + irqs->last_ts = ts; + return; + } + } else { + /* + * The anomalies must be consecutives, so at this + * point, we reset the anomalies counter. + */ + irqs->anomalies = 0; + } + + /* + * The interrupt is considered stable enough to try to predict + * the next event on it. + */ + irqs->valid = 1; + + /* + * Online average algorithm: + * + * new_average = average + ((value - average) / count) + * + * The variance computation depends on the new average + * to be computed here first. + * + */ + irqs->avg = irqs->avg + (diff >> IRQ_TIMINGS_SHIFT); + + /* + * Online variance algorithm: + * + * new_variance = variance + (value - average) x (value - new_average) + * + * Warning: irqs->avg is updated with the line above, hence + * 'interval - irqs->avg' is no longer equal to 'diff' + */ + irqs->variance = irqs->variance + (diff * (interval - irqs->avg)); + + /* + * Update the next event + */ + irqs->next_evt = ts + irqs->avg; +} + +/** + * irq_timings_next_event - Return when the next event is supposed to arrive + * + * During the last busy cycle, the number of interrupts is incremented + * and stored in the irq_timings structure. This information is + * necessary to: + * + * - know if the index in the table wrapped up: + * + * If more than the array size interrupts happened during the + * last busy/idle cycle, the index wrapped up and we have to + * begin with the next element in the array which is the last one + * in the sequence, otherwise it is a the index 0. + * + * - have an indication of the interrupts activity on this CPU + * (eg. irq/sec) + * + * The values are 'consumed' after inserting in the statistical model, + * thus the count is reinitialized. + * + * The array of values **must** be browsed in the time direction, the + * timestamp must increase between an element and the next one. + * + * Returns a nanosec time based estimation of the earliest interrupt, + * U64_MAX otherwise. + */ +u64 irq_timings_next_event(u64 now) +{ + struct irq_timings *irqts = this_cpu_ptr(&irq_timings); + struct irqt_stat *irqs; + struct irqt_stat __percpu *s; + u64 ts, next_evt = U64_MAX; + int i, irq = 0; + + /* + * This function must be called with the local irq disabled in + * order to prevent the timings circular buffer to be updated + * while we are reading it. + */ + WARN_ON_ONCE(!irqs_disabled()); + + /* + * Number of elements in the circular buffer: If it happens it + * was flushed before, then the number of elements could be + * smaller than IRQ_TIMINGS_SIZE, so the count is used, + * otherwise the array size is used as we wrapped. The index + * begins from zero when we did not wrap. That could be done + * in a nicer way with the proper circular array structure + * type but with the cost of extra computation in the + * interrupt handler hot path. We choose efficiency. + * + * Inject measured irq/timestamp to the statistical model + * while decrementing the counter because we consume the data + * from our circular buffer. + */ + for (i = irqts->count & IRQ_TIMINGS_MASK, + irqts->count = min(IRQ_TIMINGS_SIZE, irqts->count); + irqts->count > 0; irqts->count--, i = (i + 1) & IRQ_TIMINGS_MASK) { + + irq = irq_timing_decode(irqts->values[i], &ts); + + s = idr_find(&irqt_stats, irq); + if (s) { + irqs = this_cpu_ptr(s); + irqs_update(irqs, ts); + } + } + + /* + * Look in the list of interrupts' statistics, the earliest + * next event. + */ + idr_for_each_entry(&irqt_stats, s, i) { + + irqs = this_cpu_ptr(s); + + if (!irqs->valid) + continue; + + if (irqs->next_evt <= now) { + irq = i; + next_evt = now; + + /* + * This interrupt mustn't use in the future + * until new events occur and update the + * statistics. + */ + irqs->valid = 0; + break; + } + + if (irqs->next_evt < next_evt) { + irq = i; + next_evt = irqs->next_evt; + } + } + + return next_evt; +} + +void irq_timings_free(int irq) +{ + struct irqt_stat __percpu *s; + + s = idr_find(&irqt_stats, irq); + if (s) { + free_percpu(s); + idr_remove(&irqt_stats, irq); + } +} + +int irq_timings_alloc(int irq) +{ + struct irqt_stat __percpu *s; + int id; + + /* + * Some platforms can have the same private interrupt per cpu, + * so this function may be be called several times with the + * same interrupt number. Just bail out in case the per cpu + * stat structure is already allocated. + */ + s = idr_find(&irqt_stats, irq); + if (s) + return 0; + + s = alloc_percpu(*s); + if (!s) + return -ENOMEM; + + idr_preload(GFP_KERNEL); + id = idr_alloc(&irqt_stats, s, irq, irq + 1, GFP_NOWAIT); + idr_preload_end(); + + if (id < 0) { + free_percpu(s); + return id; + } + + return 0; +} diff --git a/kernel/jump_label.c b/kernel/jump_label.c index a9b8cf500591..d11c506a6ac3 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -15,6 +15,7 @@ #include <linux/static_key.h> #include <linux/jump_label_ratelimit.h> #include <linux/bug.h> +#include <linux/cpu.h> #ifdef HAVE_JUMP_LABEL @@ -124,6 +125,7 @@ void static_key_slow_inc(struct static_key *key) return; } + cpus_read_lock(); jump_label_lock(); if (atomic_read(&key->enabled) == 0) { atomic_set(&key->enabled, -1); @@ -133,12 +135,14 @@ void static_key_slow_inc(struct static_key *key) atomic_inc(&key->enabled); } jump_label_unlock(); + cpus_read_unlock(); } EXPORT_SYMBOL_GPL(static_key_slow_inc); static void __static_key_slow_dec(struct static_key *key, unsigned long rate_limit, struct delayed_work *work) { + cpus_read_lock(); /* * The negative count check is valid even when a negative * key->enabled is in use by static_key_slow_inc(); a @@ -149,6 +153,7 @@ static void __static_key_slow_dec(struct static_key *key, if (!atomic_dec_and_mutex_lock(&key->enabled, &jump_label_mutex)) { WARN(atomic_read(&key->enabled) < 0, "jump label: negative count!\n"); + cpus_read_unlock(); return; } @@ -159,6 +164,7 @@ static void __static_key_slow_dec(struct static_key *key, jump_label_update(key); } jump_label_unlock(); + cpus_read_unlock(); } static void jump_label_update_timeout(struct work_struct *work) @@ -236,12 +242,28 @@ void __weak __init_or_module arch_jump_label_transform_static(struct jump_entry static inline struct jump_entry *static_key_entries(struct static_key *key) { - return (struct jump_entry *)((unsigned long)key->entries & ~JUMP_TYPE_MASK); + WARN_ON_ONCE(key->type & JUMP_TYPE_LINKED); + return (struct jump_entry *)(key->type & ~JUMP_TYPE_MASK); } static inline bool static_key_type(struct static_key *key) { - return (unsigned long)key->entries & JUMP_TYPE_MASK; + return key->type & JUMP_TYPE_TRUE; +} + +static inline bool static_key_linked(struct static_key *key) +{ + return key->type & JUMP_TYPE_LINKED; +} + +static inline void static_key_clear_linked(struct static_key *key) +{ + key->type &= ~JUMP_TYPE_LINKED; +} + +static inline void static_key_set_linked(struct static_key *key) +{ + key->type |= JUMP_TYPE_LINKED; } static inline struct static_key *jump_entry_key(struct jump_entry *entry) @@ -254,6 +276,26 @@ static bool jump_entry_branch(struct jump_entry *entry) return (unsigned long)entry->key & 1UL; } +/*** + * A 'struct static_key' uses a union such that it either points directly + * to a table of 'struct jump_entry' or to a linked list of modules which in + * turn point to 'struct jump_entry' tables. + * + * The two lower bits of the pointer are used to keep track of which pointer + * type is in use and to store the initial branch direction, we use an access + * function which preserves these bits. + */ +static void static_key_set_entries(struct static_key *key, + struct jump_entry *entries) +{ + unsigned long type; + + WARN_ON_ONCE((unsigned long)entries & JUMP_TYPE_MASK); + type = key->type & JUMP_TYPE_MASK; + key->entries = entries; + key->type |= type; +} + static enum jump_label_type jump_label_type(struct jump_entry *entry) { struct static_key *key = jump_entry_key(entry); @@ -298,6 +340,7 @@ void __init jump_label_init(void) if (static_key_initialized) return; + cpus_read_lock(); jump_label_lock(); jump_label_sort_entries(iter_start, iter_stop); @@ -313,16 +356,11 @@ void __init jump_label_init(void) continue; key = iterk; - /* - * Set key->entries to iter, but preserve JUMP_LABEL_TRUE_BRANCH. - */ - *((unsigned long *)&key->entries) += (unsigned long)iter; -#ifdef CONFIG_MODULES - key->next = NULL; -#endif + static_key_set_entries(key, iter); } static_key_initialized = true; jump_label_unlock(); + cpus_read_unlock(); } #ifdef CONFIG_MODULES @@ -343,6 +381,29 @@ struct static_key_mod { struct module *mod; }; +static inline struct static_key_mod *static_key_mod(struct static_key *key) +{ + WARN_ON_ONCE(!(key->type & JUMP_TYPE_LINKED)); + return (struct static_key_mod *)(key->type & ~JUMP_TYPE_MASK); +} + +/*** + * key->type and key->next are the same via union. + * This sets key->next and preserves the type bits. + * + * See additional comments above static_key_set_entries(). + */ +static void static_key_set_mod(struct static_key *key, + struct static_key_mod *mod) +{ + unsigned long type; + + WARN_ON_ONCE((unsigned long)mod & JUMP_TYPE_MASK); + type = key->type & JUMP_TYPE_MASK; + key->next = mod; + key->type |= type; +} + static int __jump_label_mod_text_reserved(void *start, void *end) { struct module *mod; @@ -365,11 +426,23 @@ static void __jump_label_mod_update(struct static_key *key) { struct static_key_mod *mod; - for (mod = key->next; mod; mod = mod->next) { - struct module *m = mod->mod; + for (mod = static_key_mod(key); mod; mod = mod->next) { + struct jump_entry *stop; + struct module *m; - __jump_label_update(key, mod->entries, - m->jump_entries + m->num_jump_entries); + /* + * NULL if the static_key is defined in a module + * that does not use it + */ + if (!mod->entries) + continue; + + m = mod->mod; + if (!m) + stop = __stop___jump_table; + else + stop = m->jump_entries + m->num_jump_entries; + __jump_label_update(key, mod->entries, stop); } } @@ -404,7 +477,7 @@ static int jump_label_add_module(struct module *mod) struct jump_entry *iter_stop = iter_start + mod->num_jump_entries; struct jump_entry *iter; struct static_key *key = NULL; - struct static_key_mod *jlm; + struct static_key_mod *jlm, *jlm2; /* if the module doesn't have jump label entries, just return */ if (iter_start == iter_stop) @@ -421,20 +494,32 @@ static int jump_label_add_module(struct module *mod) key = iterk; if (within_module(iter->key, mod)) { - /* - * Set key->entries to iter, but preserve JUMP_LABEL_TRUE_BRANCH. - */ - *((unsigned long *)&key->entries) += (unsigned long)iter; - key->next = NULL; + static_key_set_entries(key, iter); continue; } jlm = kzalloc(sizeof(struct static_key_mod), GFP_KERNEL); if (!jlm) return -ENOMEM; + if (!static_key_linked(key)) { + jlm2 = kzalloc(sizeof(struct static_key_mod), + GFP_KERNEL); + if (!jlm2) { + kfree(jlm); + return -ENOMEM; + } + preempt_disable(); + jlm2->mod = __module_address((unsigned long)key); + preempt_enable(); + jlm2->entries = static_key_entries(key); + jlm2->next = NULL; + static_key_set_mod(key, jlm2); + static_key_set_linked(key); + } jlm->mod = mod; jlm->entries = iter; - jlm->next = key->next; - key->next = jlm; + jlm->next = static_key_mod(key); + static_key_set_mod(key, jlm); + static_key_set_linked(key); /* Only update if we've changed from our initial state */ if (jump_label_type(iter) != jump_label_init_type(iter)) @@ -461,16 +546,34 @@ static void jump_label_del_module(struct module *mod) if (within_module(iter->key, mod)) continue; + /* No memory during module load */ + if (WARN_ON(!static_key_linked(key))) + continue; + prev = &key->next; - jlm = key->next; + jlm = static_key_mod(key); while (jlm && jlm->mod != mod) { prev = &jlm->next; jlm = jlm->next; } - if (jlm) { + /* No memory during module load */ + if (WARN_ON(!jlm)) + continue; + + if (prev == &key->next) + static_key_set_mod(key, jlm->next); + else *prev = jlm->next; + + kfree(jlm); + + jlm = static_key_mod(key); + /* if only one etry is left, fold it back into the static_key */ + if (jlm->next == NULL) { + static_key_set_entries(key, jlm->entries); + static_key_clear_linked(key); kfree(jlm); } } @@ -495,26 +598,28 @@ jump_label_module_notify(struct notifier_block *self, unsigned long val, struct module *mod = data; int ret = 0; + cpus_read_lock(); + jump_label_lock(); + switch (val) { case MODULE_STATE_COMING: - jump_label_lock(); ret = jump_label_add_module(mod); - if (ret) + if (ret) { + WARN(1, "Failed to allocatote memory: jump_label may not work properly.\n"); jump_label_del_module(mod); - jump_label_unlock(); + } break; case MODULE_STATE_GOING: - jump_label_lock(); jump_label_del_module(mod); - jump_label_unlock(); break; case MODULE_STATE_LIVE: - jump_label_lock(); jump_label_invalidate_module_init(mod); - jump_label_unlock(); break; } + jump_label_unlock(); + cpus_read_unlock(); + return notifier_from_errno(ret); } @@ -561,11 +666,14 @@ int jump_label_text_reserved(void *start, void *end) static void jump_label_update(struct static_key *key) { struct jump_entry *stop = __stop___jump_table; - struct jump_entry *entry = static_key_entries(key); + struct jump_entry *entry; #ifdef CONFIG_MODULES struct module *mod; - __jump_label_mod_update(key); + if (static_key_linked(key)) { + __jump_label_mod_update(key); + return; + } preempt_disable(); mod = __module_address((unsigned long)key); @@ -573,6 +681,7 @@ static void jump_label_update(struct static_key *key) stop = mod->jump_entries + mod->num_jump_entries; preempt_enable(); #endif + entry = static_key_entries(key); /* if there are no users, entry can be NULL */ if (entry) __jump_label_update(key, entry, stop); diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index fafd1a3ef0da..127e7cfafa55 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -23,16 +23,11 @@ #include <linux/mm.h> #include <linux/ctype.h> #include <linux/slab.h> +#include <linux/filter.h> #include <linux/compiler.h> #include <asm/sections.h> -#ifdef CONFIG_KALLSYMS_ALL -#define all_var 1 -#else -#define all_var 0 -#endif - /* * These will be re-linked against their real values * during the second link stage. @@ -81,7 +76,7 @@ static inline int is_kernel(unsigned long addr) static int is_ksym_addr(unsigned long addr) { - if (all_var) + if (IS_ENABLED(CONFIG_KALLSYMS_ALL)) return is_kernel(addr); return is_kernel_text(addr) || is_kernel_inittext(addr); @@ -279,7 +274,7 @@ static unsigned long get_symbol_pos(unsigned long addr, if (!symbol_end) { if (is_kernel_inittext(addr)) symbol_end = (unsigned long)_einittext; - else if (all_var) + else if (IS_ENABLED(CONFIG_KALLSYMS_ALL)) symbol_end = (unsigned long)_end; else symbol_end = (unsigned long)_etext; @@ -300,10 +295,11 @@ int kallsyms_lookup_size_offset(unsigned long addr, unsigned long *symbolsize, unsigned long *offset) { char namebuf[KSYM_NAME_LEN]; + if (is_ksym_addr(addr)) return !!get_symbol_pos(addr, symbolsize, offset); - - return !!module_address_lookup(addr, symbolsize, offset, NULL, namebuf); + return !!module_address_lookup(addr, symbolsize, offset, NULL, namebuf) || + !!__bpf_address_lookup(addr, symbolsize, offset, namebuf); } /* @@ -318,6 +314,8 @@ const char *kallsyms_lookup(unsigned long addr, unsigned long *offset, char **modname, char *namebuf) { + const char *ret; + namebuf[KSYM_NAME_LEN - 1] = 0; namebuf[0] = 0; @@ -333,9 +331,13 @@ const char *kallsyms_lookup(unsigned long addr, return namebuf; } - /* See if it's in a module. */ - return module_address_lookup(addr, symbolsize, offset, modname, - namebuf); + /* See if it's in a module or a BPF JITed image. */ + ret = module_address_lookup(addr, symbolsize, offset, + modname, namebuf); + if (!ret) + ret = bpf_address_lookup(addr, symbolsize, + offset, modname, namebuf); + return ret; } int lookup_symbol_name(unsigned long addr, char *symname) @@ -471,6 +473,7 @@ EXPORT_SYMBOL(__print_symbol); /* To avoid using get_symbol_offset for every symbol, we carry prefix along. */ struct kallsym_iter { loff_t pos; + loff_t pos_mod_end; unsigned long value; unsigned int nameoff; /* If iterating in core kernel symbols. */ char type; @@ -481,13 +484,27 @@ struct kallsym_iter { static int get_ksymbol_mod(struct kallsym_iter *iter) { - if (module_get_kallsym(iter->pos - kallsyms_num_syms, &iter->value, - &iter->type, iter->name, iter->module_name, - &iter->exported) < 0) + int ret = module_get_kallsym(iter->pos - kallsyms_num_syms, + &iter->value, &iter->type, + iter->name, iter->module_name, + &iter->exported); + if (ret < 0) { + iter->pos_mod_end = iter->pos; return 0; + } + return 1; } +static int get_ksymbol_bpf(struct kallsym_iter *iter) +{ + iter->module_name[0] = '\0'; + iter->exported = 0; + return bpf_get_kallsym(iter->pos - iter->pos_mod_end, + &iter->value, &iter->type, + iter->name) < 0 ? 0 : 1; +} + /* Returns space to next name. */ static unsigned long get_ksymbol_core(struct kallsym_iter *iter) { @@ -508,16 +525,30 @@ static void reset_iter(struct kallsym_iter *iter, loff_t new_pos) iter->name[0] = '\0'; iter->nameoff = get_symbol_offset(new_pos); iter->pos = new_pos; + if (new_pos == 0) + iter->pos_mod_end = 0; +} + +static int update_iter_mod(struct kallsym_iter *iter, loff_t pos) +{ + iter->pos = pos; + + if (iter->pos_mod_end > 0 && + iter->pos_mod_end < iter->pos) + return get_ksymbol_bpf(iter); + + if (!get_ksymbol_mod(iter)) + return get_ksymbol_bpf(iter); + + return 1; } /* Returns false if pos at or past end of file. */ static int update_iter(struct kallsym_iter *iter, loff_t pos) { /* Module symbols can be accessed randomly. */ - if (pos >= kallsyms_num_syms) { - iter->pos = pos; - return get_ksymbol_mod(iter); - } + if (pos >= kallsyms_num_syms) + return update_iter_mod(iter, pos); /* If we're not on the desired position, reset to new position. */ if (pos != iter->pos) diff --git a/kernel/kcmp.c b/kernel/kcmp.c index 3a47fa998fe0..ea34ed8bb952 100644 --- a/kernel/kcmp.c +++ b/kernel/kcmp.c @@ -11,6 +11,10 @@ #include <linux/bug.h> #include <linux/err.h> #include <linux/kcmp.h> +#include <linux/capability.h> +#include <linux/list.h> +#include <linux/eventpoll.h> +#include <linux/file.h> #include <asm/unistd.h> @@ -94,6 +98,56 @@ static int kcmp_lock(struct mutex *m1, struct mutex *m2) return err; } +#ifdef CONFIG_EPOLL +static int kcmp_epoll_target(struct task_struct *task1, + struct task_struct *task2, + unsigned long idx1, + struct kcmp_epoll_slot __user *uslot) +{ + struct file *filp, *filp_epoll, *filp_tgt; + struct kcmp_epoll_slot slot; + struct files_struct *files; + + if (copy_from_user(&slot, uslot, sizeof(slot))) + return -EFAULT; + + filp = get_file_raw_ptr(task1, idx1); + if (!filp) + return -EBADF; + + files = get_files_struct(task2); + if (!files) + return -EBADF; + + spin_lock(&files->file_lock); + filp_epoll = fcheck_files(files, slot.efd); + if (filp_epoll) + get_file(filp_epoll); + else + filp_tgt = ERR_PTR(-EBADF); + spin_unlock(&files->file_lock); + put_files_struct(files); + + if (filp_epoll) { + filp_tgt = get_epoll_tfile_raw_ptr(filp_epoll, slot.tfd, slot.toff); + fput(filp_epoll); + } else + + if (IS_ERR(filp_tgt)) + return PTR_ERR(filp_tgt); + + return kcmp_ptr(filp, filp_tgt, KCMP_FILE); +} +#else +static int kcmp_epoll_target(struct task_struct *task1, + struct task_struct *task2, + unsigned long idx1, + struct kcmp_epoll_slot __user *uslot) +{ + return -EOPNOTSUPP; +} +#endif + SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type, unsigned long, idx1, unsigned long, idx2) { @@ -165,6 +219,9 @@ SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type, ret = -EOPNOTSUPP; #endif break; + case KCMP_EPOLL_TFD: + ret = kcmp_epoll_target(task1, task2, idx1, (void *)idx2); + break; default: ret = -EINVAL; break; diff --git a/kernel/kcov.c b/kernel/kcov.c index 85e5546cd791..cd771993f96f 100644 --- a/kernel/kcov.c +++ b/kernel/kcov.c @@ -60,15 +60,8 @@ void notrace __sanitizer_cov_trace_pc(void) /* * We are interested in code coverage as a function of a syscall inputs, * so we ignore code executed in interrupts. - * The checks for whether we are in an interrupt are open-coded, because - * 1. We can't use in_interrupt() here, since it also returns true - * when we are inside local_bh_disable() section. - * 2. We don't want to use (in_irq() | in_serving_softirq() | in_nmi()), - * since that leads to slower generated code (three separate tests, - * one for each of the flags). */ - if (!t || (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_OFFSET - | NMI_MASK))) + if (!t || !in_task()) return; mode = READ_ONCE(t->kcov_mode); if (mode == KCOV_MODE_TRACE) { diff --git a/kernel/kexec.c b/kernel/kexec.c index 980936a90ee6..e62ec4dc6620 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -144,6 +144,14 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments, if (ret) goto out; + /* + * Some architecture(like S390) may touch the crash memory before + * machine_kexec_prepare(), we must copy vmcoreinfo data after it. + */ + ret = kimage_crash_copy_vmcoreinfo(image); + if (ret) + goto out; + for (i = 0; i < nr_segments; i++) { ret = kimage_load_segment(image, &image->segment[i]); if (ret) diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index 5617cc412444..1ae7c41c33c1 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -38,6 +38,7 @@ #include <linux/syscore_ops.h> #include <linux/compiler.h> #include <linux/hugetlb.h> +#include <linux/frame.h> #include <asm/page.h> #include <asm/sections.h> @@ -51,12 +52,6 @@ DEFINE_MUTEX(kexec_mutex); /* Per cpu memory for storing cpu states in case of system crash. */ note_buf_t __percpu *crash_notes; -/* vmcoreinfo stuff */ -static unsigned char vmcoreinfo_data[VMCOREINFO_BYTES]; -u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; -size_t vmcoreinfo_size; -size_t vmcoreinfo_max_size = sizeof(vmcoreinfo_data); - /* Flag to indicate we are going to kexec a new kernel */ bool kexec_in_progress = false; @@ -487,6 +482,40 @@ struct page *kimage_alloc_control_pages(struct kimage *image, return pages; } +int kimage_crash_copy_vmcoreinfo(struct kimage *image) +{ + struct page *vmcoreinfo_page; + void *safecopy; + + if (image->type != KEXEC_TYPE_CRASH) + return 0; + + /* + * For kdump, allocate one vmcoreinfo safe copy from the + * crash memory. as we have arch_kexec_protect_crashkres() + * after kexec syscall, we naturally protect it from write + * (even read) access under kernel direct mapping. But on + * the other hand, we still need to operate it when crash + * happens to generate vmcoreinfo note, hereby we rely on + * vmap for this purpose. + */ + vmcoreinfo_page = kimage_alloc_control_pages(image, 0); + if (!vmcoreinfo_page) { + pr_warn("Could not allocate vmcoreinfo buffer\n"); + return -ENOMEM; + } + safecopy = vmap(&vmcoreinfo_page, 1, VM_MAP, PAGE_KERNEL); + if (!safecopy) { + pr_warn("Could not vmap vmcoreinfo buffer\n"); + return -ENOMEM; + } + + image->vmcoreinfo_data_copy = safecopy; + crash_update_vmcoreinfo_safecopy(safecopy); + + return 0; +} + static int kimage_add_entry(struct kimage *image, kimage_entry_t entry) { if (*image->entry != 0) @@ -574,6 +603,11 @@ void kimage_free(struct kimage *image) if (!image) return; + if (image->vmcoreinfo_data_copy) { + crash_update_vmcoreinfo_safecopy(NULL); + vunmap(image->vmcoreinfo_data_copy); + } + kimage_free_extra_pages(image); for_each_kimage_entry(image, ptr, entry) { if (entry & IND_INDIRECTION) { @@ -880,7 +914,7 @@ int kexec_load_disabled; * only when panic_cpu holds the current CPU number; this is the only CPU * which processes crash_kexec routines. */ -void __crash_kexec(struct pt_regs *regs) +void __noclone __crash_kexec(struct pt_regs *regs) { /* Take the kexec_mutex here to prevent sys_kexec_load * running on one cpu from replacing the crash kernel @@ -902,6 +936,7 @@ void __crash_kexec(struct pt_regs *regs) mutex_unlock(&kexec_mutex); } } +STACK_FRAME_NON_STANDARD(__crash_kexec); void crash_kexec(struct pt_regs *regs) { @@ -916,7 +951,7 @@ void crash_kexec(struct pt_regs *regs) old_cpu = atomic_cmpxchg(&panic_cpu, PANIC_CPU_INVALID, this_cpu); if (old_cpu == PANIC_CPU_INVALID) { /* This is the 1st CPU which comes here, so go ahead. */ - printk_nmi_flush_on_panic(); + printk_safe_flush_on_panic(); __crash_kexec(regs); /* @@ -996,34 +1031,6 @@ unlock: return ret; } -static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, - size_t data_len) -{ - struct elf_note note; - - note.n_namesz = strlen(name) + 1; - note.n_descsz = data_len; - note.n_type = type; - memcpy(buf, ¬e, sizeof(note)); - buf += (sizeof(note) + 3)/4; - memcpy(buf, name, note.n_namesz); - buf += (note.n_namesz + 3)/4; - memcpy(buf, data, note.n_descsz); - buf += (note.n_descsz + 3)/4; - - return buf; -} - -static void final_note(u32 *buf) -{ - struct elf_note note; - - note.n_namesz = 0; - note.n_descsz = 0; - note.n_type = 0; - memcpy(buf, ¬e, sizeof(note)); -} - void crash_save_cpu(struct pt_regs *regs, int cpu) { struct elf_prstatus prstatus; @@ -1085,403 +1092,6 @@ subsys_initcall(crash_notes_memory_init); /* - * parsing the "crashkernel" commandline - * - * this code is intended to be called from architecture specific code - */ - - -/* - * This function parses command lines in the format - * - * crashkernel=ramsize-range:size[,...][@offset] - * - * The function returns 0 on success and -EINVAL on failure. - */ -static int __init parse_crashkernel_mem(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - char *cur = cmdline, *tmp; - - /* for each entry of the comma-separated list */ - do { - unsigned long long start, end = ULLONG_MAX, size; - - /* get the start of the range */ - start = memparse(cur, &tmp); - if (cur == tmp) { - pr_warn("crashkernel: Memory value expected\n"); - return -EINVAL; - } - cur = tmp; - if (*cur != '-') { - pr_warn("crashkernel: '-' expected\n"); - return -EINVAL; - } - cur++; - - /* if no ':' is here, than we read the end */ - if (*cur != ':') { - end = memparse(cur, &tmp); - if (cur == tmp) { - pr_warn("crashkernel: Memory value expected\n"); - return -EINVAL; - } - cur = tmp; - if (end <= start) { - pr_warn("crashkernel: end <= start\n"); - return -EINVAL; - } - } - - if (*cur != ':') { - pr_warn("crashkernel: ':' expected\n"); - return -EINVAL; - } - cur++; - - size = memparse(cur, &tmp); - if (cur == tmp) { - pr_warn("Memory value expected\n"); - return -EINVAL; - } - cur = tmp; - if (size >= system_ram) { - pr_warn("crashkernel: invalid size\n"); - return -EINVAL; - } - - /* match ? */ - if (system_ram >= start && system_ram < end) { - *crash_size = size; - break; - } - } while (*cur++ == ','); - - if (*crash_size > 0) { - while (*cur && *cur != ' ' && *cur != '@') - cur++; - if (*cur == '@') { - cur++; - *crash_base = memparse(cur, &tmp); - if (cur == tmp) { - pr_warn("Memory value expected after '@'\n"); - return -EINVAL; - } - } - } - - return 0; -} - -/* - * That function parses "simple" (old) crashkernel command lines like - * - * crashkernel=size[@offset] - * - * It returns 0 on success and -EINVAL on failure. - */ -static int __init parse_crashkernel_simple(char *cmdline, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - char *cur = cmdline; - - *crash_size = memparse(cmdline, &cur); - if (cmdline == cur) { - pr_warn("crashkernel: memory value expected\n"); - return -EINVAL; - } - - if (*cur == '@') - *crash_base = memparse(cur+1, &cur); - else if (*cur != ' ' && *cur != '\0') { - pr_warn("crashkernel: unrecognized char: %c\n", *cur); - return -EINVAL; - } - - return 0; -} - -#define SUFFIX_HIGH 0 -#define SUFFIX_LOW 1 -#define SUFFIX_NULL 2 -static __initdata char *suffix_tbl[] = { - [SUFFIX_HIGH] = ",high", - [SUFFIX_LOW] = ",low", - [SUFFIX_NULL] = NULL, -}; - -/* - * That function parses "suffix" crashkernel command lines like - * - * crashkernel=size,[high|low] - * - * It returns 0 on success and -EINVAL on failure. - */ -static int __init parse_crashkernel_suffix(char *cmdline, - unsigned long long *crash_size, - const char *suffix) -{ - char *cur = cmdline; - - *crash_size = memparse(cmdline, &cur); - if (cmdline == cur) { - pr_warn("crashkernel: memory value expected\n"); - return -EINVAL; - } - - /* check with suffix */ - if (strncmp(cur, suffix, strlen(suffix))) { - pr_warn("crashkernel: unrecognized char: %c\n", *cur); - return -EINVAL; - } - cur += strlen(suffix); - if (*cur != ' ' && *cur != '\0') { - pr_warn("crashkernel: unrecognized char: %c\n", *cur); - return -EINVAL; - } - - return 0; -} - -static __init char *get_last_crashkernel(char *cmdline, - const char *name, - const char *suffix) -{ - char *p = cmdline, *ck_cmdline = NULL; - - /* find crashkernel and use the last one if there are more */ - p = strstr(p, name); - while (p) { - char *end_p = strchr(p, ' '); - char *q; - - if (!end_p) - end_p = p + strlen(p); - - if (!suffix) { - int i; - - /* skip the one with any known suffix */ - for (i = 0; suffix_tbl[i]; i++) { - q = end_p - strlen(suffix_tbl[i]); - if (!strncmp(q, suffix_tbl[i], - strlen(suffix_tbl[i]))) - goto next; - } - ck_cmdline = p; - } else { - q = end_p - strlen(suffix); - if (!strncmp(q, suffix, strlen(suffix))) - ck_cmdline = p; - } -next: - p = strstr(p+1, name); - } - - if (!ck_cmdline) - return NULL; - - return ck_cmdline; -} - -static int __init __parse_crashkernel(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base, - const char *name, - const char *suffix) -{ - char *first_colon, *first_space; - char *ck_cmdline; - - BUG_ON(!crash_size || !crash_base); - *crash_size = 0; - *crash_base = 0; - - ck_cmdline = get_last_crashkernel(cmdline, name, suffix); - - if (!ck_cmdline) - return -EINVAL; - - ck_cmdline += strlen(name); - - if (suffix) - return parse_crashkernel_suffix(ck_cmdline, crash_size, - suffix); - /* - * if the commandline contains a ':', then that's the extended - * syntax -- if not, it must be the classic syntax - */ - first_colon = strchr(ck_cmdline, ':'); - first_space = strchr(ck_cmdline, ' '); - if (first_colon && (!first_space || first_colon < first_space)) - return parse_crashkernel_mem(ck_cmdline, system_ram, - crash_size, crash_base); - - return parse_crashkernel_simple(ck_cmdline, crash_size, crash_base); -} - -/* - * That function is the entry point for command line parsing and should be - * called from the arch-specific code. - */ -int __init parse_crashkernel(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, - "crashkernel=", NULL); -} - -int __init parse_crashkernel_high(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, - "crashkernel=", suffix_tbl[SUFFIX_HIGH]); -} - -int __init parse_crashkernel_low(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, - "crashkernel=", suffix_tbl[SUFFIX_LOW]); -} - -static void update_vmcoreinfo_note(void) -{ - u32 *buf = vmcoreinfo_note; - - if (!vmcoreinfo_size) - return; - buf = append_elf_note(buf, VMCOREINFO_NOTE_NAME, 0, vmcoreinfo_data, - vmcoreinfo_size); - final_note(buf); -} - -void crash_save_vmcoreinfo(void) -{ - vmcoreinfo_append_str("CRASHTIME=%ld\n", get_seconds()); - update_vmcoreinfo_note(); -} - -void vmcoreinfo_append_str(const char *fmt, ...) -{ - va_list args; - char buf[0x50]; - size_t r; - - va_start(args, fmt); - r = vscnprintf(buf, sizeof(buf), fmt, args); - va_end(args); - - r = min(r, vmcoreinfo_max_size - vmcoreinfo_size); - - memcpy(&vmcoreinfo_data[vmcoreinfo_size], buf, r); - - vmcoreinfo_size += r; -} - -/* - * provide an empty default implementation here -- architecture - * code may override this - */ -void __weak arch_crash_save_vmcoreinfo(void) -{} - -phys_addr_t __weak paddr_vmcoreinfo_note(void) -{ - return __pa((unsigned long)(char *)&vmcoreinfo_note); -} - -static int __init crash_save_vmcoreinfo_init(void) -{ - VMCOREINFO_OSRELEASE(init_uts_ns.name.release); - VMCOREINFO_PAGESIZE(PAGE_SIZE); - - VMCOREINFO_SYMBOL(init_uts_ns); - VMCOREINFO_SYMBOL(node_online_map); -#ifdef CONFIG_MMU - VMCOREINFO_SYMBOL(swapper_pg_dir); -#endif - VMCOREINFO_SYMBOL(_stext); - VMCOREINFO_SYMBOL(vmap_area_list); - -#ifndef CONFIG_NEED_MULTIPLE_NODES - VMCOREINFO_SYMBOL(mem_map); - VMCOREINFO_SYMBOL(contig_page_data); -#endif -#ifdef CONFIG_SPARSEMEM - VMCOREINFO_SYMBOL(mem_section); - VMCOREINFO_LENGTH(mem_section, NR_SECTION_ROOTS); - VMCOREINFO_STRUCT_SIZE(mem_section); - VMCOREINFO_OFFSET(mem_section, section_mem_map); -#endif - VMCOREINFO_STRUCT_SIZE(page); - VMCOREINFO_STRUCT_SIZE(pglist_data); - VMCOREINFO_STRUCT_SIZE(zone); - VMCOREINFO_STRUCT_SIZE(free_area); - VMCOREINFO_STRUCT_SIZE(list_head); - VMCOREINFO_SIZE(nodemask_t); - VMCOREINFO_OFFSET(page, flags); - VMCOREINFO_OFFSET(page, _refcount); - VMCOREINFO_OFFSET(page, mapping); - VMCOREINFO_OFFSET(page, lru); - VMCOREINFO_OFFSET(page, _mapcount); - VMCOREINFO_OFFSET(page, private); - VMCOREINFO_OFFSET(page, compound_dtor); - VMCOREINFO_OFFSET(page, compound_order); - VMCOREINFO_OFFSET(page, compound_head); - VMCOREINFO_OFFSET(pglist_data, node_zones); - VMCOREINFO_OFFSET(pglist_data, nr_zones); -#ifdef CONFIG_FLAT_NODE_MEM_MAP - VMCOREINFO_OFFSET(pglist_data, node_mem_map); -#endif - VMCOREINFO_OFFSET(pglist_data, node_start_pfn); - VMCOREINFO_OFFSET(pglist_data, node_spanned_pages); - VMCOREINFO_OFFSET(pglist_data, node_id); - VMCOREINFO_OFFSET(zone, free_area); - VMCOREINFO_OFFSET(zone, vm_stat); - VMCOREINFO_OFFSET(zone, spanned_pages); - VMCOREINFO_OFFSET(free_area, free_list); - VMCOREINFO_OFFSET(list_head, next); - VMCOREINFO_OFFSET(list_head, prev); - VMCOREINFO_OFFSET(vmap_area, va_start); - VMCOREINFO_OFFSET(vmap_area, list); - VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER); - log_buf_kexec_setup(); - VMCOREINFO_LENGTH(free_area.free_list, MIGRATE_TYPES); - VMCOREINFO_NUMBER(NR_FREE_PAGES); - VMCOREINFO_NUMBER(PG_lru); - VMCOREINFO_NUMBER(PG_private); - VMCOREINFO_NUMBER(PG_swapcache); - VMCOREINFO_NUMBER(PG_slab); -#ifdef CONFIG_MEMORY_FAILURE - VMCOREINFO_NUMBER(PG_hwpoison); -#endif - VMCOREINFO_NUMBER(PG_head_mask); - VMCOREINFO_NUMBER(PAGE_BUDDY_MAPCOUNT_VALUE); -#ifdef CONFIG_HUGETLB_PAGE - VMCOREINFO_NUMBER(HUGETLB_PAGE_DTOR); -#endif - - arch_crash_save_vmcoreinfo(); - update_vmcoreinfo_note(); - - return 0; -} - -subsys_initcall(crash_save_vmcoreinfo_init); - -/* * Move into place and start executing a preloaded standalone * executable. If nothing was preloaded return an error. */ diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c index b56a558e406d..9f48f4412297 100644 --- a/kernel/kexec_file.c +++ b/kernel/kexec_file.c @@ -26,13 +26,6 @@ #include <linux/vmalloc.h> #include "kexec_internal.h" -/* - * Declare these symbols weak so that if architecture provides a purgatory, - * these will be overridden. - */ -char __weak kexec_purgatory[0]; -size_t __weak kexec_purgatory_size = 0; - static int kexec_calculate_store_digests(struct kimage *image); /* Architectures can provide this probe function */ @@ -162,16 +155,10 @@ kimage_file_prepare_segments(struct kimage *image, int kernel_fd, int initrd_fd, } if (cmdline_len) { - image->cmdline_buf = kzalloc(cmdline_len, GFP_KERNEL); - if (!image->cmdline_buf) { - ret = -ENOMEM; - goto out; - } - - ret = copy_from_user(image->cmdline_buf, cmdline_ptr, - cmdline_len); - if (ret) { - ret = -EFAULT; + image->cmdline_buf = memdup_user(cmdline_ptr, cmdline_len); + if (IS_ERR(image->cmdline_buf)) { + ret = PTR_ERR(image->cmdline_buf); + image->cmdline_buf = NULL; goto out; } @@ -304,6 +291,14 @@ SYSCALL_DEFINE5(kexec_file_load, int, kernel_fd, int, initrd_fd, if (ret) goto out; + /* + * Some architecture(like S390) may touch the crash memory before + * machine_kexec_prepare(), we must copy vmcoreinfo data after it. + */ + ret = kimage_crash_copy_vmcoreinfo(image); + if (ret) + goto out; + ret = kexec_calculate_store_digests(image); if (ret) goto out; @@ -614,13 +609,13 @@ static int kexec_calculate_store_digests(struct kimage *image) ret = crypto_shash_final(desc, digest); if (ret) goto out_free_digest; - ret = kexec_purgatory_get_set_symbol(image, "sha_regions", - sha_regions, sha_region_sz, 0); + ret = kexec_purgatory_get_set_symbol(image, "purgatory_sha_regions", + sha_regions, sha_region_sz, 0); if (ret) goto out_free_digest; - ret = kexec_purgatory_get_set_symbol(image, "sha256_digest", - digest, SHA256_DIGEST_SIZE, 0); + ret = kexec_purgatory_get_set_symbol(image, "purgatory_sha256_digest", + digest, SHA256_DIGEST_SIZE, 0); if (ret) goto out_free_digest; } diff --git a/kernel/kexec_internal.h b/kernel/kexec_internal.h index 4cef7e4706b0..50dfcb039a41 100644 --- a/kernel/kexec_internal.h +++ b/kernel/kexec_internal.h @@ -15,12 +15,10 @@ int kimage_is_destination_range(struct kimage *image, extern struct mutex kexec_mutex; #ifdef CONFIG_KEXEC_FILE -struct kexec_sha_region { - unsigned long start; - unsigned long len; -}; - +#include <linux/purgatory.h> void kimage_file_post_load_cleanup(struct kimage *image); +extern char kexec_purgatory[]; +extern size_t kexec_purgatory_size; #else /* CONFIG_KEXEC_FILE */ static inline void kimage_file_post_load_cleanup(struct kimage *image) { } #endif /* CONFIG_KEXEC_FILE */ diff --git a/kernel/kmod.c b/kernel/kmod.c index d45c96073afb..6d016c5d97c8 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -20,6 +20,8 @@ */ #include <linux/module.h> #include <linux/sched.h> +#include <linux/sched/task.h> +#include <linux/binfmts.h> #include <linux/syscalls.h> #include <linux/unistd.h> #include <linux/kmod.h> @@ -43,8 +45,6 @@ #include <trace/events/module.h> -extern int max_threads; - #define CAP_BSET (void *)1 #define CAP_PI (void *)2 @@ -54,6 +54,21 @@ static DEFINE_SPINLOCK(umh_sysctl_lock); static DECLARE_RWSEM(umhelper_sem); #ifdef CONFIG_MODULES +/* + * Assuming: + * + * threads = div64_u64((u64) totalram_pages * (u64) PAGE_SIZE, + * (u64) THREAD_SIZE * 8UL); + * + * If you need less than 50 threads would mean we're dealing with systems + * smaller than 3200 pages. This assuems you are capable of having ~13M memory, + * and this would only be an be an upper limit, after which the OOM killer + * would take effect. Systems like these are very unlikely if modules are + * enabled. + */ +#define MAX_KMOD_CONCURRENT 50 +static atomic_t kmod_concurrent_max = ATOMIC_INIT(MAX_KMOD_CONCURRENT); +static DECLARE_WAIT_QUEUE_HEAD(kmod_wq); /* modprobe_path is set via /proc/sys. @@ -125,11 +140,7 @@ int __request_module(bool wait, const char *fmt, ...) { va_list args; char module_name[MODULE_NAME_LEN]; - unsigned int max_modprobes; int ret; - static atomic_t kmod_concurrent = ATOMIC_INIT(0); -#define MAX_KMOD_CONCURRENT 50 /* Completely arbitrary value - KAO */ - static int kmod_loop_msg; /* * We don't allow synchronous module loading from async. Module @@ -152,40 +163,25 @@ int __request_module(bool wait, const char *fmt, ...) if (ret) return ret; - /* If modprobe needs a service that is in a module, we get a recursive - * loop. Limit the number of running kmod threads to max_threads/2 or - * MAX_KMOD_CONCURRENT, whichever is the smaller. A cleaner method - * would be to run the parents of this process, counting how many times - * kmod was invoked. That would mean accessing the internals of the - * process tables to get the command line, proc_pid_cmdline is static - * and it is not worth changing the proc code just to handle this case. - * KAO. - * - * "trace the ppid" is simple, but will fail if someone's - * parent exits. I think this is as good as it gets. --RR - */ - max_modprobes = min(max_threads/2, MAX_KMOD_CONCURRENT); - atomic_inc(&kmod_concurrent); - if (atomic_read(&kmod_concurrent) > max_modprobes) { - /* We may be blaming an innocent here, but unlikely */ - if (kmod_loop_msg < 5) { - printk(KERN_ERR - "request_module: runaway loop modprobe %s\n", - module_name); - kmod_loop_msg++; - } - atomic_dec(&kmod_concurrent); - return -ENOMEM; + if (atomic_dec_if_positive(&kmod_concurrent_max) < 0) { + pr_warn_ratelimited("request_module: kmod_concurrent_max (%u) close to 0 (max_modprobes: %u), for module %s, throttling...", + atomic_read(&kmod_concurrent_max), + MAX_KMOD_CONCURRENT, module_name); + wait_event_interruptible(kmod_wq, + atomic_dec_if_positive(&kmod_concurrent_max) >= 0); } trace_module_request(module_name, wait, _RET_IP_); ret = call_modprobe(module_name, wait ? UMH_WAIT_PROC : UMH_WAIT_EXEC); - atomic_dec(&kmod_concurrent); + atomic_inc(&kmod_concurrent_max); + wake_up(&kmod_wq); + return ret; } EXPORT_SYMBOL(__request_module); + #endif /* CONFIG_MODULES */ static void call_usermodehelper_freeinfo(struct subprocess_info *info) @@ -516,7 +512,7 @@ static void helper_unlock(void) * Function must be runnable in either a process context or the * context in which call_usermodehelper_exec is called. */ -struct subprocess_info *call_usermodehelper_setup(char *path, char **argv, +struct subprocess_info *call_usermodehelper_setup(const char *path, char **argv, char **envp, gfp_t gfp_mask, int (*init)(struct subprocess_info *info, struct cred *new), void (*cleanup)(struct subprocess_info *info), @@ -528,7 +524,12 @@ struct subprocess_info *call_usermodehelper_setup(char *path, char **argv, goto out; INIT_WORK(&sub_info->work, call_usermodehelper_exec_work); + +#ifdef CONFIG_STATIC_USERMODEHELPER + sub_info->path = CONFIG_STATIC_USERMODEHELPER_PATH; +#else sub_info->path = path; +#endif sub_info->argv = argv; sub_info->envp = envp; @@ -566,6 +567,15 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) retval = -EBUSY; goto out; } + + /* + * If there is no binary for us to call, then just return and get out of + * here. This allows us to set STATIC_USERMODEHELPER_PATH to "" and + * disable all call_usermodehelper() calls. + */ + if (strlen(sub_info->path) == 0) + goto out; + /* * Set the completion pointer only if there is a waiter. * This makes it possible to use umh_complete to free @@ -613,7 +623,7 @@ EXPORT_SYMBOL(call_usermodehelper_exec); * This function is the equivalent to use call_usermodehelper_setup() and * call_usermodehelper_exec(). */ -int call_usermodehelper(char *path, char **argv, char **envp, int wait) +int call_usermodehelper(const char *path, char **argv, char **envp, int wait) { struct subprocess_info *info; gfp_t gfp_mask = (wait == UMH_NO_WAIT) ? GFP_ATOMIC : GFP_KERNEL; diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 43460104f119..a1606a4224e1 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -58,15 +58,6 @@ #define KPROBE_TABLE_SIZE (1 << KPROBE_HASH_BITS) -/* - * Some oddball architectures like 64bit powerpc have function descriptors - * so this must be overridable. - */ -#ifndef kprobe_lookup_name -#define kprobe_lookup_name(name, addr) \ - addr = ((kprobe_opcode_t *)(kallsyms_lookup_name(name))) -#endif - static int kprobes_initialized; static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE]; static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE]; @@ -81,6 +72,12 @@ static struct { raw_spinlock_t lock ____cacheline_aligned_in_smp; } kretprobe_table_locks[KPROBE_TABLE_SIZE]; +kprobe_opcode_t * __weak kprobe_lookup_name(const char *name, + unsigned int __unused) +{ + return ((kprobe_opcode_t *)(kallsyms_lookup_name(name))); +} + static raw_spinlock_t *kretprobe_table_lock_ptr(unsigned long hash) { return &(kretprobe_table_locks[hash].lock); @@ -125,7 +122,7 @@ static void *alloc_insn_page(void) return module_alloc(PAGE_SIZE); } -static void free_insn_page(void *page) +void __weak free_insn_page(void *page) { module_memfree(page); } @@ -149,9 +146,11 @@ kprobe_opcode_t *__get_insn_slot(struct kprobe_insn_cache *c) struct kprobe_insn_page *kip; kprobe_opcode_t *slot = NULL; + /* Since the slot array is not protected by rcu, we need a mutex */ mutex_lock(&c->mutex); retry: - list_for_each_entry(kip, &c->pages, list) { + rcu_read_lock(); + list_for_each_entry_rcu(kip, &c->pages, list) { if (kip->nused < slots_per_page(c)) { int i; for (i = 0; i < slots_per_page(c); i++) { @@ -159,6 +158,7 @@ kprobe_opcode_t *__get_insn_slot(struct kprobe_insn_cache *c) kip->slot_used[i] = SLOT_USED; kip->nused++; slot = kip->insns + (i * c->insn_size); + rcu_read_unlock(); goto out; } } @@ -167,6 +167,7 @@ kprobe_opcode_t *__get_insn_slot(struct kprobe_insn_cache *c) WARN_ON(1); } } + rcu_read_unlock(); /* If there are any garbage slots, collect it and try again. */ if (c->nr_garbage && collect_garbage_slots(c) == 0) @@ -193,7 +194,7 @@ kprobe_opcode_t *__get_insn_slot(struct kprobe_insn_cache *c) kip->nused = 1; kip->ngarbage = 0; kip->cache = c; - list_add(&kip->list, &c->pages); + list_add_rcu(&kip->list, &c->pages); slot = kip->insns; out: mutex_unlock(&c->mutex); @@ -213,7 +214,8 @@ static int collect_one_slot(struct kprobe_insn_page *kip, int idx) * next time somebody inserts a probe. */ if (!list_is_singular(&kip->list)) { - list_del(&kip->list); + list_del_rcu(&kip->list); + synchronize_rcu(); kip->cache->free(kip->insns); kfree(kip); } @@ -235,8 +237,7 @@ static int collect_garbage_slots(struct kprobe_insn_cache *c) continue; kip->ngarbage = 0; /* we will collect all garbages */ for (i = 0; i < slots_per_page(c); i++) { - if (kip->slot_used[i] == SLOT_DIRTY && - collect_one_slot(kip, i)) + if (kip->slot_used[i] == SLOT_DIRTY && collect_one_slot(kip, i)) break; } } @@ -248,29 +249,60 @@ void __free_insn_slot(struct kprobe_insn_cache *c, kprobe_opcode_t *slot, int dirty) { struct kprobe_insn_page *kip; + long idx; mutex_lock(&c->mutex); - list_for_each_entry(kip, &c->pages, list) { - long idx = ((long)slot - (long)kip->insns) / - (c->insn_size * sizeof(kprobe_opcode_t)); - if (idx >= 0 && idx < slots_per_page(c)) { - WARN_ON(kip->slot_used[idx] != SLOT_USED); - if (dirty) { - kip->slot_used[idx] = SLOT_DIRTY; - kip->ngarbage++; - if (++c->nr_garbage > slots_per_page(c)) - collect_garbage_slots(c); - } else - collect_one_slot(kip, idx); + rcu_read_lock(); + list_for_each_entry_rcu(kip, &c->pages, list) { + idx = ((long)slot - (long)kip->insns) / + (c->insn_size * sizeof(kprobe_opcode_t)); + if (idx >= 0 && idx < slots_per_page(c)) goto out; - } } - /* Could not free this slot. */ + /* Could not find this slot. */ WARN_ON(1); + kip = NULL; out: + rcu_read_unlock(); + /* Mark and sweep: this may sleep */ + if (kip) { + /* Check double free */ + WARN_ON(kip->slot_used[idx] != SLOT_USED); + if (dirty) { + kip->slot_used[idx] = SLOT_DIRTY; + kip->ngarbage++; + if (++c->nr_garbage > slots_per_page(c)) + collect_garbage_slots(c); + } else { + collect_one_slot(kip, idx); + } + } mutex_unlock(&c->mutex); } +/* + * Check given address is on the page of kprobe instruction slots. + * This will be used for checking whether the address on a stack + * is on a text area or not. + */ +bool __is_insn_slot_addr(struct kprobe_insn_cache *c, unsigned long addr) +{ + struct kprobe_insn_page *kip; + bool ret = false; + + rcu_read_lock(); + list_for_each_entry_rcu(kip, &c->pages, list) { + if (addr >= (unsigned long)kip->insns && + addr < (unsigned long)kip->insns + PAGE_SIZE) { + ret = true; + break; + } + } + rcu_read_unlock(); + + return ret; +} + #ifdef CONFIG_OPTPROBES /* For optimized_kprobe buffer */ struct kprobe_insn_cache kprobe_optinsn_slots = { @@ -451,11 +483,6 @@ static DECLARE_DELAYED_WORK(optimizing_work, kprobe_optimizer); */ static void do_optimize_kprobes(void) { - /* Optimization never be done when disarmed */ - if (kprobes_all_disarmed || !kprobes_allow_optimization || - list_empty(&optimizing_list)) - return; - /* * The optimization/unoptimization refers online_cpus via * stop_machine() and cpu-hotplug modifies online_cpus. @@ -463,14 +490,19 @@ static void do_optimize_kprobes(void) * This combination can cause a deadlock (cpu-hotplug try to lock * text_mutex but stop_machine can not be done because online_cpus * has been changed) - * To avoid this deadlock, we need to call get_online_cpus() + * To avoid this deadlock, caller must have locked cpu hotplug * for preventing cpu-hotplug outside of text_mutex locking. */ - get_online_cpus(); + lockdep_assert_cpus_held(); + + /* Optimization never be done when disarmed */ + if (kprobes_all_disarmed || !kprobes_allow_optimization || + list_empty(&optimizing_list)) + return; + mutex_lock(&text_mutex); arch_optimize_kprobes(&optimizing_list); mutex_unlock(&text_mutex); - put_online_cpus(); } /* @@ -481,12 +513,13 @@ static void do_unoptimize_kprobes(void) { struct optimized_kprobe *op, *tmp; + /* See comment in do_optimize_kprobes() */ + lockdep_assert_cpus_held(); + /* Unoptimization must be done anytime */ if (list_empty(&unoptimizing_list)) return; - /* Ditto to do_optimize_kprobes */ - get_online_cpus(); mutex_lock(&text_mutex); arch_unoptimize_kprobes(&unoptimizing_list, &freeing_list); /* Loop free_list for disarming */ @@ -505,7 +538,6 @@ static void do_unoptimize_kprobes(void) list_del_init(&op->list); } mutex_unlock(&text_mutex); - put_online_cpus(); } /* Reclaim all kprobes on the free_list */ @@ -530,6 +562,7 @@ static void kick_kprobe_optimizer(void) static void kprobe_optimizer(struct work_struct *work) { mutex_lock(&kprobe_mutex); + cpus_read_lock(); /* Lock modules while optimizing kprobes */ mutex_lock(&module_mutex); @@ -555,6 +588,7 @@ static void kprobe_optimizer(struct work_struct *work) do_free_cleaned_kprobes(); mutex_unlock(&module_mutex); + cpus_read_unlock(); mutex_unlock(&kprobe_mutex); /* Step 5: Kick optimizer again if needed */ @@ -563,7 +597,7 @@ static void kprobe_optimizer(struct work_struct *work) } /* Wait for completing optimization and unoptimization */ -static void wait_for_kprobe_optimizer(void) +void wait_for_kprobe_optimizer(void) { mutex_lock(&kprobe_mutex); @@ -618,9 +652,8 @@ static void optimize_kprobe(struct kprobe *p) /* Short cut to direct unoptimizing */ static void force_unoptimize_kprobe(struct optimized_kprobe *op) { - get_online_cpus(); + lockdep_assert_cpus_held(); arch_unoptimize_kprobe(op); - put_online_cpus(); if (kprobe_disabled(&op->kp)) arch_disarm_kprobe(&op->kp); } @@ -711,13 +744,20 @@ static void kill_optimized_kprobe(struct kprobe *p) arch_remove_optimized_kprobe(op); } +static inline +void __prepare_optimized_kprobe(struct optimized_kprobe *op, struct kprobe *p) +{ + if (!kprobe_ftrace(p)) + arch_prepare_optimized_kprobe(op, p); +} + /* Try to prepare optimized instructions */ static void prepare_optimized_kprobe(struct kprobe *p) { struct optimized_kprobe *op; op = container_of(p, struct optimized_kprobe, kp); - arch_prepare_optimized_kprobe(op, p); + __prepare_optimized_kprobe(op, p); } /* Allocate new optimized_kprobe and try to prepare optimized instructions */ @@ -731,7 +771,7 @@ static struct kprobe *alloc_aggr_kprobe(struct kprobe *p) INIT_LIST_HEAD(&op->list); op->kp.addr = p->addr; - arch_prepare_optimized_kprobe(op, p); + __prepare_optimized_kprobe(op, p); return &op->kp; } @@ -752,6 +792,7 @@ static void try_to_optimize_kprobe(struct kprobe *p) return; /* For preparing optimization, jump_label_text_reserved() is called */ + cpus_read_lock(); jump_label_lock(); mutex_lock(&text_mutex); @@ -773,6 +814,7 @@ static void try_to_optimize_kprobe(struct kprobe *p) out: mutex_unlock(&text_mutex); jump_label_unlock(); + cpus_read_unlock(); } #ifdef CONFIG_SYSCTL @@ -787,6 +829,7 @@ static void optimize_all_kprobes(void) if (kprobes_allow_optimization) goto out; + cpus_read_lock(); kprobes_allow_optimization = true; for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; @@ -794,6 +837,7 @@ static void optimize_all_kprobes(void) if (!kprobe_disabled(p)) optimize_kprobe(p); } + cpus_read_unlock(); printk(KERN_INFO "Kprobes globally optimized\n"); out: mutex_unlock(&kprobe_mutex); @@ -812,6 +856,7 @@ static void unoptimize_all_kprobes(void) return; } + cpus_read_lock(); kprobes_allow_optimization = false; for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; @@ -820,6 +865,7 @@ static void unoptimize_all_kprobes(void) unoptimize_kprobe(p, false); } } + cpus_read_unlock(); mutex_unlock(&kprobe_mutex); /* Wait for unoptimizing completion */ @@ -971,14 +1017,11 @@ static void arm_kprobe(struct kprobe *kp) arm_kprobe_ftrace(kp); return; } - /* - * Here, since __arm_kprobe() doesn't use stop_machine(), - * this doesn't cause deadlock on text_mutex. So, we don't - * need get_online_cpus(). - */ + cpus_read_lock(); mutex_lock(&text_mutex); __arm_kprobe(kp); mutex_unlock(&text_mutex); + cpus_read_unlock(); } /* Disarm a kprobe with text_mutex */ @@ -988,10 +1031,12 @@ static void disarm_kprobe(struct kprobe *kp, bool reopt) disarm_kprobe_ftrace(kp); return; } - /* Ditto */ + + cpus_read_lock(); mutex_lock(&text_mutex); __disarm_kprobe(kp, reopt); mutex_unlock(&text_mutex); + cpus_read_unlock(); } /* @@ -1259,13 +1304,10 @@ static int register_aggr_kprobe(struct kprobe *orig_p, struct kprobe *p) int ret = 0; struct kprobe *ap = orig_p; + cpus_read_lock(); + /* For preparing optimization, jump_label_text_reserved() is called */ jump_label_lock(); - /* - * Get online CPUs to avoid text_mutex deadlock.with stop machine, - * which is invoked by unoptimize_kprobe() in add_new_kprobe() - */ - get_online_cpus(); mutex_lock(&text_mutex); if (!kprobe_aggrprobe(orig_p)) { @@ -1313,8 +1355,8 @@ static int register_aggr_kprobe(struct kprobe *orig_p, struct kprobe *p) out: mutex_unlock(&text_mutex); - put_online_cpus(); jump_label_unlock(); + cpus_read_unlock(); if (ret == 0 && kprobe_disabled(ap) && !kprobe_disabled(p)) { ap->flags &= ~KPROBE_FLAG_DISABLED; @@ -1356,21 +1398,19 @@ bool within_kprobe_blacklist(unsigned long addr) * This returns encoded errors if it fails to look up symbol or invalid * combination of parameters. */ -static kprobe_opcode_t *kprobe_addr(struct kprobe *p) +static kprobe_opcode_t *_kprobe_addr(kprobe_opcode_t *addr, + const char *symbol_name, unsigned int offset) { - kprobe_opcode_t *addr = p->addr; - - if ((p->symbol_name && p->addr) || - (!p->symbol_name && !p->addr)) + if ((symbol_name && addr) || (!symbol_name && !addr)) goto invalid; - if (p->symbol_name) { - kprobe_lookup_name(p->symbol_name, addr); + if (symbol_name) { + addr = kprobe_lookup_name(symbol_name, offset); if (!addr) return ERR_PTR(-ENOENT); } - addr = (kprobe_opcode_t *)(((char *)addr) + p->offset); + addr = (kprobe_opcode_t *)(((char *)addr) + offset); if (addr) return addr; @@ -1378,6 +1418,11 @@ invalid: return ERR_PTR(-EINVAL); } +static kprobe_opcode_t *kprobe_addr(struct kprobe *p) +{ + return _kprobe_addr(p->addr, p->symbol_name, p->offset); +} + /* Check passed kprobe is valid and return kprobe in kprobe_table. */ static struct kprobe *__get_valid_kprobe(struct kprobe *p) { @@ -1513,9 +1558,12 @@ int register_kprobe(struct kprobe *p) goto out; } - mutex_lock(&text_mutex); /* Avoiding text modification */ + cpus_read_lock(); + /* Prevent text modification */ + mutex_lock(&text_mutex); ret = prepare_kprobe(p); mutex_unlock(&text_mutex); + cpus_read_unlock(); if (ret) goto out; @@ -1528,7 +1576,6 @@ int register_kprobe(struct kprobe *p) /* Try to optimize kprobe */ try_to_optimize_kprobe(p); - out: mutex_unlock(&kprobe_mutex); @@ -1705,6 +1752,13 @@ void unregister_kprobes(struct kprobe **kps, int num) } EXPORT_SYMBOL_GPL(unregister_kprobes); +int __weak kprobe_exceptions_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + return NOTIFY_DONE; +} +NOKPROBE_SYMBOL(kprobe_exceptions_notify); + static struct notifier_block kprobe_exceptions_nb = { .notifier_call = kprobe_exceptions_notify, .priority = 0x7fffffff /* we need to be notified first */ @@ -1717,24 +1771,13 @@ unsigned long __weak arch_deref_entry_point(void *entry) int register_jprobes(struct jprobe **jps, int num) { - struct jprobe *jp; int ret = 0, i; if (num <= 0) return -EINVAL; + for (i = 0; i < num; i++) { - unsigned long addr, offset; - jp = jps[i]; - addr = arch_deref_entry_point(jp->entry); - - /* Verify probepoint is a function entry point */ - if (kallsyms_lookup_size_offset(addr, NULL, &offset) && - offset == 0) { - jp->kp.pre_handler = setjmp_pre_handler; - jp->kp.break_handler = longjmp_break_handler; - ret = register_kprobe(&jp->kp); - } else - ret = -EINVAL; + ret = register_jprobe(jps[i]); if (ret < 0) { if (i > 0) @@ -1742,13 +1785,30 @@ int register_jprobes(struct jprobe **jps, int num) break; } } + return ret; } EXPORT_SYMBOL_GPL(register_jprobes); int register_jprobe(struct jprobe *jp) { - return register_jprobes(&jp, 1); + unsigned long addr, offset; + struct kprobe *kp = &jp->kp; + + /* + * Verify probepoint as well as the jprobe handler are + * valid function entry points. + */ + addr = arch_deref_entry_point(jp->entry); + + if (kallsyms_lookup_size_offset(addr, NULL, &offset) && offset == 0 && + kprobe_on_func_entry(kp->addr, kp->symbol_name, kp->offset)) { + kp->pre_handler = setjmp_pre_handler; + kp->break_handler = longjmp_break_handler; + return register_kprobe(kp); + } + + return -EINVAL; } EXPORT_SYMBOL_GPL(register_jprobe); @@ -1834,6 +1894,25 @@ static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs) } NOKPROBE_SYMBOL(pre_handler_kretprobe); +bool __weak arch_kprobe_on_func_entry(unsigned long offset) +{ + return !offset; +} + +bool kprobe_on_func_entry(kprobe_opcode_t *addr, const char *sym, unsigned long offset) +{ + kprobe_opcode_t *kp_addr = _kprobe_addr(addr, sym, offset); + + if (IS_ERR(kp_addr)) + return false; + + if (!kallsyms_lookup_size_offset((unsigned long)kp_addr, NULL, &offset) || + !arch_kprobe_on_func_entry(offset)) + return false; + + return true; +} + int register_kretprobe(struct kretprobe *rp) { int ret = 0; @@ -1841,6 +1920,9 @@ int register_kretprobe(struct kretprobe *rp) int i; void *addr; + if (!kprobe_on_func_entry(rp->kp.addr, rp->kp.symbol_name, rp->kp.offset)) + return -EINVAL; + if (kretprobe_blacklist_size) { addr = kprobe_addr(&rp->kp); if (IS_ERR(addr)) @@ -2112,6 +2194,12 @@ static int kprobes_module_callback(struct notifier_block *nb, * The vaddr this probe is installed will soon * be vfreed buy not synced to disk. Hence, * disarming the breakpoint isn't needed. + * + * Note, this will also move any optimized probes + * that are pending to be removed from their + * corresponding lists to the freeing_list and + * will not be touched by the delayed + * kprobe_optimizer work handler. */ kill_kprobe(p); } @@ -2151,8 +2239,8 @@ static int __init init_kprobes(void) if (kretprobe_blacklist_size) { /* lookup the function address from its name */ for (i = 0; kretprobe_blacklist[i].name != NULL; i++) { - kprobe_lookup_name(kretprobe_blacklist[i].name, - kretprobe_blacklist[i].addr); + kretprobe_blacklist[i].addr = + kprobe_lookup_name(kretprobe_blacklist[i].name, 0); if (!kretprobe_blacklist[i].addr) printk("kretprobe: lookup failed: %s\n", kretprobe_blacklist[i].name); diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index ee1bc1bb8feb..46ba853656f6 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -125,16 +125,20 @@ static ssize_t kexec_crash_size_store(struct kobject *kobj, } KERNEL_ATTR_RW(kexec_crash_size); +#endif /* CONFIG_KEXEC_CORE */ + +#ifdef CONFIG_CRASH_CORE + static ssize_t vmcoreinfo_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { phys_addr_t vmcore_base = paddr_vmcoreinfo_note(); return sprintf(buf, "%pa %x\n", &vmcore_base, - (unsigned int)sizeof(vmcoreinfo_note)); + (unsigned int)VMCOREINFO_NOTE_SIZE); } KERNEL_ATTR_RO(vmcoreinfo); -#endif /* CONFIG_KEXEC_CORE */ +#endif /* CONFIG_CRASH_CORE */ /* whether file capabilities are enabled */ static ssize_t fscaps_show(struct kobject *kobj, @@ -195,7 +199,7 @@ static ssize_t notes_read(struct file *filp, struct kobject *kobj, return count; } -static struct bin_attribute notes_attr = { +static struct bin_attribute notes_attr __ro_after_init = { .attr = { .name = "notes", .mode = S_IRUGO, @@ -219,6 +223,8 @@ static struct attribute * kernel_attrs[] = { &kexec_loaded_attr.attr, &kexec_crash_loaded_attr.attr, &kexec_crash_size_attr.attr, +#endif +#ifdef CONFIG_CRASH_CORE &vmcoreinfo_attr.attr, #endif #ifndef CONFIG_TINY_RCU @@ -228,7 +234,7 @@ static struct attribute * kernel_attrs[] = { NULL }; -static struct attribute_group kernel_attr_group = { +static const struct attribute_group kernel_attr_group = { .attrs = kernel_attrs, }; diff --git a/kernel/kthread.c b/kernel/kthread.c index 2318fba86277..26db528c1d88 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -5,7 +5,9 @@ * even if we're invoked from userspace (think modprobe, hotplug cpu, * etc.). */ +#include <uapi/linux/sched/types.h> #include <linux/sched.h> +#include <linux/sched/task.h> #include <linux/kthread.h> #include <linux/completion.h> #include <linux/err.h> @@ -18,6 +20,7 @@ #include <linux/freezer.h> #include <linux/ptrace.h> #include <linux/uaccess.h> +#include <linux/cgroup.h> #include <trace/events/sched.h> static DEFINE_SPINLOCK(kthread_create_lock); @@ -223,6 +226,7 @@ static int kthread(void *_create) ret = -EINTR; if (!test_bit(KTHREAD_SHOULD_STOP, &self->flags)) { + cgroup_kthread_ready(); __kthread_parkme(self); ret = threadfn(data); } @@ -536,6 +540,7 @@ int kthreadd(void *unused) set_mems_allowed(node_states[N_MEMORY]); current->flags |= PF_NOFREEZE; + cgroup_init_kthreadd(); for (;;) { set_current_state(TASK_INTERRUPTIBLE); @@ -850,7 +855,6 @@ void __kthread_queue_delayed_work(struct kthread_worker *worker, list_add(&work->node, &worker->delayed_work_list); work->worker = worker; - timer_stats_timer_set_start_info(&dwork->timer); timer->expires = jiffies + delay; add_timer(timer); } diff --git a/kernel/latencytop.c b/kernel/latencytop.c index b5c30d9f46c5..96b4179cee6a 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -55,6 +55,8 @@ #include <linux/latencytop.h> #include <linux/export.h> #include <linux/sched.h> +#include <linux/sched/debug.h> +#include <linux/sched/stat.h> #include <linux/list.h> #include <linux/stacktrace.h> diff --git a/kernel/livepatch/Kconfig b/kernel/livepatch/Kconfig index 045022557936..ec4565122e65 100644 --- a/kernel/livepatch/Kconfig +++ b/kernel/livepatch/Kconfig @@ -10,6 +10,7 @@ config LIVEPATCH depends on SYSFS depends on KALLSYMS_ALL depends on HAVE_LIVEPATCH + depends on !TRIM_UNUSED_KSYMS help Say Y here if you want to support kernel live patching. This option has no runtime impact until a kernel "patch" diff --git a/kernel/livepatch/Makefile b/kernel/livepatch/Makefile index e8780c0901d9..2b8bdb1925da 100644 --- a/kernel/livepatch/Makefile +++ b/kernel/livepatch/Makefile @@ -1,3 +1,3 @@ obj-$(CONFIG_LIVEPATCH) += livepatch.o -livepatch-objs := core.o +livepatch-objs := core.o patch.o transition.o diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c index af4643873e71..b9628e43c78f 100644 --- a/kernel/livepatch/core.c +++ b/kernel/livepatch/core.c @@ -24,61 +24,31 @@ #include <linux/kernel.h> #include <linux/mutex.h> #include <linux/slab.h> -#include <linux/ftrace.h> #include <linux/list.h> #include <linux/kallsyms.h> #include <linux/livepatch.h> #include <linux/elf.h> #include <linux/moduleloader.h> +#include <linux/completion.h> #include <asm/cacheflush.h> - -/** - * struct klp_ops - structure for tracking registered ftrace ops structs - * - * A single ftrace_ops is shared between all enabled replacement functions - * (klp_func structs) which have the same old_addr. This allows the switch - * between function versions to happen instantaneously by updating the klp_ops - * struct's func_stack list. The winner is the klp_func at the top of the - * func_stack (front of the list). - * - * @node: node for the global klp_ops list - * @func_stack: list head for the stack of klp_func's (active func is on top) - * @fops: registered ftrace ops struct - */ -struct klp_ops { - struct list_head node; - struct list_head func_stack; - struct ftrace_ops fops; -}; +#include "core.h" +#include "patch.h" +#include "transition.h" /* - * The klp_mutex protects the global lists and state transitions of any - * structure reachable from them. References to any structure must be obtained - * under mutex protection (except in klp_ftrace_handler(), which uses RCU to - * ensure it gets consistent data). + * klp_mutex is a coarse lock which serializes access to klp data. All + * accesses to klp-related variables and structures must have mutex protection, + * except within the following functions which carefully avoid the need for it: + * + * - klp_ftrace_handler() + * - klp_update_patch_state() */ -static DEFINE_MUTEX(klp_mutex); +DEFINE_MUTEX(klp_mutex); static LIST_HEAD(klp_patches); -static LIST_HEAD(klp_ops); static struct kobject *klp_root_kobj; -static struct klp_ops *klp_find_ops(unsigned long old_addr) -{ - struct klp_ops *ops; - struct klp_func *func; - - list_for_each_entry(ops, &klp_ops, node) { - func = list_first_entry(&ops->func_stack, struct klp_func, - stack_node); - if (func->old_addr == old_addr) - return ops; - } - - return NULL; -} - static bool klp_is_module(struct klp_object *obj) { return obj->name; @@ -117,7 +87,6 @@ static void klp_find_object_module(struct klp_object *obj) mutex_unlock(&module_mutex); } -/* klp_mutex must be held by caller */ static bool klp_is_patch_registered(struct klp_patch *patch) { struct klp_patch *mypatch; @@ -182,7 +151,10 @@ static int klp_find_object_symbol(const char *objname, const char *name, }; mutex_lock(&module_mutex); - kallsyms_on_each_symbol(klp_find_callback, &args); + if (objname) + module_kallsyms_on_each_symbol(klp_find_callback, &args); + else + kallsyms_on_each_symbol(klp_find_callback, &args); mutex_unlock(&module_mutex); /* @@ -233,7 +205,7 @@ static int klp_resolve_symbols(Elf_Shdr *relasec, struct module *pmod) for (i = 0; i < relasec->sh_size / sizeof(Elf_Rela); i++) { sym = pmod->core_kallsyms.symtab + ELF_R_SYM(relas[i].r_info); if (sym->st_shndx != SHN_LIVEPATCH) { - pr_err("symbol %s is not marked as a livepatch symbol", + pr_err("symbol %s is not marked as a livepatch symbol\n", strtab + sym->st_name); return -EINVAL; } @@ -243,7 +215,7 @@ static int klp_resolve_symbols(Elf_Shdr *relasec, struct module *pmod) ".klp.sym.%55[^.].%127[^,],%lu", objname, symname, &sympos); if (cnt != 3) { - pr_err("symbol %s has an incorrectly formatted name", + pr_err("symbol %s has an incorrectly formatted name\n", strtab + sym->st_name); return -EINVAL; } @@ -288,7 +260,7 @@ static int klp_write_object_relocations(struct module *pmod, */ cnt = sscanf(secname, ".klp.rela.%55[^.]", sec_objname); if (cnt != 1) { - pr_err("section %s has an incorrectly formatted name", + pr_err("section %s has an incorrectly formatted name\n", secname); ret = -EINVAL; break; @@ -311,191 +283,30 @@ static int klp_write_object_relocations(struct module *pmod, return ret; } -static void notrace klp_ftrace_handler(unsigned long ip, - unsigned long parent_ip, - struct ftrace_ops *fops, - struct pt_regs *regs) -{ - struct klp_ops *ops; - struct klp_func *func; - - ops = container_of(fops, struct klp_ops, fops); - - rcu_read_lock(); - func = list_first_or_null_rcu(&ops->func_stack, struct klp_func, - stack_node); - if (WARN_ON_ONCE(!func)) - goto unlock; - - klp_arch_set_pc(regs, (unsigned long)func->new_func); -unlock: - rcu_read_unlock(); -} - -/* - * Convert a function address into the appropriate ftrace location. - * - * Usually this is just the address of the function, but on some architectures - * it's more complicated so allow them to provide a custom behaviour. - */ -#ifndef klp_get_ftrace_location -static unsigned long klp_get_ftrace_location(unsigned long faddr) -{ - return faddr; -} -#endif - -static void klp_disable_func(struct klp_func *func) -{ - struct klp_ops *ops; - - if (WARN_ON(func->state != KLP_ENABLED)) - return; - if (WARN_ON(!func->old_addr)) - return; - - ops = klp_find_ops(func->old_addr); - if (WARN_ON(!ops)) - return; - - if (list_is_singular(&ops->func_stack)) { - unsigned long ftrace_loc; - - ftrace_loc = klp_get_ftrace_location(func->old_addr); - if (WARN_ON(!ftrace_loc)) - return; - - WARN_ON(unregister_ftrace_function(&ops->fops)); - WARN_ON(ftrace_set_filter_ip(&ops->fops, ftrace_loc, 1, 0)); - - list_del_rcu(&func->stack_node); - list_del(&ops->node); - kfree(ops); - } else { - list_del_rcu(&func->stack_node); - } - - func->state = KLP_DISABLED; -} - -static int klp_enable_func(struct klp_func *func) -{ - struct klp_ops *ops; - int ret; - - if (WARN_ON(!func->old_addr)) - return -EINVAL; - - if (WARN_ON(func->state != KLP_DISABLED)) - return -EINVAL; - - ops = klp_find_ops(func->old_addr); - if (!ops) { - unsigned long ftrace_loc; - - ftrace_loc = klp_get_ftrace_location(func->old_addr); - if (!ftrace_loc) { - pr_err("failed to find location for function '%s'\n", - func->old_name); - return -EINVAL; - } - - ops = kzalloc(sizeof(*ops), GFP_KERNEL); - if (!ops) - return -ENOMEM; - - ops->fops.func = klp_ftrace_handler; - ops->fops.flags = FTRACE_OPS_FL_SAVE_REGS | - FTRACE_OPS_FL_DYNAMIC | - FTRACE_OPS_FL_IPMODIFY; - - list_add(&ops->node, &klp_ops); - - INIT_LIST_HEAD(&ops->func_stack); - list_add_rcu(&func->stack_node, &ops->func_stack); - - ret = ftrace_set_filter_ip(&ops->fops, ftrace_loc, 0, 0); - if (ret) { - pr_err("failed to set ftrace filter for function '%s' (%d)\n", - func->old_name, ret); - goto err; - } - - ret = register_ftrace_function(&ops->fops); - if (ret) { - pr_err("failed to register ftrace handler for function '%s' (%d)\n", - func->old_name, ret); - ftrace_set_filter_ip(&ops->fops, ftrace_loc, 1, 0); - goto err; - } - - - } else { - list_add_rcu(&func->stack_node, &ops->func_stack); - } - - func->state = KLP_ENABLED; - - return 0; - -err: - list_del_rcu(&func->stack_node); - list_del(&ops->node); - kfree(ops); - return ret; -} - -static void klp_disable_object(struct klp_object *obj) -{ - struct klp_func *func; - - klp_for_each_func(obj, func) - if (func->state == KLP_ENABLED) - klp_disable_func(func); - - obj->state = KLP_DISABLED; -} - -static int klp_enable_object(struct klp_object *obj) -{ - struct klp_func *func; - int ret; - - if (WARN_ON(obj->state != KLP_DISABLED)) - return -EINVAL; - - if (WARN_ON(!klp_is_object_loaded(obj))) - return -EINVAL; - - klp_for_each_func(obj, func) { - ret = klp_enable_func(func); - if (ret) { - klp_disable_object(obj); - return ret; - } - } - obj->state = KLP_ENABLED; - - return 0; -} - static int __klp_disable_patch(struct klp_patch *patch) { - struct klp_object *obj; + if (klp_transition_patch) + return -EBUSY; /* enforce stacking: only the last enabled patch can be disabled */ if (!list_is_last(&patch->list, &klp_patches) && - list_next_entry(patch, list)->state == KLP_ENABLED) + list_next_entry(patch, list)->enabled) return -EBUSY; - pr_notice("disabling patch '%s'\n", patch->mod->name); + klp_init_transition(patch, KLP_UNPATCHED); - klp_for_each_object(patch, obj) { - if (obj->state == KLP_ENABLED) - klp_disable_object(obj); - } + /* + * Enforce the order of the func->transition writes in + * klp_init_transition() and the TIF_PATCH_PENDING writes in + * klp_start_transition(). In the rare case where klp_ftrace_handler() + * is called shortly after klp_update_patch_state() switches the task, + * this ensures the handler sees that func->transition is set. + */ + smp_wmb(); - patch->state = KLP_DISABLED; + klp_start_transition(); + klp_try_complete_transition(); + patch->enabled = false; return 0; } @@ -519,7 +330,7 @@ int klp_disable_patch(struct klp_patch *patch) goto err; } - if (patch->state == KLP_DISABLED) { + if (!patch->enabled) { ret = -EINVAL; goto err; } @@ -537,32 +348,61 @@ static int __klp_enable_patch(struct klp_patch *patch) struct klp_object *obj; int ret; - if (WARN_ON(patch->state != KLP_DISABLED)) + if (klp_transition_patch) + return -EBUSY; + + if (WARN_ON(patch->enabled)) return -EINVAL; /* enforce stacking: only the first disabled patch can be enabled */ if (patch->list.prev != &klp_patches && - list_prev_entry(patch, list)->state == KLP_DISABLED) + !list_prev_entry(patch, list)->enabled) return -EBUSY; + /* + * A reference is taken on the patch module to prevent it from being + * unloaded. + * + * Note: For immediate (no consistency model) patches we don't allow + * patch modules to unload since there is no safe/sane method to + * determine if a thread is still running in the patched code contained + * in the patch module once the ftrace registration is successful. + */ + if (!try_module_get(patch->mod)) + return -ENODEV; + pr_notice("enabling patch '%s'\n", patch->mod->name); + klp_init_transition(patch, KLP_PATCHED); + + /* + * Enforce the order of the func->transition writes in + * klp_init_transition() and the ops->func_stack writes in + * klp_patch_object(), so that klp_ftrace_handler() will see the + * func->transition updates before the handler is registered and the + * new funcs become visible to the handler. + */ + smp_wmb(); + klp_for_each_object(patch, obj) { if (!klp_is_object_loaded(obj)) continue; - ret = klp_enable_object(obj); - if (ret) - goto unregister; + ret = klp_patch_object(obj); + if (ret) { + pr_warn("failed to enable patch '%s'\n", + patch->mod->name); + + klp_cancel_transition(); + return ret; + } } - patch->state = KLP_ENABLED; + klp_start_transition(); + klp_try_complete_transition(); + patch->enabled = true; return 0; - -unregister: - WARN_ON(__klp_disable_patch(patch)); - return ret; } /** @@ -599,6 +439,7 @@ EXPORT_SYMBOL_GPL(klp_enable_patch); * /sys/kernel/livepatch * /sys/kernel/livepatch/<patch> * /sys/kernel/livepatch/<patch>/enabled + * /sys/kernel/livepatch/<patch>/transition * /sys/kernel/livepatch/<patch>/<object> * /sys/kernel/livepatch/<patch>/<object>/<function,sympos> */ @@ -608,26 +449,34 @@ static ssize_t enabled_store(struct kobject *kobj, struct kobj_attribute *attr, { struct klp_patch *patch; int ret; - unsigned long val; + bool enabled; - ret = kstrtoul(buf, 10, &val); + ret = kstrtobool(buf, &enabled); if (ret) - return -EINVAL; - - if (val != KLP_DISABLED && val != KLP_ENABLED) - return -EINVAL; + return ret; patch = container_of(kobj, struct klp_patch, kobj); mutex_lock(&klp_mutex); - if (val == patch->state) { + if (!klp_is_patch_registered(patch)) { + /* + * Module with the patch could either disappear meanwhile or is + * not properly initialized yet. + */ + ret = -EINVAL; + goto err; + } + + if (patch->enabled == enabled) { /* already in requested state */ ret = -EINVAL; goto err; } - if (val == KLP_ENABLED) { + if (patch == klp_transition_patch) { + klp_reverse_transition(); + } else if (enabled) { ret = __klp_enable_patch(patch); if (ret) goto err; @@ -652,21 +501,33 @@ static ssize_t enabled_show(struct kobject *kobj, struct klp_patch *patch; patch = container_of(kobj, struct klp_patch, kobj); - return snprintf(buf, PAGE_SIZE-1, "%d\n", patch->state); + return snprintf(buf, PAGE_SIZE-1, "%d\n", patch->enabled); +} + +static ssize_t transition_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct klp_patch *patch; + + patch = container_of(kobj, struct klp_patch, kobj); + return snprintf(buf, PAGE_SIZE-1, "%d\n", + patch == klp_transition_patch); } static struct kobj_attribute enabled_kobj_attr = __ATTR_RW(enabled); +static struct kobj_attribute transition_kobj_attr = __ATTR_RO(transition); static struct attribute *klp_patch_attrs[] = { &enabled_kobj_attr.attr, + &transition_kobj_attr.attr, NULL }; static void klp_kobj_release_patch(struct kobject *kobj) { - /* - * Once we have a consistency model we'll need to module_put() the - * patch module here. See klp_register_patch() for more details. - */ + struct klp_patch *patch; + + patch = container_of(kobj, struct klp_patch, kobj); + complete(&patch->finish); } static struct kobj_type klp_ktype_patch = { @@ -737,7 +598,6 @@ static void klp_free_patch(struct klp_patch *patch) klp_free_objects_limited(patch, NULL); if (!list_empty(&patch->list)) list_del(&patch->list); - kobject_put(&patch->kobj); } static int klp_init_func(struct klp_object *obj, struct klp_func *func) @@ -746,7 +606,8 @@ static int klp_init_func(struct klp_object *obj, struct klp_func *func) return -EINVAL; INIT_LIST_HEAD(&func->stack_node); - func->state = KLP_DISABLED; + func->patched = false; + func->transition = false; /* The format for the sysfs directory is <function,sympos> where sympos * is the nth occurrence of this symbol in kallsyms for the patched @@ -787,6 +648,22 @@ static int klp_init_object_loaded(struct klp_patch *patch, &func->old_addr); if (ret) return ret; + + ret = kallsyms_lookup_size_offset(func->old_addr, + &func->old_size, NULL); + if (!ret) { + pr_err("kallsyms size lookup failed for '%s'\n", + func->old_name); + return -ENOENT; + } + + ret = kallsyms_lookup_size_offset((unsigned long)func->new_func, + &func->new_size, NULL); + if (!ret) { + pr_err("kallsyms size lookup failed for '%s' replacement\n", + func->old_name); + return -ENOENT; + } } return 0; @@ -801,7 +678,7 @@ static int klp_init_object(struct klp_patch *patch, struct klp_object *obj) if (!obj->funcs) return -EINVAL; - obj->state = KLP_DISABLED; + obj->patched = false; obj->mod = NULL; klp_find_object_module(obj); @@ -842,12 +719,15 @@ static int klp_init_patch(struct klp_patch *patch) mutex_lock(&klp_mutex); - patch->state = KLP_DISABLED; + patch->enabled = false; + init_completion(&patch->finish); ret = kobject_init_and_add(&patch->kobj, &klp_ktype_patch, klp_root_kobj, "%s", patch->mod->name); - if (ret) - goto unlock; + if (ret) { + mutex_unlock(&klp_mutex); + return ret; + } klp_for_each_object(patch, obj) { ret = klp_init_object(patch, obj); @@ -863,9 +743,12 @@ static int klp_init_patch(struct klp_patch *patch) free: klp_free_objects_limited(patch, obj); - kobject_put(&patch->kobj); -unlock: + mutex_unlock(&klp_mutex); + + kobject_put(&patch->kobj); + wait_for_completion(&patch->finish); + return ret; } @@ -879,23 +762,29 @@ unlock: */ int klp_unregister_patch(struct klp_patch *patch) { - int ret = 0; + int ret; mutex_lock(&klp_mutex); if (!klp_is_patch_registered(patch)) { ret = -EINVAL; - goto out; + goto err; } - if (patch->state == KLP_ENABLED) { + if (patch->enabled) { ret = -EBUSY; - goto out; + goto err; } klp_free_patch(patch); -out: + mutex_unlock(&klp_mutex); + + kobject_put(&patch->kobj); + wait_for_completion(&patch->finish); + + return 0; +err: mutex_unlock(&klp_mutex); return ret; } @@ -908,17 +797,18 @@ EXPORT_SYMBOL_GPL(klp_unregister_patch); * Initializes the data structure associated with the patch and * creates the sysfs interface. * + * There is no need to take the reference on the patch module here. It is done + * later when the patch is enabled. + * * Return: 0 on success, otherwise error */ int klp_register_patch(struct klp_patch *patch) { - int ret; - if (!patch || !patch->mod) return -EINVAL; if (!is_livepatch_module(patch->mod)) { - pr_err("module %s is not marked as a livepatch module", + pr_err("module %s is not marked as a livepatch module\n", patch->mod->name); return -EINVAL; } @@ -927,20 +817,16 @@ int klp_register_patch(struct klp_patch *patch) return -ENODEV; /* - * A reference is taken on the patch module to prevent it from being - * unloaded. Right now, we don't allow patch modules to unload since - * there is currently no method to determine if a thread is still - * running in the patched code contained in the patch module once - * the ftrace registration is successful. + * Architectures without reliable stack traces have to set + * patch->immediate because there's currently no way to patch kthreads + * with the consistency model. */ - if (!try_module_get(patch->mod)) - return -ENODEV; - - ret = klp_init_patch(patch); - if (ret) - module_put(patch->mod); + if (!klp_have_reliable_stack() && !patch->immediate) { + pr_err("This architecture doesn't have support for the livepatch consistency model.\n"); + return -ENOSYS; + } - return ret; + return klp_init_patch(patch); } EXPORT_SYMBOL_GPL(klp_register_patch); @@ -975,13 +861,17 @@ int klp_module_coming(struct module *mod) goto err; } - if (patch->state == KLP_DISABLED) + /* + * Only patch the module if the patch is enabled or is + * in transition. + */ + if (!patch->enabled && patch != klp_transition_patch) break; pr_notice("applying patch '%s' to loading module '%s'\n", patch->mod->name, obj->mod->name); - ret = klp_enable_object(obj); + ret = klp_patch_object(obj); if (ret) { pr_warn("failed to apply patch '%s' to module '%s' (%d)\n", patch->mod->name, obj->mod->name, ret); @@ -1032,10 +922,14 @@ void klp_module_going(struct module *mod) if (!klp_is_module(obj) || strcmp(obj->name, mod->name)) continue; - if (patch->state != KLP_DISABLED) { + /* + * Only unpatch the module if the patch is enabled or + * is in transition. + */ + if (patch->enabled || patch == klp_transition_patch) { pr_notice("reverting patch '%s' on unloading module '%s'\n", patch->mod->name, obj->mod->name); - klp_disable_object(obj); + klp_unpatch_object(obj); } klp_free_object_loaded(obj); diff --git a/kernel/livepatch/core.h b/kernel/livepatch/core.h new file mode 100644 index 000000000000..c74f24c47837 --- /dev/null +++ b/kernel/livepatch/core.h @@ -0,0 +1,6 @@ +#ifndef _LIVEPATCH_CORE_H +#define _LIVEPATCH_CORE_H + +extern struct mutex klp_mutex; + +#endif /* _LIVEPATCH_CORE_H */ diff --git a/kernel/livepatch/patch.c b/kernel/livepatch/patch.c new file mode 100644 index 000000000000..52c4e907c14b --- /dev/null +++ b/kernel/livepatch/patch.c @@ -0,0 +1,276 @@ +/* + * patch.c - livepatch patching functions + * + * Copyright (C) 2014 Seth Jennings <[email protected]> + * Copyright (C) 2014 SUSE + * Copyright (C) 2015 Josh Poimboeuf <[email protected]> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/livepatch.h> +#include <linux/list.h> +#include <linux/ftrace.h> +#include <linux/rculist.h> +#include <linux/slab.h> +#include <linux/bug.h> +#include <linux/printk.h> +#include "patch.h" +#include "transition.h" + +static LIST_HEAD(klp_ops); + +struct klp_ops *klp_find_ops(unsigned long old_addr) +{ + struct klp_ops *ops; + struct klp_func *func; + + list_for_each_entry(ops, &klp_ops, node) { + func = list_first_entry(&ops->func_stack, struct klp_func, + stack_node); + if (func->old_addr == old_addr) + return ops; + } + + return NULL; +} + +static void notrace klp_ftrace_handler(unsigned long ip, + unsigned long parent_ip, + struct ftrace_ops *fops, + struct pt_regs *regs) +{ + struct klp_ops *ops; + struct klp_func *func; + int patch_state; + + ops = container_of(fops, struct klp_ops, fops); + + /* + * A variant of synchronize_sched() is used to allow patching functions + * where RCU is not watching, see klp_synchronize_transition(). + */ + preempt_disable_notrace(); + + func = list_first_or_null_rcu(&ops->func_stack, struct klp_func, + stack_node); + + /* + * func should never be NULL because preemption should be disabled here + * and unregister_ftrace_function() does the equivalent of a + * synchronize_sched() before the func_stack removal. + */ + if (WARN_ON_ONCE(!func)) + goto unlock; + + /* + * In the enable path, enforce the order of the ops->func_stack and + * func->transition reads. The corresponding write barrier is in + * __klp_enable_patch(). + * + * (Note that this barrier technically isn't needed in the disable + * path. In the rare case where klp_update_patch_state() runs before + * this handler, its TIF_PATCH_PENDING read and this func->transition + * read need to be ordered. But klp_update_patch_state() already + * enforces that.) + */ + smp_rmb(); + + if (unlikely(func->transition)) { + + /* + * Enforce the order of the func->transition and + * current->patch_state reads. Otherwise we could read an + * out-of-date task state and pick the wrong function. The + * corresponding write barrier is in klp_init_transition(). + */ + smp_rmb(); + + patch_state = current->patch_state; + + WARN_ON_ONCE(patch_state == KLP_UNDEFINED); + + if (patch_state == KLP_UNPATCHED) { + /* + * Use the previously patched version of the function. + * If no previous patches exist, continue with the + * original function. + */ + func = list_entry_rcu(func->stack_node.next, + struct klp_func, stack_node); + + if (&func->stack_node == &ops->func_stack) + goto unlock; + } + } + + klp_arch_set_pc(regs, (unsigned long)func->new_func); +unlock: + preempt_enable_notrace(); +} + +/* + * Convert a function address into the appropriate ftrace location. + * + * Usually this is just the address of the function, but on some architectures + * it's more complicated so allow them to provide a custom behaviour. + */ +#ifndef klp_get_ftrace_location +static unsigned long klp_get_ftrace_location(unsigned long faddr) +{ + return faddr; +} +#endif + +static void klp_unpatch_func(struct klp_func *func) +{ + struct klp_ops *ops; + + if (WARN_ON(!func->patched)) + return; + if (WARN_ON(!func->old_addr)) + return; + + ops = klp_find_ops(func->old_addr); + if (WARN_ON(!ops)) + return; + + if (list_is_singular(&ops->func_stack)) { + unsigned long ftrace_loc; + + ftrace_loc = klp_get_ftrace_location(func->old_addr); + if (WARN_ON(!ftrace_loc)) + return; + + WARN_ON(unregister_ftrace_function(&ops->fops)); + WARN_ON(ftrace_set_filter_ip(&ops->fops, ftrace_loc, 1, 0)); + + list_del_rcu(&func->stack_node); + list_del(&ops->node); + kfree(ops); + } else { + list_del_rcu(&func->stack_node); + } + + func->patched = false; +} + +static int klp_patch_func(struct klp_func *func) +{ + struct klp_ops *ops; + int ret; + + if (WARN_ON(!func->old_addr)) + return -EINVAL; + + if (WARN_ON(func->patched)) + return -EINVAL; + + ops = klp_find_ops(func->old_addr); + if (!ops) { + unsigned long ftrace_loc; + + ftrace_loc = klp_get_ftrace_location(func->old_addr); + if (!ftrace_loc) { + pr_err("failed to find location for function '%s'\n", + func->old_name); + return -EINVAL; + } + + ops = kzalloc(sizeof(*ops), GFP_KERNEL); + if (!ops) + return -ENOMEM; + + ops->fops.func = klp_ftrace_handler; + ops->fops.flags = FTRACE_OPS_FL_SAVE_REGS | + FTRACE_OPS_FL_DYNAMIC | + FTRACE_OPS_FL_IPMODIFY; + + list_add(&ops->node, &klp_ops); + + INIT_LIST_HEAD(&ops->func_stack); + list_add_rcu(&func->stack_node, &ops->func_stack); + + ret = ftrace_set_filter_ip(&ops->fops, ftrace_loc, 0, 0); + if (ret) { + pr_err("failed to set ftrace filter for function '%s' (%d)\n", + func->old_name, ret); + goto err; + } + + ret = register_ftrace_function(&ops->fops); + if (ret) { + pr_err("failed to register ftrace handler for function '%s' (%d)\n", + func->old_name, ret); + ftrace_set_filter_ip(&ops->fops, ftrace_loc, 1, 0); + goto err; + } + + + } else { + list_add_rcu(&func->stack_node, &ops->func_stack); + } + + func->patched = true; + + return 0; + +err: + list_del_rcu(&func->stack_node); + list_del(&ops->node); + kfree(ops); + return ret; +} + +void klp_unpatch_object(struct klp_object *obj) +{ + struct klp_func *func; + + klp_for_each_func(obj, func) + if (func->patched) + klp_unpatch_func(func); + + obj->patched = false; +} + +int klp_patch_object(struct klp_object *obj) +{ + struct klp_func *func; + int ret; + + if (WARN_ON(obj->patched)) + return -EINVAL; + + klp_for_each_func(obj, func) { + ret = klp_patch_func(func); + if (ret) { + klp_unpatch_object(obj); + return ret; + } + } + obj->patched = true; + + return 0; +} + +void klp_unpatch_objects(struct klp_patch *patch) +{ + struct klp_object *obj; + + klp_for_each_object(patch, obj) + if (obj->patched) + klp_unpatch_object(obj); +} diff --git a/kernel/livepatch/patch.h b/kernel/livepatch/patch.h new file mode 100644 index 000000000000..0db227170c36 --- /dev/null +++ b/kernel/livepatch/patch.h @@ -0,0 +1,33 @@ +#ifndef _LIVEPATCH_PATCH_H +#define _LIVEPATCH_PATCH_H + +#include <linux/livepatch.h> +#include <linux/list.h> +#include <linux/ftrace.h> + +/** + * struct klp_ops - structure for tracking registered ftrace ops structs + * + * A single ftrace_ops is shared between all enabled replacement functions + * (klp_func structs) which have the same old_addr. This allows the switch + * between function versions to happen instantaneously by updating the klp_ops + * struct's func_stack list. The winner is the klp_func at the top of the + * func_stack (front of the list). + * + * @node: node for the global klp_ops list + * @func_stack: list head for the stack of klp_func's (active func is on top) + * @fops: registered ftrace ops struct + */ +struct klp_ops { + struct list_head node; + struct list_head func_stack; + struct ftrace_ops fops; +}; + +struct klp_ops *klp_find_ops(unsigned long old_addr); + +int klp_patch_object(struct klp_object *obj); +void klp_unpatch_object(struct klp_object *obj); +void klp_unpatch_objects(struct klp_patch *patch); + +#endif /* _LIVEPATCH_PATCH_H */ diff --git a/kernel/livepatch/transition.c b/kernel/livepatch/transition.c new file mode 100644 index 000000000000..b004a1fb6032 --- /dev/null +++ b/kernel/livepatch/transition.c @@ -0,0 +1,579 @@ +/* + * transition.c - Kernel Live Patching transition functions + * + * Copyright (C) 2015-2016 Josh Poimboeuf <[email protected]> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/cpu.h> +#include <linux/stacktrace.h> +#include "core.h" +#include "patch.h" +#include "transition.h" +#include "../sched/sched.h" + +#define MAX_STACK_ENTRIES 100 +#define STACK_ERR_BUF_SIZE 128 + +struct klp_patch *klp_transition_patch; + +static int klp_target_state = KLP_UNDEFINED; + +/* + * This work can be performed periodically to finish patching or unpatching any + * "straggler" tasks which failed to transition in the first attempt. + */ +static void klp_transition_work_fn(struct work_struct *work) +{ + mutex_lock(&klp_mutex); + + if (klp_transition_patch) + klp_try_complete_transition(); + + mutex_unlock(&klp_mutex); +} +static DECLARE_DELAYED_WORK(klp_transition_work, klp_transition_work_fn); + +/* + * This function is just a stub to implement a hard force + * of synchronize_sched(). This requires synchronizing + * tasks even in userspace and idle. + */ +static void klp_sync(struct work_struct *work) +{ +} + +/* + * We allow to patch also functions where RCU is not watching, + * e.g. before user_exit(). We can not rely on the RCU infrastructure + * to do the synchronization. Instead hard force the sched synchronization. + * + * This approach allows to use RCU functions for manipulating func_stack + * safely. + */ +static void klp_synchronize_transition(void) +{ + schedule_on_each_cpu(klp_sync); +} + +/* + * The transition to the target patch state is complete. Clean up the data + * structures. + */ +static void klp_complete_transition(void) +{ + struct klp_object *obj; + struct klp_func *func; + struct task_struct *g, *task; + unsigned int cpu; + bool immediate_func = false; + + if (klp_target_state == KLP_UNPATCHED) { + /* + * All tasks have transitioned to KLP_UNPATCHED so we can now + * remove the new functions from the func_stack. + */ + klp_unpatch_objects(klp_transition_patch); + + /* + * Make sure klp_ftrace_handler() can no longer see functions + * from this patch on the ops->func_stack. Otherwise, after + * func->transition gets cleared, the handler may choose a + * removed function. + */ + klp_synchronize_transition(); + } + + if (klp_transition_patch->immediate) + goto done; + + klp_for_each_object(klp_transition_patch, obj) { + klp_for_each_func(obj, func) { + func->transition = false; + if (func->immediate) + immediate_func = true; + } + } + + if (klp_target_state == KLP_UNPATCHED && !immediate_func) + module_put(klp_transition_patch->mod); + + /* Prevent klp_ftrace_handler() from seeing KLP_UNDEFINED state */ + if (klp_target_state == KLP_PATCHED) + klp_synchronize_transition(); + + read_lock(&tasklist_lock); + for_each_process_thread(g, task) { + WARN_ON_ONCE(test_tsk_thread_flag(task, TIF_PATCH_PENDING)); + task->patch_state = KLP_UNDEFINED; + } + read_unlock(&tasklist_lock); + + for_each_possible_cpu(cpu) { + task = idle_task(cpu); + WARN_ON_ONCE(test_tsk_thread_flag(task, TIF_PATCH_PENDING)); + task->patch_state = KLP_UNDEFINED; + } + +done: + klp_target_state = KLP_UNDEFINED; + klp_transition_patch = NULL; +} + +/* + * This is called in the error path, to cancel a transition before it has + * started, i.e. klp_init_transition() has been called but + * klp_start_transition() hasn't. If the transition *has* been started, + * klp_reverse_transition() should be used instead. + */ +void klp_cancel_transition(void) +{ + if (WARN_ON_ONCE(klp_target_state != KLP_PATCHED)) + return; + + klp_target_state = KLP_UNPATCHED; + klp_complete_transition(); +} + +/* + * Switch the patched state of the task to the set of functions in the target + * patch state. + * + * NOTE: If task is not 'current', the caller must ensure the task is inactive. + * Otherwise klp_ftrace_handler() might read the wrong 'patch_state' value. + */ +void klp_update_patch_state(struct task_struct *task) +{ + /* + * A variant of synchronize_sched() is used to allow patching functions + * where RCU is not watching, see klp_synchronize_transition(). + */ + preempt_disable_notrace(); + + /* + * This test_and_clear_tsk_thread_flag() call also serves as a read + * barrier (smp_rmb) for two cases: + * + * 1) Enforce the order of the TIF_PATCH_PENDING read and the + * klp_target_state read. The corresponding write barrier is in + * klp_init_transition(). + * + * 2) Enforce the order of the TIF_PATCH_PENDING read and a future read + * of func->transition, if klp_ftrace_handler() is called later on + * the same CPU. See __klp_disable_patch(). + */ + if (test_and_clear_tsk_thread_flag(task, TIF_PATCH_PENDING)) + task->patch_state = READ_ONCE(klp_target_state); + + preempt_enable_notrace(); +} + +/* + * Determine whether the given stack trace includes any references to a + * to-be-patched or to-be-unpatched function. + */ +static int klp_check_stack_func(struct klp_func *func, + struct stack_trace *trace) +{ + unsigned long func_addr, func_size, address; + struct klp_ops *ops; + int i; + + if (func->immediate) + return 0; + + for (i = 0; i < trace->nr_entries; i++) { + address = trace->entries[i]; + + if (klp_target_state == KLP_UNPATCHED) { + /* + * Check for the to-be-unpatched function + * (the func itself). + */ + func_addr = (unsigned long)func->new_func; + func_size = func->new_size; + } else { + /* + * Check for the to-be-patched function + * (the previous func). + */ + ops = klp_find_ops(func->old_addr); + + if (list_is_singular(&ops->func_stack)) { + /* original function */ + func_addr = func->old_addr; + func_size = func->old_size; + } else { + /* previously patched function */ + struct klp_func *prev; + + prev = list_next_entry(func, stack_node); + func_addr = (unsigned long)prev->new_func; + func_size = prev->new_size; + } + } + + if (address >= func_addr && address < func_addr + func_size) + return -EAGAIN; + } + + return 0; +} + +/* + * Determine whether it's safe to transition the task to the target patch state + * by looking for any to-be-patched or to-be-unpatched functions on its stack. + */ +static int klp_check_stack(struct task_struct *task, char *err_buf) +{ + static unsigned long entries[MAX_STACK_ENTRIES]; + struct stack_trace trace; + struct klp_object *obj; + struct klp_func *func; + int ret; + + trace.skip = 0; + trace.nr_entries = 0; + trace.max_entries = MAX_STACK_ENTRIES; + trace.entries = entries; + ret = save_stack_trace_tsk_reliable(task, &trace); + WARN_ON_ONCE(ret == -ENOSYS); + if (ret) { + snprintf(err_buf, STACK_ERR_BUF_SIZE, + "%s: %s:%d has an unreliable stack\n", + __func__, task->comm, task->pid); + return ret; + } + + klp_for_each_object(klp_transition_patch, obj) { + if (!obj->patched) + continue; + klp_for_each_func(obj, func) { + ret = klp_check_stack_func(func, &trace); + if (ret) { + snprintf(err_buf, STACK_ERR_BUF_SIZE, + "%s: %s:%d is sleeping on function %s\n", + __func__, task->comm, task->pid, + func->old_name); + return ret; + } + } + } + + return 0; +} + +/* + * Try to safely switch a task to the target patch state. If it's currently + * running, or it's sleeping on a to-be-patched or to-be-unpatched function, or + * if the stack is unreliable, return false. + */ +static bool klp_try_switch_task(struct task_struct *task) +{ + struct rq *rq; + struct rq_flags flags; + int ret; + bool success = false; + char err_buf[STACK_ERR_BUF_SIZE]; + + err_buf[0] = '\0'; + + /* check if this task has already switched over */ + if (task->patch_state == klp_target_state) + return true; + + /* + * For arches which don't have reliable stack traces, we have to rely + * on other methods (e.g., switching tasks at kernel exit). + */ + if (!klp_have_reliable_stack()) + return false; + + /* + * Now try to check the stack for any to-be-patched or to-be-unpatched + * functions. If all goes well, switch the task to the target patch + * state. + */ + rq = task_rq_lock(task, &flags); + + if (task_running(rq, task) && task != current) { + snprintf(err_buf, STACK_ERR_BUF_SIZE, + "%s: %s:%d is running\n", __func__, task->comm, + task->pid); + goto done; + } + + ret = klp_check_stack(task, err_buf); + if (ret) + goto done; + + success = true; + + clear_tsk_thread_flag(task, TIF_PATCH_PENDING); + task->patch_state = klp_target_state; + +done: + task_rq_unlock(rq, task, &flags); + + /* + * Due to console deadlock issues, pr_debug() can't be used while + * holding the task rq lock. Instead we have to use a temporary buffer + * and print the debug message after releasing the lock. + */ + if (err_buf[0] != '\0') + pr_debug("%s", err_buf); + + return success; + +} + +/* + * Try to switch all remaining tasks to the target patch state by walking the + * stacks of sleeping tasks and looking for any to-be-patched or + * to-be-unpatched functions. If such functions are found, the task can't be + * switched yet. + * + * If any tasks are still stuck in the initial patch state, schedule a retry. + */ +void klp_try_complete_transition(void) +{ + unsigned int cpu; + struct task_struct *g, *task; + bool complete = true; + + WARN_ON_ONCE(klp_target_state == KLP_UNDEFINED); + + /* + * If the patch can be applied or reverted immediately, skip the + * per-task transitions. + */ + if (klp_transition_patch->immediate) + goto success; + + /* + * Try to switch the tasks to the target patch state by walking their + * stacks and looking for any to-be-patched or to-be-unpatched + * functions. If such functions are found on a stack, or if the stack + * is deemed unreliable, the task can't be switched yet. + * + * Usually this will transition most (or all) of the tasks on a system + * unless the patch includes changes to a very common function. + */ + read_lock(&tasklist_lock); + for_each_process_thread(g, task) + if (!klp_try_switch_task(task)) + complete = false; + read_unlock(&tasklist_lock); + + /* + * Ditto for the idle "swapper" tasks. + */ + get_online_cpus(); + for_each_possible_cpu(cpu) { + task = idle_task(cpu); + if (cpu_online(cpu)) { + if (!klp_try_switch_task(task)) + complete = false; + } else if (task->patch_state != klp_target_state) { + /* offline idle tasks can be switched immediately */ + clear_tsk_thread_flag(task, TIF_PATCH_PENDING); + task->patch_state = klp_target_state; + } + } + put_online_cpus(); + + if (!complete) { + /* + * Some tasks weren't able to be switched over. Try again + * later and/or wait for other methods like kernel exit + * switching. + */ + schedule_delayed_work(&klp_transition_work, + round_jiffies_relative(HZ)); + return; + } + +success: + pr_notice("'%s': %s complete\n", klp_transition_patch->mod->name, + klp_target_state == KLP_PATCHED ? "patching" : "unpatching"); + + /* we're done, now cleanup the data structures */ + klp_complete_transition(); +} + +/* + * Start the transition to the specified target patch state so tasks can begin + * switching to it. + */ +void klp_start_transition(void) +{ + struct task_struct *g, *task; + unsigned int cpu; + + WARN_ON_ONCE(klp_target_state == KLP_UNDEFINED); + + pr_notice("'%s': %s...\n", klp_transition_patch->mod->name, + klp_target_state == KLP_PATCHED ? "patching" : "unpatching"); + + /* + * If the patch can be applied or reverted immediately, skip the + * per-task transitions. + */ + if (klp_transition_patch->immediate) + return; + + /* + * Mark all normal tasks as needing a patch state update. They'll + * switch either in klp_try_complete_transition() or as they exit the + * kernel. + */ + read_lock(&tasklist_lock); + for_each_process_thread(g, task) + if (task->patch_state != klp_target_state) + set_tsk_thread_flag(task, TIF_PATCH_PENDING); + read_unlock(&tasklist_lock); + + /* + * Mark all idle tasks as needing a patch state update. They'll switch + * either in klp_try_complete_transition() or at the idle loop switch + * point. + */ + for_each_possible_cpu(cpu) { + task = idle_task(cpu); + if (task->patch_state != klp_target_state) + set_tsk_thread_flag(task, TIF_PATCH_PENDING); + } +} + +/* + * Initialize the global target patch state and all tasks to the initial patch + * state, and initialize all function transition states to true in preparation + * for patching or unpatching. + */ +void klp_init_transition(struct klp_patch *patch, int state) +{ + struct task_struct *g, *task; + unsigned int cpu; + struct klp_object *obj; + struct klp_func *func; + int initial_state = !state; + + WARN_ON_ONCE(klp_target_state != KLP_UNDEFINED); + + klp_transition_patch = patch; + + /* + * Set the global target patch state which tasks will switch to. This + * has no effect until the TIF_PATCH_PENDING flags get set later. + */ + klp_target_state = state; + + /* + * If the patch can be applied or reverted immediately, skip the + * per-task transitions. + */ + if (patch->immediate) + return; + + /* + * Initialize all tasks to the initial patch state to prepare them for + * switching to the target state. + */ + read_lock(&tasklist_lock); + for_each_process_thread(g, task) { + WARN_ON_ONCE(task->patch_state != KLP_UNDEFINED); + task->patch_state = initial_state; + } + read_unlock(&tasklist_lock); + + /* + * Ditto for the idle "swapper" tasks. + */ + for_each_possible_cpu(cpu) { + task = idle_task(cpu); + WARN_ON_ONCE(task->patch_state != KLP_UNDEFINED); + task->patch_state = initial_state; + } + + /* + * Enforce the order of the task->patch_state initializations and the + * func->transition updates to ensure that klp_ftrace_handler() doesn't + * see a func in transition with a task->patch_state of KLP_UNDEFINED. + * + * Also enforce the order of the klp_target_state write and future + * TIF_PATCH_PENDING writes to ensure klp_update_patch_state() doesn't + * set a task->patch_state to KLP_UNDEFINED. + */ + smp_wmb(); + + /* + * Set the func transition states so klp_ftrace_handler() will know to + * switch to the transition logic. + * + * When patching, the funcs aren't yet in the func_stack and will be + * made visible to the ftrace handler shortly by the calls to + * klp_patch_object(). + * + * When unpatching, the funcs are already in the func_stack and so are + * already visible to the ftrace handler. + */ + klp_for_each_object(patch, obj) + klp_for_each_func(obj, func) + func->transition = true; +} + +/* + * This function can be called in the middle of an existing transition to + * reverse the direction of the target patch state. This can be done to + * effectively cancel an existing enable or disable operation if there are any + * tasks which are stuck in the initial patch state. + */ +void klp_reverse_transition(void) +{ + unsigned int cpu; + struct task_struct *g, *task; + + klp_transition_patch->enabled = !klp_transition_patch->enabled; + + klp_target_state = !klp_target_state; + + /* + * Clear all TIF_PATCH_PENDING flags to prevent races caused by + * klp_update_patch_state() running in parallel with + * klp_start_transition(). + */ + read_lock(&tasklist_lock); + for_each_process_thread(g, task) + clear_tsk_thread_flag(task, TIF_PATCH_PENDING); + read_unlock(&tasklist_lock); + + for_each_possible_cpu(cpu) + clear_tsk_thread_flag(idle_task(cpu), TIF_PATCH_PENDING); + + /* Let any remaining calls to klp_update_patch_state() complete */ + klp_synchronize_transition(); + + klp_start_transition(); +} + +/* Called from copy_process() during fork */ +void klp_copy_process(struct task_struct *child) +{ + child->patch_state = current->patch_state; + + /* TIF_PATCH_PENDING gets copied in setup_thread_stack() */ +} diff --git a/kernel/livepatch/transition.h b/kernel/livepatch/transition.h new file mode 100644 index 000000000000..ce09b326546c --- /dev/null +++ b/kernel/livepatch/transition.h @@ -0,0 +1,14 @@ +#ifndef _LIVEPATCH_TRANSITION_H +#define _LIVEPATCH_TRANSITION_H + +#include <linux/livepatch.h> + +extern struct klp_patch *klp_transition_patch; + +void klp_init_transition(struct klp_patch *patch, int state); +void klp_cancel_transition(void); +void klp_start_transition(void); +void klp_try_complete_transition(void); +void klp_reverse_transition(void); + +#endif /* _LIVEPATCH_TRANSITION_H */ diff --git a/kernel/locking/Makefile b/kernel/locking/Makefile index 6f88e352cd4f..760158d9d98d 100644 --- a/kernel/locking/Makefile +++ b/kernel/locking/Makefile @@ -28,3 +28,4 @@ obj-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o obj-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem-xadd.o obj-$(CONFIG_QUEUED_RWLOCKS) += qrwlock.o obj-$(CONFIG_LOCK_TORTURE_TEST) += locktorture.o +obj-$(CONFIG_WW_MUTEX_SELFTEST) += test-ww_mutex.o diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 7c38f8f3d97b..7d2499bec5fe 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -28,6 +28,9 @@ #define DISABLE_BRANCH_PROFILING #include <linux/mutex.h> #include <linux/sched.h> +#include <linux/sched/clock.h> +#include <linux/sched/task.h> +#include <linux/sched/mm.h> #include <linux/delay.h> #include <linux/module.h> #include <linux/proc_fs.h> @@ -658,6 +661,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) struct lockdep_subclass_key *key; struct hlist_head *hash_head; struct lock_class *class; + bool is_static = false; if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) { debug_locks_off(); @@ -671,10 +675,23 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) /* * Static locks do not have their class-keys yet - for them the key - * is the lock object itself: + * is the lock object itself. If the lock is in the per cpu area, + * the canonical address of the lock (per cpu offset removed) is + * used. */ - if (unlikely(!lock->key)) - lock->key = (void *)lock; + if (unlikely(!lock->key)) { + unsigned long can_addr, addr = (unsigned long)lock; + + if (__is_kernel_percpu_address(addr, &can_addr)) + lock->key = (void *)can_addr; + else if (__is_module_percpu_address(addr, &can_addr)) + lock->key = (void *)can_addr; + else if (static_obj(lock)) + lock->key = (void *)lock; + else + return ERR_PTR(-EINVAL); + is_static = true; + } /* * NOTE: the class-key must be unique. For dynamic locks, a static @@ -706,7 +723,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) } } - return NULL; + return is_static || static_obj(lock->key) ? NULL : ERR_PTR(-EINVAL); } /* @@ -724,19 +741,18 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) DEBUG_LOCKS_WARN_ON(!irqs_disabled()); class = look_up_lock_class(lock, subclass); - if (likely(class)) + if (likely(!IS_ERR_OR_NULL(class))) goto out_set_class_cache; /* * Debug-check: all keys must be persistent! - */ - if (!static_obj(lock->key)) { + */ + if (IS_ERR(class)) { debug_locks_off(); printk("INFO: trying to register non-static key.\n"); printk("the code is fine but needs lockdep annotation.\n"); printk("turning off the locking correctness validator.\n"); dump_stack(); - return NULL; } @@ -1141,18 +1157,18 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth, if (debug_locks_silent) return 0; - printk("\n"); - printk("======================================================\n"); - printk("[ INFO: possible circular locking dependency detected ]\n"); + pr_warn("\n"); + pr_warn("======================================================\n"); + pr_warn("WARNING: possible circular locking dependency detected\n"); print_kernel_ident(); - printk("-------------------------------------------------------\n"); - printk("%s/%d is trying to acquire lock:\n", + pr_warn("------------------------------------------------------\n"); + pr_warn("%s/%d is trying to acquire lock:\n", curr->comm, task_pid_nr(curr)); print_lock(check_src); - printk("\nbut task is already holding lock:\n"); + pr_warn("\nbut task is already holding lock:\n"); print_lock(check_tgt); - printk("\nwhich lock already depends on the new lock.\n\n"); - printk("\nthe existing dependency chain (in reverse order) is:\n"); + pr_warn("\nwhich lock already depends on the new lock.\n\n"); + pr_warn("\nthe existing dependency chain (in reverse order) is:\n"); print_circular_bug_entry(entry, depth); @@ -1479,13 +1495,13 @@ print_bad_irq_dependency(struct task_struct *curr, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n"); - printk("======================================================\n"); - printk("[ INFO: %s-safe -> %s-unsafe lock order detected ]\n", + pr_warn("\n"); + pr_warn("=====================================================\n"); + pr_warn("WARNING: %s-safe -> %s-unsafe lock order detected\n", irqclass, irqclass); print_kernel_ident(); - printk("------------------------------------------------------\n"); - printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] is trying to acquire:\n", + pr_warn("-----------------------------------------------------\n"); + pr_warn("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] is trying to acquire:\n", curr->comm, task_pid_nr(curr), curr->hardirq_context, hardirq_count() >> HARDIRQ_SHIFT, curr->softirq_context, softirq_count() >> SOFTIRQ_SHIFT, @@ -1493,46 +1509,46 @@ print_bad_irq_dependency(struct task_struct *curr, curr->softirqs_enabled); print_lock(next); - printk("\nand this task is already holding:\n"); + pr_warn("\nand this task is already holding:\n"); print_lock(prev); - printk("which would create a new lock dependency:\n"); + pr_warn("which would create a new lock dependency:\n"); print_lock_name(hlock_class(prev)); - printk(KERN_CONT " ->"); + pr_cont(" ->"); print_lock_name(hlock_class(next)); - printk(KERN_CONT "\n"); + pr_cont("\n"); - printk("\nbut this new dependency connects a %s-irq-safe lock:\n", + pr_warn("\nbut this new dependency connects a %s-irq-safe lock:\n", irqclass); print_lock_name(backwards_entry->class); - printk("\n... which became %s-irq-safe at:\n", irqclass); + pr_warn("\n... which became %s-irq-safe at:\n", irqclass); print_stack_trace(backwards_entry->class->usage_traces + bit1, 1); - printk("\nto a %s-irq-unsafe lock:\n", irqclass); + pr_warn("\nto a %s-irq-unsafe lock:\n", irqclass); print_lock_name(forwards_entry->class); - printk("\n... which became %s-irq-unsafe at:\n", irqclass); - printk("..."); + pr_warn("\n... which became %s-irq-unsafe at:\n", irqclass); + pr_warn("..."); print_stack_trace(forwards_entry->class->usage_traces + bit2, 1); - printk("\nother info that might help us debug this:\n\n"); + pr_warn("\nother info that might help us debug this:\n\n"); print_irq_lock_scenario(backwards_entry, forwards_entry, hlock_class(prev), hlock_class(next)); lockdep_print_held_locks(curr); - printk("\nthe dependencies between %s-irq-safe lock and the holding lock:\n", irqclass); + pr_warn("\nthe dependencies between %s-irq-safe lock and the holding lock:\n", irqclass); if (!save_trace(&prev_root->trace)) return 0; print_shortest_lock_dependencies(backwards_entry, prev_root); - printk("\nthe dependencies between the lock to be acquired"); - printk(" and %s-irq-unsafe lock:\n", irqclass); + pr_warn("\nthe dependencies between the lock to be acquired"); + pr_warn(" and %s-irq-unsafe lock:\n", irqclass); if (!save_trace(&next_root->trace)) return 0; print_shortest_lock_dependencies(forwards_entry, next_root); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -1708,22 +1724,22 @@ print_deadlock_bug(struct task_struct *curr, struct held_lock *prev, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n"); - printk("=============================================\n"); - printk("[ INFO: possible recursive locking detected ]\n"); + pr_warn("\n"); + pr_warn("============================================\n"); + pr_warn("WARNING: possible recursive locking detected\n"); print_kernel_ident(); - printk("---------------------------------------------\n"); - printk("%s/%d is trying to acquire lock:\n", + pr_warn("--------------------------------------------\n"); + pr_warn("%s/%d is trying to acquire lock:\n", curr->comm, task_pid_nr(curr)); print_lock(next); - printk("\nbut task is already holding lock:\n"); + pr_warn("\nbut task is already holding lock:\n"); print_lock(prev); - printk("\nother info that might help us debug this:\n"); + pr_warn("\nother info that might help us debug this:\n"); print_deadlock_scenario(next, prev); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -2058,21 +2074,21 @@ static void print_collision(struct task_struct *curr, struct held_lock *hlock_next, struct lock_chain *chain) { - printk("\n"); - printk("======================\n"); - printk("[chain_key collision ]\n"); + pr_warn("\n"); + pr_warn("============================\n"); + pr_warn("WARNING: chain_key collision\n"); print_kernel_ident(); - printk("----------------------\n"); - printk("%s/%d: ", current->comm, task_pid_nr(current)); - printk("Hash chain already cached but the contents don't match!\n"); + pr_warn("----------------------------\n"); + pr_warn("%s/%d: ", current->comm, task_pid_nr(current)); + pr_warn("Hash chain already cached but the contents don't match!\n"); - printk("Held locks:"); + pr_warn("Held locks:"); print_chain_keys_held_locks(curr, hlock_next); - printk("Locks in cached chain:"); + pr_warn("Locks in cached chain:"); print_chain_keys_chain(chain); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); } #endif @@ -2203,7 +2219,7 @@ cache_hit: * Important for check_no_collision(). */ if (unlikely(nr_chain_hlocks > MAX_LOCKDEP_CHAIN_HLOCKS)) { - if (debug_locks_off_graph_unlock()) + if (!debug_locks_off_graph_unlock()) return 0; print_lockdep_off("BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!"); @@ -2357,16 +2373,16 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n"); - printk("=================================\n"); - printk("[ INFO: inconsistent lock state ]\n"); + pr_warn("\n"); + pr_warn("================================\n"); + pr_warn("WARNING: inconsistent lock state\n"); print_kernel_ident(); - printk("---------------------------------\n"); + pr_warn("--------------------------------\n"); - printk("inconsistent {%s} -> {%s} usage.\n", + pr_warn("inconsistent {%s} -> {%s} usage.\n", usage_str[prev_bit], usage_str[new_bit]); - printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] takes:\n", + pr_warn("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] takes:\n", curr->comm, task_pid_nr(curr), trace_hardirq_context(curr), hardirq_count() >> HARDIRQ_SHIFT, trace_softirq_context(curr), softirq_count() >> SOFTIRQ_SHIFT, @@ -2374,16 +2390,16 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this, trace_softirqs_enabled(curr)); print_lock(this); - printk("{%s} state was registered at:\n", usage_str[prev_bit]); + pr_warn("{%s} state was registered at:\n", usage_str[prev_bit]); print_stack_trace(hlock_class(this)->usage_traces + prev_bit, 1); print_irqtrace_events(curr); - printk("\nother info that might help us debug this:\n"); + pr_warn("\nother info that might help us debug this:\n"); print_usage_bug_scenario(this); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -2422,28 +2438,28 @@ print_irq_inversion_bug(struct task_struct *curr, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n"); - printk("=========================================================\n"); - printk("[ INFO: possible irq lock inversion dependency detected ]\n"); + pr_warn("\n"); + pr_warn("========================================================\n"); + pr_warn("WARNING: possible irq lock inversion dependency detected\n"); print_kernel_ident(); - printk("---------------------------------------------------------\n"); - printk("%s/%d just changed the state of lock:\n", + pr_warn("--------------------------------------------------------\n"); + pr_warn("%s/%d just changed the state of lock:\n", curr->comm, task_pid_nr(curr)); print_lock(this); if (forwards) - printk("but this lock took another, %s-unsafe lock in the past:\n", irqclass); + pr_warn("but this lock took another, %s-unsafe lock in the past:\n", irqclass); else - printk("but this lock was taken by another, %s-safe lock in the past:\n", irqclass); + pr_warn("but this lock was taken by another, %s-safe lock in the past:\n", irqclass); print_lock_name(other->class); - printk("\n\nand interrupts could create inverse lock ordering between them.\n\n"); + pr_warn("\n\nand interrupts could create inverse lock ordering between them.\n\n"); - printk("\nother info that might help us debug this:\n"); + pr_warn("\nother info that might help us debug this:\n"); /* Find a middle lock (if one exists) */ depth = get_lock_depth(other); do { if (depth == 0 && (entry != root)) { - printk("lockdep:%s bad path found in chain graph\n", __func__); + pr_warn("lockdep:%s bad path found in chain graph\n", __func__); break; } middle = entry; @@ -2459,12 +2475,12 @@ print_irq_inversion_bug(struct task_struct *curr, lockdep_print_held_locks(curr); - printk("\nthe shortest dependencies between 2nd lock and 1st lock:\n"); + pr_warn("\nthe shortest dependencies between 2nd lock and 1st lock:\n"); if (!save_trace(&root->trace)) return 0; print_shortest_lock_dependencies(other, root); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -2861,6 +2877,8 @@ static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags) if (unlikely(!debug_locks)) return; + gfp_mask = current_gfp_context(gfp_mask); + /* no reclaim without waiting on it */ if (!(gfp_mask & __GFP_DIRECT_RECLAIM)) return; @@ -2870,7 +2888,7 @@ static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags) return; /* We're only interested __GFP_FS allocations for now */ - if (!(gfp_mask & __GFP_FS)) + if (!(gfp_mask & __GFP_FS) || (curr->flags & PF_MEMALLOC_NOFS)) return; /* @@ -2879,6 +2897,10 @@ static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags) if (DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))) return; + /* Disable lockdep if explicitly requested */ + if (gfp_mask & __GFP_NOLOCKDEP) + return; + mark_held_locks(curr, RECLAIM_FS); } @@ -3167,25 +3189,25 @@ print_lock_nested_lock_not_held(struct task_struct *curr, if (debug_locks_silent) return 0; - printk("\n"); - printk("==================================\n"); - printk("[ BUG: Nested lock was not taken ]\n"); + pr_warn("\n"); + pr_warn("==================================\n"); + pr_warn("WARNING: Nested lock was not taken\n"); print_kernel_ident(); - printk("----------------------------------\n"); + pr_warn("----------------------------------\n"); - printk("%s/%d is trying to lock:\n", curr->comm, task_pid_nr(curr)); + pr_warn("%s/%d is trying to lock:\n", curr->comm, task_pid_nr(curr)); print_lock(hlock); - printk("\nbut this task is not holding:\n"); - printk("%s\n", hlock->nest_lock->name); + pr_warn("\nbut this task is not holding:\n"); + pr_warn("%s\n", hlock->nest_lock->name); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); - printk("\nother info that might help us debug this:\n"); + pr_warn("\nother info that might help us debug this:\n"); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -3260,10 +3282,17 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (depth) { hlock = curr->held_locks + depth - 1; if (hlock->class_idx == class_idx && nest_lock) { - if (hlock->references) + if (hlock->references) { + /* + * Check: unsigned int references:12, overflow. + */ + if (DEBUG_LOCKS_WARN_ON(hlock->references == (1 << 12)-1)) + return 0; + hlock->references++; - else + } else { hlock->references = 2; + } return 1; } @@ -3373,21 +3402,21 @@ print_unlock_imbalance_bug(struct task_struct *curr, struct lockdep_map *lock, if (debug_locks_silent) return 0; - printk("\n"); - printk("=====================================\n"); - printk("[ BUG: bad unlock balance detected! ]\n"); + pr_warn("\n"); + pr_warn("=====================================\n"); + pr_warn("WARNING: bad unlock balance detected!\n"); print_kernel_ident(); - printk("-------------------------------------\n"); - printk("%s/%d is trying to release lock (", + pr_warn("-------------------------------------\n"); + pr_warn("%s/%d is trying to release lock (", curr->comm, task_pid_nr(curr)); print_lockdep_cache(lock); - printk(KERN_CONT ") at:\n"); + pr_cont(") at:\n"); print_ip_sym(ip); - printk("but there are no more locks to release!\n"); - printk("\nother info that might help us debug this:\n"); + pr_warn("but there are no more locks to release!\n"); + pr_warn("\nother info that might help us debug this:\n"); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -3410,7 +3439,7 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) * Clearly if the lock hasn't been acquired _ever_, we're not * holding it either, so report failure. */ - if (!class) + if (IS_ERR_OR_NULL(class)) return 0; /* @@ -3428,13 +3457,67 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) return 0; } +/* @depth must not be zero */ +static struct held_lock *find_held_lock(struct task_struct *curr, + struct lockdep_map *lock, + unsigned int depth, int *idx) +{ + struct held_lock *ret, *hlock, *prev_hlock; + int i; + + i = depth - 1; + hlock = curr->held_locks + i; + ret = hlock; + if (match_held_lock(hlock, lock)) + goto out; + + ret = NULL; + for (i--, prev_hlock = hlock--; + i >= 0; + i--, prev_hlock = hlock--) { + /* + * We must not cross into another context: + */ + if (prev_hlock->irq_context != hlock->irq_context) { + ret = NULL; + break; + } + if (match_held_lock(hlock, lock)) { + ret = hlock; + break; + } + } + +out: + *idx = i; + return ret; +} + +static int reacquire_held_locks(struct task_struct *curr, unsigned int depth, + int idx) +{ + struct held_lock *hlock; + + for (hlock = curr->held_locks + idx; idx < depth; idx++, hlock++) { + if (!__lock_acquire(hlock->instance, + hlock_class(hlock)->subclass, + hlock->trylock, + hlock->read, hlock->check, + hlock->hardirqs_off, + hlock->nest_lock, hlock->acquire_ip, + hlock->references, hlock->pin_count)) + return 1; + } + return 0; +} + static int __lock_set_class(struct lockdep_map *lock, const char *name, struct lock_class_key *key, unsigned int subclass, unsigned long ip) { struct task_struct *curr = current; - struct held_lock *hlock, *prev_hlock; + struct held_lock *hlock; struct lock_class *class; unsigned int depth; int i; @@ -3447,21 +3530,10 @@ __lock_set_class(struct lockdep_map *lock, const char *name, if (DEBUG_LOCKS_WARN_ON(!depth)) return 0; - prev_hlock = NULL; - for (i = depth-1; i >= 0; i--) { - hlock = curr->held_locks + i; - /* - * We must not cross into another context: - */ - if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) - break; - if (match_held_lock(hlock, lock)) - goto found_it; - prev_hlock = hlock; - } - return print_unlock_imbalance_bug(curr, lock, ip); + hlock = find_held_lock(curr, lock, depth, &i); + if (!hlock) + return print_unlock_imbalance_bug(curr, lock, ip); -found_it: lockdep_init_map(lock, name, key, 0); class = register_lock_class(lock, subclass, 0); hlock->class_idx = class - lock_classes + 1; @@ -3469,15 +3541,46 @@ found_it: curr->lockdep_depth = i; curr->curr_chain_key = hlock->prev_chain_key; - for (; i < depth; i++) { - hlock = curr->held_locks + i; - if (!__lock_acquire(hlock->instance, - hlock_class(hlock)->subclass, hlock->trylock, - hlock->read, hlock->check, hlock->hardirqs_off, - hlock->nest_lock, hlock->acquire_ip, - hlock->references, hlock->pin_count)) - return 0; - } + if (reacquire_held_locks(curr, depth, i)) + return 0; + + /* + * I took it apart and put it back together again, except now I have + * these 'spare' parts.. where shall I put them. + */ + if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth)) + return 0; + return 1; +} + +static int __lock_downgrade(struct lockdep_map *lock, unsigned long ip) +{ + struct task_struct *curr = current; + struct held_lock *hlock; + unsigned int depth; + int i; + + depth = curr->lockdep_depth; + /* + * This function is about (re)setting the class of a held lock, + * yet we're not actually holding any locks. Naughty user! + */ + if (DEBUG_LOCKS_WARN_ON(!depth)) + return 0; + + hlock = find_held_lock(curr, lock, depth, &i); + if (!hlock) + return print_unlock_imbalance_bug(curr, lock, ip); + + curr->lockdep_depth = i; + curr->curr_chain_key = hlock->prev_chain_key; + + WARN(hlock->read, "downgrading a read lock"); + hlock->read = 1; + hlock->acquire_ip = ip; + + if (reacquire_held_locks(curr, depth, i)) + return 0; /* * I took it apart and put it back together again, except now I have @@ -3499,7 +3602,7 @@ static int __lock_release(struct lockdep_map *lock, int nested, unsigned long ip) { struct task_struct *curr = current; - struct held_lock *hlock, *prev_hlock; + struct held_lock *hlock; unsigned int depth; int i; @@ -3518,21 +3621,10 @@ __lock_release(struct lockdep_map *lock, int nested, unsigned long ip) * Check whether the lock exists in the current stack * of held locks: */ - prev_hlock = NULL; - for (i = depth-1; i >= 0; i--) { - hlock = curr->held_locks + i; - /* - * We must not cross into another context: - */ - if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) - break; - if (match_held_lock(hlock, lock)) - goto found_it; - prev_hlock = hlock; - } - return print_unlock_imbalance_bug(curr, lock, ip); + hlock = find_held_lock(curr, lock, depth, &i); + if (!hlock) + return print_unlock_imbalance_bug(curr, lock, ip); -found_it: if (hlock->instance == lock) lock_release_holdtime(hlock); @@ -3559,15 +3651,8 @@ found_it: curr->lockdep_depth = i; curr->curr_chain_key = hlock->prev_chain_key; - for (i++; i < depth; i++) { - hlock = curr->held_locks + i; - if (!__lock_acquire(hlock->instance, - hlock_class(hlock)->subclass, hlock->trylock, - hlock->read, hlock->check, hlock->hardirqs_off, - hlock->nest_lock, hlock->acquire_ip, - hlock->references, hlock->pin_count)) - return 0; - } + if (reacquire_held_locks(curr, depth, i + 1)) + return 0; /* * We had N bottles of beer on the wall, we drank one, but now @@ -3732,6 +3817,23 @@ void lock_set_class(struct lockdep_map *lock, const char *name, } EXPORT_SYMBOL_GPL(lock_set_class); +void lock_downgrade(struct lockdep_map *lock, unsigned long ip) +{ + unsigned long flags; + + if (unlikely(current->lockdep_recursion)) + return; + + raw_local_irq_save(flags); + current->lockdep_recursion = 1; + check_flags(flags); + if (__lock_downgrade(lock, ip)) + check_chain_key(current); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(lock_downgrade); + /* * We are not always called with irqs disabled - do that here, * and also avoid lockdep recursion: @@ -3852,13 +3954,15 @@ EXPORT_SYMBOL_GPL(lock_unpin_lock); void lockdep_set_current_reclaim_state(gfp_t gfp_mask) { - current->lockdep_reclaim_gfp = gfp_mask; + current->lockdep_reclaim_gfp = current_gfp_context(gfp_mask); } +EXPORT_SYMBOL_GPL(lockdep_set_current_reclaim_state); void lockdep_clear_current_reclaim_state(void) { current->lockdep_reclaim_gfp = 0; } +EXPORT_SYMBOL_GPL(lockdep_clear_current_reclaim_state); #ifdef CONFIG_LOCK_STAT static int @@ -3870,21 +3974,21 @@ print_lock_contention_bug(struct task_struct *curr, struct lockdep_map *lock, if (debug_locks_silent) return 0; - printk("\n"); - printk("=================================\n"); - printk("[ BUG: bad contention detected! ]\n"); + pr_warn("\n"); + pr_warn("=================================\n"); + pr_warn("WARNING: bad contention detected!\n"); print_kernel_ident(); - printk("---------------------------------\n"); - printk("%s/%d is trying to contend lock (", + pr_warn("---------------------------------\n"); + pr_warn("%s/%d is trying to contend lock (", curr->comm, task_pid_nr(curr)); print_lockdep_cache(lock); - printk(KERN_CONT ") at:\n"); + pr_cont(") at:\n"); print_ip_sym(ip); - printk("but there are no locks held!\n"); - printk("\nother info that might help us debug this:\n"); + pr_warn("but there are no locks held!\n"); + pr_warn("\nother info that might help us debug this:\n"); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); return 0; @@ -3894,7 +3998,7 @@ static void __lock_contended(struct lockdep_map *lock, unsigned long ip) { struct task_struct *curr = current; - struct held_lock *hlock, *prev_hlock; + struct held_lock *hlock; struct lock_class_stats *stats; unsigned int depth; int i, contention_point, contending_point; @@ -3907,22 +4011,12 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) if (DEBUG_LOCKS_WARN_ON(!depth)) return; - prev_hlock = NULL; - for (i = depth-1; i >= 0; i--) { - hlock = curr->held_locks + i; - /* - * We must not cross into another context: - */ - if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) - break; - if (match_held_lock(hlock, lock)) - goto found_it; - prev_hlock = hlock; + hlock = find_held_lock(curr, lock, depth, &i); + if (!hlock) { + print_lock_contention_bug(curr, lock, ip); + return; } - print_lock_contention_bug(curr, lock, ip); - return; -found_it: if (hlock->instance != lock) return; @@ -3946,7 +4040,7 @@ static void __lock_acquired(struct lockdep_map *lock, unsigned long ip) { struct task_struct *curr = current; - struct held_lock *hlock, *prev_hlock; + struct held_lock *hlock; struct lock_class_stats *stats; unsigned int depth; u64 now, waittime = 0; @@ -3960,22 +4054,12 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) if (DEBUG_LOCKS_WARN_ON(!depth)) return; - prev_hlock = NULL; - for (i = depth-1; i >= 0; i--) { - hlock = curr->held_locks + i; - /* - * We must not cross into another context: - */ - if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) - break; - if (match_held_lock(hlock, lock)) - goto found_it; - prev_hlock = hlock; + hlock = find_held_lock(curr, lock, depth, &i); + if (!hlock) { + print_lock_contention_bug(curr, lock, _RET_IP_); + return; } - print_lock_contention_bug(curr, lock, _RET_IP_); - return; -found_it: if (hlock->instance != lock) return; @@ -4163,7 +4247,7 @@ void lockdep_reset_lock(struct lockdep_map *lock) * If the class exists we look it up and zap it: */ class = look_up_lock_class(lock, j); - if (class) + if (!IS_ERR_OR_NULL(class)) zap_class(class); } /* @@ -4234,17 +4318,17 @@ print_freed_lock_bug(struct task_struct *curr, const void *mem_from, if (debug_locks_silent) return; - printk("\n"); - printk("=========================\n"); - printk("[ BUG: held lock freed! ]\n"); + pr_warn("\n"); + pr_warn("=========================\n"); + pr_warn("WARNING: held lock freed!\n"); print_kernel_ident(); - printk("-------------------------\n"); - printk("%s/%d is freeing memory %p-%p, with a lock still held there!\n", + pr_warn("-------------------------\n"); + pr_warn("%s/%d is freeing memory %p-%p, with a lock still held there!\n", curr->comm, task_pid_nr(curr), mem_from, mem_to-1); print_lock(hlock); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); } @@ -4292,14 +4376,14 @@ static void print_held_locks_bug(void) if (debug_locks_silent) return; - printk("\n"); - printk("=====================================\n"); - printk("[ BUG: %s/%d still has locks held! ]\n", + pr_warn("\n"); + pr_warn("====================================\n"); + pr_warn("WARNING: %s/%d still has locks held!\n", current->comm, task_pid_nr(current)); print_kernel_ident(); - printk("-------------------------------------\n"); + pr_warn("------------------------------------\n"); lockdep_print_held_locks(current); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); } @@ -4318,10 +4402,10 @@ void debug_show_all_locks(void) int unlock = 1; if (unlikely(!debug_locks)) { - printk("INFO: lockdep is turned off.\n"); + pr_warn("INFO: lockdep is turned off.\n"); return; } - printk("\nShowing all locks held in the system:\n"); + pr_warn("\nShowing all locks held in the system:\n"); /* * Here we try to get the tasklist_lock as hard as possible, @@ -4332,18 +4416,18 @@ void debug_show_all_locks(void) retry: if (!read_trylock(&tasklist_lock)) { if (count == 10) - printk("hm, tasklist_lock locked, retrying... "); + pr_warn("hm, tasklist_lock locked, retrying... "); if (count) { count--; - printk(" #%d", 10-count); + pr_cont(" #%d", 10-count); mdelay(200); goto retry; } - printk(" ignoring it.\n"); + pr_cont(" ignoring it.\n"); unlock = 0; } else { if (count != 10) - printk(KERN_CONT " locked it.\n"); + pr_cont(" locked it.\n"); } do_each_thread(g, p) { @@ -4361,8 +4445,8 @@ retry: unlock = 1; } while_each_thread(g, p); - printk("\n"); - printk("=============================================\n\n"); + pr_warn("\n"); + pr_warn("=============================================\n\n"); if (unlock) read_unlock(&tasklist_lock); @@ -4391,12 +4475,12 @@ asmlinkage __visible void lockdep_sys_exit(void) if (unlikely(curr->lockdep_depth)) { if (!debug_locks_off()) return; - printk("\n"); - printk("================================================\n"); - printk("[ BUG: lock held when returning to user space! ]\n"); + pr_warn("\n"); + pr_warn("================================================\n"); + pr_warn("WARNING: lock held when returning to user space!\n"); print_kernel_ident(); - printk("------------------------------------------------\n"); - printk("%s/%d is leaving the kernel with locks still held!\n", + pr_warn("------------------------------------------------\n"); + pr_warn("%s/%d is leaving the kernel with locks still held!\n", curr->comm, curr->pid); lockdep_print_held_locks(curr); } @@ -4406,19 +4490,15 @@ void lockdep_rcu_suspicious(const char *file, const int line, const char *s) { struct task_struct *curr = current; -#ifndef CONFIG_PROVE_RCU_REPEATEDLY - if (!debug_locks_off()) - return; -#endif /* #ifdef CONFIG_PROVE_RCU_REPEATEDLY */ /* Note: the following can be executed concurrently, so be careful. */ - printk("\n"); - printk("===============================\n"); - printk("[ INFO: suspicious RCU usage. ]\n"); + pr_warn("\n"); + pr_warn("=============================\n"); + pr_warn("WARNING: suspicious RCU usage\n"); print_kernel_ident(); - printk("-------------------------------\n"); - printk("%s:%d %s!\n", file, line, s); - printk("\nother info that might help us debug this:\n\n"); - printk("\n%srcu_scheduler_active = %d, debug_locks = %d\n", + pr_warn("-----------------------------\n"); + pr_warn("%s:%d %s!\n", file, line, s); + pr_warn("\nother info that might help us debug this:\n\n"); + pr_warn("\n%srcu_scheduler_active = %d, debug_locks = %d\n", !rcu_lockdep_current_cpu_online() ? "RCU used illegally from offline CPU!\n" : !rcu_is_watching() @@ -4445,10 +4525,10 @@ void lockdep_rcu_suspicious(const char *file, const int line, const char *s) * rcu_read_lock_bh() and so on from extended quiescent states. */ if (!rcu_is_watching()) - printk("RCU used illegally from extended quiescent state!\n"); + pr_warn("RCU used illegally from extended quiescent state!\n"); lockdep_print_held_locks(curr); - printk("\nstack backtrace:\n"); + pr_warn("\nstack backtrace:\n"); dump_stack(); } EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious); diff --git a/kernel/locking/lockdep_internals.h b/kernel/locking/lockdep_internals.h index c2b88490d857..c08fbd2f5ba9 100644 --- a/kernel/locking/lockdep_internals.h +++ b/kernel/locking/lockdep_internals.h @@ -46,13 +46,13 @@ enum { (LOCKF_USED_IN_HARDIRQ_READ | LOCKF_USED_IN_SOFTIRQ_READ) /* - * CONFIG_PROVE_LOCKING_SMALL is defined for sparc. Sparc requires .text, + * CONFIG_LOCKDEP_SMALL is defined for sparc. Sparc requires .text, * .data and .bss to fit in required 32MB limit for the kernel. With - * PROVE_LOCKING we could go over this limit and cause system boot-up problems. + * CONFIG_LOCKDEP we could go over this limit and cause system boot-up problems. * So, reduce the static allocations for lockdeps related structures so that * everything fits in current required size limit. */ -#ifdef CONFIG_PROVE_LOCKING_SMALL +#ifdef CONFIG_LOCKDEP_SMALL /* * MAX_LOCKDEP_ENTRIES is the maximum number of lock dependencies * we track. diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c index f8c5af52a131..f24582d4dad3 100644 --- a/kernel/locking/locktorture.c +++ b/kernel/locking/locktorture.c @@ -32,6 +32,8 @@ #include <linux/smp.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <uapi/linux/sched/types.h> +#include <linux/rtmutex.h> #include <linux/atomic.h> #include <linux/moduleparam.h> #include <linux/delay.h> @@ -372,6 +374,78 @@ static struct lock_torture_ops mutex_lock_ops = { .name = "mutex_lock" }; +#include <linux/ww_mutex.h> +static DEFINE_WW_CLASS(torture_ww_class); +static DEFINE_WW_MUTEX(torture_ww_mutex_0, &torture_ww_class); +static DEFINE_WW_MUTEX(torture_ww_mutex_1, &torture_ww_class); +static DEFINE_WW_MUTEX(torture_ww_mutex_2, &torture_ww_class); + +static int torture_ww_mutex_lock(void) +__acquires(torture_ww_mutex_0) +__acquires(torture_ww_mutex_1) +__acquires(torture_ww_mutex_2) +{ + LIST_HEAD(list); + struct reorder_lock { + struct list_head link; + struct ww_mutex *lock; + } locks[3], *ll, *ln; + struct ww_acquire_ctx ctx; + + locks[0].lock = &torture_ww_mutex_0; + list_add(&locks[0].link, &list); + + locks[1].lock = &torture_ww_mutex_1; + list_add(&locks[1].link, &list); + + locks[2].lock = &torture_ww_mutex_2; + list_add(&locks[2].link, &list); + + ww_acquire_init(&ctx, &torture_ww_class); + + list_for_each_entry(ll, &list, link) { + int err; + + err = ww_mutex_lock(ll->lock, &ctx); + if (!err) + continue; + + ln = ll; + list_for_each_entry_continue_reverse(ln, &list, link) + ww_mutex_unlock(ln->lock); + + if (err != -EDEADLK) + return err; + + ww_mutex_lock_slow(ll->lock, &ctx); + list_move(&ll->link, &list); + } + + ww_acquire_fini(&ctx); + return 0; +} + +static void torture_ww_mutex_unlock(void) +__releases(torture_ww_mutex_0) +__releases(torture_ww_mutex_1) +__releases(torture_ww_mutex_2) +{ + ww_mutex_unlock(&torture_ww_mutex_0); + ww_mutex_unlock(&torture_ww_mutex_1); + ww_mutex_unlock(&torture_ww_mutex_2); +} + +static struct lock_torture_ops ww_mutex_lock_ops = { + .writelock = torture_ww_mutex_lock, + .write_delay = torture_mutex_delay, + .task_boost = torture_boost_dummy, + .writeunlock = torture_ww_mutex_unlock, + .readlock = NULL, + .read_delay = NULL, + .readunlock = NULL, + .name = "ww_mutex_lock" +}; + #ifdef CONFIG_RT_MUTEXES static DEFINE_RT_MUTEX(torture_rtmutex); @@ -780,6 +854,10 @@ static void lock_torture_cleanup(void) else lock_torture_print_module_parms(cxt.cur_ops, "End of test: SUCCESS"); + + kfree(cxt.lwsa); + kfree(cxt.lrsa); + end: torture_cleanup_end(); } @@ -793,6 +871,7 @@ static int __init lock_torture_init(void) &spin_lock_ops, &spin_lock_irq_ops, &rw_lock_ops, &rw_lock_irq_ops, &mutex_lock_ops, + &ww_mutex_lock_ops, #ifdef CONFIG_RT_MUTEXES &rtmutex_lock_ops, #endif @@ -924,6 +1003,8 @@ static int __init lock_torture_init(void) GFP_KERNEL); if (reader_tasks == NULL) { VERBOSE_TOROUT_ERRSTRING("reader_tasks: Out of memory"); + kfree(writer_tasks); + writer_tasks = NULL; firsterr = -ENOMEM; goto unwind; } diff --git a/kernel/locking/mutex-debug.h b/kernel/locking/mutex-debug.h index a459faa48987..4174417d5309 100644 --- a/kernel/locking/mutex-debug.h +++ b/kernel/locking/mutex-debug.h @@ -26,20 +26,3 @@ extern void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, extern void debug_mutex_unlock(struct mutex *lock); extern void debug_mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key); - -#define spin_lock_mutex(lock, flags) \ - do { \ - struct mutex *l = container_of(lock, struct mutex, wait_lock); \ - \ - DEBUG_LOCKS_WARN_ON(in_interrupt()); \ - local_irq_save(flags); \ - arch_spin_lock(&(lock)->rlock.raw_lock);\ - DEBUG_LOCKS_WARN_ON(l->magic != l); \ - } while (0) - -#define spin_unlock_mutex(lock, flags) \ - do { \ - arch_spin_unlock(&(lock)->rlock.raw_lock); \ - local_irq_restore(flags); \ - preempt_check_resched(); \ - } while (0) diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c index 9b349619f431..858a07590e39 100644 --- a/kernel/locking/mutex.c +++ b/kernel/locking/mutex.c @@ -19,8 +19,10 @@ */ #include <linux/mutex.h> #include <linux/ww_mutex.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> #include <linux/sched/rt.h> +#include <linux/sched/wake_q.h> +#include <linux/sched/debug.h> #include <linux/export.h> #include <linux/spinlock.h> #include <linux/interrupt.h> @@ -50,16 +52,17 @@ EXPORT_SYMBOL(__mutex_init); /* * @owner: contains: 'struct task_struct *' to the current lock owner, * NULL means not owned. Since task_struct pointers are aligned at - * ARCH_MIN_TASKALIGN (which is at least sizeof(void *)), we have low - * bits to store extra state. + * at least L1_CACHE_BYTES, we have low bits to store extra state. * * Bit0 indicates a non-empty waiter list; unlock must issue a wakeup. * Bit1 indicates unlock needs to hand the lock to the top-waiter + * Bit2 indicates handoff has been done and we're waiting for pickup. */ #define MUTEX_FLAG_WAITERS 0x01 #define MUTEX_FLAG_HANDOFF 0x02 +#define MUTEX_FLAG_PICKUP 0x04 -#define MUTEX_FLAGS 0x03 +#define MUTEX_FLAGS 0x07 static inline struct task_struct *__owner_task(unsigned long owner) { @@ -72,38 +75,29 @@ static inline unsigned long __owner_flags(unsigned long owner) } /* - * Actual trylock that will work on any unlocked state. - * - * When setting the owner field, we must preserve the low flag bits. - * - * Be careful with @handoff, only set that in a wait-loop (where you set - * HANDOFF) to avoid recursive lock attempts. + * Trylock variant that retuns the owning task on failure. */ -static inline bool __mutex_trylock(struct mutex *lock, const bool handoff) +static inline struct task_struct *__mutex_trylock_or_owner(struct mutex *lock) { unsigned long owner, curr = (unsigned long)current; owner = atomic_long_read(&lock->owner); for (;;) { /* must loop, can race against a flag */ unsigned long old, flags = __owner_flags(owner); + unsigned long task = owner & ~MUTEX_FLAGS; - if (__owner_task(owner)) { - if (handoff && unlikely(__owner_task(owner) == current)) { - /* - * Provide ACQUIRE semantics for the lock-handoff. - * - * We cannot easily use load-acquire here, since - * the actual load is a failed cmpxchg, which - * doesn't imply any barriers. - * - * Also, this is a fairly unlikely scenario, and - * this contains the cost. - */ - smp_mb(); /* ACQUIRE */ - return true; - } + if (task) { + if (likely(task != curr)) + break; - return false; + if (likely(!(flags & MUTEX_FLAG_PICKUP))) + break; + + flags &= ~MUTEX_FLAG_PICKUP; + } else { +#ifdef CONFIG_DEBUG_MUTEXES + DEBUG_LOCKS_WARN_ON(flags & MUTEX_FLAG_PICKUP); +#endif } /* @@ -111,15 +105,24 @@ static inline bool __mutex_trylock(struct mutex *lock, const bool handoff) * past the point where we acquire it. This would be possible * if we (accidentally) set the bit on an unlocked mutex. */ - if (handoff) - flags &= ~MUTEX_FLAG_HANDOFF; + flags &= ~MUTEX_FLAG_HANDOFF; old = atomic_long_cmpxchg_acquire(&lock->owner, owner, curr | flags); if (old == owner) - return true; + return NULL; owner = old; } + + return __owner_task(owner); +} + +/* + * Actual trylock that will work on any unlocked state. + */ +static inline bool __mutex_trylock(struct mutex *lock) +{ + return !__mutex_trylock_or_owner(lock); } #ifndef CONFIG_DEBUG_LOCK_ALLOC @@ -171,9 +174,9 @@ static inline bool __mutex_waiter_is_first(struct mutex *lock, struct mutex_wait /* * Give up ownership to a specific task, when @task = NULL, this is equivalent - * to a regular unlock. Clears HANDOFF, preserves WAITERS. Provides RELEASE - * semantics like a regular unlock, the __mutex_trylock() provides matching - * ACQUIRE semantics for the handoff. + * to a regular unlock. Sets PICKUP on a handoff, clears HANDOF, preserves + * WAITERS. Provides RELEASE semantics like a regular unlock, the + * __mutex_trylock() provides a matching ACQUIRE semantics for the handoff. */ static void __mutex_handoff(struct mutex *lock, struct task_struct *task) { @@ -184,10 +187,13 @@ static void __mutex_handoff(struct mutex *lock, struct task_struct *task) #ifdef CONFIG_DEBUG_MUTEXES DEBUG_LOCKS_WARN_ON(__owner_task(owner) != current); + DEBUG_LOCKS_WARN_ON(owner & MUTEX_FLAG_PICKUP); #endif new = (owner & MUTEX_FLAG_WAITERS); new |= (unsigned long)task; + if (task) + new |= MUTEX_FLAG_PICKUP; old = atomic_long_cmpxchg_release(&lock->owner, owner, new); if (old == owner) @@ -221,9 +227,9 @@ static void __sched __mutex_lock_slowpath(struct mutex *lock); * (or statically defined) before it can be locked. memset()-ing * the mutex to 0 is not allowed. * - * ( The CONFIG_DEBUG_MUTEXES .config option turns on debugging - * checks that will enforce the restrictions and will also do - * deadlock debugging. ) + * (The CONFIG_DEBUG_MUTEXES .config option turns on debugging + * checks that will enforce the restrictions and will also do + * deadlock debugging) * * This function is similar to (but not equivalent to) down(). */ @@ -237,8 +243,8 @@ void __sched mutex_lock(struct mutex *lock) EXPORT_SYMBOL(mutex_lock); #endif -static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww, - struct ww_acquire_ctx *ww_ctx) +static __always_inline void +ww_mutex_lock_acquired(struct ww_mutex *ww, struct ww_acquire_ctx *ww_ctx) { #ifdef CONFIG_DEBUG_MUTEXES /* @@ -277,17 +283,50 @@ static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww, ww_ctx->acquired++; } +static inline bool __sched +__ww_ctx_stamp_after(struct ww_acquire_ctx *a, struct ww_acquire_ctx *b) +{ + return a->stamp - b->stamp <= LONG_MAX && + (a->stamp != b->stamp || a > b); +} + +/* + * Wake up any waiters that may have to back off when the lock is held by the + * given context. + * + * Due to the invariants on the wait list, this can only affect the first + * waiter with a context. + * + * The current task must not be on the wait list. + */ +static void __sched +__ww_mutex_wakeup_for_backoff(struct mutex *lock, struct ww_acquire_ctx *ww_ctx) +{ + struct mutex_waiter *cur; + + lockdep_assert_held(&lock->wait_lock); + + list_for_each_entry(cur, &lock->wait_list, list) { + if (!cur->ww_ctx) + continue; + + if (cur->ww_ctx->acquired > 0 && + __ww_ctx_stamp_after(cur->ww_ctx, ww_ctx)) { + debug_mutex_wake_waiter(lock, cur); + wake_up_process(cur->task); + } + + break; + } +} + /* * After acquiring lock with fastpath or when we lost out in contested * slowpath, set ctx and wake up any waiters so they can recheck. */ static __always_inline void -ww_mutex_set_context_fastpath(struct ww_mutex *lock, - struct ww_acquire_ctx *ctx) +ww_mutex_set_context_fastpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { - unsigned long flags; - struct mutex_waiter *cur; - ww_mutex_lock_acquired(lock, ctx); lock->ctx = ctx; @@ -311,46 +350,79 @@ ww_mutex_set_context_fastpath(struct ww_mutex *lock, * Uh oh, we raced in fastpath, wake up everyone in this case, * so they can see the new lock->ctx. */ - spin_lock_mutex(&lock->base.wait_lock, flags); - list_for_each_entry(cur, &lock->base.wait_list, list) { - debug_mutex_wake_waiter(&lock->base, cur); - wake_up_process(cur->task); - } - spin_unlock_mutex(&lock->base.wait_lock, flags); + spin_lock(&lock->base.wait_lock); + __ww_mutex_wakeup_for_backoff(&lock->base, ctx); + spin_unlock(&lock->base.wait_lock); } /* - * After acquiring lock in the slowpath set ctx and wake up any - * waiters so they can recheck. + * After acquiring lock in the slowpath set ctx. + * + * Unlike for the fast path, the caller ensures that waiters are woken up where + * necessary. * * Callers must hold the mutex wait_lock. */ static __always_inline void -ww_mutex_set_context_slowpath(struct ww_mutex *lock, - struct ww_acquire_ctx *ctx) +ww_mutex_set_context_slowpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { - struct mutex_waiter *cur; - ww_mutex_lock_acquired(lock, ctx); lock->ctx = ctx; +} + +#ifdef CONFIG_MUTEX_SPIN_ON_OWNER + +static inline +bool ww_mutex_spin_on_owner(struct mutex *lock, struct ww_acquire_ctx *ww_ctx, + struct mutex_waiter *waiter) +{ + struct ww_mutex *ww; + + ww = container_of(lock, struct ww_mutex, base); /* - * Give any possible sleeping processes the chance to wake up, - * so they can recheck if they have to back off. + * If ww->ctx is set the contents are undefined, only + * by acquiring wait_lock there is a guarantee that + * they are not invalid when reading. + * + * As such, when deadlock detection needs to be + * performed the optimistic spinning cannot be done. + * + * Check this in every inner iteration because we may + * be racing against another thread's ww_mutex_lock. */ - list_for_each_entry(cur, &lock->base.wait_list, list) { - debug_mutex_wake_waiter(&lock->base, cur); - wake_up_process(cur->task); - } + if (ww_ctx->acquired > 0 && READ_ONCE(ww->ctx)) + return false; + + /* + * If we aren't on the wait list yet, cancel the spin + * if there are waiters. We want to avoid stealing the + * lock from a waiter with an earlier stamp, since the + * other thread may already own a lock that we also + * need. + */ + if (!waiter && (atomic_long_read(&lock->owner) & MUTEX_FLAG_WAITERS)) + return false; + + /* + * Similarly, stop spinning if we are no longer the + * first waiter. + */ + if (waiter && !__mutex_waiter_is_first(lock, waiter)) + return false; + + return true; } -#ifdef CONFIG_MUTEX_SPIN_ON_OWNER /* - * Look out! "owner" is an entirely speculative pointer - * access and not reliable. + * Look out! "owner" is an entirely speculative pointer access and not + * reliable. + * + * "noinline" so that this function shows up on perf profiles. */ static noinline -bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) +bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner, + struct ww_acquire_ctx *ww_ctx, struct mutex_waiter *waiter) { bool ret = true; @@ -373,6 +445,11 @@ bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) break; } + if (ww_ctx && !ww_mutex_spin_on_owner(lock, ww_ctx, waiter)) { + ret = false; + break; + } + cpu_relax(); } rcu_read_unlock(); @@ -431,12 +508,10 @@ static inline int mutex_can_spin_on_owner(struct mutex *lock) * with the spinner at the head of the OSQ, if present, until the owner is * changed to itself. */ -static bool mutex_optimistic_spin(struct mutex *lock, - struct ww_acquire_ctx *ww_ctx, - const bool use_ww_ctx, const bool waiter) +static __always_inline bool +mutex_optimistic_spin(struct mutex *lock, struct ww_acquire_ctx *ww_ctx, + const bool use_ww_ctx, struct mutex_waiter *waiter) { - struct task_struct *task = current; - if (!waiter) { /* * The purpose of the mutex_can_spin_on_owner() function is @@ -460,40 +535,17 @@ static bool mutex_optimistic_spin(struct mutex *lock, for (;;) { struct task_struct *owner; - if (use_ww_ctx && ww_ctx->acquired > 0) { - struct ww_mutex *ww; - - ww = container_of(lock, struct ww_mutex, base); - /* - * If ww->ctx is set the contents are undefined, only - * by acquiring wait_lock there is a guarantee that - * they are not invalid when reading. - * - * As such, when deadlock detection needs to be - * performed the optimistic spinning cannot be done. - */ - if (READ_ONCE(ww->ctx)) - goto fail_unlock; - } + /* Try to acquire the mutex... */ + owner = __mutex_trylock_or_owner(lock); + if (!owner) + break; /* - * If there's an owner, wait for it to either + * There's an owner, wait for it to either * release the lock or go to sleep. */ - owner = __mutex_owner(lock); - if (owner) { - if (waiter && owner == task) { - smp_mb(); /* ACQUIRE */ - break; - } - - if (!mutex_spin_on_owner(lock, owner)) - goto fail_unlock; - } - - /* Try to acquire the mutex if it is unlocked. */ - if (__mutex_trylock(lock, waiter)) - break; + if (!mutex_spin_on_owner(lock, owner, ww_ctx, waiter)) + goto fail_unlock; /* * The cpu_relax() call is a compiler barrier which forces @@ -532,9 +584,9 @@ fail: return false; } #else -static bool mutex_optimistic_spin(struct mutex *lock, - struct ww_acquire_ctx *ww_ctx, - const bool use_ww_ctx, const bool waiter) +static __always_inline bool +mutex_optimistic_spin(struct mutex *lock, struct ww_acquire_ctx *ww_ctx, + const bool use_ww_ctx, struct mutex_waiter *waiter) { return false; } @@ -594,23 +646,88 @@ void __sched ww_mutex_unlock(struct ww_mutex *lock) EXPORT_SYMBOL(ww_mutex_unlock); static inline int __sched -__ww_mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx) +__ww_mutex_lock_check_stamp(struct mutex *lock, struct mutex_waiter *waiter, + struct ww_acquire_ctx *ctx) { struct ww_mutex *ww = container_of(lock, struct ww_mutex, base); struct ww_acquire_ctx *hold_ctx = READ_ONCE(ww->ctx); + struct mutex_waiter *cur; + + if (hold_ctx && __ww_ctx_stamp_after(ctx, hold_ctx)) + goto deadlock; + + /* + * If there is a waiter in front of us that has a context, then its + * stamp is earlier than ours and we must back off. + */ + cur = waiter; + list_for_each_entry_continue_reverse(cur, &lock->wait_list, list) { + if (cur->ww_ctx) + goto deadlock; + } + + return 0; - if (!hold_ctx) +deadlock: +#ifdef CONFIG_DEBUG_MUTEXES + DEBUG_LOCKS_WARN_ON(ctx->contending_lock); + ctx->contending_lock = ww; +#endif + return -EDEADLK; +} + +static inline int __sched +__ww_mutex_add_waiter(struct mutex_waiter *waiter, + struct mutex *lock, + struct ww_acquire_ctx *ww_ctx) +{ + struct mutex_waiter *cur; + struct list_head *pos; + + if (!ww_ctx) { + list_add_tail(&waiter->list, &lock->wait_list); return 0; + } - if (ctx->stamp - hold_ctx->stamp <= LONG_MAX && - (ctx->stamp != hold_ctx->stamp || ctx > hold_ctx)) { + /* + * Add the waiter before the first waiter with a higher stamp. + * Waiters without a context are skipped to avoid starving + * them. + */ + pos = &lock->wait_list; + list_for_each_entry_reverse(cur, &lock->wait_list, list) { + if (!cur->ww_ctx) + continue; + + if (__ww_ctx_stamp_after(ww_ctx, cur->ww_ctx)) { + /* Back off immediately if necessary. */ + if (ww_ctx->acquired > 0) { #ifdef CONFIG_DEBUG_MUTEXES - DEBUG_LOCKS_WARN_ON(ctx->contending_lock); - ctx->contending_lock = ww; + struct ww_mutex *ww; + + ww = container_of(lock, struct ww_mutex, base); + DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock); + ww_ctx->contending_lock = ww; #endif - return -EDEADLK; + return -EDEADLK; + } + + break; + } + + pos = &cur->list; + + /* + * Wake up the waiter so that it gets a chance to back + * off. + */ + if (cur->ww_ctx->acquired > 0) { + debug_mutex_wake_waiter(lock, cur); + wake_up_process(cur->task); + } } + list_add_tail(&waiter->list, pos); return 0; } @@ -622,15 +739,15 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, struct lockdep_map *nest_lock, unsigned long ip, struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx) { - struct task_struct *task = current; struct mutex_waiter waiter; - unsigned long flags; bool first = false; struct ww_mutex *ww; int ret; - if (use_ww_ctx) { - ww = container_of(lock, struct ww_mutex, base); + might_sleep(); + + ww = container_of(lock, struct ww_mutex, base); + if (use_ww_ctx && ww_ctx) { if (unlikely(ww_ctx == READ_ONCE(ww->ctx))) return -EALREADY; } @@ -638,36 +755,54 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, preempt_disable(); mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip); - if (__mutex_trylock(lock, false) || - mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, false)) { + if (__mutex_trylock(lock) || + mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, NULL)) { /* got the lock, yay! */ lock_acquired(&lock->dep_map, ip); - if (use_ww_ctx) + if (use_ww_ctx && ww_ctx) ww_mutex_set_context_fastpath(ww, ww_ctx); preempt_enable(); return 0; } - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); /* * After waiting to acquire the wait_lock, try again. */ - if (__mutex_trylock(lock, false)) + if (__mutex_trylock(lock)) { + if (use_ww_ctx && ww_ctx) + __ww_mutex_wakeup_for_backoff(lock, ww_ctx); + goto skip_wait; + } debug_mutex_lock_common(lock, &waiter); - debug_mutex_add_waiter(lock, &waiter, task); + debug_mutex_add_waiter(lock, &waiter, current); + + lock_contended(&lock->dep_map, ip); - /* add waiting tasks to the end of the waitqueue (FIFO): */ - list_add_tail(&waiter.list, &lock->wait_list); - waiter.task = task; + if (!use_ww_ctx) { + /* add waiting tasks to the end of the waitqueue (FIFO): */ + list_add_tail(&waiter.list, &lock->wait_list); + +#ifdef CONFIG_DEBUG_MUTEXES + waiter.ww_ctx = MUTEX_POISON_WW_CTX; +#endif + } else { + /* Add in stamp order, waking up waiters that must back off. */ + ret = __ww_mutex_add_waiter(&waiter, lock, ww_ctx); + if (ret) + goto err_early_backoff; + + waiter.ww_ctx = ww_ctx; + } + + waiter.task = current; if (__mutex_waiter_is_first(lock, &waiter)) __mutex_set_flag(lock, MUTEX_FLAG_WAITERS); - lock_contended(&lock->dep_map, ip); - - set_task_state(task, state); + set_current_state(state); for (;;) { /* * Once we hold wait_lock, we're serialized against @@ -675,7 +810,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, * before testing the error conditions to make sure we pick up * the handoff. */ - if (__mutex_trylock(lock, first)) + if (__mutex_trylock(lock)) goto acquired; /* @@ -683,42 +818,47 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, * wait_lock. This ensures the lock cancellation is ordered * against mutex_unlock() and wake-ups do not go missing. */ - if (unlikely(signal_pending_state(state, task))) { + if (unlikely(signal_pending_state(state, current))) { ret = -EINTR; goto err; } - if (use_ww_ctx && ww_ctx->acquired > 0) { - ret = __ww_mutex_lock_check_stamp(lock, ww_ctx); + if (use_ww_ctx && ww_ctx && ww_ctx->acquired > 0) { + ret = __ww_mutex_lock_check_stamp(lock, &waiter, ww_ctx); if (ret) goto err; } - spin_unlock_mutex(&lock->wait_lock, flags); + spin_unlock(&lock->wait_lock); schedule_preempt_disabled(); - if (!first && __mutex_waiter_is_first(lock, &waiter)) { - first = true; - __mutex_set_flag(lock, MUTEX_FLAG_HANDOFF); + /* + * ww_mutex needs to always recheck its position since its waiter + * list is not FIFO ordered. + */ + if ((use_ww_ctx && ww_ctx) || !first) { + first = __mutex_waiter_is_first(lock, &waiter); + if (first) + __mutex_set_flag(lock, MUTEX_FLAG_HANDOFF); } - set_task_state(task, state); + set_current_state(state); /* * Here we order against unlock; we must either see it change * state back to RUNNING and fall through the next schedule(), * or we must see its unlock and acquire. */ - if ((first && mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, true)) || - __mutex_trylock(lock, first)) + if (__mutex_trylock(lock) || + (first && mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, &waiter))) break; - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); } - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); acquired: - __set_task_state(task, TASK_RUNNING); + __set_current_state(TASK_RUNNING); - mutex_remove_waiter(lock, &waiter, task); + mutex_remove_waiter(lock, &waiter, current); if (likely(list_empty(&lock->wait_list))) __mutex_clear_flag(lock, MUTEX_FLAGS); @@ -728,30 +868,44 @@ skip_wait: /* got the lock - cleanup and rejoice! */ lock_acquired(&lock->dep_map, ip); - if (use_ww_ctx) + if (use_ww_ctx && ww_ctx) ww_mutex_set_context_slowpath(ww, ww_ctx); - spin_unlock_mutex(&lock->wait_lock, flags); + spin_unlock(&lock->wait_lock); preempt_enable(); return 0; err: - __set_task_state(task, TASK_RUNNING); - mutex_remove_waiter(lock, &waiter, task); - spin_unlock_mutex(&lock->wait_lock, flags); + __set_current_state(TASK_RUNNING); + mutex_remove_waiter(lock, &waiter, current); +err_early_backoff: + spin_unlock(&lock->wait_lock); debug_mutex_free_waiter(&waiter); mutex_release(&lock->dep_map, 1, ip); preempt_enable(); return ret; } +static int __sched +__mutex_lock(struct mutex *lock, long state, unsigned int subclass, + struct lockdep_map *nest_lock, unsigned long ip) +{ + return __mutex_lock_common(lock, state, subclass, nest_lock, ip, NULL, false); +} + +static int __sched +__ww_mutex_lock(struct mutex *lock, long state, unsigned int subclass, + struct lockdep_map *nest_lock, unsigned long ip, + struct ww_acquire_ctx *ww_ctx) +{ + return __mutex_lock_common(lock, state, subclass, nest_lock, ip, ww_ctx, true); +} + #ifdef CONFIG_DEBUG_LOCK_ALLOC void __sched mutex_lock_nested(struct mutex *lock, unsigned int subclass) { - might_sleep(); - __mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, - subclass, NULL, _RET_IP_, NULL, 0); + __mutex_lock(lock, TASK_UNINTERRUPTIBLE, subclass, NULL, _RET_IP_); } EXPORT_SYMBOL_GPL(mutex_lock_nested); @@ -759,30 +913,38 @@ EXPORT_SYMBOL_GPL(mutex_lock_nested); void __sched _mutex_lock_nest_lock(struct mutex *lock, struct lockdep_map *nest) { - might_sleep(); - __mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, - 0, nest, _RET_IP_, NULL, 0); + __mutex_lock(lock, TASK_UNINTERRUPTIBLE, 0, nest, _RET_IP_); } EXPORT_SYMBOL_GPL(_mutex_lock_nest_lock); int __sched mutex_lock_killable_nested(struct mutex *lock, unsigned int subclass) { - might_sleep(); - return __mutex_lock_common(lock, TASK_KILLABLE, - subclass, NULL, _RET_IP_, NULL, 0); + return __mutex_lock(lock, TASK_KILLABLE, subclass, NULL, _RET_IP_); } EXPORT_SYMBOL_GPL(mutex_lock_killable_nested); int __sched mutex_lock_interruptible_nested(struct mutex *lock, unsigned int subclass) { - might_sleep(); - return __mutex_lock_common(lock, TASK_INTERRUPTIBLE, - subclass, NULL, _RET_IP_, NULL, 0); + return __mutex_lock(lock, TASK_INTERRUPTIBLE, subclass, NULL, _RET_IP_); } EXPORT_SYMBOL_GPL(mutex_lock_interruptible_nested); +void __sched +mutex_lock_io_nested(struct mutex *lock, unsigned int subclass) +{ + int token; + + might_sleep(); + + token = io_schedule_prepare(); + __mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, + subclass, NULL, _RET_IP_, NULL, 0); + io_schedule_finish(token); +} +EXPORT_SYMBOL_GPL(mutex_lock_io_nested); + static inline int ww_mutex_deadlock_injection(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { @@ -810,35 +972,37 @@ ww_mutex_deadlock_injection(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) } int __sched -__ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { int ret; might_sleep(); - ret = __mutex_lock_common(&lock->base, TASK_UNINTERRUPTIBLE, - 0, &ctx->dep_map, _RET_IP_, ctx, 1); - if (!ret && ctx->acquired > 1) + ret = __ww_mutex_lock(&lock->base, TASK_UNINTERRUPTIBLE, + 0, ctx ? &ctx->dep_map : NULL, _RET_IP_, + ctx); + if (!ret && ctx && ctx->acquired > 1) return ww_mutex_deadlock_injection(lock, ctx); return ret; } -EXPORT_SYMBOL_GPL(__ww_mutex_lock); +EXPORT_SYMBOL_GPL(ww_mutex_lock); int __sched -__ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { int ret; might_sleep(); - ret = __mutex_lock_common(&lock->base, TASK_INTERRUPTIBLE, - 0, &ctx->dep_map, _RET_IP_, ctx, 1); + ret = __ww_mutex_lock(&lock->base, TASK_INTERRUPTIBLE, + 0, ctx ? &ctx->dep_map : NULL, _RET_IP_, + ctx); - if (!ret && ctx->acquired > 1) + if (!ret && ctx && ctx->acquired > 1) return ww_mutex_deadlock_injection(lock, ctx); return ret; } -EXPORT_SYMBOL_GPL(__ww_mutex_lock_interruptible); +EXPORT_SYMBOL_GPL(ww_mutex_lock_interruptible); #endif @@ -848,8 +1012,8 @@ EXPORT_SYMBOL_GPL(__ww_mutex_lock_interruptible); static noinline void __sched __mutex_unlock_slowpath(struct mutex *lock, unsigned long ip) { struct task_struct *next = NULL; - unsigned long owner, flags; DEFINE_WAKE_Q(wake_q); + unsigned long owner; mutex_release(&lock->dep_map, 1, ip); @@ -866,6 +1030,7 @@ static noinline void __sched __mutex_unlock_slowpath(struct mutex *lock, unsigne #ifdef CONFIG_DEBUG_MUTEXES DEBUG_LOCKS_WARN_ON(__owner_task(owner) != current); + DEBUG_LOCKS_WARN_ON(owner & MUTEX_FLAG_PICKUP); #endif if (owner & MUTEX_FLAG_HANDOFF) @@ -883,7 +1048,7 @@ static noinline void __sched __mutex_unlock_slowpath(struct mutex *lock, unsigne owner = old; } - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); debug_mutex_unlock(lock); if (!list_empty(&lock->wait_list)) { /* get the first entry from the wait-list: */ @@ -900,7 +1065,7 @@ static noinline void __sched __mutex_unlock_slowpath(struct mutex *lock, unsigne if (owner & MUTEX_FLAG_HANDOFF) __mutex_handoff(lock, next); - spin_unlock_mutex(&lock->wait_lock, flags); + spin_unlock(&lock->wait_lock); wake_up_q(&wake_q); } @@ -950,40 +1115,47 @@ int __sched mutex_lock_killable(struct mutex *lock) } EXPORT_SYMBOL(mutex_lock_killable); +void __sched mutex_lock_io(struct mutex *lock) +{ + int token; + + token = io_schedule_prepare(); + mutex_lock(lock); + io_schedule_finish(token); +} +EXPORT_SYMBOL_GPL(mutex_lock_io); + static noinline void __sched __mutex_lock_slowpath(struct mutex *lock) { - __mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, 0, - NULL, _RET_IP_, NULL, 0); + __mutex_lock(lock, TASK_UNINTERRUPTIBLE, 0, NULL, _RET_IP_); } static noinline int __sched __mutex_lock_killable_slowpath(struct mutex *lock) { - return __mutex_lock_common(lock, TASK_KILLABLE, 0, - NULL, _RET_IP_, NULL, 0); + return __mutex_lock(lock, TASK_KILLABLE, 0, NULL, _RET_IP_); } static noinline int __sched __mutex_lock_interruptible_slowpath(struct mutex *lock) { - return __mutex_lock_common(lock, TASK_INTERRUPTIBLE, 0, - NULL, _RET_IP_, NULL, 0); + return __mutex_lock(lock, TASK_INTERRUPTIBLE, 0, NULL, _RET_IP_); } static noinline int __sched __ww_mutex_lock_slowpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { - return __mutex_lock_common(&lock->base, TASK_UNINTERRUPTIBLE, 0, - NULL, _RET_IP_, ctx, 1); + return __ww_mutex_lock(&lock->base, TASK_UNINTERRUPTIBLE, 0, NULL, + _RET_IP_, ctx); } static noinline int __sched __ww_mutex_lock_interruptible_slowpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { - return __mutex_lock_common(&lock->base, TASK_INTERRUPTIBLE, 0, - NULL, _RET_IP_, ctx, 1); + return __ww_mutex_lock(&lock->base, TASK_INTERRUPTIBLE, 0, NULL, + _RET_IP_, ctx); } #endif @@ -1004,7 +1176,7 @@ __ww_mutex_lock_interruptible_slowpath(struct ww_mutex *lock, */ int __sched mutex_trylock(struct mutex *lock) { - bool locked = __mutex_trylock(lock, false); + bool locked = __mutex_trylock(lock); if (locked) mutex_acquire(&lock->dep_map, 0, 1, _RET_IP_); @@ -1015,32 +1187,34 @@ EXPORT_SYMBOL(mutex_trylock); #ifndef CONFIG_DEBUG_LOCK_ALLOC int __sched -__ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { might_sleep(); if (__mutex_trylock_fast(&lock->base)) { - ww_mutex_set_context_fastpath(lock, ctx); + if (ctx) + ww_mutex_set_context_fastpath(lock, ctx); return 0; } return __ww_mutex_lock_slowpath(lock, ctx); } -EXPORT_SYMBOL(__ww_mutex_lock); +EXPORT_SYMBOL(ww_mutex_lock); int __sched -__ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { might_sleep(); if (__mutex_trylock_fast(&lock->base)) { - ww_mutex_set_context_fastpath(lock, ctx); + if (ctx) + ww_mutex_set_context_fastpath(lock, ctx); return 0; } return __ww_mutex_lock_interruptible_slowpath(lock, ctx); } -EXPORT_SYMBOL(__ww_mutex_lock_interruptible); +EXPORT_SYMBOL(ww_mutex_lock_interruptible); #endif diff --git a/kernel/locking/mutex.h b/kernel/locking/mutex.h index 4410a4af42a3..6ebc1902f779 100644 --- a/kernel/locking/mutex.h +++ b/kernel/locking/mutex.h @@ -9,10 +9,6 @@ * !CONFIG_DEBUG_MUTEXES case. Most of them are NOPs: */ -#define spin_lock_mutex(lock, flags) \ - do { spin_lock(lock); (void)(flags); } while (0) -#define spin_unlock_mutex(lock, flags) \ - do { spin_unlock(lock); (void)(flags); } while (0) #define mutex_remove_waiter(lock, waiter, task) \ __list_del((waiter)->list.prev, (waiter)->list.next) diff --git a/kernel/locking/percpu-rwsem.c b/kernel/locking/percpu-rwsem.c index ce182599cf2e..883cf1b92d90 100644 --- a/kernel/locking/percpu-rwsem.c +++ b/kernel/locking/percpu-rwsem.c @@ -1,7 +1,6 @@ #include <linux/atomic.h> #include <linux/rwsem.h> #include <linux/percpu.h> -#include <linux/wait.h> #include <linux/lockdep.h> #include <linux/percpu-rwsem.h> #include <linux/rcupdate.h> @@ -18,7 +17,7 @@ int __percpu_init_rwsem(struct percpu_rw_semaphore *sem, /* ->rw_sem represents the whole percpu_rw_semaphore for lockdep */ rcu_sync_init(&sem->rss, RCU_SCHED_SYNC); __init_rwsem(&sem->rw_sem, name, rwsem_key); - init_waitqueue_head(&sem->writer); + rcuwait_init(&sem->writer); sem->readers_block = 0; return 0; } @@ -103,7 +102,7 @@ void __percpu_up_read(struct percpu_rw_semaphore *sem) __this_cpu_dec(*sem->read_count); /* Prod writer to recheck readers_active */ - wake_up(&sem->writer); + rcuwait_wake_up(&sem->writer); } EXPORT_SYMBOL_GPL(__percpu_up_read); @@ -160,7 +159,7 @@ void percpu_down_write(struct percpu_rw_semaphore *sem) */ /* Wait for all now active readers to complete. */ - wait_event(sem->writer, readers_active_check(sem)); + rcuwait_wait_event(&sem->writer, readers_active_check(sem)); } EXPORT_SYMBOL_GPL(percpu_down_write); diff --git a/kernel/locking/qrwlock.c b/kernel/locking/qrwlock.c index cc3ed0ccdfa2..2655f26ec882 100644 --- a/kernel/locking/qrwlock.c +++ b/kernel/locking/qrwlock.c @@ -20,6 +20,7 @@ #include <linux/cpumask.h> #include <linux/percpu.h> #include <linux/hardirq.h> +#include <linux/spinlock.h> #include <asm/qrwlock.h> /* diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c index b2caec7315af..fd24153e8a48 100644 --- a/kernel/locking/qspinlock.c +++ b/kernel/locking/qspinlock.c @@ -28,6 +28,7 @@ #include <linux/percpu.h> #include <linux/hardirq.h> #include <linux/mutex.h> +#include <linux/prefetch.h> #include <asm/byteorder.h> #include <asm/qspinlock.h> diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h index e3b5520005db..4ccfcaae5b89 100644 --- a/kernel/locking/qspinlock_paravirt.h +++ b/kernel/locking/qspinlock_paravirt.h @@ -193,7 +193,8 @@ void __init __pv_init_lock_hash(void) */ pv_lock_hash = alloc_large_system_hash("PV qspinlock", sizeof(struct pv_hash_entry), - pv_hash_size, 0, HASH_EARLY, + pv_hash_size, 0, + HASH_EARLY | HASH_ZERO, &pv_lock_hash_bits, NULL, pv_hash_size, pv_hash_size); } @@ -263,7 +264,7 @@ pv_wait_early(struct pv_node *prev, int loop) if ((loop & PV_PREV_CHECK_MASK) != 0) return false; - return READ_ONCE(prev->state) != vcpu_running; + return READ_ONCE(prev->state) != vcpu_running || vcpu_is_preempted(prev->cpu); } /* diff --git a/kernel/locking/qspinlock_stat.h b/kernel/locking/qspinlock_stat.h index e852be4851fc..4a30ef63c607 100644 --- a/kernel/locking/qspinlock_stat.h +++ b/kernel/locking/qspinlock_stat.h @@ -63,6 +63,7 @@ enum qlock_stats { */ #include <linux/debugfs.h> #include <linux/sched.h> +#include <linux/sched/clock.h> #include <linux/fs.h> static const char * const qstat_names[qstat_num + 1] = { diff --git a/kernel/locking/rtmutex-debug.c b/kernel/locking/rtmutex-debug.c index 62b6cee8ea7f..ac35e648b0e5 100644 --- a/kernel/locking/rtmutex-debug.c +++ b/kernel/locking/rtmutex-debug.c @@ -18,6 +18,7 @@ */ #include <linux/sched.h> #include <linux/sched/rt.h> +#include <linux/sched/debug.h> #include <linux/delay.h> #include <linux/export.h> #include <linux/spinlock.h> @@ -101,10 +102,11 @@ void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) return; } - printk("\n============================================\n"); - printk( "[ BUG: circular locking deadlock detected! ]\n"); - printk("%s\n", print_tainted()); - printk( "--------------------------------------------\n"); + pr_warn("\n"); + pr_warn("============================================\n"); + pr_warn("WARNING: circular locking deadlock detected!\n"); + pr_warn("%s\n", print_tainted()); + pr_warn("--------------------------------------------\n"); printk("%s/%d is deadlocking current task %s/%d\n\n", task->comm, task_pid_nr(task), current->comm, task_pid_nr(current)); @@ -164,21 +166,16 @@ void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter) memset(waiter, 0x22, sizeof(*waiter)); } -void debug_rt_mutex_init(struct rt_mutex *lock, const char *name) +void debug_rt_mutex_init(struct rt_mutex *lock, const char *name, struct lock_class_key *key) { /* * Make sure we are not reinitializing a held lock: */ debug_check_no_locks_freed((void *)lock, sizeof(*lock)); lock->name = name; -} - -void -rt_mutex_deadlock_account_lock(struct rt_mutex *lock, struct task_struct *task) -{ -} -void rt_mutex_deadlock_account_unlock(struct task_struct *task) -{ +#ifdef CONFIG_DEBUG_LOCK_ALLOC + lockdep_init_map(&lock->dep_map, name, key, 0); +#endif } diff --git a/kernel/locking/rtmutex-debug.h b/kernel/locking/rtmutex-debug.h index d0519c3432b6..5078c6ddf4a5 100644 --- a/kernel/locking/rtmutex-debug.h +++ b/kernel/locking/rtmutex-debug.h @@ -9,12 +9,9 @@ * This file contains macros used solely by rtmutex.c. Debug version. */ -extern void -rt_mutex_deadlock_account_lock(struct rt_mutex *lock, struct task_struct *task); -extern void rt_mutex_deadlock_account_unlock(struct task_struct *task); extern void debug_rt_mutex_init_waiter(struct rt_mutex_waiter *waiter); extern void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter); -extern void debug_rt_mutex_init(struct rt_mutex *lock, const char *name); +extern void debug_rt_mutex_init(struct rt_mutex *lock, const char *name, struct lock_class_key *key); extern void debug_rt_mutex_lock(struct rt_mutex *lock); extern void debug_rt_mutex_unlock(struct rt_mutex *lock); extern void debug_rt_mutex_proxy_lock(struct rt_mutex *lock, diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index 2f443ed2320a..649dc9d3951a 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -12,9 +12,11 @@ */ #include <linux/spinlock.h> #include <linux/export.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> #include <linux/sched/rt.h> #include <linux/sched/deadline.h> +#include <linux/sched/wake_q.h> +#include <linux/sched/debug.h> #include <linux/timer.h> #include "rtmutex_common.h" @@ -222,6 +224,12 @@ static inline bool unlock_rt_mutex_safe(struct rt_mutex *lock, } #endif +/* + * Only use with rt_mutex_waiter_{less,equal}() + */ +#define task_to_waiter(p) \ + &(struct rt_mutex_waiter){ .prio = (p)->prio, .deadline = (p)->dl.deadline } + static inline int rt_mutex_waiter_less(struct rt_mutex_waiter *left, struct rt_mutex_waiter *right) @@ -236,12 +244,30 @@ rt_mutex_waiter_less(struct rt_mutex_waiter *left, * then right waiter has a dl_prio() too. */ if (dl_prio(left->prio)) - return dl_time_before(left->task->dl.deadline, - right->task->dl.deadline); + return dl_time_before(left->deadline, right->deadline); return 0; } +static inline int +rt_mutex_waiter_equal(struct rt_mutex_waiter *left, + struct rt_mutex_waiter *right) +{ + if (left->prio != right->prio) + return 0; + + /* + * If both waiters have dl_prio(), we check the deadlines of the + * associated tasks. + * If left waiter has a dl_prio(), and we didn't return 0 above, + * then right waiter has a dl_prio() too. + */ + if (dl_prio(left->prio)) + return left->deadline == right->deadline; + + return 1; +} + static void rt_mutex_enqueue(struct rt_mutex *lock, struct rt_mutex_waiter *waiter) { @@ -320,72 +346,16 @@ rt_mutex_dequeue_pi(struct task_struct *task, struct rt_mutex_waiter *waiter) RB_CLEAR_NODE(&waiter->pi_tree_entry); } -/* - * Calculate task priority from the waiter tree priority - * - * Return task->normal_prio when the waiter tree is empty or when - * the waiter is not allowed to do priority boosting - */ -int rt_mutex_getprio(struct task_struct *task) -{ - if (likely(!task_has_pi_waiters(task))) - return task->normal_prio; - - return min(task_top_pi_waiter(task)->prio, - task->normal_prio); -} - -struct task_struct *rt_mutex_get_top_task(struct task_struct *task) -{ - if (likely(!task_has_pi_waiters(task))) - return NULL; - - return task_top_pi_waiter(task)->task; -} - -/* - * Called by sched_setscheduler() to get the priority which will be - * effective after the change. - */ -int rt_mutex_get_effective_prio(struct task_struct *task, int newprio) +static void rt_mutex_adjust_prio(struct task_struct *p) { - if (!task_has_pi_waiters(task)) - return newprio; - - if (task_top_pi_waiter(task)->task->prio <= newprio) - return task_top_pi_waiter(task)->task->prio; - return newprio; -} + struct task_struct *pi_task = NULL; -/* - * Adjust the priority of a task, after its pi_waiters got modified. - * - * This can be both boosting and unboosting. task->pi_lock must be held. - */ -static void __rt_mutex_adjust_prio(struct task_struct *task) -{ - int prio = rt_mutex_getprio(task); - - if (task->prio != prio || dl_prio(prio)) - rt_mutex_setprio(task, prio); -} + lockdep_assert_held(&p->pi_lock); -/* - * Adjust task priority (undo boosting). Called from the exit path of - * rt_mutex_slowunlock() and rt_mutex_slowlock(). - * - * (Note: We do this outside of the protection of lock->wait_lock to - * allow the lock to be taken while or before we readjust the priority - * of task. We do not use the spin_xx_mutex() variants here as we are - * outside of the debug path.) - */ -void rt_mutex_adjust_prio(struct task_struct *task) -{ - unsigned long flags; + if (task_has_pi_waiters(p)) + pi_task = task_top_pi_waiter(p)->task; - raw_spin_lock_irqsave(&task->pi_lock, flags); - __rt_mutex_adjust_prio(task); - raw_spin_unlock_irqrestore(&task->pi_lock, flags); + rt_mutex_setprio(p, pi_task); } /* @@ -608,7 +578,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, * enabled we continue, but stop the requeueing in the chain * walk. */ - if (waiter->prio == task->prio) { + if (rt_mutex_waiter_equal(waiter, task_to_waiter(task))) { if (!detect_deadlock) goto out_unlock_pi; else @@ -704,7 +674,26 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, /* [7] Requeue the waiter in the lock waiter tree. */ rt_mutex_dequeue(lock, waiter); + + /* + * Update the waiter prio fields now that we're dequeued. + * + * These values can have changed through either: + * + * sys_sched_set_scheduler() / sys_sched_setattr() + * + * or + * + * DL CBS enforcement advancing the effective deadline. + * + * Even though pi_waiters also uses these fields, and that tree is only + * updated in [11], we can do this here, since we hold [L], which + * serializes all pi_waiters access and rb_erase() does not care about + * the values of the node being removed. + */ waiter->prio = task->prio; + waiter->deadline = task->dl.deadline; + rt_mutex_enqueue(lock, waiter); /* [8] Release the task */ @@ -745,7 +734,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, */ rt_mutex_dequeue_pi(task, prerequeue_top_waiter); rt_mutex_enqueue_pi(task, waiter); - __rt_mutex_adjust_prio(task); + rt_mutex_adjust_prio(task); } else if (prerequeue_top_waiter == waiter) { /* @@ -761,7 +750,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, rt_mutex_dequeue_pi(task, waiter); waiter = rt_mutex_top_waiter(lock); rt_mutex_enqueue_pi(task, waiter); - __rt_mutex_adjust_prio(task); + rt_mutex_adjust_prio(task); } else { /* * Nothing changed. No need to do any priority @@ -831,6 +820,8 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, struct rt_mutex_waiter *waiter) { + lockdep_assert_held(&lock->wait_lock); + /* * Before testing whether we can acquire @lock, we set the * RT_MUTEX_HAS_WAITERS bit in @lock->owner. This forces all @@ -890,7 +881,8 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, * the top waiter priority (kernel view), * @task lost. */ - if (task->prio >= rt_mutex_top_waiter(lock)->prio) + if (!rt_mutex_waiter_less(task_to_waiter(task), + rt_mutex_top_waiter(lock))) return 0; /* @@ -936,8 +928,6 @@ takeit: */ rt_mutex_set_owner(lock, task); - rt_mutex_deadlock_account_lock(lock, task); - return 1; } @@ -958,6 +948,8 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, struct rt_mutex *next_lock; int chain_walk = 0, res; + lockdep_assert_held(&lock->wait_lock); + /* * Early deadlock detection. We really don't want the task to * enqueue on itself just to untangle the mess later. It's not @@ -971,10 +963,10 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, return -EDEADLK; raw_spin_lock(&task->pi_lock); - __rt_mutex_adjust_prio(task); waiter->task = task; waiter->lock = lock; waiter->prio = task->prio; + waiter->deadline = task->dl.deadline; /* Get the top priority waiter on the lock */ if (rt_mutex_has_waiters(lock)) @@ -993,7 +985,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, rt_mutex_dequeue_pi(owner, top_waiter); rt_mutex_enqueue_pi(owner, waiter); - __rt_mutex_adjust_prio(owner); + rt_mutex_adjust_prio(owner); if (owner->pi_blocked_on) chain_walk = 1; } else if (rt_mutex_cond_detect_deadlock(waiter, chwalk)) { @@ -1045,12 +1037,14 @@ static void mark_wakeup_next_waiter(struct wake_q_head *wake_q, waiter = rt_mutex_top_waiter(lock); /* - * Remove it from current->pi_waiters. We do not adjust a - * possible priority boost right now. We execute wakeup in the - * boosted mode and go back to normal after releasing - * lock->wait_lock. + * Remove it from current->pi_waiters and deboost. + * + * We must in fact deboost here in order to ensure we call + * rt_mutex_setprio() to update p->pi_top_task before the + * task unblocks. */ rt_mutex_dequeue_pi(current, waiter); + rt_mutex_adjust_prio(current); /* * As we are waking up the top waiter, and the waiter stays @@ -1062,9 +1056,19 @@ static void mark_wakeup_next_waiter(struct wake_q_head *wake_q, */ lock->owner = (void *) RT_MUTEX_HAS_WAITERS; - raw_spin_unlock(¤t->pi_lock); - + /* + * We deboosted before waking the top waiter task such that we don't + * run two tasks with the 'same' priority (and ensure the + * p->pi_top_task pointer points to a blocked task). This however can + * lead to priority inversion if we would get preempted after the + * deboost but before waking our donor task, hence the preempt_disable() + * before unlock. + * + * Pairs with preempt_enable() in rt_mutex_postunlock(); + */ + preempt_disable(); wake_q_add(wake_q, waiter->task); + raw_spin_unlock(¤t->pi_lock); } /* @@ -1080,6 +1084,8 @@ static void remove_waiter(struct rt_mutex *lock, struct task_struct *owner = rt_mutex_owner(lock); struct rt_mutex *next_lock; + lockdep_assert_held(&lock->wait_lock); + raw_spin_lock(¤t->pi_lock); rt_mutex_dequeue(lock, waiter); current->pi_blocked_on = NULL; @@ -1099,7 +1105,7 @@ static void remove_waiter(struct rt_mutex *lock, if (rt_mutex_has_waiters(lock)) rt_mutex_enqueue_pi(owner, rt_mutex_top_waiter(lock)); - __rt_mutex_adjust_prio(owner); + rt_mutex_adjust_prio(owner); /* Store the lock on which owner is blocked or NULL */ next_lock = task_blocked_on_lock(owner); @@ -1138,8 +1144,7 @@ void rt_mutex_adjust_pi(struct task_struct *task) raw_spin_lock_irqsave(&task->pi_lock, flags); waiter = task->pi_blocked_on; - if (!waiter || (waiter->prio == task->prio && - !dl_prio(task->prio))) { + if (!waiter || rt_mutex_waiter_equal(waiter, task_to_waiter(task))) { raw_spin_unlock_irqrestore(&task->pi_lock, flags); return; } @@ -1153,6 +1158,14 @@ void rt_mutex_adjust_pi(struct task_struct *task) next_lock, NULL, task); } +void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) +{ + debug_rt_mutex_init_waiter(waiter); + RB_CLEAR_NODE(&waiter->pi_tree_entry); + RB_CLEAR_NODE(&waiter->tree_entry); + waiter->task = NULL; +} + /** * __rt_mutex_slowlock() - Perform the wait-wake-try-to-take loop * @lock: the rt_mutex to take @@ -1179,7 +1192,7 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state, * TASK_INTERRUPTIBLE checks for signals and * timeout. Ignored otherwise. */ - if (unlikely(state == TASK_INTERRUPTIBLE)) { + if (likely(state == TASK_INTERRUPTIBLE)) { /* Signal pending? */ if (signal_pending(current)) ret = -EINTR; @@ -1235,9 +1248,7 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, unsigned long flags; int ret = 0; - debug_rt_mutex_init_waiter(&waiter); - RB_CLEAR_NODE(&waiter.pi_tree_entry); - RB_CLEAR_NODE(&waiter.tree_entry); + rt_mutex_init_waiter(&waiter); /* * Technically we could use raw_spin_[un]lock_irq() here, but this can @@ -1328,7 +1339,8 @@ static inline int rt_mutex_slowtrylock(struct rt_mutex *lock) /* * Slow path to release a rt-mutex. - * Return whether the current task needs to undo a potential priority boosting. + * + * Return whether the current task needs to call rt_mutex_postunlock(). */ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock, struct wake_q_head *wake_q) @@ -1340,8 +1352,6 @@ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock, debug_rt_mutex_unlock(lock); - rt_mutex_deadlock_account_unlock(current); - /* * We must be careful here if the fast path is enabled. If we * have no waiters queued we cannot set owner to NULL here @@ -1388,11 +1398,9 @@ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock, * Queue the next waiter for wakeup once we release the wait_lock. */ mark_wakeup_next_waiter(wake_q, lock); - raw_spin_unlock_irqrestore(&lock->wait_lock, flags); - /* check PI boosting */ - return true; + return true; /* call rt_mutex_postunlock() */ } /* @@ -1407,11 +1415,10 @@ rt_mutex_fastlock(struct rt_mutex *lock, int state, struct hrtimer_sleeper *timeout, enum rtmutex_chainwalk chwalk)) { - if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) { - rt_mutex_deadlock_account_lock(lock, current); + if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) return 0; - } else - return slowfn(lock, state, NULL, RT_MUTEX_MIN_CHAINWALK); + + return slowfn(lock, state, NULL, RT_MUTEX_MIN_CHAINWALK); } static inline int @@ -1423,24 +1430,33 @@ rt_mutex_timed_fastlock(struct rt_mutex *lock, int state, enum rtmutex_chainwalk chwalk)) { if (chwalk == RT_MUTEX_MIN_CHAINWALK && - likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) { - rt_mutex_deadlock_account_lock(lock, current); + likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) return 0; - } else - return slowfn(lock, state, timeout, chwalk); + + return slowfn(lock, state, timeout, chwalk); } static inline int rt_mutex_fasttrylock(struct rt_mutex *lock, int (*slowfn)(struct rt_mutex *lock)) { - if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) { - rt_mutex_deadlock_account_lock(lock, current); + if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) return 1; - } + return slowfn(lock); } +/* + * Performs the wakeup of the the top-waiter and re-enables preemption. + */ +void rt_mutex_postunlock(struct wake_q_head *wake_q) +{ + wake_up_q(wake_q); + + /* Pairs with preempt_disable() in rt_mutex_slowunlock() */ + preempt_enable(); +} + static inline void rt_mutex_fastunlock(struct rt_mutex *lock, bool (*slowfn)(struct rt_mutex *lock, @@ -1448,18 +1464,11 @@ rt_mutex_fastunlock(struct rt_mutex *lock, { DEFINE_WAKE_Q(wake_q); - if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) { - rt_mutex_deadlock_account_unlock(current); - - } else { - bool deboost = slowfn(lock, &wake_q); - - wake_up_q(&wake_q); + if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) + return; - /* Undo pi boosting if necessary: */ - if (deboost) - rt_mutex_adjust_prio(current); - } + if (slowfn(lock, &wake_q)) + rt_mutex_postunlock(&wake_q); } /** @@ -1471,6 +1480,7 @@ void __sched rt_mutex_lock(struct rt_mutex *lock) { might_sleep(); + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, rt_mutex_slowlock); } EXPORT_SYMBOL_GPL(rt_mutex_lock); @@ -1486,23 +1496,25 @@ EXPORT_SYMBOL_GPL(rt_mutex_lock); */ int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock) { + int ret; + might_sleep(); - return rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE, rt_mutex_slowlock); + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); + ret = rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE, rt_mutex_slowlock); + if (ret) + mutex_release(&lock->dep_map, 1, _RET_IP_); + + return ret; } EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible); /* - * Futex variant with full deadlock detection. + * Futex variant, must not use fastpath. */ -int rt_mutex_timed_futex_lock(struct rt_mutex *lock, - struct hrtimer_sleeper *timeout) +int __sched rt_mutex_futex_trylock(struct rt_mutex *lock) { - might_sleep(); - - return rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout, - RT_MUTEX_FULL_CHAINWALK, - rt_mutex_slowlock); + return rt_mutex_slowtrylock(lock); } /** @@ -1521,11 +1533,18 @@ int rt_mutex_timed_futex_lock(struct rt_mutex *lock, int rt_mutex_timed_lock(struct rt_mutex *lock, struct hrtimer_sleeper *timeout) { + int ret; + might_sleep(); - return rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout, + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); + ret = rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout, RT_MUTEX_MIN_CHAINWALK, rt_mutex_slowlock); + if (ret) + mutex_release(&lock->dep_map, 1, _RET_IP_); + + return ret; } EXPORT_SYMBOL_GPL(rt_mutex_timed_lock); @@ -1542,10 +1561,16 @@ EXPORT_SYMBOL_GPL(rt_mutex_timed_lock); */ int __sched rt_mutex_trylock(struct rt_mutex *lock) { + int ret; + if (WARN_ON_ONCE(in_irq() || in_nmi() || in_serving_softirq())) return 0; - return rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock); + ret = rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock); + if (ret) + mutex_acquire(&lock->dep_map, 0, 1, _RET_IP_); + + return ret; } EXPORT_SYMBOL_GPL(rt_mutex_trylock); @@ -1556,25 +1581,49 @@ EXPORT_SYMBOL_GPL(rt_mutex_trylock); */ void __sched rt_mutex_unlock(struct rt_mutex *lock) { + mutex_release(&lock->dep_map, 1, _RET_IP_); rt_mutex_fastunlock(lock, rt_mutex_slowunlock); } EXPORT_SYMBOL_GPL(rt_mutex_unlock); /** - * rt_mutex_futex_unlock - Futex variant of rt_mutex_unlock - * @lock: the rt_mutex to be unlocked - * - * Returns: true/false indicating whether priority adjustment is - * required or not. + * Futex variant, that since futex variants do not use the fast-path, can be + * simple and will not need to retry. */ -bool __sched rt_mutex_futex_unlock(struct rt_mutex *lock, - struct wake_q_head *wqh) +bool __sched __rt_mutex_futex_unlock(struct rt_mutex *lock, + struct wake_q_head *wake_q) { - if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) { - rt_mutex_deadlock_account_unlock(current); - return false; + lockdep_assert_held(&lock->wait_lock); + + debug_rt_mutex_unlock(lock); + + if (!rt_mutex_has_waiters(lock)) { + lock->owner = NULL; + return false; /* done */ } - return rt_mutex_slowunlock(lock, wqh); + + /* + * We've already deboosted, mark_wakeup_next_waiter() will + * retain preempt_disabled when we drop the wait_lock, to + * avoid inversion prior to the wakeup. preempt_disable() + * therein pairs with rt_mutex_postunlock(). + */ + mark_wakeup_next_waiter(wake_q, lock); + + return true; /* call postunlock() */ +} + +void __sched rt_mutex_futex_unlock(struct rt_mutex *lock) +{ + DEFINE_WAKE_Q(wake_q); + bool postunlock; + + raw_spin_lock_irq(&lock->wait_lock); + postunlock = __rt_mutex_futex_unlock(lock, &wake_q); + raw_spin_unlock_irq(&lock->wait_lock); + + if (postunlock) + rt_mutex_postunlock(&wake_q); } /** @@ -1592,7 +1641,6 @@ void rt_mutex_destroy(struct rt_mutex *lock) lock->magic = NULL; #endif } - EXPORT_SYMBOL_GPL(rt_mutex_destroy); /** @@ -1604,14 +1652,16 @@ EXPORT_SYMBOL_GPL(rt_mutex_destroy); * * Initializing of a locked rt lock is not allowed */ -void __rt_mutex_init(struct rt_mutex *lock, const char *name) +void __rt_mutex_init(struct rt_mutex *lock, const char *name, + struct lock_class_key *key) { lock->owner = NULL; raw_spin_lock_init(&lock->wait_lock); lock->waiters = RB_ROOT; lock->waiters_leftmost = NULL; - debug_rt_mutex_init(lock, name); + if (name && key) + debug_rt_mutex_init(lock, name, key); } EXPORT_SYMBOL_GPL(__rt_mutex_init); @@ -1632,10 +1682,9 @@ EXPORT_SYMBOL_GPL(__rt_mutex_init); void rt_mutex_init_proxy_locked(struct rt_mutex *lock, struct task_struct *proxy_owner) { - __rt_mutex_init(lock, NULL); + __rt_mutex_init(lock, NULL, NULL); debug_rt_mutex_proxy_lock(lock, proxy_owner); rt_mutex_set_owner(lock, proxy_owner); - rt_mutex_deadlock_account_lock(lock, proxy_owner); } /** @@ -1655,34 +1704,16 @@ void rt_mutex_proxy_unlock(struct rt_mutex *lock, { debug_rt_mutex_proxy_unlock(lock); rt_mutex_set_owner(lock, NULL); - rt_mutex_deadlock_account_unlock(proxy_owner); } -/** - * rt_mutex_start_proxy_lock() - Start lock acquisition for another task - * @lock: the rt_mutex to take - * @waiter: the pre-initialized rt_mutex_waiter - * @task: the task to prepare - * - * Returns: - * 0 - task blocked on lock - * 1 - acquired the lock for task, caller should wake it up - * <0 - error - * - * Special API call for FUTEX_REQUEUE_PI support. - */ -int rt_mutex_start_proxy_lock(struct rt_mutex *lock, +int __rt_mutex_start_proxy_lock(struct rt_mutex *lock, struct rt_mutex_waiter *waiter, struct task_struct *task) { int ret; - raw_spin_lock_irq(&lock->wait_lock); - - if (try_to_take_rt_mutex(lock, task, NULL)) { - raw_spin_unlock_irq(&lock->wait_lock); + if (try_to_take_rt_mutex(lock, task, NULL)) return 1; - } /* We enforce deadlock detection for futexes */ ret = task_blocks_on_rt_mutex(lock, waiter, task, @@ -1701,14 +1732,38 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock, if (unlikely(ret)) remove_waiter(lock, waiter); - raw_spin_unlock_irq(&lock->wait_lock); - debug_rt_mutex_print_deadlock(waiter); return ret; } /** + * rt_mutex_start_proxy_lock() - Start lock acquisition for another task + * @lock: the rt_mutex to take + * @waiter: the pre-initialized rt_mutex_waiter + * @task: the task to prepare + * + * Returns: + * 0 - task blocked on lock + * 1 - acquired the lock for task, caller should wake it up + * <0 - error + * + * Special API call for FUTEX_REQUEUE_PI support. + */ +int rt_mutex_start_proxy_lock(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + struct task_struct *task) +{ + int ret; + + raw_spin_lock_irq(&lock->wait_lock); + ret = __rt_mutex_start_proxy_lock(lock, waiter, task); + raw_spin_unlock_irq(&lock->wait_lock); + + return ret; +} + +/** * rt_mutex_next_owner - return the next owner of the lock * * @lock: the rt lock query @@ -1729,36 +1784,87 @@ struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock) } /** - * rt_mutex_finish_proxy_lock() - Complete lock acquisition + * rt_mutex_wait_proxy_lock() - Wait for lock acquisition * @lock: the rt_mutex we were woken on * @to: the timeout, null if none. hrtimer should already have * been started. * @waiter: the pre-initialized rt_mutex_waiter * - * Complete the lock acquisition started our behalf by another thread. + * Wait for the the lock acquisition started on our behalf by + * rt_mutex_start_proxy_lock(). Upon failure, the caller must call + * rt_mutex_cleanup_proxy_lock(). * * Returns: * 0 - success * <0 - error, one of -EINTR, -ETIMEDOUT * - * Special API call for PI-futex requeue support + * Special API call for PI-futex support */ -int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, +int rt_mutex_wait_proxy_lock(struct rt_mutex *lock, struct hrtimer_sleeper *to, struct rt_mutex_waiter *waiter) { int ret; raw_spin_lock_irq(&lock->wait_lock); - - set_current_state(TASK_INTERRUPTIBLE); - /* sleep on the mutex */ + set_current_state(TASK_INTERRUPTIBLE); ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter); + /* + * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might + * have to fix that up. + */ + fixup_rt_mutex_waiters(lock); + raw_spin_unlock_irq(&lock->wait_lock); - if (unlikely(ret)) - remove_waiter(lock, waiter); + return ret; +} +/** + * rt_mutex_cleanup_proxy_lock() - Cleanup failed lock acquisition + * @lock: the rt_mutex we were woken on + * @waiter: the pre-initialized rt_mutex_waiter + * + * Attempt to clean up after a failed rt_mutex_wait_proxy_lock(). + * + * Unless we acquired the lock; we're still enqueued on the wait-list and can + * in fact still be granted ownership until we're removed. Therefore we can + * find we are in fact the owner and must disregard the + * rt_mutex_wait_proxy_lock() failure. + * + * Returns: + * true - did the cleanup, we done. + * false - we acquired the lock after rt_mutex_wait_proxy_lock() returned, + * caller should disregards its return value. + * + * Special API call for PI-futex support + */ +bool rt_mutex_cleanup_proxy_lock(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter) +{ + bool cleanup = false; + + raw_spin_lock_irq(&lock->wait_lock); + /* + * Do an unconditional try-lock, this deals with the lock stealing + * state where __rt_mutex_futex_unlock() -> mark_wakeup_next_waiter() + * sets a NULL owner. + * + * We're not interested in the return value, because the subsequent + * test on rt_mutex_owner() will infer that. If the trylock succeeded, + * we will own the lock and it will have removed the waiter. If we + * failed the trylock, we're still not owner and we need to remove + * ourselves. + */ + try_to_take_rt_mutex(lock, current, waiter); + /* + * Unless we're the owner; we're still enqueued on the wait_list. + * So check if we became owner, if not, take us off the wait_list. + */ + if (rt_mutex_owner(lock) != current) { + remove_waiter(lock, waiter); + cleanup = true; + } /* * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might * have to fix that up. @@ -1767,5 +1873,5 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, raw_spin_unlock_irq(&lock->wait_lock); - return ret; + return cleanup; } diff --git a/kernel/locking/rtmutex.h b/kernel/locking/rtmutex.h index c4060584c407..5c253caffe91 100644 --- a/kernel/locking/rtmutex.h +++ b/kernel/locking/rtmutex.h @@ -11,15 +11,13 @@ */ #define rt_mutex_deadlock_check(l) (0) -#define rt_mutex_deadlock_account_lock(m, t) do { } while (0) -#define rt_mutex_deadlock_account_unlock(l) do { } while (0) #define debug_rt_mutex_init_waiter(w) do { } while (0) #define debug_rt_mutex_free_waiter(w) do { } while (0) #define debug_rt_mutex_lock(l) do { } while (0) #define debug_rt_mutex_proxy_lock(l,p) do { } while (0) #define debug_rt_mutex_proxy_unlock(l) do { } while (0) #define debug_rt_mutex_unlock(l) do { } while (0) -#define debug_rt_mutex_init(m, n) do { } while (0) +#define debug_rt_mutex_init(m, n, k) do { } while (0) #define debug_rt_mutex_deadlock(d, a ,l) do { } while (0) #define debug_rt_mutex_print_deadlock(w) do { } while (0) #define debug_rt_mutex_reset_waiter(w) do { } while (0) diff --git a/kernel/locking/rtmutex_common.h b/kernel/locking/rtmutex_common.h index 990134617b4c..72ad45a9a794 100644 --- a/kernel/locking/rtmutex_common.h +++ b/kernel/locking/rtmutex_common.h @@ -13,6 +13,7 @@ #define __KERNEL_RTMUTEX_COMMON_H #include <linux/rtmutex.h> +#include <linux/sched/wake_q.h> /* * This is the control structure for tasks blocked on a rt_mutex, @@ -33,6 +34,7 @@ struct rt_mutex_waiter { struct rt_mutex *deadlock_lock; #endif int prio; + u64 deadline; }; /* @@ -102,16 +104,26 @@ extern void rt_mutex_init_proxy_locked(struct rt_mutex *lock, struct task_struct *proxy_owner); extern void rt_mutex_proxy_unlock(struct rt_mutex *lock, struct task_struct *proxy_owner); +extern void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter); +extern int __rt_mutex_start_proxy_lock(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + struct task_struct *task); extern int rt_mutex_start_proxy_lock(struct rt_mutex *lock, struct rt_mutex_waiter *waiter, struct task_struct *task); -extern int rt_mutex_finish_proxy_lock(struct rt_mutex *lock, - struct hrtimer_sleeper *to, - struct rt_mutex_waiter *waiter); -extern int rt_mutex_timed_futex_lock(struct rt_mutex *l, struct hrtimer_sleeper *to); -extern bool rt_mutex_futex_unlock(struct rt_mutex *lock, - struct wake_q_head *wqh); -extern void rt_mutex_adjust_prio(struct task_struct *task); +extern int rt_mutex_wait_proxy_lock(struct rt_mutex *lock, + struct hrtimer_sleeper *to, + struct rt_mutex_waiter *waiter); +extern bool rt_mutex_cleanup_proxy_lock(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter); + +extern int rt_mutex_futex_trylock(struct rt_mutex *l); + +extern void rt_mutex_futex_unlock(struct rt_mutex *lock); +extern bool __rt_mutex_futex_unlock(struct rt_mutex *lock, + struct wake_q_head *wqh); + +extern void rt_mutex_postunlock(struct wake_q_head *wake_q); #ifdef CONFIG_DEBUG_RT_MUTEXES # include "rtmutex-debug.h" diff --git a/kernel/locking/rwsem-spinlock.c b/kernel/locking/rwsem-spinlock.c index 1591f6b3539f..20819df98125 100644 --- a/kernel/locking/rwsem-spinlock.c +++ b/kernel/locking/rwsem-spinlock.c @@ -6,7 +6,8 @@ * - Derived also from comments by Linus */ #include <linux/rwsem.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> #include <linux/export.h> enum rwsem_waiter_type { @@ -128,7 +129,6 @@ __rwsem_wake_one_writer(struct rw_semaphore *sem) void __sched __down_read(struct rw_semaphore *sem) { struct rwsem_waiter waiter; - struct task_struct *tsk; unsigned long flags; raw_spin_lock_irqsave(&sem->wait_lock, flags); @@ -140,13 +140,12 @@ void __sched __down_read(struct rw_semaphore *sem) goto out; } - tsk = current; - set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE); /* set up my own style of waitqueue */ - waiter.task = tsk; + waiter.task = current; waiter.type = RWSEM_WAITING_FOR_READ; - get_task_struct(tsk); + get_task_struct(current); list_add_tail(&waiter.list, &sem->wait_list); @@ -158,10 +157,10 @@ void __sched __down_read(struct rw_semaphore *sem) if (!waiter.task) break; schedule(); - set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE); } - __set_task_state(tsk, TASK_RUNNING); + __set_current_state(TASK_RUNNING); out: ; } @@ -194,15 +193,13 @@ int __down_read_trylock(struct rw_semaphore *sem) int __sched __down_write_common(struct rw_semaphore *sem, int state) { struct rwsem_waiter waiter; - struct task_struct *tsk; unsigned long flags; int ret = 0; raw_spin_lock_irqsave(&sem->wait_lock, flags); /* set up my own style of waitqueue */ - tsk = current; - waiter.task = tsk; + waiter.task = current; waiter.type = RWSEM_WAITING_FOR_WRITE; list_add_tail(&waiter.list, &sem->wait_list); @@ -216,23 +213,29 @@ int __sched __down_write_common(struct rw_semaphore *sem, int state) */ if (sem->count == 0) break; - if (signal_pending_state(state, current)) { - ret = -EINTR; - goto out; - } - set_task_state(tsk, state); + if (signal_pending_state(state, current)) + goto out_nolock; + + set_current_state(state); raw_spin_unlock_irqrestore(&sem->wait_lock, flags); schedule(); raw_spin_lock_irqsave(&sem->wait_lock, flags); } /* got the lock */ sem->count = -1; -out: list_del(&waiter.list); raw_spin_unlock_irqrestore(&sem->wait_lock, flags); return ret; + +out_nolock: + list_del(&waiter.list); + if (!list_empty(&sem->wait_list) && sem->count >= 0) + __rwsem_do_wake(sem, 0); + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); + + return -EINTR; } void __sched __down_write(struct rw_semaphore *sem) diff --git a/kernel/locking/rwsem-xadd.c b/kernel/locking/rwsem-xadd.c index 631506004f9e..34e727f18e49 100644 --- a/kernel/locking/rwsem-xadd.c +++ b/kernel/locking/rwsem-xadd.c @@ -10,10 +10,12 @@ * and Davidlohr Bueso <[email protected]>. Based on mutexes. */ #include <linux/rwsem.h> -#include <linux/sched.h> #include <linux/init.h> #include <linux/export.h> +#include <linux/sched/signal.h> #include <linux/sched/rt.h> +#include <linux/sched/wake_q.h> +#include <linux/sched/debug.h> #include <linux/osq_lock.h> #include "rwsem.h" @@ -224,10 +226,9 @@ struct rw_semaphore __sched *rwsem_down_read_failed(struct rw_semaphore *sem) { long count, adjustment = -RWSEM_ACTIVE_READ_BIAS; struct rwsem_waiter waiter; - struct task_struct *tsk = current; DEFINE_WAKE_Q(wake_q); - waiter.task = tsk; + waiter.task = current; waiter.type = RWSEM_WAITING_FOR_READ; raw_spin_lock_irq(&sem->wait_lock); @@ -254,13 +255,13 @@ struct rw_semaphore __sched *rwsem_down_read_failed(struct rw_semaphore *sem) /* wait to be given the lock */ while (true) { - set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE); if (!waiter.task) break; schedule(); } - __set_task_state(tsk, TASK_RUNNING); + __set_current_state(TASK_RUNNING); return sem; } EXPORT_SYMBOL(rwsem_down_read_failed); @@ -503,8 +504,6 @@ __rwsem_down_write_failed_common(struct rw_semaphore *sem, int state) * wake any read locks that were queued ahead of us. */ if (count > RWSEM_WAITING_BIAS) { - DEFINE_WAKE_Q(wake_q); - __rwsem_mark_wake(sem, RWSEM_WAKE_READERS, &wake_q); /* * The wakeup is normally called _after_ the wait_lock @@ -514,6 +513,11 @@ __rwsem_down_write_failed_common(struct rw_semaphore *sem, int state) * for attempting rwsem_try_write_lock(). */ wake_up_q(&wake_q); + + /* + * Reinitialize wake_q after use. + */ + wake_q_init(&wake_q); } } else diff --git a/kernel/locking/rwsem.c b/kernel/locking/rwsem.c index 45ba475d4be3..4d48b1c4870d 100644 --- a/kernel/locking/rwsem.c +++ b/kernel/locking/rwsem.c @@ -7,6 +7,7 @@ #include <linux/types.h> #include <linux/kernel.h> #include <linux/sched.h> +#include <linux/sched/debug.h> #include <linux/export.h> #include <linux/rwsem.h> #include <linux/atomic.h> @@ -123,10 +124,8 @@ EXPORT_SYMBOL(up_write); */ void downgrade_write(struct rw_semaphore *sem) { - /* - * lockdep: a downgraded write will live on as a write - * dependency. - */ + lock_downgrade(&sem->dep_map, _RET_IP_); + rwsem_set_reader_owned(sem); __downgrade_write(sem); } diff --git a/kernel/locking/semaphore.c b/kernel/locking/semaphore.c index b8120abe594b..561acdd39960 100644 --- a/kernel/locking/semaphore.c +++ b/kernel/locking/semaphore.c @@ -29,6 +29,7 @@ #include <linux/kernel.h> #include <linux/export.h> #include <linux/sched.h> +#include <linux/sched/debug.h> #include <linux/semaphore.h> #include <linux/spinlock.h> #include <linux/ftrace.h> @@ -204,19 +205,18 @@ struct semaphore_waiter { static inline int __sched __down_common(struct semaphore *sem, long state, long timeout) { - struct task_struct *task = current; struct semaphore_waiter waiter; list_add_tail(&waiter.list, &sem->wait_list); - waiter.task = task; + waiter.task = current; waiter.up = false; for (;;) { - if (signal_pending_state(state, task)) + if (signal_pending_state(state, current)) goto interrupted; if (unlikely(timeout <= 0)) goto timed_out; - __set_task_state(task, state); + __set_current_state(state); raw_spin_unlock_irq(&sem->lock); timeout = schedule_timeout(timeout); raw_spin_lock_irq(&sem->lock); diff --git a/kernel/locking/spinlock.c b/kernel/locking/spinlock.c index db3ccb1dd614..4b082b5cac9e 100644 --- a/kernel/locking/spinlock.c +++ b/kernel/locking/spinlock.c @@ -363,14 +363,6 @@ void __lockfunc _raw_spin_lock_nested(raw_spinlock_t *lock, int subclass) } EXPORT_SYMBOL(_raw_spin_lock_nested); -void __lockfunc _raw_spin_lock_bh_nested(raw_spinlock_t *lock, int subclass) -{ - __local_bh_disable_ip(_RET_IP_, SOFTIRQ_LOCK_OFFSET); - spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - LOCK_CONTENDED(lock, do_raw_spin_trylock, do_raw_spin_lock); -} -EXPORT_SYMBOL(_raw_spin_lock_bh_nested); - unsigned long __lockfunc _raw_spin_lock_irqsave_nested(raw_spinlock_t *lock, int subclass) { diff --git a/kernel/locking/spinlock_debug.c b/kernel/locking/spinlock_debug.c index 0374a596cffa..9aa0fccd5d43 100644 --- a/kernel/locking/spinlock_debug.c +++ b/kernel/locking/spinlock_debug.c @@ -103,38 +103,14 @@ static inline void debug_spin_unlock(raw_spinlock_t *lock) lock->owner_cpu = -1; } -static void __spin_lock_debug(raw_spinlock_t *lock) -{ - u64 i; - u64 loops = loops_per_jiffy * HZ; - - for (i = 0; i < loops; i++) { - if (arch_spin_trylock(&lock->raw_lock)) - return; - __delay(1); - } - /* lockup suspected: */ - spin_dump(lock, "lockup suspected"); -#ifdef CONFIG_SMP - trigger_all_cpu_backtrace(); -#endif - - /* - * The trylock above was causing a livelock. Give the lower level arch - * specific lock code a chance to acquire the lock. We have already - * printed a warning/backtrace at this point. The non-debug arch - * specific code might actually succeed in acquiring the lock. If it is - * not successful, the end-result is the same - there is no forward - * progress. - */ - arch_spin_lock(&lock->raw_lock); -} - +/* + * We are now relying on the NMI watchdog to detect lockup instead of doing + * the detection here with an unfair lock which can cause problem of its own. + */ void do_raw_spin_lock(raw_spinlock_t *lock) { debug_spin_lock_before(lock); - if (unlikely(!arch_spin_trylock(&lock->raw_lock))) - __spin_lock_debug(lock); + arch_spin_lock(&lock->raw_lock); debug_spin_lock_after(lock); } @@ -172,32 +148,6 @@ static void rwlock_bug(rwlock_t *lock, const char *msg) #define RWLOCK_BUG_ON(cond, lock, msg) if (unlikely(cond)) rwlock_bug(lock, msg) -#if 0 /* __write_lock_debug() can lock up - maybe this can too? */ -static void __read_lock_debug(rwlock_t *lock) -{ - u64 i; - u64 loops = loops_per_jiffy * HZ; - int print_once = 1; - - for (;;) { - for (i = 0; i < loops; i++) { - if (arch_read_trylock(&lock->raw_lock)) - return; - __delay(1); - } - /* lockup suspected: */ - if (print_once) { - print_once = 0; - printk(KERN_EMERG "BUG: read-lock lockup on CPU#%d, " - "%s/%d, %p\n", - raw_smp_processor_id(), current->comm, - current->pid, lock); - dump_stack(); - } - } -} -#endif - void do_raw_read_lock(rwlock_t *lock) { RWLOCK_BUG_ON(lock->magic != RWLOCK_MAGIC, lock, "bad magic"); @@ -247,32 +197,6 @@ static inline void debug_write_unlock(rwlock_t *lock) lock->owner_cpu = -1; } -#if 0 /* This can cause lockups */ -static void __write_lock_debug(rwlock_t *lock) -{ - u64 i; - u64 loops = loops_per_jiffy * HZ; - int print_once = 1; - - for (;;) { - for (i = 0; i < loops; i++) { - if (arch_write_trylock(&lock->raw_lock)) - return; - __delay(1); - } - /* lockup suspected: */ - if (print_once) { - print_once = 0; - printk(KERN_EMERG "BUG: write-lock lockup on CPU#%d, " - "%s/%d, %p\n", - raw_smp_processor_id(), current->comm, - current->pid, lock); - dump_stack(); - } - } -} -#endif - void do_raw_write_lock(rwlock_t *lock) { debug_write_lock_before(lock); diff --git a/kernel/locking/test-ww_mutex.c b/kernel/locking/test-ww_mutex.c new file mode 100644 index 000000000000..39f56c870051 --- /dev/null +++ b/kernel/locking/test-ww_mutex.c @@ -0,0 +1,645 @@ +/* + * Module-based API test facility for ww_mutexes + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, you can access it online at + * http://www.gnu.org/licenses/gpl-2.0.html. + */ + +#include <linux/kernel.h> + +#include <linux/completion.h> +#include <linux/delay.h> +#include <linux/kthread.h> +#include <linux/module.h> +#include <linux/random.h> +#include <linux/slab.h> +#include <linux/ww_mutex.h> + +static DEFINE_WW_CLASS(ww_class); +struct workqueue_struct *wq; + +struct test_mutex { + struct work_struct work; + struct ww_mutex mutex; + struct completion ready, go, done; + unsigned int flags; +}; + +#define TEST_MTX_SPIN BIT(0) +#define TEST_MTX_TRY BIT(1) +#define TEST_MTX_CTX BIT(2) +#define __TEST_MTX_LAST BIT(3) + +static void test_mutex_work(struct work_struct *work) +{ + struct test_mutex *mtx = container_of(work, typeof(*mtx), work); + + complete(&mtx->ready); + wait_for_completion(&mtx->go); + + if (mtx->flags & TEST_MTX_TRY) { + while (!ww_mutex_trylock(&mtx->mutex)) + cond_resched(); + } else { + ww_mutex_lock(&mtx->mutex, NULL); + } + complete(&mtx->done); + ww_mutex_unlock(&mtx->mutex); +} + +static int __test_mutex(unsigned int flags) +{ +#define TIMEOUT (HZ / 16) + struct test_mutex mtx; + struct ww_acquire_ctx ctx; + int ret; + + ww_mutex_init(&mtx.mutex, &ww_class); + ww_acquire_init(&ctx, &ww_class); + + INIT_WORK_ONSTACK(&mtx.work, test_mutex_work); + init_completion(&mtx.ready); + init_completion(&mtx.go); + init_completion(&mtx.done); + mtx.flags = flags; + + schedule_work(&mtx.work); + + wait_for_completion(&mtx.ready); + ww_mutex_lock(&mtx.mutex, (flags & TEST_MTX_CTX) ? &ctx : NULL); + complete(&mtx.go); + if (flags & TEST_MTX_SPIN) { + unsigned long timeout = jiffies + TIMEOUT; + + ret = 0; + do { + if (completion_done(&mtx.done)) { + ret = -EINVAL; + break; + } + cond_resched(); + } while (time_before(jiffies, timeout)); + } else { + ret = wait_for_completion_timeout(&mtx.done, TIMEOUT); + } + ww_mutex_unlock(&mtx.mutex); + ww_acquire_fini(&ctx); + + if (ret) { + pr_err("%s(flags=%x): mutual exclusion failure\n", + __func__, flags); + ret = -EINVAL; + } + + flush_work(&mtx.work); + destroy_work_on_stack(&mtx.work); + return ret; +#undef TIMEOUT +} + +static int test_mutex(void) +{ + int ret; + int i; + + for (i = 0; i < __TEST_MTX_LAST; i++) { + ret = __test_mutex(i); + if (ret) + return ret; + } + + return 0; +} + +static int test_aa(void) +{ + struct ww_mutex mutex; + struct ww_acquire_ctx ctx; + int ret; + + ww_mutex_init(&mutex, &ww_class); + ww_acquire_init(&ctx, &ww_class); + + ww_mutex_lock(&mutex, &ctx); + + if (ww_mutex_trylock(&mutex)) { + pr_err("%s: trylocked itself!\n", __func__); + ww_mutex_unlock(&mutex); + ret = -EINVAL; + goto out; + } + + ret = ww_mutex_lock(&mutex, &ctx); + if (ret != -EALREADY) { + pr_err("%s: missed deadlock for recursing, ret=%d\n", + __func__, ret); + if (!ret) + ww_mutex_unlock(&mutex); + ret = -EINVAL; + goto out; + } + + ret = 0; +out: + ww_mutex_unlock(&mutex); + ww_acquire_fini(&ctx); + return ret; +} + +struct test_abba { + struct work_struct work; + struct ww_mutex a_mutex; + struct ww_mutex b_mutex; + struct completion a_ready; + struct completion b_ready; + bool resolve; + int result; +}; + +static void test_abba_work(struct work_struct *work) +{ + struct test_abba *abba = container_of(work, typeof(*abba), work); + struct ww_acquire_ctx ctx; + int err; + + ww_acquire_init(&ctx, &ww_class); + ww_mutex_lock(&abba->b_mutex, &ctx); + + complete(&abba->b_ready); + wait_for_completion(&abba->a_ready); + + err = ww_mutex_lock(&abba->a_mutex, &ctx); + if (abba->resolve && err == -EDEADLK) { + ww_mutex_unlock(&abba->b_mutex); + ww_mutex_lock_slow(&abba->a_mutex, &ctx); + err = ww_mutex_lock(&abba->b_mutex, &ctx); + } + + if (!err) + ww_mutex_unlock(&abba->a_mutex); + ww_mutex_unlock(&abba->b_mutex); + ww_acquire_fini(&ctx); + + abba->result = err; +} + +static int test_abba(bool resolve) +{ + struct test_abba abba; + struct ww_acquire_ctx ctx; + int err, ret; + + ww_mutex_init(&abba.a_mutex, &ww_class); + ww_mutex_init(&abba.b_mutex, &ww_class); + INIT_WORK_ONSTACK(&abba.work, test_abba_work); + init_completion(&abba.a_ready); + init_completion(&abba.b_ready); + abba.resolve = resolve; + + schedule_work(&abba.work); + + ww_acquire_init(&ctx, &ww_class); + ww_mutex_lock(&abba.a_mutex, &ctx); + + complete(&abba.a_ready); + wait_for_completion(&abba.b_ready); + + err = ww_mutex_lock(&abba.b_mutex, &ctx); + if (resolve && err == -EDEADLK) { + ww_mutex_unlock(&abba.a_mutex); + ww_mutex_lock_slow(&abba.b_mutex, &ctx); + err = ww_mutex_lock(&abba.a_mutex, &ctx); + } + + if (!err) + ww_mutex_unlock(&abba.b_mutex); + ww_mutex_unlock(&abba.a_mutex); + ww_acquire_fini(&ctx); + + flush_work(&abba.work); + destroy_work_on_stack(&abba.work); + + ret = 0; + if (resolve) { + if (err || abba.result) { + pr_err("%s: failed to resolve ABBA deadlock, A err=%d, B err=%d\n", + __func__, err, abba.result); + ret = -EINVAL; + } + } else { + if (err != -EDEADLK && abba.result != -EDEADLK) { + pr_err("%s: missed ABBA deadlock, A err=%d, B err=%d\n", + __func__, err, abba.result); + ret = -EINVAL; + } + } + return ret; +} + +struct test_cycle { + struct work_struct work; + struct ww_mutex a_mutex; + struct ww_mutex *b_mutex; + struct completion *a_signal; + struct completion b_signal; + int result; +}; + +static void test_cycle_work(struct work_struct *work) +{ + struct test_cycle *cycle = container_of(work, typeof(*cycle), work); + struct ww_acquire_ctx ctx; + int err; + + ww_acquire_init(&ctx, &ww_class); + ww_mutex_lock(&cycle->a_mutex, &ctx); + + complete(cycle->a_signal); + wait_for_completion(&cycle->b_signal); + + err = ww_mutex_lock(cycle->b_mutex, &ctx); + if (err == -EDEADLK) { + ww_mutex_unlock(&cycle->a_mutex); + ww_mutex_lock_slow(cycle->b_mutex, &ctx); + err = ww_mutex_lock(&cycle->a_mutex, &ctx); + } + + if (!err) + ww_mutex_unlock(cycle->b_mutex); + ww_mutex_unlock(&cycle->a_mutex); + ww_acquire_fini(&ctx); + + cycle->result = err; +} + +static int __test_cycle(unsigned int nthreads) +{ + struct test_cycle *cycles; + unsigned int n, last = nthreads - 1; + int ret; + + cycles = kmalloc_array(nthreads, sizeof(*cycles), GFP_KERNEL); + if (!cycles) + return -ENOMEM; + + for (n = 0; n < nthreads; n++) { + struct test_cycle *cycle = &cycles[n]; + + ww_mutex_init(&cycle->a_mutex, &ww_class); + if (n == last) + cycle->b_mutex = &cycles[0].a_mutex; + else + cycle->b_mutex = &cycles[n + 1].a_mutex; + + if (n == 0) + cycle->a_signal = &cycles[last].b_signal; + else + cycle->a_signal = &cycles[n - 1].b_signal; + init_completion(&cycle->b_signal); + + INIT_WORK(&cycle->work, test_cycle_work); + cycle->result = 0; + } + + for (n = 0; n < nthreads; n++) + queue_work(wq, &cycles[n].work); + + flush_workqueue(wq); + + ret = 0; + for (n = 0; n < nthreads; n++) { + struct test_cycle *cycle = &cycles[n]; + + if (!cycle->result) + continue; + + pr_err("cylic deadlock not resolved, ret[%d/%d] = %d\n", + n, nthreads, cycle->result); + ret = -EINVAL; + break; + } + + for (n = 0; n < nthreads; n++) + ww_mutex_destroy(&cycles[n].a_mutex); + kfree(cycles); + return ret; +} + +static int test_cycle(unsigned int ncpus) +{ + unsigned int n; + int ret; + + for (n = 2; n <= ncpus + 1; n++) { + ret = __test_cycle(n); + if (ret) + return ret; + } + + return 0; +} + +struct stress { + struct work_struct work; + struct ww_mutex *locks; + unsigned long timeout; + int nlocks; +}; + +static int *get_random_order(int count) +{ + int *order; + int n, r, tmp; + + order = kmalloc_array(count, sizeof(*order), GFP_TEMPORARY); + if (!order) + return order; + + for (n = 0; n < count; n++) + order[n] = n; + + for (n = count - 1; n > 1; n--) { + r = get_random_int() % (n + 1); + if (r != n) { + tmp = order[n]; + order[n] = order[r]; + order[r] = tmp; + } + } + + return order; +} + +static void dummy_load(struct stress *stress) +{ + usleep_range(1000, 2000); +} + +static void stress_inorder_work(struct work_struct *work) +{ + struct stress *stress = container_of(work, typeof(*stress), work); + const int nlocks = stress->nlocks; + struct ww_mutex *locks = stress->locks; + struct ww_acquire_ctx ctx; + int *order; + + order = get_random_order(nlocks); + if (!order) + return; + + do { + int contended = -1; + int n, err; + + ww_acquire_init(&ctx, &ww_class); +retry: + err = 0; + for (n = 0; n < nlocks; n++) { + if (n == contended) + continue; + + err = ww_mutex_lock(&locks[order[n]], &ctx); + if (err < 0) + break; + } + if (!err) + dummy_load(stress); + + if (contended > n) + ww_mutex_unlock(&locks[order[contended]]); + contended = n; + while (n--) + ww_mutex_unlock(&locks[order[n]]); + + if (err == -EDEADLK) { + ww_mutex_lock_slow(&locks[order[contended]], &ctx); + goto retry; + } + + if (err) { + pr_err_once("stress (%s) failed with %d\n", + __func__, err); + break; + } + + ww_acquire_fini(&ctx); + } while (!time_after(jiffies, stress->timeout)); + + kfree(order); + kfree(stress); +} + +struct reorder_lock { + struct list_head link; + struct ww_mutex *lock; +}; + +static void stress_reorder_work(struct work_struct *work) +{ + struct stress *stress = container_of(work, typeof(*stress), work); + LIST_HEAD(locks); + struct ww_acquire_ctx ctx; + struct reorder_lock *ll, *ln; + int *order; + int n, err; + + order = get_random_order(stress->nlocks); + if (!order) + return; + + for (n = 0; n < stress->nlocks; n++) { + ll = kmalloc(sizeof(*ll), GFP_KERNEL); + if (!ll) + goto out; + + ll->lock = &stress->locks[order[n]]; + list_add(&ll->link, &locks); + } + kfree(order); + order = NULL; + + do { + ww_acquire_init(&ctx, &ww_class); + + list_for_each_entry(ll, &locks, link) { + err = ww_mutex_lock(ll->lock, &ctx); + if (!err) + continue; + + ln = ll; + list_for_each_entry_continue_reverse(ln, &locks, link) + ww_mutex_unlock(ln->lock); + + if (err != -EDEADLK) { + pr_err_once("stress (%s) failed with %d\n", + __func__, err); + break; + } + + ww_mutex_lock_slow(ll->lock, &ctx); + list_move(&ll->link, &locks); /* restarts iteration */ + } + + dummy_load(stress); + list_for_each_entry(ll, &locks, link) + ww_mutex_unlock(ll->lock); + + ww_acquire_fini(&ctx); + } while (!time_after(jiffies, stress->timeout)); + +out: + list_for_each_entry_safe(ll, ln, &locks, link) + kfree(ll); + kfree(order); + kfree(stress); +} + +static void stress_one_work(struct work_struct *work) +{ + struct stress *stress = container_of(work, typeof(*stress), work); + const int nlocks = stress->nlocks; + struct ww_mutex *lock = stress->locks + (get_random_int() % nlocks); + int err; + + do { + err = ww_mutex_lock(lock, NULL); + if (!err) { + dummy_load(stress); + ww_mutex_unlock(lock); + } else { + pr_err_once("stress (%s) failed with %d\n", + __func__, err); + break; + } + } while (!time_after(jiffies, stress->timeout)); + + kfree(stress); +} + +#define STRESS_INORDER BIT(0) +#define STRESS_REORDER BIT(1) +#define STRESS_ONE BIT(2) +#define STRESS_ALL (STRESS_INORDER | STRESS_REORDER | STRESS_ONE) + +static int stress(int nlocks, int nthreads, unsigned int flags) +{ + struct ww_mutex *locks; + int n; + + locks = kmalloc_array(nlocks, sizeof(*locks), GFP_KERNEL); + if (!locks) + return -ENOMEM; + + for (n = 0; n < nlocks; n++) + ww_mutex_init(&locks[n], &ww_class); + + for (n = 0; nthreads; n++) { + struct stress *stress; + void (*fn)(struct work_struct *work); + + fn = NULL; + switch (n & 3) { + case 0: + if (flags & STRESS_INORDER) + fn = stress_inorder_work; + break; + case 1: + if (flags & STRESS_REORDER) + fn = stress_reorder_work; + break; + case 2: + if (flags & STRESS_ONE) + fn = stress_one_work; + break; + } + + if (!fn) + continue; + + stress = kmalloc(sizeof(*stress), GFP_KERNEL); + if (!stress) + break; + + INIT_WORK(&stress->work, fn); + stress->locks = locks; + stress->nlocks = nlocks; + stress->timeout = jiffies + 2*HZ; + + queue_work(wq, &stress->work); + nthreads--; + } + + flush_workqueue(wq); + + for (n = 0; n < nlocks; n++) + ww_mutex_destroy(&locks[n]); + kfree(locks); + + return 0; +} + +static int __init test_ww_mutex_init(void) +{ + int ncpus = num_online_cpus(); + int ret; + + wq = alloc_workqueue("test-ww_mutex", WQ_UNBOUND, 0); + if (!wq) + return -ENOMEM; + + ret = test_mutex(); + if (ret) + return ret; + + ret = test_aa(); + if (ret) + return ret; + + ret = test_abba(false); + if (ret) + return ret; + + ret = test_abba(true); + if (ret) + return ret; + + ret = test_cycle(ncpus); + if (ret) + return ret; + + ret = stress(16, 2*ncpus, STRESS_INORDER); + if (ret) + return ret; + + ret = stress(16, 2*ncpus, STRESS_REORDER); + if (ret) + return ret; + + ret = stress(4095, hweight32(STRESS_ALL)*ncpus, STRESS_ALL); + if (ret) + return ret; + + return 0; +} + +static void __exit test_ww_mutex_exit(void) +{ + destroy_workqueue(wq); +} + +module_init(test_ww_mutex_init); +module_exit(test_ww_mutex_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Intel Corporation"); diff --git a/kernel/membarrier.c b/kernel/membarrier.c index 536c727a56e9..9f9284f37f8d 100644 --- a/kernel/membarrier.c +++ b/kernel/membarrier.c @@ -16,6 +16,7 @@ #include <linux/syscalls.h> #include <linux/membarrier.h> +#include <linux/tick.h> /* * Bitmask made from a "or" of all commands within enum membarrier_cmd, @@ -51,6 +52,9 @@ */ SYSCALL_DEFINE2(membarrier, int, cmd, int, flags) { + /* MEMBARRIER_CMD_SHARED is not compatible with nohz_full. */ + if (tick_nohz_full_enabled()) + return -ENOSYS; if (unlikely(flags)) return -EINVAL; switch (cmd) { diff --git a/kernel/memremap.c b/kernel/memremap.c index 9ecedc28b928..124bed776532 100644 --- a/kernel/memremap.c +++ b/kernel/memremap.c @@ -182,18 +182,6 @@ struct page_map { struct vmem_altmap altmap; }; -void get_zone_device_page(struct page *page) -{ - percpu_ref_get(page->pgmap->ref); -} -EXPORT_SYMBOL(get_zone_device_page); - -void put_zone_device_page(struct page *page) -{ - put_dev_pagemap(page->pgmap); -} -EXPORT_SYMBOL(put_zone_device_page); - static void pgmap_radix_release(struct resource *res) { resource_size_t key, align_start, align_size, align_end; @@ -237,6 +225,10 @@ static void devm_memremap_pages_release(struct device *dev, void *data) struct resource *res = &page_map->res; resource_size_t align_start, align_size; struct dev_pagemap *pgmap = &page_map->pgmap; + unsigned long pfn; + + for_each_device_pfn(pfn, page_map) + put_page(pfn_to_page(pfn)); if (percpu_ref_tryget_live(pgmap->ref)) { dev_WARN(dev, "%s: page mapping is still live!\n", __func__); @@ -246,9 +238,11 @@ static void devm_memremap_pages_release(struct device *dev, void *data) /* pages are dead and unused, undo the arch mapping */ align_start = res->start & ~(SECTION_SIZE - 1); align_size = ALIGN(resource_size(res), SECTION_SIZE); + mem_hotplug_begin(); arch_remove_memory(align_start, align_size); mem_hotplug_done(); + untrack_pfn(NULL, PHYS_PFN(align_start), align_size); pgmap_radix_release(res); dev_WARN_ONCE(dev, pgmap->altmap && pgmap->altmap->alloc, @@ -275,7 +269,10 @@ struct dev_pagemap *find_dev_pagemap(resource_size_t phys) * * Notes: * 1/ @ref must be 'live' on entry and 'dead' before devm_memunmap_pages() time - * (or devm release event). + * (or devm release event). The expected order of events is that @ref has + * been through percpu_ref_kill() before devm_memremap_pages_release(). The + * wait for the completion of all references being dropped and + * percpu_ref_exit() must occur after devm_memremap_pages_release(). * * 2/ @res is expected to be a host memory range that could feasibly be * treated as a "System RAM" range, i.e. not a device mmio range, but @@ -361,7 +358,11 @@ void *devm_memremap_pages(struct device *dev, struct resource *res, goto err_pfn_remap; mem_hotplug_begin(); - error = arch_add_memory(nid, align_start, align_size, true); + error = arch_add_memory(nid, align_start, align_size, false); + if (!error) + move_pfn_range_to_zone(&NODE_DATA(nid)->node_zones[ZONE_DEVICE], + align_start >> PAGE_SHIFT, + align_size >> PAGE_SHIFT); mem_hotplug_done(); if (error) goto err_add_memory; @@ -377,6 +378,7 @@ void *devm_memremap_pages(struct device *dev, struct resource *res, */ list_del(&page->lru); page->pgmap = pgmap; + percpu_ref_get(ref); } devres_add(dev, page_map); return __va(res->start); diff --git a/kernel/module.c b/kernel/module.c index 3d8f126208e3..40f983cbea81 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -17,6 +17,7 @@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ #include <linux/export.h> +#include <linux/extable.h> #include <linux/moduleloader.h> #include <linux/trace_events.h> #include <linux/init.h> @@ -48,6 +49,7 @@ #include <linux/rculist.h> #include <linux/uaccess.h> #include <asm/cacheflush.h> +#include <linux/set_memory.h> #include <asm/mmu_context.h> #include <linux/license.h> #include <asm/sections.h> @@ -61,6 +63,7 @@ #include <linux/pfn.h> #include <linux/bsearch.h> #include <linux/dynamic_debug.h> +#include <linux/audit.h> #include <uapi/linux/module.h> #include "module-internal.h" @@ -74,9 +77,9 @@ /* * Modules' sections will be aligned on page boundaries * to ensure complete separation of code and data, but - * only when CONFIG_DEBUG_SET_MODULE_RONX=y + * only when CONFIG_STRICT_MODULE_RWX=y */ -#ifdef CONFIG_DEBUG_SET_MODULE_RONX +#ifdef CONFIG_STRICT_MODULE_RWX # define debug_align(X) ALIGN(X, PAGE_SIZE) #else # define debug_align(X) (X) @@ -297,6 +300,7 @@ int unregister_module_notifier(struct notifier_block *nb) EXPORT_SYMBOL(unregister_module_notifier); struct load_info { + const char *name; Elf_Ehdr *hdr; unsigned long len; Elf_Shdr *sechdrs; @@ -597,7 +601,7 @@ static struct module *find_module_all(const char *name, size_t len, module_assert_mutex_or_preempt(); - list_for_each_entry(mod, &modules, list) { + list_for_each_entry_rcu(mod, &modules, list) { if (!even_unformed && mod->state == MODULE_STATE_UNFORMED) continue; if (strlen(mod->name) == len && !memcmp(mod->name, name, len)) @@ -663,16 +667,7 @@ static void percpu_modcopy(struct module *mod, memcpy(per_cpu_ptr(mod->percpu, cpu), from, size); } -/** - * is_module_percpu_address - test whether address is from module static percpu - * @addr: address to test - * - * Test whether @addr belongs to module static percpu area. - * - * RETURNS: - * %true if @addr is from module static percpu area - */ -bool is_module_percpu_address(unsigned long addr) +bool __is_module_percpu_address(unsigned long addr, unsigned long *can_addr) { struct module *mod; unsigned int cpu; @@ -686,9 +681,15 @@ bool is_module_percpu_address(unsigned long addr) continue; for_each_possible_cpu(cpu) { void *start = per_cpu_ptr(mod->percpu, cpu); - - if ((void *)addr >= start && - (void *)addr < start + mod->percpu_size) { + void *va = (void *)addr; + + if (va >= start && va < start + mod->percpu_size) { + if (can_addr) { + *can_addr = (unsigned long) (va - start); + *can_addr += (unsigned long) + per_cpu_ptr(mod->percpu, + get_boot_cpu_id()); + } preempt_enable(); return true; } @@ -699,6 +700,20 @@ bool is_module_percpu_address(unsigned long addr) return false; } +/** + * is_module_percpu_address - test whether address is from module static percpu + * @addr: address to test + * + * Test whether @addr belongs to module static percpu area. + * + * RETURNS: + * %true if @addr is from module static percpu area + */ +bool is_module_percpu_address(unsigned long addr) +{ + return __is_module_percpu_address(addr, NULL); +} + #else /* ... !CONFIG_SMP */ static inline void __percpu *mod_percpu(struct module *mod) @@ -730,6 +745,11 @@ bool is_module_percpu_address(unsigned long addr) return false; } +bool __is_module_percpu_address(unsigned long addr, unsigned long *can_addr) +{ + return false; +} + #endif /* CONFIG_SMP */ #define MODINFO_ATTR(field) \ @@ -945,6 +965,8 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user, return -EFAULT; name[MODULE_NAME_LEN-1] = '\0'; + audit_log_kern_module(name); + if (mutex_lock_interruptible(&module_mutex) != 0) return -EINTR; @@ -1179,10 +1201,7 @@ static ssize_t store_uevent(struct module_attribute *mattr, struct module_kobject *mk, const char *buffer, size_t count) { - enum kobject_action action; - - if (kobject_action_type(buffer, count, &action) == 0) - kobject_uevent(&mk->kobj, action); + kobject_synth_uevent(&mk->kobj, buffer, count); return count; } @@ -1255,12 +1274,13 @@ static u32 resolve_rel_crc(const s32 *crc) return *(u32 *)((void *)crc + *crc); } -static int check_version(Elf_Shdr *sechdrs, - unsigned int versindex, +static int check_version(const struct load_info *info, const char *symname, struct module *mod, const s32 *crc) { + Elf_Shdr *sechdrs = info->sechdrs; + unsigned int versindex = info->index.vers; unsigned int i, num_versions; struct modversion_info *versions; @@ -1294,17 +1314,16 @@ static int check_version(Elf_Shdr *sechdrs, } /* Broken toolchain. Warn once, then let it go.. */ - pr_warn_once("%s: no symbol version for %s\n", mod->name, symname); + pr_warn_once("%s: no symbol version for %s\n", info->name, symname); return 1; bad_version: pr_warn("%s: disagrees about version of symbol %s\n", - mod->name, symname); + info->name, symname); return 0; } -static inline int check_modstruct_version(Elf_Shdr *sechdrs, - unsigned int versindex, +static inline int check_modstruct_version(const struct load_info *info, struct module *mod) { const s32 *crc; @@ -1320,8 +1339,8 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, BUG(); } preempt_enable(); - return check_version(sechdrs, versindex, - VMLINUX_SYMBOL_STR(module_layout), mod, crc); + return check_version(info, VMLINUX_SYMBOL_STR(module_layout), + mod, crc); } /* First part is kernel version, which we ignore if module has crcs. */ @@ -1335,8 +1354,7 @@ static inline int same_magic(const char *amagic, const char *bmagic, return strcmp(amagic, bmagic) == 0; } #else -static inline int check_version(Elf_Shdr *sechdrs, - unsigned int versindex, +static inline int check_version(const struct load_info *info, const char *symname, struct module *mod, const s32 *crc) @@ -1344,8 +1362,7 @@ static inline int check_version(Elf_Shdr *sechdrs, return 1; } -static inline int check_modstruct_version(Elf_Shdr *sechdrs, - unsigned int versindex, +static inline int check_modstruct_version(const struct load_info *info, struct module *mod) { return 1; @@ -1381,7 +1398,7 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod, if (!sym) goto unlock; - if (!check_version(info->sechdrs, info->index.vers, name, mod, crc)) { + if (!check_version(info, name, mod, crc)) { sym = ERR_PTR(-EINVAL); goto getname; } @@ -1644,31 +1661,36 @@ static inline void remove_notes_attrs(struct module *mod) } #endif /* CONFIG_KALLSYMS */ -static void add_usage_links(struct module *mod) +static void del_usage_links(struct module *mod) { #ifdef CONFIG_MODULE_UNLOAD struct module_use *use; - int nowarn; mutex_lock(&module_mutex); - list_for_each_entry(use, &mod->target_list, target_list) { - nowarn = sysfs_create_link(use->target->holders_dir, - &mod->mkobj.kobj, mod->name); - } + list_for_each_entry(use, &mod->target_list, target_list) + sysfs_remove_link(use->target->holders_dir, mod->name); mutex_unlock(&module_mutex); #endif } -static void del_usage_links(struct module *mod) +static int add_usage_links(struct module *mod) { + int ret = 0; #ifdef CONFIG_MODULE_UNLOAD struct module_use *use; mutex_lock(&module_mutex); - list_for_each_entry(use, &mod->target_list, target_list) - sysfs_remove_link(use->target->holders_dir, mod->name); + list_for_each_entry(use, &mod->target_list, target_list) { + ret = sysfs_create_link(use->target->holders_dir, + &mod->mkobj.kobj, mod->name); + if (ret) + break; + } mutex_unlock(&module_mutex); + if (ret) + del_usage_links(mod); #endif + return ret; } static int module_add_modinfo_attrs(struct module *mod) @@ -1779,13 +1801,18 @@ static int mod_sysfs_setup(struct module *mod, if (err) goto out_unreg_param; - add_usage_links(mod); + err = add_usage_links(mod); + if (err) + goto out_unreg_modinfo_attrs; + add_sect_attrs(mod, info); add_notes_attrs(mod, info); kobject_uevent(&mod->mkobj.kobj, KOBJ_ADD); return 0; +out_unreg_modinfo_attrs: + module_remove_modinfo_attrs(mod); out_unreg_param: module_param_sysfs_remove(mod); out_unreg_holders: @@ -1844,7 +1871,7 @@ static void mod_sysfs_teardown(struct module *mod) mod_sysfs_fini(mod); } -#ifdef CONFIG_DEBUG_SET_MODULE_RONX +#ifdef CONFIG_STRICT_MODULE_RWX /* * LKM RO/NX protection: protect module's text/ro-data * from modification and any data from execution. @@ -2809,6 +2836,8 @@ static int check_modinfo_livepatch(struct module *mod, struct load_info *info) if (get_modinfo(info, "livepatch")) { mod->klp = true; add_taint_module(mod, TAINT_LIVEPATCH, LOCKDEP_STILL_OK); + pr_notice_once("%s: tainting kernel with TAINT_LIVEPATCH\n", + mod->name); } return 0; @@ -2842,7 +2871,7 @@ static int copy_module_from_user(const void __user *umod, unsigned long len, /* Suck in entire file: we'll want most of it. */ info->hdr = __vmalloc(info->len, - GFP_KERNEL | __GFP_HIGHMEM | __GFP_NOWARN, PAGE_KERNEL); + GFP_KERNEL | __GFP_NOWARN, PAGE_KERNEL); if (!info->hdr) return -ENOMEM; @@ -2890,9 +2919,15 @@ static int rewrite_section_headers(struct load_info *info, int flags) info->index.vers = 0; /* Pretend no __versions section! */ else info->index.vers = find_sec(info, "__versions"); + info->sechdrs[info->index.vers].sh_flags &= ~(unsigned long)SHF_ALLOC; + info->index.info = find_sec(info, ".modinfo"); + if (!info->index.info) + info->name = "(missing .modinfo section)"; + else + info->name = get_modinfo(info, "name"); info->sechdrs[info->index.info].sh_flags &= ~(unsigned long)SHF_ALLOC; - info->sechdrs[info->index.vers].sh_flags &= ~(unsigned long)SHF_ALLOC; + return 0; } @@ -2932,21 +2967,29 @@ static struct module *setup_load_info(struct load_info *info, int flags) info->index.mod = find_sec(info, ".gnu.linkonce.this_module"); if (!info->index.mod) { - pr_warn("No module found in object\n"); + pr_warn("%s: No module found in object\n", + info->name ?: "(missing .modinfo name field)"); return ERR_PTR(-ENOEXEC); } /* This is temporary: point mod into copy of data. */ mod = (void *)info->sechdrs[info->index.mod].sh_addr; + /* + * If we didn't load the .modinfo 'name' field, fall back to + * on-disk struct mod 'name' field. + */ + if (!info->name) + info->name = mod->name; + if (info->index.sym == 0) { - pr_warn("%s: module has no symbols (stripped?)\n", mod->name); + pr_warn("%s: module has no symbols (stripped?)\n", info->name); return ERR_PTR(-ENOEXEC); } info->index.pcpu = find_pcpusec(info); /* Check module struct version now, before we try to use module. */ - if (!check_modstruct_version(info->sechdrs, info->index.vers, mod)) + if (!check_modstruct_version(info, mod)) return ERR_PTR(-ENOEXEC); return mod; @@ -2967,7 +3010,7 @@ static int check_modinfo(struct module *mod, struct load_info *info, int flags) return err; } else if (!same_magic(modmagic, vermagic, info->index.vers)) { pr_err("%s: version magic '%s' should be '%s'\n", - mod->name, modmagic, vermagic); + info->name, modmagic, vermagic); return -ENOEXEC; } @@ -3052,9 +3095,9 @@ static int find_module_sections(struct module *mod, struct load_info *info) mod->trace_events = section_objs(info, "_ftrace_events", sizeof(*mod->trace_events), &mod->num_trace_events); - mod->trace_enums = section_objs(info, "_ftrace_enum_map", - sizeof(*mod->trace_enums), - &mod->num_trace_enums); + mod->trace_evals = section_objs(info, "_ftrace_eval_map", + sizeof(*mod->trace_evals), + &mod->num_trace_evals); #endif #ifdef CONFIG_TRACING mod->trace_bprintk_fmt_start = section_objs(info, "__trace_printk_fmt", @@ -3217,7 +3260,7 @@ int __weak module_frob_arch_sections(Elf_Ehdr *hdr, /* module_blacklist is a comma-separated list of module names */ static char *module_blacklist; -static bool blacklisted(char *module_name) +static bool blacklisted(const char *module_name) { const char *p; size_t len; @@ -3247,7 +3290,7 @@ static struct module *layout_and_allocate(struct load_info *info, int flags) if (IS_ERR(mod)) return mod; - if (blacklisted(mod->name)) + if (blacklisted(info->name)) return ERR_PTR(-EPERM); err = check_modinfo(mod, info, flags); @@ -3608,6 +3651,8 @@ static int load_module(struct load_info *info, const char __user *uargs, goto free_copy; } + audit_log_kern_module(mod->name); + /* Reserve our place in the list. */ err = add_unformed_module(mod); if (err) @@ -3696,7 +3741,7 @@ static int load_module(struct load_info *info, const char __user *uargs, mod->name, after_dashes); } - /* Link in to syfs. */ + /* Link in to sysfs. */ err = mod_sysfs_setup(mod, info, mod->kp, mod->num_kp); if (err < 0) goto coming_cleanup; @@ -3719,6 +3764,7 @@ static int load_module(struct load_info *info, const char __user *uargs, mod_sysfs_teardown(mod); coming_cleanup: mod->state = MODULE_STATE_GOING; + destroy_params(mod->kp, mod->num_kp); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); klp_module_going(mod); @@ -4010,7 +4056,7 @@ unsigned long module_kallsyms_lookup_name(const char *name) /* Don't lock: we're in enough trouble already. */ preempt_disable(); - if ((colon = strchr(name, ':')) != NULL) { + if ((colon = strnchr(name, MODULE_NAME_LEN, ':')) != NULL) { if ((mod = find_module_all(name, colon - name, false)) != NULL) ret = mod_find_symname(mod, colon+1); } else { @@ -4165,22 +4211,23 @@ const struct exception_table_entry *search_module_extables(unsigned long addr) struct module *mod; preempt_disable(); - list_for_each_entry_rcu(mod, &modules, list) { - if (mod->state == MODULE_STATE_UNFORMED) - continue; - if (mod->num_exentries == 0) - continue; + mod = __module_address(addr); + if (!mod) + goto out; - e = search_extable(mod->extable, - mod->extable + mod->num_exentries - 1, - addr); - if (e) - break; - } + if (!mod->num_exentries) + goto out; + + e = search_extable(mod->extable, + mod->num_exentries, + addr); +out: preempt_enable(); - /* Now, if we found one, we are running inside it now, hence - we cannot unload the module, hence no refcnt needed. */ + /* + * Now, if we found one, we are running inside it now, hence + * we cannot unload the module, hence no refcnt needed. + */ return e; } diff --git a/kernel/notifier.c b/kernel/notifier.c index fd2c9acbcc19..6196af8a8223 100644 --- a/kernel/notifier.c +++ b/kernel/notifier.c @@ -95,7 +95,7 @@ static int notifier_call_chain(struct notifier_block **nl, if (nr_calls) (*nr_calls)++; - if ((ret & NOTIFY_STOP_MASK) == NOTIFY_STOP_MASK) + if (ret & NOTIFY_STOP_MASK) break; nb = next_nb; nr_to_call--; diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 782102e59eed..f6c5d330059a 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -26,6 +26,7 @@ #include <linux/file.h> #include <linux/syscalls.h> #include <linux/cgroup.h> +#include <linux/perf_event.h> static struct kmem_cache *nsproxy_cachep; @@ -262,6 +263,8 @@ SYSCALL_DEFINE2(setns, int, fd, int, nstype) goto out; } switch_task_namespaces(tsk, new_nsproxy); + + perf_event_namespaces(tsk); out: fput(file); return err; diff --git a/kernel/padata.c b/kernel/padata.c index 05316c9f32da..868f947166d7 100644 --- a/kernel/padata.c +++ b/kernel/padata.c @@ -154,8 +154,6 @@ EXPORT_SYMBOL(padata_do_parallel); * A pointer to the control struct of the next object that needs * serialization, if present in one of the percpu reorder queues. * - * NULL, if all percpu reorder queues are empty. - * * -EINPROGRESS, if the next object that needs serialization will * be parallel processed by another cpu and is not yet present in * the cpu's reorder queue. @@ -182,23 +180,22 @@ static struct padata_priv *padata_get_next(struct parallel_data *pd) cpu = padata_index_to_cpu(pd, next_index); next_queue = per_cpu_ptr(pd->pqueue, cpu); - padata = NULL; - reorder = &next_queue->reorder; + spin_lock(&reorder->lock); if (!list_empty(&reorder->list)) { padata = list_entry(reorder->list.next, struct padata_priv, list); - spin_lock(&reorder->lock); list_del_init(&padata->list); atomic_dec(&pd->reorder_objects); - spin_unlock(&reorder->lock); pd->processed++; + spin_unlock(&reorder->lock); goto out; } + spin_unlock(&reorder->lock); if (__this_cpu_read(pd->pqueue->cpu_index) == next_queue->cpu_index) { padata = ERR_PTR(-ENODATA); @@ -234,12 +231,11 @@ static void padata_reorder(struct parallel_data *pd) padata = padata_get_next(pd); /* - * All reorder queues are empty, or the next object that needs - * serialization is parallel processed by another cpu and is - * still on it's way to the cpu's reorder queue, nothing to - * do for now. + * If the next object that needs serialization is parallel + * processed by another cpu and is still on it's way to the + * cpu's reorder queue, nothing to do for now. */ - if (!padata || PTR_ERR(padata) == -EINPROGRESS) + if (PTR_ERR(padata) == -EINPROGRESS) break; /* @@ -353,7 +349,7 @@ static int padata_setup_cpumasks(struct parallel_data *pd, cpumask_and(pd->cpumask.pcpu, pcpumask, cpu_online_mask); if (!alloc_cpumask_var(&pd->cpumask.cbcpu, GFP_KERNEL)) { - free_cpumask_var(pd->cpumask.cbcpu); + free_cpumask_var(pd->cpumask.pcpu); return -ENOMEM; } @@ -938,29 +934,18 @@ static struct kobj_type padata_attr_type = { }; /** - * padata_alloc_possible - Allocate and initialize padata instance. - * Use the cpu_possible_mask for serial and - * parallel workers. - * - * @wq: workqueue to use for the allocated padata instance - */ -struct padata_instance *padata_alloc_possible(struct workqueue_struct *wq) -{ - return padata_alloc(wq, cpu_possible_mask, cpu_possible_mask); -} -EXPORT_SYMBOL(padata_alloc_possible); - -/** * padata_alloc - allocate and initialize a padata instance and specify * cpumasks for serial and parallel workers. * * @wq: workqueue to use for the allocated padata instance * @pcpumask: cpumask that will be used for padata parallelization * @cbcpumask: cpumask that will be used for padata serialization + * + * Must be called from a cpus_read_lock() protected region */ -struct padata_instance *padata_alloc(struct workqueue_struct *wq, - const struct cpumask *pcpumask, - const struct cpumask *cbcpumask) +static struct padata_instance *padata_alloc(struct workqueue_struct *wq, + const struct cpumask *pcpumask, + const struct cpumask *cbcpumask) { struct padata_instance *pinst; struct parallel_data *pd = NULL; @@ -969,7 +954,6 @@ struct padata_instance *padata_alloc(struct workqueue_struct *wq, if (!pinst) goto err; - get_online_cpus(); if (!alloc_cpumask_var(&pinst->cpumask.pcpu, GFP_KERNEL)) goto err_free_inst; if (!alloc_cpumask_var(&pinst->cpumask.cbcpu, GFP_KERNEL)) { @@ -993,14 +977,12 @@ struct padata_instance *padata_alloc(struct workqueue_struct *wq, pinst->flags = 0; - put_online_cpus(); - BLOCKING_INIT_NOTIFIER_HEAD(&pinst->cpumask_change_notifier); kobject_init(&pinst->kobj, &padata_attr_type); mutex_init(&pinst->lock); #ifdef CONFIG_HOTPLUG_CPU - cpuhp_state_add_instance_nocalls(hp_online, &pinst->node); + cpuhp_state_add_instance_nocalls_cpuslocked(hp_online, &pinst->node); #endif return pinst; @@ -1009,12 +991,27 @@ err_free_masks: free_cpumask_var(pinst->cpumask.cbcpu); err_free_inst: kfree(pinst); - put_online_cpus(); err: return NULL; } /** + * padata_alloc_possible - Allocate and initialize padata instance. + * Use the cpu_possible_mask for serial and + * parallel workers. + * + * @wq: workqueue to use for the allocated padata instance + * + * Must be called from a cpus_read_lock() protected region + */ +struct padata_instance *padata_alloc_possible(struct workqueue_struct *wq) +{ + lockdep_assert_cpus_held(); + return padata_alloc(wq, cpu_possible_mask, cpu_possible_mask); +} +EXPORT_SYMBOL(padata_alloc_possible); + +/** * padata_free - free a padata instance * * @padata_inst: padata instance to free diff --git a/kernel/panic.c b/kernel/panic.c index 08aa88dde7de..a58932b41700 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -9,6 +9,7 @@ * to indicate a major problem. */ #include <linux/debug_locks.h> +#include <linux/sched/debug.h> #include <linux/interrupt.h> #include <linux/kmsg_dump.h> #include <linux/kallsyms.h> @@ -188,7 +189,7 @@ void panic(const char *fmt, ...) * Bypass the panic_cpu check and call __crash_kexec directly. */ if (!_crash_kexec_post_notifiers) { - printk_nmi_flush_on_panic(); + printk_safe_flush_on_panic(); __crash_kexec(NULL); /* @@ -213,7 +214,7 @@ void panic(const char *fmt, ...) atomic_notifier_call_chain(&panic_notifier_list, 0, buf); /* Call flush even twice. It tries harder with a single online CPU */ - printk_nmi_flush_on_panic(); + printk_safe_flush_on_panic(); kmsg_dump(KMSG_DUMP_PANIC); /* @@ -273,7 +274,8 @@ void panic(const char *fmt, ...) extern int stop_a_enabled; /* Make sure the user can actually press Stop-A (L1-A) */ stop_a_enabled = 1; - pr_emerg("Press Stop-A (L1-A) to return to the boot prom\n"); + pr_emerg("Press Stop-A (L1-A) from sun keyboard or send break\n" + "twice on console to return to the boot prom\n"); } #endif #if defined(CONFIG_S390) diff --git a/kernel/params.c b/kernel/params.c index a6d6149c0fe6..60b2d8101355 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -160,58 +160,6 @@ static int parse_one(char *param, return -ENOENT; } -/* You can use " around spaces, but can't escape ". */ -/* Hyphens and underscores equivalent in parameter names. */ -static char *next_arg(char *args, char **param, char **val) -{ - unsigned int i, equals = 0; - int in_quote = 0, quoted = 0; - char *next; - - if (*args == '"') { - args++; - in_quote = 1; - quoted = 1; - } - - for (i = 0; args[i]; i++) { - if (isspace(args[i]) && !in_quote) - break; - if (equals == 0) { - if (args[i] == '=') - equals = i; - } - if (args[i] == '"') - in_quote = !in_quote; - } - - *param = args; - if (!equals) - *val = NULL; - else { - args[equals] = '\0'; - *val = args + equals + 1; - - /* Don't include quotes in value. */ - if (**val == '"') { - (*val)++; - if (args[i-1] == '"') - args[i-1] = '\0'; - } - } - if (quoted && args[i-1] == '"') - args[i-1] = '\0'; - - if (args[i]) { - args[i] = '\0'; - next = args + i + 1; - } else - next = args + i; - - /* Chew up trailing spaces. */ - return skip_spaces(next); -} - /* Args looks like "foo=bar,bar2 baz=fuz wiz". */ char *parse_args(const char *doing, char *args, diff --git a/kernel/pid.c b/kernel/pid.c index f66162f2359b..731c4e528f4e 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -38,6 +38,7 @@ #include <linux/syscalls.h> #include <linux/proc_ns.h> #include <linux/proc_fs.h> +#include <linux/sched/task.h> #define pid_hashfn(nr, ns) \ hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift) @@ -68,9 +69,7 @@ static inline int mk_pid(struct pid_namespace *pid_ns, * the scheme scales to up to 4 million PIDs, runtime. */ struct pid_namespace init_pid_ns = { - .kref = { - .refcount = ATOMIC_INIT(2), - }, + .kref = KREF_INIT(2), .pidmap = { [ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL } }, @@ -322,8 +321,10 @@ struct pid *alloc_pid(struct pid_namespace *ns) } if (unlikely(is_child_reaper(pid))) { - if (pid_ns_prepare_proc(ns)) + if (pid_ns_prepare_proc(ns)) { + disable_pid_allocation(ns); goto out_free; + } } get_pid_ns(ns); @@ -574,16 +575,13 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns) */ void __init pidhash_init(void) { - unsigned int i, pidhash_size; + unsigned int pidhash_size; pid_hash = alloc_large_system_hash("PID", sizeof(*pid_hash), 0, 18, - HASH_EARLY | HASH_SMALL, + HASH_EARLY | HASH_SMALL | HASH_ZERO, &pidhash_shift, NULL, 0, 4096); pidhash_size = 1U << pidhash_shift; - - for (i = 0; i < pidhash_size; i++) - INIT_HLIST_HEAD(&pid_hash[i]); } void __init pidmap_init(void) diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index eef2ce968636..74a5a7255b4d 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -12,12 +12,15 @@ #include <linux/pid_namespace.h> #include <linux/user_namespace.h> #include <linux/syscalls.h> +#include <linux/cred.h> #include <linux/err.h> #include <linux/acct.h> #include <linux/slab.h> #include <linux/proc_ns.h> #include <linux/reboot.h> #include <linux/export.h> +#include <linux/sched/task.h> +#include <linux/sched/signal.h> struct pid_cache { int nr_ids; @@ -274,7 +277,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) * if reparented. */ for (;;) { - set_current_state(TASK_UNINTERRUPTIBLE); + set_current_state(TASK_INTERRUPTIBLE); if (pid_ns->nr_hashed == init_pids) break; schedule(); @@ -371,6 +374,29 @@ static struct ns_common *pidns_get(struct task_struct *task) return ns ? &ns->ns : NULL; } +static struct ns_common *pidns_for_children_get(struct task_struct *task) +{ + struct pid_namespace *ns = NULL; + + task_lock(task); + if (task->nsproxy) { + ns = task->nsproxy->pid_ns_for_children; + get_pid_ns(ns); + } + task_unlock(task); + + if (ns) { + read_lock(&tasklist_lock); + if (!ns->child_reaper) { + put_pid_ns(ns); + ns = NULL; + } + read_unlock(&tasklist_lock); + } + + return ns ? &ns->ns : NULL; +} + static void pidns_put(struct ns_common *ns) { put_pid_ns(to_pid_ns(ns)); @@ -440,6 +466,17 @@ const struct proc_ns_operations pidns_operations = { .get_parent = pidns_get_parent, }; +const struct proc_ns_operations pidns_for_children_operations = { + .name = "pid_for_children", + .real_ns_name = "pid", + .type = CLONE_NEWPID, + .get = pidns_for_children_get, + .put = pidns_put, + .install = pidns_install, + .owner = pidns_owner, + .get_parent = pidns_get_parent, +}; + static __init int pid_namespaces_init(void) { pid_ns_cachep = KMEM_CACHE(pid_namespace, SLAB_PANIC); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index b26dbc48c75b..e1914c7b85b1 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -10,6 +10,8 @@ * This file is released under the GPLv2. */ +#define pr_fmt(fmt) "PM: " fmt + #include <linux/export.h> #include <linux/suspend.h> #include <linux/syscalls.h> @@ -21,6 +23,7 @@ #include <linux/fs.h> #include <linux/mount.h> #include <linux/pm.h> +#include <linux/nmi.h> #include <linux/console.h> #include <linux/cpu.h> #include <linux/freezer.h> @@ -104,7 +107,7 @@ EXPORT_SYMBOL(system_entering_hibernation); #ifdef CONFIG_PM_DEBUG static void hibernation_debug_sleep(void) { - printk(KERN_INFO "hibernation debug: Waiting for 5 seconds.\n"); + pr_info("hibernation debug: Waiting for 5 seconds.\n"); mdelay(5000); } @@ -250,10 +253,9 @@ void swsusp_show_speed(ktime_t start, ktime_t stop, centisecs = 1; /* avoid div-by-zero */ k = nr_pages * (PAGE_SIZE / 1024); kps = (k * 100) / centisecs; - printk(KERN_INFO "PM: %s %u kbytes in %u.%02u seconds (%u.%02u MB/s)\n", - msg, k, - centisecs / 100, centisecs % 100, - kps / 1000, (kps % 1000) / 10); + pr_info("%s %u kbytes in %u.%02u seconds (%u.%02u MB/s)\n", + msg, k, centisecs / 100, centisecs % 100, kps / 1000, + (kps % 1000) / 10); } /** @@ -271,8 +273,7 @@ static int create_image(int platform_mode) error = dpm_suspend_end(PMSG_FREEZE); if (error) { - printk(KERN_ERR "PM: Some devices failed to power down, " - "aborting hibernation\n"); + pr_err("Some devices failed to power down, aborting hibernation\n"); return error; } @@ -288,8 +289,7 @@ static int create_image(int platform_mode) error = syscore_suspend(); if (error) { - printk(KERN_ERR "PM: Some system devices failed to power down, " - "aborting hibernation\n"); + pr_err("Some system devices failed to power down, aborting hibernation\n"); goto Enable_irqs; } @@ -304,8 +304,8 @@ static int create_image(int platform_mode) restore_processor_state(); trace_suspend_resume(TPS("machine_suspend"), PM_EVENT_HIBERNATE, false); if (error) - printk(KERN_ERR "PM: Error %d creating hibernation image\n", - error); + pr_err("Error %d creating hibernation image\n", error); + if (!in_suspend) { events_check_enabled = false; clear_free_pages(); @@ -432,8 +432,7 @@ static int resume_target_kernel(bool platform_mode) error = dpm_suspend_end(PMSG_QUIESCE); if (error) { - printk(KERN_ERR "PM: Some devices failed to power down, " - "aborting resume\n"); + pr_err("Some devices failed to power down, aborting resume\n"); return error; } @@ -608,6 +607,22 @@ static void power_down(void) { #ifdef CONFIG_SUSPEND int error; + + if (hibernation_mode == HIBERNATION_SUSPEND) { + error = suspend_devices_and_enter(PM_SUSPEND_MEM); + if (error) { + hibernation_mode = hibernation_ops ? + HIBERNATION_PLATFORM : + HIBERNATION_SHUTDOWN; + } else { + /* Restore swap signature. */ + error = swsusp_unmark(); + if (error) + pr_err("Swap will be unusable! Try swapon -a.\n"); + + return; + } + } #endif switch (hibernation_mode) { @@ -620,32 +635,13 @@ static void power_down(void) if (pm_power_off) kernel_power_off(); break; -#ifdef CONFIG_SUSPEND - case HIBERNATION_SUSPEND: - error = suspend_devices_and_enter(PM_SUSPEND_MEM); - if (error) { - if (hibernation_ops) - hibernation_mode = HIBERNATION_PLATFORM; - else - hibernation_mode = HIBERNATION_SHUTDOWN; - power_down(); - } - /* - * Restore swap signature. - */ - error = swsusp_unmark(); - if (error) - printk(KERN_ERR "PM: Swap will be unusable! " - "Try swapon -a.\n"); - return; -#endif } kernel_halt(); /* * Valid image is on the disk, if we continue we risk serious data * corruption after resume. */ - printk(KERN_CRIT "PM: Please power down manually\n"); + pr_crit("Power down manually\n"); while (1) cpu_relax(); } @@ -655,7 +651,7 @@ static int load_image_and_restore(void) int error; unsigned int flags; - pr_debug("PM: Loading hibernation image.\n"); + pr_debug("Loading hibernation image.\n"); lock_device_hotplug(); error = create_basic_memory_bitmaps(); @@ -667,7 +663,7 @@ static int load_image_and_restore(void) if (!error) hibernation_restore(flags & SF_PLATFORM_MODE); - printk(KERN_ERR "PM: Failed to load hibernation image, recovering.\n"); + pr_err("Failed to load hibernation image, recovering.\n"); swsusp_free(); free_basic_memory_bitmaps(); Unlock: @@ -685,7 +681,7 @@ int hibernate(void) bool snapshot_test = false; if (!hibernation_available()) { - pr_debug("PM: Hibernation not available.\n"); + pr_debug("Hibernation not available.\n"); return -EPERM; } @@ -703,9 +699,9 @@ int hibernate(void) goto Exit; } - printk(KERN_INFO "PM: Syncing filesystems ... "); + pr_info("Syncing filesystems ... \n"); sys_sync(); - printk("done.\n"); + pr_info("done.\n"); error = freeze_processes(); if (error) @@ -731,7 +727,7 @@ int hibernate(void) else flags |= SF_CRC32_MODE; - pr_debug("PM: writing image.\n"); + pr_debug("Writing image.\n"); error = swsusp_write(flags); swsusp_free(); if (!error) { @@ -743,7 +739,7 @@ int hibernate(void) in_suspend = 0; pm_restore_gfp_mask(); } else { - pr_debug("PM: Image restored successfully.\n"); + pr_debug("Image restored successfully.\n"); } Free_bitmaps: @@ -751,7 +747,7 @@ int hibernate(void) Thaw: unlock_device_hotplug(); if (snapshot_test) { - pr_debug("PM: Checking hibernation image\n"); + pr_debug("Checking hibernation image\n"); error = swsusp_check(); if (!error) error = load_image_and_restore(); @@ -815,10 +811,10 @@ static int software_resume(void) goto Unlock; } - pr_debug("PM: Checking hibernation image partition %s\n", resume_file); + pr_debug("Checking hibernation image partition %s\n", resume_file); if (resume_delay) { - printk(KERN_INFO "Waiting %dsec before reading resume device...\n", + pr_info("Waiting %dsec before reading resume device ...\n", resume_delay); ssleep(resume_delay); } @@ -857,10 +853,10 @@ static int software_resume(void) } Check_image: - pr_debug("PM: Hibernation image partition %d:%d present\n", + pr_debug("Hibernation image partition %d:%d present\n", MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device)); - pr_debug("PM: Looking for hibernation image.\n"); + pr_debug("Looking for hibernation image.\n"); error = swsusp_check(); if (error) goto Unlock; @@ -879,7 +875,7 @@ static int software_resume(void) goto Close_Finish; } - pr_debug("PM: Preparing processes for restore.\n"); + pr_debug("Preparing processes for restore.\n"); error = freeze_processes(); if (error) goto Close_Finish; @@ -892,7 +888,7 @@ static int software_resume(void) /* For success case, the suspend path will release the lock */ Unlock: mutex_unlock(&pm_mutex); - pr_debug("PM: Hibernation image not present or could not be loaded.\n"); + pr_debug("Hibernation image not present or could not be loaded.\n"); return error; Close_Finish: swsusp_close(FMODE_READ); @@ -1016,7 +1012,7 @@ static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr, error = -EINVAL; if (!error) - pr_debug("PM: Hibernation mode set to '%s'\n", + pr_debug("Hibernation mode set to '%s'\n", hibernation_modes[mode]); unlock_system_sleep(); return error ? error : n; @@ -1052,7 +1048,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr, lock_system_sleep(); swsusp_resume_device = res; unlock_system_sleep(); - printk(KERN_INFO "PM: Starting manual resume from disk\n"); + pr_info("Starting manual resume from disk\n"); noresume = 0; software_resume(); return n; @@ -1112,7 +1108,7 @@ static struct attribute * g[] = { }; -static struct attribute_group attr_group = { +static const struct attribute_group attr_group = { .attrs = g, }; @@ -1156,7 +1152,7 @@ static int __init hibernate_setup(char *str) } else if (!strncmp(str, "no", 2)) { noresume = 1; nohibernate = 1; - } else if (IS_ENABLED(CONFIG_DEBUG_RODATA) + } else if (IS_ENABLED(CONFIG_STRICT_KERNEL_RWX) && !strncmp(str, "protect_image", 13)) { enable_restore_image_protection(); } diff --git a/kernel/power/main.c b/kernel/power/main.c index d401c21136d1..42bd800a6755 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -705,7 +705,7 @@ static struct attribute * g[] = { NULL, }; -static struct attribute_group attr_group = { +static const struct attribute_group attr_group = { .attrs = g, }; diff --git a/kernel/power/power.h b/kernel/power/power.h index 1dfa0da827d3..7fdc40d31b7d 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -61,12 +61,12 @@ extern int hibernation_snapshot(int platform_mode); extern int hibernation_restore(int platform_mode); extern int hibernation_platform_enter(void); -#ifdef CONFIG_DEBUG_RODATA +#ifdef CONFIG_STRICT_KERNEL_RWX /* kernel/power/snapshot.c */ extern void enable_restore_image_protection(void); #else static inline void enable_restore_image_protection(void) {} -#endif /* CONFIG_DEBUG_RODATA */ +#endif /* CONFIG_STRICT_KERNEL_RWX */ #else /* !CONFIG_HIBERNATION */ diff --git a/kernel/power/process.c b/kernel/power/process.c index 2fba066e125f..78672d324a6e 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -12,6 +12,8 @@ #include <linux/oom.h> #include <linux/suspend.h> #include <linux/module.h> +#include <linux/sched/debug.h> +#include <linux/sched/task.h> #include <linux/syscalls.h> #include <linux/freezer.h> #include <linux/delay.h> @@ -130,7 +132,7 @@ int freeze_processes(void) if (!pm_freezing) atomic_inc(&system_freezing_cnt); - pm_wakeup_clear(); + pm_wakeup_clear(true); pr_info("Freezing user space processes ... "); pm_freezing = true; error = try_to_freeze_tasks(true); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 2d8e2b227db8..222317721c5a 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -22,6 +22,7 @@ #include <linux/device.h> #include <linux/init.h> #include <linux/bootmem.h> +#include <linux/nmi.h> #include <linux/syscalls.h> #include <linux/console.h> #include <linux/highmem.h> @@ -29,6 +30,7 @@ #include <linux/slab.h> #include <linux/compiler.h> #include <linux/ktime.h> +#include <linux/set_memory.h> #include <linux/uaccess.h> #include <asm/mmu_context.h> @@ -38,7 +40,7 @@ #include "power.h" -#ifdef CONFIG_DEBUG_RODATA +#if defined(CONFIG_STRICT_KERNEL_RWX) && defined(CONFIG_ARCH_HAS_SET_MEMORY) static bool hibernate_restore_protection; static bool hibernate_restore_protection_active; @@ -73,7 +75,7 @@ static inline void hibernate_restore_protection_begin(void) {} static inline void hibernate_restore_protection_end(void) {} static inline void hibernate_restore_protect_page(void *page_address) {} static inline void hibernate_restore_unprotect_page(void *page_address) {} -#endif /* CONFIG_DEBUG_RODATA */ +#endif /* CONFIG_STRICT_KERNEL_RWX && CONFIG_ARCH_HAS_SET_MEMORY */ static int swsusp_page_is_free(struct page *); static void swsusp_set_page_forbidden(struct page *); @@ -1421,7 +1423,7 @@ static unsigned int nr_meta_pages; * Numbers of normal and highmem page frames allocated for hibernation image * before suspending devices. */ -unsigned int alloc_normal, alloc_highmem; +static unsigned int alloc_normal, alloc_highmem; /* * Memory bitmap used for marking saveable pages (during hibernation) or * hibernation image pages (during restore) @@ -1925,8 +1927,7 @@ static inline unsigned int alloc_highmem_pages(struct memory_bitmap *bm, * also be located in the high memory, because of the way in which * copy_data_pages() works. */ -static int swsusp_alloc(struct memory_bitmap *orig_bm, - struct memory_bitmap *copy_bm, +static int swsusp_alloc(struct memory_bitmap *copy_bm, unsigned int nr_pages, unsigned int nr_highmem) { if (nr_highmem > 0) { @@ -1972,7 +1973,7 @@ asmlinkage __visible int swsusp_save(void) return -ENOMEM; } - if (swsusp_alloc(&orig_bm, ©_bm, nr_pages, nr_highmem)) { + if (swsusp_alloc(©_bm, nr_pages, nr_highmem)) { printk(KERN_ERR "PM: Memory allocation failed\n"); return -ENOMEM; } diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 15e6baef5c73..3ecf275d7e44 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -72,6 +72,8 @@ static void freeze_begin(void) static void freeze_enter(void) { + trace_suspend_resume(TPS("machine_suspend"), PM_SUSPEND_FREEZE, true); + spin_lock_irq(&suspend_freeze_lock); if (pm_wakeup_pending()) goto out; @@ -84,11 +86,9 @@ static void freeze_enter(void) /* Push all the CPUs into the idle loop. */ wake_up_all_idle_cpus(); - pr_debug("PM: suspend-to-idle\n"); /* Make the current CPU wait so it can enter the idle loop too. */ wait_event(suspend_freeze_wait_head, suspend_freeze_state == FREEZE_STATE_WAKE); - pr_debug("PM: resume from suspend-to-idle\n"); cpuidle_pause(); put_online_cpus(); @@ -98,6 +98,31 @@ static void freeze_enter(void) out: suspend_freeze_state = FREEZE_STATE_NONE; spin_unlock_irq(&suspend_freeze_lock); + + trace_suspend_resume(TPS("machine_suspend"), PM_SUSPEND_FREEZE, false); +} + +static void s2idle_loop(void) +{ + pr_debug("PM: suspend-to-idle\n"); + + do { + freeze_enter(); + + if (freeze_ops && freeze_ops->wake) + freeze_ops->wake(); + + dpm_resume_noirq(PMSG_RESUME); + if (freeze_ops && freeze_ops->sync) + freeze_ops->sync(); + + if (pm_wakeup_pending()) + break; + + pm_wakeup_clear(false); + } while (!dpm_suspend_noirq(PMSG_SUSPEND)); + + pr_debug("PM: resume from suspend-to-idle\n"); } void freeze_wake(void) @@ -371,10 +396,8 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) * all the devices are suspended. */ if (state == PM_SUSPEND_FREEZE) { - trace_suspend_resume(TPS("machine_suspend"), state, true); - freeze_enter(); - trace_suspend_resume(TPS("machine_suspend"), state, false); - goto Platform_wake; + s2idle_loop(); + goto Platform_early_resume; } error = disable_nonboot_cpus(); diff --git a/kernel/power/suspend_test.c b/kernel/power/suspend_test.c index bdff5ed57f10..5db217051232 100644 --- a/kernel/power/suspend_test.c +++ b/kernel/power/suspend_test.c @@ -166,7 +166,7 @@ static int __init setup_test_suspend(char *value) return 0; } - for (i = 0; pm_labels[i]; i++) + for (i = PM_SUSPEND_MIN; i < PM_SUSPEND_MAX; i++) if (!strcmp(pm_labels[i], suspend_type)) { test_state_label = pm_labels[i]; return 0; diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 32e0c232efba..57d22571f306 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -201,7 +201,7 @@ void free_all_swap_pages(int swap) struct swsusp_extent *ext; unsigned long offset; - ext = container_of(node, struct swsusp_extent, node); + ext = rb_entry(node, struct swsusp_extent, node); rb_erase(node, &swsusp_extents); for (offset = ext->start; offset <= ext->end; offset++) swap_free(swp_entry(swap, offset)); @@ -225,14 +225,14 @@ static struct block_device *hib_resume_bdev; struct hib_bio_batch { atomic_t count; wait_queue_head_t wait; - int error; + blk_status_t error; }; static void hib_init_batch(struct hib_bio_batch *hb) { atomic_set(&hb->count, 0); init_waitqueue_head(&hb->wait); - hb->error = 0; + hb->error = BLK_STS_OK; } static void hib_end_io(struct bio *bio) @@ -240,7 +240,7 @@ static void hib_end_io(struct bio *bio) struct hib_bio_batch *hb = bio->bi_private; struct page *page = bio->bi_io_vec[0].bv_page; - if (bio->bi_error) { + if (bio->bi_status) { printk(KERN_ALERT "Read-error on swap-device (%u:%u:%Lu)\n", imajor(bio->bi_bdev->bd_inode), iminor(bio->bi_bdev->bd_inode), @@ -253,8 +253,8 @@ static void hib_end_io(struct bio *bio) flush_icache_range((unsigned long)page_address(page), (unsigned long)page_address(page) + PAGE_SIZE); - if (bio->bi_error && !hb->error) - hb->error = bio->bi_error; + if (bio->bi_status && !hb->error) + hb->error = bio->bi_status; if (atomic_dec_and_test(&hb->count)) wake_up(&hb->wait); @@ -293,10 +293,10 @@ static int hib_submit_io(int op, int op_flags, pgoff_t page_off, void *addr, return error; } -static int hib_wait_io(struct hib_bio_batch *hb) +static blk_status_t hib_wait_io(struct hib_bio_batch *hb) { wait_event(hb->wait, atomic_read(&hb->count) == 0); - return hb->error; + return blk_status_to_errno(hb->error); } /* diff --git a/kernel/printk/Makefile b/kernel/printk/Makefile index abb0042a427b..4a2ffc39eb95 100644 --- a/kernel/printk/Makefile +++ b/kernel/printk/Makefile @@ -1,3 +1,3 @@ obj-y = printk.o -obj-$(CONFIG_PRINTK_NMI) += nmi.o +obj-$(CONFIG_PRINTK) += printk_safe.o obj-$(CONFIG_A11Y_BRAILLE_CONSOLE) += braille.o diff --git a/kernel/printk/braille.c b/kernel/printk/braille.c index d5760c42f042..61d41ca41844 100644 --- a/kernel/printk/braille.c +++ b/kernel/printk/braille.c @@ -2,12 +2,13 @@ #include <linux/kernel.h> #include <linux/console.h> +#include <linux/errno.h> #include <linux/string.h> #include "console_cmdline.h" #include "braille.h" -char *_braille_console_setup(char **str, char **brl_options) +int _braille_console_setup(char **str, char **brl_options) { if (!strncmp(*str, "brl,", 4)) { *brl_options = ""; @@ -15,14 +16,14 @@ char *_braille_console_setup(char **str, char **brl_options) } else if (!strncmp(*str, "brl=", 4)) { *brl_options = *str + 4; *str = strchr(*brl_options, ','); - if (!*str) + if (!*str) { pr_err("need port name after brl=\n"); - else - *((*str)++) = 0; - } else - return NULL; + return -EINVAL; + } + *((*str)++) = 0; + } - return *str; + return 0; } int diff --git a/kernel/printk/braille.h b/kernel/printk/braille.h index 769d771145c8..749a6756843a 100644 --- a/kernel/printk/braille.h +++ b/kernel/printk/braille.h @@ -9,7 +9,14 @@ braille_set_options(struct console_cmdline *c, char *brl_options) c->brl_options = brl_options; } -char * +/* + * Setup console according to braille options. + * Return -EINVAL on syntax error, 0 on success (or no braille option was + * actually given). + * Modifies str to point to the serial options + * Sets brl_options to the parsed braille options. + */ +int _braille_console_setup(char **str, char **brl_options); int @@ -25,10 +32,10 @@ braille_set_options(struct console_cmdline *c, char *brl_options) { } -static inline char * +static inline int _braille_console_setup(char **str, char **brl_options) { - return NULL; + return 0; } static inline int diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h index 7fd2838fa417..2a7d04049af4 100644 --- a/kernel/printk/internal.h +++ b/kernel/printk/internal.h @@ -16,42 +16,57 @@ */ #include <linux/percpu.h> -typedef __printf(1, 0) int (*printk_func_t)(const char *fmt, va_list args); +#ifdef CONFIG_PRINTK -int __printf(1, 0) vprintk_default(const char *fmt, va_list args); - -#ifdef CONFIG_PRINTK_NMI +#define PRINTK_SAFE_CONTEXT_MASK 0x3fffffff +#define PRINTK_NMI_DEFERRED_CONTEXT_MASK 0x40000000 +#define PRINTK_NMI_CONTEXT_MASK 0x80000000 extern raw_spinlock_t logbuf_lock; +__printf(1, 0) int vprintk_default(const char *fmt, va_list args); +__printf(1, 0) int vprintk_deferred(const char *fmt, va_list args); +__printf(1, 0) int vprintk_func(const char *fmt, va_list args); +void __printk_safe_enter(void); +void __printk_safe_exit(void); + +#define printk_safe_enter_irqsave(flags) \ + do { \ + local_irq_save(flags); \ + __printk_safe_enter(); \ + } while (0) + +#define printk_safe_exit_irqrestore(flags) \ + do { \ + __printk_safe_exit(); \ + local_irq_restore(flags); \ + } while (0) + +#define printk_safe_enter_irq() \ + do { \ + local_irq_disable(); \ + __printk_safe_enter(); \ + } while (0) + +#define printk_safe_exit_irq() \ + do { \ + __printk_safe_exit(); \ + local_irq_enable(); \ + } while (0) + +#else + +__printf(1, 0) int vprintk_func(const char *fmt, va_list args) { return 0; } + /* - * printk() could not take logbuf_lock in NMI context. Instead, - * it temporary stores the strings into a per-CPU buffer. - * The alternative implementation is chosen transparently - * via per-CPU variable. + * In !PRINTK builds we still export logbuf_lock spin_lock, console_sem + * semaphore and some of console functions (console_unlock()/etc.), so + * printk-safe must preserve the existing local IRQ guarantees. */ -DECLARE_PER_CPU(printk_func_t, printk_func); -static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) -{ - return this_cpu_read(printk_func)(fmt, args); -} - -extern atomic_t nmi_message_lost; -static inline int get_nmi_message_lost(void) -{ - return atomic_xchg(&nmi_message_lost, 0); -} - -#else /* CONFIG_PRINTK_NMI */ - -static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) -{ - return vprintk_default(fmt, args); -} - -static inline int get_nmi_message_lost(void) -{ - return 0; -} - -#endif /* CONFIG_PRINTK_NMI */ +#define printk_safe_enter_irqsave(flags) local_irq_save(flags) +#define printk_safe_exit_irqrestore(flags) local_irq_restore(flags) + +#define printk_safe_enter_irq() local_irq_disable() +#define printk_safe_exit_irq() local_irq_enable() + +#endif /* CONFIG_PRINTK */ diff --git a/kernel/printk/nmi.c b/kernel/printk/nmi.c deleted file mode 100644 index f011aaef583c..000000000000 --- a/kernel/printk/nmi.c +++ /dev/null @@ -1,291 +0,0 @@ -/* - * nmi.c - Safe printk in NMI context - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, see <http://www.gnu.org/licenses/>. - */ - -#include <linux/preempt.h> -#include <linux/spinlock.h> -#include <linux/debug_locks.h> -#include <linux/smp.h> -#include <linux/cpumask.h> -#include <linux/irq_work.h> -#include <linux/printk.h> - -#include "internal.h" - -/* - * printk() could not take logbuf_lock in NMI context. Instead, - * it uses an alternative implementation that temporary stores - * the strings into a per-CPU buffer. The content of the buffer - * is later flushed into the main ring buffer via IRQ work. - * - * The alternative implementation is chosen transparently - * via @printk_func per-CPU variable. - * - * The implementation allows to flush the strings also from another CPU. - * There are situations when we want to make sure that all buffers - * were handled or when IRQs are blocked. - */ -DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default; -static int printk_nmi_irq_ready; -atomic_t nmi_message_lost; - -#define NMI_LOG_BUF_LEN ((1 << CONFIG_NMI_LOG_BUF_SHIFT) - \ - sizeof(atomic_t) - sizeof(struct irq_work)) - -struct nmi_seq_buf { - atomic_t len; /* length of written data */ - struct irq_work work; /* IRQ work that flushes the buffer */ - unsigned char buffer[NMI_LOG_BUF_LEN]; -}; -static DEFINE_PER_CPU(struct nmi_seq_buf, nmi_print_seq); - -/* - * Safe printk() for NMI context. It uses a per-CPU buffer to - * store the message. NMIs are not nested, so there is always only - * one writer running. But the buffer might get flushed from another - * CPU, so we need to be careful. - */ -static int vprintk_nmi(const char *fmt, va_list args) -{ - struct nmi_seq_buf *s = this_cpu_ptr(&nmi_print_seq); - int add = 0; - size_t len; - -again: - len = atomic_read(&s->len); - - /* The trailing '\0' is not counted into len. */ - if (len >= sizeof(s->buffer) - 1) { - atomic_inc(&nmi_message_lost); - return 0; - } - - /* - * Make sure that all old data have been read before the buffer was - * reseted. This is not needed when we just append data. - */ - if (!len) - smp_rmb(); - - add = vscnprintf(s->buffer + len, sizeof(s->buffer) - len, fmt, args); - - /* - * Do it once again if the buffer has been flushed in the meantime. - * Note that atomic_cmpxchg() is an implicit memory barrier that - * makes sure that the data were written before updating s->len. - */ - if (atomic_cmpxchg(&s->len, len, len + add) != len) - goto again; - - /* Get flushed in a more safe context. */ - if (add && printk_nmi_irq_ready) { - /* Make sure that IRQ work is really initialized. */ - smp_rmb(); - irq_work_queue(&s->work); - } - - return add; -} - -static void printk_nmi_flush_line(const char *text, int len) -{ - /* - * The buffers are flushed in NMI only on panic. The messages must - * go only into the ring buffer at this stage. Consoles will get - * explicitly called later when a crashdump is not generated. - */ - if (in_nmi()) - printk_deferred("%.*s", len, text); - else - printk("%.*s", len, text); - -} - -/* printk part of the temporary buffer line by line */ -static int printk_nmi_flush_buffer(const char *start, size_t len) -{ - const char *c, *end; - bool header; - - c = start; - end = start + len; - header = true; - - /* Print line by line. */ - while (c < end) { - if (*c == '\n') { - printk_nmi_flush_line(start, c - start + 1); - start = ++c; - header = true; - continue; - } - - /* Handle continuous lines or missing new line. */ - if ((c + 1 < end) && printk_get_level(c)) { - if (header) { - c = printk_skip_level(c); - continue; - } - - printk_nmi_flush_line(start, c - start); - start = c++; - header = true; - continue; - } - - header = false; - c++; - } - - /* Check if there was a partial line. Ignore pure header. */ - if (start < end && !header) { - static const char newline[] = KERN_CONT "\n"; - - printk_nmi_flush_line(start, end - start); - printk_nmi_flush_line(newline, strlen(newline)); - } - - return len; -} - -/* - * Flush data from the associated per_CPU buffer. The function - * can be called either via IRQ work or independently. - */ -static void __printk_nmi_flush(struct irq_work *work) -{ - static raw_spinlock_t read_lock = - __RAW_SPIN_LOCK_INITIALIZER(read_lock); - struct nmi_seq_buf *s = container_of(work, struct nmi_seq_buf, work); - unsigned long flags; - size_t len; - int i; - - /* - * The lock has two functions. First, one reader has to flush all - * available message to make the lockless synchronization with - * writers easier. Second, we do not want to mix messages from - * different CPUs. This is especially important when printing - * a backtrace. - */ - raw_spin_lock_irqsave(&read_lock, flags); - - i = 0; -more: - len = atomic_read(&s->len); - - /* - * This is just a paranoid check that nobody has manipulated - * the buffer an unexpected way. If we printed something then - * @len must only increase. Also it should never overflow the - * buffer size. - */ - if ((i && i >= len) || len > sizeof(s->buffer)) { - const char *msg = "printk_nmi_flush: internal error\n"; - - printk_nmi_flush_line(msg, strlen(msg)); - len = 0; - } - - if (!len) - goto out; /* Someone else has already flushed the buffer. */ - - /* Make sure that data has been written up to the @len */ - smp_rmb(); - i += printk_nmi_flush_buffer(s->buffer + i, len - i); - - /* - * Check that nothing has got added in the meantime and truncate - * the buffer. Note that atomic_cmpxchg() is an implicit memory - * barrier that makes sure that the data were copied before - * updating s->len. - */ - if (atomic_cmpxchg(&s->len, len, 0) != len) - goto more; - -out: - raw_spin_unlock_irqrestore(&read_lock, flags); -} - -/** - * printk_nmi_flush - flush all per-cpu nmi buffers. - * - * The buffers are flushed automatically via IRQ work. This function - * is useful only when someone wants to be sure that all buffers have - * been flushed at some point. - */ -void printk_nmi_flush(void) -{ - int cpu; - - for_each_possible_cpu(cpu) - __printk_nmi_flush(&per_cpu(nmi_print_seq, cpu).work); -} - -/** - * printk_nmi_flush_on_panic - flush all per-cpu nmi buffers when the system - * goes down. - * - * Similar to printk_nmi_flush() but it can be called even in NMI context when - * the system goes down. It does the best effort to get NMI messages into - * the main ring buffer. - * - * Note that it could try harder when there is only one CPU online. - */ -void printk_nmi_flush_on_panic(void) -{ - /* - * Make sure that we could access the main ring buffer. - * Do not risk a double release when more CPUs are up. - */ - if (in_nmi() && raw_spin_is_locked(&logbuf_lock)) { - if (num_online_cpus() > 1) - return; - - debug_locks_off(); - raw_spin_lock_init(&logbuf_lock); - } - - printk_nmi_flush(); -} - -void __init printk_nmi_init(void) -{ - int cpu; - - for_each_possible_cpu(cpu) { - struct nmi_seq_buf *s = &per_cpu(nmi_print_seq, cpu); - - init_irq_work(&s->work, __printk_nmi_flush); - } - - /* Make sure that IRQ works are initialized before enabling. */ - smp_wmb(); - printk_nmi_irq_ready = 1; - - /* Flush pending messages that did not have scheduled IRQ works. */ - printk_nmi_flush(); -} - -void printk_nmi_enter(void) -{ - this_cpu_write(printk_func, vprintk_nmi); -} - -void printk_nmi_exit(void) -{ - this_cpu_write(printk_func, vprintk_default); -} diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 8b2696420abb..fc47863f629c 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -32,7 +32,7 @@ #include <linux/bootmem.h> #include <linux/memblock.h> #include <linux/syscalls.h> -#include <linux/kexec.h> +#include <linux/crash_core.h> #include <linux/kdb.h> #include <linux/ratelimit.h> #include <linux/kmsg_dump.h> @@ -45,6 +45,9 @@ #include <linux/utsname.h> #include <linux/ctype.h> #include <linux/uio.h> +#include <linux/sched/clock.h> +#include <linux/sched/debug.h> +#include <linux/sched/task_stack.h> #include <linux/uaccess.h> #include <asm/sections.h> @@ -213,17 +216,36 @@ static int nr_ext_console_drivers; static int __down_trylock_console_sem(unsigned long ip) { - if (down_trylock(&console_sem)) + int lock_failed; + unsigned long flags; + + /* + * Here and in __up_console_sem() we need to be in safe mode, + * because spindump/WARN/etc from under console ->lock will + * deadlock in printk()->down_trylock_console_sem() otherwise. + */ + printk_safe_enter_irqsave(flags); + lock_failed = down_trylock(&console_sem); + printk_safe_exit_irqrestore(flags); + + if (lock_failed) return 1; mutex_acquire(&console_lock_dep_map, 0, 1, ip); return 0; } #define down_trylock_console_sem() __down_trylock_console_sem(_RET_IP_) -#define up_console_sem() do { \ - mutex_release(&console_lock_dep_map, 1, _RET_IP_);\ - up(&console_sem);\ -} while (0) +static void __up_console_sem(unsigned long ip) +{ + unsigned long flags; + + mutex_release(&console_lock_dep_map, 1, ip); + + printk_safe_enter_irqsave(flags); + up(&console_sem); + printk_safe_exit_irqrestore(flags); +} +#define up_console_sem() __up_console_sem(_RET_IP_) /* * This is used for debugging the mess that is the VT code by @@ -248,7 +270,6 @@ static struct console *exclusive_console; static struct console_cmdline console_cmdline[MAX_CMDLINECONSOLES]; -static int selected_console = -1; static int preferred_console = -1; int console_set_on_cmdline; EXPORT_SYMBOL(console_set_on_cmdline); @@ -351,6 +372,34 @@ __packed __aligned(4) */ DEFINE_RAW_SPINLOCK(logbuf_lock); +/* + * Helper macros to lock/unlock logbuf_lock and switch between + * printk-safe/unsafe modes. + */ +#define logbuf_lock_irq() \ + do { \ + printk_safe_enter_irq(); \ + raw_spin_lock(&logbuf_lock); \ + } while (0) + +#define logbuf_unlock_irq() \ + do { \ + raw_spin_unlock(&logbuf_lock); \ + printk_safe_exit_irq(); \ + } while (0) + +#define logbuf_lock_irqsave(flags) \ + do { \ + printk_safe_enter_irqsave(flags); \ + raw_spin_lock(&logbuf_lock); \ + } while (0) + +#define logbuf_unlock_irqrestore(flags) \ + do { \ + raw_spin_unlock(&logbuf_lock); \ + printk_safe_exit_irqrestore(flags); \ + } while (0) + #ifdef CONFIG_PRINTK DECLARE_WAIT_QUEUE_HEAD(log_wait); /* the next printk record to read by syslog(READ) or /proc/kmsg */ @@ -782,20 +831,21 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, ret = mutex_lock_interruptible(&user->lock); if (ret) return ret; - raw_spin_lock_irq(&logbuf_lock); + + logbuf_lock_irq(); while (user->seq == log_next_seq) { if (file->f_flags & O_NONBLOCK) { ret = -EAGAIN; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); goto out; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); ret = wait_event_interruptible(log_wait, user->seq != log_next_seq); if (ret) goto out; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); } if (user->seq < log_first_seq) { @@ -803,7 +853,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, user->idx = log_first_idx; user->seq = log_first_seq; ret = -EPIPE; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); goto out; } @@ -816,7 +866,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, user->idx = log_next(user->idx); user->seq++; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); if (len > count) { ret = -EINVAL; @@ -843,7 +893,7 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) if (offset) return -ESPIPE; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); switch (whence) { case SEEK_SET: /* the first record */ @@ -867,7 +917,7 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) default: ret = -EINVAL; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); return ret; } @@ -881,7 +931,7 @@ static unsigned int devkmsg_poll(struct file *file, poll_table *wait) poll_wait(file, &log_wait, wait); - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (user->seq < log_next_seq) { /* return error when data has vanished underneath us */ if (user->seq < log_first_seq) @@ -889,7 +939,7 @@ static unsigned int devkmsg_poll(struct file *file, poll_table *wait) else ret = POLLIN|POLLRDNORM; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); return ret; } @@ -919,10 +969,10 @@ static int devkmsg_open(struct inode *inode, struct file *file) mutex_init(&user->lock); - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); user->idx = log_first_idx; user->seq = log_first_seq; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); file->private_data = user; return 0; @@ -951,7 +1001,7 @@ const struct file_operations kmsg_fops = { .release = devkmsg_release, }; -#ifdef CONFIG_KEXEC_CORE +#ifdef CONFIG_CRASH_CORE /* * This appends the listed symbols to /proc/vmcore * @@ -960,7 +1010,7 @@ const struct file_operations kmsg_fops = { * symbols are specifically used so that utilities can access and extract the * dmesg log from a vmcore file after a crash. */ -void log_buf_kexec_setup(void) +void log_buf_vmcoreinfo_setup(void) { VMCOREINFO_SYMBOL(log_buf); VMCOREINFO_SYMBOL(log_buf_len); @@ -1064,13 +1114,13 @@ void __init setup_log_buf(int early) return; } - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); log_buf_len = new_log_buf_len; log_buf = new_log_buf; new_log_buf_len = 0; free = __LOG_BUF_LEN - log_next_idx; memcpy(log_buf, __log_buf, __LOG_BUF_LEN); - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); pr_info("log_buf_len: %d bytes\n", log_buf_len); pr_info("early log buf free: %d(%d%%)\n", @@ -1125,7 +1175,7 @@ static void boot_delay_msec(int level) unsigned long long k; unsigned long timeout; - if ((boot_delay == 0 || system_state != SYSTEM_BOOTING) + if ((boot_delay == 0 || system_state >= SYSTEM_RUNNING) || suppress_message_printing(level)) { return; } @@ -1248,7 +1298,7 @@ static int syslog_print(char __user *buf, int size) size_t n; size_t skip; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (syslog_seq < log_first_seq) { /* messages are gone, move to first one */ syslog_seq = log_first_seq; @@ -1256,7 +1306,7 @@ static int syslog_print(char __user *buf, int size) syslog_partial = 0; } if (syslog_seq == log_next_seq) { - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); break; } @@ -1275,7 +1325,7 @@ static int syslog_print(char __user *buf, int size) syslog_partial += n; } else n = 0; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); if (!n) break; @@ -1304,7 +1354,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) if (!text) return -ENOMEM; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (buf) { u64 next_seq; u64 seq; @@ -1352,12 +1402,12 @@ static int syslog_print_all(char __user *buf, int size, bool clear) idx = log_next(idx); seq++; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); if (copy_to_user(buf + len, text, textlen)) len = -EFAULT; else len += textlen; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (seq < log_first_seq) { /* messages are gone, move to next one */ @@ -1371,7 +1421,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) clear_seq = log_next_seq; clear_idx = log_next_idx; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); kfree(text); return len; @@ -1458,7 +1508,7 @@ int do_syslog(int type, char __user *buf, int len, int source) break; /* Number of chars in the log buffer */ case SYSLOG_ACTION_SIZE_UNREAD: - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (syslog_seq < log_first_seq) { /* messages are gone, move to first one */ syslog_seq = log_first_seq; @@ -1486,7 +1536,7 @@ int do_syslog(int type, char __user *buf, int len, int source) } error -= syslog_partial; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); break; /* Size of the log buffer */ case SYSLOG_ACTION_SIZE_BUFFER: @@ -1510,13 +1560,12 @@ SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len) * log_buf[start] to log_buf[end - 1]. * The console_lock must be held. */ -static void call_console_drivers(int level, - const char *ext_text, size_t ext_len, +static void call_console_drivers(const char *ext_text, size_t ext_len, const char *text, size_t len) { struct console *con; - trace_console(text, len); + trace_console_rcuidle(text, len); if (!console_drivers) return; @@ -1538,28 +1587,6 @@ static void call_console_drivers(int level, } } -/* - * Zap console related locks when oopsing. - * To leave time for slow consoles to print a full oops, - * only zap at most once every 30 seconds. - */ -static void zap_locks(void) -{ - static unsigned long oops_timestamp; - - if (time_after_eq(jiffies, oops_timestamp) && - !time_after(jiffies, oops_timestamp + 30 * HZ)) - return; - - oops_timestamp = jiffies; - - debug_locks_off(); - /* If a crash is occurring, make sure we can't deadlock */ - raw_spin_lock_init(&logbuf_lock); - /* And make sure that we print immediately */ - sema_init(&console_sem, 1); -} - int printk_delay_msec __read_mostly; static inline void printk_delay(void) @@ -1669,18 +1696,13 @@ asmlinkage int vprintk_emit(int facility, int level, const char *dict, size_t dictlen, const char *fmt, va_list args) { - static bool recursion_bug; static char textbuf[LOG_LINE_MAX]; char *text = textbuf; size_t text_len = 0; enum log_flags lflags = 0; unsigned long flags; - int this_cpu; int printed_len = 0; - int nmi_message_lost; bool in_sched = false; - /* cpu currently holding logbuf_lock in this function */ - static unsigned int logbuf_cpu = UINT_MAX; if (level == LOGLEVEL_SCHED) { level = LOGLEVEL_DEFAULT; @@ -1690,53 +1712,8 @@ asmlinkage int vprintk_emit(int facility, int level, boot_delay_msec(level); printk_delay(); - local_irq_save(flags); - this_cpu = smp_processor_id(); - - /* - * Ouch, printk recursed into itself! - */ - if (unlikely(logbuf_cpu == this_cpu)) { - /* - * If a crash is occurring during printk() on this CPU, - * then try to get the crash message out but make sure - * we can't deadlock. Otherwise just return to avoid the - * recursion and return - but flag the recursion so that - * it can be printed at the next appropriate moment: - */ - if (!oops_in_progress && !lockdep_recursing(current)) { - recursion_bug = true; - local_irq_restore(flags); - return 0; - } - zap_locks(); - } - - lockdep_off(); /* This stops the holder of console_sem just where we want him */ - raw_spin_lock(&logbuf_lock); - logbuf_cpu = this_cpu; - - if (unlikely(recursion_bug)) { - static const char recursion_msg[] = - "BUG: recent printk recursion!"; - - recursion_bug = false; - /* emit KERN_CRIT message */ - printed_len += log_store(0, 2, LOG_PREFIX|LOG_NEWLINE, 0, - NULL, 0, recursion_msg, - strlen(recursion_msg)); - } - - nmi_message_lost = get_nmi_message_lost(); - if (unlikely(nmi_message_lost)) { - text_len = scnprintf(textbuf, sizeof(textbuf), - "BAD LUCK: lost %d message(s) from NMI context!", - nmi_message_lost); - printed_len += log_store(0, 2, LOG_PREFIX|LOG_NEWLINE, 0, - NULL, 0, textbuf, text_len); - } - + logbuf_lock_irqsave(flags); /* * The printf needs to come first; we need the syslog * prefix which might be passed-in as a parameter. @@ -1779,14 +1756,10 @@ asmlinkage int vprintk_emit(int facility, int level, printed_len += log_output(facility, level, lflags, dict, dictlen, text, text_len); - logbuf_cpu = UINT_MAX; - raw_spin_unlock(&logbuf_lock); - lockdep_on(); - local_irq_restore(flags); + logbuf_unlock_irqrestore(flags); /* If called from the scheduler, we can not call up(). */ if (!in_sched) { - lockdep_off(); /* * Try to acquire and then immediately release the console * semaphore. The release will print out buffers and wake up @@ -1794,7 +1767,6 @@ asmlinkage int vprintk_emit(int facility, int level, */ if (console_trylock()) console_unlock(); - lockdep_on(); } return printed_len; @@ -1803,7 +1775,7 @@ EXPORT_SYMBOL(vprintk_emit); asmlinkage int vprintk(const char *fmt, va_list args) { - return vprintk_emit(0, LOGLEVEL_DEFAULT, NULL, 0, fmt, args); + return vprintk_func(fmt, args); } EXPORT_SYMBOL(vprintk); @@ -1895,16 +1867,12 @@ static ssize_t msg_print_ext_header(char *buf, size_t size, static ssize_t msg_print_ext_body(char *buf, size_t size, char *dict, size_t dict_len, char *text, size_t text_len) { return 0; } -static void call_console_drivers(int level, - const char *ext_text, size_t ext_len, +static void call_console_drivers(const char *ext_text, size_t ext_len, const char *text, size_t len) {} static size_t msg_print_text(const struct printk_log *msg, bool syslog, char *buf, size_t size) { return 0; } static bool suppress_message_printing(int level) { return false; } -/* Still needs to be defined for users */ -DEFINE_PER_CPU(printk_func_t, printk_func); - #endif /* CONFIG_PRINTK */ #ifdef CONFIG_EARLY_PRINTK @@ -1942,14 +1910,14 @@ static int __add_preferred_console(char *name, int idx, char *options, i++, c++) { if (strcmp(c->name, name) == 0 && c->index == idx) { if (!brl_options) - selected_console = i; + preferred_console = i; return 0; } } if (i == MAX_CMDLINECONSOLES) return -E2BIG; if (!brl_options) - selected_console = i; + preferred_console = i; strlcpy(c->name, name, sizeof(c->name)); c->options = options; braille_set_options(c, brl_options); @@ -2062,15 +2030,16 @@ void resume_console(void) * @cpu: unused * * If printk() is called from a CPU that is not online yet, the messages - * will be spooled but will not show up on the console. This function is - * called when a new CPU comes online (or fails to come up), and ensures - * that any such output gets printed. + * will be printed on the console only if there are CON_ANYTIME consoles. + * This function is called when a new CPU comes online (or fails to come + * up) or goes offline. */ static int console_cpu_notify(unsigned int cpu) { if (!cpuhp_tasks_frozen) { - console_lock(); - console_unlock(); + /* If trylock fails, someone else is doing the printing */ + if (console_trylock()) + console_unlock(); } return 0; } @@ -2192,7 +2161,7 @@ void console_unlock(void) } /* - * Console drivers are called under logbuf_lock, so + * Console drivers are called with interrupts disabled, so * @console_may_schedule should be cleared before; however, we may * end up dumping a lot of lines, for example, if called from * console registration path, and should invoke cond_resched() @@ -2200,11 +2169,15 @@ void console_unlock(void) * scheduling stall on a slow console leading to RCU stall and * softlockup warnings which exacerbate the issue with more * messages practically incapacitating the system. + * + * console_trylock() is not able to detect the preemptive + * context reliably. Therefore the value must be stored before + * and cleared after the the "again" goto label. */ do_cond_resched = console_may_schedule; +again: console_may_schedule = 0; -again: /* * We released the console_sem lock, so we need to recheck if * cpu is online and (if not) is there at least one CON_ANYTIME @@ -2220,9 +2193,9 @@ again: struct printk_log *msg; size_t ext_len = 0; size_t len; - int level; - raw_spin_lock_irqsave(&logbuf_lock, flags); + printk_safe_enter_irqsave(flags); + raw_spin_lock(&logbuf_lock); if (seen_seq != log_next_seq) { wake_klogd = true; seen_seq = log_next_seq; @@ -2243,8 +2216,7 @@ skip: break; msg = log_from_idx(console_idx); - level = msg->level; - if (suppress_message_printing(level)) { + if (suppress_message_printing(msg->level)) { /* * Skip record we have buffered and already printed * directly to the console when we received it, and @@ -2270,9 +2242,9 @@ skip: raw_spin_unlock(&logbuf_lock); stop_critical_timings(); /* don't trace print latency */ - call_console_drivers(level, ext_text, ext_len, text, len); + call_console_drivers(ext_text, ext_len, text, len); start_critical_timings(); - local_irq_restore(flags); + printk_safe_exit_irqrestore(flags); if (do_cond_resched) cond_resched(); @@ -2295,7 +2267,8 @@ skip: */ raw_spin_lock(&logbuf_lock); retry = console_seq != log_next_seq; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + raw_spin_unlock(&logbuf_lock); + printk_safe_exit_irqrestore(flags); if (retry && console_trylock()) goto again; @@ -2440,6 +2413,7 @@ void register_console(struct console *newcon) unsigned long flags; struct console *bcon = NULL; struct console_cmdline *c; + static bool has_preferred; if (console_drivers) for_each_console(bcon) @@ -2466,15 +2440,15 @@ void register_console(struct console *newcon) if (console_drivers && console_drivers->flags & CON_BOOT) bcon = console_drivers; - if (preferred_console < 0 || bcon || !console_drivers) - preferred_console = selected_console; + if (!has_preferred || bcon || !console_drivers) + has_preferred = preferred_console >= 0; /* * See if we want to use this console driver. If we * didn't select a console we take the first one * that registers here. */ - if (preferred_console < 0) { + if (!has_preferred) { if (newcon->index < 0) newcon->index = 0; if (newcon->setup == NULL || @@ -2482,7 +2456,7 @@ void register_console(struct console *newcon) newcon->flags |= CON_ENABLED; if (newcon->device) { newcon->flags |= CON_CONSDEV; - preferred_console = 0; + has_preferred = true; } } } @@ -2515,9 +2489,9 @@ void register_console(struct console *newcon) } newcon->flags |= CON_ENABLED; - if (i == selected_console) { + if (i == preferred_console) { newcon->flags |= CON_CONSDEV; - preferred_console = selected_console; + has_preferred = true; } break; } @@ -2558,10 +2532,10 @@ void register_console(struct console *newcon) * console_unlock(); will print out the buffered messages * for us. */ - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); console_seq = syslog_seq; console_idx = syslog_idx; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); /* * We're about to replay the log buffer. Only do this to the * just-registered console to avoid excessive message spam to @@ -2642,6 +2616,30 @@ int unregister_console(struct console *console) EXPORT_SYMBOL(unregister_console); /* + * Initialize the console device. This is called *early*, so + * we can't necessarily depend on lots of kernel help here. + * Just do some early initializations, and do the complex setup + * later. + */ +void __init console_init(void) +{ + initcall_t *call; + + /* Setup the default TTY line discipline. */ + n_tty_init(); + + /* + * set up the console device so that later boot sequences can + * inform about problems etc.. + */ + call = __con_initcall_start; + while (call < __con_initcall_end) { + (*call)(); + call++; + } +} + +/* * Some boot consoles access data that is in the init section and which will * be discarded after the initcalls have been run. To make sure that no code * will access this data, unregister the boot consoles in a late initcall. @@ -2722,16 +2720,13 @@ void wake_up_klogd(void) preempt_enable(); } -int printk_deferred(const char *fmt, ...) +int vprintk_deferred(const char *fmt, va_list args) { - va_list args; int r; - preempt_disable(); - va_start(args, fmt); r = vprintk_emit(0, LOGLEVEL_SCHED, NULL, 0, fmt, args); - va_end(args); + preempt_disable(); __this_cpu_or(printk_pending, PRINTK_PENDING_OUTPUT); irq_work_queue(this_cpu_ptr(&wake_up_klogd_work)); preempt_enable(); @@ -2739,6 +2734,18 @@ int printk_deferred(const char *fmt, ...) return r; } +int printk_deferred(const char *fmt, ...) +{ + va_list args; + int r; + + va_start(args, fmt); + r = vprintk_deferred(fmt, args); + va_end(args); + + return r; +} + /* * printk rate limiting, lifted from the networking subsystem. * @@ -2860,12 +2867,12 @@ void kmsg_dump(enum kmsg_dump_reason reason) /* initialize iterator with data about the stored records */ dumper->active = true; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); dumper->cur_seq = clear_seq; dumper->cur_idx = clear_idx; dumper->next_seq = log_next_seq; dumper->next_idx = log_next_idx; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); /* invoke dumper which will iterate over records */ dumper->dump(dumper, reason); @@ -2950,9 +2957,9 @@ bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, unsigned long flags; bool ret; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); ret = kmsg_dump_get_line_nolock(dumper, syslog, line, size, len); - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); return ret; } @@ -2991,7 +2998,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, if (!dumper->active) goto out; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); if (dumper->cur_seq < log_first_seq) { /* messages are gone, move to first available one */ dumper->cur_seq = log_first_seq; @@ -3000,7 +3007,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, /* last entry */ if (dumper->cur_seq >= dumper->next_seq) { - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); goto out; } @@ -3042,7 +3049,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, dumper->next_seq = next_seq; dumper->next_idx = next_idx; ret = true; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); out: if (len) *len = l; @@ -3080,9 +3087,9 @@ void kmsg_dump_rewind(struct kmsg_dumper *dumper) { unsigned long flags; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); kmsg_dump_rewind_nolock(dumper); - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); } EXPORT_SYMBOL_GPL(kmsg_dump_rewind); diff --git a/kernel/printk/printk_safe.c b/kernel/printk/printk_safe.c new file mode 100644 index 000000000000..3cdaeaef9ce1 --- /dev/null +++ b/kernel/printk/printk_safe.c @@ -0,0 +1,407 @@ +/* + * printk_safe.c - Safe printk for printk-deadlock-prone contexts + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/preempt.h> +#include <linux/spinlock.h> +#include <linux/debug_locks.h> +#include <linux/smp.h> +#include <linux/cpumask.h> +#include <linux/irq_work.h> +#include <linux/printk.h> + +#include "internal.h" + +/* + * printk() could not take logbuf_lock in NMI context. Instead, + * it uses an alternative implementation that temporary stores + * the strings into a per-CPU buffer. The content of the buffer + * is later flushed into the main ring buffer via IRQ work. + * + * The alternative implementation is chosen transparently + * by examinig current printk() context mask stored in @printk_context + * per-CPU variable. + * + * The implementation allows to flush the strings also from another CPU. + * There are situations when we want to make sure that all buffers + * were handled or when IRQs are blocked. + */ +static int printk_safe_irq_ready; + +#define SAFE_LOG_BUF_LEN ((1 << CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT) - \ + sizeof(atomic_t) - \ + sizeof(atomic_t) - \ + sizeof(struct irq_work)) + +struct printk_safe_seq_buf { + atomic_t len; /* length of written data */ + atomic_t message_lost; + struct irq_work work; /* IRQ work that flushes the buffer */ + unsigned char buffer[SAFE_LOG_BUF_LEN]; +}; + +static DEFINE_PER_CPU(struct printk_safe_seq_buf, safe_print_seq); +static DEFINE_PER_CPU(int, printk_context); + +#ifdef CONFIG_PRINTK_NMI +static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq); +#endif + +/* Get flushed in a more safe context. */ +static void queue_flush_work(struct printk_safe_seq_buf *s) +{ + if (printk_safe_irq_ready) { + /* Make sure that IRQ work is really initialized. */ + smp_rmb(); + irq_work_queue(&s->work); + } +} + +/* + * Add a message to per-CPU context-dependent buffer. NMI and printk-safe + * have dedicated buffers, because otherwise printk-safe preempted by + * NMI-printk would have overwritten the NMI messages. + * + * The messages are fushed from irq work (or from panic()), possibly, + * from other CPU, concurrently with printk_safe_log_store(). Should this + * happen, printk_safe_log_store() will notice the buffer->len mismatch + * and repeat the write. + */ +static __printf(2, 0) int printk_safe_log_store(struct printk_safe_seq_buf *s, + const char *fmt, va_list args) +{ + int add; + size_t len; + +again: + len = atomic_read(&s->len); + + /* The trailing '\0' is not counted into len. */ + if (len >= sizeof(s->buffer) - 1) { + atomic_inc(&s->message_lost); + queue_flush_work(s); + return 0; + } + + /* + * Make sure that all old data have been read before the buffer + * was reset. This is not needed when we just append data. + */ + if (!len) + smp_rmb(); + + add = vscnprintf(s->buffer + len, sizeof(s->buffer) - len, fmt, args); + if (!add) + return 0; + + /* + * Do it once again if the buffer has been flushed in the meantime. + * Note that atomic_cmpxchg() is an implicit memory barrier that + * makes sure that the data were written before updating s->len. + */ + if (atomic_cmpxchg(&s->len, len, len + add) != len) + goto again; + + queue_flush_work(s); + return add; +} + +static inline void printk_safe_flush_line(const char *text, int len) +{ + /* + * Avoid any console drivers calls from here, because we may be + * in NMI or printk_safe context (when in panic). The messages + * must go only into the ring buffer at this stage. Consoles will + * get explicitly called later when a crashdump is not generated. + */ + printk_deferred("%.*s", len, text); +} + +/* printk part of the temporary buffer line by line */ +static int printk_safe_flush_buffer(const char *start, size_t len) +{ + const char *c, *end; + bool header; + + c = start; + end = start + len; + header = true; + + /* Print line by line. */ + while (c < end) { + if (*c == '\n') { + printk_safe_flush_line(start, c - start + 1); + start = ++c; + header = true; + continue; + } + + /* Handle continuous lines or missing new line. */ + if ((c + 1 < end) && printk_get_level(c)) { + if (header) { + c = printk_skip_level(c); + continue; + } + + printk_safe_flush_line(start, c - start); + start = c++; + header = true; + continue; + } + + header = false; + c++; + } + + /* Check if there was a partial line. Ignore pure header. */ + if (start < end && !header) { + static const char newline[] = KERN_CONT "\n"; + + printk_safe_flush_line(start, end - start); + printk_safe_flush_line(newline, strlen(newline)); + } + + return len; +} + +static void report_message_lost(struct printk_safe_seq_buf *s) +{ + int lost = atomic_xchg(&s->message_lost, 0); + + if (lost) + printk_deferred("Lost %d message(s)!\n", lost); +} + +/* + * Flush data from the associated per-CPU buffer. The function + * can be called either via IRQ work or independently. + */ +static void __printk_safe_flush(struct irq_work *work) +{ + static raw_spinlock_t read_lock = + __RAW_SPIN_LOCK_INITIALIZER(read_lock); + struct printk_safe_seq_buf *s = + container_of(work, struct printk_safe_seq_buf, work); + unsigned long flags; + size_t len; + int i; + + /* + * The lock has two functions. First, one reader has to flush all + * available message to make the lockless synchronization with + * writers easier. Second, we do not want to mix messages from + * different CPUs. This is especially important when printing + * a backtrace. + */ + raw_spin_lock_irqsave(&read_lock, flags); + + i = 0; +more: + len = atomic_read(&s->len); + + /* + * This is just a paranoid check that nobody has manipulated + * the buffer an unexpected way. If we printed something then + * @len must only increase. Also it should never overflow the + * buffer size. + */ + if ((i && i >= len) || len > sizeof(s->buffer)) { + const char *msg = "printk_safe_flush: internal error\n"; + + printk_safe_flush_line(msg, strlen(msg)); + len = 0; + } + + if (!len) + goto out; /* Someone else has already flushed the buffer. */ + + /* Make sure that data has been written up to the @len */ + smp_rmb(); + i += printk_safe_flush_buffer(s->buffer + i, len - i); + + /* + * Check that nothing has got added in the meantime and truncate + * the buffer. Note that atomic_cmpxchg() is an implicit memory + * barrier that makes sure that the data were copied before + * updating s->len. + */ + if (atomic_cmpxchg(&s->len, len, 0) != len) + goto more; + +out: + report_message_lost(s); + raw_spin_unlock_irqrestore(&read_lock, flags); +} + +/** + * printk_safe_flush - flush all per-cpu nmi buffers. + * + * The buffers are flushed automatically via IRQ work. This function + * is useful only when someone wants to be sure that all buffers have + * been flushed at some point. + */ +void printk_safe_flush(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { +#ifdef CONFIG_PRINTK_NMI + __printk_safe_flush(&per_cpu(nmi_print_seq, cpu).work); +#endif + __printk_safe_flush(&per_cpu(safe_print_seq, cpu).work); + } +} + +/** + * printk_safe_flush_on_panic - flush all per-cpu nmi buffers when the system + * goes down. + * + * Similar to printk_safe_flush() but it can be called even in NMI context when + * the system goes down. It does the best effort to get NMI messages into + * the main ring buffer. + * + * Note that it could try harder when there is only one CPU online. + */ +void printk_safe_flush_on_panic(void) +{ + /* + * Make sure that we could access the main ring buffer. + * Do not risk a double release when more CPUs are up. + */ + if (in_nmi() && raw_spin_is_locked(&logbuf_lock)) { + if (num_online_cpus() > 1) + return; + + debug_locks_off(); + raw_spin_lock_init(&logbuf_lock); + } + + printk_safe_flush(); +} + +#ifdef CONFIG_PRINTK_NMI +/* + * Safe printk() for NMI context. It uses a per-CPU buffer to + * store the message. NMIs are not nested, so there is always only + * one writer running. But the buffer might get flushed from another + * CPU, so we need to be careful. + */ +static __printf(1, 0) int vprintk_nmi(const char *fmt, va_list args) +{ + struct printk_safe_seq_buf *s = this_cpu_ptr(&nmi_print_seq); + + return printk_safe_log_store(s, fmt, args); +} + +void printk_nmi_enter(void) +{ + /* + * The size of the extra per-CPU buffer is limited. Use it only when + * the main one is locked. If this CPU is not in the safe context, + * the lock must be taken on another CPU and we could wait for it. + */ + if ((this_cpu_read(printk_context) & PRINTK_SAFE_CONTEXT_MASK) && + raw_spin_is_locked(&logbuf_lock)) { + this_cpu_or(printk_context, PRINTK_NMI_CONTEXT_MASK); + } else { + this_cpu_or(printk_context, PRINTK_NMI_DEFERRED_CONTEXT_MASK); + } +} + +void printk_nmi_exit(void) +{ + this_cpu_and(printk_context, + ~(PRINTK_NMI_CONTEXT_MASK | + PRINTK_NMI_DEFERRED_CONTEXT_MASK)); +} + +#else + +static __printf(1, 0) int vprintk_nmi(const char *fmt, va_list args) +{ + return 0; +} + +#endif /* CONFIG_PRINTK_NMI */ + +/* + * Lock-less printk(), to avoid deadlocks should the printk() recurse + * into itself. It uses a per-CPU buffer to store the message, just like + * NMI. + */ +static __printf(1, 0) int vprintk_safe(const char *fmt, va_list args) +{ + struct printk_safe_seq_buf *s = this_cpu_ptr(&safe_print_seq); + + return printk_safe_log_store(s, fmt, args); +} + +/* Can be preempted by NMI. */ +void __printk_safe_enter(void) +{ + this_cpu_inc(printk_context); +} + +/* Can be preempted by NMI. */ +void __printk_safe_exit(void) +{ + this_cpu_dec(printk_context); +} + +__printf(1, 0) int vprintk_func(const char *fmt, va_list args) +{ + /* Use extra buffer in NMI when logbuf_lock is taken or in safe mode. */ + if (this_cpu_read(printk_context) & PRINTK_NMI_CONTEXT_MASK) + return vprintk_nmi(fmt, args); + + /* Use extra buffer to prevent a recursion deadlock in safe mode. */ + if (this_cpu_read(printk_context) & PRINTK_SAFE_CONTEXT_MASK) + return vprintk_safe(fmt, args); + + /* + * Use the main logbuf when logbuf_lock is available in NMI. + * But avoid calling console drivers that might have their own locks. + */ + if (this_cpu_read(printk_context) & PRINTK_NMI_DEFERRED_CONTEXT_MASK) + return vprintk_deferred(fmt, args); + + /* No obstacles. */ + return vprintk_default(fmt, args); +} + +void __init printk_safe_init(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + struct printk_safe_seq_buf *s; + + s = &per_cpu(safe_print_seq, cpu); + init_irq_work(&s->work, __printk_safe_flush); + +#ifdef CONFIG_PRINTK_NMI + s = &per_cpu(nmi_print_seq, cpu); + init_irq_work(&s->work, __printk_safe_flush); +#endif + } + + /* Make sure that IRQ works are initialized before enabling. */ + smp_wmb(); + printk_safe_irq_ready = 1; + + /* Flush pending messages that did not have scheduled IRQ works. */ + printk_safe_flush(); +} diff --git a/kernel/profile.c b/kernel/profile.c index f67ce0aa6bc4..9aa2a4445b0d 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -25,6 +25,8 @@ #include <linux/mutex.h> #include <linux/slab.h> #include <linux/vmalloc.h> +#include <linux/sched/stat.h> + #include <asm/sections.h> #include <asm/irq_regs.h> #include <asm/ptrace.h> diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 49ba7c1ade9d..60f356d91060 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -10,6 +10,9 @@ #include <linux/capability.h> #include <linux/export.h> #include <linux/sched.h> +#include <linux/sched/mm.h> +#include <linux/sched/coredump.h> +#include <linux/sched/task.h> #include <linux/errno.h> #include <linux/mm.h> #include <linux/highmem.h> @@ -57,19 +60,25 @@ int ptrace_access_vm(struct task_struct *tsk, unsigned long addr, } +void __ptrace_link(struct task_struct *child, struct task_struct *new_parent, + const struct cred *ptracer_cred) +{ + BUG_ON(!list_empty(&child->ptrace_entry)); + list_add(&child->ptrace_entry, &new_parent->ptraced); + child->parent = new_parent; + child->ptracer_cred = get_cred(ptracer_cred); +} + /* * ptrace a task: make the debugger its new parent and * move it to the ptrace list. * * Must be called with the tasklist lock write-held. */ -void __ptrace_link(struct task_struct *child, struct task_struct *new_parent) +static void ptrace_link(struct task_struct *child, struct task_struct *new_parent) { - BUG_ON(!list_empty(&child->ptrace_entry)); - list_add(&child->ptrace_entry, &new_parent->ptraced); - child->parent = new_parent; rcu_read_lock(); - child->ptracer_cred = get_cred(__task_cred(new_parent)); + __ptrace_link(child, new_parent, __task_cred(new_parent)); rcu_read_unlock(); } @@ -181,11 +190,17 @@ static void ptrace_unfreeze_traced(struct task_struct *task) WARN_ON(!task->ptrace || task->parent != current); + /* + * PTRACE_LISTEN can allow ptrace_trap_notify to wake us up remotely. + * Recheck state under the lock to close this race. + */ spin_lock_irq(&task->sighand->siglock); - if (__fatal_signal_pending(task)) - wake_up_state(task, __TASK_TRACED); - else - task->state = TASK_TRACED; + if (task->state == __TASK_TRACED) { + if (__fatal_signal_pending(task)) + wake_up_state(task, __TASK_TRACED); + else + task->state = TASK_TRACED; + } spin_unlock_irq(&task->sighand->siglock); } @@ -377,7 +392,7 @@ static int ptrace_attach(struct task_struct *task, long request, flags |= PT_SEIZED; task->ptrace = flags; - __ptrace_link(task, current); + ptrace_link(task, current); /* SEIZE doesn't trap tracee on attach */ if (!seize) @@ -450,7 +465,7 @@ static int ptrace_traceme(void) */ if (!ret && !(current->real_parent->flags & PF_EXITING)) { current->ptrace = PT_PTRACED; - __ptrace_link(current, current->real_parent); + ptrace_link(current, current->real_parent); } } write_unlock_irq(&tasklist_lock); diff --git a/kernel/rcu/Kconfig b/kernel/rcu/Kconfig new file mode 100644 index 000000000000..be90c945063f --- /dev/null +++ b/kernel/rcu/Kconfig @@ -0,0 +1,242 @@ +# +# RCU-related configuration options +# + +menu "RCU Subsystem" + +config TREE_RCU + bool + default y if !PREEMPT && SMP + help + This option selects the RCU implementation that is + designed for very large SMP system with hundreds or + thousands of CPUs. It also scales down nicely to + smaller systems. + +config PREEMPT_RCU + bool + default y if PREEMPT + help + This option selects the RCU implementation that is + designed for very large SMP systems with hundreds or + thousands of CPUs, but for which real-time response + is also required. It also scales down nicely to + smaller systems. + + Select this option if you are unsure. + +config TINY_RCU + bool + default y if !PREEMPT && !SMP + help + This option selects the RCU implementation that is + designed for UP systems from which real-time response + is not required. This option greatly reduces the + memory footprint of RCU. + +config RCU_EXPERT + bool "Make expert-level adjustments to RCU configuration" + default n + help + This option needs to be enabled if you wish to make + expert-level adjustments to RCU configuration. By default, + no such adjustments can be made, which has the often-beneficial + side-effect of preventing "make oldconfig" from asking you all + sorts of detailed questions about how you would like numerous + obscure RCU options to be set up. + + Say Y if you need to make expert-level adjustments to RCU. + + Say N if you are unsure. + +config SRCU + bool + help + This option selects the sleepable version of RCU. This version + permits arbitrary sleeping or blocking within RCU read-side critical + sections. + +config TINY_SRCU + bool + default y if SRCU && TINY_RCU + help + This option selects the single-CPU non-preemptible version of SRCU. + +config TREE_SRCU + bool + default y if SRCU && !TINY_RCU + help + This option selects the full-fledged version of SRCU. + +config TASKS_RCU + bool + default n + select SRCU + help + This option enables a task-based RCU implementation that uses + only voluntary context switch (not preemption!), idle, and + user-mode execution as quiescent states. + +config RCU_STALL_COMMON + def_bool ( TREE_RCU || PREEMPT_RCU ) + help + This option enables RCU CPU stall code that is common between + the TINY and TREE variants of RCU. The purpose is to allow + the tiny variants to disable RCU CPU stall warnings, while + making these warnings mandatory for the tree variants. + +config RCU_NEED_SEGCBLIST + def_bool ( TREE_RCU || PREEMPT_RCU || TREE_SRCU ) + +config CONTEXT_TRACKING + bool + +config CONTEXT_TRACKING_FORCE + bool "Force context tracking" + depends on CONTEXT_TRACKING + default y if !NO_HZ_FULL + help + The major pre-requirement for full dynticks to work is to + support the context tracking subsystem. But there are also + other dependencies to provide in order to make the full + dynticks working. + + This option stands for testing when an arch implements the + context tracking backend but doesn't yet fullfill all the + requirements to make the full dynticks feature working. + Without the full dynticks, there is no way to test the support + for context tracking and the subsystems that rely on it: RCU + userspace extended quiescent state and tickless cputime + accounting. This option copes with the absence of the full + dynticks subsystem by forcing the context tracking on all + CPUs in the system. + + Say Y only if you're working on the development of an + architecture backend for the context tracking. + + Say N otherwise, this option brings an overhead that you + don't want in production. + + +config RCU_FANOUT + int "Tree-based hierarchical RCU fanout value" + range 2 64 if 64BIT + range 2 32 if !64BIT + depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT + default 64 if 64BIT + default 32 if !64BIT + help + This option controls the fanout of hierarchical implementations + of RCU, allowing RCU to work efficiently on machines with + large numbers of CPUs. This value must be at least the fourth + root of NR_CPUS, which allows NR_CPUS to be insanely large. + The default value of RCU_FANOUT should be used for production + systems, but if you are stress-testing the RCU implementation + itself, small RCU_FANOUT values allow you to test large-system + code paths on small(er) systems. + + Select a specific number if testing RCU itself. + Take the default if unsure. + +config RCU_FANOUT_LEAF + int "Tree-based hierarchical RCU leaf-level fanout value" + range 2 64 if 64BIT + range 2 32 if !64BIT + depends on (TREE_RCU || PREEMPT_RCU) && RCU_EXPERT + default 16 + help + This option controls the leaf-level fanout of hierarchical + implementations of RCU, and allows trading off cache misses + against lock contention. Systems that synchronize their + scheduling-clock interrupts for energy-efficiency reasons will + want the default because the smaller leaf-level fanout keeps + lock contention levels acceptably low. Very large systems + (hundreds or thousands of CPUs) will instead want to set this + value to the maximum value possible in order to reduce the + number of cache misses incurred during RCU's grace-period + initialization. These systems tend to run CPU-bound, and thus + are not helped by synchronized interrupts, and thus tend to + skew them, which reduces lock contention enough that large + leaf-level fanouts work well. That said, setting leaf-level + fanout to a large number will likely cause problematic + lock contention on the leaf-level rcu_node structures unless + you boot with the skew_tick kernel parameter. + + Select a specific number if testing RCU itself. + + Select the maximum permissible value for large systems, but + please understand that you may also need to set the skew_tick + kernel boot parameter to avoid contention on the rcu_node + structure's locks. + + Take the default if unsure. + +config RCU_FAST_NO_HZ + bool "Accelerate last non-dyntick-idle CPU's grace periods" + depends on NO_HZ_COMMON && SMP && RCU_EXPERT + default n + help + This option permits CPUs to enter dynticks-idle state even if + they have RCU callbacks queued, and prevents RCU from waking + these CPUs up more than roughly once every four jiffies (by + default, you can adjust this using the rcutree.rcu_idle_gp_delay + parameter), thus improving energy efficiency. On the other + hand, this option increases the duration of RCU grace periods, + for example, slowing down synchronize_rcu(). + + Say Y if energy efficiency is critically important, and you + don't care about increased grace-period durations. + + Say N if you are unsure. + +config RCU_BOOST + bool "Enable RCU priority boosting" + depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT + default n + help + This option boosts the priority of preempted RCU readers that + block the current preemptible RCU grace period for too long. + This option also prevents heavy loads from blocking RCU + callback invocation for all flavors of RCU. + + Say Y here if you are working with real-time apps or heavy loads + Say N here if you are unsure. + +config RCU_BOOST_DELAY + int "Milliseconds to delay boosting after RCU grace-period start" + range 0 3000 + depends on RCU_BOOST + default 500 + help + This option specifies the time to wait after the beginning of + a given grace period before priority-boosting preempted RCU + readers blocking that grace period. Note that any RCU reader + blocking an expedited RCU grace period is boosted immediately. + + Accept the default if unsure. + +config RCU_NOCB_CPU + bool "Offload RCU callback processing from boot-selected CPUs" + depends on TREE_RCU || PREEMPT_RCU + depends on RCU_EXPERT || NO_HZ_FULL + default n + help + Use this option to reduce OS jitter for aggressive HPC or + real-time workloads. It can also be used to offload RCU + callback invocation to energy-efficient CPUs in battery-powered + asymmetric multiprocessors. + + This option offloads callback invocation from the set of + CPUs specified at boot time by the rcu_nocbs parameter. + For each such CPU, a kthread ("rcuox/N") will be created to + invoke callbacks, where the "N" is the CPU being offloaded, + and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and + "s" for RCU-sched. Nothing prevents this kthread from running + on the specified CPUs, but (1) the kthreads may be preempted + between each callback, and (2) affinity or cgroups can be used + to force the kthreads to run on whatever set of CPUs is desired. + + Say Y here if you want to help to debug reduced OS jitter. + Say N here if you are unsure. + +endmenu # "RCU Subsystem" diff --git a/kernel/rcu/Kconfig.debug b/kernel/rcu/Kconfig.debug new file mode 100644 index 000000000000..0ec7d1d33a14 --- /dev/null +++ b/kernel/rcu/Kconfig.debug @@ -0,0 +1,82 @@ +# +# RCU-related debugging configuration options +# + +menu "RCU Debugging" + +config PROVE_RCU + def_bool PROVE_LOCKING + +config TORTURE_TEST + tristate + default n + +config RCU_PERF_TEST + tristate "performance tests for RCU" + depends on DEBUG_KERNEL + select TORTURE_TEST + select SRCU + select TASKS_RCU + default n + help + This option provides a kernel module that runs performance + tests on the RCU infrastructure. The kernel module may be built + after the fact on the running kernel to be tested, if desired. + + Say Y here if you want RCU performance tests to be built into + the kernel. + Say M if you want the RCU performance tests to build as a module. + Say N if you are unsure. + +config RCU_TORTURE_TEST + tristate "torture tests for RCU" + depends on DEBUG_KERNEL + select TORTURE_TEST + select SRCU + select TASKS_RCU + default n + help + This option provides a kernel module that runs torture tests + on the RCU infrastructure. The kernel module may be built + after the fact on the running kernel to be tested, if desired. + + Say Y here if you want RCU torture tests to be built into + the kernel. + Say M if you want the RCU torture tests to build as a module. + Say N if you are unsure. + +config RCU_CPU_STALL_TIMEOUT + int "RCU CPU stall timeout in seconds" + depends on RCU_STALL_COMMON + range 3 300 + default 21 + help + If a given RCU grace period extends more than the specified + number of seconds, a CPU stall warning is printed. If the + RCU grace period persists, additional CPU stall warnings are + printed at more widely spaced intervals. + +config RCU_TRACE + bool "Enable tracing for RCU" + depends on DEBUG_KERNEL + default y if TREE_RCU + select TRACE_CLOCK + help + This option enables additional tracepoints for ftrace-style + event tracing. + + Say Y here if you want to enable RCU tracing + Say N if you are unsure. + +config RCU_EQS_DEBUG + bool "Provide debugging asserts for adding NO_HZ support to an arch" + depends on DEBUG_KERNEL + help + This option provides consistency checks in RCU's handling of + NO_HZ. These checks have proven quite helpful in detecting + bugs in arch-specific NO_HZ code. + + Say N here if you need ultimate kernel/user switch latencies + Say Y if you are unsure + +endmenu # "RCU Debugging" diff --git a/kernel/rcu/Makefile b/kernel/rcu/Makefile index 18dfc485225c..13c0fc852767 100644 --- a/kernel/rcu/Makefile +++ b/kernel/rcu/Makefile @@ -3,10 +3,11 @@ KCOV_INSTRUMENT := n obj-y += update.o sync.o -obj-$(CONFIG_SRCU) += srcu.o +obj-$(CONFIG_TREE_SRCU) += srcutree.o +obj-$(CONFIG_TINY_SRCU) += srcutiny.o obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o obj-$(CONFIG_RCU_PERF_TEST) += rcuperf.o obj-$(CONFIG_TREE_RCU) += tree.o obj-$(CONFIG_PREEMPT_RCU) += tree.o -obj-$(CONFIG_TREE_RCU_TRACE) += tree_trace.o obj-$(CONFIG_TINY_RCU) += tiny.o +obj-$(CONFIG_RCU_NEED_SEGCBLIST) += rcu_segcblist.o diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h index 0d6ff3e471be..808b8c85f626 100644 --- a/kernel/rcu/rcu.h +++ b/kernel/rcu/rcu.h @@ -56,6 +56,83 @@ #define DYNTICK_TASK_EXIT_IDLE (DYNTICK_TASK_NEST_VALUE + \ DYNTICK_TASK_FLAG) + +/* + * Grace-period counter management. + */ + +#define RCU_SEQ_CTR_SHIFT 2 +#define RCU_SEQ_STATE_MASK ((1 << RCU_SEQ_CTR_SHIFT) - 1) + +/* + * Return the counter portion of a sequence number previously returned + * by rcu_seq_snap() or rcu_seq_current(). + */ +static inline unsigned long rcu_seq_ctr(unsigned long s) +{ + return s >> RCU_SEQ_CTR_SHIFT; +} + +/* + * Return the state portion of a sequence number previously returned + * by rcu_seq_snap() or rcu_seq_current(). + */ +static inline int rcu_seq_state(unsigned long s) +{ + return s & RCU_SEQ_STATE_MASK; +} + +/* + * Set the state portion of the pointed-to sequence number. + * The caller is responsible for preventing conflicting updates. + */ +static inline void rcu_seq_set_state(unsigned long *sp, int newstate) +{ + WARN_ON_ONCE(newstate & ~RCU_SEQ_STATE_MASK); + WRITE_ONCE(*sp, (*sp & ~RCU_SEQ_STATE_MASK) + newstate); +} + +/* Adjust sequence number for start of update-side operation. */ +static inline void rcu_seq_start(unsigned long *sp) +{ + WRITE_ONCE(*sp, *sp + 1); + smp_mb(); /* Ensure update-side operation after counter increment. */ + WARN_ON_ONCE(rcu_seq_state(*sp) != 1); +} + +/* Adjust sequence number for end of update-side operation. */ +static inline void rcu_seq_end(unsigned long *sp) +{ + smp_mb(); /* Ensure update-side operation before counter increment. */ + WARN_ON_ONCE(!rcu_seq_state(*sp)); + WRITE_ONCE(*sp, (*sp | RCU_SEQ_STATE_MASK) + 1); +} + +/* Take a snapshot of the update side's sequence number. */ +static inline unsigned long rcu_seq_snap(unsigned long *sp) +{ + unsigned long s; + + s = (READ_ONCE(*sp) + 2 * RCU_SEQ_STATE_MASK + 1) & ~RCU_SEQ_STATE_MASK; + smp_mb(); /* Above access must not bleed into critical section. */ + return s; +} + +/* Return the current value the update side's sequence number, no ordering. */ +static inline unsigned long rcu_seq_current(unsigned long *sp) +{ + return READ_ONCE(*sp); +} + +/* + * Given a snapshot from rcu_seq_snap(), determine whether or not a + * full update-side operation has occurred. + */ +static inline bool rcu_seq_done(unsigned long *sp, unsigned long s) +{ + return ULONG_CMP_GE(READ_ONCE(*sp), s); +} + /* * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally * by call_rcu() and rcu callback execution, and are therefore not part of the @@ -109,12 +186,12 @@ static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head) rcu_lock_acquire(&rcu_callback_map); if (__is_kfree_rcu_offset(offset)) { - RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset)); + RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset);) kfree((void *)head - offset); rcu_lock_release(&rcu_callback_map); return true; } else { - RCU_TRACE(trace_rcu_invoke_callback(rn, head)); + RCU_TRACE(trace_rcu_invoke_callback(rn, head);) head->func(head); rcu_lock_release(&rcu_callback_map); return false; @@ -135,6 +212,18 @@ int rcu_jiffies_till_stall_check(void); */ #define TPS(x) tracepoint_string(x) +/* + * Dump the ftrace buffer, but only one time per callsite per boot. + */ +#define rcu_ftrace_dump(oops_dump_mode) \ +do { \ + static atomic_t ___rfd_beenhere = ATOMIC_INIT(0); \ + \ + if (!atomic_read(&___rfd_beenhere) && \ + !atomic_xchg(&___rfd_beenhere, 1)) \ + ftrace_dump(oops_dump_mode); \ +} while (0) + void rcu_early_boot_tests(void); void rcu_test_sync_prims(void); @@ -144,4 +233,341 @@ void rcu_test_sync_prims(void); */ extern void resched_cpu(int cpu); +#if defined(SRCU) || !defined(TINY_RCU) + +#include <linux/rcu_node_tree.h> + +extern int rcu_num_lvls; +extern int num_rcu_lvl[]; +extern int rcu_num_nodes; +static bool rcu_fanout_exact; +static int rcu_fanout_leaf; + +/* + * Compute the per-level fanout, either using the exact fanout specified + * or balancing the tree, depending on the rcu_fanout_exact boot parameter. + */ +static inline void rcu_init_levelspread(int *levelspread, const int *levelcnt) +{ + int i; + + if (rcu_fanout_exact) { + levelspread[rcu_num_lvls - 1] = rcu_fanout_leaf; + for (i = rcu_num_lvls - 2; i >= 0; i--) + levelspread[i] = RCU_FANOUT; + } else { + int ccur; + int cprv; + + cprv = nr_cpu_ids; + for (i = rcu_num_lvls - 1; i >= 0; i--) { + ccur = levelcnt[i]; + levelspread[i] = (cprv + ccur - 1) / ccur; + cprv = ccur; + } + } +} + +/* + * Do a full breadth-first scan of the rcu_node structures for the + * specified rcu_state structure. + */ +#define rcu_for_each_node_breadth_first(rsp, rnp) \ + for ((rnp) = &(rsp)->node[0]; \ + (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++) + +/* + * Do a breadth-first scan of the non-leaf rcu_node structures for the + * specified rcu_state structure. Note that if there is a singleton + * rcu_node tree with but one rcu_node structure, this loop is a no-op. + */ +#define rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) \ + for ((rnp) = &(rsp)->node[0]; \ + (rnp) < (rsp)->level[rcu_num_lvls - 1]; (rnp)++) + +/* + * Scan the leaves of the rcu_node hierarchy for the specified rcu_state + * structure. Note that if there is a singleton rcu_node tree with but + * one rcu_node structure, this loop -will- visit the rcu_node structure. + * It is still a leaf node, even if it is also the root node. + */ +#define rcu_for_each_leaf_node(rsp, rnp) \ + for ((rnp) = (rsp)->level[rcu_num_lvls - 1]; \ + (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++) + +/* + * Iterate over all possible CPUs in a leaf RCU node. + */ +#define for_each_leaf_node_possible_cpu(rnp, cpu) \ + for ((cpu) = cpumask_next(rnp->grplo - 1, cpu_possible_mask); \ + cpu <= rnp->grphi; \ + cpu = cpumask_next((cpu), cpu_possible_mask)) + +/* + * Wrappers for the rcu_node::lock acquire and release. + * + * Because the rcu_nodes form a tree, the tree traversal locking will observe + * different lock values, this in turn means that an UNLOCK of one level + * followed by a LOCK of another level does not imply a full memory barrier; + * and most importantly transitivity is lost. + * + * In order to restore full ordering between tree levels, augment the regular + * lock acquire functions with smp_mb__after_unlock_lock(). + * + * As ->lock of struct rcu_node is a __private field, therefore one should use + * these wrappers rather than directly call raw_spin_{lock,unlock}* on ->lock. + */ +#define raw_spin_lock_rcu_node(p) \ +do { \ + raw_spin_lock(&ACCESS_PRIVATE(p, lock)); \ + smp_mb__after_unlock_lock(); \ +} while (0) + +#define raw_spin_unlock_rcu_node(p) raw_spin_unlock(&ACCESS_PRIVATE(p, lock)) + +#define raw_spin_lock_irq_rcu_node(p) \ +do { \ + raw_spin_lock_irq(&ACCESS_PRIVATE(p, lock)); \ + smp_mb__after_unlock_lock(); \ +} while (0) + +#define raw_spin_unlock_irq_rcu_node(p) \ + raw_spin_unlock_irq(&ACCESS_PRIVATE(p, lock)) + +#define raw_spin_lock_irqsave_rcu_node(p, flags) \ +do { \ + raw_spin_lock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ + smp_mb__after_unlock_lock(); \ +} while (0) + +#define raw_spin_unlock_irqrestore_rcu_node(p, flags) \ + raw_spin_unlock_irqrestore(&ACCESS_PRIVATE(p, lock), flags) \ + +#define raw_spin_trylock_rcu_node(p) \ +({ \ + bool ___locked = raw_spin_trylock(&ACCESS_PRIVATE(p, lock)); \ + \ + if (___locked) \ + smp_mb__after_unlock_lock(); \ + ___locked; \ +}) + +#endif /* #if defined(SRCU) || !defined(TINY_RCU) */ + +#ifdef CONFIG_TINY_RCU +/* Tiny RCU doesn't expedite, as its purpose in life is instead to be tiny. */ +static inline bool rcu_gp_is_normal(void) /* Internal RCU use. */ +{ + return true; +} +static inline bool rcu_gp_is_expedited(void) /* Internal RCU use. */ +{ + return false; +} + +static inline void rcu_expedite_gp(void) +{ +} + +static inline void rcu_unexpedite_gp(void) +{ +} +#else /* #ifdef CONFIG_TINY_RCU */ +bool rcu_gp_is_normal(void); /* Internal RCU use. */ +bool rcu_gp_is_expedited(void); /* Internal RCU use. */ +void rcu_expedite_gp(void); +void rcu_unexpedite_gp(void); +void rcupdate_announce_bootup_oddness(void); +#endif /* #else #ifdef CONFIG_TINY_RCU */ + +#define RCU_SCHEDULER_INACTIVE 0 +#define RCU_SCHEDULER_INIT 1 +#define RCU_SCHEDULER_RUNNING 2 + +#ifdef CONFIG_TINY_RCU +static inline void rcu_request_urgent_qs_task(struct task_struct *t) { } +#else /* #ifdef CONFIG_TINY_RCU */ +void rcu_request_urgent_qs_task(struct task_struct *t); +#endif /* #else #ifdef CONFIG_TINY_RCU */ + +enum rcutorture_type { + RCU_FLAVOR, + RCU_BH_FLAVOR, + RCU_SCHED_FLAVOR, + RCU_TASKS_FLAVOR, + SRCU_FLAVOR, + INVALID_RCU_FLAVOR +}; + +#if defined(CONFIG_TREE_RCU) || defined(CONFIG_PREEMPT_RCU) +void rcutorture_get_gp_data(enum rcutorture_type test_type, int *flags, + unsigned long *gpnum, unsigned long *completed); +void rcutorture_record_test_transition(void); +void rcutorture_record_progress(unsigned long vernum); +void do_trace_rcu_torture_read(const char *rcutorturename, + struct rcu_head *rhp, + unsigned long secs, + unsigned long c_old, + unsigned long c); +#else +static inline void rcutorture_get_gp_data(enum rcutorture_type test_type, + int *flags, + unsigned long *gpnum, + unsigned long *completed) +{ + *flags = 0; + *gpnum = 0; + *completed = 0; +} +static inline void rcutorture_record_test_transition(void) +{ +} +static inline void rcutorture_record_progress(unsigned long vernum) +{ +} +#ifdef CONFIG_RCU_TRACE +void do_trace_rcu_torture_read(const char *rcutorturename, + struct rcu_head *rhp, + unsigned long secs, + unsigned long c_old, + unsigned long c); +#else +#define do_trace_rcu_torture_read(rcutorturename, rhp, secs, c_old, c) \ + do { } while (0) +#endif +#endif + +#ifdef CONFIG_TINY_SRCU + +static inline void srcutorture_get_gp_data(enum rcutorture_type test_type, + struct srcu_struct *sp, int *flags, + unsigned long *gpnum, + unsigned long *completed) +{ + if (test_type != SRCU_FLAVOR) + return; + *flags = 0; + *completed = sp->srcu_idx; + *gpnum = *completed; +} + +#elif defined(CONFIG_TREE_SRCU) + +void srcutorture_get_gp_data(enum rcutorture_type test_type, + struct srcu_struct *sp, int *flags, + unsigned long *gpnum, unsigned long *completed); + +#endif + +#ifdef CONFIG_TINY_RCU + +/* + * Return the number of grace periods started. + */ +static inline unsigned long rcu_batches_started(void) +{ + return 0; +} + +/* + * Return the number of bottom-half grace periods started. + */ +static inline unsigned long rcu_batches_started_bh(void) +{ + return 0; +} + +/* + * Return the number of sched grace periods started. + */ +static inline unsigned long rcu_batches_started_sched(void) +{ + return 0; +} + +/* + * Return the number of grace periods completed. + */ +static inline unsigned long rcu_batches_completed(void) +{ + return 0; +} + +/* + * Return the number of bottom-half grace periods completed. + */ +static inline unsigned long rcu_batches_completed_bh(void) +{ + return 0; +} + +/* + * Return the number of sched grace periods completed. + */ +static inline unsigned long rcu_batches_completed_sched(void) +{ + return 0; +} + +/* + * Return the number of expedited grace periods completed. + */ +static inline unsigned long rcu_exp_batches_completed(void) +{ + return 0; +} + +/* + * Return the number of expedited sched grace periods completed. + */ +static inline unsigned long rcu_exp_batches_completed_sched(void) +{ + return 0; +} + +static inline unsigned long srcu_batches_completed(struct srcu_struct *sp) +{ + return 0; +} + +static inline void rcu_force_quiescent_state(void) +{ +} + +static inline void rcu_bh_force_quiescent_state(void) +{ +} + +static inline void rcu_sched_force_quiescent_state(void) +{ +} + +static inline void show_rcu_gp_kthreads(void) +{ +} + +#else /* #ifdef CONFIG_TINY_RCU */ +extern unsigned long rcutorture_testseq; +extern unsigned long rcutorture_vernum; +unsigned long rcu_batches_started(void); +unsigned long rcu_batches_started_bh(void); +unsigned long rcu_batches_started_sched(void); +unsigned long rcu_batches_completed(void); +unsigned long rcu_batches_completed_bh(void); +unsigned long rcu_batches_completed_sched(void); +unsigned long rcu_exp_batches_completed(void); +unsigned long rcu_exp_batches_completed_sched(void); +unsigned long srcu_batches_completed(struct srcu_struct *sp); +void show_rcu_gp_kthreads(void); +void rcu_force_quiescent_state(void); +void rcu_bh_force_quiescent_state(void); +void rcu_sched_force_quiescent_state(void); +#endif /* #else #ifdef CONFIG_TINY_RCU */ + +#ifdef CONFIG_RCU_NOCB_CPU +bool rcu_is_nocb_cpu(int cpu); +#else +static inline bool rcu_is_nocb_cpu(int cpu) { return false; } +#endif + #endif /* __LINUX_RCU_H */ diff --git a/kernel/rcu/rcu_segcblist.c b/kernel/rcu/rcu_segcblist.c new file mode 100644 index 000000000000..2b62a38b080f --- /dev/null +++ b/kernel/rcu/rcu_segcblist.c @@ -0,0 +1,505 @@ +/* + * RCU segmented callback lists, function definitions + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, you can access it online at + * http://www.gnu.org/licenses/gpl-2.0.html. + * + * Copyright IBM Corporation, 2017 + * + * Authors: Paul E. McKenney <[email protected]> + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/interrupt.h> + +#include "rcu_segcblist.h" + +/* Initialize simple callback list. */ +void rcu_cblist_init(struct rcu_cblist *rclp) +{ + rclp->head = NULL; + rclp->tail = &rclp->head; + rclp->len = 0; + rclp->len_lazy = 0; +} + +/* + * Debug function to actually count the number of callbacks. + * If the number exceeds the limit specified, return -1. + */ +long rcu_cblist_count_cbs(struct rcu_cblist *rclp, long lim) +{ + int cnt = 0; + struct rcu_head **rhpp = &rclp->head; + + for (;;) { + if (!*rhpp) + return cnt; + if (++cnt > lim) + return -1; + rhpp = &(*rhpp)->next; + } +} + +/* + * Dequeue the oldest rcu_head structure from the specified callback + * list. This function assumes that the callback is non-lazy, but + * the caller can later invoke rcu_cblist_dequeued_lazy() if it + * finds otherwise (and if it cares about laziness). This allows + * different users to have different ways of determining laziness. + */ +struct rcu_head *rcu_cblist_dequeue(struct rcu_cblist *rclp) +{ + struct rcu_head *rhp; + + rhp = rclp->head; + if (!rhp) + return NULL; + rclp->len--; + rclp->head = rhp->next; + if (!rclp->head) + rclp->tail = &rclp->head; + return rhp; +} + +/* + * Initialize an rcu_segcblist structure. + */ +void rcu_segcblist_init(struct rcu_segcblist *rsclp) +{ + int i; + + BUILD_BUG_ON(RCU_NEXT_TAIL + 1 != ARRAY_SIZE(rsclp->gp_seq)); + BUILD_BUG_ON(ARRAY_SIZE(rsclp->tails) != ARRAY_SIZE(rsclp->gp_seq)); + rsclp->head = NULL; + for (i = 0; i < RCU_CBLIST_NSEGS; i++) + rsclp->tails[i] = &rsclp->head; + rsclp->len = 0; + rsclp->len_lazy = 0; +} + +/* + * Disable the specified rcu_segcblist structure, so that callbacks can + * no longer be posted to it. This structure must be empty. + */ +void rcu_segcblist_disable(struct rcu_segcblist *rsclp) +{ + WARN_ON_ONCE(!rcu_segcblist_empty(rsclp)); + WARN_ON_ONCE(rcu_segcblist_n_cbs(rsclp)); + WARN_ON_ONCE(rcu_segcblist_n_lazy_cbs(rsclp)); + rsclp->tails[RCU_NEXT_TAIL] = NULL; +} + +/* + * Is the specified segment of the specified rcu_segcblist structure + * empty of callbacks? + */ +bool rcu_segcblist_segempty(struct rcu_segcblist *rsclp, int seg) +{ + if (seg == RCU_DONE_TAIL) + return &rsclp->head == rsclp->tails[RCU_DONE_TAIL]; + return rsclp->tails[seg - 1] == rsclp->tails[seg]; +} + +/* + * Does the specified rcu_segcblist structure contain callbacks that + * are ready to be invoked? + */ +bool rcu_segcblist_ready_cbs(struct rcu_segcblist *rsclp) +{ + return rcu_segcblist_is_enabled(rsclp) && + &rsclp->head != rsclp->tails[RCU_DONE_TAIL]; +} + +/* + * Does the specified rcu_segcblist structure contain callbacks that + * are still pending, that is, not yet ready to be invoked? + */ +bool rcu_segcblist_pend_cbs(struct rcu_segcblist *rsclp) +{ + return rcu_segcblist_is_enabled(rsclp) && + !rcu_segcblist_restempty(rsclp, RCU_DONE_TAIL); +} + +/* + * Dequeue and return the first ready-to-invoke callback. If there + * are no ready-to-invoke callbacks, return NULL. Disables interrupts + * to avoid interference. Does not protect from interference from other + * CPUs or tasks. + */ +struct rcu_head *rcu_segcblist_dequeue(struct rcu_segcblist *rsclp) +{ + unsigned long flags; + int i; + struct rcu_head *rhp; + + local_irq_save(flags); + if (!rcu_segcblist_ready_cbs(rsclp)) { + local_irq_restore(flags); + return NULL; + } + rhp = rsclp->head; + BUG_ON(!rhp); + rsclp->head = rhp->next; + for (i = RCU_DONE_TAIL; i < RCU_CBLIST_NSEGS; i++) { + if (rsclp->tails[i] != &rhp->next) + break; + rsclp->tails[i] = &rsclp->head; + } + smp_mb(); /* Dequeue before decrement for rcu_barrier(). */ + WRITE_ONCE(rsclp->len, rsclp->len - 1); + local_irq_restore(flags); + return rhp; +} + +/* + * Account for the fact that a previously dequeued callback turned out + * to be marked as lazy. + */ +void rcu_segcblist_dequeued_lazy(struct rcu_segcblist *rsclp) +{ + unsigned long flags; + + local_irq_save(flags); + rsclp->len_lazy--; + local_irq_restore(flags); +} + +/* + * Return a pointer to the first callback in the specified rcu_segcblist + * structure. This is useful for diagnostics. + */ +struct rcu_head *rcu_segcblist_first_cb(struct rcu_segcblist *rsclp) +{ + if (rcu_segcblist_is_enabled(rsclp)) + return rsclp->head; + return NULL; +} + +/* + * Return a pointer to the first pending callback in the specified + * rcu_segcblist structure. This is useful just after posting a given + * callback -- if that callback is the first pending callback, then + * you cannot rely on someone else having already started up the required + * grace period. + */ +struct rcu_head *rcu_segcblist_first_pend_cb(struct rcu_segcblist *rsclp) +{ + if (rcu_segcblist_is_enabled(rsclp)) + return *rsclp->tails[RCU_DONE_TAIL]; + return NULL; +} + +/* + * Does the specified rcu_segcblist structure contain callbacks that + * have not yet been processed beyond having been posted, that is, + * does it contain callbacks in its last segment? + */ +bool rcu_segcblist_new_cbs(struct rcu_segcblist *rsclp) +{ + return rcu_segcblist_is_enabled(rsclp) && + !rcu_segcblist_restempty(rsclp, RCU_NEXT_READY_TAIL); +} + +/* + * Enqueue the specified callback onto the specified rcu_segcblist + * structure, updating accounting as needed. Note that the ->len + * field may be accessed locklessly, hence the WRITE_ONCE(). + * The ->len field is used by rcu_barrier() and friends to determine + * if it must post a callback on this structure, and it is OK + * for rcu_barrier() to sometimes post callbacks needlessly, but + * absolutely not OK for it to ever miss posting a callback. + */ +void rcu_segcblist_enqueue(struct rcu_segcblist *rsclp, + struct rcu_head *rhp, bool lazy) +{ + WRITE_ONCE(rsclp->len, rsclp->len + 1); /* ->len sampled locklessly. */ + if (lazy) + rsclp->len_lazy++; + smp_mb(); /* Ensure counts are updated before callback is enqueued. */ + rhp->next = NULL; + *rsclp->tails[RCU_NEXT_TAIL] = rhp; + rsclp->tails[RCU_NEXT_TAIL] = &rhp->next; +} + +/* + * Entrain the specified callback onto the specified rcu_segcblist at + * the end of the last non-empty segment. If the entire rcu_segcblist + * is empty, make no change, but return false. + * + * This is intended for use by rcu_barrier()-like primitives, -not- + * for normal grace-period use. IMPORTANT: The callback you enqueue + * will wait for all prior callbacks, NOT necessarily for a grace + * period. You have been warned. + */ +bool rcu_segcblist_entrain(struct rcu_segcblist *rsclp, + struct rcu_head *rhp, bool lazy) +{ + int i; + + if (rcu_segcblist_n_cbs(rsclp) == 0) + return false; + WRITE_ONCE(rsclp->len, rsclp->len + 1); + if (lazy) + rsclp->len_lazy++; + smp_mb(); /* Ensure counts are updated before callback is entrained. */ + rhp->next = NULL; + for (i = RCU_NEXT_TAIL; i > RCU_DONE_TAIL; i--) + if (rsclp->tails[i] != rsclp->tails[i - 1]) + break; + *rsclp->tails[i] = rhp; + for (; i <= RCU_NEXT_TAIL; i++) + rsclp->tails[i] = &rhp->next; + return true; +} + +/* + * Extract only the counts from the specified rcu_segcblist structure, + * and place them in the specified rcu_cblist structure. This function + * supports both callback orphaning and invocation, hence the separation + * of counts and callbacks. (Callbacks ready for invocation must be + * orphaned and adopted separately from pending callbacks, but counts + * apply to all callbacks. Locking must be used to make sure that + * both orphaned-callbacks lists are consistent.) + */ +void rcu_segcblist_extract_count(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp) +{ + rclp->len_lazy += rsclp->len_lazy; + rclp->len += rsclp->len; + rsclp->len_lazy = 0; + WRITE_ONCE(rsclp->len, 0); /* ->len sampled locklessly. */ +} + +/* + * Extract only those callbacks ready to be invoked from the specified + * rcu_segcblist structure and place them in the specified rcu_cblist + * structure. + */ +void rcu_segcblist_extract_done_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp) +{ + int i; + + if (!rcu_segcblist_ready_cbs(rsclp)) + return; /* Nothing to do. */ + *rclp->tail = rsclp->head; + rsclp->head = *rsclp->tails[RCU_DONE_TAIL]; + *rsclp->tails[RCU_DONE_TAIL] = NULL; + rclp->tail = rsclp->tails[RCU_DONE_TAIL]; + for (i = RCU_CBLIST_NSEGS - 1; i >= RCU_DONE_TAIL; i--) + if (rsclp->tails[i] == rsclp->tails[RCU_DONE_TAIL]) + rsclp->tails[i] = &rsclp->head; +} + +/* + * Extract only those callbacks still pending (not yet ready to be + * invoked) from the specified rcu_segcblist structure and place them in + * the specified rcu_cblist structure. Note that this loses information + * about any callbacks that might have been partway done waiting for + * their grace period. Too bad! They will have to start over. + */ +void rcu_segcblist_extract_pend_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp) +{ + int i; + + if (!rcu_segcblist_pend_cbs(rsclp)) + return; /* Nothing to do. */ + *rclp->tail = *rsclp->tails[RCU_DONE_TAIL]; + rclp->tail = rsclp->tails[RCU_NEXT_TAIL]; + *rsclp->tails[RCU_DONE_TAIL] = NULL; + for (i = RCU_DONE_TAIL + 1; i < RCU_CBLIST_NSEGS; i++) + rsclp->tails[i] = rsclp->tails[RCU_DONE_TAIL]; +} + +/* + * Insert counts from the specified rcu_cblist structure in the + * specified rcu_segcblist structure. + */ +void rcu_segcblist_insert_count(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp) +{ + rsclp->len_lazy += rclp->len_lazy; + /* ->len sampled locklessly. */ + WRITE_ONCE(rsclp->len, rsclp->len + rclp->len); + rclp->len_lazy = 0; + rclp->len = 0; +} + +/* + * Move callbacks from the specified rcu_cblist to the beginning of the + * done-callbacks segment of the specified rcu_segcblist. + */ +void rcu_segcblist_insert_done_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp) +{ + int i; + + if (!rclp->head) + return; /* No callbacks to move. */ + *rclp->tail = rsclp->head; + rsclp->head = rclp->head; + for (i = RCU_DONE_TAIL; i < RCU_CBLIST_NSEGS; i++) + if (&rsclp->head == rsclp->tails[i]) + rsclp->tails[i] = rclp->tail; + else + break; + rclp->head = NULL; + rclp->tail = &rclp->head; +} + +/* + * Move callbacks from the specified rcu_cblist to the end of the + * new-callbacks segment of the specified rcu_segcblist. + */ +void rcu_segcblist_insert_pend_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp) +{ + if (!rclp->head) + return; /* Nothing to do. */ + *rsclp->tails[RCU_NEXT_TAIL] = rclp->head; + rsclp->tails[RCU_NEXT_TAIL] = rclp->tail; + rclp->head = NULL; + rclp->tail = &rclp->head; +} + +/* + * Advance the callbacks in the specified rcu_segcblist structure based + * on the current value passed in for the grace-period counter. + */ +void rcu_segcblist_advance(struct rcu_segcblist *rsclp, unsigned long seq) +{ + int i, j; + + WARN_ON_ONCE(!rcu_segcblist_is_enabled(rsclp)); + if (rcu_segcblist_restempty(rsclp, RCU_DONE_TAIL)) + return; + + /* + * Find all callbacks whose ->gp_seq numbers indicate that they + * are ready to invoke, and put them into the RCU_DONE_TAIL segment. + */ + for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) { + if (ULONG_CMP_LT(seq, rsclp->gp_seq[i])) + break; + rsclp->tails[RCU_DONE_TAIL] = rsclp->tails[i]; + } + + /* If no callbacks moved, nothing more need be done. */ + if (i == RCU_WAIT_TAIL) + return; + + /* Clean up tail pointers that might have been misordered above. */ + for (j = RCU_WAIT_TAIL; j < i; j++) + rsclp->tails[j] = rsclp->tails[RCU_DONE_TAIL]; + + /* + * Callbacks moved, so clean up the misordered ->tails[] pointers + * that now point into the middle of the list of ready-to-invoke + * callbacks. The overall effect is to copy down the later pointers + * into the gap that was created by the now-ready segments. + */ + for (j = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++, j++) { + if (rsclp->tails[j] == rsclp->tails[RCU_NEXT_TAIL]) + break; /* No more callbacks. */ + rsclp->tails[j] = rsclp->tails[i]; + rsclp->gp_seq[j] = rsclp->gp_seq[i]; + } +} + +/* + * "Accelerate" callbacks based on more-accurate grace-period information. + * The reason for this is that RCU does not synchronize the beginnings and + * ends of grace periods, and that callbacks are posted locally. This in + * turn means that the callbacks must be labelled conservatively early + * on, as getting exact information would degrade both performance and + * scalability. When more accurate grace-period information becomes + * available, previously posted callbacks can be "accelerated", marking + * them to complete at the end of the earlier grace period. + * + * This function operates on an rcu_segcblist structure, and also the + * grace-period sequence number seq at which new callbacks would become + * ready to invoke. Returns true if there are callbacks that won't be + * ready to invoke until seq, false otherwise. + */ +bool rcu_segcblist_accelerate(struct rcu_segcblist *rsclp, unsigned long seq) +{ + int i; + + WARN_ON_ONCE(!rcu_segcblist_is_enabled(rsclp)); + if (rcu_segcblist_restempty(rsclp, RCU_DONE_TAIL)) + return false; + + /* + * Find the segment preceding the oldest segment of callbacks + * whose ->gp_seq[] completion is at or after that passed in via + * "seq", skipping any empty segments. This oldest segment, along + * with any later segments, can be merged in with any newly arrived + * callbacks in the RCU_NEXT_TAIL segment, and assigned "seq" + * as their ->gp_seq[] grace-period completion sequence number. + */ + for (i = RCU_NEXT_READY_TAIL; i > RCU_DONE_TAIL; i--) + if (rsclp->tails[i] != rsclp->tails[i - 1] && + ULONG_CMP_LT(rsclp->gp_seq[i], seq)) + break; + + /* + * If all the segments contain callbacks that correspond to + * earlier grace-period sequence numbers than "seq", leave. + * Assuming that the rcu_segcblist structure has enough + * segments in its arrays, this can only happen if some of + * the non-done segments contain callbacks that really are + * ready to invoke. This situation will get straightened + * out by the next call to rcu_segcblist_advance(). + * + * Also advance to the oldest segment of callbacks whose + * ->gp_seq[] completion is at or after that passed in via "seq", + * skipping any empty segments. + */ + if (++i >= RCU_NEXT_TAIL) + return false; + + /* + * Merge all later callbacks, including newly arrived callbacks, + * into the segment located by the for-loop above. Assign "seq" + * as the ->gp_seq[] value in order to correctly handle the case + * where there were no pending callbacks in the rcu_segcblist + * structure other than in the RCU_NEXT_TAIL segment. + */ + for (; i < RCU_NEXT_TAIL; i++) { + rsclp->tails[i] = rsclp->tails[RCU_NEXT_TAIL]; + rsclp->gp_seq[i] = seq; + } + return true; +} + +/* + * Scan the specified rcu_segcblist structure for callbacks that need + * a grace period later than the one specified by "seq". We don't look + * at the RCU_DONE_TAIL or RCU_NEXT_TAIL segments because they don't + * have a grace-period sequence number. + */ +bool rcu_segcblist_future_gp_needed(struct rcu_segcblist *rsclp, + unsigned long seq) +{ + int i; + + for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) + if (rsclp->tails[i - 1] != rsclp->tails[i] && + ULONG_CMP_LT(seq, rsclp->gp_seq[i])) + return true; + return false; +} diff --git a/kernel/rcu/rcu_segcblist.h b/kernel/rcu/rcu_segcblist.h new file mode 100644 index 000000000000..6e36e36478cd --- /dev/null +++ b/kernel/rcu/rcu_segcblist.h @@ -0,0 +1,164 @@ +/* + * RCU segmented callback lists, internal-to-rcu header file + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, you can access it online at + * http://www.gnu.org/licenses/gpl-2.0.html. + * + * Copyright IBM Corporation, 2017 + * + * Authors: Paul E. McKenney <[email protected]> + */ + +#include <linux/rcu_segcblist.h> + +/* + * Account for the fact that a previously dequeued callback turned out + * to be marked as lazy. + */ +static inline void rcu_cblist_dequeued_lazy(struct rcu_cblist *rclp) +{ + rclp->len_lazy--; +} + +/* + * Interim function to return rcu_cblist head pointer. Longer term, the + * rcu_cblist will be used more pervasively, removing the need for this + * function. + */ +static inline struct rcu_head *rcu_cblist_head(struct rcu_cblist *rclp) +{ + return rclp->head; +} + +/* + * Interim function to return rcu_cblist head pointer. Longer term, the + * rcu_cblist will be used more pervasively, removing the need for this + * function. + */ +static inline struct rcu_head **rcu_cblist_tail(struct rcu_cblist *rclp) +{ + WARN_ON_ONCE(!rclp->head); + return rclp->tail; +} + +void rcu_cblist_init(struct rcu_cblist *rclp); +long rcu_cblist_count_cbs(struct rcu_cblist *rclp, long lim); +struct rcu_head *rcu_cblist_dequeue(struct rcu_cblist *rclp); + +/* + * Is the specified rcu_segcblist structure empty? + * + * But careful! The fact that the ->head field is NULL does not + * necessarily imply that there are no callbacks associated with + * this structure. When callbacks are being invoked, they are + * removed as a group. If callback invocation must be preempted, + * the remaining callbacks will be added back to the list. Either + * way, the counts are updated later. + * + * So it is often the case that rcu_segcblist_n_cbs() should be used + * instead. + */ +static inline bool rcu_segcblist_empty(struct rcu_segcblist *rsclp) +{ + return !rsclp->head; +} + +/* Return number of callbacks in segmented callback list. */ +static inline long rcu_segcblist_n_cbs(struct rcu_segcblist *rsclp) +{ + return READ_ONCE(rsclp->len); +} + +/* Return number of lazy callbacks in segmented callback list. */ +static inline long rcu_segcblist_n_lazy_cbs(struct rcu_segcblist *rsclp) +{ + return rsclp->len_lazy; +} + +/* Return number of lazy callbacks in segmented callback list. */ +static inline long rcu_segcblist_n_nonlazy_cbs(struct rcu_segcblist *rsclp) +{ + return rsclp->len - rsclp->len_lazy; +} + +/* + * Is the specified rcu_segcblist enabled, for example, not corresponding + * to an offline or callback-offloaded CPU? + */ +static inline bool rcu_segcblist_is_enabled(struct rcu_segcblist *rsclp) +{ + return !!rsclp->tails[RCU_NEXT_TAIL]; +} + +/* + * Are all segments following the specified segment of the specified + * rcu_segcblist structure empty of callbacks? (The specified + * segment might well contain callbacks.) + */ +static inline bool rcu_segcblist_restempty(struct rcu_segcblist *rsclp, int seg) +{ + return !*rsclp->tails[seg]; +} + +/* + * Interim function to return rcu_segcblist head pointer. Longer term, the + * rcu_segcblist will be used more pervasively, removing the need for this + * function. + */ +static inline struct rcu_head *rcu_segcblist_head(struct rcu_segcblist *rsclp) +{ + return rsclp->head; +} + +/* + * Interim function to return rcu_segcblist head pointer. Longer term, the + * rcu_segcblist will be used more pervasively, removing the need for this + * function. + */ +static inline struct rcu_head **rcu_segcblist_tail(struct rcu_segcblist *rsclp) +{ + WARN_ON_ONCE(rcu_segcblist_empty(rsclp)); + return rsclp->tails[RCU_NEXT_TAIL]; +} + +void rcu_segcblist_init(struct rcu_segcblist *rsclp); +void rcu_segcblist_disable(struct rcu_segcblist *rsclp); +bool rcu_segcblist_segempty(struct rcu_segcblist *rsclp, int seg); +bool rcu_segcblist_ready_cbs(struct rcu_segcblist *rsclp); +bool rcu_segcblist_pend_cbs(struct rcu_segcblist *rsclp); +struct rcu_head *rcu_segcblist_dequeue(struct rcu_segcblist *rsclp); +void rcu_segcblist_dequeued_lazy(struct rcu_segcblist *rsclp); +struct rcu_head *rcu_segcblist_first_cb(struct rcu_segcblist *rsclp); +struct rcu_head *rcu_segcblist_first_pend_cb(struct rcu_segcblist *rsclp); +bool rcu_segcblist_new_cbs(struct rcu_segcblist *rsclp); +void rcu_segcblist_enqueue(struct rcu_segcblist *rsclp, + struct rcu_head *rhp, bool lazy); +bool rcu_segcblist_entrain(struct rcu_segcblist *rsclp, + struct rcu_head *rhp, bool lazy); +void rcu_segcblist_extract_count(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp); +void rcu_segcblist_extract_done_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp); +void rcu_segcblist_extract_pend_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp); +void rcu_segcblist_insert_count(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp); +void rcu_segcblist_insert_done_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp); +void rcu_segcblist_insert_pend_cbs(struct rcu_segcblist *rsclp, + struct rcu_cblist *rclp); +void rcu_segcblist_advance(struct rcu_segcblist *rsclp, unsigned long seq); +bool rcu_segcblist_accelerate(struct rcu_segcblist *rsclp, unsigned long seq); +bool rcu_segcblist_future_gp_needed(struct rcu_segcblist *rsclp, + unsigned long seq); diff --git a/kernel/rcu/rcuperf.c b/kernel/rcu/rcuperf.c index 123ccbd22449..3cc18110b612 100644 --- a/kernel/rcu/rcuperf.c +++ b/kernel/rcu/rcuperf.c @@ -30,6 +30,7 @@ #include <linux/rcupdate.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <uapi/linux/sched/types.h> #include <linux/atomic.h> #include <linux/bitops.h> #include <linux/completion.h> @@ -47,6 +48,8 @@ #include <linux/torture.h> #include <linux/vmalloc.h> +#include "rcu.h" + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <[email protected]>"); @@ -58,12 +61,16 @@ MODULE_AUTHOR("Paul E. McKenney <[email protected]>"); #define VERBOSE_PERFOUT_ERRSTRING(s) \ do { if (verbose) pr_alert("%s" PERF_FLAG "!!! %s\n", perf_type, s); } while (0) +torture_param(bool, gp_async, false, "Use asynchronous GP wait primitives"); +torture_param(int, gp_async_max, 1000, "Max # outstanding waits per reader"); torture_param(bool, gp_exp, false, "Use expedited GP wait primitives"); torture_param(int, holdoff, 10, "Holdoff time before test start (s)"); -torture_param(int, nreaders, -1, "Number of RCU reader threads"); +torture_param(int, nreaders, 0, "Number of RCU reader threads"); torture_param(int, nwriters, -1, "Number of RCU updater threads"); -torture_param(bool, shutdown, false, "Shutdown at end of performance tests."); +torture_param(bool, shutdown, !IS_ENABLED(MODULE), + "Shutdown at end of performance tests."); torture_param(bool, verbose, true, "Enable verbose debugging printk()s"); +torture_param(int, writer_holdoff, 0, "Holdoff (us) between GPs, zero to disable"); static char *perf_type = "rcu"; module_param(perf_type, charp, 0444); @@ -85,13 +92,16 @@ static u64 t_rcu_perf_writer_started; static u64 t_rcu_perf_writer_finished; static unsigned long b_rcu_perf_writer_started; static unsigned long b_rcu_perf_writer_finished; +static DEFINE_PER_CPU(atomic_t, n_async_inflight); static int rcu_perf_writer_state; #define RTWS_INIT 0 -#define RTWS_EXP_SYNC 1 -#define RTWS_SYNC 2 -#define RTWS_IDLE 2 -#define RTWS_STOPPING 3 +#define RTWS_ASYNC 1 +#define RTWS_BARRIER 2 +#define RTWS_EXP_SYNC 3 +#define RTWS_SYNC 4 +#define RTWS_IDLE 5 +#define RTWS_STOPPING 6 #define MAX_MEAS 10000 #define MIN_MEAS 100 @@ -113,6 +123,8 @@ struct rcu_perf_ops { unsigned long (*started)(void); unsigned long (*completed)(void); unsigned long (*exp_completed)(void); + void (*async)(struct rcu_head *head, rcu_callback_t func); + void (*gp_barrier)(void); void (*sync)(void); void (*exp_sync)(void); const char *name; @@ -152,6 +164,8 @@ static struct rcu_perf_ops rcu_ops = { .started = rcu_batches_started, .completed = rcu_batches_completed, .exp_completed = rcu_exp_batches_completed, + .async = call_rcu, + .gp_barrier = rcu_barrier, .sync = synchronize_rcu, .exp_sync = synchronize_rcu_expedited, .name = "rcu" @@ -180,6 +194,8 @@ static struct rcu_perf_ops rcu_bh_ops = { .started = rcu_batches_started_bh, .completed = rcu_batches_completed_bh, .exp_completed = rcu_exp_batches_completed_sched, + .async = call_rcu_bh, + .gp_barrier = rcu_barrier_bh, .sync = synchronize_rcu_bh, .exp_sync = synchronize_rcu_bh_expedited, .name = "rcu_bh" @@ -207,6 +223,16 @@ static unsigned long srcu_perf_completed(void) return srcu_batches_completed(srcu_ctlp); } +static void srcu_call_rcu(struct rcu_head *head, rcu_callback_t func) +{ + call_srcu(srcu_ctlp, head, func); +} + +static void srcu_rcu_barrier(void) +{ + srcu_barrier(srcu_ctlp); +} + static void srcu_perf_synchronize(void) { synchronize_srcu(srcu_ctlp); @@ -225,11 +251,42 @@ static struct rcu_perf_ops srcu_ops = { .started = NULL, .completed = srcu_perf_completed, .exp_completed = srcu_perf_completed, + .async = srcu_call_rcu, + .gp_barrier = srcu_rcu_barrier, .sync = srcu_perf_synchronize, .exp_sync = srcu_perf_synchronize_expedited, .name = "srcu" }; +static struct srcu_struct srcud; + +static void srcu_sync_perf_init(void) +{ + srcu_ctlp = &srcud; + init_srcu_struct(srcu_ctlp); +} + +static void srcu_sync_perf_cleanup(void) +{ + cleanup_srcu_struct(srcu_ctlp); +} + +static struct rcu_perf_ops srcud_ops = { + .ptype = SRCU_FLAVOR, + .init = srcu_sync_perf_init, + .cleanup = srcu_sync_perf_cleanup, + .readlock = srcu_perf_read_lock, + .readunlock = srcu_perf_read_unlock, + .started = NULL, + .completed = srcu_perf_completed, + .exp_completed = srcu_perf_completed, + .async = srcu_call_rcu, + .gp_barrier = srcu_rcu_barrier, + .sync = srcu_perf_synchronize, + .exp_sync = srcu_perf_synchronize_expedited, + .name = "srcud" +}; + /* * Definitions for sched perf testing. */ @@ -253,6 +310,8 @@ static struct rcu_perf_ops sched_ops = { .started = rcu_batches_started_sched, .completed = rcu_batches_completed_sched, .exp_completed = rcu_exp_batches_completed_sched, + .async = call_rcu_sched, + .gp_barrier = rcu_barrier_sched, .sync = synchronize_sched, .exp_sync = synchronize_sched_expedited, .name = "sched" @@ -280,6 +339,8 @@ static struct rcu_perf_ops tasks_ops = { .readunlock = tasks_perf_read_unlock, .started = rcu_no_completed, .completed = rcu_no_completed, + .async = call_rcu_tasks, + .gp_barrier = rcu_barrier_tasks, .sync = synchronize_rcu_tasks, .exp_sync = synchronize_rcu_tasks, .name = "tasks" @@ -343,6 +404,15 @@ rcu_perf_reader(void *arg) } /* + * Callback function for asynchronous grace periods from rcu_perf_writer(). + */ +static void rcu_perf_async_cb(struct rcu_head *rhp) +{ + atomic_dec(this_cpu_ptr(&n_async_inflight)); + kfree(rhp); +} + +/* * RCU perf writer kthread. Repeatedly does a grace period. */ static int @@ -351,6 +421,7 @@ rcu_perf_writer(void *arg) int i = 0; int i_max; long me = (long)arg; + struct rcu_head *rhp = NULL; struct sched_param sp; bool started = false, done = false, alldone = false; u64 t; @@ -379,9 +450,27 @@ rcu_perf_writer(void *arg) } do { + if (writer_holdoff) + udelay(writer_holdoff); wdp = &wdpp[i]; *wdp = ktime_get_mono_fast_ns(); - if (gp_exp) { + if (gp_async) { +retry: + if (!rhp) + rhp = kmalloc(sizeof(*rhp), GFP_KERNEL); + if (rhp && atomic_read(this_cpu_ptr(&n_async_inflight)) < gp_async_max) { + rcu_perf_writer_state = RTWS_ASYNC; + atomic_inc(this_cpu_ptr(&n_async_inflight)); + cur_ops->async(rhp, rcu_perf_async_cb); + rhp = NULL; + } else if (!kthread_should_stop()) { + rcu_perf_writer_state = RTWS_BARRIER; + cur_ops->gp_barrier(); + goto retry; + } else { + kfree(rhp); /* Because we are stopping. */ + } + } else if (gp_exp) { rcu_perf_writer_state = RTWS_EXP_SYNC; cur_ops->exp_sync(); } else { @@ -428,6 +517,10 @@ rcu_perf_writer(void *arg) i++; rcu_perf_wait_shutdown(); } while (!torture_must_stop()); + if (gp_async) { + rcu_perf_writer_state = RTWS_BARRIER; + cur_ops->gp_barrier(); + } rcu_perf_writer_state = RTWS_STOPPING; writer_n_durations[me] = i_max; torture_kthread_stopping("rcu_perf_writer"); @@ -451,6 +544,17 @@ rcu_perf_cleanup(void) u64 *wdp; u64 *wdpp; + /* + * Would like warning at start, but everything is expedited + * during the mid-boot phase, so have to wait till the end. + */ + if (rcu_gp_is_expedited() && !rcu_gp_is_normal() && !gp_exp) + VERBOSE_PERFOUT_ERRSTRING("All grace periods expedited, no normal ones to measure!"); + if (rcu_gp_is_normal() && gp_exp) + VERBOSE_PERFOUT_ERRSTRING("All grace periods normal, no expedited ones to measure!"); + if (gp_exp && gp_async) + VERBOSE_PERFOUT_ERRSTRING("No expedited async GPs, so went with async!"); + if (torture_cleanup_begin()) return; @@ -553,7 +657,7 @@ rcu_perf_init(void) long i; int firsterr = 0; static struct rcu_perf_ops *perf_ops[] = { - &rcu_ops, &rcu_bh_ops, &srcu_ops, &sched_ops, + &rcu_ops, &rcu_bh_ops, &srcu_ops, &srcud_ops, &sched_ops, RCUPERF_TASKS_OPS }; @@ -623,16 +727,6 @@ rcu_perf_init(void) firsterr = -ENOMEM; goto unwind; } - if (rcu_gp_is_expedited() && !rcu_gp_is_normal() && !gp_exp) { - VERBOSE_PERFOUT_ERRSTRING("All grace periods expedited, no normal ones to measure!"); - firsterr = -EINVAL; - goto unwind; - } - if (rcu_gp_is_normal() && gp_exp) { - VERBOSE_PERFOUT_ERRSTRING("All grace periods normal, no expedited ones to measure!"); - firsterr = -EINVAL; - goto unwind; - } for (i = 0; i < nrealwriters; i++) { writer_durations[i] = kcalloc(MAX_MEAS, sizeof(*writer_durations[i]), diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 87c51225ceec..b8f7f8ce8575 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -32,7 +32,8 @@ #include <linux/smp.h> #include <linux/rcupdate.h> #include <linux/interrupt.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <uapi/linux/sched/types.h> #include <linux/atomic.h> #include <linux/bitops.h> #include <linux/completion.h> @@ -51,6 +52,8 @@ #include <linux/torture.h> #include <linux/vmalloc.h> +#include "rcu.h" + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <[email protected]> and Josh Triplett <[email protected]>"); @@ -558,19 +561,44 @@ static void srcu_torture_barrier(void) static void srcu_torture_stats(void) { - int cpu; - int idx = srcu_ctlp->completed & 0x1; + int __maybe_unused cpu; + int idx; - pr_alert("%s%s per-CPU(idx=%d):", +#ifdef CONFIG_TREE_SRCU + idx = srcu_ctlp->srcu_idx & 0x1; + pr_alert("%s%s Tree SRCU per-CPU(idx=%d):", torture_type, TORTURE_FLAG, idx); for_each_possible_cpu(cpu) { + unsigned long l0, l1; + unsigned long u0, u1; long c0, c1; + struct srcu_data *counts; - c0 = (long)per_cpu_ptr(srcu_ctlp->per_cpu_ref, cpu)->c[!idx]; - c1 = (long)per_cpu_ptr(srcu_ctlp->per_cpu_ref, cpu)->c[idx]; + counts = per_cpu_ptr(srcu_ctlp->sda, cpu); + u0 = counts->srcu_unlock_count[!idx]; + u1 = counts->srcu_unlock_count[idx]; + + /* + * Make sure that a lock is always counted if the corresponding + * unlock is counted. + */ + smp_rmb(); + + l0 = counts->srcu_lock_count[!idx]; + l1 = counts->srcu_lock_count[idx]; + + c0 = l0 - u0; + c1 = l1 - u1; pr_cont(" %d(%ld,%ld)", cpu, c0, c1); } pr_cont("\n"); +#elif defined(CONFIG_TINY_SRCU) + idx = READ_ONCE(srcu_ctlp->srcu_idx) & 0x1; + pr_alert("%s%s Tiny SRCU per-CPU(idx=%d): (%hd,%hd)\n", + torture_type, TORTURE_FLAG, idx, + READ_ONCE(srcu_ctlp->srcu_lock_nesting[!idx]), + READ_ONCE(srcu_ctlp->srcu_lock_nesting[idx])); +#endif } static void srcu_torture_synchronize_expedited(void) @@ -1317,12 +1345,14 @@ rcu_torture_stats_print(void) cur_ops->stats(); if (rtcv_snap == rcu_torture_current_version && rcu_torture_current != NULL) { - int __maybe_unused flags; - unsigned long __maybe_unused gpnum; - unsigned long __maybe_unused completed; + int __maybe_unused flags = 0; + unsigned long __maybe_unused gpnum = 0; + unsigned long __maybe_unused completed = 0; rcutorture_get_gp_data(cur_ops->ttype, &flags, &gpnum, &completed); + srcutorture_get_gp_data(cur_ops->ttype, srcu_ctlp, + &flags, &gpnum, &completed); wtp = READ_ONCE(writer_task); pr_alert("??? Writer stall state %s(%d) g%lu c%lu f%#x ->state %#lx\n", rcu_torture_writer_state_getname(), diff --git a/kernel/rcu/srcu.c b/kernel/rcu/srcu.c deleted file mode 100644 index 9b9cdd549caa..000000000000 --- a/kernel/rcu/srcu.c +++ /dev/null @@ -1,675 +0,0 @@ -/* - * Sleepable Read-Copy Update mechanism for mutual exclusion. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, you can access it online at - * http://www.gnu.org/licenses/gpl-2.0.html. - * - * Copyright (C) IBM Corporation, 2006 - * Copyright (C) Fujitsu, 2012 - * - * Author: Paul McKenney <[email protected]> - * Lai Jiangshan <[email protected]> - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU/ *.txt - * - */ - -#include <linux/export.h> -#include <linux/mutex.h> -#include <linux/percpu.h> -#include <linux/preempt.h> -#include <linux/rcupdate.h> -#include <linux/sched.h> -#include <linux/smp.h> -#include <linux/delay.h> -#include <linux/srcu.h> - -#include "rcu.h" - -/* - * Initialize an rcu_batch structure to empty. - */ -static inline void rcu_batch_init(struct rcu_batch *b) -{ - b->head = NULL; - b->tail = &b->head; -} - -/* - * Enqueue a callback onto the tail of the specified rcu_batch structure. - */ -static inline void rcu_batch_queue(struct rcu_batch *b, struct rcu_head *head) -{ - *b->tail = head; - b->tail = &head->next; -} - -/* - * Is the specified rcu_batch structure empty? - */ -static inline bool rcu_batch_empty(struct rcu_batch *b) -{ - return b->tail == &b->head; -} - -/* - * Remove the callback at the head of the specified rcu_batch structure - * and return a pointer to it, or return NULL if the structure is empty. - */ -static inline struct rcu_head *rcu_batch_dequeue(struct rcu_batch *b) -{ - struct rcu_head *head; - - if (rcu_batch_empty(b)) - return NULL; - - head = b->head; - b->head = head->next; - if (b->tail == &head->next) - rcu_batch_init(b); - - return head; -} - -/* - * Move all callbacks from the rcu_batch structure specified by "from" to - * the structure specified by "to". - */ -static inline void rcu_batch_move(struct rcu_batch *to, struct rcu_batch *from) -{ - if (!rcu_batch_empty(from)) { - *to->tail = from->head; - to->tail = from->tail; - rcu_batch_init(from); - } -} - -static int init_srcu_struct_fields(struct srcu_struct *sp) -{ - sp->completed = 0; - spin_lock_init(&sp->queue_lock); - sp->running = false; - rcu_batch_init(&sp->batch_queue); - rcu_batch_init(&sp->batch_check0); - rcu_batch_init(&sp->batch_check1); - rcu_batch_init(&sp->batch_done); - INIT_DELAYED_WORK(&sp->work, process_srcu); - sp->per_cpu_ref = alloc_percpu(struct srcu_struct_array); - return sp->per_cpu_ref ? 0 : -ENOMEM; -} - -#ifdef CONFIG_DEBUG_LOCK_ALLOC - -int __init_srcu_struct(struct srcu_struct *sp, const char *name, - struct lock_class_key *key) -{ - /* Don't re-initialize a lock while it is held. */ - debug_check_no_locks_freed((void *)sp, sizeof(*sp)); - lockdep_init_map(&sp->dep_map, name, key, 0); - return init_srcu_struct_fields(sp); -} -EXPORT_SYMBOL_GPL(__init_srcu_struct); - -#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ - -/** - * init_srcu_struct - initialize a sleep-RCU structure - * @sp: structure to initialize. - * - * Must invoke this on a given srcu_struct before passing that srcu_struct - * to any other function. Each srcu_struct represents a separate domain - * of SRCU protection. - */ -int init_srcu_struct(struct srcu_struct *sp) -{ - return init_srcu_struct_fields(sp); -} -EXPORT_SYMBOL_GPL(init_srcu_struct); - -#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */ - -/* - * Returns approximate total of the readers' ->seq[] values for the - * rank of per-CPU counters specified by idx. - */ -static unsigned long srcu_readers_seq_idx(struct srcu_struct *sp, int idx) -{ - int cpu; - unsigned long sum = 0; - unsigned long t; - - for_each_possible_cpu(cpu) { - t = READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->seq[idx]); - sum += t; - } - return sum; -} - -/* - * Returns approximate number of readers active on the specified rank - * of the per-CPU ->c[] counters. - */ -static unsigned long srcu_readers_active_idx(struct srcu_struct *sp, int idx) -{ - int cpu; - unsigned long sum = 0; - unsigned long t; - - for_each_possible_cpu(cpu) { - t = READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[idx]); - sum += t; - } - return sum; -} - -/* - * Return true if the number of pre-existing readers is determined to - * be stably zero. An example unstable zero can occur if the call - * to srcu_readers_active_idx() misses an __srcu_read_lock() increment, - * but due to task migration, sees the corresponding __srcu_read_unlock() - * decrement. This can happen because srcu_readers_active_idx() takes - * time to sum the array, and might in fact be interrupted or preempted - * partway through the summation. - */ -static bool srcu_readers_active_idx_check(struct srcu_struct *sp, int idx) -{ - unsigned long seq; - - seq = srcu_readers_seq_idx(sp, idx); - - /* - * The following smp_mb() A pairs with the smp_mb() B located in - * __srcu_read_lock(). This pairing ensures that if an - * __srcu_read_lock() increments its counter after the summation - * in srcu_readers_active_idx(), then the corresponding SRCU read-side - * critical section will see any changes made prior to the start - * of the current SRCU grace period. - * - * Also, if the above call to srcu_readers_seq_idx() saw the - * increment of ->seq[], then the call to srcu_readers_active_idx() - * must see the increment of ->c[]. - */ - smp_mb(); /* A */ - - /* - * Note that srcu_readers_active_idx() can incorrectly return - * zero even though there is a pre-existing reader throughout. - * To see this, suppose that task A is in a very long SRCU - * read-side critical section that started on CPU 0, and that - * no other reader exists, so that the sum of the counters - * is equal to one. Then suppose that task B starts executing - * srcu_readers_active_idx(), summing up to CPU 1, and then that - * task C starts reading on CPU 0, so that its increment is not - * summed, but finishes reading on CPU 2, so that its decrement - * -is- summed. Then when task B completes its sum, it will - * incorrectly get zero, despite the fact that task A has been - * in its SRCU read-side critical section the whole time. - * - * We therefore do a validation step should srcu_readers_active_idx() - * return zero. - */ - if (srcu_readers_active_idx(sp, idx) != 0) - return false; - - /* - * The remainder of this function is the validation step. - * The following smp_mb() D pairs with the smp_mb() C in - * __srcu_read_unlock(). If the __srcu_read_unlock() was seen - * by srcu_readers_active_idx() above, then any destructive - * operation performed after the grace period will happen after - * the corresponding SRCU read-side critical section. - * - * Note that there can be at most NR_CPUS worth of readers using - * the old index, which is not enough to overflow even a 32-bit - * integer. (Yes, this does mean that systems having more than - * a billion or so CPUs need to be 64-bit systems.) Therefore, - * the sum of the ->seq[] counters cannot possibly overflow. - * Therefore, the only way that the return values of the two - * calls to srcu_readers_seq_idx() can be equal is if there were - * no increments of the corresponding rank of ->seq[] counts - * in the interim. But the missed-increment scenario laid out - * above includes an increment of the ->seq[] counter by - * the corresponding __srcu_read_lock(). Therefore, if this - * scenario occurs, the return values from the two calls to - * srcu_readers_seq_idx() will differ, and thus the validation - * step below suffices. - */ - smp_mb(); /* D */ - - return srcu_readers_seq_idx(sp, idx) == seq; -} - -/** - * srcu_readers_active - returns true if there are readers. and false - * otherwise - * @sp: which srcu_struct to count active readers (holding srcu_read_lock). - * - * Note that this is not an atomic primitive, and can therefore suffer - * severe errors when invoked on an active srcu_struct. That said, it - * can be useful as an error check at cleanup time. - */ -static bool srcu_readers_active(struct srcu_struct *sp) -{ - int cpu; - unsigned long sum = 0; - - for_each_possible_cpu(cpu) { - sum += READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[0]); - sum += READ_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[1]); - } - return sum; -} - -/** - * cleanup_srcu_struct - deconstruct a sleep-RCU structure - * @sp: structure to clean up. - * - * Must invoke this after you are finished using a given srcu_struct that - * was initialized via init_srcu_struct(), else you leak memory. - */ -void cleanup_srcu_struct(struct srcu_struct *sp) -{ - if (WARN_ON(srcu_readers_active(sp))) - return; /* Leakage unless caller handles error. */ - free_percpu(sp->per_cpu_ref); - sp->per_cpu_ref = NULL; -} -EXPORT_SYMBOL_GPL(cleanup_srcu_struct); - -/* - * Counts the new reader in the appropriate per-CPU element of the - * srcu_struct. Must be called from process context. - * Returns an index that must be passed to the matching srcu_read_unlock(). - */ -int __srcu_read_lock(struct srcu_struct *sp) -{ - int idx; - - idx = READ_ONCE(sp->completed) & 0x1; - __this_cpu_inc(sp->per_cpu_ref->c[idx]); - smp_mb(); /* B */ /* Avoid leaking the critical section. */ - __this_cpu_inc(sp->per_cpu_ref->seq[idx]); - return idx; -} -EXPORT_SYMBOL_GPL(__srcu_read_lock); - -/* - * Removes the count for the old reader from the appropriate per-CPU - * element of the srcu_struct. Note that this may well be a different - * CPU than that which was incremented by the corresponding srcu_read_lock(). - * Must be called from process context. - */ -void __srcu_read_unlock(struct srcu_struct *sp, int idx) -{ - smp_mb(); /* C */ /* Avoid leaking the critical section. */ - this_cpu_dec(sp->per_cpu_ref->c[idx]); -} -EXPORT_SYMBOL_GPL(__srcu_read_unlock); - -/* - * We use an adaptive strategy for synchronize_srcu() and especially for - * synchronize_srcu_expedited(). We spin for a fixed time period - * (defined below) to allow SRCU readers to exit their read-side critical - * sections. If there are still some readers after 10 microseconds, - * we repeatedly block for 1-millisecond time periods. This approach - * has done well in testing, so there is no need for a config parameter. - */ -#define SRCU_RETRY_CHECK_DELAY 5 -#define SYNCHRONIZE_SRCU_TRYCOUNT 2 -#define SYNCHRONIZE_SRCU_EXP_TRYCOUNT 12 - -/* - * @@@ Wait until all pre-existing readers complete. Such readers - * will have used the index specified by "idx". - * the caller should ensures the ->completed is not changed while checking - * and idx = (->completed & 1) ^ 1 - */ -static bool try_check_zero(struct srcu_struct *sp, int idx, int trycount) -{ - for (;;) { - if (srcu_readers_active_idx_check(sp, idx)) - return true; - if (--trycount <= 0) - return false; - udelay(SRCU_RETRY_CHECK_DELAY); - } -} - -/* - * Increment the ->completed counter so that future SRCU readers will - * use the other rank of the ->c[] and ->seq[] arrays. This allows - * us to wait for pre-existing readers in a starvation-free manner. - */ -static void srcu_flip(struct srcu_struct *sp) -{ - sp->completed++; -} - -/* - * Enqueue an SRCU callback on the specified srcu_struct structure, - * initiating grace-period processing if it is not already running. - * - * Note that all CPUs must agree that the grace period extended beyond - * all pre-existing SRCU read-side critical section. On systems with - * more than one CPU, this means that when "func()" is invoked, each CPU - * is guaranteed to have executed a full memory barrier since the end of - * its last corresponding SRCU read-side critical section whose beginning - * preceded the call to call_rcu(). It also means that each CPU executing - * an SRCU read-side critical section that continues beyond the start of - * "func()" must have executed a memory barrier after the call_rcu() - * but before the beginning of that SRCU read-side critical section. - * Note that these guarantees include CPUs that are offline, idle, or - * executing in user mode, as well as CPUs that are executing in the kernel. - * - * Furthermore, if CPU A invoked call_rcu() and CPU B invoked the - * resulting SRCU callback function "func()", then both CPU A and CPU - * B are guaranteed to execute a full memory barrier during the time - * interval between the call to call_rcu() and the invocation of "func()". - * This guarantee applies even if CPU A and CPU B are the same CPU (but - * again only if the system has more than one CPU). - * - * Of course, these guarantees apply only for invocations of call_srcu(), - * srcu_read_lock(), and srcu_read_unlock() that are all passed the same - * srcu_struct structure. - */ -void call_srcu(struct srcu_struct *sp, struct rcu_head *head, - rcu_callback_t func) -{ - unsigned long flags; - - head->next = NULL; - head->func = func; - spin_lock_irqsave(&sp->queue_lock, flags); - rcu_batch_queue(&sp->batch_queue, head); - if (!sp->running) { - sp->running = true; - queue_delayed_work(system_power_efficient_wq, &sp->work, 0); - } - spin_unlock_irqrestore(&sp->queue_lock, flags); -} -EXPORT_SYMBOL_GPL(call_srcu); - -static void srcu_advance_batches(struct srcu_struct *sp, int trycount); -static void srcu_reschedule(struct srcu_struct *sp); - -/* - * Helper function for synchronize_srcu() and synchronize_srcu_expedited(). - */ -static void __synchronize_srcu(struct srcu_struct *sp, int trycount) -{ - struct rcu_synchronize rcu; - struct rcu_head *head = &rcu.head; - bool done = false; - - RCU_LOCKDEP_WARN(lock_is_held(&sp->dep_map) || - lock_is_held(&rcu_bh_lock_map) || - lock_is_held(&rcu_lock_map) || - lock_is_held(&rcu_sched_lock_map), - "Illegal synchronize_srcu() in same-type SRCU (or in RCU) read-side critical section"); - - might_sleep(); - init_completion(&rcu.completion); - - head->next = NULL; - head->func = wakeme_after_rcu; - spin_lock_irq(&sp->queue_lock); - if (!sp->running) { - /* steal the processing owner */ - sp->running = true; - rcu_batch_queue(&sp->batch_check0, head); - spin_unlock_irq(&sp->queue_lock); - - srcu_advance_batches(sp, trycount); - if (!rcu_batch_empty(&sp->batch_done)) { - BUG_ON(sp->batch_done.head != head); - rcu_batch_dequeue(&sp->batch_done); - done = true; - } - /* give the processing owner to work_struct */ - srcu_reschedule(sp); - } else { - rcu_batch_queue(&sp->batch_queue, head); - spin_unlock_irq(&sp->queue_lock); - } - - if (!done) - wait_for_completion(&rcu.completion); -} - -/** - * synchronize_srcu - wait for prior SRCU read-side critical-section completion - * @sp: srcu_struct with which to synchronize. - * - * Wait for the count to drain to zero of both indexes. To avoid the - * possible starvation of synchronize_srcu(), it waits for the count of - * the index=((->completed & 1) ^ 1) to drain to zero at first, - * and then flip the completed and wait for the count of the other index. - * - * Can block; must be called from process context. - * - * Note that it is illegal to call synchronize_srcu() from the corresponding - * SRCU read-side critical section; doing so will result in deadlock. - * However, it is perfectly legal to call synchronize_srcu() on one - * srcu_struct from some other srcu_struct's read-side critical section, - * as long as the resulting graph of srcu_structs is acyclic. - * - * There are memory-ordering constraints implied by synchronize_srcu(). - * On systems with more than one CPU, when synchronize_srcu() returns, - * each CPU is guaranteed to have executed a full memory barrier since - * the end of its last corresponding SRCU-sched read-side critical section - * whose beginning preceded the call to synchronize_srcu(). In addition, - * each CPU having an SRCU read-side critical section that extends beyond - * the return from synchronize_srcu() is guaranteed to have executed a - * full memory barrier after the beginning of synchronize_srcu() and before - * the beginning of that SRCU read-side critical section. Note that these - * guarantees include CPUs that are offline, idle, or executing in user mode, - * as well as CPUs that are executing in the kernel. - * - * Furthermore, if CPU A invoked synchronize_srcu(), which returned - * to its caller on CPU B, then both CPU A and CPU B are guaranteed - * to have executed a full memory barrier during the execution of - * synchronize_srcu(). This guarantee applies even if CPU A and CPU B - * are the same CPU, but again only if the system has more than one CPU. - * - * Of course, these memory-ordering guarantees apply only when - * synchronize_srcu(), srcu_read_lock(), and srcu_read_unlock() are - * passed the same srcu_struct structure. - */ -void synchronize_srcu(struct srcu_struct *sp) -{ - __synchronize_srcu(sp, (rcu_gp_is_expedited() && !rcu_gp_is_normal()) - ? SYNCHRONIZE_SRCU_EXP_TRYCOUNT - : SYNCHRONIZE_SRCU_TRYCOUNT); -} -EXPORT_SYMBOL_GPL(synchronize_srcu); - -/** - * synchronize_srcu_expedited - Brute-force SRCU grace period - * @sp: srcu_struct with which to synchronize. - * - * Wait for an SRCU grace period to elapse, but be more aggressive about - * spinning rather than blocking when waiting. - * - * Note that synchronize_srcu_expedited() has the same deadlock and - * memory-ordering properties as does synchronize_srcu(). - */ -void synchronize_srcu_expedited(struct srcu_struct *sp) -{ - __synchronize_srcu(sp, SYNCHRONIZE_SRCU_EXP_TRYCOUNT); -} -EXPORT_SYMBOL_GPL(synchronize_srcu_expedited); - -/** - * srcu_barrier - Wait until all in-flight call_srcu() callbacks complete. - * @sp: srcu_struct on which to wait for in-flight callbacks. - */ -void srcu_barrier(struct srcu_struct *sp) -{ - synchronize_srcu(sp); -} -EXPORT_SYMBOL_GPL(srcu_barrier); - -/** - * srcu_batches_completed - return batches completed. - * @sp: srcu_struct on which to report batch completion. - * - * Report the number of batches, correlated with, but not necessarily - * precisely the same as, the number of grace periods that have elapsed. - */ -unsigned long srcu_batches_completed(struct srcu_struct *sp) -{ - return sp->completed; -} -EXPORT_SYMBOL_GPL(srcu_batches_completed); - -#define SRCU_CALLBACK_BATCH 10 -#define SRCU_INTERVAL 1 - -/* - * Move any new SRCU callbacks to the first stage of the SRCU grace - * period pipeline. - */ -static void srcu_collect_new(struct srcu_struct *sp) -{ - if (!rcu_batch_empty(&sp->batch_queue)) { - spin_lock_irq(&sp->queue_lock); - rcu_batch_move(&sp->batch_check0, &sp->batch_queue); - spin_unlock_irq(&sp->queue_lock); - } -} - -/* - * Core SRCU state machine. Advance callbacks from ->batch_check0 to - * ->batch_check1 and then to ->batch_done as readers drain. - */ -static void srcu_advance_batches(struct srcu_struct *sp, int trycount) -{ - int idx = 1 ^ (sp->completed & 1); - - /* - * Because readers might be delayed for an extended period after - * fetching ->completed for their index, at any point in time there - * might well be readers using both idx=0 and idx=1. We therefore - * need to wait for readers to clear from both index values before - * invoking a callback. - */ - - if (rcu_batch_empty(&sp->batch_check0) && - rcu_batch_empty(&sp->batch_check1)) - return; /* no callbacks need to be advanced */ - - if (!try_check_zero(sp, idx, trycount)) - return; /* failed to advance, will try after SRCU_INTERVAL */ - - /* - * The callbacks in ->batch_check1 have already done with their - * first zero check and flip back when they were enqueued on - * ->batch_check0 in a previous invocation of srcu_advance_batches(). - * (Presumably try_check_zero() returned false during that - * invocation, leaving the callbacks stranded on ->batch_check1.) - * They are therefore ready to invoke, so move them to ->batch_done. - */ - rcu_batch_move(&sp->batch_done, &sp->batch_check1); - - if (rcu_batch_empty(&sp->batch_check0)) - return; /* no callbacks need to be advanced */ - srcu_flip(sp); - - /* - * The callbacks in ->batch_check0 just finished their - * first check zero and flip, so move them to ->batch_check1 - * for future checking on the other idx. - */ - rcu_batch_move(&sp->batch_check1, &sp->batch_check0); - - /* - * SRCU read-side critical sections are normally short, so check - * at least twice in quick succession after a flip. - */ - trycount = trycount < 2 ? 2 : trycount; - if (!try_check_zero(sp, idx^1, trycount)) - return; /* failed to advance, will try after SRCU_INTERVAL */ - - /* - * The callbacks in ->batch_check1 have now waited for all - * pre-existing readers using both idx values. They are therefore - * ready to invoke, so move them to ->batch_done. - */ - rcu_batch_move(&sp->batch_done, &sp->batch_check1); -} - -/* - * Invoke a limited number of SRCU callbacks that have passed through - * their grace period. If there are more to do, SRCU will reschedule - * the workqueue. - */ -static void srcu_invoke_callbacks(struct srcu_struct *sp) -{ - int i; - struct rcu_head *head; - - for (i = 0; i < SRCU_CALLBACK_BATCH; i++) { - head = rcu_batch_dequeue(&sp->batch_done); - if (!head) - break; - local_bh_disable(); - head->func(head); - local_bh_enable(); - } -} - -/* - * Finished one round of SRCU grace period. Start another if there are - * more SRCU callbacks queued, otherwise put SRCU into not-running state. - */ -static void srcu_reschedule(struct srcu_struct *sp) -{ - bool pending = true; - - if (rcu_batch_empty(&sp->batch_done) && - rcu_batch_empty(&sp->batch_check1) && - rcu_batch_empty(&sp->batch_check0) && - rcu_batch_empty(&sp->batch_queue)) { - spin_lock_irq(&sp->queue_lock); - if (rcu_batch_empty(&sp->batch_done) && - rcu_batch_empty(&sp->batch_check1) && - rcu_batch_empty(&sp->batch_check0) && - rcu_batch_empty(&sp->batch_queue)) { - sp->running = false; - pending = false; - } - spin_unlock_irq(&sp->queue_lock); - } - - if (pending) - queue_delayed_work(system_power_efficient_wq, - &sp->work, SRCU_INTERVAL); -} - -/* - * This is the work-queue function that handles SRCU grace periods. - */ -void process_srcu(struct work_struct *work) -{ - struct srcu_struct *sp; - - sp = container_of(work, struct srcu_struct, work.work); - - srcu_collect_new(sp); - srcu_advance_batches(sp, 1); - srcu_invoke_callbacks(sp); - srcu_reschedule(sp); -} -EXPORT_SYMBOL_GPL(process_srcu); diff --git a/kernel/rcu/srcutiny.c b/kernel/rcu/srcutiny.c new file mode 100644 index 000000000000..1a1c1047d2ed --- /dev/null +++ b/kernel/rcu/srcutiny.c @@ -0,0 +1,195 @@ +/* + * Sleepable Read-Copy Update mechanism for mutual exclusion, + * tiny version for non-preemptible single-CPU use. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, you can access it online at + * http://www.gnu.org/licenses/gpl-2.0.html. + * + * Copyright (C) IBM Corporation, 2017 + * + * Author: Paul McKenney <[email protected]> + */ + +#include <linux/export.h> +#include <linux/mutex.h> +#include <linux/preempt.h> +#include <linux/rcupdate_wait.h> +#include <linux/sched.h> +#include <linux/delay.h> +#include <linux/srcu.h> + +#include <linux/rcu_node_tree.h> +#include "rcu_segcblist.h" +#include "rcu.h" + +static int init_srcu_struct_fields(struct srcu_struct *sp) +{ + sp->srcu_lock_nesting[0] = 0; + sp->srcu_lock_nesting[1] = 0; + init_swait_queue_head(&sp->srcu_wq); + sp->srcu_cb_head = NULL; + sp->srcu_cb_tail = &sp->srcu_cb_head; + sp->srcu_gp_running = false; + sp->srcu_gp_waiting = false; + sp->srcu_idx = 0; + INIT_WORK(&sp->srcu_work, srcu_drive_gp); + return 0; +} + +#ifdef CONFIG_DEBUG_LOCK_ALLOC + +int __init_srcu_struct(struct srcu_struct *sp, const char *name, + struct lock_class_key *key) +{ + /* Don't re-initialize a lock while it is held. */ + debug_check_no_locks_freed((void *)sp, sizeof(*sp)); + lockdep_init_map(&sp->dep_map, name, key, 0); + return init_srcu_struct_fields(sp); +} +EXPORT_SYMBOL_GPL(__init_srcu_struct); + +#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +/* + * init_srcu_struct - initialize a sleep-RCU structure + * @sp: structure to initialize. + * + * Must invoke this on a given srcu_struct before passing that srcu_struct + * to any other function. Each srcu_struct represents a separate domain + * of SRCU protection. + */ +int init_srcu_struct(struct srcu_struct *sp) +{ + return init_srcu_struct_fields(sp); +} +EXPORT_SYMBOL_GPL(init_srcu_struct); + +#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +/* + * cleanup_srcu_struct - deconstruct a sleep-RCU structure + * @sp: structure to clean up. + * + * Must invoke this after you are finished using a given srcu_struct that + * was initialized via init_srcu_struct(), else you leak memory. + */ +void cleanup_srcu_struct(struct srcu_struct *sp) +{ + WARN_ON(sp->srcu_lock_nesting[0] || sp->srcu_lock_nesting[1]); + flush_work(&sp->srcu_work); + WARN_ON(sp->srcu_gp_running); + WARN_ON(sp->srcu_gp_waiting); + WARN_ON(sp->srcu_cb_head); + WARN_ON(&sp->srcu_cb_head != sp->srcu_cb_tail); +} +EXPORT_SYMBOL_GPL(cleanup_srcu_struct); + +/* + * Removes the count for the old reader from the appropriate element of + * the srcu_struct. + */ +void __srcu_read_unlock(struct srcu_struct *sp, int idx) +{ + int newval = sp->srcu_lock_nesting[idx] - 1; + + WRITE_ONCE(sp->srcu_lock_nesting[idx], newval); + if (!newval && READ_ONCE(sp->srcu_gp_waiting)) + swake_up(&sp->srcu_wq); +} +EXPORT_SYMBOL_GPL(__srcu_read_unlock); + +/* + * Workqueue handler to drive one grace period and invoke any callbacks + * that become ready as a result. Single-CPU and !PREEMPT operation + * means that we get away with murder on synchronization. ;-) + */ +void srcu_drive_gp(struct work_struct *wp) +{ + int idx; + struct rcu_head *lh; + struct rcu_head *rhp; + struct srcu_struct *sp; + + sp = container_of(wp, struct srcu_struct, srcu_work); + if (sp->srcu_gp_running || !READ_ONCE(sp->srcu_cb_head)) + return; /* Already running or nothing to do. */ + + /* Remove recently arrived callbacks and wait for readers. */ + WRITE_ONCE(sp->srcu_gp_running, true); + local_irq_disable(); + lh = sp->srcu_cb_head; + sp->srcu_cb_head = NULL; + sp->srcu_cb_tail = &sp->srcu_cb_head; + local_irq_enable(); + idx = sp->srcu_idx; + WRITE_ONCE(sp->srcu_idx, !sp->srcu_idx); + WRITE_ONCE(sp->srcu_gp_waiting, true); /* srcu_read_unlock() wakes! */ + swait_event(sp->srcu_wq, !READ_ONCE(sp->srcu_lock_nesting[idx])); + WRITE_ONCE(sp->srcu_gp_waiting, false); /* srcu_read_unlock() cheap. */ + + /* Invoke the callbacks we removed above. */ + while (lh) { + rhp = lh; + lh = lh->next; + local_bh_disable(); + rhp->func(rhp); + local_bh_enable(); + } + + /* + * Enable rescheduling, and if there are more callbacks, + * reschedule ourselves. This can race with a call_srcu() + * at interrupt level, but the ->srcu_gp_running checks will + * straighten that out. + */ + WRITE_ONCE(sp->srcu_gp_running, false); + if (READ_ONCE(sp->srcu_cb_head)) + schedule_work(&sp->srcu_work); +} +EXPORT_SYMBOL_GPL(srcu_drive_gp); + +/* + * Enqueue an SRCU callback on the specified srcu_struct structure, + * initiating grace-period processing if it is not already running. + */ +void call_srcu(struct srcu_struct *sp, struct rcu_head *rhp, + rcu_callback_t func) +{ + unsigned long flags; + + rhp->func = func; + rhp->next = NULL; + local_irq_save(flags); + *sp->srcu_cb_tail = rhp; + sp->srcu_cb_tail = &rhp->next; + local_irq_restore(flags); + if (!READ_ONCE(sp->srcu_gp_running)) + schedule_work(&sp->srcu_work); +} +EXPORT_SYMBOL_GPL(call_srcu); + +/* + * synchronize_srcu - wait for prior SRCU read-side critical-section completion + */ +void synchronize_srcu(struct srcu_struct *sp) +{ + struct rcu_synchronize rs; + + init_rcu_head_on_stack(&rs.head); + init_completion(&rs.completion); + call_srcu(sp, &rs.head, wakeme_after_rcu); + wait_for_completion(&rs.completion); + destroy_rcu_head_on_stack(&rs.head); +} +EXPORT_SYMBOL_GPL(synchronize_srcu); diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c new file mode 100644 index 000000000000..d0ca524bf042 --- /dev/null +++ b/kernel/rcu/srcutree.c @@ -0,0 +1,1227 @@ +/* + * Sleepable Read-Copy Update mechanism for mutual exclusion. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, you can access it online at + * http://www.gnu.org/licenses/gpl-2.0.html. + * + * Copyright (C) IBM Corporation, 2006 + * Copyright (C) Fujitsu, 2012 + * + * Author: Paul McKenney <[email protected]> + * Lai Jiangshan <[email protected]> + * + * For detailed explanation of Read-Copy Update mechanism see - + * Documentation/RCU/ *.txt + * + */ + +#include <linux/export.h> +#include <linux/mutex.h> +#include <linux/percpu.h> +#include <linux/preempt.h> +#include <linux/rcupdate_wait.h> +#include <linux/sched.h> +#include <linux/smp.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/srcu.h> + +#include "rcu.h" +#include "rcu_segcblist.h" + +/* Holdoff in nanoseconds for auto-expediting. */ +#define DEFAULT_SRCU_EXP_HOLDOFF (25 * 1000) +static ulong exp_holdoff = DEFAULT_SRCU_EXP_HOLDOFF; +module_param(exp_holdoff, ulong, 0444); + +/* Overflow-check frequency. N bits roughly says every 2**N grace periods. */ +static ulong counter_wrap_check = (ULONG_MAX >> 2); +module_param(counter_wrap_check, ulong, 0444); + +static void srcu_invoke_callbacks(struct work_struct *work); +static void srcu_reschedule(struct srcu_struct *sp, unsigned long delay); + +/* + * Initialize SRCU combining tree. Note that statically allocated + * srcu_struct structures might already have srcu_read_lock() and + * srcu_read_unlock() running against them. So if the is_static parameter + * is set, don't initialize ->srcu_lock_count[] and ->srcu_unlock_count[]. + */ +static void init_srcu_struct_nodes(struct srcu_struct *sp, bool is_static) +{ + int cpu; + int i; + int level = 0; + int levelspread[RCU_NUM_LVLS]; + struct srcu_data *sdp; + struct srcu_node *snp; + struct srcu_node *snp_first; + + /* Work out the overall tree geometry. */ + sp->level[0] = &sp->node[0]; + for (i = 1; i < rcu_num_lvls; i++) + sp->level[i] = sp->level[i - 1] + num_rcu_lvl[i - 1]; + rcu_init_levelspread(levelspread, num_rcu_lvl); + + /* Each pass through this loop initializes one srcu_node structure. */ + rcu_for_each_node_breadth_first(sp, snp) { + raw_spin_lock_init(&ACCESS_PRIVATE(snp, lock)); + WARN_ON_ONCE(ARRAY_SIZE(snp->srcu_have_cbs) != + ARRAY_SIZE(snp->srcu_data_have_cbs)); + for (i = 0; i < ARRAY_SIZE(snp->srcu_have_cbs); i++) { + snp->srcu_have_cbs[i] = 0; + snp->srcu_data_have_cbs[i] = 0; + } + snp->srcu_gp_seq_needed_exp = 0; + snp->grplo = -1; + snp->grphi = -1; + if (snp == &sp->node[0]) { + /* Root node, special case. */ + snp->srcu_parent = NULL; + continue; + } + + /* Non-root node. */ + if (snp == sp->level[level + 1]) + level++; + snp->srcu_parent = sp->level[level - 1] + + (snp - sp->level[level]) / + levelspread[level - 1]; + } + + /* + * Initialize the per-CPU srcu_data array, which feeds into the + * leaves of the srcu_node tree. + */ + WARN_ON_ONCE(ARRAY_SIZE(sdp->srcu_lock_count) != + ARRAY_SIZE(sdp->srcu_unlock_count)); + level = rcu_num_lvls - 1; + snp_first = sp->level[level]; + for_each_possible_cpu(cpu) { + sdp = per_cpu_ptr(sp->sda, cpu); + raw_spin_lock_init(&ACCESS_PRIVATE(sdp, lock)); + rcu_segcblist_init(&sdp->srcu_cblist); + sdp->srcu_cblist_invoking = false; + sdp->srcu_gp_seq_needed = sp->srcu_gp_seq; + sdp->srcu_gp_seq_needed_exp = sp->srcu_gp_seq; + sdp->mynode = &snp_first[cpu / levelspread[level]]; + for (snp = sdp->mynode; snp != NULL; snp = snp->srcu_parent) { + if (snp->grplo < 0) + snp->grplo = cpu; + snp->grphi = cpu; + } + sdp->cpu = cpu; + INIT_DELAYED_WORK(&sdp->work, srcu_invoke_callbacks); + sdp->sp = sp; + sdp->grpmask = 1 << (cpu - sdp->mynode->grplo); + if (is_static) + continue; + + /* Dynamically allocated, better be no srcu_read_locks()! */ + for (i = 0; i < ARRAY_SIZE(sdp->srcu_lock_count); i++) { + sdp->srcu_lock_count[i] = 0; + sdp->srcu_unlock_count[i] = 0; + } + } +} + +/* + * Initialize non-compile-time initialized fields, including the + * associated srcu_node and srcu_data structures. The is_static + * parameter is passed through to init_srcu_struct_nodes(), and + * also tells us that ->sda has already been wired up to srcu_data. + */ +static int init_srcu_struct_fields(struct srcu_struct *sp, bool is_static) +{ + mutex_init(&sp->srcu_cb_mutex); + mutex_init(&sp->srcu_gp_mutex); + sp->srcu_idx = 0; + sp->srcu_gp_seq = 0; + sp->srcu_barrier_seq = 0; + mutex_init(&sp->srcu_barrier_mutex); + atomic_set(&sp->srcu_barrier_cpu_cnt, 0); + INIT_DELAYED_WORK(&sp->work, process_srcu); + if (!is_static) + sp->sda = alloc_percpu(struct srcu_data); + init_srcu_struct_nodes(sp, is_static); + sp->srcu_gp_seq_needed_exp = 0; + sp->srcu_last_gp_end = ktime_get_mono_fast_ns(); + smp_store_release(&sp->srcu_gp_seq_needed, 0); /* Init done. */ + return sp->sda ? 0 : -ENOMEM; +} + +#ifdef CONFIG_DEBUG_LOCK_ALLOC + +int __init_srcu_struct(struct srcu_struct *sp, const char *name, + struct lock_class_key *key) +{ + /* Don't re-initialize a lock while it is held. */ + debug_check_no_locks_freed((void *)sp, sizeof(*sp)); + lockdep_init_map(&sp->dep_map, name, key, 0); + raw_spin_lock_init(&ACCESS_PRIVATE(sp, lock)); + return init_srcu_struct_fields(sp, false); +} +EXPORT_SYMBOL_GPL(__init_srcu_struct); + +#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +/** + * init_srcu_struct - initialize a sleep-RCU structure + * @sp: structure to initialize. + * + * Must invoke this on a given srcu_struct before passing that srcu_struct + * to any other function. Each srcu_struct represents a separate domain + * of SRCU protection. + */ +int init_srcu_struct(struct srcu_struct *sp) +{ + raw_spin_lock_init(&ACCESS_PRIVATE(sp, lock)); + return init_srcu_struct_fields(sp, false); +} +EXPORT_SYMBOL_GPL(init_srcu_struct); + +#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +/* + * First-use initialization of statically allocated srcu_struct + * structure. Wiring up the combining tree is more than can be + * done with compile-time initialization, so this check is added + * to each update-side SRCU primitive. Use sp->lock, which -is- + * compile-time initialized, to resolve races involving multiple + * CPUs trying to garner first-use privileges. + */ +static void check_init_srcu_struct(struct srcu_struct *sp) +{ + unsigned long flags; + + WARN_ON_ONCE(rcu_scheduler_active == RCU_SCHEDULER_INIT); + /* The smp_load_acquire() pairs with the smp_store_release(). */ + if (!rcu_seq_state(smp_load_acquire(&sp->srcu_gp_seq_needed))) /*^^^*/ + return; /* Already initialized. */ + raw_spin_lock_irqsave_rcu_node(sp, flags); + if (!rcu_seq_state(sp->srcu_gp_seq_needed)) { + raw_spin_unlock_irqrestore_rcu_node(sp, flags); + return; + } + init_srcu_struct_fields(sp, true); + raw_spin_unlock_irqrestore_rcu_node(sp, flags); +} + +/* + * Returns approximate total of the readers' ->srcu_lock_count[] values + * for the rank of per-CPU counters specified by idx. + */ +static unsigned long srcu_readers_lock_idx(struct srcu_struct *sp, int idx) +{ + int cpu; + unsigned long sum = 0; + + for_each_possible_cpu(cpu) { + struct srcu_data *cpuc = per_cpu_ptr(sp->sda, cpu); + + sum += READ_ONCE(cpuc->srcu_lock_count[idx]); + } + return sum; +} + +/* + * Returns approximate total of the readers' ->srcu_unlock_count[] values + * for the rank of per-CPU counters specified by idx. + */ +static unsigned long srcu_readers_unlock_idx(struct srcu_struct *sp, int idx) +{ + int cpu; + unsigned long sum = 0; + + for_each_possible_cpu(cpu) { + struct srcu_data *cpuc = per_cpu_ptr(sp->sda, cpu); + + sum += READ_ONCE(cpuc->srcu_unlock_count[idx]); + } + return sum; +} + +/* + * Return true if the number of pre-existing readers is determined to + * be zero. + */ +static bool srcu_readers_active_idx_check(struct srcu_struct *sp, int idx) +{ + unsigned long unlocks; + + unlocks = srcu_readers_unlock_idx(sp, idx); + + /* + * Make sure that a lock is always counted if the corresponding + * unlock is counted. Needs to be a smp_mb() as the read side may + * contain a read from a variable that is written to before the + * synchronize_srcu() in the write side. In this case smp_mb()s + * A and B act like the store buffering pattern. + * + * This smp_mb() also pairs with smp_mb() C to prevent accesses + * after the synchronize_srcu() from being executed before the + * grace period ends. + */ + smp_mb(); /* A */ + + /* + * If the locks are the same as the unlocks, then there must have + * been no readers on this index at some time in between. This does + * not mean that there are no more readers, as one could have read + * the current index but not have incremented the lock counter yet. + * + * So suppose that the updater is preempted here for so long + * that more than ULONG_MAX non-nested readers come and go in + * the meantime. It turns out that this cannot result in overflow + * because if a reader modifies its unlock count after we read it + * above, then that reader's next load of ->srcu_idx is guaranteed + * to get the new value, which will cause it to operate on the + * other bank of counters, where it cannot contribute to the + * overflow of these counters. This means that there is a maximum + * of 2*NR_CPUS increments, which cannot overflow given current + * systems, especially not on 64-bit systems. + * + * OK, how about nesting? This does impose a limit on nesting + * of floor(ULONG_MAX/NR_CPUS/2), which should be sufficient, + * especially on 64-bit systems. + */ + return srcu_readers_lock_idx(sp, idx) == unlocks; +} + +/** + * srcu_readers_active - returns true if there are readers. and false + * otherwise + * @sp: which srcu_struct to count active readers (holding srcu_read_lock). + * + * Note that this is not an atomic primitive, and can therefore suffer + * severe errors when invoked on an active srcu_struct. That said, it + * can be useful as an error check at cleanup time. + */ +static bool srcu_readers_active(struct srcu_struct *sp) +{ + int cpu; + unsigned long sum = 0; + + for_each_possible_cpu(cpu) { + struct srcu_data *cpuc = per_cpu_ptr(sp->sda, cpu); + + sum += READ_ONCE(cpuc->srcu_lock_count[0]); + sum += READ_ONCE(cpuc->srcu_lock_count[1]); + sum -= READ_ONCE(cpuc->srcu_unlock_count[0]); + sum -= READ_ONCE(cpuc->srcu_unlock_count[1]); + } + return sum; +} + +#define SRCU_INTERVAL 1 + +/* + * Return grace-period delay, zero if there are expedited grace + * periods pending, SRCU_INTERVAL otherwise. + */ +static unsigned long srcu_get_delay(struct srcu_struct *sp) +{ + if (ULONG_CMP_LT(READ_ONCE(sp->srcu_gp_seq), + READ_ONCE(sp->srcu_gp_seq_needed_exp))) + return 0; + return SRCU_INTERVAL; +} + +/** + * cleanup_srcu_struct - deconstruct a sleep-RCU structure + * @sp: structure to clean up. + * + * Must invoke this after you are finished using a given srcu_struct that + * was initialized via init_srcu_struct(), else you leak memory. + */ +void cleanup_srcu_struct(struct srcu_struct *sp) +{ + int cpu; + + if (WARN_ON(!srcu_get_delay(sp))) + return; /* Leakage unless caller handles error. */ + if (WARN_ON(srcu_readers_active(sp))) + return; /* Leakage unless caller handles error. */ + flush_delayed_work(&sp->work); + for_each_possible_cpu(cpu) + flush_delayed_work(&per_cpu_ptr(sp->sda, cpu)->work); + if (WARN_ON(rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)) != SRCU_STATE_IDLE) || + WARN_ON(srcu_readers_active(sp))) { + pr_info("cleanup_srcu_struct: Active srcu_struct %p state: %d\n", sp, rcu_seq_state(READ_ONCE(sp->srcu_gp_seq))); + return; /* Caller forgot to stop doing call_srcu()? */ + } + free_percpu(sp->sda); + sp->sda = NULL; +} +EXPORT_SYMBOL_GPL(cleanup_srcu_struct); + +/* + * Counts the new reader in the appropriate per-CPU element of the + * srcu_struct. + * Returns an index that must be passed to the matching srcu_read_unlock(). + */ +int __srcu_read_lock(struct srcu_struct *sp) +{ + int idx; + + idx = READ_ONCE(sp->srcu_idx) & 0x1; + this_cpu_inc(sp->sda->srcu_lock_count[idx]); + smp_mb(); /* B */ /* Avoid leaking the critical section. */ + return idx; +} +EXPORT_SYMBOL_GPL(__srcu_read_lock); + +/* + * Removes the count for the old reader from the appropriate per-CPU + * element of the srcu_struct. Note that this may well be a different + * CPU than that which was incremented by the corresponding srcu_read_lock(). + */ +void __srcu_read_unlock(struct srcu_struct *sp, int idx) +{ + smp_mb(); /* C */ /* Avoid leaking the critical section. */ + this_cpu_inc(sp->sda->srcu_unlock_count[idx]); +} +EXPORT_SYMBOL_GPL(__srcu_read_unlock); + +/* + * We use an adaptive strategy for synchronize_srcu() and especially for + * synchronize_srcu_expedited(). We spin for a fixed time period + * (defined below) to allow SRCU readers to exit their read-side critical + * sections. If there are still some readers after a few microseconds, + * we repeatedly block for 1-millisecond time periods. + */ +#define SRCU_RETRY_CHECK_DELAY 5 + +/* + * Start an SRCU grace period. + */ +static void srcu_gp_start(struct srcu_struct *sp) +{ + struct srcu_data *sdp = this_cpu_ptr(sp->sda); + int state; + + lockdep_assert_held(&sp->lock); + WARN_ON_ONCE(ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed)); + rcu_segcblist_advance(&sdp->srcu_cblist, + rcu_seq_current(&sp->srcu_gp_seq)); + (void)rcu_segcblist_accelerate(&sdp->srcu_cblist, + rcu_seq_snap(&sp->srcu_gp_seq)); + smp_mb(); /* Order prior store to ->srcu_gp_seq_needed vs. GP start. */ + rcu_seq_start(&sp->srcu_gp_seq); + state = rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)); + WARN_ON_ONCE(state != SRCU_STATE_SCAN1); +} + +/* + * Track online CPUs to guide callback workqueue placement. + */ +DEFINE_PER_CPU(bool, srcu_online); + +void srcu_online_cpu(unsigned int cpu) +{ + WRITE_ONCE(per_cpu(srcu_online, cpu), true); +} + +void srcu_offline_cpu(unsigned int cpu) +{ + WRITE_ONCE(per_cpu(srcu_online, cpu), false); +} + +/* + * Place the workqueue handler on the specified CPU if online, otherwise + * just run it whereever. This is useful for placing workqueue handlers + * that are to invoke the specified CPU's callbacks. + */ +static bool srcu_queue_delayed_work_on(int cpu, struct workqueue_struct *wq, + struct delayed_work *dwork, + unsigned long delay) +{ + bool ret; + + preempt_disable(); + if (READ_ONCE(per_cpu(srcu_online, cpu))) + ret = queue_delayed_work_on(cpu, wq, dwork, delay); + else + ret = queue_delayed_work(wq, dwork, delay); + preempt_enable(); + return ret; +} + +/* + * Schedule callback invocation for the specified srcu_data structure, + * if possible, on the corresponding CPU. + */ +static void srcu_schedule_cbs_sdp(struct srcu_data *sdp, unsigned long delay) +{ + srcu_queue_delayed_work_on(sdp->cpu, system_power_efficient_wq, + &sdp->work, delay); +} + +/* + * Schedule callback invocation for all srcu_data structures associated + * with the specified srcu_node structure that have callbacks for the + * just-completed grace period, the one corresponding to idx. If possible, + * schedule this invocation on the corresponding CPUs. + */ +static void srcu_schedule_cbs_snp(struct srcu_struct *sp, struct srcu_node *snp, + unsigned long mask, unsigned long delay) +{ + int cpu; + + for (cpu = snp->grplo; cpu <= snp->grphi; cpu++) { + if (!(mask & (1 << (cpu - snp->grplo)))) + continue; + srcu_schedule_cbs_sdp(per_cpu_ptr(sp->sda, cpu), delay); + } +} + +/* + * Note the end of an SRCU grace period. Initiates callback invocation + * and starts a new grace period if needed. + * + * The ->srcu_cb_mutex acquisition does not protect any data, but + * instead prevents more than one grace period from starting while we + * are initiating callback invocation. This allows the ->srcu_have_cbs[] + * array to have a finite number of elements. + */ +static void srcu_gp_end(struct srcu_struct *sp) +{ + unsigned long cbdelay; + bool cbs; + int cpu; + unsigned long flags; + unsigned long gpseq; + int idx; + int idxnext; + unsigned long mask; + struct srcu_data *sdp; + struct srcu_node *snp; + + /* Prevent more than one additional grace period. */ + mutex_lock(&sp->srcu_cb_mutex); + + /* End the current grace period. */ + raw_spin_lock_irq_rcu_node(sp); + idx = rcu_seq_state(sp->srcu_gp_seq); + WARN_ON_ONCE(idx != SRCU_STATE_SCAN2); + cbdelay = srcu_get_delay(sp); + sp->srcu_last_gp_end = ktime_get_mono_fast_ns(); + rcu_seq_end(&sp->srcu_gp_seq); + gpseq = rcu_seq_current(&sp->srcu_gp_seq); + if (ULONG_CMP_LT(sp->srcu_gp_seq_needed_exp, gpseq)) + sp->srcu_gp_seq_needed_exp = gpseq; + raw_spin_unlock_irq_rcu_node(sp); + mutex_unlock(&sp->srcu_gp_mutex); + /* A new grace period can start at this point. But only one. */ + + /* Initiate callback invocation as needed. */ + idx = rcu_seq_ctr(gpseq) % ARRAY_SIZE(snp->srcu_have_cbs); + idxnext = (idx + 1) % ARRAY_SIZE(snp->srcu_have_cbs); + rcu_for_each_node_breadth_first(sp, snp) { + raw_spin_lock_irq_rcu_node(snp); + cbs = false; + if (snp >= sp->level[rcu_num_lvls - 1]) + cbs = snp->srcu_have_cbs[idx] == gpseq; + snp->srcu_have_cbs[idx] = gpseq; + rcu_seq_set_state(&snp->srcu_have_cbs[idx], 1); + if (ULONG_CMP_LT(snp->srcu_gp_seq_needed_exp, gpseq)) + snp->srcu_gp_seq_needed_exp = gpseq; + mask = snp->srcu_data_have_cbs[idx]; + snp->srcu_data_have_cbs[idx] = 0; + raw_spin_unlock_irq_rcu_node(snp); + if (cbs) + srcu_schedule_cbs_snp(sp, snp, mask, cbdelay); + + /* Occasionally prevent srcu_data counter wrap. */ + if (!(gpseq & counter_wrap_check)) + for (cpu = snp->grplo; cpu <= snp->grphi; cpu++) { + sdp = per_cpu_ptr(sp->sda, cpu); + raw_spin_lock_irqsave_rcu_node(sdp, flags); + if (ULONG_CMP_GE(gpseq, + sdp->srcu_gp_seq_needed + 100)) + sdp->srcu_gp_seq_needed = gpseq; + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); + } + } + + /* Callback initiation done, allow grace periods after next. */ + mutex_unlock(&sp->srcu_cb_mutex); + + /* Start a new grace period if needed. */ + raw_spin_lock_irq_rcu_node(sp); + gpseq = rcu_seq_current(&sp->srcu_gp_seq); + if (!rcu_seq_state(gpseq) && + ULONG_CMP_LT(gpseq, sp->srcu_gp_seq_needed)) { + srcu_gp_start(sp); + raw_spin_unlock_irq_rcu_node(sp); + /* Throttle expedited grace periods: Should be rare! */ + srcu_reschedule(sp, rcu_seq_ctr(gpseq) & 0x3ff + ? 0 : SRCU_INTERVAL); + } else { + raw_spin_unlock_irq_rcu_node(sp); + } +} + +/* + * Funnel-locking scheme to scalably mediate many concurrent expedited + * grace-period requests. This function is invoked for the first known + * expedited request for a grace period that has already been requested, + * but without expediting. To start a completely new grace period, + * whether expedited or not, use srcu_funnel_gp_start() instead. + */ +static void srcu_funnel_exp_start(struct srcu_struct *sp, struct srcu_node *snp, + unsigned long s) +{ + unsigned long flags; + + for (; snp != NULL; snp = snp->srcu_parent) { + if (rcu_seq_done(&sp->srcu_gp_seq, s) || + ULONG_CMP_GE(READ_ONCE(snp->srcu_gp_seq_needed_exp), s)) + return; + raw_spin_lock_irqsave_rcu_node(snp, flags); + if (ULONG_CMP_GE(snp->srcu_gp_seq_needed_exp, s)) { + raw_spin_unlock_irqrestore_rcu_node(snp, flags); + return; + } + WRITE_ONCE(snp->srcu_gp_seq_needed_exp, s); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); + } + raw_spin_lock_irqsave_rcu_node(sp, flags); + if (!ULONG_CMP_LT(sp->srcu_gp_seq_needed_exp, s)) + sp->srcu_gp_seq_needed_exp = s; + raw_spin_unlock_irqrestore_rcu_node(sp, flags); +} + +/* + * Funnel-locking scheme to scalably mediate many concurrent grace-period + * requests. The winner has to do the work of actually starting grace + * period s. Losers must either ensure that their desired grace-period + * number is recorded on at least their leaf srcu_node structure, or they + * must take steps to invoke their own callbacks. + */ +static void srcu_funnel_gp_start(struct srcu_struct *sp, struct srcu_data *sdp, + unsigned long s, bool do_norm) +{ + unsigned long flags; + int idx = rcu_seq_ctr(s) % ARRAY_SIZE(sdp->mynode->srcu_have_cbs); + struct srcu_node *snp = sdp->mynode; + unsigned long snp_seq; + + /* Each pass through the loop does one level of the srcu_node tree. */ + for (; snp != NULL; snp = snp->srcu_parent) { + if (rcu_seq_done(&sp->srcu_gp_seq, s) && snp != sdp->mynode) + return; /* GP already done and CBs recorded. */ + raw_spin_lock_irqsave_rcu_node(snp, flags); + if (ULONG_CMP_GE(snp->srcu_have_cbs[idx], s)) { + snp_seq = snp->srcu_have_cbs[idx]; + if (snp == sdp->mynode && snp_seq == s) + snp->srcu_data_have_cbs[idx] |= sdp->grpmask; + raw_spin_unlock_irqrestore_rcu_node(snp, flags); + if (snp == sdp->mynode && snp_seq != s) { + srcu_schedule_cbs_sdp(sdp, do_norm + ? SRCU_INTERVAL + : 0); + return; + } + if (!do_norm) + srcu_funnel_exp_start(sp, snp, s); + return; + } + snp->srcu_have_cbs[idx] = s; + if (snp == sdp->mynode) + snp->srcu_data_have_cbs[idx] |= sdp->grpmask; + if (!do_norm && ULONG_CMP_LT(snp->srcu_gp_seq_needed_exp, s)) + snp->srcu_gp_seq_needed_exp = s; + raw_spin_unlock_irqrestore_rcu_node(snp, flags); + } + + /* Top of tree, must ensure the grace period will be started. */ + raw_spin_lock_irqsave_rcu_node(sp, flags); + if (ULONG_CMP_LT(sp->srcu_gp_seq_needed, s)) { + /* + * Record need for grace period s. Pair with load + * acquire setting up for initialization. + */ + smp_store_release(&sp->srcu_gp_seq_needed, s); /*^^^*/ + } + if (!do_norm && ULONG_CMP_LT(sp->srcu_gp_seq_needed_exp, s)) + sp->srcu_gp_seq_needed_exp = s; + + /* If grace period not already done and none in progress, start it. */ + if (!rcu_seq_done(&sp->srcu_gp_seq, s) && + rcu_seq_state(sp->srcu_gp_seq) == SRCU_STATE_IDLE) { + WARN_ON_ONCE(ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed)); + srcu_gp_start(sp); + queue_delayed_work(system_power_efficient_wq, &sp->work, + srcu_get_delay(sp)); + } + raw_spin_unlock_irqrestore_rcu_node(sp, flags); +} + +/* + * Wait until all readers counted by array index idx complete, but + * loop an additional time if there is an expedited grace period pending. + * The caller must ensure that ->srcu_idx is not changed while checking. + */ +static bool try_check_zero(struct srcu_struct *sp, int idx, int trycount) +{ + for (;;) { + if (srcu_readers_active_idx_check(sp, idx)) + return true; + if (--trycount + !srcu_get_delay(sp) <= 0) + return false; + udelay(SRCU_RETRY_CHECK_DELAY); + } +} + +/* + * Increment the ->srcu_idx counter so that future SRCU readers will + * use the other rank of the ->srcu_(un)lock_count[] arrays. This allows + * us to wait for pre-existing readers in a starvation-free manner. + */ +static void srcu_flip(struct srcu_struct *sp) +{ + /* + * Ensure that if this updater saw a given reader's increment + * from __srcu_read_lock(), that reader was using an old value + * of ->srcu_idx. Also ensure that if a given reader sees the + * new value of ->srcu_idx, this updater's earlier scans cannot + * have seen that reader's increments (which is OK, because this + * grace period need not wait on that reader). + */ + smp_mb(); /* E */ /* Pairs with B and C. */ + + WRITE_ONCE(sp->srcu_idx, sp->srcu_idx + 1); + + /* + * Ensure that if the updater misses an __srcu_read_unlock() + * increment, that task's next __srcu_read_lock() will see the + * above counter update. Note that both this memory barrier + * and the one in srcu_readers_active_idx_check() provide the + * guarantee for __srcu_read_lock(). + */ + smp_mb(); /* D */ /* Pairs with C. */ +} + +/* + * If SRCU is likely idle, return true, otherwise return false. + * + * Note that it is OK for several current from-idle requests for a new + * grace period from idle to specify expediting because they will all end + * up requesting the same grace period anyhow. So no loss. + * + * Note also that if any CPU (including the current one) is still invoking + * callbacks, this function will nevertheless say "idle". This is not + * ideal, but the overhead of checking all CPUs' callback lists is even + * less ideal, especially on large systems. Furthermore, the wakeup + * can happen before the callback is fully removed, so we have no choice + * but to accept this type of error. + * + * This function is also subject to counter-wrap errors, but let's face + * it, if this function was preempted for enough time for the counters + * to wrap, it really doesn't matter whether or not we expedite the grace + * period. The extra overhead of a needlessly expedited grace period is + * negligible when amoritized over that time period, and the extra latency + * of a needlessly non-expedited grace period is similarly negligible. + */ +static bool srcu_might_be_idle(struct srcu_struct *sp) +{ + unsigned long curseq; + unsigned long flags; + struct srcu_data *sdp; + unsigned long t; + + /* If the local srcu_data structure has callbacks, not idle. */ + local_irq_save(flags); + sdp = this_cpu_ptr(sp->sda); + if (rcu_segcblist_pend_cbs(&sdp->srcu_cblist)) { + local_irq_restore(flags); + return false; /* Callbacks already present, so not idle. */ + } + local_irq_restore(flags); + + /* + * No local callbacks, so probabalistically probe global state. + * Exact information would require acquiring locks, which would + * kill scalability, hence the probabalistic nature of the probe. + */ + + /* First, see if enough time has passed since the last GP. */ + t = ktime_get_mono_fast_ns(); + if (exp_holdoff == 0 || + time_in_range_open(t, sp->srcu_last_gp_end, + sp->srcu_last_gp_end + exp_holdoff)) + return false; /* Too soon after last GP. */ + + /* Next, check for probable idleness. */ + curseq = rcu_seq_current(&sp->srcu_gp_seq); + smp_mb(); /* Order ->srcu_gp_seq with ->srcu_gp_seq_needed. */ + if (ULONG_CMP_LT(curseq, READ_ONCE(sp->srcu_gp_seq_needed))) + return false; /* Grace period in progress, so not idle. */ + smp_mb(); /* Order ->srcu_gp_seq with prior access. */ + if (curseq != rcu_seq_current(&sp->srcu_gp_seq)) + return false; /* GP # changed, so not idle. */ + return true; /* With reasonable probability, idle! */ +} + +/* + * SRCU callback function to leak a callback. + */ +static void srcu_leak_callback(struct rcu_head *rhp) +{ +} + +/* + * Enqueue an SRCU callback on the srcu_data structure associated with + * the current CPU and the specified srcu_struct structure, initiating + * grace-period processing if it is not already running. + * + * Note that all CPUs must agree that the grace period extended beyond + * all pre-existing SRCU read-side critical section. On systems with + * more than one CPU, this means that when "func()" is invoked, each CPU + * is guaranteed to have executed a full memory barrier since the end of + * its last corresponding SRCU read-side critical section whose beginning + * preceded the call to call_rcu(). It also means that each CPU executing + * an SRCU read-side critical section that continues beyond the start of + * "func()" must have executed a memory barrier after the call_rcu() + * but before the beginning of that SRCU read-side critical section. + * Note that these guarantees include CPUs that are offline, idle, or + * executing in user mode, as well as CPUs that are executing in the kernel. + * + * Furthermore, if CPU A invoked call_rcu() and CPU B invoked the + * resulting SRCU callback function "func()", then both CPU A and CPU + * B are guaranteed to execute a full memory barrier during the time + * interval between the call to call_rcu() and the invocation of "func()". + * This guarantee applies even if CPU A and CPU B are the same CPU (but + * again only if the system has more than one CPU). + * + * Of course, these guarantees apply only for invocations of call_srcu(), + * srcu_read_lock(), and srcu_read_unlock() that are all passed the same + * srcu_struct structure. + */ +void __call_srcu(struct srcu_struct *sp, struct rcu_head *rhp, + rcu_callback_t func, bool do_norm) +{ + unsigned long flags; + bool needexp = false; + bool needgp = false; + unsigned long s; + struct srcu_data *sdp; + + check_init_srcu_struct(sp); + if (debug_rcu_head_queue(rhp)) { + /* Probable double call_srcu(), so leak the callback. */ + WRITE_ONCE(rhp->func, srcu_leak_callback); + WARN_ONCE(1, "call_srcu(): Leaked duplicate callback\n"); + return; + } + rhp->func = func; + local_irq_save(flags); + sdp = this_cpu_ptr(sp->sda); + raw_spin_lock_rcu_node(sdp); + rcu_segcblist_enqueue(&sdp->srcu_cblist, rhp, false); + rcu_segcblist_advance(&sdp->srcu_cblist, + rcu_seq_current(&sp->srcu_gp_seq)); + s = rcu_seq_snap(&sp->srcu_gp_seq); + (void)rcu_segcblist_accelerate(&sdp->srcu_cblist, s); + if (ULONG_CMP_LT(sdp->srcu_gp_seq_needed, s)) { + sdp->srcu_gp_seq_needed = s; + needgp = true; + } + if (!do_norm && ULONG_CMP_LT(sdp->srcu_gp_seq_needed_exp, s)) { + sdp->srcu_gp_seq_needed_exp = s; + needexp = true; + } + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); + if (needgp) + srcu_funnel_gp_start(sp, sdp, s, do_norm); + else if (needexp) + srcu_funnel_exp_start(sp, sdp->mynode, s); +} + +/** + * call_srcu() - Queue a callback for invocation after an SRCU grace period + * @sp: srcu_struct in queue the callback + * @head: structure to be used for queueing the SRCU callback. + * @func: function to be invoked after the SRCU grace period + * + * The callback function will be invoked some time after a full SRCU + * grace period elapses, in other words after all pre-existing SRCU + * read-side critical sections have completed. However, the callback + * function might well execute concurrently with other SRCU read-side + * critical sections that started after call_srcu() was invoked. SRCU + * read-side critical sections are delimited by srcu_read_lock() and + * srcu_read_unlock(), and may be nested. + * + * The callback will be invoked from process context, but must nevertheless + * be fast and must not block. + */ +void call_srcu(struct srcu_struct *sp, struct rcu_head *rhp, + rcu_callback_t func) +{ + __call_srcu(sp, rhp, func, true); +} +EXPORT_SYMBOL_GPL(call_srcu); + +/* + * Helper function for synchronize_srcu() and synchronize_srcu_expedited(). + */ +static void __synchronize_srcu(struct srcu_struct *sp, bool do_norm) +{ + struct rcu_synchronize rcu; + + RCU_LOCKDEP_WARN(lock_is_held(&sp->dep_map) || + lock_is_held(&rcu_bh_lock_map) || + lock_is_held(&rcu_lock_map) || + lock_is_held(&rcu_sched_lock_map), + "Illegal synchronize_srcu() in same-type SRCU (or in RCU) read-side critical section"); + + if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE) + return; + might_sleep(); + check_init_srcu_struct(sp); + init_completion(&rcu.completion); + init_rcu_head_on_stack(&rcu.head); + __call_srcu(sp, &rcu.head, wakeme_after_rcu, do_norm); + wait_for_completion(&rcu.completion); + destroy_rcu_head_on_stack(&rcu.head); +} + +/** + * synchronize_srcu_expedited - Brute-force SRCU grace period + * @sp: srcu_struct with which to synchronize. + * + * Wait for an SRCU grace period to elapse, but be more aggressive about + * spinning rather than blocking when waiting. + * + * Note that synchronize_srcu_expedited() has the same deadlock and + * memory-ordering properties as does synchronize_srcu(). + */ +void synchronize_srcu_expedited(struct srcu_struct *sp) +{ + __synchronize_srcu(sp, rcu_gp_is_normal()); +} +EXPORT_SYMBOL_GPL(synchronize_srcu_expedited); + +/** + * synchronize_srcu - wait for prior SRCU read-side critical-section completion + * @sp: srcu_struct with which to synchronize. + * + * Wait for the count to drain to zero of both indexes. To avoid the + * possible starvation of synchronize_srcu(), it waits for the count of + * the index=((->srcu_idx & 1) ^ 1) to drain to zero at first, + * and then flip the srcu_idx and wait for the count of the other index. + * + * Can block; must be called from process context. + * + * Note that it is illegal to call synchronize_srcu() from the corresponding + * SRCU read-side critical section; doing so will result in deadlock. + * However, it is perfectly legal to call synchronize_srcu() on one + * srcu_struct from some other srcu_struct's read-side critical section, + * as long as the resulting graph of srcu_structs is acyclic. + * + * There are memory-ordering constraints implied by synchronize_srcu(). + * On systems with more than one CPU, when synchronize_srcu() returns, + * each CPU is guaranteed to have executed a full memory barrier since + * the end of its last corresponding SRCU-sched read-side critical section + * whose beginning preceded the call to synchronize_srcu(). In addition, + * each CPU having an SRCU read-side critical section that extends beyond + * the return from synchronize_srcu() is guaranteed to have executed a + * full memory barrier after the beginning of synchronize_srcu() and before + * the beginning of that SRCU read-side critical section. Note that these + * guarantees include CPUs that are offline, idle, or executing in user mode, + * as well as CPUs that are executing in the kernel. + * + * Furthermore, if CPU A invoked synchronize_srcu(), which returned + * to its caller on CPU B, then both CPU A and CPU B are guaranteed + * to have executed a full memory barrier during the execution of + * synchronize_srcu(). This guarantee applies even if CPU A and CPU B + * are the same CPU, but again only if the system has more than one CPU. + * + * Of course, these memory-ordering guarantees apply only when + * synchronize_srcu(), srcu_read_lock(), and srcu_read_unlock() are + * passed the same srcu_struct structure. + * + * If SRCU is likely idle, expedite the first request. This semantic + * was provided by Classic SRCU, and is relied upon by its users, so TREE + * SRCU must also provide it. Note that detecting idleness is heuristic + * and subject to both false positives and negatives. + */ +void synchronize_srcu(struct srcu_struct *sp) +{ + if (srcu_might_be_idle(sp) || rcu_gp_is_expedited()) + synchronize_srcu_expedited(sp); + else + __synchronize_srcu(sp, true); +} +EXPORT_SYMBOL_GPL(synchronize_srcu); + +/* + * Callback function for srcu_barrier() use. + */ +static void srcu_barrier_cb(struct rcu_head *rhp) +{ + struct srcu_data *sdp; + struct srcu_struct *sp; + + sdp = container_of(rhp, struct srcu_data, srcu_barrier_head); + sp = sdp->sp; + if (atomic_dec_and_test(&sp->srcu_barrier_cpu_cnt)) + complete(&sp->srcu_barrier_completion); +} + +/** + * srcu_barrier - Wait until all in-flight call_srcu() callbacks complete. + * @sp: srcu_struct on which to wait for in-flight callbacks. + */ +void srcu_barrier(struct srcu_struct *sp) +{ + int cpu; + struct srcu_data *sdp; + unsigned long s = rcu_seq_snap(&sp->srcu_barrier_seq); + + check_init_srcu_struct(sp); + mutex_lock(&sp->srcu_barrier_mutex); + if (rcu_seq_done(&sp->srcu_barrier_seq, s)) { + smp_mb(); /* Force ordering following return. */ + mutex_unlock(&sp->srcu_barrier_mutex); + return; /* Someone else did our work for us. */ + } + rcu_seq_start(&sp->srcu_barrier_seq); + init_completion(&sp->srcu_barrier_completion); + + /* Initial count prevents reaching zero until all CBs are posted. */ + atomic_set(&sp->srcu_barrier_cpu_cnt, 1); + + /* + * Each pass through this loop enqueues a callback, but only + * on CPUs already having callbacks enqueued. Note that if + * a CPU already has callbacks enqueue, it must have already + * registered the need for a future grace period, so all we + * need do is enqueue a callback that will use the same + * grace period as the last callback already in the queue. + */ + for_each_possible_cpu(cpu) { + sdp = per_cpu_ptr(sp->sda, cpu); + raw_spin_lock_irq_rcu_node(sdp); + atomic_inc(&sp->srcu_barrier_cpu_cnt); + sdp->srcu_barrier_head.func = srcu_barrier_cb; + debug_rcu_head_queue(&sdp->srcu_barrier_head); + if (!rcu_segcblist_entrain(&sdp->srcu_cblist, + &sdp->srcu_barrier_head, 0)) { + debug_rcu_head_unqueue(&sdp->srcu_barrier_head); + atomic_dec(&sp->srcu_barrier_cpu_cnt); + } + raw_spin_unlock_irq_rcu_node(sdp); + } + + /* Remove the initial count, at which point reaching zero can happen. */ + if (atomic_dec_and_test(&sp->srcu_barrier_cpu_cnt)) + complete(&sp->srcu_barrier_completion); + wait_for_completion(&sp->srcu_barrier_completion); + + rcu_seq_end(&sp->srcu_barrier_seq); + mutex_unlock(&sp->srcu_barrier_mutex); +} +EXPORT_SYMBOL_GPL(srcu_barrier); + +/** + * srcu_batches_completed - return batches completed. + * @sp: srcu_struct on which to report batch completion. + * + * Report the number of batches, correlated with, but not necessarily + * precisely the same as, the number of grace periods that have elapsed. + */ +unsigned long srcu_batches_completed(struct srcu_struct *sp) +{ + return sp->srcu_idx; +} +EXPORT_SYMBOL_GPL(srcu_batches_completed); + +/* + * Core SRCU state machine. Push state bits of ->srcu_gp_seq + * to SRCU_STATE_SCAN2, and invoke srcu_gp_end() when scan has + * completed in that state. + */ +static void srcu_advance_state(struct srcu_struct *sp) +{ + int idx; + + mutex_lock(&sp->srcu_gp_mutex); + + /* + * Because readers might be delayed for an extended period after + * fetching ->srcu_idx for their index, at any point in time there + * might well be readers using both idx=0 and idx=1. We therefore + * need to wait for readers to clear from both index values before + * invoking a callback. + * + * The load-acquire ensures that we see the accesses performed + * by the prior grace period. + */ + idx = rcu_seq_state(smp_load_acquire(&sp->srcu_gp_seq)); /* ^^^ */ + if (idx == SRCU_STATE_IDLE) { + raw_spin_lock_irq_rcu_node(sp); + if (ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed)) { + WARN_ON_ONCE(rcu_seq_state(sp->srcu_gp_seq)); + raw_spin_unlock_irq_rcu_node(sp); + mutex_unlock(&sp->srcu_gp_mutex); + return; + } + idx = rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)); + if (idx == SRCU_STATE_IDLE) + srcu_gp_start(sp); + raw_spin_unlock_irq_rcu_node(sp); + if (idx != SRCU_STATE_IDLE) { + mutex_unlock(&sp->srcu_gp_mutex); + return; /* Someone else started the grace period. */ + } + } + + if (rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)) == SRCU_STATE_SCAN1) { + idx = 1 ^ (sp->srcu_idx & 1); + if (!try_check_zero(sp, idx, 1)) { + mutex_unlock(&sp->srcu_gp_mutex); + return; /* readers present, retry later. */ + } + srcu_flip(sp); + rcu_seq_set_state(&sp->srcu_gp_seq, SRCU_STATE_SCAN2); + } + + if (rcu_seq_state(READ_ONCE(sp->srcu_gp_seq)) == SRCU_STATE_SCAN2) { + + /* + * SRCU read-side critical sections are normally short, + * so check at least twice in quick succession after a flip. + */ + idx = 1 ^ (sp->srcu_idx & 1); + if (!try_check_zero(sp, idx, 2)) { + mutex_unlock(&sp->srcu_gp_mutex); + return; /* readers present, retry later. */ + } + srcu_gp_end(sp); /* Releases ->srcu_gp_mutex. */ + } +} + +/* + * Invoke a limited number of SRCU callbacks that have passed through + * their grace period. If there are more to do, SRCU will reschedule + * the workqueue. Note that needed memory barriers have been executed + * in this task's context by srcu_readers_active_idx_check(). + */ +static void srcu_invoke_callbacks(struct work_struct *work) +{ + bool more; + struct rcu_cblist ready_cbs; + struct rcu_head *rhp; + struct srcu_data *sdp; + struct srcu_struct *sp; + + sdp = container_of(work, struct srcu_data, work.work); + sp = sdp->sp; + rcu_cblist_init(&ready_cbs); + raw_spin_lock_irq_rcu_node(sdp); + rcu_segcblist_advance(&sdp->srcu_cblist, + rcu_seq_current(&sp->srcu_gp_seq)); + if (sdp->srcu_cblist_invoking || + !rcu_segcblist_ready_cbs(&sdp->srcu_cblist)) { + raw_spin_unlock_irq_rcu_node(sdp); + return; /* Someone else on the job or nothing to do. */ + } + + /* We are on the job! Extract and invoke ready callbacks. */ + sdp->srcu_cblist_invoking = true; + rcu_segcblist_extract_done_cbs(&sdp->srcu_cblist, &ready_cbs); + raw_spin_unlock_irq_rcu_node(sdp); + rhp = rcu_cblist_dequeue(&ready_cbs); + for (; rhp != NULL; rhp = rcu_cblist_dequeue(&ready_cbs)) { + debug_rcu_head_unqueue(rhp); + local_bh_disable(); + rhp->func(rhp); + local_bh_enable(); + } + + /* + * Update counts, accelerate new callbacks, and if needed, + * schedule another round of callback invocation. + */ + raw_spin_lock_irq_rcu_node(sdp); + rcu_segcblist_insert_count(&sdp->srcu_cblist, &ready_cbs); + (void)rcu_segcblist_accelerate(&sdp->srcu_cblist, + rcu_seq_snap(&sp->srcu_gp_seq)); + sdp->srcu_cblist_invoking = false; + more = rcu_segcblist_ready_cbs(&sdp->srcu_cblist); + raw_spin_unlock_irq_rcu_node(sdp); + if (more) + srcu_schedule_cbs_sdp(sdp, 0); +} + +/* + * Finished one round of SRCU grace period. Start another if there are + * more SRCU callbacks queued, otherwise put SRCU into not-running state. + */ +static void srcu_reschedule(struct srcu_struct *sp, unsigned long delay) +{ + bool pushgp = true; + + raw_spin_lock_irq_rcu_node(sp); + if (ULONG_CMP_GE(sp->srcu_gp_seq, sp->srcu_gp_seq_needed)) { + if (!WARN_ON_ONCE(rcu_seq_state(sp->srcu_gp_seq))) { + /* All requests fulfilled, time to go idle. */ + pushgp = false; + } + } else if (!rcu_seq_state(sp->srcu_gp_seq)) { + /* Outstanding request and no GP. Start one. */ + srcu_gp_start(sp); + } + raw_spin_unlock_irq_rcu_node(sp); + + if (pushgp) + queue_delayed_work(system_power_efficient_wq, &sp->work, delay); +} + +/* + * This is the work-queue function that handles SRCU grace periods. + */ +void process_srcu(struct work_struct *work) +{ + struct srcu_struct *sp; + + sp = container_of(work, struct srcu_struct, work.work); + + srcu_advance_state(sp); + srcu_reschedule(sp, srcu_get_delay(sp)); +} +EXPORT_SYMBOL_GPL(process_srcu); + +void srcutorture_get_gp_data(enum rcutorture_type test_type, + struct srcu_struct *sp, int *flags, + unsigned long *gpnum, unsigned long *completed) +{ + if (test_type != SRCU_FLAVOR) + return; + *flags = 0; + *completed = rcu_seq_ctr(sp->srcu_gp_seq); + *gpnum = rcu_seq_ctr(sp->srcu_gp_seq_needed); +} +EXPORT_SYMBOL_GPL(srcutorture_get_gp_data); + +static int __init srcu_bootup_announce(void) +{ + pr_info("Hierarchical SRCU implementation.\n"); + if (exp_holdoff != DEFAULT_SRCU_EXP_HOLDOFF) + pr_info("\tNon-default auto-expedite holdoff of %lu ns.\n", exp_holdoff); + return 0; +} +early_initcall(srcu_bootup_announce); diff --git a/kernel/rcu/tiny.c b/kernel/rcu/tiny.c index b23a4d076f3d..f8488965250f 100644 --- a/kernel/rcu/tiny.c +++ b/kernel/rcu/tiny.c @@ -25,7 +25,7 @@ #include <linux/completion.h> #include <linux/interrupt.h> #include <linux/notifier.h> -#include <linux/rcupdate.h> +#include <linux/rcupdate_wait.h> #include <linux/kernel.h> #include <linux/export.h> #include <linux/mutex.h> @@ -35,32 +35,40 @@ #include <linux/time.h> #include <linux/cpu.h> #include <linux/prefetch.h> -#include <linux/trace_events.h> #include "rcu.h" -/* Forward declarations for tiny_plugin.h. */ -struct rcu_ctrlblk; -static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp); -static void rcu_process_callbacks(struct softirq_action *unused); -static void __call_rcu(struct rcu_head *head, - rcu_callback_t func, - struct rcu_ctrlblk *rcp); +/* Global control variables for rcupdate callback mechanism. */ +struct rcu_ctrlblk { + struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */ + struct rcu_head **donetail; /* ->next pointer of last "done" CB. */ + struct rcu_head **curtail; /* ->next pointer of last CB. */ +}; -#include "tiny_plugin.h" +/* Definition for rcupdate control block. */ +static struct rcu_ctrlblk rcu_sched_ctrlblk = { + .donetail = &rcu_sched_ctrlblk.rcucblist, + .curtail = &rcu_sched_ctrlblk.rcucblist, +}; -#if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) +static struct rcu_ctrlblk rcu_bh_ctrlblk = { + .donetail = &rcu_bh_ctrlblk.rcucblist, + .curtail = &rcu_bh_ctrlblk.rcucblist, +}; -/* - * Test whether RCU thinks that the current CPU is idle. - */ -bool notrace __rcu_is_watching(void) +#include "tiny_plugin.h" + +void rcu_barrier_bh(void) { - return true; + wait_rcu_gp(call_rcu_bh); } -EXPORT_SYMBOL(__rcu_is_watching); +EXPORT_SYMBOL(rcu_barrier_bh); -#endif /* defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) */ +void rcu_barrier_sched(void) +{ + wait_rcu_gp(call_rcu_sched); +} +EXPORT_SYMBOL(rcu_barrier_sched); /* * Helper function for rcu_sched_qs() and rcu_bh_qs(). @@ -69,7 +77,6 @@ EXPORT_SYMBOL(__rcu_is_watching); */ static int rcu_qsctr_help(struct rcu_ctrlblk *rcp) { - RCU_TRACE(reset_cpu_stall_ticks(rcp)); if (rcp->donetail != rcp->curtail) { rcp->donetail = rcp->curtail; return 1; @@ -115,7 +122,6 @@ void rcu_bh_qs(void) */ void rcu_check_callbacks(int user) { - RCU_TRACE(check_cpu_stalls()); if (user) rcu_sched_qs(); else if (!in_softirq()) @@ -130,10 +136,8 @@ void rcu_check_callbacks(int user) */ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) { - const char *rn = NULL; struct rcu_head *next, *list; unsigned long flags; - RCU_TRACE(int cb_count = 0); /* Move the ready-to-invoke callbacks to a local list. */ local_irq_save(flags); @@ -142,7 +146,6 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) local_irq_restore(flags); return; } - RCU_TRACE(trace_rcu_batch_start(rcp->name, 0, rcp->qlen, -1)); list = rcp->rcucblist; rcp->rcucblist = *rcp->donetail; *rcp->donetail = NULL; @@ -152,22 +155,15 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) local_irq_restore(flags); /* Invoke the callbacks on the local list. */ - RCU_TRACE(rn = rcp->name); while (list) { next = list->next; prefetch(next); debug_rcu_head_unqueue(list); local_bh_disable(); - __rcu_reclaim(rn, list); + __rcu_reclaim("", list); local_bh_enable(); list = next; - RCU_TRACE(cb_count++); } - RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count)); - RCU_TRACE(trace_rcu_batch_end(rcp->name, - cb_count, 0, need_resched(), - is_idle_task(current), - false)); } static __latent_entropy void rcu_process_callbacks(struct softirq_action *unused) @@ -211,7 +207,6 @@ static void __call_rcu(struct rcu_head *head, local_irq_save(flags); *rcp->curtail = head; rcp->curtail = &head->next; - RCU_TRACE(rcp->qlen++); local_irq_restore(flags); if (unlikely(is_idle_task(current))) { @@ -244,8 +239,5 @@ EXPORT_SYMBOL_GPL(call_rcu_bh); void __init rcu_init(void) { open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); - RCU_TRACE(reset_cpu_stall_ticks(&rcu_sched_ctrlblk)); - RCU_TRACE(reset_cpu_stall_ticks(&rcu_bh_ctrlblk)); - rcu_early_boot_tests(); } diff --git a/kernel/rcu/tiny_plugin.h b/kernel/rcu/tiny_plugin.h index c64b827ecbca..f0a01b2a3062 100644 --- a/kernel/rcu/tiny_plugin.h +++ b/kernel/rcu/tiny_plugin.h @@ -22,37 +22,7 @@ * Author: Paul E. McKenney <[email protected]> */ -#include <linux/kthread.h> -#include <linux/init.h> -#include <linux/debugfs.h> -#include <linux/seq_file.h> - -/* Global control variables for rcupdate callback mechanism. */ -struct rcu_ctrlblk { - struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */ - struct rcu_head **donetail; /* ->next pointer of last "done" CB. */ - struct rcu_head **curtail; /* ->next pointer of last CB. */ - RCU_TRACE(long qlen); /* Number of pending CBs. */ - RCU_TRACE(unsigned long gp_start); /* Start time for stalls. */ - RCU_TRACE(unsigned long ticks_this_gp); /* Statistic for stalls. */ - RCU_TRACE(unsigned long jiffies_stall); /* Jiffies at next stall. */ - RCU_TRACE(const char *name); /* Name of RCU type. */ -}; - -/* Definition for rcupdate control block. */ -static struct rcu_ctrlblk rcu_sched_ctrlblk = { - .donetail = &rcu_sched_ctrlblk.rcucblist, - .curtail = &rcu_sched_ctrlblk.rcucblist, - RCU_TRACE(.name = "rcu_sched") -}; - -static struct rcu_ctrlblk rcu_bh_ctrlblk = { - .donetail = &rcu_bh_ctrlblk.rcucblist, - .curtail = &rcu_bh_ctrlblk.rcucblist, - RCU_TRACE(.name = "rcu_bh") -}; - -#ifdef CONFIG_DEBUG_LOCK_ALLOC +#if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU) #include <linux/kernel_stat.h> int rcu_scheduler_active __read_mostly; @@ -65,105 +35,13 @@ EXPORT_SYMBOL_GPL(rcu_scheduler_active); * to RCU_SCHEDULER_RUNNING, skipping the RCU_SCHEDULER_INIT stage. * The reason for this is that Tiny RCU does not need kthreads, so does * not have to care about the fact that the scheduler is half-initialized - * at a certain phase of the boot process. + * at a certain phase of the boot process. Unless SRCU is in the mix. */ void __init rcu_scheduler_starting(void) { WARN_ON(nr_context_switches() > 0); - rcu_scheduler_active = RCU_SCHEDULER_RUNNING; -} - -#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ - -#ifdef CONFIG_RCU_TRACE - -static void rcu_trace_sub_qlen(struct rcu_ctrlblk *rcp, int n) -{ - unsigned long flags; - - local_irq_save(flags); - rcp->qlen -= n; - local_irq_restore(flags); -} - -/* - * Dump statistics for TINY_RCU, such as they are. - */ -static int show_tiny_stats(struct seq_file *m, void *unused) -{ - seq_printf(m, "rcu_sched: qlen: %ld\n", rcu_sched_ctrlblk.qlen); - seq_printf(m, "rcu_bh: qlen: %ld\n", rcu_bh_ctrlblk.qlen); - return 0; -} - -static int show_tiny_stats_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_tiny_stats, NULL); -} - -static const struct file_operations show_tiny_stats_fops = { - .owner = THIS_MODULE, - .open = show_tiny_stats_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static struct dentry *rcudir; - -static int __init rcutiny_trace_init(void) -{ - struct dentry *retval; - - rcudir = debugfs_create_dir("rcu", NULL); - if (!rcudir) - goto free_out; - retval = debugfs_create_file("rcudata", 0444, rcudir, - NULL, &show_tiny_stats_fops); - if (!retval) - goto free_out; - return 0; -free_out: - debugfs_remove_recursive(rcudir); - return 1; -} -device_initcall(rcutiny_trace_init); - -static void check_cpu_stall(struct rcu_ctrlblk *rcp) -{ - unsigned long j; - unsigned long js; - - if (rcu_cpu_stall_suppress) - return; - rcp->ticks_this_gp++; - j = jiffies; - js = READ_ONCE(rcp->jiffies_stall); - if (rcp->rcucblist && ULONG_CMP_GE(j, js)) { - pr_err("INFO: %s stall on CPU (%lu ticks this GP) idle=%llx (t=%lu jiffies q=%ld)\n", - rcp->name, rcp->ticks_this_gp, DYNTICK_TASK_EXIT_IDLE, - jiffies - rcp->gp_start, rcp->qlen); - dump_stack(); - WRITE_ONCE(rcp->jiffies_stall, - jiffies + 3 * rcu_jiffies_till_stall_check() + 3); - } else if (ULONG_CMP_GE(j, js)) { - WRITE_ONCE(rcp->jiffies_stall, - jiffies + rcu_jiffies_till_stall_check()); - } -} - -static void reset_cpu_stall_ticks(struct rcu_ctrlblk *rcp) -{ - rcp->ticks_this_gp = 0; - rcp->gp_start = jiffies; - WRITE_ONCE(rcp->jiffies_stall, - jiffies + rcu_jiffies_till_stall_check()); -} - -static void check_cpu_stalls(void) -{ - RCU_TRACE(check_cpu_stall(&rcu_bh_ctrlblk)); - RCU_TRACE(check_cpu_stall(&rcu_sched_ctrlblk)); + rcu_scheduler_active = IS_ENABLED(CONFIG_SRCU) + ? RCU_SCHEDULER_INIT : RCU_SCHEDULER_RUNNING; } -#endif /* #ifdef CONFIG_RCU_TRACE */ +#endif /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SRCU) */ diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index cb4e2056ccf3..51d4c3acf32d 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -32,9 +32,10 @@ #include <linux/init.h> #include <linux/spinlock.h> #include <linux/smp.h> -#include <linux/rcupdate.h> +#include <linux/rcupdate_wait.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <linux/sched/debug.h> #include <linux/nmi.h> #include <linux/atomic.h> #include <linux/bitops.h> @@ -49,12 +50,14 @@ #include <linux/kernel_stat.h> #include <linux/wait.h> #include <linux/kthread.h> +#include <uapi/linux/sched/types.h> #include <linux/prefetch.h> #include <linux/delay.h> #include <linux/stop_machine.h> #include <linux/random.h> #include <linux/trace_events.h> #include <linux/suspend.h> +#include <linux/ftrace.h> #include "tree.h" #include "rcu.h" @@ -95,8 +98,8 @@ struct rcu_state sname##_state = { \ .gpnum = 0UL - 300UL, \ .completed = 0UL - 300UL, \ .orphan_lock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.orphan_lock), \ - .orphan_nxttail = &sname##_state.orphan_nxtlist, \ - .orphan_donetail = &sname##_state.orphan_donelist, \ + .orphan_pend = RCU_CBLIST_INITIALIZER(sname##_state.orphan_pend), \ + .orphan_done = RCU_CBLIST_INITIALIZER(sname##_state.orphan_done), \ .barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \ .name = RCU_STATE_NAME(sname), \ .abbr = sabbr, \ @@ -121,7 +124,7 @@ static int rcu_fanout_leaf = RCU_FANOUT_LEAF; module_param(rcu_fanout_leaf, int, 0444); int rcu_num_lvls __read_mostly = RCU_NUM_LVLS; /* Number of rcu_nodes at specified level. */ -static int num_rcu_lvl[] = NUM_RCU_LVL_INIT; +int num_rcu_lvl[] = NUM_RCU_LVL_INIT; int rcu_num_nodes __read_mostly = NUM_RCU_NODES; /* Total # rcu_nodes in use. */ /* panic() on RCU Stall sysctl. */ int sysctl_panic_on_rcu_stall __read_mostly; @@ -165,39 +168,21 @@ static void rcu_report_exp_rdp(struct rcu_state *rsp, static void sync_sched_exp_online_cleanup(int cpu); /* rcuc/rcub kthread realtime priority */ -#ifdef CONFIG_RCU_KTHREAD_PRIO -static int kthread_prio = CONFIG_RCU_KTHREAD_PRIO; -#else /* #ifdef CONFIG_RCU_KTHREAD_PRIO */ static int kthread_prio = IS_ENABLED(CONFIG_RCU_BOOST) ? 1 : 0; -#endif /* #else #ifdef CONFIG_RCU_KTHREAD_PRIO */ module_param(kthread_prio, int, 0644); /* Delay in jiffies for grace-period initialization delays, debug only. */ -#ifdef CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT -static int gp_preinit_delay = CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT_DELAY; -module_param(gp_preinit_delay, int, 0644); -#else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT */ -static const int gp_preinit_delay; -#endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_PREINIT */ - -#ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT -static int gp_init_delay = CONFIG_RCU_TORTURE_TEST_SLOW_INIT_DELAY; -module_param(gp_init_delay, int, 0644); -#else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT */ -static const int gp_init_delay; -#endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_INIT */ - -#ifdef CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP -static int gp_cleanup_delay = CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP_DELAY; -module_param(gp_cleanup_delay, int, 0644); -#else /* #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP */ -static const int gp_cleanup_delay; -#endif /* #else #ifdef CONFIG_RCU_TORTURE_TEST_SLOW_CLEANUP */ +static int gp_preinit_delay; +module_param(gp_preinit_delay, int, 0444); +static int gp_init_delay; +module_param(gp_init_delay, int, 0444); +static int gp_cleanup_delay; +module_param(gp_cleanup_delay, int, 0444); /* * Number of grace periods between delays, normalized by the duration of - * the delay. The longer the the delay, the more the grace periods between + * the delay. The longer the delay, the more the grace periods between * each delay. The reason for this normalization is that it means that, * for non-zero delays, the overall slowdown of grace periods is constant * regardless of the duration of the delay. This arrangement balances @@ -247,6 +232,7 @@ static int rcu_gp_in_progress(struct rcu_state *rsp) */ void rcu_sched_qs(void) { + RCU_LOCKDEP_WARN(preemptible(), "rcu_sched_qs() invoked with preemption enabled!!!"); if (!__this_cpu_read(rcu_sched_data.cpu_no_qs.s)) return; trace_rcu_grace_period(TPS("rcu_sched"), @@ -262,6 +248,7 @@ void rcu_sched_qs(void) void rcu_bh_qs(void) { + RCU_LOCKDEP_WARN(preemptible(), "rcu_bh_qs() invoked with preemption enabled!!!"); if (__this_cpu_read(rcu_bh_data.cpu_no_qs.s)) { trace_rcu_grace_period(TPS("rcu_bh"), __this_cpu_read(rcu_bh_data.gpnum), @@ -270,19 +257,179 @@ void rcu_bh_qs(void) } } -static DEFINE_PER_CPU(int, rcu_sched_qs_mask); +/* + * Steal a bit from the bottom of ->dynticks for idle entry/exit + * control. Initially this is for TLB flushing. + */ +#define RCU_DYNTICK_CTRL_MASK 0x1 +#define RCU_DYNTICK_CTRL_CTR (RCU_DYNTICK_CTRL_MASK + 1) +#ifndef rcu_eqs_special_exit +#define rcu_eqs_special_exit() do { } while (0) +#endif static DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { .dynticks_nesting = DYNTICK_TASK_EXIT_IDLE, - .dynticks = ATOMIC_INIT(1), -#ifdef CONFIG_NO_HZ_FULL_SYSIDLE - .dynticks_idle_nesting = DYNTICK_TASK_NEST_VALUE, - .dynticks_idle = ATOMIC_INIT(1), -#endif /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ + .dynticks = ATOMIC_INIT(RCU_DYNTICK_CTRL_CTR), }; -DEFINE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr); -EXPORT_PER_CPU_SYMBOL_GPL(rcu_qs_ctr); +/* + * There's a few places, currently just in the tracing infrastructure, + * that uses rcu_irq_enter() to make sure RCU is watching. But there's + * a small location where that will not even work. In those cases + * rcu_irq_enter_disabled() needs to be checked to make sure rcu_irq_enter() + * can be called. + */ +static DEFINE_PER_CPU(bool, disable_rcu_irq_enter); + +bool rcu_irq_enter_disabled(void) +{ + return this_cpu_read(disable_rcu_irq_enter); +} + +/* + * Record entry into an extended quiescent state. This is only to be + * called when not already in an extended quiescent state. + */ +static void rcu_dynticks_eqs_enter(void) +{ + struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + int seq; + + /* + * CPUs seeing atomic_add_return() must see prior RCU read-side + * critical sections, and we also must force ordering with the + * next idle sojourn. + */ + seq = atomic_add_return(RCU_DYNTICK_CTRL_CTR, &rdtp->dynticks); + /* Better be in an extended quiescent state! */ + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + (seq & RCU_DYNTICK_CTRL_CTR)); + /* Better not have special action (TLB flush) pending! */ + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + (seq & RCU_DYNTICK_CTRL_MASK)); +} + +/* + * Record exit from an extended quiescent state. This is only to be + * called from an extended quiescent state. + */ +static void rcu_dynticks_eqs_exit(void) +{ + struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + int seq; + + /* + * CPUs seeing atomic_add_return() must see prior idle sojourns, + * and we also must force ordering with the next RCU read-side + * critical section. + */ + seq = atomic_add_return(RCU_DYNTICK_CTRL_CTR, &rdtp->dynticks); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && + !(seq & RCU_DYNTICK_CTRL_CTR)); + if (seq & RCU_DYNTICK_CTRL_MASK) { + atomic_andnot(RCU_DYNTICK_CTRL_MASK, &rdtp->dynticks); + smp_mb__after_atomic(); /* _exit after clearing mask. */ + /* Prefer duplicate flushes to losing a flush. */ + rcu_eqs_special_exit(); + } +} + +/* + * Reset the current CPU's ->dynticks counter to indicate that the + * newly onlined CPU is no longer in an extended quiescent state. + * This will either leave the counter unchanged, or increment it + * to the next non-quiescent value. + * + * The non-atomic test/increment sequence works because the upper bits + * of the ->dynticks counter are manipulated only by the corresponding CPU, + * or when the corresponding CPU is offline. + */ +static void rcu_dynticks_eqs_online(void) +{ + struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + + if (atomic_read(&rdtp->dynticks) & RCU_DYNTICK_CTRL_CTR) + return; + atomic_add(RCU_DYNTICK_CTRL_CTR, &rdtp->dynticks); +} + +/* + * Is the current CPU in an extended quiescent state? + * + * No ordering, as we are sampling CPU-local information. + */ +bool rcu_dynticks_curr_cpu_in_eqs(void) +{ + struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + + return !(atomic_read(&rdtp->dynticks) & RCU_DYNTICK_CTRL_CTR); +} + +/* + * Snapshot the ->dynticks counter with full ordering so as to allow + * stable comparison of this counter with past and future snapshots. + */ +int rcu_dynticks_snap(struct rcu_dynticks *rdtp) +{ + int snap = atomic_add_return(0, &rdtp->dynticks); + + return snap & ~RCU_DYNTICK_CTRL_MASK; +} + +/* + * Return true if the snapshot returned from rcu_dynticks_snap() + * indicates that RCU is in an extended quiescent state. + */ +static bool rcu_dynticks_in_eqs(int snap) +{ + return !(snap & RCU_DYNTICK_CTRL_CTR); +} + +/* + * Return true if the CPU corresponding to the specified rcu_dynticks + * structure has spent some time in an extended quiescent state since + * rcu_dynticks_snap() returned the specified snapshot. + */ +static bool rcu_dynticks_in_eqs_since(struct rcu_dynticks *rdtp, int snap) +{ + return snap != rcu_dynticks_snap(rdtp); +} + +/* + * Do a double-increment of the ->dynticks counter to emulate a + * momentary idle-CPU quiescent state. + */ +static void rcu_dynticks_momentary_idle(void) +{ + struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + int special = atomic_add_return(2 * RCU_DYNTICK_CTRL_CTR, + &rdtp->dynticks); + + /* It is illegal to call this from idle state. */ + WARN_ON_ONCE(!(special & RCU_DYNTICK_CTRL_CTR)); +} + +/* + * Set the special (bottom) bit of the specified CPU so that it + * will take special action (such as flushing its TLB) on the + * next exit from an extended quiescent state. Returns true if + * the bit was successfully set, or false if the CPU was not in + * an extended quiescent state. + */ +bool rcu_eqs_special_set(int cpu) +{ + int old; + int new; + struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu); + + do { + old = atomic_read(&rdtp->dynticks); + if (old & RCU_DYNTICK_CTRL_CTR) + return false; + new = old | RCU_DYNTICK_CTRL_MASK; + } while (atomic_cmpxchg(&rdtp->dynticks, old, new) != old); + return true; +} /* * Let the RCU core know that this CPU has gone through the scheduler, @@ -291,48 +438,14 @@ EXPORT_PER_CPU_SYMBOL_GPL(rcu_qs_ctr); * memory barriers to let the RCU core know about it, regardless of what * this CPU might (or might not) do in the near future. * - * We inform the RCU core by emulating a zero-duration dyntick-idle - * period, which we in turn do by incrementing the ->dynticks counter - * by two. + * We inform the RCU core by emulating a zero-duration dyntick-idle period. * * The caller must have disabled interrupts. */ static void rcu_momentary_dyntick_idle(void) { - struct rcu_data *rdp; - struct rcu_dynticks *rdtp; - int resched_mask; - struct rcu_state *rsp; - - /* - * Yes, we can lose flag-setting operations. This is OK, because - * the flag will be set again after some delay. - */ - resched_mask = raw_cpu_read(rcu_sched_qs_mask); - raw_cpu_write(rcu_sched_qs_mask, 0); - - /* Find the flavor that needs a quiescent state. */ - for_each_rcu_flavor(rsp) { - rdp = raw_cpu_ptr(rsp->rda); - if (!(resched_mask & rsp->flavor_mask)) - continue; - smp_mb(); /* rcu_sched_qs_mask before cond_resched_completed. */ - if (READ_ONCE(rdp->mynode->completed) != - READ_ONCE(rdp->cond_resched_completed)) - continue; - - /* - * Pretend to be momentarily idle for the quiescent state. - * This allows the grace-period kthread to record the - * quiescent state, with no need for this CPU to do anything - * further. - */ - rdtp = this_cpu_ptr(&rcu_dynticks); - smp_mb__before_atomic(); /* Earlier stuff before QS. */ - atomic_add(2, &rdtp->dynticks); /* QS. */ - smp_mb__after_atomic(); /* Later stuff after QS. */ - break; - } + raw_cpu_write(rcu_dynticks.rcu_need_heavy_qs, false); + rcu_dynticks_momentary_idle(); } /* @@ -340,14 +453,22 @@ static void rcu_momentary_dyntick_idle(void) * and requires special handling for preemptible RCU. * The caller must have disabled interrupts. */ -void rcu_note_context_switch(void) +void rcu_note_context_switch(bool preempt) { barrier(); /* Avoid RCU read-side critical sections leaking down. */ trace_rcu_utilization(TPS("Start context switch")); rcu_sched_qs(); - rcu_preempt_note_context_switch(); - if (unlikely(raw_cpu_read(rcu_sched_qs_mask))) + rcu_preempt_note_context_switch(preempt); + /* Load rcu_urgent_qs before other flags. */ + if (!smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) + goto out; + this_cpu_write(rcu_dynticks.rcu_urgent_qs, false); + if (unlikely(raw_cpu_read(rcu_dynticks.rcu_need_heavy_qs))) rcu_momentary_dyntick_idle(); + this_cpu_inc(rcu_dynticks.rcu_qs_ctr); + if (!preempt) + rcu_note_voluntary_context_switch_lite(current); +out: trace_rcu_utilization(TPS("End context switch")); barrier(); /* Avoid RCU read-side critical sections leaking up. */ } @@ -370,35 +491,35 @@ void rcu_all_qs(void) { unsigned long flags; + if (!raw_cpu_read(rcu_dynticks.rcu_urgent_qs)) + return; + preempt_disable(); + /* Load rcu_urgent_qs before other flags. */ + if (!smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) { + preempt_enable(); + return; + } + this_cpu_write(rcu_dynticks.rcu_urgent_qs, false); barrier(); /* Avoid RCU read-side critical sections leaking down. */ - if (unlikely(raw_cpu_read(rcu_sched_qs_mask))) { + if (unlikely(raw_cpu_read(rcu_dynticks.rcu_need_heavy_qs))) { local_irq_save(flags); rcu_momentary_dyntick_idle(); local_irq_restore(flags); } - if (unlikely(raw_cpu_read(rcu_sched_data.cpu_no_qs.b.exp))) { - /* - * Yes, we just checked a per-CPU variable with preemption - * enabled, so we might be migrated to some other CPU at - * this point. That is OK because in that case, the - * migration will supply the needed quiescent state. - * We might end up needlessly disabling preemption and - * invoking rcu_sched_qs() on the destination CPU, but - * the probability and cost are both quite low, so this - * should not be a problem in practice. - */ - preempt_disable(); + if (unlikely(raw_cpu_read(rcu_sched_data.cpu_no_qs.b.exp))) rcu_sched_qs(); - preempt_enable(); - } - this_cpu_inc(rcu_qs_ctr); + this_cpu_inc(rcu_dynticks.rcu_qs_ctr); barrier(); /* Avoid RCU read-side critical sections leaking up. */ + preempt_enable(); } EXPORT_SYMBOL_GPL(rcu_all_qs); -static long blimit = 10; /* Maximum callbacks per rcu_do_batch. */ -static long qhimark = 10000; /* If this many pending, ignore blimit. */ -static long qlowmark = 100; /* Once only this many pending, use blimit. */ +#define DEFAULT_RCU_BLIMIT 10 /* Maximum callbacks per rcu_do_batch. */ +static long blimit = DEFAULT_RCU_BLIMIT; +#define DEFAULT_RCU_QHIMARK 10000 /* If this many pending, ignore blimit. */ +static long qhimark = DEFAULT_RCU_QHIMARK; +#define DEFAULT_RCU_QLOMARK 100 /* Once only this many pending, use blimit. */ +static long qlowmark = DEFAULT_RCU_QLOMARK; module_param(blimit, long, 0444); module_param(qhimark, long, 0444); @@ -421,10 +542,7 @@ module_param(jiffies_till_sched_qs, ulong, 0644); static bool rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp); -static void force_qs_rnp(struct rcu_state *rsp, - int (*f)(struct rcu_data *rsp, bool *isidle, - unsigned long *maxj), - bool *isidle, unsigned long *maxj); +static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *rsp)); static void force_quiescent_state(struct rcu_state *rsp); static int rcu_pending(void); @@ -581,15 +699,11 @@ void rcutorture_get_gp_data(enum rcutorture_type test_type, int *flags, default: break; } - if (rsp != NULL) { - *flags = READ_ONCE(rsp->gp_flags); - *gpnum = READ_ONCE(rsp->gpnum); - *completed = READ_ONCE(rsp->completed); + if (rsp == NULL) return; - } - *flags = 0; - *gpnum = 0; - *completed = 0; + *flags = READ_ONCE(rsp->gp_flags); + *gpnum = READ_ONCE(rsp->gpnum); + *completed = READ_ONCE(rsp->completed); } EXPORT_SYMBOL_GPL(rcutorture_get_gp_data); @@ -605,16 +719,6 @@ void rcutorture_record_progress(unsigned long vernum) EXPORT_SYMBOL_GPL(rcutorture_record_progress); /* - * Does the CPU have callbacks ready to be invoked? - */ -static int -cpu_has_callbacks_ready_to_invoke(struct rcu_data *rdp) -{ - return &rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL] && - rdp->nxttail[RCU_DONE_TAIL] != NULL; -} - -/* * Return the root node of the specified rcu_state structure. */ static struct rcu_node *rcu_get_root(struct rcu_state *rsp) @@ -633,6 +737,7 @@ static int rcu_future_needs_gp(struct rcu_state *rsp) int idx = (READ_ONCE(rnp->completed) + 1) & 0x1; int *fp = &rnp->need_future_gp[idx]; + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_future_needs_gp() invoked with irqs enabled!!!"); return READ_ONCE(*fp); } @@ -644,44 +749,41 @@ static int rcu_future_needs_gp(struct rcu_state *rsp) static bool cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp) { - int i; - + RCU_LOCKDEP_WARN(!irqs_disabled(), "cpu_needs_another_gp() invoked with irqs enabled!!!"); if (rcu_gp_in_progress(rsp)) return false; /* No, a grace period is already in progress. */ if (rcu_future_needs_gp(rsp)) return true; /* Yes, a no-CBs CPU needs one. */ - if (!rdp->nxttail[RCU_NEXT_TAIL]) + if (!rcu_segcblist_is_enabled(&rdp->cblist)) return false; /* No, this is a no-CBs (or offline) CPU. */ - if (*rdp->nxttail[RCU_NEXT_READY_TAIL]) + if (!rcu_segcblist_restempty(&rdp->cblist, RCU_NEXT_READY_TAIL)) return true; /* Yes, CPU has newly registered callbacks. */ - for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) - if (rdp->nxttail[i - 1] != rdp->nxttail[i] && - ULONG_CMP_LT(READ_ONCE(rsp->completed), - rdp->nxtcompleted[i])) - return true; /* Yes, CBs for future grace period. */ + if (rcu_segcblist_future_gp_needed(&rdp->cblist, + READ_ONCE(rsp->completed))) + return true; /* Yes, CBs for future grace period. */ return false; /* No grace period needed. */ } /* - * rcu_eqs_enter_common - current CPU is moving towards extended quiescent state + * rcu_eqs_enter_common - current CPU is entering an extended quiescent state * - * If the new value of the ->dynticks_nesting counter now is zero, - * we really have entered idle, and must do the appropriate accounting. - * The caller must have disabled interrupts. + * Enter idle, doing appropriate accounting. The caller must have + * disabled interrupts. */ -static void rcu_eqs_enter_common(long long oldval, bool user) +static void rcu_eqs_enter_common(bool user) { struct rcu_state *rsp; struct rcu_data *rdp; struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); - trace_rcu_dyntick(TPS("Start"), oldval, rdtp->dynticks_nesting); + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_eqs_enter_common() invoked with irqs enabled!!!"); + trace_rcu_dyntick(TPS("Start"), rdtp->dynticks_nesting, 0); if (IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)) { struct task_struct *idle __maybe_unused = idle_task(smp_processor_id()); - trace_rcu_dyntick(TPS("Error on entry: not idle task"), oldval, 0); + trace_rcu_dyntick(TPS("Error on entry: not idle task"), rdtp->dynticks_nesting, 0); rcu_ftrace_dump(DUMP_ORIG); WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s", current->pid, current->comm, @@ -692,12 +794,10 @@ static void rcu_eqs_enter_common(long long oldval, bool user) do_nocb_deferred_wakeup(rdp); } rcu_prepare_for_idle(); - /* CPUs seeing atomic_inc() must see prior RCU read-side crit sects */ - smp_mb__before_atomic(); /* See above. */ - atomic_inc(&rdtp->dynticks); - smp_mb__after_atomic(); /* Force ordering with next sojourn. */ - WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && - atomic_read(&rdtp->dynticks) & 0x1); + __this_cpu_inc(disable_rcu_irq_enter); + rdtp->dynticks_nesting = 0; /* Breaks tracing momentarily. */ + rcu_dynticks_eqs_enter(); /* After this, tracing works again. */ + __this_cpu_dec(disable_rcu_irq_enter); rcu_dynticks_task_enter(); /* @@ -718,19 +818,15 @@ static void rcu_eqs_enter_common(long long oldval, bool user) */ static void rcu_eqs_enter(bool user) { - long long oldval; struct rcu_dynticks *rdtp; rdtp = this_cpu_ptr(&rcu_dynticks); - oldval = rdtp->dynticks_nesting; WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && - (oldval & DYNTICK_TASK_NEST_MASK) == 0); - if ((oldval & DYNTICK_TASK_NEST_MASK) == DYNTICK_TASK_NEST_VALUE) { - rdtp->dynticks_nesting = 0; - rcu_eqs_enter_common(oldval, user); - } else { + (rdtp->dynticks_nesting & DYNTICK_TASK_NEST_MASK) == 0); + if ((rdtp->dynticks_nesting & DYNTICK_TASK_NEST_MASK) == DYNTICK_TASK_NEST_VALUE) + rcu_eqs_enter_common(user); + else rdtp->dynticks_nesting -= DYNTICK_TASK_NEST_VALUE; - } } /** @@ -751,7 +847,6 @@ void rcu_idle_enter(void) local_irq_save(flags); rcu_eqs_enter(false); - rcu_sysidle_enter(0); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(rcu_idle_enter); @@ -789,20 +884,18 @@ void rcu_user_enter(void) */ void rcu_irq_exit(void) { - long long oldval; struct rcu_dynticks *rdtp; RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_exit() invoked with irqs enabled!!!"); rdtp = this_cpu_ptr(&rcu_dynticks); - oldval = rdtp->dynticks_nesting; - rdtp->dynticks_nesting--; WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && - rdtp->dynticks_nesting < 0); - if (rdtp->dynticks_nesting) - trace_rcu_dyntick(TPS("--="), oldval, rdtp->dynticks_nesting); - else - rcu_eqs_enter_common(oldval, true); - rcu_sysidle_enter(1); + rdtp->dynticks_nesting < 1); + if (rdtp->dynticks_nesting <= 1) { + rcu_eqs_enter_common(true); + } else { + trace_rcu_dyntick(TPS("--="), rdtp->dynticks_nesting, rdtp->dynticks_nesting - 1); + rdtp->dynticks_nesting--; + } } /* @@ -826,15 +919,10 @@ void rcu_irq_exit_irqson(void) */ static void rcu_eqs_exit_common(long long oldval, int user) { - struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); + RCU_TRACE(struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);) rcu_dynticks_task_exit(); - smp_mb__before_atomic(); /* Force ordering w/previous sojourn. */ - atomic_inc(&rdtp->dynticks); - /* CPUs seeing atomic_inc() must see later RCU read-side crit sects */ - smp_mb__after_atomic(); /* See above. */ - WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && - !(atomic_read(&rdtp->dynticks) & 0x1)); + rcu_dynticks_eqs_exit(); rcu_cleanup_after_idle(); trace_rcu_dyntick(TPS("End"), oldval, rdtp->dynticks_nesting); if (IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && @@ -860,6 +948,7 @@ static void rcu_eqs_exit(bool user) struct rcu_dynticks *rdtp; long long oldval; + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_eqs_exit() invoked with irqs enabled!!!"); rdtp = this_cpu_ptr(&rcu_dynticks); oldval = rdtp->dynticks_nesting; WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && oldval < 0); @@ -888,7 +977,6 @@ void rcu_idle_exit(void) local_irq_save(flags); rcu_eqs_exit(false); - rcu_sysidle_exit(0); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(rcu_idle_exit); @@ -940,7 +1028,6 @@ void rcu_irq_enter(void) trace_rcu_dyntick(TPS("++="), oldval, rdtp->dynticks_nesting); else rcu_eqs_exit_common(oldval, true); - rcu_sysidle_exit(1); } /* @@ -980,12 +1067,8 @@ void rcu_nmi_enter(void) * to be in the outermost NMI handler that interrupted an RCU-idle * period (observation due to Andy Lutomirski). */ - if (!(atomic_read(&rdtp->dynticks) & 0x1)) { - smp_mb__before_atomic(); /* Force delay from prior write. */ - atomic_inc(&rdtp->dynticks); - /* atomic_inc() before later RCU read-side crit sects */ - smp_mb__after_atomic(); /* See above. */ - WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1)); + if (rcu_dynticks_curr_cpu_in_eqs()) { + rcu_dynticks_eqs_exit(); incby = 1; } rdtp->dynticks_nmi_nesting += incby; @@ -1010,7 +1093,7 @@ void rcu_nmi_exit(void) * to us!) */ WARN_ON_ONCE(rdtp->dynticks_nmi_nesting <= 0); - WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1)); + WARN_ON_ONCE(rcu_dynticks_curr_cpu_in_eqs()); /* * If the nesting level is not 1, the CPU wasn't RCU-idle, so @@ -1023,30 +1106,15 @@ void rcu_nmi_exit(void) /* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */ rdtp->dynticks_nmi_nesting = 0; - /* CPUs seeing atomic_inc() must see prior RCU read-side crit sects */ - smp_mb__before_atomic(); /* See above. */ - atomic_inc(&rdtp->dynticks); - smp_mb__after_atomic(); /* Force delay to next write. */ - WARN_ON_ONCE(atomic_read(&rdtp->dynticks) & 0x1); -} - -/** - * __rcu_is_watching - are RCU read-side critical sections safe? - * - * Return true if RCU is watching the running CPU, which means that - * this CPU can safely enter RCU read-side critical sections. Unlike - * rcu_is_watching(), the caller of __rcu_is_watching() must have at - * least disabled preemption. - */ -bool notrace __rcu_is_watching(void) -{ - return atomic_read(this_cpu_ptr(&rcu_dynticks.dynticks)) & 0x1; + rcu_dynticks_eqs_enter(); } /** * rcu_is_watching - see if RCU thinks that the current CPU is idle * - * If the current CPU is in its idle loop and is neither in an interrupt + * Return true if RCU is watching the running CPU, which means that this + * CPU can safely enter RCU read-side critical sections. In other words, + * if the current CPU is in its idle loop and is neither in an interrupt * or NMI handler, return true. */ bool notrace rcu_is_watching(void) @@ -1054,12 +1122,30 @@ bool notrace rcu_is_watching(void) bool ret; preempt_disable_notrace(); - ret = __rcu_is_watching(); + ret = !rcu_dynticks_curr_cpu_in_eqs(); preempt_enable_notrace(); return ret; } EXPORT_SYMBOL_GPL(rcu_is_watching); +/* + * If a holdout task is actually running, request an urgent quiescent + * state from its CPU. This is unsynchronized, so migrations can cause + * the request to go to the wrong CPU. Which is OK, all that will happen + * is that the CPU's next context switch will be a bit slower and next + * time around this task will generate another request. + */ +void rcu_request_urgent_qs_task(struct task_struct *t) +{ + int cpu; + + barrier(); + cpu = task_cpu(t); + if (!task_curr(t)) + return; /* This task is not running on that CPU. */ + smp_store_release(per_cpu_ptr(&rcu_dynticks.rcu_urgent_qs, cpu), true); +} + #if defined(CONFIG_PROVE_RCU) && defined(CONFIG_HOTPLUG_CPU) /* @@ -1120,12 +1206,10 @@ static int rcu_is_cpu_rrupt_from_idle(void) * credit them with an implicit quiescent state. Return 1 if this CPU * is in dynticks idle mode, which is an extended quiescent state. */ -static int dyntick_save_progress_counter(struct rcu_data *rdp, - bool *isidle, unsigned long *maxj) +static int dyntick_save_progress_counter(struct rcu_data *rdp) { - rdp->dynticks_snap = atomic_add_return(0, &rdp->dynticks->dynticks); - rcu_sysidle_check_cpu(rdp, isidle, maxj); - if ((rdp->dynticks_snap & 0x1) == 0) { + rdp->dynticks_snap = rcu_dynticks_snap(rdp->dynticks); + if (rcu_dynticks_in_eqs(rdp->dynticks_snap)) { trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti")); if (ULONG_CMP_LT(READ_ONCE(rdp->gpnum) + ULONG_MAX / 4, rdp->mynode->gpnum)) @@ -1141,15 +1225,13 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp, * idle state since the last call to dyntick_save_progress_counter() * for this same CPU, or by virtue of having been offline. */ -static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, - bool *isidle, unsigned long *maxj) +static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) { - unsigned int curr; - int *rcrmp; - unsigned int snap; - - curr = (unsigned int)atomic_add_return(0, &rdp->dynticks->dynticks); - snap = (unsigned int)rdp->dynticks_snap; + unsigned long jtsq; + bool *rnhqp; + bool *ruqp; + unsigned long rjtsc; + struct rcu_node *rnp; /* * If the CPU passed through or entered a dynticks idle phase with @@ -1159,27 +1241,43 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, * read-side critical section that started before the beginning * of the current RCU grace period. */ - if ((curr & 0x1) == 0 || UINT_CMP_GE(curr, snap + 2)) { + if (rcu_dynticks_in_eqs_since(rdp->dynticks, rdp->dynticks_snap)) { trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti")); rdp->dynticks_fqs++; return 1; } + /* Compute and saturate jiffies_till_sched_qs. */ + jtsq = jiffies_till_sched_qs; + rjtsc = rcu_jiffies_till_stall_check(); + if (jtsq > rjtsc / 2) { + WRITE_ONCE(jiffies_till_sched_qs, rjtsc); + jtsq = rjtsc / 2; + } else if (jtsq < 1) { + WRITE_ONCE(jiffies_till_sched_qs, 1); + jtsq = 1; + } + /* - * Check for the CPU being offline, but only if the grace period - * is old enough. We don't need to worry about the CPU changing - * state: If we see it offline even once, it has been through a - * quiescent state. - * - * The reason for insisting that the grace period be at least - * one jiffy old is that CPUs that are not quite online and that - * have just gone offline can still execute RCU read-side critical - * sections. + * Has this CPU encountered a cond_resched_rcu_qs() since the + * beginning of the grace period? For this to be the case, + * the CPU has to have noticed the current grace period. This + * might not be the case for nohz_full CPUs looping in the kernel. */ - if (ULONG_CMP_GE(rdp->rsp->gp_start + 2, jiffies)) - return 0; /* Grace period is not old enough. */ - barrier(); - if (cpu_is_offline(rdp->cpu)) { + rnp = rdp->mynode; + ruqp = per_cpu_ptr(&rcu_dynticks.rcu_urgent_qs, rdp->cpu); + if (time_after(jiffies, rdp->rsp->gp_start + jtsq) && + READ_ONCE(rdp->rcu_qs_ctr_snap) != per_cpu(rcu_dynticks.rcu_qs_ctr, rdp->cpu) && + READ_ONCE(rdp->gpnum) == rnp->gpnum && !rdp->gpwrap) { + trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("rqc")); + return 1; + } else { + /* Load rcu_qs_ctr before store to rcu_urgent_qs. */ + smp_store_release(ruqp, true); + } + + /* Check for the CPU being offline. */ + if (!(rdp->grpmask & rcu_rnp_online_cpus(rnp))) { trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("ofl")); rdp->offline_fqs++; return 1; @@ -1192,7 +1290,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, * in-kernel CPU-bound tasks cannot advance grace periods. * So if the grace period is old enough, make the CPU pay attention. * Note that the unsynchronized assignments to the per-CPU - * rcu_sched_qs_mask variable are safe. Yes, setting of + * rcu_need_heavy_qs variable are safe. Yes, setting of * bits can be lost, but they will be set again on the next * force-quiescent-state pass. So lost bit sets do not result * in incorrect behavior, merely in a grace period lasting @@ -1206,25 +1304,22 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, * is set too high, we override with half of the RCU CPU stall * warning delay. */ - rcrmp = &per_cpu(rcu_sched_qs_mask, rdp->cpu); - if (ULONG_CMP_GE(jiffies, - rdp->rsp->gp_start + jiffies_till_sched_qs) || - ULONG_CMP_GE(jiffies, rdp->rsp->jiffies_resched)) { - if (!(READ_ONCE(*rcrmp) & rdp->rsp->flavor_mask)) { - WRITE_ONCE(rdp->cond_resched_completed, - READ_ONCE(rdp->mynode->completed)); - smp_mb(); /* ->cond_resched_completed before *rcrmp. */ - WRITE_ONCE(*rcrmp, - READ_ONCE(*rcrmp) + rdp->rsp->flavor_mask); - } + rnhqp = &per_cpu(rcu_dynticks.rcu_need_heavy_qs, rdp->cpu); + if (!READ_ONCE(*rnhqp) && + (time_after(jiffies, rdp->rsp->gp_start + jtsq) || + time_after(jiffies, rdp->rsp->jiffies_resched))) { + WRITE_ONCE(*rnhqp, true); + /* Store rcu_need_heavy_qs before rcu_urgent_qs. */ + smp_store_release(ruqp, true); rdp->rsp->jiffies_resched += 5; /* Re-enable beating. */ } - /* And if it has been a really long time, kick the CPU as well. */ - if (ULONG_CMP_GE(jiffies, - rdp->rsp->gp_start + 2 * jiffies_till_sched_qs) || - ULONG_CMP_GE(jiffies, rdp->rsp->gp_start + jiffies_till_sched_qs)) - resched_cpu(rdp->cpu); /* Force CPU into scheduler. */ + /* + * If more than halfway to RCU CPU stall-warning time, do + * a resched_cpu() to try to loosen things up a bit. + */ + if (jiffies - rdp->rsp->gp_start > rcu_jiffies_till_stall_check() / 2) + resched_cpu(rdp->cpu); return 0; } @@ -1277,7 +1372,10 @@ static void rcu_check_gp_kthread_starvation(struct rcu_state *rsp) } /* - * Dump stacks of all tasks running on stalled CPUs. + * Dump stacks of all tasks running on stalled CPUs. First try using + * NMIs, but fall back to manual remote stack tracing on architectures + * that don't support NMI-based stack dumps. The NMI-triggered stack + * traces are more accurate because they are printed by the target CPU. */ static void rcu_dump_cpu_stacks(struct rcu_state *rsp) { @@ -1287,11 +1385,10 @@ static void rcu_dump_cpu_stacks(struct rcu_state *rsp) rcu_for_each_leaf_node(rsp, rnp) { raw_spin_lock_irqsave_rcu_node(rnp, flags); - if (rnp->qsmask != 0) { - for_each_leaf_node_possible_cpu(rnp, cpu) - if (rnp->qsmask & leaf_node_cpu_bit(rnp, cpu)) + for_each_leaf_node_possible_cpu(rnp, cpu) + if (rnp->qsmask & leaf_node_cpu_bit(rnp, cpu)) + if (!trigger_single_cpu_backtrace(cpu)) dump_cpu_task(cpu); - } raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } } @@ -1373,12 +1470,16 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum) print_cpu_stall_info_end(); for_each_possible_cpu(cpu) - totqlen += per_cpu_ptr(rsp->rda, cpu)->qlen; + totqlen += rcu_segcblist_n_cbs(&per_cpu_ptr(rsp->rda, + cpu)->cblist); pr_cont("(detected by %d, t=%ld jiffies, g=%ld, c=%ld, q=%lu)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start), (long)rsp->gpnum, (long)rsp->completed, totqlen); if (ndetected) { rcu_dump_cpu_stacks(rsp); + + /* Complain about tasks blocking the grace period. */ + rcu_print_detail_task_stall(rsp); } else { if (READ_ONCE(rsp->gpnum) != gpnum || READ_ONCE(rsp->completed) == gpnum) { @@ -1395,9 +1496,6 @@ static void print_other_cpu_stall(struct rcu_state *rsp, unsigned long gpnum) } } - /* Complain about tasks blocking the grace period. */ - rcu_print_detail_task_stall(rsp); - rcu_check_gp_kthread_starvation(rsp); panic_on_rcu_stall(); @@ -1427,7 +1525,8 @@ static void print_cpu_stall(struct rcu_state *rsp) print_cpu_stall_info(rsp, smp_processor_id()); print_cpu_stall_info_end(); for_each_possible_cpu(cpu) - totqlen += per_cpu_ptr(rsp->rda, cpu)->qlen; + totqlen += rcu_segcblist_n_cbs(&per_cpu_ptr(rsp->rda, + cpu)->cblist); pr_cont(" (t=%lu jiffies g=%ld c=%ld q=%lu)\n", jiffies - rsp->gp_start, (long)rsp->gpnum, (long)rsp->completed, totqlen); @@ -1530,30 +1629,6 @@ void rcu_cpu_stall_reset(void) } /* - * Initialize the specified rcu_data structure's default callback list - * to empty. The default callback list is the one that is not used by - * no-callbacks CPUs. - */ -static void init_default_callback_list(struct rcu_data *rdp) -{ - int i; - - rdp->nxtlist = NULL; - for (i = 0; i < RCU_NEXT_SIZE; i++) - rdp->nxttail[i] = &rdp->nxtlist; -} - -/* - * Initialize the specified rcu_data structure's callback list to empty. - */ -static void init_callback_list(struct rcu_data *rdp) -{ - if (init_nocb_callback_list(rdp)) - return; - init_default_callback_list(rdp); -} - -/* * Determine the value that ->completed will have at the end of the * next subsequent grace period. This is used to tag callbacks so that * a CPU can invoke callbacks in a timely fashion even if that CPU has @@ -1565,6 +1640,8 @@ static void init_callback_list(struct rcu_data *rdp) static unsigned long rcu_cbs_completed(struct rcu_state *rsp, struct rcu_node *rnp) { + lockdep_assert_held(&rnp->lock); + /* * If RCU is idle, we just wait for the next grace period. * But we can only be sure that RCU is idle if we are looking @@ -1607,10 +1684,11 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp, unsigned long *c_out) { unsigned long c; - int i; bool ret = false; struct rcu_node *rnp_root = rcu_get_root(rdp->rsp); + lockdep_assert_held(&rnp->lock); + /* * Pick up grace-period number for new callbacks. If this * grace period is already marked as needed, return to the caller. @@ -1653,13 +1731,11 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp, /* * Get a new grace-period number. If there really is no grace * period in progress, it will be smaller than the one we obtained - * earlier. Adjust callbacks as needed. Note that even no-CBs - * CPUs have a ->nxtcompleted[] array, so no no-CBs checks needed. + * earlier. Adjust callbacks as needed. */ c = rcu_cbs_completed(rdp->rsp, rnp_root); - for (i = RCU_DONE_TAIL; i < RCU_NEXT_TAIL; i++) - if (ULONG_CMP_LT(c, rdp->nxtcompleted[i])) - rdp->nxtcompleted[i] = c; + if (!rcu_is_nocb_cpu(rdp->cpu)) + (void)rcu_segcblist_accelerate(&rdp->cblist, c); /* * If the needed for the required grace period is already @@ -1691,9 +1767,7 @@ out: /* * Clean up any old requests for the just-ended grace period. Also return - * whether any additional grace periods have been requested. Also invoke - * rcu_nocb_gp_cleanup() in order to wake up any no-callbacks kthreads - * waiting for this grace period to complete. + * whether any additional grace periods have been requested. */ static int rcu_future_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp) { @@ -1739,57 +1813,29 @@ static void rcu_gp_kthread_wake(struct rcu_state *rsp) static bool rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { - unsigned long c; - int i; - bool ret; - - /* If the CPU has no callbacks, nothing to do. */ - if (!rdp->nxttail[RCU_NEXT_TAIL] || !*rdp->nxttail[RCU_DONE_TAIL]) - return false; + bool ret = false; - /* - * Starting from the sublist containing the callbacks most - * recently assigned a ->completed number and working down, find the - * first sublist that is not assignable to an upcoming grace period. - * Such a sublist has something in it (first two tests) and has - * a ->completed number assigned that will complete sooner than - * the ->completed number for newly arrived callbacks (last test). - * - * The key point is that any later sublist can be assigned the - * same ->completed number as the newly arrived callbacks, which - * means that the callbacks in any of these later sublist can be - * grouped into a single sublist, whether or not they have already - * been assigned a ->completed number. - */ - c = rcu_cbs_completed(rsp, rnp); - for (i = RCU_NEXT_TAIL - 1; i > RCU_DONE_TAIL; i--) - if (rdp->nxttail[i] != rdp->nxttail[i - 1] && - !ULONG_CMP_GE(rdp->nxtcompleted[i], c)) - break; + lockdep_assert_held(&rnp->lock); - /* - * If there are no sublist for unassigned callbacks, leave. - * At the same time, advance "i" one sublist, so that "i" will - * index into the sublist where all the remaining callbacks should - * be grouped into. - */ - if (++i >= RCU_NEXT_TAIL) + /* If no pending (not yet ready to invoke) callbacks, nothing to do. */ + if (!rcu_segcblist_pend_cbs(&rdp->cblist)) return false; /* - * Assign all subsequent callbacks' ->completed number to the next - * full grace period and group them all in the sublist initially - * indexed by "i". + * Callbacks are often registered with incomplete grace-period + * information. Something about the fact that getting exact + * information requires acquiring a global lock... RCU therefore + * makes a conservative estimate of the grace period number at which + * a given callback will become ready to invoke. The following + * code checks this estimate and improves it when possible, thus + * accelerating callback invocation to an earlier grace-period + * number. */ - for (; i <= RCU_NEXT_TAIL; i++) { - rdp->nxttail[i] = rdp->nxttail[RCU_NEXT_TAIL]; - rdp->nxtcompleted[i] = c; - } - /* Record any needed additional grace periods. */ - ret = rcu_start_future_gp(rnp, rdp, NULL); + if (rcu_segcblist_accelerate(&rdp->cblist, rcu_cbs_completed(rsp, rnp))) + ret = rcu_start_future_gp(rnp, rdp, NULL); /* Trace depending on how much we were able to accelerate. */ - if (!*rdp->nxttail[RCU_WAIT_TAIL]) + if (rcu_segcblist_restempty(&rdp->cblist, RCU_WAIT_TAIL)) trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccWaitCB")); else trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccReadyCB")); @@ -1809,32 +1855,17 @@ static bool rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp, static bool rcu_advance_cbs(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { - int i, j; + lockdep_assert_held(&rnp->lock); - /* If the CPU has no callbacks, nothing to do. */ - if (!rdp->nxttail[RCU_NEXT_TAIL] || !*rdp->nxttail[RCU_DONE_TAIL]) + /* If no pending (not yet ready to invoke) callbacks, nothing to do. */ + if (!rcu_segcblist_pend_cbs(&rdp->cblist)) return false; /* * Find all callbacks whose ->completed numbers indicate that they * are ready to invoke, and put them into the RCU_DONE_TAIL sublist. */ - for (i = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++) { - if (ULONG_CMP_LT(rnp->completed, rdp->nxtcompleted[i])) - break; - rdp->nxttail[RCU_DONE_TAIL] = rdp->nxttail[i]; - } - /* Clean up any sublist tail pointers that were misordered above. */ - for (j = RCU_WAIT_TAIL; j < i; j++) - rdp->nxttail[j] = rdp->nxttail[RCU_DONE_TAIL]; - - /* Copy down callbacks to fill in empty sublists. */ - for (j = RCU_WAIT_TAIL; i < RCU_NEXT_TAIL; i++, j++) { - if (rdp->nxttail[j] == rdp->nxttail[RCU_NEXT_TAIL]) - break; - rdp->nxttail[j] = rdp->nxttail[i]; - rdp->nxtcompleted[j] = rdp->nxtcompleted[i]; - } + rcu_segcblist_advance(&rdp->cblist, rnp->completed); /* Classify any remaining callbacks. */ return rcu_accelerate_cbs(rsp, rnp, rdp); @@ -1852,6 +1883,8 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, bool ret; bool need_gp; + lockdep_assert_held(&rnp->lock); + /* Handle the ends of any preceding grace periods first. */ if (rdp->completed == rnp->completed && !unlikely(READ_ONCE(rdp->gpwrap))) { @@ -1879,7 +1912,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpustart")); need_gp = !!(rnp->qsmask & rdp->grpmask); rdp->cpu_no_qs.b.norm = need_gp; - rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); + rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_dynticks.rcu_qs_ctr); rdp->core_needs_qs = need_gp; zero_cpu_stall_ticks(rdp); WRITE_ONCE(rdp->gpwrap, false); @@ -2058,25 +2091,16 @@ static bool rcu_gp_fqs_check_wake(struct rcu_state *rsp, int *gfp) */ static void rcu_gp_fqs(struct rcu_state *rsp, bool first_time) { - bool isidle = false; - unsigned long maxj; struct rcu_node *rnp = rcu_get_root(rsp); WRITE_ONCE(rsp->gp_activity, jiffies); rsp->n_force_qs++; if (first_time) { /* Collect dyntick-idle snapshots. */ - if (is_sysidle_rcu_state(rsp)) { - isidle = true; - maxj = jiffies - ULONG_MAX / 4; - } - force_qs_rnp(rsp, dyntick_save_progress_counter, - &isidle, &maxj); - rcu_sysidle_report_gp(rsp, isidle, maxj); + force_qs_rnp(rsp, dyntick_save_progress_counter); } else { /* Handle dyntick-idle and offline CPUs. */ - isidle = true; - force_qs_rnp(rsp, rcu_implicit_dynticks_qs, &isidle, &maxj); + force_qs_rnp(rsp, rcu_implicit_dynticks_qs); } /* Clear flag to prevent immediate re-entry. */ if (READ_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) { @@ -2284,6 +2308,7 @@ static bool rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { + lockdep_assert_held(&rnp->lock); if (!rsp->gp_kthread || !cpu_needs_another_gp(rsp, rdp)) { /* * Either we have not yet spawned the grace-period @@ -2345,6 +2370,7 @@ static bool rcu_start_gp(struct rcu_state *rsp) static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) __releases(rcu_get_root(rsp)->lock) { + lockdep_assert_held(&rcu_get_root(rsp)->lock); WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); WRITE_ONCE(rsp->gp_flags, READ_ONCE(rsp->gp_flags) | RCU_GP_FLAG_FQS); raw_spin_unlock_irqrestore_rcu_node(rcu_get_root(rsp), flags); @@ -2369,6 +2395,8 @@ rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp, unsigned long oldmask = 0; struct rcu_node *rnp_c; + lockdep_assert_held(&rnp->lock); + /* Walk up the rcu_node hierarchy. */ for (;;) { if (!(rnp->qsmask & mask) || rnp->gpnum != gps) { @@ -2429,6 +2457,7 @@ static void rcu_report_unblock_qs_rnp(struct rcu_state *rsp, unsigned long mask; struct rcu_node *rnp_p; + lockdep_assert_held(&rnp->lock); if (rcu_state_p == &rcu_sched_state || rsp != rcu_state_p || rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) { raw_spin_unlock_irqrestore_rcu_node(rnp, flags); @@ -2467,10 +2496,8 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp) rnp = rdp->mynode; raw_spin_lock_irqsave_rcu_node(rnp, flags); - if ((rdp->cpu_no_qs.b.norm && - rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) || - rdp->gpnum != rnp->gpnum || rnp->completed == rnp->gpnum || - rdp->gpwrap) { + if (rdp->cpu_no_qs.b.norm || rdp->gpnum != rnp->gpnum || + rnp->completed == rnp->gpnum || rdp->gpwrap) { /* * The grace period in which this quiescent state was @@ -2479,7 +2506,7 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp) * within the current grace period. */ rdp->cpu_no_qs.b.norm = true; /* need qs for new gp. */ - rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr); + rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_dynticks.rcu_qs_ctr); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); return; } @@ -2525,8 +2552,7 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) * Was there a quiescent state since the beginning of the grace * period? If no, then exit and wait for the next call. */ - if (rdp->cpu_no_qs.b.norm && - rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) + if (rdp->cpu_no_qs.b.norm) return; /* @@ -2545,6 +2571,8 @@ static void rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { + lockdep_assert_held(&rsp->orphan_lock); + /* No-CBs CPUs do not have orphanable callbacks. */ if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || rcu_is_nocb_cpu(rdp->cpu)) return; @@ -2554,13 +2582,8 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, * because _rcu_barrier() excludes CPU-hotplug operations, so it * cannot be running now. Thus no memory barrier is required. */ - if (rdp->nxtlist != NULL) { - rsp->qlen_lazy += rdp->qlen_lazy; - rsp->qlen += rdp->qlen; - rdp->n_cbs_orphaned += rdp->qlen; - rdp->qlen_lazy = 0; - WRITE_ONCE(rdp->qlen, 0); - } + rdp->n_cbs_orphaned += rcu_segcblist_n_cbs(&rdp->cblist); + rcu_segcblist_extract_count(&rdp->cblist, &rsp->orphan_done); /* * Next, move those callbacks still needing a grace period to @@ -2568,31 +2591,18 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, * Some of the callbacks might have gone partway through a grace * period, but that is too bad. They get to start over because we * cannot assume that grace periods are synchronized across CPUs. - * We don't bother updating the ->nxttail[] array yet, instead - * we just reset the whole thing later on. */ - if (*rdp->nxttail[RCU_DONE_TAIL] != NULL) { - *rsp->orphan_nxttail = *rdp->nxttail[RCU_DONE_TAIL]; - rsp->orphan_nxttail = rdp->nxttail[RCU_NEXT_TAIL]; - *rdp->nxttail[RCU_DONE_TAIL] = NULL; - } + rcu_segcblist_extract_pend_cbs(&rdp->cblist, &rsp->orphan_pend); /* * Then move the ready-to-invoke callbacks to the orphanage, * where some other CPU will pick them up. These will not be * required to pass though another grace period: They are done. */ - if (rdp->nxtlist != NULL) { - *rsp->orphan_donetail = rdp->nxtlist; - rsp->orphan_donetail = rdp->nxttail[RCU_DONE_TAIL]; - } + rcu_segcblist_extract_done_cbs(&rdp->cblist, &rsp->orphan_done); - /* - * Finally, initialize the rcu_data structure's list to empty and - * disallow further callbacks on this CPU. - */ - init_callback_list(rdp); - rdp->nxttail[RCU_NEXT_TAIL] = NULL; + /* Finally, disallow further callbacks on this CPU. */ + rcu_segcblist_disable(&rdp->cblist); } /* @@ -2601,22 +2611,20 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, */ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags) { - int i; struct rcu_data *rdp = raw_cpu_ptr(rsp->rda); + lockdep_assert_held(&rsp->orphan_lock); + /* No-CBs CPUs are handled specially. */ if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || rcu_nocb_adopt_orphan_cbs(rsp, rdp, flags)) return; /* Do the accounting first. */ - rdp->qlen_lazy += rsp->qlen_lazy; - rdp->qlen += rsp->qlen; - rdp->n_cbs_adopted += rsp->qlen; - if (rsp->qlen_lazy != rsp->qlen) + rdp->n_cbs_adopted += rsp->orphan_done.len; + if (rsp->orphan_done.len_lazy != rsp->orphan_done.len) rcu_idle_count_callbacks_posted(); - rsp->qlen_lazy = 0; - rsp->qlen = 0; + rcu_segcblist_insert_count(&rdp->cblist, &rsp->orphan_done); /* * We do not need a memory barrier here because the only way we @@ -2624,24 +2632,13 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags) * we are the task doing the rcu_barrier(). */ - /* First adopt the ready-to-invoke callbacks. */ - if (rsp->orphan_donelist != NULL) { - *rsp->orphan_donetail = *rdp->nxttail[RCU_DONE_TAIL]; - *rdp->nxttail[RCU_DONE_TAIL] = rsp->orphan_donelist; - for (i = RCU_NEXT_SIZE - 1; i >= RCU_DONE_TAIL; i--) - if (rdp->nxttail[i] == rdp->nxttail[RCU_DONE_TAIL]) - rdp->nxttail[i] = rsp->orphan_donetail; - rsp->orphan_donelist = NULL; - rsp->orphan_donetail = &rsp->orphan_donelist; - } - - /* And then adopt the callbacks that still need a grace period. */ - if (rsp->orphan_nxtlist != NULL) { - *rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_nxtlist; - rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_nxttail; - rsp->orphan_nxtlist = NULL; - rsp->orphan_nxttail = &rsp->orphan_nxtlist; - } + /* First adopt the ready-to-invoke callbacks, then the done ones. */ + rcu_segcblist_insert_done_cbs(&rdp->cblist, &rsp->orphan_done); + WARN_ON_ONCE(rsp->orphan_done.head); + rcu_segcblist_insert_pend_cbs(&rdp->cblist, &rsp->orphan_pend); + WARN_ON_ONCE(rsp->orphan_pend.head); + WARN_ON_ONCE(rcu_segcblist_empty(&rdp->cblist) != + !rcu_segcblist_n_cbs(&rdp->cblist)); } /* @@ -2649,14 +2646,14 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags) */ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) { - RCU_TRACE(unsigned long mask); - RCU_TRACE(struct rcu_data *rdp = this_cpu_ptr(rsp->rda)); - RCU_TRACE(struct rcu_node *rnp = rdp->mynode); + RCU_TRACE(unsigned long mask;) + RCU_TRACE(struct rcu_data *rdp = this_cpu_ptr(rsp->rda);) + RCU_TRACE(struct rcu_node *rnp = rdp->mynode;) if (!IS_ENABLED(CONFIG_HOTPLUG_CPU)) return; - RCU_TRACE(mask = rdp->grpmask); + RCU_TRACE(mask = rdp->grpmask;) trace_rcu_grace_period(rsp->name, rnp->gpnum + 1 - !!(rnp->qsmask & mask), TPS("cpuofl")); @@ -2684,6 +2681,7 @@ static void rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf) long mask; struct rcu_node *rnp = rnp_leaf; + lockdep_assert_held(&rnp->lock); if (!IS_ENABLED(CONFIG_HOTPLUG_CPU) || rnp->qsmaskinit || rcu_preempt_has_tasks(rnp)) return; @@ -2729,9 +2727,11 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) rcu_adopt_orphan_cbs(rsp, flags); raw_spin_unlock_irqrestore(&rsp->orphan_lock, flags); - WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL, - "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n", - cpu, rdp->qlen, rdp->nxtlist); + WARN_ONCE(rcu_segcblist_n_cbs(&rdp->cblist) != 0 || + !rcu_segcblist_empty(&rdp->cblist), + "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, 1stCB=%p\n", + cpu, rcu_segcblist_n_cbs(&rdp->cblist), + rcu_segcblist_first_cb(&rdp->cblist)); } /* @@ -2741,14 +2741,17 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) { unsigned long flags; - struct rcu_head *next, *list, **tail; - long bl, count, count_lazy; - int i; + struct rcu_head *rhp; + struct rcu_cblist rcl = RCU_CBLIST_INITIALIZER(rcl); + long bl, count; /* If no callbacks are ready, just return. */ - if (!cpu_has_callbacks_ready_to_invoke(rdp)) { - trace_rcu_batch_start(rsp->name, rdp->qlen_lazy, rdp->qlen, 0); - trace_rcu_batch_end(rsp->name, 0, !!READ_ONCE(rdp->nxtlist), + if (!rcu_segcblist_ready_cbs(&rdp->cblist)) { + trace_rcu_batch_start(rsp->name, + rcu_segcblist_n_lazy_cbs(&rdp->cblist), + rcu_segcblist_n_cbs(&rdp->cblist), 0); + trace_rcu_batch_end(rsp->name, 0, + !rcu_segcblist_empty(&rdp->cblist), need_resched(), is_idle_task(current), rcu_is_callbacks_kthread()); return; @@ -2756,73 +2759,61 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) /* * Extract the list of ready callbacks, disabling to prevent - * races with call_rcu() from interrupt handlers. + * races with call_rcu() from interrupt handlers. Leave the + * callback counts, as rcu_barrier() needs to be conservative. */ local_irq_save(flags); WARN_ON_ONCE(cpu_is_offline(smp_processor_id())); bl = rdp->blimit; - trace_rcu_batch_start(rsp->name, rdp->qlen_lazy, rdp->qlen, bl); - list = rdp->nxtlist; - rdp->nxtlist = *rdp->nxttail[RCU_DONE_TAIL]; - *rdp->nxttail[RCU_DONE_TAIL] = NULL; - tail = rdp->nxttail[RCU_DONE_TAIL]; - for (i = RCU_NEXT_SIZE - 1; i >= 0; i--) - if (rdp->nxttail[i] == rdp->nxttail[RCU_DONE_TAIL]) - rdp->nxttail[i] = &rdp->nxtlist; + trace_rcu_batch_start(rsp->name, rcu_segcblist_n_lazy_cbs(&rdp->cblist), + rcu_segcblist_n_cbs(&rdp->cblist), bl); + rcu_segcblist_extract_done_cbs(&rdp->cblist, &rcl); local_irq_restore(flags); /* Invoke callbacks. */ - count = count_lazy = 0; - while (list) { - next = list->next; - prefetch(next); - debug_rcu_head_unqueue(list); - if (__rcu_reclaim(rsp->name, list)) - count_lazy++; - list = next; - /* Stop only if limit reached and CPU has something to do. */ - if (++count >= bl && + rhp = rcu_cblist_dequeue(&rcl); + for (; rhp; rhp = rcu_cblist_dequeue(&rcl)) { + debug_rcu_head_unqueue(rhp); + if (__rcu_reclaim(rsp->name, rhp)) + rcu_cblist_dequeued_lazy(&rcl); + /* + * Stop only if limit reached and CPU has something to do. + * Note: The rcl structure counts down from zero. + */ + if (-rcl.len >= bl && (need_resched() || (!is_idle_task(current) && !rcu_is_callbacks_kthread()))) break; } local_irq_save(flags); - trace_rcu_batch_end(rsp->name, count, !!list, need_resched(), - is_idle_task(current), - rcu_is_callbacks_kthread()); - - /* Update count, and requeue any remaining callbacks. */ - if (list != NULL) { - *tail = rdp->nxtlist; - rdp->nxtlist = list; - for (i = 0; i < RCU_NEXT_SIZE; i++) - if (&rdp->nxtlist == rdp->nxttail[i]) - rdp->nxttail[i] = tail; - else - break; - } + count = -rcl.len; + trace_rcu_batch_end(rsp->name, count, !!rcl.head, need_resched(), + is_idle_task(current), rcu_is_callbacks_kthread()); + + /* Update counts and requeue any remaining callbacks. */ + rcu_segcblist_insert_done_cbs(&rdp->cblist, &rcl); smp_mb(); /* List handling before counting for rcu_barrier(). */ - rdp->qlen_lazy -= count_lazy; - WRITE_ONCE(rdp->qlen, rdp->qlen - count); rdp->n_cbs_invoked += count; + rcu_segcblist_insert_count(&rdp->cblist, &rcl); /* Reinstate batch limit if we have worked down the excess. */ - if (rdp->blimit == LONG_MAX && rdp->qlen <= qlowmark) + count = rcu_segcblist_n_cbs(&rdp->cblist); + if (rdp->blimit == LONG_MAX && count <= qlowmark) rdp->blimit = blimit; /* Reset ->qlen_last_fqs_check trigger if enough CBs have drained. */ - if (rdp->qlen == 0 && rdp->qlen_last_fqs_check != 0) { + if (count == 0 && rdp->qlen_last_fqs_check != 0) { rdp->qlen_last_fqs_check = 0; rdp->n_force_qs_snap = rsp->n_force_qs; - } else if (rdp->qlen < rdp->qlen_last_fqs_check - qhimark) - rdp->qlen_last_fqs_check = rdp->qlen; - WARN_ON_ONCE((rdp->nxtlist == NULL) != (rdp->qlen == 0)); + } else if (count < rdp->qlen_last_fqs_check - qhimark) + rdp->qlen_last_fqs_check = count; + WARN_ON_ONCE(rcu_segcblist_empty(&rdp->cblist) != (count == 0)); local_irq_restore(flags); /* Re-invoke RCU core processing if there are callbacks remaining. */ - if (cpu_has_callbacks_ready_to_invoke(rdp)) + if (rcu_segcblist_ready_cbs(&rdp->cblist)) invoke_rcu_core(); } @@ -2881,10 +2872,7 @@ void rcu_check_callbacks(int user) * * The caller must have suppressed start of new grace periods. */ -static void force_qs_rnp(struct rcu_state *rsp, - int (*f)(struct rcu_data *rsp, bool *isidle, - unsigned long *maxj), - bool *isidle, unsigned long *maxj) +static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *rsp)) { int cpu; unsigned long flags; @@ -2923,7 +2911,7 @@ static void force_qs_rnp(struct rcu_state *rsp, for_each_leaf_node_possible_cpu(rnp, cpu) { unsigned long bit = leaf_node_cpu_bit(rnp, cpu); if ((rnp->qsmask & bit) != 0) { - if (f(per_cpu_ptr(rsp->rda, cpu), isidle, maxj)) + if (f(per_cpu_ptr(rsp->rda, cpu))) mask |= bit; } } @@ -2988,7 +2976,7 @@ __rcu_process_callbacks(struct rcu_state *rsp) bool needwake; struct rcu_data *rdp = raw_cpu_ptr(rsp->rda); - WARN_ON_ONCE(rdp->beenonline == 0); + WARN_ON_ONCE(!rdp->beenonline); /* Update RCU state based on any recent quiescent states. */ rcu_check_quiescent_state(rsp, rdp); @@ -3006,7 +2994,7 @@ __rcu_process_callbacks(struct rcu_state *rsp) } /* If there are callbacks ready, invoke them. */ - if (cpu_has_callbacks_ready_to_invoke(rdp)) + if (rcu_segcblist_ready_cbs(&rdp->cblist)) invoke_rcu_callbacks(rsp, rdp); /* Do any needed deferred wakeups of rcuo kthreads. */ @@ -3078,7 +3066,8 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp, * invoking force_quiescent_state() if the newly enqueued callback * is the only one waiting for a grace period to complete. */ - if (unlikely(rdp->qlen > rdp->qlen_last_fqs_check + qhimark)) { + if (unlikely(rcu_segcblist_n_cbs(&rdp->cblist) > + rdp->qlen_last_fqs_check + qhimark)) { /* Are we ignoring a completed grace period? */ note_gp_changes(rsp, rdp); @@ -3096,10 +3085,10 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp, /* Give the grace period a kick. */ rdp->blimit = LONG_MAX; if (rsp->n_force_qs == rdp->n_force_qs_snap && - *rdp->nxttail[RCU_DONE_TAIL] != head) + rcu_segcblist_first_pend_cb(&rdp->cblist) != head) force_quiescent_state(rsp); rdp->n_force_qs_snap = rsp->n_force_qs; - rdp->qlen_last_fqs_check = rdp->qlen; + rdp->qlen_last_fqs_check = rcu_segcblist_n_cbs(&rdp->cblist); } } } @@ -3128,9 +3117,14 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func, WARN_ON_ONCE((unsigned long)head & (sizeof(void *) - 1)); if (debug_rcu_head_queue(head)) { - /* Probable double call_rcu(), so leak the callback. */ + /* + * Probable double call_rcu(), so leak the callback. + * Use rcu:rcu_callback trace event to find the previous + * time callback was passed to __call_rcu(). + */ + WARN_ONCE(1, "__call_rcu(): Double-freed CB %p->%pF()!!!\n", + head, head->func); WRITE_ONCE(head->func, rcu_leak_callback); - WARN_ONCE(1, "__call_rcu(): Leaked duplicate callback\n"); return; } head->func = func; @@ -3139,7 +3133,7 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func, rdp = this_cpu_ptr(rsp->rda); /* Add the callback to our list. */ - if (unlikely(rdp->nxttail[RCU_NEXT_TAIL] == NULL) || cpu != -1) { + if (unlikely(!rcu_segcblist_is_enabled(&rdp->cblist)) || cpu != -1) { int offline; if (cpu != -1) @@ -3158,31 +3152,45 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func, */ BUG_ON(cpu != -1); WARN_ON_ONCE(!rcu_is_watching()); - if (!likely(rdp->nxtlist)) - init_default_callback_list(rdp); + if (rcu_segcblist_empty(&rdp->cblist)) + rcu_segcblist_init(&rdp->cblist); } - WRITE_ONCE(rdp->qlen, rdp->qlen + 1); - if (lazy) - rdp->qlen_lazy++; - else + rcu_segcblist_enqueue(&rdp->cblist, head, lazy); + if (!lazy) rcu_idle_count_callbacks_posted(); - smp_mb(); /* Count before adding callback for rcu_barrier(). */ - *rdp->nxttail[RCU_NEXT_TAIL] = head; - rdp->nxttail[RCU_NEXT_TAIL] = &head->next; if (__is_kfree_rcu_offset((unsigned long)func)) trace_rcu_kfree_callback(rsp->name, head, (unsigned long)func, - rdp->qlen_lazy, rdp->qlen); + rcu_segcblist_n_lazy_cbs(&rdp->cblist), + rcu_segcblist_n_cbs(&rdp->cblist)); else - trace_rcu_callback(rsp->name, head, rdp->qlen_lazy, rdp->qlen); + trace_rcu_callback(rsp->name, head, + rcu_segcblist_n_lazy_cbs(&rdp->cblist), + rcu_segcblist_n_cbs(&rdp->cblist)); /* Go handle any RCU core processing required. */ __call_rcu_core(rsp, rdp, head, flags); local_irq_restore(flags); } -/* - * Queue an RCU-sched callback for invocation after a grace period. +/** + * call_rcu_sched() - Queue an RCU for invocation after sched grace period. + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all currently executing RCU + * read-side critical sections have completed. call_rcu_sched() assumes + * that the read-side critical sections end on enabling of preemption + * or on voluntary preemption. + * RCU read-side critical sections are delimited by : + * - rcu_read_lock_sched() and rcu_read_unlock_sched(), OR + * - anything that disables preemption. + * + * These may be nested. + * + * See the description of call_rcu() for more detailed information on + * memory ordering guarantees. */ void call_rcu_sched(struct rcu_head *head, rcu_callback_t func) { @@ -3190,8 +3198,26 @@ void call_rcu_sched(struct rcu_head *head, rcu_callback_t func) } EXPORT_SYMBOL_GPL(call_rcu_sched); -/* - * Queue an RCU callback for invocation after a quicker grace period. +/** + * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period. + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all currently executing RCU + * read-side critical sections have completed. call_rcu_bh() assumes + * that the read-side critical sections end on completion of a softirq + * handler. This means that read-side critical sections in process + * context must not be interrupted by softirqs. This interface is to be + * used when most of the read-side critical sections are in softirq context. + * RCU read-side critical sections are delimited by : + * - rcu_read_lock() and rcu_read_unlock(), if in interrupt context. + * OR + * - rcu_read_lock_bh() and rcu_read_unlock_bh(), if in process context. + * These may be nested. + * + * See the description of call_rcu() for more detailed information on + * memory ordering guarantees. */ void call_rcu_bh(struct rcu_head *head, rcu_callback_t func) { @@ -3267,12 +3293,6 @@ static inline int rcu_blocking_is_gp(void) * to have executed a full memory barrier during the execution of * synchronize_sched() -- even if CPU A and CPU B are the same CPU (but * again only if the system has more than one CPU). - * - * This primitive provides the guarantees made by the (now removed) - * synchronize_kernel() API. In contrast, synchronize_rcu() only - * guarantees that rcu_read_lock() sections will have completed. - * In "classic RCU", these two guarantees happen to be one and - * the same, but can differ in realtime RCU implementations. */ void synchronize_sched(void) { @@ -3420,41 +3440,6 @@ void cond_synchronize_sched(unsigned long oldstate) } EXPORT_SYMBOL_GPL(cond_synchronize_sched); -/* Adjust sequence number for start of update-side operation. */ -static void rcu_seq_start(unsigned long *sp) -{ - WRITE_ONCE(*sp, *sp + 1); - smp_mb(); /* Ensure update-side operation after counter increment. */ - WARN_ON_ONCE(!(*sp & 0x1)); -} - -/* Adjust sequence number for end of update-side operation. */ -static void rcu_seq_end(unsigned long *sp) -{ - smp_mb(); /* Ensure update-side operation before counter increment. */ - WRITE_ONCE(*sp, *sp + 1); - WARN_ON_ONCE(*sp & 0x1); -} - -/* Take a snapshot of the update side's sequence number. */ -static unsigned long rcu_seq_snap(unsigned long *sp) -{ - unsigned long s; - - s = (READ_ONCE(*sp) + 3) & ~0x1; - smp_mb(); /* Above access must not bleed into critical section. */ - return s; -} - -/* - * Given a snapshot from rcu_seq_snap(), determine whether or not a - * full update-side operation has occurred. - */ -static bool rcu_seq_done(unsigned long *sp, unsigned long s) -{ - return ULONG_CMP_GE(READ_ONCE(*sp), s); -} - /* * Check to see if there is any immediate RCU-related work to be done * by the current CPU, for the specified type of RCU, returning 1 if so. @@ -3478,17 +3463,15 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) /* Is the RCU core waiting for a quiescent state from this CPU? */ if (rcu_scheduler_fully_active && rdp->core_needs_qs && rdp->cpu_no_qs.b.norm && - rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) { + rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_dynticks.rcu_qs_ctr)) { rdp->n_rp_core_needs_qs++; - } else if (rdp->core_needs_qs && - (!rdp->cpu_no_qs.b.norm || - rdp->rcu_qs_ctr_snap != __this_cpu_read(rcu_qs_ctr))) { + } else if (rdp->core_needs_qs && !rdp->cpu_no_qs.b.norm) { rdp->n_rp_report_qs++; return 1; } /* Does this CPU have callbacks ready to invoke? */ - if (cpu_has_callbacks_ready_to_invoke(rdp)) { + if (rcu_segcblist_ready_cbs(&rdp->cblist)) { rdp->n_rp_cb_ready++; return 1; } @@ -3552,10 +3535,10 @@ static bool __maybe_unused rcu_cpu_has_callbacks(bool *all_lazy) for_each_rcu_flavor(rsp) { rdp = this_cpu_ptr(rsp->rda); - if (!rdp->nxtlist) + if (rcu_segcblist_empty(&rdp->cblist)) continue; hc = true; - if (rdp->qlen != rdp->qlen_lazy || !all_lazy) { + if (rcu_segcblist_n_nonlazy_cbs(&rdp->cblist) || !all_lazy) { al = false; break; } @@ -3602,8 +3585,14 @@ static void rcu_barrier_func(void *type) struct rcu_data *rdp = raw_cpu_ptr(rsp->rda); _rcu_barrier_trace(rsp, "IRQ", -1, rsp->barrier_sequence); - atomic_inc(&rsp->barrier_cpu_count); - rsp->call(&rdp->barrier_head, rcu_barrier_callback); + rdp->barrier_head.func = rcu_barrier_callback; + debug_rcu_head_queue(&rdp->barrier_head); + if (rcu_segcblist_entrain(&rdp->cblist, &rdp->barrier_head, 0)) { + atomic_inc(&rsp->barrier_cpu_count); + } else { + debug_rcu_head_unqueue(&rdp->barrier_head); + _rcu_barrier_trace(rsp, "IRQNQ", -1, rsp->barrier_sequence); + } } /* @@ -3664,7 +3653,7 @@ static void _rcu_barrier(struct rcu_state *rsp) __call_rcu(&rdp->barrier_head, rcu_barrier_callback, rsp, cpu, 0); } - } else if (READ_ONCE(rdp->qlen)) { + } else if (rcu_segcblist_n_cbs(&rdp->cblist)) { _rcu_barrier_trace(rsp, "OnlineQ", cpu, rsp->barrier_sequence); smp_call_function_single(cpu, rcu_barrier_func, rsp, 1); @@ -3722,6 +3711,7 @@ static void rcu_init_new_rnp(struct rcu_node *rnp_leaf) long mask; struct rcu_node *rnp = rnp_leaf; + lockdep_assert_held(&rnp->lock); for (;;) { mask = rnp->grpmask; rnp = rnp->parent; @@ -3748,7 +3738,7 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->grpmask = leaf_node_cpu_bit(rdp->mynode, cpu); rdp->dynticks = &per_cpu(rcu_dynticks, cpu); WARN_ON_ONCE(rdp->dynticks->dynticks_nesting != DYNTICK_TASK_EXIT_IDLE); - WARN_ON_ONCE(atomic_read(&rdp->dynticks->dynticks) != 1); + WARN_ON_ONCE(rcu_dynticks_in_eqs(rcu_dynticks_snap(rdp->dynticks))); rdp->cpu = cpu; rdp->rsp = rsp; rcu_boot_init_nocb_percpu_data(rdp); @@ -3765,7 +3755,6 @@ static void rcu_init_percpu_data(int cpu, struct rcu_state *rsp) { unsigned long flags; - unsigned long mask; struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rcu_get_root(rsp); @@ -3774,12 +3763,11 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->qlen_last_fqs_check = 0; rdp->n_force_qs_snap = rsp->n_force_qs; rdp->blimit = blimit; - if (!rdp->nxtlist) - init_callback_list(rdp); /* Re-enable callbacks on this CPU. */ + if (rcu_segcblist_empty(&rdp->cblist) && /* No early-boot CBs? */ + !init_nocb_callback_list(rdp)) + rcu_segcblist_init(&rdp->cblist); /* Re-enable callbacks. */ rdp->dynticks->dynticks_nesting = DYNTICK_TASK_EXIT_IDLE; - rcu_sysidle_init_percpu_data(rdp->dynticks); - atomic_set(&rdp->dynticks->dynticks, - (atomic_read(&rdp->dynticks->dynticks) & ~0x1) + 1); + rcu_dynticks_eqs_online(); raw_spin_unlock_rcu_node(rnp); /* irqs remain disabled. */ /* @@ -3788,7 +3776,6 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) * of the next grace period. */ rnp = rdp->mynode; - mask = rdp->grpmask; raw_spin_lock_rcu_node(rnp); /* irqs already disabled. */ if (!rdp->beenonline) WRITE_ONCE(rsp->ncpus, READ_ONCE(rsp->ncpus) + 1); @@ -3796,12 +3783,16 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->gpnum = rnp->completed; /* Make CPU later note any new GP. */ rdp->completed = rnp->completed; rdp->cpu_no_qs.b.norm = true; - rdp->rcu_qs_ctr_snap = per_cpu(rcu_qs_ctr, cpu); + rdp->rcu_qs_ctr_snap = per_cpu(rcu_dynticks.rcu_qs_ctr, cpu); rdp->core_needs_qs = false; trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl")); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } +/* + * Invoked early in the CPU-online process, when pretty much all + * services are available. The incoming CPU is not present. + */ int rcutree_prepare_cpu(unsigned int cpu) { struct rcu_state *rsp; @@ -3815,6 +3806,9 @@ int rcutree_prepare_cpu(unsigned int cpu) return 0; } +/* + * Update RCU priority boot kthread affinity for CPU-hotplug changes. + */ static void rcutree_affinity_setting(unsigned int cpu, int outgoing) { struct rcu_data *rdp = per_cpu_ptr(rcu_state_p->rda, cpu); @@ -3822,20 +3816,34 @@ static void rcutree_affinity_setting(unsigned int cpu, int outgoing) rcu_boost_kthread_setaffinity(rdp->mynode, outgoing); } +/* + * Near the end of the CPU-online process. Pretty much all services + * enabled, and the CPU is now very much alive. + */ int rcutree_online_cpu(unsigned int cpu) { sync_sched_exp_online_cleanup(cpu); rcutree_affinity_setting(cpu, -1); + if (IS_ENABLED(CONFIG_TREE_SRCU)) + srcu_online_cpu(cpu); return 0; } +/* + * Near the beginning of the process. The CPU is still very much alive + * with pretty much all services enabled. + */ int rcutree_offline_cpu(unsigned int cpu) { rcutree_affinity_setting(cpu, cpu); + if (IS_ENABLED(CONFIG_TREE_SRCU)) + srcu_offline_cpu(cpu); return 0; } - +/* + * Near the end of the offline process. We do only tracing here. + */ int rcutree_dying_cpu(unsigned int cpu) { struct rcu_state *rsp; @@ -3845,6 +3853,9 @@ int rcutree_dying_cpu(unsigned int cpu) return 0; } +/* + * The outgoing CPU is gone and we are running elsewhere. + */ int rcutree_dead_cpu(unsigned int cpu) { struct rcu_state *rsp; @@ -3862,6 +3873,10 @@ int rcutree_dead_cpu(unsigned int cpu) * incoming CPUs are not allowed to use RCU read-side critical sections * until this function is called. Failing to observe this restriction * will result in lockdep splats. + * + * Note that this function is special in that it is invoked directly + * from the incoming CPU rather than from the cpuhp_step mechanism. + * This is because this function must be invoked at a precise location. */ void rcu_cpu_starting(unsigned int cpu) { @@ -3872,7 +3887,7 @@ void rcu_cpu_starting(unsigned int cpu) struct rcu_state *rsp; for_each_rcu_flavor(rsp) { - rdp = this_cpu_ptr(rsp->rda); + rdp = per_cpu_ptr(rsp->rda, cpu); rnp = rdp->mynode; mask = rdp->grpmask; raw_spin_lock_irqsave_rcu_node(rnp, flags); @@ -3887,9 +3902,6 @@ void rcu_cpu_starting(unsigned int cpu) * The CPU is exiting the idle loop into the arch_cpu_idle_dead() * function. We now remove it from the rcu_node tree's ->qsmaskinit * bit masks. - * The CPU is exiting the idle loop into the arch_cpu_idle_dead() - * function. We now remove it from the rcu_node tree's ->qsmaskinit - * bit masks. */ static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp) { @@ -3905,6 +3917,14 @@ static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp) raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } +/* + * The outgoing function has no further need of RCU, so remove it from + * the list of CPUs that RCU must track. + * + * Note that this function is special in that it is invoked directly + * from the outgoing CPU rather than from the cpuhp_step mechanism. + * This is because this function must be invoked at a precise location. + */ void rcu_report_dead(unsigned int cpu) { struct rcu_state *rsp; @@ -3919,6 +3939,10 @@ void rcu_report_dead(unsigned int cpu) } #endif +/* + * On non-huge systems, use expedited RCU grace periods to make suspend + * and hibernation run faster. + */ static int rcu_pm_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -3989,7 +4013,7 @@ early_initcall(rcu_spawn_gp_kthread); * task is booting the system, and such primitives are no-ops). After this * function is called, any synchronous grace-period primitives are run as * expedited, with the requesting task driving the grace period forward. - * A later core_initcall() rcu_exp_runtime_mode() will switch to full + * A later core_initcall() rcu_set_runtime_mode() will switch to full * runtime RCU functionality. */ void rcu_scheduler_starting(void) @@ -4002,31 +4026,6 @@ void rcu_scheduler_starting(void) } /* - * Compute the per-level fanout, either using the exact fanout specified - * or balancing the tree, depending on the rcu_fanout_exact boot parameter. - */ -static void __init rcu_init_levelspread(int *levelspread, const int *levelcnt) -{ - int i; - - if (rcu_fanout_exact) { - levelspread[rcu_num_lvls - 1] = rcu_fanout_leaf; - for (i = rcu_num_lvls - 2; i >= 0; i--) - levelspread[i] = RCU_FANOUT; - } else { - int ccur; - int cprv; - - cprv = nr_cpu_ids; - for (i = rcu_num_lvls - 1; i >= 0; i--) { - ccur = levelcnt[i]; - levelspread[i] = (cprv + ccur - 1) / ccur; - cprv = ccur; - } - } -} - -/* * Helper function for rcu_init() that initializes one rcu_state structure. */ static void __init rcu_init_one(struct rcu_state *rsp) @@ -4035,9 +4034,7 @@ static void __init rcu_init_one(struct rcu_state *rsp) static const char * const fqs[] = RCU_FQS_NAME_INIT; static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS]; - static u8 fl_mask = 0x1; - int levelcnt[RCU_NUM_LVLS]; /* # nodes in each level. */ int levelspread[RCU_NUM_LVLS]; /* kids/node in each level. */ int cpustride = 1; int i; @@ -4052,20 +4049,16 @@ static void __init rcu_init_one(struct rcu_state *rsp) /* Initialize the level-tracking arrays. */ - for (i = 0; i < rcu_num_lvls; i++) - levelcnt[i] = num_rcu_lvl[i]; for (i = 1; i < rcu_num_lvls; i++) - rsp->level[i] = rsp->level[i - 1] + levelcnt[i - 1]; - rcu_init_levelspread(levelspread, levelcnt); - rsp->flavor_mask = fl_mask; - fl_mask <<= 1; + rsp->level[i] = rsp->level[i - 1] + num_rcu_lvl[i - 1]; + rcu_init_levelspread(levelspread, num_rcu_lvl); /* Initialize the elements themselves, starting from the leaves. */ for (i = rcu_num_lvls - 1; i >= 0; i--) { cpustride *= levelspread[i]; rnp = rsp->level[i]; - for (j = 0; j < levelcnt[i]; j++, rnp++) { + for (j = 0; j < num_rcu_lvl[i]; j++, rnp++) { raw_spin_lock_init(&ACCESS_PRIVATE(rnp, lock)); lockdep_set_class_and_name(&ACCESS_PRIVATE(rnp, lock), &rcu_node_class[i], buf[i]); @@ -4238,6 +4231,8 @@ void __init rcu_init(void) for_each_online_cpu(cpu) { rcutree_prepare_cpu(cpu); rcu_cpu_starting(cpu); + if (IS_ENABLED(CONFIG_TREE_SRCU)) + srcu_online_cpu(cpu); } } diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index fe98dd24adf8..9af0f31d6847 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -24,85 +24,15 @@ #include <linux/cache.h> #include <linux/spinlock.h> +#include <linux/rtmutex.h> #include <linux/threads.h> #include <linux/cpumask.h> #include <linux/seqlock.h> #include <linux/swait.h> #include <linux/stop_machine.h> +#include <linux/rcu_node_tree.h> -/* - * Define shape of hierarchy based on NR_CPUS, CONFIG_RCU_FANOUT, and - * CONFIG_RCU_FANOUT_LEAF. - * In theory, it should be possible to add more levels straightforwardly. - * In practice, this did work well going from three levels to four. - * Of course, your mileage may vary. - */ - -#ifdef CONFIG_RCU_FANOUT -#define RCU_FANOUT CONFIG_RCU_FANOUT -#else /* #ifdef CONFIG_RCU_FANOUT */ -# ifdef CONFIG_64BIT -# define RCU_FANOUT 64 -# else -# define RCU_FANOUT 32 -# endif -#endif /* #else #ifdef CONFIG_RCU_FANOUT */ - -#ifdef CONFIG_RCU_FANOUT_LEAF -#define RCU_FANOUT_LEAF CONFIG_RCU_FANOUT_LEAF -#else /* #ifdef CONFIG_RCU_FANOUT_LEAF */ -# ifdef CONFIG_64BIT -# define RCU_FANOUT_LEAF 64 -# else -# define RCU_FANOUT_LEAF 32 -# endif -#endif /* #else #ifdef CONFIG_RCU_FANOUT_LEAF */ - -#define RCU_FANOUT_1 (RCU_FANOUT_LEAF) -#define RCU_FANOUT_2 (RCU_FANOUT_1 * RCU_FANOUT) -#define RCU_FANOUT_3 (RCU_FANOUT_2 * RCU_FANOUT) -#define RCU_FANOUT_4 (RCU_FANOUT_3 * RCU_FANOUT) - -#if NR_CPUS <= RCU_FANOUT_1 -# define RCU_NUM_LVLS 1 -# define NUM_RCU_LVL_0 1 -# define NUM_RCU_NODES NUM_RCU_LVL_0 -# define NUM_RCU_LVL_INIT { NUM_RCU_LVL_0 } -# define RCU_NODE_NAME_INIT { "rcu_node_0" } -# define RCU_FQS_NAME_INIT { "rcu_node_fqs_0" } -#elif NR_CPUS <= RCU_FANOUT_2 -# define RCU_NUM_LVLS 2 -# define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) -# define NUM_RCU_NODES (NUM_RCU_LVL_0 + NUM_RCU_LVL_1) -# define NUM_RCU_LVL_INIT { NUM_RCU_LVL_0, NUM_RCU_LVL_1 } -# define RCU_NODE_NAME_INIT { "rcu_node_0", "rcu_node_1" } -# define RCU_FQS_NAME_INIT { "rcu_node_fqs_0", "rcu_node_fqs_1" } -#elif NR_CPUS <= RCU_FANOUT_3 -# define RCU_NUM_LVLS 3 -# define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) -# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) -# define NUM_RCU_NODES (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2) -# define NUM_RCU_LVL_INIT { NUM_RCU_LVL_0, NUM_RCU_LVL_1, NUM_RCU_LVL_2 } -# define RCU_NODE_NAME_INIT { "rcu_node_0", "rcu_node_1", "rcu_node_2" } -# define RCU_FQS_NAME_INIT { "rcu_node_fqs_0", "rcu_node_fqs_1", "rcu_node_fqs_2" } -#elif NR_CPUS <= RCU_FANOUT_4 -# define RCU_NUM_LVLS 4 -# define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_3) -# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) -# define NUM_RCU_LVL_3 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) -# define NUM_RCU_NODES (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3) -# define NUM_RCU_LVL_INIT { NUM_RCU_LVL_0, NUM_RCU_LVL_1, NUM_RCU_LVL_2, NUM_RCU_LVL_3 } -# define RCU_NODE_NAME_INIT { "rcu_node_0", "rcu_node_1", "rcu_node_2", "rcu_node_3" } -# define RCU_FQS_NAME_INIT { "rcu_node_fqs_0", "rcu_node_fqs_1", "rcu_node_fqs_2", "rcu_node_fqs_3" } -#else -# error "CONFIG_RCU_FANOUT insufficient for NR_CPUS" -#endif /* #if (NR_CPUS) <= RCU_FANOUT_1 */ - -extern int rcu_num_lvls; -extern int rcu_num_nodes; +#include "rcu_segcblist.h" /* * Dynticks per-CPU state. @@ -112,14 +42,9 @@ struct rcu_dynticks { /* Process level is worth LLONG_MAX/2. */ int dynticks_nmi_nesting; /* Track NMI nesting level. */ atomic_t dynticks; /* Even value for idle, else odd. */ -#ifdef CONFIG_NO_HZ_FULL_SYSIDLE - long long dynticks_idle_nesting; - /* irq/process nesting level from idle. */ - atomic_t dynticks_idle; /* Even value for idle, else odd. */ - /* "Idle" excludes userspace execution. */ - unsigned long dynticks_idle_jiffies; - /* End of last non-NMI non-idle period. */ -#endif /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ + bool rcu_need_heavy_qs; /* GP old, need heavy quiescent state. */ + unsigned long rcu_qs_ctr; /* Light universal quiescent state ctr. */ + bool rcu_urgent_qs; /* GP old need light quiescent state. */ #ifdef CONFIG_RCU_FAST_NO_HZ bool all_lazy; /* Are all CPU's CBs lazy? */ unsigned long nonlazy_posted; @@ -227,19 +152,6 @@ struct rcu_node { /* Number of tasks boosted for expedited GP. */ unsigned long n_normal_boosts; /* Number of tasks boosted for normal GP. */ - unsigned long n_balk_blkd_tasks; - /* Refused to boost: no blocked tasks. */ - unsigned long n_balk_exp_gp_tasks; - /* Refused to boost: nothing blocking GP. */ - unsigned long n_balk_boost_tasks; - /* Refused to boost: already boosting. */ - unsigned long n_balk_notblocked; - /* Refused to boost: RCU RS CS still running. */ - unsigned long n_balk_notyet; - /* Refused to boost: not yet time. */ - unsigned long n_balk_nos; - /* Refused to boost: not sure why, though. */ - /* This can happen due to race conditions. */ #ifdef CONFIG_RCU_NOCB_CPU struct swait_queue_head nocb_gp_wq[2]; /* Place for rcu_nocb_kthread() to wait GP. */ @@ -261,41 +173,6 @@ struct rcu_node { #define leaf_node_cpu_bit(rnp, cpu) (1UL << ((cpu) - (rnp)->grplo)) /* - * Do a full breadth-first scan of the rcu_node structures for the - * specified rcu_state structure. - */ -#define rcu_for_each_node_breadth_first(rsp, rnp) \ - for ((rnp) = &(rsp)->node[0]; \ - (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++) - -/* - * Do a breadth-first scan of the non-leaf rcu_node structures for the - * specified rcu_state structure. Note that if there is a singleton - * rcu_node tree with but one rcu_node structure, this loop is a no-op. - */ -#define rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) \ - for ((rnp) = &(rsp)->node[0]; \ - (rnp) < (rsp)->level[rcu_num_lvls - 1]; (rnp)++) - -/* - * Scan the leaves of the rcu_node hierarchy for the specified rcu_state - * structure. Note that if there is a singleton rcu_node tree with but - * one rcu_node structure, this loop -will- visit the rcu_node structure. - * It is still a leaf node, even if it is also the root node. - */ -#define rcu_for_each_leaf_node(rsp, rnp) \ - for ((rnp) = (rsp)->level[rcu_num_lvls - 1]; \ - (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++) - -/* - * Iterate over all possible CPUs in a leaf RCU node. - */ -#define for_each_leaf_node_possible_cpu(rnp, cpu) \ - for ((cpu) = cpumask_next(rnp->grplo - 1, cpu_possible_mask); \ - cpu <= rnp->grphi; \ - cpu = cpumask_next((cpu), cpu_possible_mask)) - -/* * Union to allow "aggregate OR" operation on the need for a quiescent * state by the normal and expedited grace periods. */ @@ -335,34 +212,9 @@ struct rcu_data { /* period it is aware of. */ /* 2) batch handling */ - /* - * If nxtlist is not NULL, it is partitioned as follows. - * Any of the partitions might be empty, in which case the - * pointer to that partition will be equal to the pointer for - * the following partition. When the list is empty, all of - * the nxttail elements point to the ->nxtlist pointer itself, - * which in that case is NULL. - * - * [nxtlist, *nxttail[RCU_DONE_TAIL]): - * Entries that batch # <= ->completed - * The grace period for these entries has completed, and - * the other grace-period-completed entries may be moved - * here temporarily in rcu_process_callbacks(). - * [*nxttail[RCU_DONE_TAIL], *nxttail[RCU_WAIT_TAIL]): - * Entries that batch # <= ->completed - 1: waiting for current GP - * [*nxttail[RCU_WAIT_TAIL], *nxttail[RCU_NEXT_READY_TAIL]): - * Entries known to have arrived before current GP ended - * [*nxttail[RCU_NEXT_READY_TAIL], *nxttail[RCU_NEXT_TAIL]): - * Entries that might have arrived after current GP ended - * Note that the value of *nxttail[RCU_NEXT_TAIL] will - * always be NULL, as this is the end of the list. - */ - struct rcu_head *nxtlist; - struct rcu_head **nxttail[RCU_NEXT_SIZE]; - unsigned long nxtcompleted[RCU_NEXT_SIZE]; - /* grace periods for sublists. */ - long qlen_lazy; /* # of lazy queued callbacks */ - long qlen; /* # of queued callbacks, incl lazy */ + struct rcu_segcblist cblist; /* Segmented callback list, with */ + /* different callbacks waiting for */ + /* different grace periods. */ long qlen_last_fqs_check; /* qlen at last check for QS forcing */ unsigned long n_cbs_invoked; /* count of RCU cbs invoked. */ @@ -439,9 +291,9 @@ struct rcu_data { }; /* Values for nocb_defer_wakeup field in struct rcu_data. */ -#define RCU_NOGP_WAKE_NOT 0 -#define RCU_NOGP_WAKE 1 -#define RCU_NOGP_WAKE_FORCE 2 +#define RCU_NOCB_WAKE_NOT 0 +#define RCU_NOCB_WAKE 1 +#define RCU_NOCB_WAKE_FORCE 2 #define RCU_JIFFIES_TILL_FORCE_QS (1 + (HZ > 250) + (HZ > 500)) /* For jiffies_till_first_fqs and */ @@ -481,7 +333,6 @@ struct rcu_state { struct rcu_node *level[RCU_NUM_LVLS + 1]; /* Hierarchy levels (+1 to */ /* shut bogus gcc warning) */ - u8 flavor_mask; /* bit in flavor mask. */ struct rcu_data __percpu *rda; /* pointer of percu rcu_data. */ call_rcu_func_t call; /* call_rcu() flavor. */ int ncpus; /* # CPUs seen so far. */ @@ -501,14 +352,11 @@ struct rcu_state { raw_spinlock_t orphan_lock ____cacheline_internodealigned_in_smp; /* Protect following fields. */ - struct rcu_head *orphan_nxtlist; /* Orphaned callbacks that */ + struct rcu_cblist orphan_pend; /* Orphaned callbacks that */ /* need a grace period. */ - struct rcu_head **orphan_nxttail; /* Tail of above. */ - struct rcu_head *orphan_donelist; /* Orphaned callbacks that */ + struct rcu_cblist orphan_done; /* Orphaned callbacks that */ /* are ready to invoke. */ - struct rcu_head **orphan_donetail; /* Tail of above. */ - long qlen_lazy; /* Number of lazy callbacks. */ - long qlen; /* Total number of callbacks. */ + /* (Contains counts.) */ /* End of fields guarded by orphan_lock. */ struct mutex barrier_mutex; /* Guards barrier fields. */ @@ -521,7 +369,6 @@ struct rcu_state { struct mutex exp_mutex; /* Serialize expedited GP. */ struct mutex exp_wake_mutex; /* Serialize wakeup. */ unsigned long expedited_sequence; /* Take a ticket. */ - atomic_long_t expedited_normal; /* # fallbacks to normal. */ atomic_t expedited_need_qs; /* # CPUs left to check in. */ struct swait_queue_head expedited_wq; /* Wait for check-ins. */ int ncpus_snap; /* # CPUs seen last time. */ @@ -595,6 +442,9 @@ extern struct rcu_state rcu_bh_state; extern struct rcu_state rcu_preempt_state; #endif /* #ifdef CONFIG_PREEMPT_RCU */ +int rcu_dynticks_snap(struct rcu_dynticks *rdtp); +bool rcu_eqs_special_set(int cpu); + #ifdef CONFIG_RCU_BOOST DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status); DECLARE_PER_CPU(int, rcu_cpu_kthread_cpu); @@ -606,7 +456,7 @@ DECLARE_PER_CPU(char, rcu_cpu_has_work); /* Forward declarations for rcutree_plugin.h */ static void rcu_bootup_announce(void); -static void rcu_preempt_note_context_switch(void); +static void rcu_preempt_note_context_switch(bool preempt); static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp); #ifdef CONFIG_HOTPLUG_CPU static bool rcu_preempt_has_tasks(struct rcu_node *rnp); @@ -658,101 +508,17 @@ static void __init rcu_organize_nocb_kthreads(struct rcu_state *rsp); #endif /* #ifdef CONFIG_RCU_NOCB_CPU */ static void __maybe_unused rcu_kick_nohz_cpu(int cpu); static bool init_nocb_callback_list(struct rcu_data *rdp); -static void rcu_sysidle_enter(int irq); -static void rcu_sysidle_exit(int irq); -static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, - unsigned long *maxj); -static bool is_sysidle_rcu_state(struct rcu_state *rsp); -static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, - unsigned long maxj); static void rcu_bind_gp_kthread(void); -static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp); static bool rcu_nohz_full_cpu(struct rcu_state *rsp); static void rcu_dynticks_task_enter(void); static void rcu_dynticks_task_exit(void); -#endif /* #ifndef RCU_TREE_NONCORE */ +#ifdef CONFIG_SRCU +void srcu_online_cpu(unsigned int cpu); +void srcu_offline_cpu(unsigned int cpu); +#else /* #ifdef CONFIG_SRCU */ +void srcu_online_cpu(unsigned int cpu) { } +void srcu_offline_cpu(unsigned int cpu) { } +#endif /* #else #ifdef CONFIG_SRCU */ -#ifdef CONFIG_RCU_TRACE -/* Read out queue lengths for tracing. */ -static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll) -{ -#ifdef CONFIG_RCU_NOCB_CPU - *ql = atomic_long_read(&rdp->nocb_q_count); - *qll = atomic_long_read(&rdp->nocb_q_count_lazy); -#else /* #ifdef CONFIG_RCU_NOCB_CPU */ - *ql = 0; - *qll = 0; -#endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */ -} -#endif /* #ifdef CONFIG_RCU_TRACE */ - -/* - * Place this after a lock-acquisition primitive to guarantee that - * an UNLOCK+LOCK pair act as a full barrier. This guarantee applies - * if the UNLOCK and LOCK are executed by the same CPU or if the - * UNLOCK and LOCK operate on the same lock variable. - */ -#ifdef CONFIG_PPC -#define smp_mb__after_unlock_lock() smp_mb() /* Full ordering for lock. */ -#else /* #ifdef CONFIG_PPC */ -#define smp_mb__after_unlock_lock() do { } while (0) -#endif /* #else #ifdef CONFIG_PPC */ - -/* - * Wrappers for the rcu_node::lock acquire and release. - * - * Because the rcu_nodes form a tree, the tree traversal locking will observe - * different lock values, this in turn means that an UNLOCK of one level - * followed by a LOCK of another level does not imply a full memory barrier; - * and most importantly transitivity is lost. - * - * In order to restore full ordering between tree levels, augment the regular - * lock acquire functions with smp_mb__after_unlock_lock(). - * - * As ->lock of struct rcu_node is a __private field, therefore one should use - * these wrappers rather than directly call raw_spin_{lock,unlock}* on ->lock. - */ -static inline void raw_spin_lock_rcu_node(struct rcu_node *rnp) -{ - raw_spin_lock(&ACCESS_PRIVATE(rnp, lock)); - smp_mb__after_unlock_lock(); -} - -static inline void raw_spin_unlock_rcu_node(struct rcu_node *rnp) -{ - raw_spin_unlock(&ACCESS_PRIVATE(rnp, lock)); -} - -static inline void raw_spin_lock_irq_rcu_node(struct rcu_node *rnp) -{ - raw_spin_lock_irq(&ACCESS_PRIVATE(rnp, lock)); - smp_mb__after_unlock_lock(); -} - -static inline void raw_spin_unlock_irq_rcu_node(struct rcu_node *rnp) -{ - raw_spin_unlock_irq(&ACCESS_PRIVATE(rnp, lock)); -} - -#define raw_spin_lock_irqsave_rcu_node(rnp, flags) \ -do { \ - typecheck(unsigned long, flags); \ - raw_spin_lock_irqsave(&ACCESS_PRIVATE(rnp, lock), flags); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define raw_spin_unlock_irqrestore_rcu_node(rnp, flags) \ -do { \ - typecheck(unsigned long, flags); \ - raw_spin_unlock_irqrestore(&ACCESS_PRIVATE(rnp, lock), flags); \ -} while (0) - -static inline bool raw_spin_trylock_rcu_node(struct rcu_node *rnp) -{ - bool locked = raw_spin_trylock(&ACCESS_PRIVATE(rnp, lock)); - - if (locked) - smp_mb__after_unlock_lock(); - return locked; -} +#endif /* #ifndef RCU_TREE_NONCORE */ diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h index e59e1849b89a..dd21ca47e4b4 100644 --- a/kernel/rcu/tree_exp.h +++ b/kernel/rcu/tree_exp.h @@ -20,16 +20,26 @@ * Authors: Paul E. McKenney <[email protected]> */ -/* Wrapper functions for expedited grace periods. */ +/* + * Record the start of an expedited grace period. + */ static void rcu_exp_gp_seq_start(struct rcu_state *rsp) { rcu_seq_start(&rsp->expedited_sequence); } + +/* + * Record the end of an expedited grace period. + */ static void rcu_exp_gp_seq_end(struct rcu_state *rsp) { rcu_seq_end(&rsp->expedited_sequence); smp_mb(); /* Ensure that consecutive grace periods serialize. */ } + +/* + * Take a snapshot of the expedited-grace-period counter. + */ static unsigned long rcu_exp_gp_seq_snap(struct rcu_state *rsp) { unsigned long s; @@ -39,6 +49,12 @@ static unsigned long rcu_exp_gp_seq_snap(struct rcu_state *rsp) trace_rcu_exp_grace_period(rsp->name, s, TPS("snap")); return s; } + +/* + * Given a counter snapshot from rcu_exp_gp_seq_snap(), return true + * if a full expedited grace period has elapsed since that snapshot + * was taken. + */ static bool rcu_exp_gp_seq_done(struct rcu_state *rsp, unsigned long s) { return rcu_seq_done(&rsp->expedited_sequence, s); @@ -131,7 +147,7 @@ static void __maybe_unused sync_exp_reset_tree(struct rcu_state *rsp) * * Caller must hold the rcu_state's exp_mutex. */ -static int sync_rcu_preempt_exp_done(struct rcu_node *rnp) +static bool sync_rcu_preempt_exp_done(struct rcu_node *rnp) { return rnp->exp_tasks == NULL && READ_ONCE(rnp->expmask) == 0; @@ -276,7 +292,7 @@ static bool exp_funnel_lock(struct rcu_state *rsp, unsigned long s) trace_rcu_exp_funnel_lock(rsp->name, rnp->level, rnp->grplo, rnp->grphi, TPS("wait")); - wait_event(rnp->exp_wq[(s >> 1) & 0x3], + wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3], sync_exp_work_done(rsp, &rdp->exp_workdone2, s)); return true; @@ -315,6 +331,8 @@ static void sync_sched_exp_handler(void *data) return; } __this_cpu_write(rcu_sched_data.cpu_no_qs.b.exp, true); + /* Store .exp before .rcu_urgent_qs. */ + smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true); resched_cpu(smp_processor_id()); } @@ -356,12 +374,11 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp, mask_ofl_test = 0; for_each_leaf_node_possible_cpu(rnp, cpu) { struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); - struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu); rdp->exp_dynticks_snap = - atomic_add_return(0, &rdtp->dynticks); + rcu_dynticks_snap(rdp->dynticks); if (raw_smp_processor_id() == cpu || - !(rdp->exp_dynticks_snap & 0x1) || + rcu_dynticks_in_eqs(rdp->exp_dynticks_snap) || !(rnp->qsmaskinitnext & rdp->grpmask)) mask_ofl_test |= rdp->grpmask; } @@ -380,13 +397,12 @@ static void sync_rcu_exp_select_cpus(struct rcu_state *rsp, for_each_leaf_node_possible_cpu(rnp, cpu) { unsigned long mask = leaf_node_cpu_bit(rnp, cpu); struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); - struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu); if (!(mask_ofl_ipi & mask)) continue; retry_ipi: - if (atomic_add_return(0, &rdtp->dynticks) != - rdp->exp_dynticks_snap) { + if (rcu_dynticks_in_eqs_since(rdp->dynticks, + rdp->exp_dynticks_snap)) { mask_ofl_test |= mask; continue; } @@ -517,7 +533,8 @@ static void rcu_exp_wait_wake(struct rcu_state *rsp, unsigned long s) rnp->exp_seq_rq = s; spin_unlock(&rnp->exp_lock); } - wake_up_all(&rnp->exp_wq[(rsp->expedited_sequence >> 1) & 0x3]); + smp_mb(); /* All above changes before wakeup. */ + wake_up_all(&rnp->exp_wq[rcu_seq_ctr(rsp->expedited_sequence) & 0x3]); } trace_rcu_exp_grace_period(rsp->name, s, TPS("endwake")); mutex_unlock(&rsp->exp_wake_mutex); @@ -595,9 +612,9 @@ static void _synchronize_rcu_expedited(struct rcu_state *rsp, /* Wait for expedited grace period to complete. */ rdp = per_cpu_ptr(rsp->rda, raw_smp_processor_id()); rnp = rcu_get_root(rsp); - wait_event(rnp->exp_wq[(s >> 1) & 0x3], - sync_exp_work_done(rsp, - &rdp->exp_workdone0, s)); + wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3], + sync_exp_work_done(rsp, &rdp->exp_workdone0, s)); + smp_mb(); /* Workqueue actions happen before return. */ /* Let the next expedited grace period start. */ mutex_unlock(&rsp->exp_mutex); @@ -623,6 +640,11 @@ void synchronize_sched_expedited(void) { struct rcu_state *rsp = &rcu_sched_state; + RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) || + lock_is_held(&rcu_lock_map) || + lock_is_held(&rcu_sched_lock_map), + "Illegal synchronize_sched_expedited() in RCU read-side critical section"); + /* If only one CPU, this is automatically a grace period. */ if (rcu_blocking_is_gp()) return; @@ -692,6 +714,11 @@ void synchronize_rcu_expedited(void) { struct rcu_state *rsp = rcu_state_p; + RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) || + lock_is_held(&rcu_lock_map) || + lock_is_held(&rcu_sched_lock_map), + "Illegal synchronize_rcu_expedited() in RCU read-side critical section"); + if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE) return; _synchronize_rcu_expedited(rsp, sync_rcu_exp_handler); @@ -711,15 +738,3 @@ void synchronize_rcu_expedited(void) EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); #endif /* #else #ifdef CONFIG_PREEMPT_RCU */ - -/* - * Switch to run-time mode once Tree RCU has fully initialized. - */ -static int __init rcu_exp_runtime_mode(void) -{ - rcu_test_sync_prims(); - rcu_scheduler_active = RCU_SCHEDULER_RUNNING; - rcu_test_sync_prims(); - return 0; -} -core_initcall(rcu_exp_runtime_mode); diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 56583e764ebf..908b309d60d7 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -27,7 +27,9 @@ #include <linux/delay.h> #include <linux/gfp.h> #include <linux/oom.h> +#include <linux/sched/debug.h> #include <linux/smpboot.h> +#include <uapi/linux/sched/types.h> #include "../time/tick-internal.h" #ifdef CONFIG_RCU_BOOST @@ -68,7 +70,7 @@ static bool __read_mostly rcu_nocb_poll; /* Offload kthread are to poll. */ static void __init rcu_bootup_announce_oddness(void) { if (IS_ENABLED(CONFIG_RCU_TRACE)) - pr_info("\tRCU debugfs-based tracing is enabled.\n"); + pr_info("\tRCU event tracing is enabled.\n"); if ((IS_ENABLED(CONFIG_64BIT) && RCU_FANOUT != 64) || (!IS_ENABLED(CONFIG_64BIT) && RCU_FANOUT != 32)) pr_info("\tCONFIG_RCU_FANOUT set to non-default value of %d\n", @@ -88,8 +90,32 @@ static void __init rcu_bootup_announce_oddness(void) pr_info("\tBoot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); if (nr_cpu_ids != NR_CPUS) pr_info("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids); - if (IS_ENABLED(CONFIG_RCU_BOOST)) - pr_info("\tRCU kthread priority: %d.\n", kthread_prio); +#ifdef CONFIG_RCU_BOOST + pr_info("\tRCU priority boosting: priority %d delay %d ms.\n", kthread_prio, CONFIG_RCU_BOOST_DELAY); +#endif + if (blimit != DEFAULT_RCU_BLIMIT) + pr_info("\tBoot-time adjustment of callback invocation limit to %ld.\n", blimit); + if (qhimark != DEFAULT_RCU_QHIMARK) + pr_info("\tBoot-time adjustment of callback high-water mark to %ld.\n", qhimark); + if (qlowmark != DEFAULT_RCU_QLOMARK) + pr_info("\tBoot-time adjustment of callback low-water mark to %ld.\n", qlowmark); + if (jiffies_till_first_fqs != ULONG_MAX) + pr_info("\tBoot-time adjustment of first FQS scan delay to %ld jiffies.\n", jiffies_till_first_fqs); + if (jiffies_till_next_fqs != ULONG_MAX) + pr_info("\tBoot-time adjustment of subsequent FQS scan delay to %ld jiffies.\n", jiffies_till_next_fqs); + if (rcu_kick_kthreads) + pr_info("\tKick kthreads if too-long grace period.\n"); + if (IS_ENABLED(CONFIG_DEBUG_OBJECTS_RCU_HEAD)) + pr_info("\tRCU callback double-/use-after-free debug enabled.\n"); + if (gp_preinit_delay) + pr_info("\tRCU debug GP pre-init slowdown %d jiffies.\n", gp_preinit_delay); + if (gp_init_delay) + pr_info("\tRCU debug GP init slowdown %d jiffies.\n", gp_init_delay); + if (gp_cleanup_delay) + pr_info("\tRCU debug GP init slowdown %d jiffies.\n", gp_cleanup_delay); + if (IS_ENABLED(CONFIG_RCU_EQS_DEBUG)) + pr_info("\tRCU debug extended QS entry/exit.\n"); + rcupdate_announce_bootup_oddness(); } #ifdef CONFIG_PREEMPT_RCU @@ -153,6 +179,8 @@ static void rcu_preempt_ctxt_queue(struct rcu_node *rnp, struct rcu_data *rdp) (rnp->expmask & rdp->grpmask ? RCU_EXP_BLKD : 0); struct task_struct *t = current; + lockdep_assert_held(&rnp->lock); + /* * Decide where to queue the newly blocked task. In theory, * this could be an if-statement. In practice, when I tried @@ -261,6 +289,7 @@ static void rcu_preempt_ctxt_queue(struct rcu_node *rnp, struct rcu_data *rdp) */ static void rcu_preempt_qs(void) { + RCU_LOCKDEP_WARN(preemptible(), "rcu_preempt_qs() invoked with preemption enabled!!!\n"); if (__this_cpu_read(rcu_data_p->cpu_no_qs.s)) { trace_rcu_grace_period(TPS("rcu_preempt"), __this_cpu_read(rcu_data_p->gpnum), @@ -284,12 +313,14 @@ static void rcu_preempt_qs(void) * * Caller must disable interrupts. */ -static void rcu_preempt_note_context_switch(void) +static void rcu_preempt_note_context_switch(bool preempt) { struct task_struct *t = current; struct rcu_data *rdp; struct rcu_node *rnp; + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_preempt_note_context_switch() invoked with interrupts enabled!!!\n"); + WARN_ON_ONCE(!preempt && t->rcu_read_lock_nesting > 0); if (t->rcu_read_lock_nesting > 0 && !t->rcu_read_unlock_special.b.blocked) { @@ -605,6 +636,7 @@ static int rcu_print_task_exp_stall(struct rcu_node *rnp) */ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) { + RCU_LOCKDEP_WARN(preemptible(), "rcu_preempt_check_blocked_tasks() invoked with preemption enabled!!!\n"); WARN_ON_ONCE(rcu_preempt_blocked_readers_cgp(rnp)); if (rcu_preempt_has_tasks(rnp)) rnp->gp_tasks = rnp->blkd_tasks.next; @@ -641,8 +673,37 @@ static void rcu_preempt_do_callbacks(void) #endif /* #ifdef CONFIG_RCU_BOOST */ -/* - * Queue a preemptible-RCU callback for invocation after a grace period. +/** + * call_rcu() - Queue an RCU callback for invocation after a grace period. + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all pre-existing RCU read-side + * critical sections have completed. However, the callback function + * might well execute concurrently with RCU read-side critical sections + * that started after call_rcu() was invoked. RCU read-side critical + * sections are delimited by rcu_read_lock() and rcu_read_unlock(), + * and may be nested. + * + * Note that all CPUs must agree that the grace period extended beyond + * all pre-existing RCU read-side critical section. On systems with more + * than one CPU, this means that when "func()" is invoked, each CPU is + * guaranteed to have executed a full memory barrier since the end of its + * last RCU read-side critical section whose beginning preceded the call + * to call_rcu(). It also means that each CPU executing an RCU read-side + * critical section that continues beyond the start of "func()" must have + * executed a memory barrier after the call_rcu() but before the beginning + * of that RCU read-side critical section. Note that these guarantees + * include CPUs that are offline, idle, or executing in user mode, as + * well as CPUs that are executing in the kernel. + * + * Furthermore, if CPU A invoked call_rcu() and CPU B invoked the + * resulting RCU callback function "func()", then both CPU A and CPU B are + * guaranteed to execute a full memory barrier during the time interval + * between the call to call_rcu() and the invocation of "func()" -- even + * if CPU A and CPU B are the same CPU (but again only if the system has + * more than one CPU). */ void call_rcu(struct rcu_head *head, rcu_callback_t func) { @@ -661,8 +722,13 @@ EXPORT_SYMBOL_GPL(call_rcu); * synchronize_rcu() was waiting. RCU read-side critical sections are * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested. * - * See the description of synchronize_sched() for more detailed information - * on memory ordering guarantees. + * See the description of synchronize_sched() for more detailed + * information on memory-ordering guarantees. However, please note + * that -only- the memory-ordering guarantees apply. For example, + * synchronize_rcu() is -not- guaranteed to wait on things like code + * protected by preempt_disable(), instead, synchronize_rcu() is -only- + * guaranteed to wait on RCU read-side critical sections, that is, sections + * of code protected by rcu_read_lock(). */ void synchronize_rcu(void) { @@ -736,7 +802,7 @@ static void __init rcu_bootup_announce(void) * Because preemptible RCU does not exist, we never have to check for * CPUs being in quiescent states. */ -static void rcu_preempt_note_context_switch(void) +static void rcu_preempt_note_context_switch(bool preempt) { } @@ -833,33 +899,6 @@ void exit_rcu(void) #include "../locking/rtmutex_common.h" -#ifdef CONFIG_RCU_TRACE - -static void rcu_initiate_boost_trace(struct rcu_node *rnp) -{ - if (!rcu_preempt_has_tasks(rnp)) - rnp->n_balk_blkd_tasks++; - else if (rnp->exp_tasks == NULL && rnp->gp_tasks == NULL) - rnp->n_balk_exp_gp_tasks++; - else if (rnp->gp_tasks != NULL && rnp->boost_tasks != NULL) - rnp->n_balk_boost_tasks++; - else if (rnp->gp_tasks != NULL && rnp->qsmask != 0) - rnp->n_balk_notblocked++; - else if (rnp->gp_tasks != NULL && - ULONG_CMP_LT(jiffies, rnp->boost_time)) - rnp->n_balk_notyet++; - else - rnp->n_balk_nos++; -} - -#else /* #ifdef CONFIG_RCU_TRACE */ - -static void rcu_initiate_boost_trace(struct rcu_node *rnp) -{ -} - -#endif /* #else #ifdef CONFIG_RCU_TRACE */ - static void rcu_wake_cond(struct task_struct *t, int status) { /* @@ -990,8 +1029,8 @@ static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags) { struct task_struct *t; + lockdep_assert_held(&rnp->lock); if (!rcu_preempt_blocked_readers_cgp(rnp) && rnp->exp_tasks == NULL) { - rnp->n_balk_exp_gp_tasks++; raw_spin_unlock_irqrestore_rcu_node(rnp, flags); return; } @@ -1007,7 +1046,6 @@ static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags) if (t) rcu_wake_cond(t, rnp->boost_kthread_status); } else { - rcu_initiate_boost_trace(rnp); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } } @@ -1258,8 +1296,7 @@ static void rcu_prepare_kthreads(int cpu) int rcu_needs_cpu(u64 basemono, u64 *nextevt) { *nextevt = KTIME_MAX; - return IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL) - ? 0 : rcu_cpu_has_callbacks(NULL); + return rcu_cpu_has_callbacks(NULL); } /* @@ -1348,10 +1385,10 @@ static bool __maybe_unused rcu_try_advance_all_cbs(void) */ if ((rdp->completed != rnp->completed || unlikely(READ_ONCE(rdp->gpwrap))) && - rdp->nxttail[RCU_DONE_TAIL] != rdp->nxttail[RCU_NEXT_TAIL]) + rcu_segcblist_pend_cbs(&rdp->cblist)) note_gp_changes(rsp, rdp); - if (cpu_has_callbacks_ready_to_invoke(rdp)) + if (rcu_segcblist_ready_cbs(&rdp->cblist)) cbs_ready = true; } return cbs_ready; @@ -1370,10 +1407,7 @@ int rcu_needs_cpu(u64 basemono, u64 *nextevt) struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); unsigned long dj; - if (IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL)) { - *nextevt = KTIME_MAX; - return 0; - } + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_needs_cpu() invoked with irqs enabled!!!"); /* Snapshot to detect later posting of non-lazy callback. */ rdtp->nonlazy_posted_snap = rdtp->nonlazy_posted; @@ -1422,8 +1456,8 @@ static void rcu_prepare_for_idle(void) struct rcu_state *rsp; int tne; - if (IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL) || - rcu_is_nocb_cpu(smp_processor_id())) + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_prepare_for_idle() invoked with irqs enabled!!!"); + if (rcu_is_nocb_cpu(smp_processor_id())) return; /* Handle nohz enablement switches conservatively. */ @@ -1459,7 +1493,7 @@ static void rcu_prepare_for_idle(void) rdtp->last_accelerate = jiffies; for_each_rcu_flavor(rsp) { rdp = this_cpu_ptr(rsp->rda); - if (!*rdp->nxttail[RCU_DONE_TAIL]) + if (rcu_segcblist_pend_cbs(&rdp->cblist)) continue; rnp = rdp->mynode; raw_spin_lock_rcu_node(rnp); /* irqs already disabled. */ @@ -1477,8 +1511,8 @@ static void rcu_prepare_for_idle(void) */ static void rcu_cleanup_after_idle(void) { - if (IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL) || - rcu_is_nocb_cpu(smp_processor_id())) + RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_cleanup_after_idle() invoked with irqs enabled!!!"); + if (rcu_is_nocb_cpu(smp_processor_id())) return; if (rcu_try_advance_all_cbs()) invoke_rcu_core(); @@ -1527,7 +1561,7 @@ static void rcu_oom_notify_cpu(void *unused) for_each_rcu_flavor(rsp) { rdp = raw_cpu_ptr(rsp->rda); - if (rdp->qlen_lazy != 0) { + if (rcu_segcblist_n_lazy_cbs(&rdp->cblist)) { atomic_inc(&oom_callback_count); rsp->call(&rdp->oom_head, rcu_oom_callback); } @@ -1643,7 +1677,7 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu) "o."[!!(rdp->grpmask & rdp->mynode->qsmaskinit)], "N."[!!(rdp->grpmask & rdp->mynode->qsmaskinitnext)], ticks_value, ticks_title, - atomic_read(&rdtp->dynticks) & 0xfff, + rcu_dynticks_snap(rdtp) & 0xfff, rdtp->dynticks_nesting, rdtp->dynticks_nmi_nesting, rdp->softirq_snap, kstat_softirqs_cpu(RCU_SOFTIRQ, cpu), READ_ONCE(rsp->n_force_qs) - rsp->n_force_qs_gpstart, @@ -1707,7 +1741,7 @@ __setup("rcu_nocbs=", rcu_nocb_setup); static int __init parse_rcu_nocb_poll(char *arg) { - rcu_nocb_poll = 1; + rcu_nocb_poll = true; return 0; } early_param("rcu_nocb_poll", parse_rcu_nocb_poll); @@ -1745,7 +1779,6 @@ static void rcu_init_one_nocb(struct rcu_node *rnp) init_swait_queue_head(&rnp->nocb_gp_wq[1]); } -#ifndef CONFIG_RCU_NOCB_CPU_ALL /* Is the specified CPU a no-CBs CPU? */ bool rcu_is_nocb_cpu(int cpu) { @@ -1753,7 +1786,6 @@ bool rcu_is_nocb_cpu(int cpu) return cpumask_test_cpu(cpu, rcu_nocb_mask); return false; } -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_ALL */ /* * Kick the leader kthread for this NOCB group. @@ -1767,6 +1799,7 @@ static void wake_nocb_leader(struct rcu_data *rdp, bool force) if (READ_ONCE(rdp_leader->nocb_leader_sleep) || force) { /* Prior smp_mb__after_atomic() orders against prior enqueue. */ WRITE_ONCE(rdp_leader->nocb_leader_sleep, false); + smp_mb(); /* ->nocb_leader_sleep before swake_up(). */ swake_up(&rdp_leader->nocb_wq); } } @@ -1858,7 +1891,9 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmpty")); } else { - rdp->nocb_defer_wakeup = RCU_NOGP_WAKE; + WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOCB_WAKE); + /* Store ->nocb_defer_wakeup before ->rcu_urgent_qs. */ + smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmptyIsDeferred")); } @@ -1870,7 +1905,9 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeOvf")); } else { - rdp->nocb_defer_wakeup = RCU_NOGP_WAKE_FORCE; + WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOCB_WAKE_FORCE); + /* Store ->nocb_defer_wakeup before ->rcu_urgent_qs. */ + smp_store_release(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs), true); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeOvfIsDeferred")); } @@ -1928,30 +1965,26 @@ static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp, struct rcu_data *rdp, unsigned long flags) { - long ql = rsp->qlen; - long qll = rsp->qlen_lazy; + long ql = rsp->orphan_done.len; + long qll = rsp->orphan_done.len_lazy; /* If this is not a no-CBs CPU, tell the caller to do it the old way. */ if (!rcu_is_nocb_cpu(smp_processor_id())) return false; - rsp->qlen = 0; - rsp->qlen_lazy = 0; /* First, enqueue the donelist, if any. This preserves CB ordering. */ - if (rsp->orphan_donelist != NULL) { - __call_rcu_nocb_enqueue(rdp, rsp->orphan_donelist, - rsp->orphan_donetail, ql, qll, flags); - ql = qll = 0; - rsp->orphan_donelist = NULL; - rsp->orphan_donetail = &rsp->orphan_donelist; + if (rsp->orphan_done.head) { + __call_rcu_nocb_enqueue(rdp, rcu_cblist_head(&rsp->orphan_done), + rcu_cblist_tail(&rsp->orphan_done), + ql, qll, flags); } - if (rsp->orphan_nxtlist != NULL) { - __call_rcu_nocb_enqueue(rdp, rsp->orphan_nxtlist, - rsp->orphan_nxttail, ql, qll, flags); - ql = qll = 0; - rsp->orphan_nxtlist = NULL; - rsp->orphan_nxttail = &rsp->orphan_nxtlist; + if (rsp->orphan_pend.head) { + __call_rcu_nocb_enqueue(rdp, rcu_cblist_head(&rsp->orphan_pend), + rcu_cblist_tail(&rsp->orphan_pend), + ql, qll, flags); } + rcu_cblist_init(&rsp->orphan_done); + rcu_cblist_init(&rsp->orphan_pend); return true; } @@ -2021,6 +2054,7 @@ wait_again: * nocb_gp_head, where they await a grace period. */ gotcbs = false; + smp_mb(); /* wakeup before ->nocb_head reads. */ for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower) { rdp->nocb_gp_head = READ_ONCE(rdp->nocb_head); if (!rdp->nocb_gp_head) @@ -2199,8 +2233,8 @@ static void do_nocb_deferred_wakeup(struct rcu_data *rdp) if (!rcu_nocb_need_deferred_wakeup(rdp)) return; ndw = READ_ONCE(rdp->nocb_defer_wakeup); - WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE_NOT); - wake_nocb_leader(rdp, ndw == RCU_NOGP_WAKE_FORCE); + WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOCB_WAKE_NOT); + wake_nocb_leader(rdp, ndw == RCU_NOCB_WAKE_FORCE); trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWake")); } @@ -2210,10 +2244,6 @@ void __init rcu_init_nohz(void) bool need_rcu_nocb_mask = true; struct rcu_state *rsp; -#ifdef CONFIG_RCU_NOCB_CPU_NONE - need_rcu_nocb_mask = false; -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ - #if defined(CONFIG_NO_HZ_FULL) if (tick_nohz_full_running && cpumask_weight(tick_nohz_full_mask)) need_rcu_nocb_mask = true; @@ -2229,14 +2259,6 @@ void __init rcu_init_nohz(void) if (!have_rcu_nocb_mask) return; -#ifdef CONFIG_RCU_NOCB_CPU_ZERO - pr_info("\tOffload RCU callbacks from CPU 0\n"); - cpumask_set_cpu(0, rcu_nocb_mask); -#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */ -#ifdef CONFIG_RCU_NOCB_CPU_ALL - pr_info("\tOffload RCU callbacks from all CPUs\n"); - cpumask_copy(rcu_nocb_mask, cpu_possible_mask); -#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ #if defined(CONFIG_NO_HZ_FULL) if (tick_nohz_full_running) cpumask_or(rcu_nocb_mask, rcu_nocb_mask, tick_nohz_full_mask); @@ -2366,8 +2388,9 @@ static void __init rcu_organize_nocb_kthreads(struct rcu_state *rsp) } /* - * Each pass through this loop sets up one rcu_data structure and - * spawns one rcu_nocb_kthread(). + * Each pass through this loop sets up one rcu_data structure. + * Should the corresponding CPU come online in the future, then + * we will spawn the needed set of rcu_nocb_kthread() kthreads. */ for_each_cpu(cpu, rcu_nocb_mask) { rdp = per_cpu_ptr(rsp->rda, cpu); @@ -2392,16 +2415,16 @@ static bool init_nocb_callback_list(struct rcu_data *rdp) return false; /* If there are early-boot callbacks, move them to nocb lists. */ - if (rdp->nxtlist) { - rdp->nocb_head = rdp->nxtlist; - rdp->nocb_tail = rdp->nxttail[RCU_NEXT_TAIL]; - atomic_long_set(&rdp->nocb_q_count, rdp->qlen); - atomic_long_set(&rdp->nocb_q_count_lazy, rdp->qlen_lazy); - rdp->nxtlist = NULL; - rdp->qlen = 0; - rdp->qlen_lazy = 0; + if (!rcu_segcblist_empty(&rdp->cblist)) { + rdp->nocb_head = rcu_segcblist_head(&rdp->cblist); + rdp->nocb_tail = rcu_segcblist_tail(&rdp->cblist); + atomic_long_set(&rdp->nocb_q_count, + rcu_segcblist_n_cbs(&rdp->cblist)); + atomic_long_set(&rdp->nocb_q_count_lazy, + rcu_segcblist_n_lazy_cbs(&rdp->cblist)); + rcu_segcblist_init(&rdp->cblist); } - rdp->nxttail[RCU_NEXT_TAIL] = NULL; + rcu_segcblist_disable(&rdp->cblist); return true; } @@ -2488,421 +2511,6 @@ static void __maybe_unused rcu_kick_nohz_cpu(int cpu) #endif /* #ifdef CONFIG_NO_HZ_FULL */ } - -#ifdef CONFIG_NO_HZ_FULL_SYSIDLE - -static int full_sysidle_state; /* Current system-idle state. */ -#define RCU_SYSIDLE_NOT 0 /* Some CPU is not idle. */ -#define RCU_SYSIDLE_SHORT 1 /* All CPUs idle for brief period. */ -#define RCU_SYSIDLE_LONG 2 /* All CPUs idle for long enough. */ -#define RCU_SYSIDLE_FULL 3 /* All CPUs idle, ready for sysidle. */ -#define RCU_SYSIDLE_FULL_NOTED 4 /* Actually entered sysidle state. */ - -/* - * Invoked to note exit from irq or task transition to idle. Note that - * usermode execution does -not- count as idle here! After all, we want - * to detect full-system idle states, not RCU quiescent states and grace - * periods. The caller must have disabled interrupts. - */ -static void rcu_sysidle_enter(int irq) -{ - unsigned long j; - struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); - - /* If there are no nohz_full= CPUs, no need to track this. */ - if (!tick_nohz_full_enabled()) - return; - - /* Adjust nesting, check for fully idle. */ - if (irq) { - rdtp->dynticks_idle_nesting--; - WARN_ON_ONCE(rdtp->dynticks_idle_nesting < 0); - if (rdtp->dynticks_idle_nesting != 0) - return; /* Still not fully idle. */ - } else { - if ((rdtp->dynticks_idle_nesting & DYNTICK_TASK_NEST_MASK) == - DYNTICK_TASK_NEST_VALUE) { - rdtp->dynticks_idle_nesting = 0; - } else { - rdtp->dynticks_idle_nesting -= DYNTICK_TASK_NEST_VALUE; - WARN_ON_ONCE(rdtp->dynticks_idle_nesting < 0); - return; /* Still not fully idle. */ - } - } - - /* Record start of fully idle period. */ - j = jiffies; - WRITE_ONCE(rdtp->dynticks_idle_jiffies, j); - smp_mb__before_atomic(); - atomic_inc(&rdtp->dynticks_idle); - smp_mb__after_atomic(); - WARN_ON_ONCE(atomic_read(&rdtp->dynticks_idle) & 0x1); -} - -/* - * Unconditionally force exit from full system-idle state. This is - * invoked when a normal CPU exits idle, but must be called separately - * for the timekeeping CPU (tick_do_timer_cpu). The reason for this - * is that the timekeeping CPU is permitted to take scheduling-clock - * interrupts while the system is in system-idle state, and of course - * rcu_sysidle_exit() has no way of distinguishing a scheduling-clock - * interrupt from any other type of interrupt. - */ -void rcu_sysidle_force_exit(void) -{ - int oldstate = READ_ONCE(full_sysidle_state); - int newoldstate; - - /* - * Each pass through the following loop attempts to exit full - * system-idle state. If contention proves to be a problem, - * a trylock-based contention tree could be used here. - */ - while (oldstate > RCU_SYSIDLE_SHORT) { - newoldstate = cmpxchg(&full_sysidle_state, - oldstate, RCU_SYSIDLE_NOT); - if (oldstate == newoldstate && - oldstate == RCU_SYSIDLE_FULL_NOTED) { - rcu_kick_nohz_cpu(tick_do_timer_cpu); - return; /* We cleared it, done! */ - } - oldstate = newoldstate; - } - smp_mb(); /* Order initial oldstate fetch vs. later non-idle work. */ -} - -/* - * Invoked to note entry to irq or task transition from idle. Note that - * usermode execution does -not- count as idle here! The caller must - * have disabled interrupts. - */ -static void rcu_sysidle_exit(int irq) -{ - struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks); - - /* If there are no nohz_full= CPUs, no need to track this. */ - if (!tick_nohz_full_enabled()) - return; - - /* Adjust nesting, check for already non-idle. */ - if (irq) { - rdtp->dynticks_idle_nesting++; - WARN_ON_ONCE(rdtp->dynticks_idle_nesting <= 0); - if (rdtp->dynticks_idle_nesting != 1) - return; /* Already non-idle. */ - } else { - /* - * Allow for irq misnesting. Yes, it really is possible - * to enter an irq handler then never leave it, and maybe - * also vice versa. Handle both possibilities. - */ - if (rdtp->dynticks_idle_nesting & DYNTICK_TASK_NEST_MASK) { - rdtp->dynticks_idle_nesting += DYNTICK_TASK_NEST_VALUE; - WARN_ON_ONCE(rdtp->dynticks_idle_nesting <= 0); - return; /* Already non-idle. */ - } else { - rdtp->dynticks_idle_nesting = DYNTICK_TASK_EXIT_IDLE; - } - } - - /* Record end of idle period. */ - smp_mb__before_atomic(); - atomic_inc(&rdtp->dynticks_idle); - smp_mb__after_atomic(); - WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks_idle) & 0x1)); - - /* - * If we are the timekeeping CPU, we are permitted to be non-idle - * during a system-idle state. This must be the case, because - * the timekeeping CPU has to take scheduling-clock interrupts - * during the time that the system is transitioning to full - * system-idle state. This means that the timekeeping CPU must - * invoke rcu_sysidle_force_exit() directly if it does anything - * more than take a scheduling-clock interrupt. - */ - if (smp_processor_id() == tick_do_timer_cpu) - return; - - /* Update system-idle state: We are clearly no longer fully idle! */ - rcu_sysidle_force_exit(); -} - -/* - * Check to see if the current CPU is idle. Note that usermode execution - * does not count as idle. The caller must have disabled interrupts, - * and must be running on tick_do_timer_cpu. - */ -static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, - unsigned long *maxj) -{ - int cur; - unsigned long j; - struct rcu_dynticks *rdtp = rdp->dynticks; - - /* If there are no nohz_full= CPUs, don't check system-wide idleness. */ - if (!tick_nohz_full_enabled()) - return; - - /* - * If some other CPU has already reported non-idle, if this is - * not the flavor of RCU that tracks sysidle state, or if this - * is an offline or the timekeeping CPU, nothing to do. - */ - if (!*isidle || rdp->rsp != rcu_state_p || - cpu_is_offline(rdp->cpu) || rdp->cpu == tick_do_timer_cpu) - return; - /* Verify affinity of current kthread. */ - WARN_ON_ONCE(smp_processor_id() != tick_do_timer_cpu); - - /* Pick up current idle and NMI-nesting counter and check. */ - cur = atomic_read(&rdtp->dynticks_idle); - if (cur & 0x1) { - *isidle = false; /* We are not idle! */ - return; - } - smp_mb(); /* Read counters before timestamps. */ - - /* Pick up timestamps. */ - j = READ_ONCE(rdtp->dynticks_idle_jiffies); - /* If this CPU entered idle more recently, update maxj timestamp. */ - if (ULONG_CMP_LT(*maxj, j)) - *maxj = j; -} - -/* - * Is this the flavor of RCU that is handling full-system idle? - */ -static bool is_sysidle_rcu_state(struct rcu_state *rsp) -{ - return rsp == rcu_state_p; -} - -/* - * Return a delay in jiffies based on the number of CPUs, rcu_node - * leaf fanout, and jiffies tick rate. The idea is to allow larger - * systems more time to transition to full-idle state in order to - * avoid the cache thrashing that otherwise occur on the state variable. - * Really small systems (less than a couple of tens of CPUs) should - * instead use a single global atomically incremented counter, and later - * versions of this will automatically reconfigure themselves accordingly. - */ -static unsigned long rcu_sysidle_delay(void) -{ - if (nr_cpu_ids <= CONFIG_NO_HZ_FULL_SYSIDLE_SMALL) - return 0; - return DIV_ROUND_UP(nr_cpu_ids * HZ, rcu_fanout_leaf * 1000); -} - -/* - * Advance the full-system-idle state. This is invoked when all of - * the non-timekeeping CPUs are idle. - */ -static void rcu_sysidle(unsigned long j) -{ - /* Check the current state. */ - switch (READ_ONCE(full_sysidle_state)) { - case RCU_SYSIDLE_NOT: - - /* First time all are idle, so note a short idle period. */ - WRITE_ONCE(full_sysidle_state, RCU_SYSIDLE_SHORT); - break; - - case RCU_SYSIDLE_SHORT: - - /* - * Idle for a bit, time to advance to next state? - * cmpxchg failure means race with non-idle, let them win. - */ - if (ULONG_CMP_GE(jiffies, j + rcu_sysidle_delay())) - (void)cmpxchg(&full_sysidle_state, - RCU_SYSIDLE_SHORT, RCU_SYSIDLE_LONG); - break; - - case RCU_SYSIDLE_LONG: - - /* - * Do an additional check pass before advancing to full. - * cmpxchg failure means race with non-idle, let them win. - */ - if (ULONG_CMP_GE(jiffies, j + rcu_sysidle_delay())) - (void)cmpxchg(&full_sysidle_state, - RCU_SYSIDLE_LONG, RCU_SYSIDLE_FULL); - break; - - default: - break; - } -} - -/* - * Found a non-idle non-timekeeping CPU, so kick the system-idle state - * back to the beginning. - */ -static void rcu_sysidle_cancel(void) -{ - smp_mb(); - if (full_sysidle_state > RCU_SYSIDLE_SHORT) - WRITE_ONCE(full_sysidle_state, RCU_SYSIDLE_NOT); -} - -/* - * Update the sysidle state based on the results of a force-quiescent-state - * scan of the CPUs' dyntick-idle state. - */ -static void rcu_sysidle_report(struct rcu_state *rsp, int isidle, - unsigned long maxj, bool gpkt) -{ - if (rsp != rcu_state_p) - return; /* Wrong flavor, ignore. */ - if (gpkt && nr_cpu_ids <= CONFIG_NO_HZ_FULL_SYSIDLE_SMALL) - return; /* Running state machine from timekeeping CPU. */ - if (isidle) - rcu_sysidle(maxj); /* More idle! */ - else - rcu_sysidle_cancel(); /* Idle is over. */ -} - -/* - * Wrapper for rcu_sysidle_report() when called from the grace-period - * kthread's context. - */ -static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, - unsigned long maxj) -{ - /* If there are no nohz_full= CPUs, no need to track this. */ - if (!tick_nohz_full_enabled()) - return; - - rcu_sysidle_report(rsp, isidle, maxj, true); -} - -/* Callback and function for forcing an RCU grace period. */ -struct rcu_sysidle_head { - struct rcu_head rh; - int inuse; -}; - -static void rcu_sysidle_cb(struct rcu_head *rhp) -{ - struct rcu_sysidle_head *rshp; - - /* - * The following memory barrier is needed to replace the - * memory barriers that would normally be in the memory - * allocator. - */ - smp_mb(); /* grace period precedes setting inuse. */ - - rshp = container_of(rhp, struct rcu_sysidle_head, rh); - WRITE_ONCE(rshp->inuse, 0); -} - -/* - * Check to see if the system is fully idle, other than the timekeeping CPU. - * The caller must have disabled interrupts. This is not intended to be - * called unless tick_nohz_full_enabled(). - */ -bool rcu_sys_is_idle(void) -{ - static struct rcu_sysidle_head rsh; - int rss = READ_ONCE(full_sysidle_state); - - if (WARN_ON_ONCE(smp_processor_id() != tick_do_timer_cpu)) - return false; - - /* Handle small-system case by doing a full scan of CPUs. */ - if (nr_cpu_ids <= CONFIG_NO_HZ_FULL_SYSIDLE_SMALL) { - int oldrss = rss - 1; - - /* - * One pass to advance to each state up to _FULL. - * Give up if any pass fails to advance the state. - */ - while (rss < RCU_SYSIDLE_FULL && oldrss < rss) { - int cpu; - bool isidle = true; - unsigned long maxj = jiffies - ULONG_MAX / 4; - struct rcu_data *rdp; - - /* Scan all the CPUs looking for nonidle CPUs. */ - for_each_possible_cpu(cpu) { - rdp = per_cpu_ptr(rcu_state_p->rda, cpu); - rcu_sysidle_check_cpu(rdp, &isidle, &maxj); - if (!isidle) - break; - } - rcu_sysidle_report(rcu_state_p, isidle, maxj, false); - oldrss = rss; - rss = READ_ONCE(full_sysidle_state); - } - } - - /* If this is the first observation of an idle period, record it. */ - if (rss == RCU_SYSIDLE_FULL) { - rss = cmpxchg(&full_sysidle_state, - RCU_SYSIDLE_FULL, RCU_SYSIDLE_FULL_NOTED); - return rss == RCU_SYSIDLE_FULL; - } - - smp_mb(); /* ensure rss load happens before later caller actions. */ - - /* If already fully idle, tell the caller (in case of races). */ - if (rss == RCU_SYSIDLE_FULL_NOTED) - return true; - - /* - * If we aren't there yet, and a grace period is not in flight, - * initiate a grace period. Either way, tell the caller that - * we are not there yet. We use an xchg() rather than an assignment - * to make up for the memory barriers that would otherwise be - * provided by the memory allocator. - */ - if (nr_cpu_ids > CONFIG_NO_HZ_FULL_SYSIDLE_SMALL && - !rcu_gp_in_progress(rcu_state_p) && - !rsh.inuse && xchg(&rsh.inuse, 1) == 0) - call_rcu(&rsh.rh, rcu_sysidle_cb); - return false; -} - -/* - * Initialize dynticks sysidle state for CPUs coming online. - */ -static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp) -{ - rdtp->dynticks_idle_nesting = DYNTICK_TASK_NEST_VALUE; -} - -#else /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ - -static void rcu_sysidle_enter(int irq) -{ -} - -static void rcu_sysidle_exit(int irq) -{ -} - -static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, - unsigned long *maxj) -{ -} - -static bool is_sysidle_rcu_state(struct rcu_state *rsp) -{ - return false; -} - -static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, - unsigned long maxj) -{ -} - -static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp) -{ -} - -#endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ - /* * Is this CPU a NO_HZ_FULL CPU that should ignore RCU so that the * grace-period kthread will do force_quiescent_state() processing? @@ -2933,13 +2541,7 @@ static void rcu_bind_gp_kthread(void) if (!tick_nohz_full_enabled()) return; -#ifdef CONFIG_NO_HZ_FULL_SYSIDLE - cpu = tick_do_timer_cpu; - if (cpu >= 0 && cpu < nr_cpu_ids) - set_cpus_allowed_ptr(current, cpumask_of(cpu)); -#else /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ housekeeping_affine(current); -#endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ } /* Record the current task on dyntick-idle entry. */ diff --git a/kernel/rcu/tree_trace.c b/kernel/rcu/tree_trace.c deleted file mode 100644 index b1f28972872c..000000000000 --- a/kernel/rcu/tree_trace.c +++ /dev/null @@ -1,495 +0,0 @@ -/* - * Read-Copy Update tracing for hierarchical implementation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, you can access it online at - * http://www.gnu.org/licenses/gpl-2.0.html. - * - * Copyright IBM Corporation, 2008 - * Author: Paul E. McKenney - * - * Papers: http://www.rdrop.com/users/paulmck/RCU - * - * For detailed explanation of Read-Copy Update mechanism see - - * Documentation/RCU - * - */ -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/init.h> -#include <linux/spinlock.h> -#include <linux/smp.h> -#include <linux/rcupdate.h> -#include <linux/interrupt.h> -#include <linux/sched.h> -#include <linux/atomic.h> -#include <linux/bitops.h> -#include <linux/completion.h> -#include <linux/percpu.h> -#include <linux/notifier.h> -#include <linux/cpu.h> -#include <linux/mutex.h> -#include <linux/debugfs.h> -#include <linux/seq_file.h> - -#define RCU_TREE_NONCORE -#include "tree.h" - -DECLARE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr); - -static int r_open(struct inode *inode, struct file *file, - const struct seq_operations *op) -{ - int ret = seq_open(file, op); - if (!ret) { - struct seq_file *m = (struct seq_file *)file->private_data; - m->private = inode->i_private; - } - return ret; -} - -static void *r_start(struct seq_file *m, loff_t *pos) -{ - struct rcu_state *rsp = (struct rcu_state *)m->private; - *pos = cpumask_next(*pos - 1, cpu_possible_mask); - if ((*pos) < nr_cpu_ids) - return per_cpu_ptr(rsp->rda, *pos); - return NULL; -} - -static void *r_next(struct seq_file *m, void *v, loff_t *pos) -{ - (*pos)++; - return r_start(m, pos); -} - -static void r_stop(struct seq_file *m, void *v) -{ -} - -static int show_rcubarrier(struct seq_file *m, void *v) -{ - struct rcu_state *rsp = (struct rcu_state *)m->private; - seq_printf(m, "bcc: %d bseq: %lu\n", - atomic_read(&rsp->barrier_cpu_count), - rsp->barrier_sequence); - return 0; -} - -static int rcubarrier_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_rcubarrier, inode->i_private); -} - -static const struct file_operations rcubarrier_fops = { - .owner = THIS_MODULE, - .open = rcubarrier_open, - .read = seq_read, - .llseek = no_llseek, - .release = single_release, -}; - -#ifdef CONFIG_RCU_BOOST - -static char convert_kthread_status(unsigned int kthread_status) -{ - if (kthread_status > RCU_KTHREAD_MAX) - return '?'; - return "SRWOY"[kthread_status]; -} - -#endif /* #ifdef CONFIG_RCU_BOOST */ - -static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) -{ - long ql, qll; - - if (!rdp->beenonline) - return; - seq_printf(m, "%3d%cc=%ld g=%ld cnq=%d/%d:%d", - rdp->cpu, - cpu_is_offline(rdp->cpu) ? '!' : ' ', - ulong2long(rdp->completed), ulong2long(rdp->gpnum), - rdp->cpu_no_qs.b.norm, - rdp->rcu_qs_ctr_snap == per_cpu(rcu_qs_ctr, rdp->cpu), - rdp->core_needs_qs); - seq_printf(m, " dt=%d/%llx/%d df=%lu", - atomic_read(&rdp->dynticks->dynticks), - rdp->dynticks->dynticks_nesting, - rdp->dynticks->dynticks_nmi_nesting, - rdp->dynticks_fqs); - seq_printf(m, " of=%lu", rdp->offline_fqs); - rcu_nocb_q_lengths(rdp, &ql, &qll); - qll += rdp->qlen_lazy; - ql += rdp->qlen; - seq_printf(m, " ql=%ld/%ld qs=%c%c%c%c", - qll, ql, - ".N"[rdp->nxttail[RCU_NEXT_READY_TAIL] != - rdp->nxttail[RCU_NEXT_TAIL]], - ".R"[rdp->nxttail[RCU_WAIT_TAIL] != - rdp->nxttail[RCU_NEXT_READY_TAIL]], - ".W"[rdp->nxttail[RCU_DONE_TAIL] != - rdp->nxttail[RCU_WAIT_TAIL]], - ".D"[&rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]]); -#ifdef CONFIG_RCU_BOOST - seq_printf(m, " kt=%d/%c ktl=%x", - per_cpu(rcu_cpu_has_work, rdp->cpu), - convert_kthread_status(per_cpu(rcu_cpu_kthread_status, - rdp->cpu)), - per_cpu(rcu_cpu_kthread_loops, rdp->cpu) & 0xffff); -#endif /* #ifdef CONFIG_RCU_BOOST */ - seq_printf(m, " b=%ld", rdp->blimit); - seq_printf(m, " ci=%lu nci=%lu co=%lu ca=%lu\n", - rdp->n_cbs_invoked, rdp->n_nocbs_invoked, - rdp->n_cbs_orphaned, rdp->n_cbs_adopted); -} - -static int show_rcudata(struct seq_file *m, void *v) -{ - print_one_rcu_data(m, (struct rcu_data *)v); - return 0; -} - -static const struct seq_operations rcudate_op = { - .start = r_start, - .next = r_next, - .stop = r_stop, - .show = show_rcudata, -}; - -static int rcudata_open(struct inode *inode, struct file *file) -{ - return r_open(inode, file, &rcudate_op); -} - -static const struct file_operations rcudata_fops = { - .owner = THIS_MODULE, - .open = rcudata_open, - .read = seq_read, - .llseek = no_llseek, - .release = seq_release, -}; - -static int show_rcuexp(struct seq_file *m, void *v) -{ - int cpu; - struct rcu_state *rsp = (struct rcu_state *)m->private; - struct rcu_data *rdp; - unsigned long s0 = 0, s1 = 0, s2 = 0, s3 = 0; - - for_each_possible_cpu(cpu) { - rdp = per_cpu_ptr(rsp->rda, cpu); - s0 += atomic_long_read(&rdp->exp_workdone0); - s1 += atomic_long_read(&rdp->exp_workdone1); - s2 += atomic_long_read(&rdp->exp_workdone2); - s3 += atomic_long_read(&rdp->exp_workdone3); - } - seq_printf(m, "s=%lu wd0=%lu wd1=%lu wd2=%lu wd3=%lu n=%lu enq=%d sc=%lu\n", - rsp->expedited_sequence, s0, s1, s2, s3, - atomic_long_read(&rsp->expedited_normal), - atomic_read(&rsp->expedited_need_qs), - rsp->expedited_sequence / 2); - return 0; -} - -static int rcuexp_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_rcuexp, inode->i_private); -} - -static const struct file_operations rcuexp_fops = { - .owner = THIS_MODULE, - .open = rcuexp_open, - .read = seq_read, - .llseek = no_llseek, - .release = single_release, -}; - -#ifdef CONFIG_RCU_BOOST - -static void print_one_rcu_node_boost(struct seq_file *m, struct rcu_node *rnp) -{ - seq_printf(m, "%d:%d tasks=%c%c%c%c kt=%c ntb=%lu neb=%lu nnb=%lu ", - rnp->grplo, rnp->grphi, - "T."[list_empty(&rnp->blkd_tasks)], - "N."[!rnp->gp_tasks], - "E."[!rnp->exp_tasks], - "B."[!rnp->boost_tasks], - convert_kthread_status(rnp->boost_kthread_status), - rnp->n_tasks_boosted, rnp->n_exp_boosts, - rnp->n_normal_boosts); - seq_printf(m, "j=%04x bt=%04x\n", - (int)(jiffies & 0xffff), - (int)(rnp->boost_time & 0xffff)); - seq_printf(m, " balk: nt=%lu egt=%lu bt=%lu nb=%lu ny=%lu nos=%lu\n", - rnp->n_balk_blkd_tasks, - rnp->n_balk_exp_gp_tasks, - rnp->n_balk_boost_tasks, - rnp->n_balk_notblocked, - rnp->n_balk_notyet, - rnp->n_balk_nos); -} - -static int show_rcu_node_boost(struct seq_file *m, void *unused) -{ - struct rcu_node *rnp; - - rcu_for_each_leaf_node(&rcu_preempt_state, rnp) - print_one_rcu_node_boost(m, rnp); - return 0; -} - -static int rcu_node_boost_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_rcu_node_boost, NULL); -} - -static const struct file_operations rcu_node_boost_fops = { - .owner = THIS_MODULE, - .open = rcu_node_boost_open, - .read = seq_read, - .llseek = no_llseek, - .release = single_release, -}; - -#endif /* #ifdef CONFIG_RCU_BOOST */ - -static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) -{ - unsigned long gpnum; - int level = 0; - struct rcu_node *rnp; - - gpnum = rsp->gpnum; - seq_printf(m, "c=%ld g=%ld s=%d jfq=%ld j=%x ", - ulong2long(rsp->completed), ulong2long(gpnum), - rsp->gp_state, - (long)(rsp->jiffies_force_qs - jiffies), - (int)(jiffies & 0xffff)); - seq_printf(m, "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld/%ld\n", - rsp->n_force_qs, rsp->n_force_qs_ngp, - rsp->n_force_qs - rsp->n_force_qs_ngp, - READ_ONCE(rsp->n_force_qs_lh), rsp->qlen_lazy, rsp->qlen); - for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < rcu_num_nodes; rnp++) { - if (rnp->level != level) { - seq_puts(m, "\n"); - level = rnp->level; - } - seq_printf(m, "%lx/%lx->%lx %c%c>%c %d:%d ^%d ", - rnp->qsmask, rnp->qsmaskinit, rnp->qsmaskinitnext, - ".G"[rnp->gp_tasks != NULL], - ".E"[rnp->exp_tasks != NULL], - ".T"[!list_empty(&rnp->blkd_tasks)], - rnp->grplo, rnp->grphi, rnp->grpnum); - } - seq_puts(m, "\n"); -} - -static int show_rcuhier(struct seq_file *m, void *v) -{ - struct rcu_state *rsp = (struct rcu_state *)m->private; - print_one_rcu_state(m, rsp); - return 0; -} - -static int rcuhier_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_rcuhier, inode->i_private); -} - -static const struct file_operations rcuhier_fops = { - .owner = THIS_MODULE, - .open = rcuhier_open, - .read = seq_read, - .llseek = no_llseek, - .release = single_release, -}; - -static void show_one_rcugp(struct seq_file *m, struct rcu_state *rsp) -{ - unsigned long flags; - unsigned long completed; - unsigned long gpnum; - unsigned long gpage; - unsigned long gpmax; - struct rcu_node *rnp = &rsp->node[0]; - - raw_spin_lock_irqsave_rcu_node(rnp, flags); - completed = READ_ONCE(rsp->completed); - gpnum = READ_ONCE(rsp->gpnum); - if (completed == gpnum) - gpage = 0; - else - gpage = jiffies - rsp->gp_start; - gpmax = rsp->gp_max; - raw_spin_unlock_irqrestore(&rnp->lock, flags); - seq_printf(m, "completed=%ld gpnum=%ld age=%ld max=%ld\n", - ulong2long(completed), ulong2long(gpnum), gpage, gpmax); -} - -static int show_rcugp(struct seq_file *m, void *v) -{ - struct rcu_state *rsp = (struct rcu_state *)m->private; - show_one_rcugp(m, rsp); - return 0; -} - -static int rcugp_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_rcugp, inode->i_private); -} - -static const struct file_operations rcugp_fops = { - .owner = THIS_MODULE, - .open = rcugp_open, - .read = seq_read, - .llseek = no_llseek, - .release = single_release, -}; - -static void print_one_rcu_pending(struct seq_file *m, struct rcu_data *rdp) -{ - if (!rdp->beenonline) - return; - seq_printf(m, "%3d%cnp=%ld ", - rdp->cpu, - cpu_is_offline(rdp->cpu) ? '!' : ' ', - rdp->n_rcu_pending); - seq_printf(m, "qsp=%ld rpq=%ld cbr=%ld cng=%ld ", - rdp->n_rp_core_needs_qs, - rdp->n_rp_report_qs, - rdp->n_rp_cb_ready, - rdp->n_rp_cpu_needs_gp); - seq_printf(m, "gpc=%ld gps=%ld nn=%ld ndw%ld\n", - rdp->n_rp_gp_completed, - rdp->n_rp_gp_started, - rdp->n_rp_nocb_defer_wakeup, - rdp->n_rp_need_nothing); -} - -static int show_rcu_pending(struct seq_file *m, void *v) -{ - print_one_rcu_pending(m, (struct rcu_data *)v); - return 0; -} - -static const struct seq_operations rcu_pending_op = { - .start = r_start, - .next = r_next, - .stop = r_stop, - .show = show_rcu_pending, -}; - -static int rcu_pending_open(struct inode *inode, struct file *file) -{ - return r_open(inode, file, &rcu_pending_op); -} - -static const struct file_operations rcu_pending_fops = { - .owner = THIS_MODULE, - .open = rcu_pending_open, - .read = seq_read, - .llseek = no_llseek, - .release = seq_release, -}; - -static int show_rcutorture(struct seq_file *m, void *unused) -{ - seq_printf(m, "rcutorture test sequence: %lu %s\n", - rcutorture_testseq >> 1, - (rcutorture_testseq & 0x1) ? "(test in progress)" : ""); - seq_printf(m, "rcutorture update version number: %lu\n", - rcutorture_vernum); - return 0; -} - -static int rcutorture_open(struct inode *inode, struct file *file) -{ - return single_open(file, show_rcutorture, NULL); -} - -static const struct file_operations rcutorture_fops = { - .owner = THIS_MODULE, - .open = rcutorture_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static struct dentry *rcudir; - -static int __init rcutree_trace_init(void) -{ - struct rcu_state *rsp; - struct dentry *retval; - struct dentry *rspdir; - - rcudir = debugfs_create_dir("rcu", NULL); - if (!rcudir) - goto free_out; - - for_each_rcu_flavor(rsp) { - rspdir = debugfs_create_dir(rsp->name, rcudir); - if (!rspdir) - goto free_out; - - retval = debugfs_create_file("rcudata", 0444, - rspdir, rsp, &rcudata_fops); - if (!retval) - goto free_out; - - retval = debugfs_create_file("rcuexp", 0444, - rspdir, rsp, &rcuexp_fops); - if (!retval) - goto free_out; - - retval = debugfs_create_file("rcu_pending", 0444, - rspdir, rsp, &rcu_pending_fops); - if (!retval) - goto free_out; - - retval = debugfs_create_file("rcubarrier", 0444, - rspdir, rsp, &rcubarrier_fops); - if (!retval) - goto free_out; - -#ifdef CONFIG_RCU_BOOST - if (rsp == &rcu_preempt_state) { - retval = debugfs_create_file("rcuboost", 0444, - rspdir, NULL, &rcu_node_boost_fops); - if (!retval) - goto free_out; - } -#endif - - retval = debugfs_create_file("rcugp", 0444, - rspdir, rsp, &rcugp_fops); - if (!retval) - goto free_out; - - retval = debugfs_create_file("rcuhier", 0444, - rspdir, rsp, &rcuhier_fops); - if (!retval) - goto free_out; - } - - retval = debugfs_create_file("rcutorture", 0444, rcudir, - NULL, &rcutorture_fops); - if (!retval) - goto free_out; - return 0; -free_out: - debugfs_remove_recursive(rcudir); - return 1; -} -device_initcall(rcutree_trace_init); diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 4f6db7e6a117..00e77c470017 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -36,7 +36,8 @@ #include <linux/spinlock.h> #include <linux/smp.h> #include <linux/interrupt.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> #include <linux/atomic.h> #include <linux/bitops.h> #include <linux/percpu.h> @@ -49,6 +50,7 @@ #include <linux/moduleparam.h> #include <linux/kthread.h> #include <linux/tick.h> +#include <linux/rcupdate_wait.h> #define CREATE_TRACE_POINTS @@ -60,7 +62,9 @@ #define MODULE_PARAM_PREFIX "rcupdate." #ifndef CONFIG_TINY_RCU +extern int rcu_expedited; /* from sysctl */ module_param(rcu_expedited, int, 0); +extern int rcu_normal; /* from sysctl */ module_param(rcu_normal, int, 0); static int rcu_normal_after_boot; module_param(rcu_normal_after_boot, int, 0); @@ -122,7 +126,7 @@ EXPORT_SYMBOL(rcu_read_lock_sched_held); * non-expedited counterparts? Intended for use within RCU. Note * that if the user specifies both rcu_expedited and rcu_normal, then * rcu_normal wins. (Except during the time period during boot from - * when the first task is spawned until the rcu_exp_runtime_mode() + * when the first task is spawned until the rcu_set_runtime_mode() * core_initcall() is invoked, at which point everything is expedited.) */ bool rcu_gp_is_normal(void) @@ -132,8 +136,7 @@ bool rcu_gp_is_normal(void) } EXPORT_SYMBOL_GPL(rcu_gp_is_normal); -static atomic_t rcu_expedited_nesting = - ATOMIC_INIT(IS_ENABLED(CONFIG_RCU_EXPEDITE_BOOT) ? 1 : 0); +static atomic_t rcu_expedited_nesting = ATOMIC_INIT(1); /* * Should normal grace-period primitives be expedited? Intended for @@ -182,14 +185,46 @@ EXPORT_SYMBOL_GPL(rcu_unexpedite_gp); */ void rcu_end_inkernel_boot(void) { - if (IS_ENABLED(CONFIG_RCU_EXPEDITE_BOOT)) - rcu_unexpedite_gp(); + rcu_unexpedite_gp(); if (rcu_normal_after_boot) WRITE_ONCE(rcu_normal, 1); } #endif /* #ifndef CONFIG_TINY_RCU */ +/* + * Test each non-SRCU synchronous grace-period wait API. This is + * useful just after a change in mode for these primitives, and + * during early boot. + */ +void rcu_test_sync_prims(void) +{ + if (!IS_ENABLED(CONFIG_PROVE_RCU)) + return; + synchronize_rcu(); + synchronize_rcu_bh(); + synchronize_sched(); + synchronize_rcu_expedited(); + synchronize_rcu_bh_expedited(); + synchronize_sched_expedited(); +} + +#if !defined(CONFIG_TINY_RCU) || defined(CONFIG_SRCU) + +/* + * Switch to run-time mode once RCU has fully initialized. + */ +static int __init rcu_set_runtime_mode(void) +{ + rcu_test_sync_prims(); + rcu_scheduler_active = RCU_SCHEDULER_RUNNING; + rcu_test_sync_prims(); + return 0; +} +core_initcall(rcu_set_runtime_mode); + +#endif /* #if !defined(CONFIG_TINY_RCU) || defined(CONFIG_SRCU) */ + #ifdef CONFIG_PREEMPT_RCU /* @@ -346,6 +381,7 @@ void __wait_rcu_gp(bool checktiny, int n, call_rcu_func_t *crcu_array, struct rcu_synchronize *rs_array) { int i; + int j; /* Initialize and register callbacks for each flavor specified. */ for (i = 0; i < n; i++) { @@ -357,7 +393,11 @@ void __wait_rcu_gp(bool checktiny, int n, call_rcu_func_t *crcu_array, } init_rcu_head_on_stack(&rs_array[i].head); init_completion(&rs_array[i].completion); - (crcu_array[i])(&rs_array[i].head, wakeme_after_rcu); + for (j = 0; j < i; j++) + if (crcu_array[j] == crcu_array[i]) + break; + if (j == i) + (crcu_array[i])(&rs_array[i].head, wakeme_after_rcu); } /* Wait for all callbacks to be invoked. */ @@ -366,7 +406,11 @@ void __wait_rcu_gp(bool checktiny, int n, call_rcu_func_t *crcu_array, (crcu_array[i] == call_rcu || crcu_array[i] == call_rcu_bh)) continue; - wait_for_completion(&rs_array[i].completion); + for (j = 0; j < i; j++) + if (crcu_array[j] == crcu_array[i]) + break; + if (j == i) + wait_for_completion(&rs_array[i].completion); destroy_rcu_head_on_stack(&rs_array[i].head); } } @@ -527,15 +571,30 @@ static DEFINE_RAW_SPINLOCK(rcu_tasks_cbs_lock); DEFINE_SRCU(tasks_rcu_exit_srcu); /* Control stall timeouts. Disable with <= 0, otherwise jiffies till stall. */ -static int rcu_task_stall_timeout __read_mostly = HZ * 60 * 10; +#define RCU_TASK_STALL_TIMEOUT (HZ * 60 * 10) +static int rcu_task_stall_timeout __read_mostly = RCU_TASK_STALL_TIMEOUT; module_param(rcu_task_stall_timeout, int, 0644); static void rcu_spawn_tasks_kthread(void); static struct task_struct *rcu_tasks_kthread_ptr; -/* - * Post an RCU-tasks callback. First call must be from process context - * after the scheduler if fully operational. +/** + * call_rcu_tasks() - Queue an RCU for invocation task-based grace period + * @rhp: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all currently executing RCU + * read-side critical sections have completed. call_rcu_tasks() assumes + * that the read-side critical sections end at a voluntary context + * switch (not a preemption!), entry into idle, or transition to usermode + * execution. As such, there are no read-side primitives analogous to + * rcu_read_lock() and rcu_read_unlock() because this primitive is intended + * to determine that all tasks have passed through a safe state, not so + * much for data-strcuture synchronization. + * + * See the description of call_rcu() for more detailed information on + * memory ordering guarantees. */ void call_rcu_tasks(struct rcu_head *rhp, rcu_callback_t func) { @@ -632,6 +691,7 @@ static void check_holdout_task(struct task_struct *t, put_task_struct(t); return; } + rcu_request_urgent_qs_task(t); if (!needreport) return; if (*firstreport) { @@ -817,23 +877,23 @@ static void rcu_spawn_tasks_kthread(void) #endif /* #ifdef CONFIG_TASKS_RCU */ +#ifndef CONFIG_TINY_RCU + /* - * Test each non-SRCU synchronous grace-period wait API. This is - * useful just after a change in mode for these primitives, and - * during early boot. + * Print any non-default Tasks RCU settings. */ -void rcu_test_sync_prims(void) +static void __init rcu_tasks_bootup_oddness(void) { - if (!IS_ENABLED(CONFIG_PROVE_RCU)) - return; - synchronize_rcu(); - synchronize_rcu_bh(); - synchronize_sched(); - synchronize_rcu_expedited(); - synchronize_rcu_bh_expedited(); - synchronize_sched_expedited(); +#ifdef CONFIG_TASKS_RCU + if (rcu_task_stall_timeout != RCU_TASK_STALL_TIMEOUT) + pr_info("\tTasks-RCU CPU stall warnings timeout set to %d (rcu_task_stall_timeout).\n", rcu_task_stall_timeout); + else + pr_info("\tTasks RCU enabled.\n"); +#endif /* #ifdef CONFIG_TASKS_RCU */ } +#endif /* #ifndef CONFIG_TINY_RCU */ + #ifdef CONFIG_PROVE_RCU /* @@ -918,3 +978,25 @@ late_initcall(rcu_verify_early_boot_tests); #else void rcu_early_boot_tests(void) {} #endif /* CONFIG_PROVE_RCU */ + +#ifndef CONFIG_TINY_RCU + +/* + * Print any significant non-default boot-time settings. + */ +void __init rcupdate_announce_bootup_oddness(void) +{ + if (rcu_normal) + pr_info("\tNo expedited grace period (rcu_normal).\n"); + else if (rcu_normal_after_boot) + pr_info("\tNo expedited grace period (rcu_normal_after_boot).\n"); + else if (rcu_expedited) + pr_info("\tAll grace periods are expedited (rcu_expedited).\n"); + if (rcu_cpu_stall_suppress) + pr_info("\tRCU CPU stall warnings suppressed (rcu_cpu_stall_suppress).\n"); + if (rcu_cpu_stall_timeout != CONFIG_RCU_CPU_STALL_TIMEOUT) + pr_info("\tRCU CPU stall warnings timeout set to %d (rcu_cpu_stall_timeout).\n", rcu_cpu_stall_timeout); + rcu_tasks_bootup_oddness(); +} + +#endif /* #ifndef CONFIG_TINY_RCU */ diff --git a/kernel/relay.c b/kernel/relay.c index 8f18d314a96a..39a9dfc69486 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -39,10 +39,10 @@ static void relay_file_mmap_close(struct vm_area_struct *vma) /* * fault() vm_op implementation for relay file mapping. */ -static int relay_buf_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +static int relay_buf_fault(struct vm_fault *vmf) { struct page *page; - struct rchan_buf *buf = vma->vm_private_data; + struct rchan_buf *buf = vmf->vma->vm_private_data; pgoff_t pgoff = vmf->pgoff; if (!buf) @@ -847,7 +847,7 @@ void relay_close(struct rchan *chan) if (chan->last_toobig) printk(KERN_WARNING "relay: one or more items not logged " - "[item size (%Zd) > sub-buffer size (%Zd)]\n", + "[item size (%zd) > sub-buffer size (%zd)]\n", chan->last_toobig, chan->subbuf_size); list_del(&chan->list); @@ -1212,7 +1212,6 @@ static ssize_t subbuf_splice_actor(struct file *in, .nr_pages = 0, .nr_pages_max = PIPE_DEF_BUFFERS, .partial = partial, - .flags = flags, .ops = &relay_pipe_buf_ops, .spd_release = relay_page_release, }; diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile index 5e59b832ae2b..53f0164ed362 100644 --- a/kernel/sched/Makefile +++ b/kernel/sched/Makefile @@ -16,10 +16,10 @@ CFLAGS_core.o := $(PROFILING) -fno-omit-frame-pointer endif obj-y += core.o loadavg.o clock.o cputime.o -obj-y += idle_task.o fair.o rt.o deadline.o stop_task.o -obj-y += wait.o swait.o completion.o idle.o -obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o -obj-$(CONFIG_SCHED_AUTOGROUP) += auto_group.o +obj-y += idle_task.o fair.o rt.o deadline.o +obj-y += wait.o wait_bit.o swait.o completion.o idle.o +obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o topology.o stop_task.o +obj-$(CONFIG_SCHED_AUTOGROUP) += autogroup.o obj-$(CONFIG_SCHEDSTATS) += stats.o obj-$(CONFIG_SCHED_DEBUG) += debug.o obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o diff --git a/kernel/sched/auto_group.c b/kernel/sched/autogroup.c index da39489d2d80..da39489d2d80 100644 --- a/kernel/sched/auto_group.c +++ b/kernel/sched/autogroup.c diff --git a/kernel/sched/auto_group.h b/kernel/sched/autogroup.h index 890c95f2587a..ce40c810cd5c 100644 --- a/kernel/sched/auto_group.h +++ b/kernel/sched/autogroup.h @@ -2,6 +2,7 @@ #include <linux/kref.h> #include <linux/rwsem.h> +#include <linux/sched/autogroup.h> struct autogroup { /* diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index e85a725e5c34..ca0f8fc945c6 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -58,10 +58,13 @@ #include <linux/percpu.h> #include <linux/ktime.h> #include <linux/sched.h> +#include <linux/nmi.h> +#include <linux/sched/clock.h> #include <linux/static_key.h> #include <linux/workqueue.h> #include <linux/compiler.h> #include <linux/tick.h> +#include <linux/init.h> /* * Scheduler clock - returns current time in nanosec units. @@ -77,91 +80,144 @@ EXPORT_SYMBOL_GPL(sched_clock); __read_mostly int sched_clock_running; -#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK -static struct static_key __sched_clock_stable = STATIC_KEY_INIT; -static int __sched_clock_stable_early; - -int sched_clock_stable(void) +void sched_clock_init(void) { - return static_key_false(&__sched_clock_stable); + sched_clock_running = 1; } -static void __set_sched_clock_stable(void) -{ - if (!sched_clock_stable()) - static_key_slow_inc(&__sched_clock_stable); - - tick_dep_clear(TICK_DEP_BIT_CLOCK_UNSTABLE); -} +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK +/* + * We must start with !__sched_clock_stable because the unstable -> stable + * transition is accurate, while the stable -> unstable transition is not. + * + * Similarly we start with __sched_clock_stable_early, thereby assuming we + * will become stable, such that there's only a single 1 -> 0 transition. + */ +static DEFINE_STATIC_KEY_FALSE(__sched_clock_stable); +static int __sched_clock_stable_early = 1; -void set_sched_clock_stable(void) -{ - __sched_clock_stable_early = 1; +/* + * We want: ktime_get_ns() + __gtod_offset == sched_clock() + __sched_clock_offset + */ +__read_mostly u64 __sched_clock_offset; +static __read_mostly u64 __gtod_offset; - smp_mb(); /* matches sched_clock_init() */ +struct sched_clock_data { + u64 tick_raw; + u64 tick_gtod; + u64 clock; +}; - if (!sched_clock_running) - return; +static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data); - __set_sched_clock_stable(); +static inline struct sched_clock_data *this_scd(void) +{ + return this_cpu_ptr(&sched_clock_data); } -static void __clear_sched_clock_stable(struct work_struct *work) +static inline struct sched_clock_data *cpu_sdc(int cpu) { - /* XXX worry about clock continuity */ - if (sched_clock_stable()) - static_key_slow_dec(&__sched_clock_stable); + return &per_cpu(sched_clock_data, cpu); +} - tick_dep_set(TICK_DEP_BIT_CLOCK_UNSTABLE); +int sched_clock_stable(void) +{ + return static_branch_likely(&__sched_clock_stable); } -static DECLARE_WORK(sched_clock_work, __clear_sched_clock_stable); +static void __scd_stamp(struct sched_clock_data *scd) +{ + scd->tick_gtod = ktime_get_ns(); + scd->tick_raw = sched_clock(); +} -void clear_sched_clock_stable(void) +static void __set_sched_clock_stable(void) { - __sched_clock_stable_early = 0; + struct sched_clock_data *scd; - smp_mb(); /* matches sched_clock_init() */ + /* + * Since we're still unstable and the tick is already running, we have + * to disable IRQs in order to get a consistent scd->tick* reading. + */ + local_irq_disable(); + scd = this_scd(); + /* + * Attempt to make the (initial) unstable->stable transition continuous. + */ + __sched_clock_offset = (scd->tick_gtod + __gtod_offset) - (scd->tick_raw); + local_irq_enable(); - if (!sched_clock_running) - return; + printk(KERN_INFO "sched_clock: Marking stable (%lld, %lld)->(%lld, %lld)\n", + scd->tick_gtod, __gtod_offset, + scd->tick_raw, __sched_clock_offset); - schedule_work(&sched_clock_work); + static_branch_enable(&__sched_clock_stable); + tick_dep_clear(TICK_DEP_BIT_CLOCK_UNSTABLE); } -struct sched_clock_data { - u64 tick_raw; - u64 tick_gtod; - u64 clock; -}; +/* + * If we ever get here, we're screwed, because we found out -- typically after + * the fact -- that TSC wasn't good. This means all our clocksources (including + * ktime) could have reported wrong values. + * + * What we do here is an attempt to fix up and continue sort of where we left + * off in a coherent manner. + * + * The only way to fully avoid random clock jumps is to boot with: + * "tsc=unstable". + */ +static void __sched_clock_work(struct work_struct *work) +{ + struct sched_clock_data *scd; + int cpu; -static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data); + /* take a current timestamp and set 'now' */ + preempt_disable(); + scd = this_scd(); + __scd_stamp(scd); + scd->clock = scd->tick_gtod + __gtod_offset; + preempt_enable(); -static inline struct sched_clock_data *this_scd(void) -{ - return this_cpu_ptr(&sched_clock_data); + /* clone to all CPUs */ + for_each_possible_cpu(cpu) + per_cpu(sched_clock_data, cpu) = *scd; + + printk(KERN_WARNING "TSC found unstable after boot, most likely due to broken BIOS. Use 'tsc=unstable'.\n"); + printk(KERN_INFO "sched_clock: Marking unstable (%lld, %lld)<-(%lld, %lld)\n", + scd->tick_gtod, __gtod_offset, + scd->tick_raw, __sched_clock_offset); + + static_branch_disable(&__sched_clock_stable); } -static inline struct sched_clock_data *cpu_sdc(int cpu) +static DECLARE_WORK(sched_clock_work, __sched_clock_work); + +static void __clear_sched_clock_stable(void) { - return &per_cpu(sched_clock_data, cpu); + if (!sched_clock_stable()) + return; + + tick_dep_set(TICK_DEP_BIT_CLOCK_UNSTABLE); + schedule_work(&sched_clock_work); } -void sched_clock_init(void) +void clear_sched_clock_stable(void) { - u64 ktime_now = ktime_to_ns(ktime_get()); - int cpu; + __sched_clock_stable_early = 0; - for_each_possible_cpu(cpu) { - struct sched_clock_data *scd = cpu_sdc(cpu); + smp_mb(); /* matches sched_clock_init_late() */ - scd->tick_raw = 0; - scd->tick_gtod = ktime_now; - scd->clock = ktime_now; - } - - sched_clock_running = 1; + if (sched_clock_running == 2) + __clear_sched_clock_stable(); +} +/* + * We run this as late_initcall() such that it runs after all built-in drivers, + * notably: acpi_processor and intel_idle, which can mark the TSC as unstable. + */ +static int __init sched_clock_init_late(void) +{ + sched_clock_running = 2; /* * Ensure that it is impossible to not do a static_key update. * @@ -173,9 +229,10 @@ void sched_clock_init(void) if (__sched_clock_stable_early) __set_sched_clock_stable(); - else - __clear_sched_clock_stable(NULL); + + return 0; } +late_initcall(sched_clock_init_late); /* * min, max except they take wrapping into account @@ -199,7 +256,7 @@ static inline u64 wrap_max(u64 x, u64 y) */ static u64 sched_clock_local(struct sched_clock_data *scd) { - u64 now, clock, old_clock, min_clock, max_clock; + u64 now, clock, old_clock, min_clock, max_clock, gtod; s64 delta; again: @@ -216,9 +273,10 @@ again: * scd->tick_gtod + TICK_NSEC); */ - clock = scd->tick_gtod + delta; - min_clock = wrap_max(scd->tick_gtod, old_clock); - max_clock = wrap_max(old_clock, scd->tick_gtod + TICK_NSEC); + gtod = scd->tick_gtod + __gtod_offset; + clock = gtod + delta; + min_clock = wrap_max(gtod, old_clock); + max_clock = wrap_max(old_clock, gtod + TICK_NSEC); clock = wrap_max(clock, min_clock); clock = wrap_min(clock, max_clock); @@ -302,7 +360,7 @@ u64 sched_clock_cpu(int cpu) u64 clock; if (sched_clock_stable()) - return sched_clock(); + return sched_clock() + __sched_clock_offset; if (unlikely(!sched_clock_running)) return 0ull; @@ -323,7 +381,6 @@ EXPORT_SYMBOL_GPL(sched_clock_cpu); void sched_clock_tick(void) { struct sched_clock_data *scd; - u64 now, now_gtod; if (sched_clock_stable()) return; @@ -334,14 +391,31 @@ void sched_clock_tick(void) WARN_ON_ONCE(!irqs_disabled()); scd = this_scd(); - now_gtod = ktime_to_ns(ktime_get()); - now = sched_clock(); - - scd->tick_raw = now; - scd->tick_gtod = now_gtod; + __scd_stamp(scd); sched_clock_local(scd); } +void sched_clock_tick_stable(void) +{ + u64 gtod, clock; + + if (!sched_clock_stable()) + return; + + /* + * Called under watchdog_lock. + * + * The watchdog just found this TSC to (still) be stable, so now is a + * good moment to update our __gtod_offset. Because once we find the + * TSC to be unstable, any computation will be computing crap. + */ + local_irq_disable(); + gtod = ktime_get_ns(); + clock = sched_clock(); + __gtod_offset = (clock + __sched_clock_offset) - gtod; + local_irq_enable(); +} + /* * We are going deep-idle (irqs are disabled): */ @@ -352,25 +426,26 @@ void sched_clock_idle_sleep_event(void) EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); /* - * We just idled delta nanoseconds (called with irqs disabled): + * We just idled; resync with ktime. */ -void sched_clock_idle_wakeup_event(u64 delta_ns) +void sched_clock_idle_wakeup_event(void) { - if (timekeeping_suspended) + unsigned long flags; + + if (sched_clock_stable()) + return; + + if (unlikely(timekeeping_suspended)) return; + local_irq_save(flags); sched_clock_tick(); - touch_softlockup_watchdog_sched(); + local_irq_restore(flags); } EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); #else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ -void sched_clock_init(void) -{ - sched_clock_running = 1; -} - u64 sched_clock_cpu(int cpu) { if (unlikely(!sched_clock_running)) @@ -378,6 +453,7 @@ u64 sched_clock_cpu(int cpu) return sched_clock(); } + #endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ /* diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c index 8d0f35debf35..13fc5ae9bf2f 100644 --- a/kernel/sched/completion.c +++ b/kernel/sched/completion.c @@ -11,7 +11,8 @@ * Waiting for completion is a typically sync point, but not an exclusion point. */ -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> #include <linux/completion.h> /** @@ -31,7 +32,8 @@ void complete(struct completion *x) unsigned long flags; spin_lock_irqsave(&x->wait.lock, flags); - x->done++; + if (x->done != UINT_MAX) + x->done++; __wake_up_locked(&x->wait, TASK_NORMAL, 1); spin_unlock_irqrestore(&x->wait.lock, flags); } @@ -51,7 +53,7 @@ void complete_all(struct completion *x) unsigned long flags; spin_lock_irqsave(&x->wait.lock, flags); - x->done += UINT_MAX/2; + x->done = UINT_MAX; __wake_up_locked(&x->wait, TASK_NORMAL, 0); spin_unlock_irqrestore(&x->wait.lock, flags); } @@ -64,7 +66,7 @@ do_wait_for_common(struct completion *x, if (!x->done) { DECLARE_WAITQUEUE(wait, current); - __add_wait_queue_tail_exclusive(&x->wait, &wait); + __add_wait_queue_entry_tail_exclusive(&x->wait, &wait); do { if (signal_pending_state(state, current)) { timeout = -ERESTARTSYS; @@ -79,7 +81,8 @@ do_wait_for_common(struct completion *x, if (!x->done) return timeout; } - x->done--; + if (x->done != UINT_MAX) + x->done--; return timeout ?: 1; } @@ -280,7 +283,7 @@ bool try_wait_for_completion(struct completion *x) spin_lock_irqsave(&x->wait.lock, flags); if (!x->done) ret = 0; - else + else if (x->done != UINT_MAX) x->done--; spin_unlock_irqrestore(&x->wait.lock, flags); return ret; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c56fb57f2991..17c667b427b4 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1,85 +1,34 @@ /* * kernel/sched/core.c * - * Kernel scheduler and related syscalls + * Core kernel scheduler code and related syscalls * * Copyright (C) 1991-2002 Linus Torvalds - * - * 1996-12-23 Modified by Dave Grothe to fix bugs in semaphores and - * make semaphores SMP safe - * 1998-11-19 Implemented schedule_timeout() and related stuff - * by Andrea Arcangeli - * 2002-01-04 New ultra-scalable O(1) scheduler by Ingo Molnar: - * hybrid priority-list and round-robin design with - * an array-switch method of distributing timeslices - * and per-CPU runqueues. Cleanups and useful suggestions - * by Davide Libenzi, preemptible kernel bits by Robert Love. - * 2003-09-03 Interactivity tuning by Con Kolivas. - * 2004-04-02 Scheduler domains code by Nick Piggin - * 2007-04-15 Work begun on replacing all interactivity tuning with a - * fair scheduling design by Con Kolivas. - * 2007-05-05 Load balancing (smp-nice) and other improvements - * by Peter Williams - * 2007-05-06 Interactivity improvements to CFS by Mike Galbraith - * 2007-07-01 Group scheduling enhancements by Srivatsa Vaddagiri - * 2007-11-29 RT balancing improvements by Steven Rostedt, Gregory Haskins, - * Thomas Gleixner, Mike Kravetz */ - -#include <linux/kasan.h> -#include <linux/mm.h> -#include <linux/module.h> -#include <linux/nmi.h> -#include <linux/init.h> -#include <linux/uaccess.h> -#include <linux/highmem.h> -#include <linux/mmu_context.h> -#include <linux/interrupt.h> -#include <linux/capability.h> -#include <linux/completion.h> -#include <linux/kernel_stat.h> -#include <linux/debug_locks.h> -#include <linux/perf_event.h> -#include <linux/security.h> -#include <linux/notifier.h> -#include <linux/profile.h> -#include <linux/freezer.h> -#include <linux/vmalloc.h> -#include <linux/blkdev.h> -#include <linux/delay.h> -#include <linux/pid_namespace.h> -#include <linux/smp.h> -#include <linux/threads.h> -#include <linux/timer.h> -#include <linux/rcupdate.h> -#include <linux/cpu.h> +#include <linux/sched.h> +#include <linux/sched/clock.h> +#include <uapi/linux/sched/types.h> +#include <linux/sched/loadavg.h> +#include <linux/sched/hotplug.h> +#include <linux/wait_bit.h> #include <linux/cpuset.h> -#include <linux/percpu.h> -#include <linux/proc_fs.h> -#include <linux/seq_file.h> -#include <linux/sysctl.h> -#include <linux/syscalls.h> -#include <linux/times.h> -#include <linux/tsacct_kern.h> -#include <linux/kprobes.h> #include <linux/delayacct.h> -#include <linux/unistd.h> -#include <linux/pagemap.h> -#include <linux/hrtimer.h> -#include <linux/tick.h> -#include <linux/ctype.h> -#include <linux/ftrace.h> -#include <linux/slab.h> #include <linux/init_task.h> #include <linux/context_tracking.h> -#include <linux/compiler.h> -#include <linux/frame.h> +#include <linux/rcupdate_wait.h> + +#include <linux/blkdev.h> +#include <linux/kprobes.h> +#include <linux/mmu_context.h> +#include <linux/module.h> +#include <linux/nmi.h> #include <linux/prefetch.h> -#include <linux/mutex.h> +#include <linux/profile.h> +#include <linux/security.h> +#include <linux/syscalls.h> #include <asm/switch_to.h> #include <asm/tlb.h> -#include <asm/irq_regs.h> #ifdef CONFIG_PARAVIRT #include <asm/paravirt.h> #endif @@ -91,27 +40,8 @@ #define CREATE_TRACE_POINTS #include <trace/events/sched.h> -DEFINE_MUTEX(sched_domains_mutex); DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); -static void update_rq_clock_task(struct rq *rq, s64 delta); - -void update_rq_clock(struct rq *rq) -{ - s64 delta; - - lockdep_assert_held(&rq->lock); - - if (rq->clock_skip_update & RQCF_ACT_SKIP) - return; - - delta = sched_clock_cpu(cpu_of(rq)) - rq->clock; - if (delta < 0) - return; - rq->clock += delta; - update_rq_clock_task(rq, delta); -} - /* * Debugging: various feature bits */ @@ -140,7 +70,7 @@ const_debug unsigned int sysctl_sched_nr_migrate = 32; const_debug unsigned int sysctl_sched_time_avg = MSEC_PER_SEC; /* - * period over which we measure -rt task cpu usage in us. + * period over which we measure -rt task CPU usage in us. * default: 1s */ unsigned int sysctl_sched_rt_period = 1000000; @@ -153,25 +83,10 @@ __read_mostly int scheduler_running; */ int sysctl_sched_rt_runtime = 950000; -/* cpus with isolated domains */ +/* CPUs with isolated domains */ cpumask_var_t cpu_isolated_map; /* - * this_rq_lock - lock this runqueue and disable interrupts. - */ -static struct rq *this_rq_lock(void) - __acquires(rq->lock) -{ - struct rq *rq; - - local_irq_disable(); - rq = this_rq(); - raw_spin_lock(&rq->lock); - - return rq; -} - -/* * __task_rq_lock - lock the rq @p resides on. */ struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) @@ -185,7 +100,7 @@ struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) rq = task_rq(p); raw_spin_lock(&rq->lock); if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) { - rf->cookie = lockdep_pin_lock(&rq->lock); + rq_pin_lock(rq, rf); return rq; } raw_spin_unlock(&rq->lock); @@ -221,11 +136,11 @@ struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) * If we observe the old cpu in task_rq_lock, the acquire of * the old rq->lock will fully serialize against the stores. * - * If we observe the new cpu in task_rq_lock, the acquire will + * If we observe the new CPU in task_rq_lock, the acquire will * pair with the WMB to ensure we must then also see migrating. */ if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) { - rf->cookie = lockdep_pin_lock(&rq->lock); + rq_pin_lock(rq, rf); return rq; } raw_spin_unlock(&rq->lock); @@ -236,6 +151,87 @@ struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) } } +/* + * RQ-clock updating methods: + */ + +static void update_rq_clock_task(struct rq *rq, s64 delta) +{ +/* + * In theory, the compile should just see 0 here, and optimize out the call + * to sched_rt_avg_update. But I don't trust it... + */ +#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) + s64 steal = 0, irq_delta = 0; +#endif +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; + + /* + * Since irq_time is only updated on {soft,}irq_exit, we might run into + * this case when a previous update_rq_clock() happened inside a + * {soft,}irq region. + * + * When this happens, we stop ->clock_task and only update the + * prev_irq_time stamp to account for the part that fit, so that a next + * update will consume the rest. This ensures ->clock_task is + * monotonic. + * + * It does however cause some slight miss-attribution of {soft,}irq + * time, a more accurate solution would be to update the irq_time using + * the current rq->clock timestamp, except that would require using + * atomic ops. + */ + if (irq_delta > delta) + irq_delta = delta; + + rq->prev_irq_time += irq_delta; + delta -= irq_delta; +#endif +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + if (static_key_false((¶virt_steal_rq_enabled))) { + steal = paravirt_steal_clock(cpu_of(rq)); + steal -= rq->prev_steal_time_rq; + + if (unlikely(steal > delta)) + steal = delta; + + rq->prev_steal_time_rq += steal; + delta -= steal; + } +#endif + + rq->clock_task += delta; + +#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) + if ((irq_delta + steal) && sched_feat(NONTASK_CAPACITY)) + sched_rt_avg_update(rq, irq_delta + steal); +#endif +} + +void update_rq_clock(struct rq *rq) +{ + s64 delta; + + lockdep_assert_held(&rq->lock); + + if (rq->clock_update_flags & RQCF_ACT_SKIP) + return; + +#ifdef CONFIG_SCHED_DEBUG + if (sched_feat(WARN_DOUBLE_CLOCK)) + SCHED_WARN_ON(rq->clock_update_flags & RQCF_UPDATED); + rq->clock_update_flags |= RQCF_UPDATED; +#endif + + delta = sched_clock_cpu(cpu_of(rq)) - rq->clock; + if (delta < 0) + return; + rq->clock += delta; + update_rq_clock_task(rq, delta); +} + + #ifdef CONFIG_SCHED_HRTICK /* * Use HR-timers to deliver accurate preemption points. @@ -254,13 +250,14 @@ static void hrtick_clear(struct rq *rq) static enum hrtimer_restart hrtick(struct hrtimer *timer) { struct rq *rq = container_of(timer, struct rq, hrtick_timer); + struct rq_flags rf; WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); update_rq_clock(rq); rq->curr->sched_class->task_tick(rq, rq->curr, 1); - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); return HRTIMER_NORESTART; } @@ -280,11 +277,12 @@ static void __hrtick_restart(struct rq *rq) static void __hrtick_start(void *arg) { struct rq *rq = arg; + struct rq_flags rf; - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); __hrtick_restart(rq); rq->hrtick_csd_pending = 0; - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); } /* @@ -458,7 +456,7 @@ void wake_up_q(struct wake_q_head *head) task = container_of(node, struct task_struct, wake_q); BUG_ON(!task); - /* task can safely be re-inserted now */ + /* Task can safely be re-inserted now: */ node = node->next; task->wake_q.next = NULL; @@ -516,12 +514,12 @@ void resched_cpu(int cpu) #ifdef CONFIG_SMP #ifdef CONFIG_NO_HZ_COMMON /* - * In the semi idle case, use the nearest busy cpu for migrating timers - * from an idle cpu. This is good for power-savings. + * In the semi idle case, use the nearest busy CPU for migrating timers + * from an idle CPU. This is good for power-savings. * * We don't do similar optimization for completely idle system, as - * selecting an idle cpu will add more delays to the timers than intended - * (as that cpu's timer base may not be uptodate wrt jiffies etc). + * selecting an idle CPU will add more delays to the timers than intended + * (as that CPU's timer base may not be uptodate wrt jiffies etc). */ int get_nohz_timer_target(void) { @@ -550,6 +548,7 @@ unlock: rcu_read_unlock(); return cpu; } + /* * When add_timer_on() enqueues a timer into the timer wheel of an * idle CPU then this timer might expire before the next timer event @@ -754,17 +753,23 @@ static void set_load_weight(struct task_struct *p) static inline void enqueue_task(struct rq *rq, struct task_struct *p, int flags) { - update_rq_clock(rq); + if (!(flags & ENQUEUE_NOCLOCK)) + update_rq_clock(rq); + if (!(flags & ENQUEUE_RESTORE)) sched_info_queued(rq, p); + p->sched_class->enqueue_task(rq, p, flags); } static inline void dequeue_task(struct rq *rq, struct task_struct *p, int flags) { - update_rq_clock(rq); + if (!(flags & DEQUEUE_NOCLOCK)) + update_rq_clock(rq); + if (!(flags & DEQUEUE_SAVE)) sched_info_dequeued(rq, p); + p->sched_class->dequeue_task(rq, p, flags); } @@ -784,90 +789,6 @@ void deactivate_task(struct rq *rq, struct task_struct *p, int flags) dequeue_task(rq, p, flags); } -static void update_rq_clock_task(struct rq *rq, s64 delta) -{ -/* - * In theory, the compile should just see 0 here, and optimize out the call - * to sched_rt_avg_update. But I don't trust it... - */ -#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) - s64 steal = 0, irq_delta = 0; -#endif -#ifdef CONFIG_IRQ_TIME_ACCOUNTING - irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; - - /* - * Since irq_time is only updated on {soft,}irq_exit, we might run into - * this case when a previous update_rq_clock() happened inside a - * {soft,}irq region. - * - * When this happens, we stop ->clock_task and only update the - * prev_irq_time stamp to account for the part that fit, so that a next - * update will consume the rest. This ensures ->clock_task is - * monotonic. - * - * It does however cause some slight miss-attribution of {soft,}irq - * time, a more accurate solution would be to update the irq_time using - * the current rq->clock timestamp, except that would require using - * atomic ops. - */ - if (irq_delta > delta) - irq_delta = delta; - - rq->prev_irq_time += irq_delta; - delta -= irq_delta; -#endif -#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING - if (static_key_false((¶virt_steal_rq_enabled))) { - steal = paravirt_steal_clock(cpu_of(rq)); - steal -= rq->prev_steal_time_rq; - - if (unlikely(steal > delta)) - steal = delta; - - rq->prev_steal_time_rq += steal; - delta -= steal; - } -#endif - - rq->clock_task += delta; - -#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) - if ((irq_delta + steal) && sched_feat(NONTASK_CAPACITY)) - sched_rt_avg_update(rq, irq_delta + steal); -#endif -} - -void sched_set_stop_task(int cpu, struct task_struct *stop) -{ - struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; - struct task_struct *old_stop = cpu_rq(cpu)->stop; - - if (stop) { - /* - * Make it appear like a SCHED_FIFO task, its something - * userspace knows about and won't get confused about. - * - * Also, it will make PI more or less work without too - * much confusion -- but then, stop work should not - * rely on PI working anyway. - */ - sched_setscheduler_nocheck(stop, SCHED_FIFO, ¶m); - - stop->sched_class = &stop_sched_class; - } - - cpu_rq(cpu)->stop = stop; - - if (old_stop) { - /* - * Reset it back to a normal scheduling class so that - * it can die in pieces. - */ - old_stop->sched_class = &rt_sched_class; - } -} - /* * __normal_prio - return the priority that is based on the static prio */ @@ -992,18 +913,19 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) * * Returns (locked) new rq. Old rq's lock is released. */ -static struct rq *move_queued_task(struct rq *rq, struct task_struct *p, int new_cpu) +static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, + struct task_struct *p, int new_cpu) { lockdep_assert_held(&rq->lock); p->on_rq = TASK_ON_RQ_MIGRATING; - dequeue_task(rq, p, 0); + dequeue_task(rq, p, DEQUEUE_NOCLOCK); set_task_cpu(p, new_cpu); - raw_spin_unlock(&rq->lock); + rq_unlock(rq, rf); rq = cpu_rq(new_cpu); - raw_spin_lock(&rq->lock); + rq_lock(rq, rf); BUG_ON(task_cpu(p) != new_cpu); enqueue_task(rq, p, 0); p->on_rq = TASK_ON_RQ_QUEUED; @@ -1018,7 +940,7 @@ struct migration_arg { }; /* - * Move (not current) task off this cpu, onto dest cpu. We're doing + * Move (not current) task off this CPU, onto the destination CPU. We're doing * this because either it can't run here any more (set_cpus_allowed() * away from this CPU, or CPU going down), or because we're * attempting to rebalance this task on exec (sched_exec). @@ -1026,16 +948,18 @@ struct migration_arg { * So we race with normal scheduler movements, but that's OK, as long * as the task is no longer on this CPU. */ -static struct rq *__migrate_task(struct rq *rq, struct task_struct *p, int dest_cpu) +static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf, + struct task_struct *p, int dest_cpu) { if (unlikely(!cpu_active(dest_cpu))) return rq; /* Affinity changed (again). */ - if (!cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p))) + if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) return rq; - rq = move_queued_task(rq, p, dest_cpu); + update_rq_clock(rq); + rq = move_queued_task(rq, rf, p, dest_cpu); return rq; } @@ -1050,10 +974,11 @@ static int migration_cpu_stop(void *data) struct migration_arg *arg = data; struct task_struct *p = arg->task; struct rq *rq = this_rq(); + struct rq_flags rf; /* - * The original target cpu might have gone down and we might - * be on another cpu but it doesn't matter. + * The original target CPU might have gone down and we might + * be on another CPU but it doesn't matter. */ local_irq_disable(); /* @@ -1064,7 +989,7 @@ static int migration_cpu_stop(void *data) sched_ttwu_pending(); raw_spin_lock(&p->pi_lock); - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); /* * If task_rq(p) != rq, it cannot be migrated here, because we're * holding rq->lock, if p->on_rq == 0 it cannot get enqueued because @@ -1072,11 +997,11 @@ static int migration_cpu_stop(void *data) */ if (task_rq(p) == rq) { if (task_on_rq_queued(p)) - rq = __migrate_task(rq, p, arg->dest_cpu); + rq = __migrate_task(rq, &rf, p, arg->dest_cpu); else p->wake_cpu = arg->dest_cpu; } - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); raw_spin_unlock(&p->pi_lock); local_irq_enable(); @@ -1109,7 +1034,7 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) * holding rq->lock. */ lockdep_assert_held(&rq->lock); - dequeue_task(rq, p, DEQUEUE_SAVE); + dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK); } if (running) put_prev_task(rq, p); @@ -1117,7 +1042,7 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) p->sched_class->set_cpus_allowed(p, new_mask); if (queued) - enqueue_task(rq, p, ENQUEUE_RESTORE); + enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); if (running) set_curr_task(rq, p); } @@ -1141,6 +1066,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, int ret = 0; rq = task_rq_lock(p, &rf); + update_rq_clock(rq); if (p->flags & PF_KTHREAD) { /* @@ -1171,7 +1097,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, if (p->flags & PF_KTHREAD) { /* * For kernel threads that do indeed end up on online && - * !active we want to ensure they are strict per-cpu threads. + * !active we want to ensure they are strict per-CPU threads. */ WARN_ON(cpumask_intersects(new_mask, cpu_online_mask) && !cpumask_intersects(new_mask, cpu_active_mask) && @@ -1195,9 +1121,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, * OK, since we're going to drop the lock immediately * afterwards anyway. */ - lockdep_unpin_lock(&rq->lock, rf.cookie); - rq = move_queued_task(rq, p, dest_cpu); - lockdep_repin_lock(&rq->lock, rf.cookie); + rq = move_queued_task(rq, &rf, p, dest_cpu); } out: task_rq_unlock(rq, p, &rf); @@ -1262,21 +1186,29 @@ static void __migrate_swap_task(struct task_struct *p, int cpu) { if (task_on_rq_queued(p)) { struct rq *src_rq, *dst_rq; + struct rq_flags srf, drf; src_rq = task_rq(p); dst_rq = cpu_rq(cpu); + rq_pin_lock(src_rq, &srf); + rq_pin_lock(dst_rq, &drf); + p->on_rq = TASK_ON_RQ_MIGRATING; deactivate_task(src_rq, p, 0); set_task_cpu(p, cpu); activate_task(dst_rq, p, 0); p->on_rq = TASK_ON_RQ_QUEUED; check_preempt_curr(dst_rq, p, 0); + + rq_unpin_lock(dst_rq, &drf); + rq_unpin_lock(src_rq, &srf); + } else { /* * Task isn't running anymore; make it appear like we migrated * it before it went to sleep. This means on wakeup we make the - * previous cpu our target instead of where it really is. + * previous CPU our target instead of where it really is. */ p->wake_cpu = cpu; } @@ -1309,10 +1241,10 @@ static int migrate_swap_stop(void *data) if (task_cpu(arg->src_task) != arg->src_cpu) goto unlock; - if (!cpumask_test_cpu(arg->dst_cpu, tsk_cpus_allowed(arg->src_task))) + if (!cpumask_test_cpu(arg->dst_cpu, &arg->src_task->cpus_allowed)) goto unlock; - if (!cpumask_test_cpu(arg->src_cpu, tsk_cpus_allowed(arg->dst_task))) + if (!cpumask_test_cpu(arg->src_cpu, &arg->dst_task->cpus_allowed)) goto unlock; __migrate_swap_task(arg->src_task, arg->dst_cpu); @@ -1353,10 +1285,10 @@ int migrate_swap(struct task_struct *cur, struct task_struct *p) if (!cpu_active(arg.src_cpu) || !cpu_active(arg.dst_cpu)) goto out; - if (!cpumask_test_cpu(arg.dst_cpu, tsk_cpus_allowed(arg.src_task))) + if (!cpumask_test_cpu(arg.dst_cpu, &arg.src_task->cpus_allowed)) goto out; - if (!cpumask_test_cpu(arg.src_cpu, tsk_cpus_allowed(arg.dst_task))) + if (!cpumask_test_cpu(arg.src_cpu, &arg.dst_task->cpus_allowed)) goto out; trace_sched_swap_numa(cur, arg.src_cpu, p, arg.dst_cpu); @@ -1508,12 +1440,12 @@ EXPORT_SYMBOL_GPL(kick_process); * * - on cpu-up we allow per-cpu kthreads on the online && !active cpu, * see __set_cpus_allowed_ptr(). At this point the newly online - * cpu isn't yet part of the sched domains, and balancing will not + * CPU isn't yet part of the sched domains, and balancing will not * see it. * - * - on cpu-down we clear cpu_active() to mask the sched domains and + * - on CPU-down we clear cpu_active() to mask the sched domains and * avoid the load balancer to place new tasks on the to be removed - * cpu. Existing tasks will remain running there and will be taken + * CPU. Existing tasks will remain running there and will be taken * off. * * This means that fallback selection must not select !active CPUs. @@ -1529,9 +1461,9 @@ static int select_fallback_rq(int cpu, struct task_struct *p) int dest_cpu; /* - * If the node that the cpu is on has been offlined, cpu_to_node() - * will return -1. There is no cpu on the node, and we should - * select the cpu on the other node. + * If the node that the CPU is on has been offlined, cpu_to_node() + * will return -1. There is no CPU on the node, and we should + * select the CPU on the other node. */ if (nid != -1) { nodemask = cpumask_of_node(nid); @@ -1540,14 +1472,14 @@ static int select_fallback_rq(int cpu, struct task_struct *p) for_each_cpu(dest_cpu, nodemask) { if (!cpu_active(dest_cpu)) continue; - if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p))) + if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) return dest_cpu; } } for (;;) { /* Any allowed, online CPU? */ - for_each_cpu(dest_cpu, tsk_cpus_allowed(p)) { + for_each_cpu(dest_cpu, &p->cpus_allowed) { if (!(p->flags & PF_KTHREAD) && !cpu_active(dest_cpu)) continue; if (!cpu_online(dest_cpu)) @@ -1563,7 +1495,7 @@ static int select_fallback_rq(int cpu, struct task_struct *p) state = possible; break; } - /* fall-through */ + /* Fall-through */ case possible: do_set_cpus_allowed(p, cpu_possible_mask); state = fail; @@ -1599,22 +1531,22 @@ int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags) { lockdep_assert_held(&p->pi_lock); - if (tsk_nr_cpus_allowed(p) > 1) + if (p->nr_cpus_allowed > 1) cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags); else - cpu = cpumask_any(tsk_cpus_allowed(p)); + cpu = cpumask_any(&p->cpus_allowed); /* * In order not to call set_task_cpu() on a blocking task we need * to rely on ttwu() to place the task on a valid ->cpus_allowed - * cpu. + * CPU. * * Since this is common to all placement strategies, this lives here. * * [ this allows ->select_task() to simply return task_cpu(p) and * not worry about this generic constraint ] */ - if (unlikely(!cpumask_test_cpu(cpu, tsk_cpus_allowed(p)) || + if (unlikely(!cpumask_test_cpu(cpu, &p->cpus_allowed) || !cpu_online(cpu))) cpu = select_fallback_rq(task_cpu(p), p); @@ -1627,6 +1559,36 @@ static void update_avg(u64 *avg, u64 sample) *avg += diff >> 3; } +void sched_set_stop_task(int cpu, struct task_struct *stop) +{ + struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; + struct task_struct *old_stop = cpu_rq(cpu)->stop; + + if (stop) { + /* + * Make it appear like a SCHED_FIFO task, its something + * userspace knows about and won't get confused about. + * + * Also, it will make PI more or less work without too + * much confusion -- but then, stop work should not + * rely on PI working anyway. + */ + sched_setscheduler_nocheck(stop, SCHED_FIFO, ¶m); + + stop->sched_class = &stop_sched_class; + } + + cpu_rq(cpu)->stop = stop; + + if (old_stop) { + /* + * Reset it back to a normal scheduling class so that + * it can die in pieces. + */ + old_stop->sched_class = &rt_sched_class; + } +} + #else static inline int __set_cpus_allowed_ptr(struct task_struct *p, @@ -1681,7 +1643,7 @@ static inline void ttwu_activate(struct rq *rq, struct task_struct *p, int en_fl activate_task(rq, p, en_flags); p->on_rq = TASK_ON_RQ_QUEUED; - /* if a worker is waking up, notify workqueue */ + /* If a worker is waking up, notify the workqueue: */ if (p->flags & PF_WQ_WORKER) wq_worker_waking_up(p, cpu_of(rq)); } @@ -1690,7 +1652,7 @@ static inline void ttwu_activate(struct rq *rq, struct task_struct *p, int en_fl * Mark the task runnable and perform wakeup-preemption. */ static void ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags, - struct pin_cookie cookie) + struct rq_flags *rf) { check_preempt_curr(rq, p, wake_flags); p->state = TASK_RUNNING; @@ -1702,9 +1664,9 @@ static void ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags, * Our task @p is fully woken up and running; so its safe to * drop the rq->lock, hereafter rq is only used for statistics. */ - lockdep_unpin_lock(&rq->lock, cookie); + rq_unpin_lock(rq, rf); p->sched_class->task_woken(rq, p); - lockdep_repin_lock(&rq->lock, cookie); + rq_repin_lock(rq, rf); } if (rq->idle_stamp) { @@ -1723,9 +1685,9 @@ static void ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags, static void ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, - struct pin_cookie cookie) + struct rq_flags *rf) { - int en_flags = ENQUEUE_WAKEUP; + int en_flags = ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK; lockdep_assert_held(&rq->lock); @@ -1738,7 +1700,7 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, #endif ttwu_activate(rq, p, en_flags); - ttwu_do_wakeup(rq, p, wake_flags, cookie); + ttwu_do_wakeup(rq, p, wake_flags, rf); } /* @@ -1757,7 +1719,7 @@ static int ttwu_remote(struct task_struct *p, int wake_flags) if (task_on_rq_queued(p)) { /* check_preempt_curr() may use rq clock */ update_rq_clock(rq); - ttwu_do_wakeup(rq, p, wake_flags, rf.cookie); + ttwu_do_wakeup(rq, p, wake_flags, &rf); ret = 1; } __task_rq_unlock(rq, &rf); @@ -1770,30 +1732,19 @@ void sched_ttwu_pending(void) { struct rq *rq = this_rq(); struct llist_node *llist = llist_del_all(&rq->wake_list); - struct pin_cookie cookie; - struct task_struct *p; - unsigned long flags; + struct task_struct *p, *t; + struct rq_flags rf; if (!llist) return; - raw_spin_lock_irqsave(&rq->lock, flags); - cookie = lockdep_pin_lock(&rq->lock); - - while (llist) { - int wake_flags = 0; - - p = llist_entry(llist, struct task_struct, wake_entry); - llist = llist_next(llist); - - if (p->sched_remote_wakeup) - wake_flags = WF_MIGRATED; + rq_lock_irqsave(rq, &rf); + update_rq_clock(rq); - ttwu_do_activate(rq, p, wake_flags, cookie); - } + llist_for_each_entry_safe(p, t, llist, wake_entry) + ttwu_do_activate(rq, p, p->sched_remote_wakeup ? WF_MIGRATED : 0, &rf); - lockdep_unpin_lock(&rq->lock, cookie); - raw_spin_unlock_irqrestore(&rq->lock, flags); + rq_unlock_irqrestore(rq, &rf); } void scheduler_ipi(void) @@ -1851,7 +1802,7 @@ static void ttwu_queue_remote(struct task_struct *p, int cpu, int wake_flags) void wake_up_if_idle(int cpu) { struct rq *rq = cpu_rq(cpu); - unsigned long flags; + struct rq_flags rf; rcu_read_lock(); @@ -1861,11 +1812,11 @@ void wake_up_if_idle(int cpu) if (set_nr_if_polling(rq->idle)) { trace_sched_wake_idle_without_ipi(cpu); } else { - raw_spin_lock_irqsave(&rq->lock, flags); + rq_lock_irqsave(rq, &rf); if (is_idle_task(rq->curr)) smp_send_reschedule(cpu); - /* Else cpu is not in idle, do nothing here */ - raw_spin_unlock_irqrestore(&rq->lock, flags); + /* Else CPU is not idle, do nothing here: */ + rq_unlock_irqrestore(rq, &rf); } out: @@ -1881,21 +1832,20 @@ bool cpus_share_cache(int this_cpu, int that_cpu) static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) { struct rq *rq = cpu_rq(cpu); - struct pin_cookie cookie; + struct rq_flags rf; #if defined(CONFIG_SMP) if (sched_feat(TTWU_QUEUE) && !cpus_share_cache(smp_processor_id(), cpu)) { - sched_clock_cpu(cpu); /* sync clocks x-cpu */ + sched_clock_cpu(cpu); /* Sync clocks across CPUs */ ttwu_queue_remote(p, cpu, wake_flags); return; } #endif - raw_spin_lock(&rq->lock); - cookie = lockdep_pin_lock(&rq->lock); - ttwu_do_activate(rq, p, wake_flags, cookie); - lockdep_unpin_lock(&rq->lock, cookie); - raw_spin_unlock(&rq->lock); + rq_lock(rq, &rf); + update_rq_clock(rq); + ttwu_do_activate(rq, p, wake_flags, &rf); + rq_unlock(rq, &rf); } /* @@ -1904,8 +1854,8 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) * MIGRATION * * The basic program-order guarantee on SMP systems is that when a task [t] - * migrates, all its activity on its old cpu [c0] happens-before any subsequent - * execution on its new cpu [c1]. + * migrates, all its activity on its old CPU [c0] happens-before any subsequent + * execution on its new CPU [c1]. * * For migration (of runnable tasks) this is provided by the following means: * @@ -1916,7 +1866,7 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) * * Transitivity guarantees that B happens after A and C after B. * Note: we only require RCpc transitivity. - * Note: the cpu doing B need not be c0 or c1 + * Note: the CPU doing B need not be c0 or c1 * * Example: * @@ -2024,7 +1974,8 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) trace_sched_waking(p); - success = 1; /* we're going to change ->state */ + /* We're going to change ->state: */ + success = 1; cpu = task_cpu(p); /* @@ -2073,7 +2024,7 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) smp_rmb(); /* - * If the owning (remote) cpu is still in the middle of schedule() with + * If the owning (remote) CPU is still in the middle of schedule() with * this task as prev, wait until its done referencing the task. * * Pairs with the smp_store_release() in finish_lock_switch(). @@ -2086,11 +2037,24 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) p->sched_contributes_to_load = !!task_contributes_to_load(p); p->state = TASK_WAKING; + if (p->in_iowait) { + delayacct_blkio_end(); + atomic_dec(&task_rq(p)->nr_iowait); + } + cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags); if (task_cpu(p) != cpu) { wake_flags |= WF_MIGRATED; set_task_cpu(p, cpu); } + +#else /* CONFIG_SMP */ + + if (p->in_iowait) { + delayacct_blkio_end(); + atomic_dec(&task_rq(p)->nr_iowait); + } + #endif /* CONFIG_SMP */ ttwu_queue(p, cpu, wake_flags); @@ -2111,7 +2075,7 @@ out: * ensure that this_rq() is locked, @p is bound to this_rq() and not * the current task. */ -static void try_to_wake_up_local(struct task_struct *p, struct pin_cookie cookie) +static void try_to_wake_up_local(struct task_struct *p, struct rq_flags *rf) { struct rq *rq = task_rq(p); @@ -2128,11 +2092,9 @@ static void try_to_wake_up_local(struct task_struct *p, struct pin_cookie cookie * disabled avoiding further scheduler activity on it and we've * not yet picked a replacement task. */ - lockdep_unpin_lock(&rq->lock, cookie); - raw_spin_unlock(&rq->lock); + rq_unlock(rq, rf); raw_spin_lock(&p->pi_lock); - raw_spin_lock(&rq->lock); - lockdep_repin_lock(&rq->lock, cookie); + rq_relock(rq, rf); } if (!(p->state & TASK_NORMAL)) @@ -2140,10 +2102,15 @@ static void try_to_wake_up_local(struct task_struct *p, struct pin_cookie cookie trace_sched_waking(p); - if (!task_on_rq_queued(p)) - ttwu_activate(rq, p, ENQUEUE_WAKEUP); + if (!task_on_rq_queued(p)) { + if (p->in_iowait) { + delayacct_blkio_end(); + atomic_dec(&rq->nr_iowait); + } + ttwu_activate(rq, p, ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK); + } - ttwu_do_wakeup(rq, p, 0, cookie); + ttwu_do_wakeup(rq, p, 0, rf); ttwu_stat(p, smp_processor_id(), 0); out: raw_spin_unlock(&p->pi_lock); @@ -2173,23 +2140,6 @@ int wake_up_state(struct task_struct *p, unsigned int state) } /* - * This function clears the sched_dl_entity static params. - */ -void __dl_clear_params(struct task_struct *p) -{ - struct sched_dl_entity *dl_se = &p->dl; - - dl_se->dl_runtime = 0; - dl_se->dl_deadline = 0; - dl_se->dl_period = 0; - dl_se->flags = 0; - dl_se->dl_bw = 0; - - dl_se->dl_throttled = 0; - dl_se->dl_yielded = 0; -} - -/* * Perform scheduler related setup for a newly forked process p. * p is forked by current. * @@ -2218,6 +2168,7 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p) RB_CLEAR_NODE(&p->dl.rb_node); init_dl_task_timer(&p->dl); + init_dl_inactive_task_timer(&p->dl); __dl_clear_params(p); INIT_LIST_HEAD(&p->rt.run_list); @@ -2427,7 +2378,7 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p) */ raw_spin_lock_irqsave(&p->pi_lock, flags); /* - * We're setting the cpu for the first time, we don't migrate, + * We're setting the CPU for the first time, we don't migrate, * so use __set_task_cpu(). */ __set_task_cpu(p, cpu); @@ -2455,7 +2406,7 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p) unsigned long to_ratio(u64 period, u64 runtime) { if (runtime == RUNTIME_INF) - return 1ULL << 20; + return BW_UNIT; /* * Doing this here saves a lot of checks in all @@ -2465,93 +2416,9 @@ unsigned long to_ratio(u64 period, u64 runtime) if (period == 0) return 0; - return div64_u64(runtime << 20, period); -} - -#ifdef CONFIG_SMP -inline struct dl_bw *dl_bw_of(int i) -{ - RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), - "sched RCU must be held"); - return &cpu_rq(i)->rd->dl_bw; -} - -static inline int dl_bw_cpus(int i) -{ - struct root_domain *rd = cpu_rq(i)->rd; - int cpus = 0; - - RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), - "sched RCU must be held"); - for_each_cpu_and(i, rd->span, cpu_active_mask) - cpus++; - - return cpus; -} -#else -inline struct dl_bw *dl_bw_of(int i) -{ - return &cpu_rq(i)->dl.dl_bw; -} - -static inline int dl_bw_cpus(int i) -{ - return 1; -} -#endif - -/* - * We must be sure that accepting a new task (or allowing changing the - * parameters of an existing one) is consistent with the bandwidth - * constraints. If yes, this function also accordingly updates the currently - * allocated bandwidth to reflect the new situation. - * - * This function is called while holding p's rq->lock. - * - * XXX we should delay bw change until the task's 0-lag point, see - * __setparam_dl(). - */ -static int dl_overflow(struct task_struct *p, int policy, - const struct sched_attr *attr) -{ - - struct dl_bw *dl_b = dl_bw_of(task_cpu(p)); - u64 period = attr->sched_period ?: attr->sched_deadline; - u64 runtime = attr->sched_runtime; - u64 new_bw = dl_policy(policy) ? to_ratio(period, runtime) : 0; - int cpus, err = -1; - - /* !deadline task may carry old deadline bandwidth */ - if (new_bw == p->dl.dl_bw && task_has_dl_policy(p)) - return 0; - - /* - * Either if a task, enters, leave, or stays -deadline but changes - * its parameters, we may need to update accordingly the total - * allocated bandwidth of the container. - */ - raw_spin_lock(&dl_b->lock); - cpus = dl_bw_cpus(task_cpu(p)); - if (dl_policy(policy) && !task_has_dl_policy(p) && - !__dl_overflow(dl_b, cpus, 0, new_bw)) { - __dl_add(dl_b, new_bw); - err = 0; - } else if (dl_policy(policy) && task_has_dl_policy(p) && - !__dl_overflow(dl_b, cpus, p->dl.dl_bw, new_bw)) { - __dl_clear(dl_b, p->dl.dl_bw); - __dl_add(dl_b, new_bw); - err = 0; - } else if (!dl_policy(policy) && task_has_dl_policy(p)) { - __dl_clear(dl_b, p->dl.dl_bw); - err = 0; - } - raw_spin_unlock(&dl_b->lock); - - return err; + return div64_u64(runtime << BW_SHIFT, period); } -extern void init_dl_bw(struct dl_bw *dl_b); - /* * wake_up_new_task - wake up a newly created task for the first time. * @@ -2570,7 +2437,7 @@ void wake_up_new_task(struct task_struct *p) /* * Fork balancing, do it here and not earlier because: * - cpus_allowed can change in the fork path - * - any previously selected cpu might disappear through hotplug + * - any previously selected CPU might disappear through hotplug * * Use __set_task_cpu() to avoid calling sched_class::migrate_task_rq, * as we're not fully set-up yet. @@ -2578,9 +2445,10 @@ void wake_up_new_task(struct task_struct *p) __set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0)); #endif rq = __task_rq_lock(p, &rf); + update_rq_clock(rq); post_init_entity_util_avg(&p->se); - activate_task(rq, p, 0); + activate_task(rq, p, ENQUEUE_NOCLOCK); p->on_rq = TASK_ON_RQ_QUEUED; trace_sched_wakeup_new(p); check_preempt_curr(rq, p, WF_FORK); @@ -2590,9 +2458,9 @@ void wake_up_new_task(struct task_struct *p) * Nothing relies on rq->lock after this, so its fine to * drop it. */ - lockdep_unpin_lock(&rq->lock, rf.cookie); + rq_unpin_lock(rq, &rf); p->sched_class->task_woken(rq, p); - lockdep_repin_lock(&rq->lock, rf.cookie); + rq_repin_lock(rq, &rf); } #endif task_rq_unlock(rq, p, &rf); @@ -2861,7 +2729,7 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev) */ static __always_inline struct rq * context_switch(struct rq *rq, struct task_struct *prev, - struct task_struct *next, struct pin_cookie cookie) + struct task_struct *next, struct rq_flags *rf) { struct mm_struct *mm, *oldmm; @@ -2878,7 +2746,7 @@ context_switch(struct rq *rq, struct task_struct *prev, if (!mm) { next->active_mm = oldmm; - atomic_inc(&oldmm->mm_count); + mmgrab(oldmm); enter_lazy_tlb(oldmm, next); } else switch_mm_irqs_off(oldmm, mm, next); @@ -2887,13 +2755,16 @@ context_switch(struct rq *rq, struct task_struct *prev, prev->active_mm = NULL; rq->prev_mm = oldmm; } + + rq->clock_update_flags &= ~(RQCF_ACT_SKIP|RQCF_REQ_SKIP); + /* * Since the runqueue lock will be released by the next * task (which is an invalid locking op but in the case * of the scheduler it's an obvious special-case), so we * do an early lockdep release here: */ - lockdep_unpin_lock(&rq->lock, cookie); + rq_unpin_lock(rq, rf); spin_release(&rq->lock.dep_map, 1, _THIS_IP_); /* Here we just switch the register state and the stack. */ @@ -2920,7 +2791,7 @@ unsigned long nr_running(void) } /* - * Check if only the current task is running on the cpu. + * Check if only the current task is running on the CPU. * * Caution: this function does not check that the caller has disabled * preemption, thus the result might have a time-of-check-to-time-of-use @@ -2949,6 +2820,36 @@ unsigned long long nr_context_switches(void) return sum; } +/* + * IO-wait accounting, and how its mostly bollocks (on SMP). + * + * The idea behind IO-wait account is to account the idle time that we could + * have spend running if it were not for IO. That is, if we were to improve the + * storage performance, we'd have a proportional reduction in IO-wait time. + * + * This all works nicely on UP, where, when a task blocks on IO, we account + * idle time as IO-wait, because if the storage were faster, it could've been + * running and we'd not be idle. + * + * This has been extended to SMP, by doing the same for each CPU. This however + * is broken. + * + * Imagine for instance the case where two tasks block on one CPU, only the one + * CPU will have IO-wait accounted, while the other has regular idle. Even + * though, if the storage were faster, both could've ran at the same time, + * utilising both CPUs. + * + * This means, that when looking globally, the current IO-wait accounting on + * SMP is a lower bound, by reason of under accounting. + * + * Worse, since the numbers are provided per CPU, they are sometimes + * interpreted per CPU, and that is nonsensical. A blocked task isn't strictly + * associated with any one particular CPU, it can wake to another CPU than it + * blocked on. This means the per CPU IO-wait number is meaningless. + * + * Task CPU affinities can make all that even more 'interesting'. + */ + unsigned long nr_iowait(void) { unsigned long i, sum = 0; @@ -2959,6 +2860,13 @@ unsigned long nr_iowait(void) return sum; } +/* + * Consumers of these two interfaces, like for example the cpufreq menu + * governor are using nonsensical data. Boosting frequency for a CPU that has + * IO-wait which might not even end up running the task when it does become + * runnable. + */ + unsigned long nr_iowait_cpu(int cpu) { struct rq *this = cpu_rq(cpu); @@ -3042,8 +2950,8 @@ unsigned long long task_sched_runtime(struct task_struct *p) * So we have a optimization chance when the task's delta_exec is 0. * Reading ->on_cpu is racy, but this is ok. * - * If we race with it leaving cpu, we'll take a lock. So we're correct. - * If we race with it entering cpu, unaccounted time is 0. This is + * If we race with it leaving CPU, we'll take a lock. So we're correct. + * If we race with it entering CPU, unaccounted time is 0. This is * indistinguishable from the read occurring a few cycles earlier. * If we see ->on_cpu without ->on_rq, the task is leaving, and has * been accounted, so we're correct here as well. @@ -3078,15 +2986,18 @@ void scheduler_tick(void) int cpu = smp_processor_id(); struct rq *rq = cpu_rq(cpu); struct task_struct *curr = rq->curr; + struct rq_flags rf; sched_clock_tick(); - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); + update_rq_clock(rq); curr->sched_class->task_tick(rq, curr, 0); cpu_load_update_active(rq); calc_global_load_tick(rq); - raw_spin_unlock(&rq->lock); + + rq_unlock(rq, &rf); perf_event_task_tick(); @@ -3201,6 +3112,15 @@ static inline void preempt_latency_start(int val) { } static inline void preempt_latency_stop(int val) { } #endif +static inline unsigned long get_preempt_disable_ip(struct task_struct *p) +{ +#ifdef CONFIG_DEBUG_PREEMPT + return p->preempt_disable_ip; +#else + return 0; +#endif +} + /* * Print scheduling while atomic bug: */ @@ -3257,31 +3177,35 @@ static inline void schedule_debug(struct task_struct *prev) * Pick up the highest-prio task: */ static inline struct task_struct * -pick_next_task(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie) +pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { - const struct sched_class *class = &fair_sched_class; + const struct sched_class *class; struct task_struct *p; /* - * Optimization: we know that if all tasks are in - * the fair class we can call that function directly: + * Optimization: we know that if all tasks are in the fair class we can + * call that function directly, but only if the @prev task wasn't of a + * higher scheduling class, because otherwise those loose the + * opportunity to pull in more work from other CPUs. */ - if (likely(prev->sched_class == class && + if (likely((prev->sched_class == &idle_sched_class || + prev->sched_class == &fair_sched_class) && rq->nr_running == rq->cfs.h_nr_running)) { - p = fair_sched_class.pick_next_task(rq, prev, cookie); + + p = fair_sched_class.pick_next_task(rq, prev, rf); if (unlikely(p == RETRY_TASK)) goto again; - /* assumes fair_sched_class->next == idle_sched_class */ + /* Assumes fair_sched_class->next == idle_sched_class */ if (unlikely(!p)) - p = idle_sched_class.pick_next_task(rq, prev, cookie); + p = idle_sched_class.pick_next_task(rq, prev, rf); return p; } again: for_each_class(class) { - p = class->pick_next_task(rq, prev, cookie); + p = class->pick_next_task(rq, prev, rf); if (p) { if (unlikely(p == RETRY_TASK)) goto again; @@ -3289,7 +3213,8 @@ again: } } - BUG(); /* the idle class will always have a runnable task */ + /* The idle class should always have a runnable task: */ + BUG(); } /* @@ -3335,7 +3260,7 @@ static void __sched notrace __schedule(bool preempt) { struct task_struct *prev, *next; unsigned long *switch_count; - struct pin_cookie cookie; + struct rq_flags rf; struct rq *rq; int cpu; @@ -3349,7 +3274,7 @@ static void __sched notrace __schedule(bool preempt) hrtick_clear(rq); local_irq_disable(); - rcu_note_context_switch(); + rcu_note_context_switch(preempt); /* * Make sure that signal_pending_state()->signal_pending() below @@ -3357,19 +3282,25 @@ static void __sched notrace __schedule(bool preempt) * done by the caller to avoid the race with signal_wake_up(). */ smp_mb__before_spinlock(); - raw_spin_lock(&rq->lock); - cookie = lockdep_pin_lock(&rq->lock); + rq_lock(rq, &rf); - rq->clock_skip_update <<= 1; /* promote REQ to ACT */ + /* Promote REQ to ACT */ + rq->clock_update_flags <<= 1; + update_rq_clock(rq); switch_count = &prev->nivcsw; if (!preempt && prev->state) { if (unlikely(signal_pending_state(prev->state, prev))) { prev->state = TASK_RUNNING; } else { - deactivate_task(rq, prev, DEQUEUE_SLEEP); + deactivate_task(rq, prev, DEQUEUE_SLEEP | DEQUEUE_NOCLOCK); prev->on_rq = 0; + if (prev->in_iowait) { + atomic_inc(&rq->nr_iowait); + delayacct_blkio_start(); + } + /* * If a worker went to sleep, notify and ask workqueue * whether it wants to wake up a task to maintain @@ -3380,19 +3311,15 @@ static void __sched notrace __schedule(bool preempt) to_wakeup = wq_worker_sleeping(prev); if (to_wakeup) - try_to_wake_up_local(to_wakeup, cookie); + try_to_wake_up_local(to_wakeup, &rf); } } switch_count = &prev->nvcsw; } - if (task_on_rq_queued(prev)) - update_rq_clock(rq); - - next = pick_next_task(rq, prev, cookie); + next = pick_next_task(rq, prev, &rf); clear_tsk_need_resched(prev); clear_preempt_need_resched(); - rq->clock_skip_update = 0; if (likely(prev != next)) { rq->nr_switches++; @@ -3400,10 +3327,12 @@ static void __sched notrace __schedule(bool preempt) ++*switch_count; trace_sched_switch(preempt, prev, next); - rq = context_switch(rq, prev, next, cookie); /* unlocks the rq */ + + /* Also unlocks the rq: */ + rq = context_switch(rq, prev, next, &rf); } else { - lockdep_unpin_lock(&rq->lock, cookie); - raw_spin_unlock_irq(&rq->lock); + rq->clock_update_flags &= ~(RQCF_ACT_SKIP|RQCF_REQ_SKIP); + rq_unlock_irq(rq, &rf); } balance_callback(rq); @@ -3426,14 +3355,18 @@ void __noreturn do_task_dead(void) smp_mb(); raw_spin_unlock_wait(¤t->pi_lock); - /* causes final put_task_struct in finish_task_switch(). */ + /* Causes final put_task_struct in finish_task_switch(): */ __set_current_state(TASK_DEAD); - current->flags |= PF_NOFREEZE; /* tell freezer to ignore us */ + + /* Tell freezer to ignore us: */ + current->flags |= PF_NOFREEZE; + __schedule(false); BUG(); - /* Avoid "noreturn function does return". */ + + /* Avoid "noreturn function does return" - but don't continue if BUG() is a NOP: */ for (;;) - cpu_relax(); /* For when BUG is null */ + cpu_relax(); } static inline void sched_submit_work(struct task_struct *tsk) @@ -3461,6 +3394,31 @@ asmlinkage __visible void __sched schedule(void) } EXPORT_SYMBOL(schedule); +/* + * synchronize_rcu_tasks() makes sure that no task is stuck in preempted + * state (have scheduled out non-voluntarily) by making sure that all + * tasks have either left the run queue or have gone into user space. + * As idle tasks do not do either, they must not ever be preempted + * (schedule out non-voluntarily). + * + * schedule_idle() is similar to schedule_preempt_disable() except that it + * never enables preemption because it does not call sched_submit_work(). + */ +void __sched schedule_idle(void) +{ + /* + * As this skips calling sched_submit_work(), which the idle task does + * regardless because that function is a nop when the task is in a + * TASK_RUNNING state, make sure this isn't used someplace that the + * current task can be in any other state. Note, idle is always in the + * TASK_RUNNING state. + */ + WARN_ON_ONCE(current->state); + do { + __schedule(false); + } while (need_resched()); +} + #ifdef CONFIG_CONTEXT_TRACKING asmlinkage __visible void __sched schedule_user(void) { @@ -3621,7 +3579,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void) exception_exit(prev_state); } -int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags, +int default_wake_function(wait_queue_entry_t *curr, unsigned mode, int wake_flags, void *key) { return try_to_wake_up(curr->private, mode, wake_flags); @@ -3630,10 +3588,25 @@ EXPORT_SYMBOL(default_wake_function); #ifdef CONFIG_RT_MUTEXES +static inline int __rt_effective_prio(struct task_struct *pi_task, int prio) +{ + if (pi_task) + prio = min(prio, pi_task->prio); + + return prio; +} + +static inline int rt_effective_prio(struct task_struct *p, int prio) +{ + struct task_struct *pi_task = rt_mutex_get_top_task(p); + + return __rt_effective_prio(pi_task, prio); +} + /* * rt_mutex_setprio - set the current priority of a task - * @p: task - * @prio: prio value (kernel-internal form) + * @p: task to boost + * @pi_task: donor task * * This function changes the 'effective' priority of a task. It does * not touch ->normal_prio like __setscheduler(). @@ -3641,16 +3614,42 @@ EXPORT_SYMBOL(default_wake_function); * Used by the rt_mutex code to implement priority inheritance * logic. Call site only calls if the priority of the task changed. */ -void rt_mutex_setprio(struct task_struct *p, int prio) +void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task) { - int oldprio, queued, running, queue_flag = DEQUEUE_SAVE | DEQUEUE_MOVE; + int prio, oldprio, queued, running, queue_flag = + DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; const struct sched_class *prev_class; struct rq_flags rf; struct rq *rq; - BUG_ON(prio > MAX_PRIO); + /* XXX used to be waiter->prio, not waiter->task->prio */ + prio = __rt_effective_prio(pi_task, p->normal_prio); + + /* + * If nothing changed; bail early. + */ + if (p->pi_top_task == pi_task && prio == p->prio && !dl_prio(prio)) + return; rq = __task_rq_lock(p, &rf); + update_rq_clock(rq); + /* + * Set under pi_lock && rq->lock, such that the value can be used under + * either lock. + * + * Note that there is loads of tricky to make this pointer cache work + * right. rt_mutex_slowunlock()+rt_mutex_postunlock() work together to + * ensure a task is de-boosted (pi_task is set to NULL) before the + * task is allowed to run again (and can exit). This ensures the pointer + * points to a blocked task -- which guaratees the task is present. + */ + p->pi_top_task = pi_task; + + /* + * For FIFO/RR we only need to set prio, if that matches we're done. + */ + if (prio == p->prio && !dl_prio(prio)) + goto out_unlock; /* * Idle task boosting is a nono in general. There is one @@ -3670,7 +3669,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio) goto out_unlock; } - trace_sched_pi_setprio(p, prio); + trace_sched_pi_setprio(p, pi_task); oldprio = p->prio; if (oldprio == prio) @@ -3694,7 +3693,6 @@ void rt_mutex_setprio(struct task_struct *p, int prio) * running task */ if (dl_prio(prio)) { - struct task_struct *pi_task = rt_mutex_get_top_task(p); if (!dl_prio(p->normal_prio) || (pi_task && dl_entity_preempt(&pi_task->dl, &p->dl))) { p->dl.dl_boosted = 1; @@ -3725,12 +3723,18 @@ void rt_mutex_setprio(struct task_struct *p, int prio) check_class_changed(rq, p, prev_class, oldprio); out_unlock: - preempt_disable(); /* avoid rq from going away on us */ + /* Avoid rq from going away on us: */ + preempt_disable(); __task_rq_unlock(rq, &rf); balance_callback(rq); preempt_enable(); } +#else +static inline int rt_effective_prio(struct task_struct *p, int prio) +{ + return prio; +} #endif void set_user_nice(struct task_struct *p, long nice) @@ -3747,6 +3751,8 @@ void set_user_nice(struct task_struct *p, long nice) * the task might be in the middle of scheduling on another CPU. */ rq = task_rq_lock(p, &rf); + update_rq_clock(rq); + /* * The RT priorities are set via sched_setscheduler(), but we still * allow the 'normal' nice value to be set - but as expected @@ -3760,7 +3766,7 @@ void set_user_nice(struct task_struct *p, long nice) queued = task_on_rq_queued(p); running = task_current(rq, p); if (queued) - dequeue_task(rq, p, DEQUEUE_SAVE); + dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK); if (running) put_prev_task(rq, p); @@ -3771,7 +3777,7 @@ void set_user_nice(struct task_struct *p, long nice) delta = p->prio - old_prio; if (queued) { - enqueue_task(rq, p, ENQUEUE_RESTORE); + enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); /* * If the task increased its priority or is running and * lowered its priority, then reschedule its CPU: @@ -3793,7 +3799,7 @@ EXPORT_SYMBOL(set_user_nice); */ int can_nice(const struct task_struct *p, const int nice) { - /* convert nice value [19,-20] to rlimit style value [1,40] */ + /* Convert nice value [19,-20] to rlimit style value [1,40]: */ int nice_rlim = nice_to_rlimit(nice); return (nice_rlim <= task_rlimit(p, RLIMIT_NICE) || @@ -3849,7 +3855,7 @@ int task_prio(const struct task_struct *p) } /** - * idle_cpu - is a given cpu idle currently? + * idle_cpu - is a given CPU idle currently? * @cpu: the processor in question. * * Return: 1 if the CPU is currently idle. 0 otherwise. @@ -3873,10 +3879,10 @@ int idle_cpu(int cpu) } /** - * idle_task - return the idle task for a given cpu. + * idle_task - return the idle task for a given CPU. * @cpu: the processor in question. * - * Return: The idle task for the cpu @cpu. + * Return: The idle task for the CPU @cpu. */ struct task_struct *idle_task(int cpu) { @@ -3895,46 +3901,6 @@ static struct task_struct *find_process_by_pid(pid_t pid) } /* - * This function initializes the sched_dl_entity of a newly becoming - * SCHED_DEADLINE task. - * - * Only the static values are considered here, the actual runtime and the - * absolute deadline will be properly calculated when the task is enqueued - * for the first time with its new policy. - */ -static void -__setparam_dl(struct task_struct *p, const struct sched_attr *attr) -{ - struct sched_dl_entity *dl_se = &p->dl; - - dl_se->dl_runtime = attr->sched_runtime; - dl_se->dl_deadline = attr->sched_deadline; - dl_se->dl_period = attr->sched_period ?: dl_se->dl_deadline; - dl_se->flags = attr->sched_flags; - dl_se->dl_bw = to_ratio(dl_se->dl_period, dl_se->dl_runtime); - - /* - * Changing the parameters of a task is 'tricky' and we're not doing - * the correct thing -- also see task_dead_dl() and switched_from_dl(). - * - * What we SHOULD do is delay the bandwidth release until the 0-lag - * point. This would include retaining the task_struct until that time - * and change dl_overflow() to not immediately decrement the current - * amount. - * - * Instead we retain the current runtime/deadline and let the new - * parameters take effect after the current reservation period lapses. - * This is safe (albeit pessimistic) because the 0-lag point is always - * before the current scheduling deadline. - * - * We can still have temporary overloads because we do not delay the - * change in bandwidth until that time; so admission control is - * not on the safe side. It does however guarantee tasks will never - * consume more than promised. - */ -} - -/* * sched_setparam() passes in -1 for its policy, to let the functions * it calls know not to change it. */ @@ -3975,10 +3941,9 @@ static void __setscheduler(struct rq *rq, struct task_struct *p, * Keep a potential priority boosting if called from * sched_setscheduler(). */ + p->prio = normal_prio(p); if (keep_boost) - p->prio = rt_mutex_get_effective_prio(p, normal_prio(p)); - else - p->prio = normal_prio(p); + p->prio = rt_effective_prio(p, p->prio); if (dl_prio(p->prio)) p->sched_class = &dl_sched_class; @@ -3988,61 +3953,8 @@ static void __setscheduler(struct rq *rq, struct task_struct *p, p->sched_class = &fair_sched_class; } -static void -__getparam_dl(struct task_struct *p, struct sched_attr *attr) -{ - struct sched_dl_entity *dl_se = &p->dl; - - attr->sched_priority = p->rt_priority; - attr->sched_runtime = dl_se->dl_runtime; - attr->sched_deadline = dl_se->dl_deadline; - attr->sched_period = dl_se->dl_period; - attr->sched_flags = dl_se->flags; -} - -/* - * This function validates the new parameters of a -deadline task. - * We ask for the deadline not being zero, and greater or equal - * than the runtime, as well as the period of being zero or - * greater than deadline. Furthermore, we have to be sure that - * user parameters are above the internal resolution of 1us (we - * check sched_runtime only since it is always the smaller one) and - * below 2^63 ns (we have to check both sched_deadline and - * sched_period, as the latter can be zero). - */ -static bool -__checkparam_dl(const struct sched_attr *attr) -{ - /* deadline != 0 */ - if (attr->sched_deadline == 0) - return false; - - /* - * Since we truncate DL_SCALE bits, make sure we're at least - * that big. - */ - if (attr->sched_runtime < (1ULL << DL_SCALE)) - return false; - - /* - * Since we use the MSB for wrap-around and sign issues, make - * sure it's not set (mind that period can be equal to zero). - */ - if (attr->sched_deadline & (1ULL << 63) || - attr->sched_period & (1ULL << 63)) - return false; - - /* runtime <= deadline <= period (if period != 0) */ - if ((attr->sched_period != 0 && - attr->sched_period < attr->sched_deadline) || - attr->sched_deadline < attr->sched_runtime) - return false; - - return true; -} - /* - * check the target process has a UID that matches the current process's + * Check the target process has a UID that matches the current process's: */ static bool check_same_owner(struct task_struct *p) { @@ -4057,20 +3969,6 @@ static bool check_same_owner(struct task_struct *p) return match; } -static bool dl_param_changed(struct task_struct *p, - const struct sched_attr *attr) -{ - struct sched_dl_entity *dl_se = &p->dl; - - if (dl_se->dl_runtime != attr->sched_runtime || - dl_se->dl_deadline != attr->sched_deadline || - dl_se->dl_period != attr->sched_period || - dl_se->flags != attr->sched_flags) - return true; - - return false; -} - static int __sched_setscheduler(struct task_struct *p, const struct sched_attr *attr, bool user, bool pi) @@ -4082,13 +3980,13 @@ static int __sched_setscheduler(struct task_struct *p, const struct sched_class *prev_class; struct rq_flags rf; int reset_on_fork; - int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE; + int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; struct rq *rq; - /* may grab non-irq protected spin_locks */ - BUG_ON(in_interrupt()); + /* The pi code expects interrupts enabled */ + BUG_ON(pi && in_interrupt()); recheck: - /* double check policy once rq lock held */ + /* Double check policy once rq lock held: */ if (policy < 0) { reset_on_fork = p->sched_reset_on_fork; policy = oldpolicy = p->policy; @@ -4099,7 +3997,8 @@ recheck: return -EINVAL; } - if (attr->sched_flags & ~(SCHED_FLAG_RESET_ON_FORK)) + if (attr->sched_flags & + ~(SCHED_FLAG_RESET_ON_FORK | SCHED_FLAG_RECLAIM)) return -EINVAL; /* @@ -4128,11 +4027,11 @@ recheck: unsigned long rlim_rtprio = task_rlimit(p, RLIMIT_RTPRIO); - /* can't set/change the rt policy */ + /* Can't set/change the rt policy: */ if (policy != p->policy && !rlim_rtprio) return -EPERM; - /* can't increase priority */ + /* Can't increase priority: */ if (attr->sched_priority > p->rt_priority && attr->sched_priority > rlim_rtprio) return -EPERM; @@ -4156,11 +4055,11 @@ recheck: return -EPERM; } - /* can't change other user's priorities */ + /* Can't change other user's priorities: */ if (!check_same_owner(p)) return -EPERM; - /* Normal users shall not reset the sched_reset_on_fork flag */ + /* Normal users shall not reset the sched_reset_on_fork flag: */ if (p->sched_reset_on_fork && !reset_on_fork) return -EPERM; } @@ -4172,16 +4071,17 @@ recheck: } /* - * make sure no PI-waiters arrive (or leave) while we are + * Make sure no PI-waiters arrive (or leave) while we are * changing the priority of the task: * * To be able to change p->policy safely, the appropriate * runqueue lock must be held. */ rq = task_rq_lock(p, &rf); + update_rq_clock(rq); /* - * Changing the policy of the stop threads its a very bad idea + * Changing the policy of the stop threads its a very bad idea: */ if (p == rq->stop) { task_rq_unlock(rq, p, &rf); @@ -4237,7 +4137,7 @@ change: #endif } - /* recheck policy now with rq lock held */ + /* Re-check policy now with rq lock held: */ if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; task_rq_unlock(rq, p, &rf); @@ -4249,7 +4149,7 @@ change: * of a SCHED_DEADLINE task) we need to check if enough bandwidth * is available. */ - if ((dl_policy(policy) || dl_task(p)) && dl_overflow(p, policy, attr)) { + if ((dl_policy(policy) || dl_task(p)) && sched_dl_overflow(p, policy, attr)) { task_rq_unlock(rq, p, &rf); return -EBUSY; } @@ -4265,7 +4165,7 @@ change: * the runqueue. This will be done when the task deboost * itself. */ - new_effective_prio = rt_mutex_get_effective_prio(p, newprio); + new_effective_prio = rt_effective_prio(p, newprio); if (new_effective_prio == oldprio) queue_flags &= ~DEQUEUE_MOVE; } @@ -4294,15 +4194,15 @@ change: set_curr_task(rq, p); check_class_changed(rq, p, prev_class, oldprio); - preempt_disable(); /* avoid rq from going away on us */ + + /* Avoid rq from going away on us: */ + preempt_disable(); task_rq_unlock(rq, p, &rf); if (pi) rt_mutex_adjust_pi(p); - /* - * Run balance callbacks after we've adjusted the PI chain. - */ + /* Run balance callbacks after we've adjusted the PI chain: */ balance_callback(rq); preempt_enable(); @@ -4395,8 +4295,7 @@ do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) /* * Mimics kernel/events/core.c perf_copy_attr(). */ -static int sched_copy_attr(struct sched_attr __user *uattr, - struct sched_attr *attr) +static int sched_copy_attr(struct sched_attr __user *uattr, struct sched_attr *attr) { u32 size; int ret; @@ -4404,19 +4303,19 @@ static int sched_copy_attr(struct sched_attr __user *uattr, if (!access_ok(VERIFY_WRITE, uattr, SCHED_ATTR_SIZE_VER0)) return -EFAULT; - /* - * zero the full structure, so that a short copy will be nice. - */ + /* Zero the full structure, so that a short copy will be nice: */ memset(attr, 0, sizeof(*attr)); ret = get_user(size, &uattr->size); if (ret) return ret; - if (size > PAGE_SIZE) /* silly large */ + /* Bail out on silly large: */ + if (size > PAGE_SIZE) goto err_size; - if (!size) /* abi compat */ + /* ABI compatibility quirk: */ + if (!size) size = SCHED_ATTR_SIZE_VER0; if (size < SCHED_ATTR_SIZE_VER0) @@ -4451,7 +4350,7 @@ static int sched_copy_attr(struct sched_attr __user *uattr, return -EFAULT; /* - * XXX: do we want to be lenient like existing syscalls; or do we want + * XXX: Do we want to be lenient like existing syscalls; or do we want * to be strict and return an error on out-of-bounds values? */ attr->sched_nice = clamp(attr->sched_nice, MIN_NICE, MAX_NICE); @@ -4471,10 +4370,8 @@ err_size: * * Return: 0 on success. An error code otherwise. */ -SYSCALL_DEFINE3(sched_setscheduler, pid_t, pid, int, policy, - struct sched_param __user *, param) +SYSCALL_DEFINE3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param) { - /* negative values for policy are not valid */ if (policy < 0) return -EINVAL; @@ -4784,10 +4681,10 @@ static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, } /** - * sys_sched_setaffinity - set the cpu affinity of a process + * sys_sched_setaffinity - set the CPU affinity of a process * @pid: pid of the process * @len: length in bytes of the bitmask pointed to by user_mask_ptr - * @user_mask_ptr: user-space pointer to the new cpu mask + * @user_mask_ptr: user-space pointer to the new CPU mask * * Return: 0 on success. An error code otherwise. */ @@ -4835,10 +4732,10 @@ out_unlock: } /** - * sys_sched_getaffinity - get the cpu affinity of a process + * sys_sched_getaffinity - get the CPU affinity of a process * @pid: pid of the process * @len: length in bytes of the bitmask pointed to by user_mask_ptr - * @user_mask_ptr: user-space pointer to hold the current cpu mask + * @user_mask_ptr: user-space pointer to hold the current CPU mask * * Return: size of CPU mask copied to user_mask_ptr on success. An * error code otherwise. @@ -4881,7 +4778,12 @@ SYSCALL_DEFINE3(sched_getaffinity, pid_t, pid, unsigned int, len, */ SYSCALL_DEFINE0(sched_yield) { - struct rq *rq = this_rq_lock(); + struct rq_flags rf; + struct rq *rq; + + local_irq_disable(); + rq = this_rq(); + rq_lock(rq, &rf); schedstat_inc(rq->yld_count); current->sched_class->yield_task(rq); @@ -4890,9 +4792,8 @@ SYSCALL_DEFINE0(sched_yield) * Since we are going to call schedule() anyway, there's * no need to preempt or enable interrupts: */ - __release(rq->lock); - spin_release(&rq->lock.dep_map, 1, _THIS_IP_); - do_raw_spin_unlock(&rq->lock); + preempt_disable(); + rq_unlock(rq, &rf); sched_preempt_enable_no_resched(); schedule(); @@ -4966,7 +4867,7 @@ EXPORT_SYMBOL(__cond_resched_softirq); * Typical broken usage is: * * while (!event) - * yield(); + * yield(); * * where one assumes that yield() will let 'the other' process run that will * make event true. If the current task is a SCHED_FIFO task that will never @@ -5057,31 +4958,48 @@ out_irq: } EXPORT_SYMBOL_GPL(yield_to); +int io_schedule_prepare(void) +{ + int old_iowait = current->in_iowait; + + current->in_iowait = 1; + blk_schedule_flush_plug(current); + + return old_iowait; +} + +void io_schedule_finish(int token) +{ + current->in_iowait = token; +} + /* * This task is about to go to sleep on IO. Increment rq->nr_iowait so * that process accounting knows that this is a task in IO wait state. */ long __sched io_schedule_timeout(long timeout) { - int old_iowait = current->in_iowait; - struct rq *rq; + int token; long ret; - current->in_iowait = 1; - blk_schedule_flush_plug(current); - - delayacct_blkio_start(); - rq = raw_rq(); - atomic_inc(&rq->nr_iowait); + token = io_schedule_prepare(); ret = schedule_timeout(timeout); - current->in_iowait = old_iowait; - atomic_dec(&rq->nr_iowait); - delayacct_blkio_end(); + io_schedule_finish(token); return ret; } EXPORT_SYMBOL(io_schedule_timeout); +void io_schedule(void) +{ + int token; + + token = io_schedule_prepare(); + schedule(); + io_schedule_finish(token); +} +EXPORT_SYMBOL(io_schedule); + /** * sys_sched_get_priority_max - return maximum RT priority. * @policy: scheduling class. @@ -5193,6 +5111,9 @@ void sched_show_task(struct task_struct *p) int ppid; unsigned long state = p->state; + /* Make sure the string lines up properly with the number of task states: */ + BUILD_BUG_ON(sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1); + if (!try_get_task_stack(p)) return; if (state) @@ -5264,7 +5185,7 @@ void init_idle_bootup_task(struct task_struct *idle) /** * init_idle - set up an idle thread for a given CPU * @idle: task in question - * @cpu: cpu the idle task belongs to + * @cpu: CPU the idle task belongs to * * NOTE: this function does not set the idle thread's NEED_RESCHED * flag, to make booting more robust. @@ -5295,7 +5216,7 @@ void init_idle(struct task_struct *idle, int cpu) #endif /* * We're having a chicken and egg problem, even though we are - * holding rq->lock, the cpu isn't yet set to this cpu so the + * holding rq->lock, the CPU isn't yet set to this CPU so the * lockdep check in task_group() will fail. * * Similar case to sched_fork(). / Alternatively we could @@ -5329,26 +5250,17 @@ void init_idle(struct task_struct *idle, int cpu) #endif } +#ifdef CONFIG_SMP + int cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial) { - int ret = 1, trial_cpus; - struct dl_bw *cur_dl_b; - unsigned long flags; + int ret = 1; if (!cpumask_weight(cur)) return ret; - rcu_read_lock_sched(); - cur_dl_b = dl_bw_of(cpumask_any(cur)); - trial_cpus = cpumask_weight(trial); - - raw_spin_lock_irqsave(&cur_dl_b->lock, flags); - if (cur_dl_b->bw != -1 && - cur_dl_b->bw * trial_cpus < cur_dl_b->total_bw) - ret = 0; - raw_spin_unlock_irqrestore(&cur_dl_b->lock, flags); - rcu_read_unlock_sched(); + ret = dl_cpuset_cpumask_can_shrink(cur, trial); return ret; } @@ -5360,7 +5272,7 @@ int task_can_attach(struct task_struct *p, /* * Kthreads which disallow setaffinity shouldn't be moved - * to a new cpuset; we don't want to change their cpu + * to a new cpuset; we don't want to change their CPU * affinity and isolating such threads by their set of * allowed nodes is unnecessary. Thus, cpusets are not * applicable for such threads. This prevents checking for @@ -5372,44 +5284,15 @@ int task_can_attach(struct task_struct *p, goto out; } -#ifdef CONFIG_SMP if (dl_task(p) && !cpumask_intersects(task_rq(p)->rd->span, - cs_cpus_allowed)) { - unsigned int dest_cpu = cpumask_any_and(cpu_active_mask, - cs_cpus_allowed); - struct dl_bw *dl_b; - bool overflow; - int cpus; - unsigned long flags; - - rcu_read_lock_sched(); - dl_b = dl_bw_of(dest_cpu); - raw_spin_lock_irqsave(&dl_b->lock, flags); - cpus = dl_bw_cpus(dest_cpu); - overflow = __dl_overflow(dl_b, cpus, 0, p->dl.dl_bw); - if (overflow) - ret = -EBUSY; - else { - /* - * We reserve space for this task in the destination - * root_domain, as we can't fail after this point. - * We will free resources in the source root_domain - * later on (see set_cpus_allowed_dl()). - */ - __dl_add(dl_b, p->dl.dl_bw); - } - raw_spin_unlock_irqrestore(&dl_b->lock, flags); - rcu_read_unlock_sched(); + cs_cpus_allowed)) + ret = dl_task_can_attach(p, cs_cpus_allowed); - } -#endif out: return ret; } -#ifdef CONFIG_SMP - -static bool sched_smp_initialized __read_mostly; +bool sched_smp_initialized __read_mostly; #ifdef CONFIG_NUMA_BALANCING /* Migrate current task p to target_cpu */ @@ -5421,7 +5304,7 @@ int migrate_task_to(struct task_struct *p, int target_cpu) if (curr_cpu == target_cpu) return 0; - if (!cpumask_test_cpu(target_cpu, tsk_cpus_allowed(p))) + if (!cpumask_test_cpu(target_cpu, &p->cpus_allowed)) return -EINVAL; /* TODO: This is not properly updating schedstats */ @@ -5452,7 +5335,7 @@ void sched_setnuma(struct task_struct *p, int nid) p->numa_preferred_nid = nid; if (queued) - enqueue_task(rq, p, ENQUEUE_RESTORE); + enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); if (running) set_curr_task(rq, p); task_rq_unlock(rq, p, &rf); @@ -5461,7 +5344,7 @@ void sched_setnuma(struct task_struct *p, int nid) #ifdef CONFIG_HOTPLUG_CPU /* - * Ensures that the idle task is using init_mm right before its cpu goes + * Ensure that the idle task is using init_mm right before its CPU goes * offline. */ void idle_task_exit(void) @@ -5471,7 +5354,7 @@ void idle_task_exit(void) BUG_ON(cpu_online(smp_processor_id())); if (mm != &init_mm) { - switch_mm_irqs_off(mm, &init_mm, current); + switch_mm(mm, &init_mm, current); finish_arch_post_lock_switch(); } mmdrop(mm); @@ -5517,11 +5400,11 @@ static struct task_struct fake_task = { * there's no concurrency possible, we hold the required locks anyway * because of lock validation efforts. */ -static void migrate_tasks(struct rq *dead_rq) +static void migrate_tasks(struct rq *dead_rq, struct rq_flags *rf) { struct rq *rq = dead_rq; struct task_struct *next, *stop = rq->stop; - struct pin_cookie cookie; + struct rq_flags orf = *rf; int dest_cpu; /* @@ -5545,16 +5428,15 @@ static void migrate_tasks(struct rq *dead_rq) for (;;) { /* * There's this thread running, bail when that's the only - * remaining thread. + * remaining thread: */ if (rq->nr_running == 1) break; /* - * pick_next_task assumes pinned rq->lock. + * pick_next_task() assumes pinned rq->lock: */ - cookie = lockdep_pin_lock(&rq->lock); - next = pick_next_task(rq, &fake_task, cookie); + next = pick_next_task(rq, &fake_task, rf); BUG_ON(!next); next->sched_class->put_prev_task(rq, next); @@ -5567,10 +5449,9 @@ static void migrate_tasks(struct rq *dead_rq) * because !cpu_active at this point, which means load-balance * will not interfere. Also, stop-machine. */ - lockdep_unpin_lock(&rq->lock, cookie); - raw_spin_unlock(&rq->lock); + rq_unlock(rq, rf); raw_spin_lock(&next->pi_lock); - raw_spin_lock(&rq->lock); + rq_relock(rq, rf); /* * Since we're inside stop-machine, _nothing_ should have @@ -5584,12 +5465,12 @@ static void migrate_tasks(struct rq *dead_rq) /* Find suitable destination for @next, with force if needed. */ dest_cpu = select_fallback_rq(dead_rq->cpu, next); - - rq = __migrate_task(rq, next, dest_cpu); + rq = __migrate_task(rq, rf, next, dest_cpu); if (rq != dead_rq) { - raw_spin_unlock(&rq->lock); + rq_unlock(rq, rf); rq = dead_rq; - raw_spin_lock(&rq->lock); + *rf = orf; + rq_relock(rq, rf); } raw_spin_unlock(&next->pi_lock); } @@ -5598,7 +5479,7 @@ static void migrate_tasks(struct rq *dead_rq) } #endif /* CONFIG_HOTPLUG_CPU */ -static void set_rq_online(struct rq *rq) +void set_rq_online(struct rq *rq) { if (!rq->online) { const struct sched_class *class; @@ -5613,7 +5494,7 @@ static void set_rq_online(struct rq *rq) } } -static void set_rq_offline(struct rq *rq) +void set_rq_offline(struct rq *rq) { if (rq->online) { const struct sched_class *class; @@ -5635,1647 +5516,10 @@ static void set_cpu_rq_start_time(unsigned int cpu) rq->age_stamp = sched_clock_cpu(cpu); } -static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */ - -#ifdef CONFIG_SCHED_DEBUG - -static __read_mostly int sched_debug_enabled; - -static int __init sched_debug_setup(char *str) -{ - sched_debug_enabled = 1; - - return 0; -} -early_param("sched_debug", sched_debug_setup); - -static inline bool sched_debug(void) -{ - return sched_debug_enabled; -} - -static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, - struct cpumask *groupmask) -{ - struct sched_group *group = sd->groups; - - cpumask_clear(groupmask); - - printk(KERN_DEBUG "%*s domain %d: ", level, "", level); - - if (!(sd->flags & SD_LOAD_BALANCE)) { - printk("does not load-balance\n"); - if (sd->parent) - printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain" - " has parent"); - return -1; - } - - printk(KERN_CONT "span %*pbl level %s\n", - cpumask_pr_args(sched_domain_span(sd)), sd->name); - - if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) { - printk(KERN_ERR "ERROR: domain->span does not contain " - "CPU%d\n", cpu); - } - if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) { - printk(KERN_ERR "ERROR: domain->groups does not contain" - " CPU%d\n", cpu); - } - - printk(KERN_DEBUG "%*s groups:", level + 1, ""); - do { - if (!group) { - printk("\n"); - printk(KERN_ERR "ERROR: group is NULL\n"); - break; - } - - if (!cpumask_weight(sched_group_cpus(group))) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: empty group\n"); - break; - } - - if (!(sd->flags & SD_OVERLAP) && - cpumask_intersects(groupmask, sched_group_cpus(group))) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: repeated CPUs\n"); - break; - } - - cpumask_or(groupmask, groupmask, sched_group_cpus(group)); - - printk(KERN_CONT " %*pbl", - cpumask_pr_args(sched_group_cpus(group))); - if (group->sgc->capacity != SCHED_CAPACITY_SCALE) { - printk(KERN_CONT " (cpu_capacity = %lu)", - group->sgc->capacity); - } - - group = group->next; - } while (group != sd->groups); - printk(KERN_CONT "\n"); - - if (!cpumask_equal(sched_domain_span(sd), groupmask)) - printk(KERN_ERR "ERROR: groups don't span domain->span\n"); - - if (sd->parent && - !cpumask_subset(groupmask, sched_domain_span(sd->parent))) - printk(KERN_ERR "ERROR: parent span is not a superset " - "of domain->span\n"); - return 0; -} - -static void sched_domain_debug(struct sched_domain *sd, int cpu) -{ - int level = 0; - - if (!sched_debug_enabled) - return; - - if (!sd) { - printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu); - return; - } - - printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu); - - for (;;) { - if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask)) - break; - level++; - sd = sd->parent; - if (!sd) - break; - } -} -#else /* !CONFIG_SCHED_DEBUG */ - -# define sched_debug_enabled 0 -# define sched_domain_debug(sd, cpu) do { } while (0) -static inline bool sched_debug(void) -{ - return false; -} -#endif /* CONFIG_SCHED_DEBUG */ - -static int sd_degenerate(struct sched_domain *sd) -{ - if (cpumask_weight(sched_domain_span(sd)) == 1) - return 1; - - /* Following flags need at least 2 groups */ - if (sd->flags & (SD_LOAD_BALANCE | - SD_BALANCE_NEWIDLE | - SD_BALANCE_FORK | - SD_BALANCE_EXEC | - SD_SHARE_CPUCAPACITY | - SD_ASYM_CPUCAPACITY | - SD_SHARE_PKG_RESOURCES | - SD_SHARE_POWERDOMAIN)) { - if (sd->groups != sd->groups->next) - return 0; - } - - /* Following flags don't use groups */ - if (sd->flags & (SD_WAKE_AFFINE)) - return 0; - - return 1; -} - -static int -sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) -{ - unsigned long cflags = sd->flags, pflags = parent->flags; - - if (sd_degenerate(parent)) - return 1; - - if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent))) - return 0; - - /* Flags needing groups don't count if only 1 group in parent */ - if (parent->groups == parent->groups->next) { - pflags &= ~(SD_LOAD_BALANCE | - SD_BALANCE_NEWIDLE | - SD_BALANCE_FORK | - SD_BALANCE_EXEC | - SD_ASYM_CPUCAPACITY | - SD_SHARE_CPUCAPACITY | - SD_SHARE_PKG_RESOURCES | - SD_PREFER_SIBLING | - SD_SHARE_POWERDOMAIN); - if (nr_node_ids == 1) - pflags &= ~SD_SERIALIZE; - } - if (~cflags & pflags) - return 0; - - return 1; -} - -static void free_rootdomain(struct rcu_head *rcu) -{ - struct root_domain *rd = container_of(rcu, struct root_domain, rcu); - - cpupri_cleanup(&rd->cpupri); - cpudl_cleanup(&rd->cpudl); - free_cpumask_var(rd->dlo_mask); - free_cpumask_var(rd->rto_mask); - free_cpumask_var(rd->online); - free_cpumask_var(rd->span); - kfree(rd); -} - -static void rq_attach_root(struct rq *rq, struct root_domain *rd) -{ - struct root_domain *old_rd = NULL; - unsigned long flags; - - raw_spin_lock_irqsave(&rq->lock, flags); - - if (rq->rd) { - old_rd = rq->rd; - - if (cpumask_test_cpu(rq->cpu, old_rd->online)) - set_rq_offline(rq); - - cpumask_clear_cpu(rq->cpu, old_rd->span); - - /* - * If we dont want to free the old_rd yet then - * set old_rd to NULL to skip the freeing later - * in this function: - */ - if (!atomic_dec_and_test(&old_rd->refcount)) - old_rd = NULL; - } - - atomic_inc(&rd->refcount); - rq->rd = rd; - - cpumask_set_cpu(rq->cpu, rd->span); - if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) - set_rq_online(rq); - - raw_spin_unlock_irqrestore(&rq->lock, flags); - - if (old_rd) - call_rcu_sched(&old_rd->rcu, free_rootdomain); -} - -static int init_rootdomain(struct root_domain *rd) -{ - memset(rd, 0, sizeof(*rd)); - - if (!zalloc_cpumask_var(&rd->span, GFP_KERNEL)) - goto out; - if (!zalloc_cpumask_var(&rd->online, GFP_KERNEL)) - goto free_span; - if (!zalloc_cpumask_var(&rd->dlo_mask, GFP_KERNEL)) - goto free_online; - if (!zalloc_cpumask_var(&rd->rto_mask, GFP_KERNEL)) - goto free_dlo_mask; - - init_dl_bw(&rd->dl_bw); - if (cpudl_init(&rd->cpudl) != 0) - goto free_dlo_mask; - - if (cpupri_init(&rd->cpupri) != 0) - goto free_rto_mask; - return 0; - -free_rto_mask: - free_cpumask_var(rd->rto_mask); -free_dlo_mask: - free_cpumask_var(rd->dlo_mask); -free_online: - free_cpumask_var(rd->online); -free_span: - free_cpumask_var(rd->span); -out: - return -ENOMEM; -} - -/* - * By default the system creates a single root-domain with all cpus as - * members (mimicking the global state we have today). - */ -struct root_domain def_root_domain; - -static void init_defrootdomain(void) -{ - init_rootdomain(&def_root_domain); - - atomic_set(&def_root_domain.refcount, 1); -} - -static struct root_domain *alloc_rootdomain(void) -{ - struct root_domain *rd; - - rd = kmalloc(sizeof(*rd), GFP_KERNEL); - if (!rd) - return NULL; - - if (init_rootdomain(rd) != 0) { - kfree(rd); - return NULL; - } - - return rd; -} - -static void free_sched_groups(struct sched_group *sg, int free_sgc) -{ - struct sched_group *tmp, *first; - - if (!sg) - return; - - first = sg; - do { - tmp = sg->next; - - if (free_sgc && atomic_dec_and_test(&sg->sgc->ref)) - kfree(sg->sgc); - - kfree(sg); - sg = tmp; - } while (sg != first); -} - -static void destroy_sched_domain(struct sched_domain *sd) -{ - /* - * If its an overlapping domain it has private groups, iterate and - * nuke them all. - */ - if (sd->flags & SD_OVERLAP) { - free_sched_groups(sd->groups, 1); - } else if (atomic_dec_and_test(&sd->groups->ref)) { - kfree(sd->groups->sgc); - kfree(sd->groups); - } - if (sd->shared && atomic_dec_and_test(&sd->shared->ref)) - kfree(sd->shared); - kfree(sd); -} - -static void destroy_sched_domains_rcu(struct rcu_head *rcu) -{ - struct sched_domain *sd = container_of(rcu, struct sched_domain, rcu); - - while (sd) { - struct sched_domain *parent = sd->parent; - destroy_sched_domain(sd); - sd = parent; - } -} - -static void destroy_sched_domains(struct sched_domain *sd) -{ - if (sd) - call_rcu(&sd->rcu, destroy_sched_domains_rcu); -} - -/* - * Keep a special pointer to the highest sched_domain that has - * SD_SHARE_PKG_RESOURCE set (Last Level Cache Domain) for this - * allows us to avoid some pointer chasing select_idle_sibling(). - * - * Also keep a unique ID per domain (we use the first cpu number in - * the cpumask of the domain), this allows us to quickly tell if - * two cpus are in the same cache domain, see cpus_share_cache(). - */ -DEFINE_PER_CPU(struct sched_domain *, sd_llc); -DEFINE_PER_CPU(int, sd_llc_size); -DEFINE_PER_CPU(int, sd_llc_id); -DEFINE_PER_CPU(struct sched_domain_shared *, sd_llc_shared); -DEFINE_PER_CPU(struct sched_domain *, sd_numa); -DEFINE_PER_CPU(struct sched_domain *, sd_asym); - -static void update_top_cache_domain(int cpu) -{ - struct sched_domain_shared *sds = NULL; - struct sched_domain *sd; - int id = cpu; - int size = 1; - - sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES); - if (sd) { - id = cpumask_first(sched_domain_span(sd)); - size = cpumask_weight(sched_domain_span(sd)); - sds = sd->shared; - } - - rcu_assign_pointer(per_cpu(sd_llc, cpu), sd); - per_cpu(sd_llc_size, cpu) = size; - per_cpu(sd_llc_id, cpu) = id; - rcu_assign_pointer(per_cpu(sd_llc_shared, cpu), sds); - - sd = lowest_flag_domain(cpu, SD_NUMA); - rcu_assign_pointer(per_cpu(sd_numa, cpu), sd); - - sd = highest_flag_domain(cpu, SD_ASYM_PACKING); - rcu_assign_pointer(per_cpu(sd_asym, cpu), sd); -} - -/* - * Attach the domain 'sd' to 'cpu' as its base domain. Callers must - * hold the hotplug lock. - */ -static void -cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) -{ - struct rq *rq = cpu_rq(cpu); - struct sched_domain *tmp; - - /* Remove the sched domains which do not contribute to scheduling. */ - for (tmp = sd; tmp; ) { - struct sched_domain *parent = tmp->parent; - if (!parent) - break; - - if (sd_parent_degenerate(tmp, parent)) { - tmp->parent = parent->parent; - if (parent->parent) - parent->parent->child = tmp; - /* - * Transfer SD_PREFER_SIBLING down in case of a - * degenerate parent; the spans match for this - * so the property transfers. - */ - if (parent->flags & SD_PREFER_SIBLING) - tmp->flags |= SD_PREFER_SIBLING; - destroy_sched_domain(parent); - } else - tmp = tmp->parent; - } - - if (sd && sd_degenerate(sd)) { - tmp = sd; - sd = sd->parent; - destroy_sched_domain(tmp); - if (sd) - sd->child = NULL; - } - - sched_domain_debug(sd, cpu); - - rq_attach_root(rq, rd); - tmp = rq->sd; - rcu_assign_pointer(rq->sd, sd); - destroy_sched_domains(tmp); - - update_top_cache_domain(cpu); -} - -/* Setup the mask of cpus configured for isolated domains */ -static int __init isolated_cpu_setup(char *str) -{ - int ret; - - alloc_bootmem_cpumask_var(&cpu_isolated_map); - ret = cpulist_parse(str, cpu_isolated_map); - if (ret) { - pr_err("sched: Error, all isolcpus= values must be between 0 and %d\n", nr_cpu_ids); - return 0; - } - return 1; -} -__setup("isolcpus=", isolated_cpu_setup); - -struct s_data { - struct sched_domain ** __percpu sd; - struct root_domain *rd; -}; - -enum s_alloc { - sa_rootdomain, - sa_sd, - sa_sd_storage, - sa_none, -}; - -/* - * Build an iteration mask that can exclude certain CPUs from the upwards - * domain traversal. - * - * Asymmetric node setups can result in situations where the domain tree is of - * unequal depth, make sure to skip domains that already cover the entire - * range. - * - * In that case build_sched_domains() will have terminated the iteration early - * and our sibling sd spans will be empty. Domains should always include the - * cpu they're built on, so check that. - * - */ -static void build_group_mask(struct sched_domain *sd, struct sched_group *sg) -{ - const struct cpumask *span = sched_domain_span(sd); - struct sd_data *sdd = sd->private; - struct sched_domain *sibling; - int i; - - for_each_cpu(i, span) { - sibling = *per_cpu_ptr(sdd->sd, i); - if (!cpumask_test_cpu(i, sched_domain_span(sibling))) - continue; - - cpumask_set_cpu(i, sched_group_mask(sg)); - } -} - -/* - * Return the canonical balance cpu for this group, this is the first cpu - * of this group that's also in the iteration mask. - */ -int group_balance_cpu(struct sched_group *sg) -{ - return cpumask_first_and(sched_group_cpus(sg), sched_group_mask(sg)); -} - -static int -build_overlap_sched_groups(struct sched_domain *sd, int cpu) -{ - struct sched_group *first = NULL, *last = NULL, *groups = NULL, *sg; - const struct cpumask *span = sched_domain_span(sd); - struct cpumask *covered = sched_domains_tmpmask; - struct sd_data *sdd = sd->private; - struct sched_domain *sibling; - int i; - - cpumask_clear(covered); - - for_each_cpu(i, span) { - struct cpumask *sg_span; - - if (cpumask_test_cpu(i, covered)) - continue; - - sibling = *per_cpu_ptr(sdd->sd, i); - - /* See the comment near build_group_mask(). */ - if (!cpumask_test_cpu(i, sched_domain_span(sibling))) - continue; - - sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), - GFP_KERNEL, cpu_to_node(cpu)); - - if (!sg) - goto fail; - - sg_span = sched_group_cpus(sg); - if (sibling->child) - cpumask_copy(sg_span, sched_domain_span(sibling->child)); - else - cpumask_set_cpu(i, sg_span); - - cpumask_or(covered, covered, sg_span); - - sg->sgc = *per_cpu_ptr(sdd->sgc, i); - if (atomic_inc_return(&sg->sgc->ref) == 1) - build_group_mask(sd, sg); - - /* - * Initialize sgc->capacity such that even if we mess up the - * domains and no possible iteration will get us here, we won't - * die on a /0 trap. - */ - sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sg_span); - sg->sgc->min_capacity = SCHED_CAPACITY_SCALE; - - /* - * Make sure the first group of this domain contains the - * canonical balance cpu. Otherwise the sched_domain iteration - * breaks. See update_sg_lb_stats(). - */ - if ((!groups && cpumask_test_cpu(cpu, sg_span)) || - group_balance_cpu(sg) == cpu) - groups = sg; - - if (!first) - first = sg; - if (last) - last->next = sg; - last = sg; - last->next = first; - } - sd->groups = groups; - - return 0; - -fail: - free_sched_groups(first, 0); - - return -ENOMEM; -} - -static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg) -{ - struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); - struct sched_domain *child = sd->child; - - if (child) - cpu = cpumask_first(sched_domain_span(child)); - - if (sg) { - *sg = *per_cpu_ptr(sdd->sg, cpu); - (*sg)->sgc = *per_cpu_ptr(sdd->sgc, cpu); - atomic_set(&(*sg)->sgc->ref, 1); /* for claim_allocations */ - } - - return cpu; -} - -/* - * build_sched_groups will build a circular linked list of the groups - * covered by the given span, and will set each group's ->cpumask correctly, - * and ->cpu_capacity to 0. - * - * Assumes the sched_domain tree is fully constructed - */ -static int -build_sched_groups(struct sched_domain *sd, int cpu) -{ - struct sched_group *first = NULL, *last = NULL; - struct sd_data *sdd = sd->private; - const struct cpumask *span = sched_domain_span(sd); - struct cpumask *covered; - int i; - - get_group(cpu, sdd, &sd->groups); - atomic_inc(&sd->groups->ref); - - if (cpu != cpumask_first(span)) - return 0; - - lockdep_assert_held(&sched_domains_mutex); - covered = sched_domains_tmpmask; - - cpumask_clear(covered); - - for_each_cpu(i, span) { - struct sched_group *sg; - int group, j; - - if (cpumask_test_cpu(i, covered)) - continue; - - group = get_group(i, sdd, &sg); - cpumask_setall(sched_group_mask(sg)); - - for_each_cpu(j, span) { - if (get_group(j, sdd, NULL) != group) - continue; - - cpumask_set_cpu(j, covered); - cpumask_set_cpu(j, sched_group_cpus(sg)); - } - - if (!first) - first = sg; - if (last) - last->next = sg; - last = sg; - } - last->next = first; - - return 0; -} - -/* - * Initialize sched groups cpu_capacity. - * - * cpu_capacity indicates the capacity of sched group, which is used while - * distributing the load between different sched groups in a sched domain. - * Typically cpu_capacity for all the groups in a sched domain will be same - * unless there are asymmetries in the topology. If there are asymmetries, - * group having more cpu_capacity will pickup more load compared to the - * group having less cpu_capacity. - */ -static void init_sched_groups_capacity(int cpu, struct sched_domain *sd) -{ - struct sched_group *sg = sd->groups; - - WARN_ON(!sg); - - do { - int cpu, max_cpu = -1; - - sg->group_weight = cpumask_weight(sched_group_cpus(sg)); - - if (!(sd->flags & SD_ASYM_PACKING)) - goto next; - - for_each_cpu(cpu, sched_group_cpus(sg)) { - if (max_cpu < 0) - max_cpu = cpu; - else if (sched_asym_prefer(cpu, max_cpu)) - max_cpu = cpu; - } - sg->asym_prefer_cpu = max_cpu; - -next: - sg = sg->next; - } while (sg != sd->groups); - - if (cpu != group_balance_cpu(sg)) - return; - - update_group_capacity(sd, cpu); -} - -/* - * Initializers for schedule domains - * Non-inlined to reduce accumulated stack pressure in build_sched_domains() - */ - -static int default_relax_domain_level = -1; -int sched_domain_level_max; - -static int __init setup_relax_domain_level(char *str) -{ - if (kstrtoint(str, 0, &default_relax_domain_level)) - pr_warn("Unable to set relax_domain_level\n"); - - return 1; -} -__setup("relax_domain_level=", setup_relax_domain_level); - -static void set_domain_attribute(struct sched_domain *sd, - struct sched_domain_attr *attr) -{ - int request; - - if (!attr || attr->relax_domain_level < 0) { - if (default_relax_domain_level < 0) - return; - else - request = default_relax_domain_level; - } else - request = attr->relax_domain_level; - if (request < sd->level) { - /* turn off idle balance on this domain */ - sd->flags &= ~(SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); - } else { - /* turn on idle balance on this domain */ - sd->flags |= (SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); - } -} - -static void __sdt_free(const struct cpumask *cpu_map); -static int __sdt_alloc(const struct cpumask *cpu_map); - -static void __free_domain_allocs(struct s_data *d, enum s_alloc what, - const struct cpumask *cpu_map) -{ - switch (what) { - case sa_rootdomain: - if (!atomic_read(&d->rd->refcount)) - free_rootdomain(&d->rd->rcu); /* fall through */ - case sa_sd: - free_percpu(d->sd); /* fall through */ - case sa_sd_storage: - __sdt_free(cpu_map); /* fall through */ - case sa_none: - break; - } -} - -static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, - const struct cpumask *cpu_map) -{ - memset(d, 0, sizeof(*d)); - - if (__sdt_alloc(cpu_map)) - return sa_sd_storage; - d->sd = alloc_percpu(struct sched_domain *); - if (!d->sd) - return sa_sd_storage; - d->rd = alloc_rootdomain(); - if (!d->rd) - return sa_sd; - return sa_rootdomain; -} - -/* - * NULL the sd_data elements we've used to build the sched_domain and - * sched_group structure so that the subsequent __free_domain_allocs() - * will not free the data we're using. - */ -static void claim_allocations(int cpu, struct sched_domain *sd) -{ - struct sd_data *sdd = sd->private; - - WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd); - *per_cpu_ptr(sdd->sd, cpu) = NULL; - - if (atomic_read(&(*per_cpu_ptr(sdd->sds, cpu))->ref)) - *per_cpu_ptr(sdd->sds, cpu) = NULL; - - if (atomic_read(&(*per_cpu_ptr(sdd->sg, cpu))->ref)) - *per_cpu_ptr(sdd->sg, cpu) = NULL; - - if (atomic_read(&(*per_cpu_ptr(sdd->sgc, cpu))->ref)) - *per_cpu_ptr(sdd->sgc, cpu) = NULL; -} - -#ifdef CONFIG_NUMA -static int sched_domains_numa_levels; -enum numa_topology_type sched_numa_topology_type; -static int *sched_domains_numa_distance; -int sched_max_numa_distance; -static struct cpumask ***sched_domains_numa_masks; -static int sched_domains_curr_level; -#endif - -/* - * SD_flags allowed in topology descriptions. - * - * These flags are purely descriptive of the topology and do not prescribe - * behaviour. Behaviour is artificial and mapped in the below sd_init() - * function: - * - * SD_SHARE_CPUCAPACITY - describes SMT topologies - * SD_SHARE_PKG_RESOURCES - describes shared caches - * SD_NUMA - describes NUMA topologies - * SD_SHARE_POWERDOMAIN - describes shared power domain - * SD_ASYM_CPUCAPACITY - describes mixed capacity topologies - * - * Odd one out, which beside describing the topology has a quirk also - * prescribes the desired behaviour that goes along with it: - * - * SD_ASYM_PACKING - describes SMT quirks - */ -#define TOPOLOGY_SD_FLAGS \ - (SD_SHARE_CPUCAPACITY | \ - SD_SHARE_PKG_RESOURCES | \ - SD_NUMA | \ - SD_ASYM_PACKING | \ - SD_ASYM_CPUCAPACITY | \ - SD_SHARE_POWERDOMAIN) - -static struct sched_domain * -sd_init(struct sched_domain_topology_level *tl, - const struct cpumask *cpu_map, - struct sched_domain *child, int cpu) -{ - struct sd_data *sdd = &tl->data; - struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); - int sd_id, sd_weight, sd_flags = 0; - -#ifdef CONFIG_NUMA - /* - * Ugly hack to pass state to sd_numa_mask()... - */ - sched_domains_curr_level = tl->numa_level; -#endif - - sd_weight = cpumask_weight(tl->mask(cpu)); - - if (tl->sd_flags) - sd_flags = (*tl->sd_flags)(); - if (WARN_ONCE(sd_flags & ~TOPOLOGY_SD_FLAGS, - "wrong sd_flags in topology description\n")) - sd_flags &= ~TOPOLOGY_SD_FLAGS; - - *sd = (struct sched_domain){ - .min_interval = sd_weight, - .max_interval = 2*sd_weight, - .busy_factor = 32, - .imbalance_pct = 125, - - .cache_nice_tries = 0, - .busy_idx = 0, - .idle_idx = 0, - .newidle_idx = 0, - .wake_idx = 0, - .forkexec_idx = 0, - - .flags = 1*SD_LOAD_BALANCE - | 1*SD_BALANCE_NEWIDLE - | 1*SD_BALANCE_EXEC - | 1*SD_BALANCE_FORK - | 0*SD_BALANCE_WAKE - | 1*SD_WAKE_AFFINE - | 0*SD_SHARE_CPUCAPACITY - | 0*SD_SHARE_PKG_RESOURCES - | 0*SD_SERIALIZE - | 0*SD_PREFER_SIBLING - | 0*SD_NUMA - | sd_flags - , - - .last_balance = jiffies, - .balance_interval = sd_weight, - .smt_gain = 0, - .max_newidle_lb_cost = 0, - .next_decay_max_lb_cost = jiffies, - .child = child, -#ifdef CONFIG_SCHED_DEBUG - .name = tl->name, -#endif - }; - - cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu)); - sd_id = cpumask_first(sched_domain_span(sd)); - - /* - * Convert topological properties into behaviour. - */ - - if (sd->flags & SD_ASYM_CPUCAPACITY) { - struct sched_domain *t = sd; - - for_each_lower_domain(t) - t->flags |= SD_BALANCE_WAKE; - } - - if (sd->flags & SD_SHARE_CPUCAPACITY) { - sd->flags |= SD_PREFER_SIBLING; - sd->imbalance_pct = 110; - sd->smt_gain = 1178; /* ~15% */ - - } else if (sd->flags & SD_SHARE_PKG_RESOURCES) { - sd->imbalance_pct = 117; - sd->cache_nice_tries = 1; - sd->busy_idx = 2; - -#ifdef CONFIG_NUMA - } else if (sd->flags & SD_NUMA) { - sd->cache_nice_tries = 2; - sd->busy_idx = 3; - sd->idle_idx = 2; - - sd->flags |= SD_SERIALIZE; - if (sched_domains_numa_distance[tl->numa_level] > RECLAIM_DISTANCE) { - sd->flags &= ~(SD_BALANCE_EXEC | - SD_BALANCE_FORK | - SD_WAKE_AFFINE); - } - -#endif - } else { - sd->flags |= SD_PREFER_SIBLING; - sd->cache_nice_tries = 1; - sd->busy_idx = 2; - sd->idle_idx = 1; - } - - /* - * For all levels sharing cache; connect a sched_domain_shared - * instance. - */ - if (sd->flags & SD_SHARE_PKG_RESOURCES) { - sd->shared = *per_cpu_ptr(sdd->sds, sd_id); - atomic_inc(&sd->shared->ref); - atomic_set(&sd->shared->nr_busy_cpus, sd_weight); - } - - sd->private = sdd; - - return sd; -} - -/* - * Topology list, bottom-up. - */ -static struct sched_domain_topology_level default_topology[] = { -#ifdef CONFIG_SCHED_SMT - { cpu_smt_mask, cpu_smt_flags, SD_INIT_NAME(SMT) }, -#endif -#ifdef CONFIG_SCHED_MC - { cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) }, -#endif - { cpu_cpu_mask, SD_INIT_NAME(DIE) }, - { NULL, }, -}; - -static struct sched_domain_topology_level *sched_domain_topology = - default_topology; - -#define for_each_sd_topology(tl) \ - for (tl = sched_domain_topology; tl->mask; tl++) - -void set_sched_topology(struct sched_domain_topology_level *tl) -{ - if (WARN_ON_ONCE(sched_smp_initialized)) - return; - - sched_domain_topology = tl; -} - -#ifdef CONFIG_NUMA - -static const struct cpumask *sd_numa_mask(int cpu) -{ - return sched_domains_numa_masks[sched_domains_curr_level][cpu_to_node(cpu)]; -} - -static void sched_numa_warn(const char *str) -{ - static int done = false; - int i,j; - - if (done) - return; - - done = true; - - printk(KERN_WARNING "ERROR: %s\n\n", str); - - for (i = 0; i < nr_node_ids; i++) { - printk(KERN_WARNING " "); - for (j = 0; j < nr_node_ids; j++) - printk(KERN_CONT "%02d ", node_distance(i,j)); - printk(KERN_CONT "\n"); - } - printk(KERN_WARNING "\n"); -} - -bool find_numa_distance(int distance) -{ - int i; - - if (distance == node_distance(0, 0)) - return true; - - for (i = 0; i < sched_domains_numa_levels; i++) { - if (sched_domains_numa_distance[i] == distance) - return true; - } - - return false; -} - -/* - * A system can have three types of NUMA topology: - * NUMA_DIRECT: all nodes are directly connected, or not a NUMA system - * NUMA_GLUELESS_MESH: some nodes reachable through intermediary nodes - * NUMA_BACKPLANE: nodes can reach other nodes through a backplane - * - * The difference between a glueless mesh topology and a backplane - * topology lies in whether communication between not directly - * connected nodes goes through intermediary nodes (where programs - * could run), or through backplane controllers. This affects - * placement of programs. - * - * The type of topology can be discerned with the following tests: - * - If the maximum distance between any nodes is 1 hop, the system - * is directly connected. - * - If for two nodes A and B, located N > 1 hops away from each other, - * there is an intermediary node C, which is < N hops away from both - * nodes A and B, the system is a glueless mesh. - */ -static void init_numa_topology_type(void) -{ - int a, b, c, n; - - n = sched_max_numa_distance; - - if (sched_domains_numa_levels <= 1) { - sched_numa_topology_type = NUMA_DIRECT; - return; - } - - for_each_online_node(a) { - for_each_online_node(b) { - /* Find two nodes furthest removed from each other. */ - if (node_distance(a, b) < n) - continue; - - /* Is there an intermediary node between a and b? */ - for_each_online_node(c) { - if (node_distance(a, c) < n && - node_distance(b, c) < n) { - sched_numa_topology_type = - NUMA_GLUELESS_MESH; - return; - } - } - - sched_numa_topology_type = NUMA_BACKPLANE; - return; - } - } -} - -static void sched_init_numa(void) -{ - int next_distance, curr_distance = node_distance(0, 0); - struct sched_domain_topology_level *tl; - int level = 0; - int i, j, k; - - sched_domains_numa_distance = kzalloc(sizeof(int) * nr_node_ids, GFP_KERNEL); - if (!sched_domains_numa_distance) - return; - - /* - * O(nr_nodes^2) deduplicating selection sort -- in order to find the - * unique distances in the node_distance() table. - * - * Assumes node_distance(0,j) includes all distances in - * node_distance(i,j) in order to avoid cubic time. - */ - next_distance = curr_distance; - for (i = 0; i < nr_node_ids; i++) { - for (j = 0; j < nr_node_ids; j++) { - for (k = 0; k < nr_node_ids; k++) { - int distance = node_distance(i, k); - - if (distance > curr_distance && - (distance < next_distance || - next_distance == curr_distance)) - next_distance = distance; - - /* - * While not a strong assumption it would be nice to know - * about cases where if node A is connected to B, B is not - * equally connected to A. - */ - if (sched_debug() && node_distance(k, i) != distance) - sched_numa_warn("Node-distance not symmetric"); - - if (sched_debug() && i && !find_numa_distance(distance)) - sched_numa_warn("Node-0 not representative"); - } - if (next_distance != curr_distance) { - sched_domains_numa_distance[level++] = next_distance; - sched_domains_numa_levels = level; - curr_distance = next_distance; - } else break; - } - - /* - * In case of sched_debug() we verify the above assumption. - */ - if (!sched_debug()) - break; - } - - if (!level) - return; - - /* - * 'level' contains the number of unique distances, excluding the - * identity distance node_distance(i,i). - * - * The sched_domains_numa_distance[] array includes the actual distance - * numbers. - */ - - /* - * Here, we should temporarily reset sched_domains_numa_levels to 0. - * If it fails to allocate memory for array sched_domains_numa_masks[][], - * the array will contain less then 'level' members. This could be - * dangerous when we use it to iterate array sched_domains_numa_masks[][] - * in other functions. - * - * We reset it to 'level' at the end of this function. - */ - sched_domains_numa_levels = 0; - - sched_domains_numa_masks = kzalloc(sizeof(void *) * level, GFP_KERNEL); - if (!sched_domains_numa_masks) - return; - - /* - * Now for each level, construct a mask per node which contains all - * cpus of nodes that are that many hops away from us. - */ - for (i = 0; i < level; i++) { - sched_domains_numa_masks[i] = - kzalloc(nr_node_ids * sizeof(void *), GFP_KERNEL); - if (!sched_domains_numa_masks[i]) - return; - - for (j = 0; j < nr_node_ids; j++) { - struct cpumask *mask = kzalloc(cpumask_size(), GFP_KERNEL); - if (!mask) - return; - - sched_domains_numa_masks[i][j] = mask; - - for_each_node(k) { - if (node_distance(j, k) > sched_domains_numa_distance[i]) - continue; - - cpumask_or(mask, mask, cpumask_of_node(k)); - } - } - } - - /* Compute default topology size */ - for (i = 0; sched_domain_topology[i].mask; i++); - - tl = kzalloc((i + level + 1) * - sizeof(struct sched_domain_topology_level), GFP_KERNEL); - if (!tl) - return; - - /* - * Copy the default topology bits.. - */ - for (i = 0; sched_domain_topology[i].mask; i++) - tl[i] = sched_domain_topology[i]; - - /* - * .. and append 'j' levels of NUMA goodness. - */ - for (j = 0; j < level; i++, j++) { - tl[i] = (struct sched_domain_topology_level){ - .mask = sd_numa_mask, - .sd_flags = cpu_numa_flags, - .flags = SDTL_OVERLAP, - .numa_level = j, - SD_INIT_NAME(NUMA) - }; - } - - sched_domain_topology = tl; - - sched_domains_numa_levels = level; - sched_max_numa_distance = sched_domains_numa_distance[level - 1]; - - init_numa_topology_type(); -} - -static void sched_domains_numa_masks_set(unsigned int cpu) -{ - int node = cpu_to_node(cpu); - int i, j; - - for (i = 0; i < sched_domains_numa_levels; i++) { - for (j = 0; j < nr_node_ids; j++) { - if (node_distance(j, node) <= sched_domains_numa_distance[i]) - cpumask_set_cpu(cpu, sched_domains_numa_masks[i][j]); - } - } -} - -static void sched_domains_numa_masks_clear(unsigned int cpu) -{ - int i, j; - - for (i = 0; i < sched_domains_numa_levels; i++) { - for (j = 0; j < nr_node_ids; j++) - cpumask_clear_cpu(cpu, sched_domains_numa_masks[i][j]); - } -} - -#else -static inline void sched_init_numa(void) { } -static void sched_domains_numa_masks_set(unsigned int cpu) { } -static void sched_domains_numa_masks_clear(unsigned int cpu) { } -#endif /* CONFIG_NUMA */ - -static int __sdt_alloc(const struct cpumask *cpu_map) -{ - struct sched_domain_topology_level *tl; - int j; - - for_each_sd_topology(tl) { - struct sd_data *sdd = &tl->data; - - sdd->sd = alloc_percpu(struct sched_domain *); - if (!sdd->sd) - return -ENOMEM; - - sdd->sds = alloc_percpu(struct sched_domain_shared *); - if (!sdd->sds) - return -ENOMEM; - - sdd->sg = alloc_percpu(struct sched_group *); - if (!sdd->sg) - return -ENOMEM; - - sdd->sgc = alloc_percpu(struct sched_group_capacity *); - if (!sdd->sgc) - return -ENOMEM; - - for_each_cpu(j, cpu_map) { - struct sched_domain *sd; - struct sched_domain_shared *sds; - struct sched_group *sg; - struct sched_group_capacity *sgc; - - sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(), - GFP_KERNEL, cpu_to_node(j)); - if (!sd) - return -ENOMEM; - - *per_cpu_ptr(sdd->sd, j) = sd; - - sds = kzalloc_node(sizeof(struct sched_domain_shared), - GFP_KERNEL, cpu_to_node(j)); - if (!sds) - return -ENOMEM; - - *per_cpu_ptr(sdd->sds, j) = sds; - - sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), - GFP_KERNEL, cpu_to_node(j)); - if (!sg) - return -ENOMEM; - - sg->next = sg; - - *per_cpu_ptr(sdd->sg, j) = sg; - - sgc = kzalloc_node(sizeof(struct sched_group_capacity) + cpumask_size(), - GFP_KERNEL, cpu_to_node(j)); - if (!sgc) - return -ENOMEM; - - *per_cpu_ptr(sdd->sgc, j) = sgc; - } - } - - return 0; -} - -static void __sdt_free(const struct cpumask *cpu_map) -{ - struct sched_domain_topology_level *tl; - int j; - - for_each_sd_topology(tl) { - struct sd_data *sdd = &tl->data; - - for_each_cpu(j, cpu_map) { - struct sched_domain *sd; - - if (sdd->sd) { - sd = *per_cpu_ptr(sdd->sd, j); - if (sd && (sd->flags & SD_OVERLAP)) - free_sched_groups(sd->groups, 0); - kfree(*per_cpu_ptr(sdd->sd, j)); - } - - if (sdd->sds) - kfree(*per_cpu_ptr(sdd->sds, j)); - if (sdd->sg) - kfree(*per_cpu_ptr(sdd->sg, j)); - if (sdd->sgc) - kfree(*per_cpu_ptr(sdd->sgc, j)); - } - free_percpu(sdd->sd); - sdd->sd = NULL; - free_percpu(sdd->sds); - sdd->sds = NULL; - free_percpu(sdd->sg); - sdd->sg = NULL; - free_percpu(sdd->sgc); - sdd->sgc = NULL; - } -} - -struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl, - const struct cpumask *cpu_map, struct sched_domain_attr *attr, - struct sched_domain *child, int cpu) -{ - struct sched_domain *sd = sd_init(tl, cpu_map, child, cpu); - - if (child) { - sd->level = child->level + 1; - sched_domain_level_max = max(sched_domain_level_max, sd->level); - child->parent = sd; - - if (!cpumask_subset(sched_domain_span(child), - sched_domain_span(sd))) { - pr_err("BUG: arch topology borken\n"); -#ifdef CONFIG_SCHED_DEBUG - pr_err(" the %s domain not a subset of the %s domain\n", - child->name, sd->name); -#endif - /* Fixup, ensure @sd has at least @child cpus. */ - cpumask_or(sched_domain_span(sd), - sched_domain_span(sd), - sched_domain_span(child)); - } - - } - set_domain_attribute(sd, attr); - - return sd; -} - -/* - * Build sched domains for a given set of cpus and attach the sched domains - * to the individual cpus - */ -static int build_sched_domains(const struct cpumask *cpu_map, - struct sched_domain_attr *attr) -{ - enum s_alloc alloc_state; - struct sched_domain *sd; - struct s_data d; - struct rq *rq = NULL; - int i, ret = -ENOMEM; - - alloc_state = __visit_domain_allocation_hell(&d, cpu_map); - if (alloc_state != sa_rootdomain) - goto error; - - /* Set up domains for cpus specified by the cpu_map. */ - for_each_cpu(i, cpu_map) { - struct sched_domain_topology_level *tl; - - sd = NULL; - for_each_sd_topology(tl) { - sd = build_sched_domain(tl, cpu_map, attr, sd, i); - if (tl == sched_domain_topology) - *per_cpu_ptr(d.sd, i) = sd; - if (tl->flags & SDTL_OVERLAP || sched_feat(FORCE_SD_OVERLAP)) - sd->flags |= SD_OVERLAP; - if (cpumask_equal(cpu_map, sched_domain_span(sd))) - break; - } - } - - /* Build the groups for the domains */ - for_each_cpu(i, cpu_map) { - for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { - sd->span_weight = cpumask_weight(sched_domain_span(sd)); - if (sd->flags & SD_OVERLAP) { - if (build_overlap_sched_groups(sd, i)) - goto error; - } else { - if (build_sched_groups(sd, i)) - goto error; - } - } - } - - /* Calculate CPU capacity for physical packages and nodes */ - for (i = nr_cpumask_bits-1; i >= 0; i--) { - if (!cpumask_test_cpu(i, cpu_map)) - continue; - - for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { - claim_allocations(i, sd); - init_sched_groups_capacity(i, sd); - } - } - - /* Attach the domains */ - rcu_read_lock(); - for_each_cpu(i, cpu_map) { - rq = cpu_rq(i); - sd = *per_cpu_ptr(d.sd, i); - - /* Use READ_ONCE()/WRITE_ONCE() to avoid load/store tearing: */ - if (rq->cpu_capacity_orig > READ_ONCE(d.rd->max_cpu_capacity)) - WRITE_ONCE(d.rd->max_cpu_capacity, rq->cpu_capacity_orig); - - cpu_attach_domain(sd, d.rd, i); - } - rcu_read_unlock(); - - if (rq && sched_debug_enabled) { - pr_info("span: %*pbl (max cpu_capacity = %lu)\n", - cpumask_pr_args(cpu_map), rq->rd->max_cpu_capacity); - } - - ret = 0; -error: - __free_domain_allocs(&d, alloc_state, cpu_map); - return ret; -} - -static cpumask_var_t *doms_cur; /* current sched domains */ -static int ndoms_cur; /* number of sched domains in 'doms_cur' */ -static struct sched_domain_attr *dattr_cur; - /* attribues of custom domains in 'doms_cur' */ - -/* - * Special case: If a kmalloc of a doms_cur partition (array of - * cpumask) fails, then fallback to a single sched domain, - * as determined by the single cpumask fallback_doms. - */ -static cpumask_var_t fallback_doms; - -/* - * arch_update_cpu_topology lets virtualized architectures update the - * cpu core maps. It is supposed to return 1 if the topology changed - * or 0 if it stayed the same. - */ -int __weak arch_update_cpu_topology(void) -{ - return 0; -} - -cpumask_var_t *alloc_sched_domains(unsigned int ndoms) -{ - int i; - cpumask_var_t *doms; - - doms = kmalloc(sizeof(*doms) * ndoms, GFP_KERNEL); - if (!doms) - return NULL; - for (i = 0; i < ndoms; i++) { - if (!alloc_cpumask_var(&doms[i], GFP_KERNEL)) { - free_sched_domains(doms, i); - return NULL; - } - } - return doms; -} - -void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms) -{ - unsigned int i; - for (i = 0; i < ndoms; i++) - free_cpumask_var(doms[i]); - kfree(doms); -} - /* - * Set up scheduler domains and groups. Callers must hold the hotplug lock. - * For now this just excludes isolated cpus, but could be used to - * exclude other special cases in the future. + * used to mark begin/end of suspend/resume: */ -static int init_sched_domains(const struct cpumask *cpu_map) -{ - int err; - - arch_update_cpu_topology(); - ndoms_cur = 1; - doms_cur = alloc_sched_domains(ndoms_cur); - if (!doms_cur) - doms_cur = &fallback_doms; - cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map); - err = build_sched_domains(doms_cur[0], NULL); - register_sched_domain_sysctl(); - - return err; -} - -/* - * Detach sched domains from a group of cpus specified in cpu_map - * These cpus will now be attached to the NULL domain - */ -static void detach_destroy_domains(const struct cpumask *cpu_map) -{ - int i; - - rcu_read_lock(); - for_each_cpu(i, cpu_map) - cpu_attach_domain(NULL, &def_root_domain, i); - rcu_read_unlock(); -} - -/* handle null as "default" */ -static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, - struct sched_domain_attr *new, int idx_new) -{ - struct sched_domain_attr tmp; - - /* fast path */ - if (!new && !cur) - return 1; - - tmp = SD_ATTR_INIT; - return !memcmp(cur ? (cur + idx_cur) : &tmp, - new ? (new + idx_new) : &tmp, - sizeof(struct sched_domain_attr)); -} - -/* - * Partition sched domains as specified by the 'ndoms_new' - * cpumasks in the array doms_new[] of cpumasks. This compares - * doms_new[] to the current sched domain partitioning, doms_cur[]. - * It destroys each deleted domain and builds each new domain. - * - * 'doms_new' is an array of cpumask_var_t's of length 'ndoms_new'. - * The masks don't intersect (don't overlap.) We should setup one - * sched domain for each mask. CPUs not in any of the cpumasks will - * not be load balanced. If the same cpumask appears both in the - * current 'doms_cur' domains and in the new 'doms_new', we can leave - * it as it is. - * - * The passed in 'doms_new' should be allocated using - * alloc_sched_domains. This routine takes ownership of it and will - * free_sched_domains it when done with it. If the caller failed the - * alloc call, then it can pass in doms_new == NULL && ndoms_new == 1, - * and partition_sched_domains() will fallback to the single partition - * 'fallback_doms', it also forces the domains to be rebuilt. - * - * If doms_new == NULL it will be replaced with cpu_online_mask. - * ndoms_new == 0 is a special case for destroying existing domains, - * and it will not create the default domain. - * - * Call with hotplug lock held - */ -void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], - struct sched_domain_attr *dattr_new) -{ - int i, j, n; - int new_topology; - - mutex_lock(&sched_domains_mutex); - - /* always unregister in case we don't destroy any domains */ - unregister_sched_domain_sysctl(); - - /* Let architecture update cpu core mappings. */ - new_topology = arch_update_cpu_topology(); - - n = doms_new ? ndoms_new : 0; - - /* Destroy deleted domains */ - for (i = 0; i < ndoms_cur; i++) { - for (j = 0; j < n && !new_topology; j++) { - if (cpumask_equal(doms_cur[i], doms_new[j]) - && dattrs_equal(dattr_cur, i, dattr_new, j)) - goto match1; - } - /* no match - a current sched domain not in new doms_new[] */ - detach_destroy_domains(doms_cur[i]); -match1: - ; - } - - n = ndoms_cur; - if (doms_new == NULL) { - n = 0; - doms_new = &fallback_doms; - cpumask_andnot(doms_new[0], cpu_active_mask, cpu_isolated_map); - WARN_ON_ONCE(dattr_new); - } - - /* Build new domains */ - for (i = 0; i < ndoms_new; i++) { - for (j = 0; j < n && !new_topology; j++) { - if (cpumask_equal(doms_new[i], doms_cur[j]) - && dattrs_equal(dattr_new, i, dattr_cur, j)) - goto match2; - } - /* no match - add a new doms_new */ - build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL); -match2: - ; - } - - /* Remember the new sched domains */ - if (doms_cur != &fallback_doms) - free_sched_domains(doms_cur, ndoms_cur); - kfree(dattr_cur); /* kfree(NULL) is safe */ - doms_cur = doms_new; - dattr_cur = dattr_new; - ndoms_cur = ndoms_new; - - register_sched_domain_sysctl(); - - mutex_unlock(&sched_domains_mutex); -} - -static int num_cpus_frozen; /* used to mark begin/end of suspend/resume */ +static int num_cpus_frozen; /* * Update cpusets according to cpu_active mask. If cpusets are @@ -7305,30 +5549,15 @@ static void cpuset_cpu_active(void) * cpuset configurations. */ } - cpuset_update_active_cpus(true); + cpuset_update_active_cpus(); } static int cpuset_cpu_inactive(unsigned int cpu) { - unsigned long flags; - struct dl_bw *dl_b; - bool overflow; - int cpus; - if (!cpuhp_tasks_frozen) { - rcu_read_lock_sched(); - dl_b = dl_bw_of(cpu); - - raw_spin_lock_irqsave(&dl_b->lock, flags); - cpus = dl_bw_cpus(cpu); - overflow = __dl_overflow(dl_b, cpus, 0, 0); - raw_spin_unlock_irqrestore(&dl_b->lock, flags); - - rcu_read_unlock_sched(); - - if (overflow) + if (dl_cpu_busy(cpu)) return -EBUSY; - cpuset_update_active_cpus(false); + cpuset_update_active_cpus(); } else { num_cpus_frozen++; partition_sched_domains(1, NULL, NULL); @@ -7339,7 +5568,7 @@ static int cpuset_cpu_inactive(unsigned int cpu) int sched_cpu_activate(unsigned int cpu) { struct rq *rq = cpu_rq(cpu); - unsigned long flags; + struct rq_flags rf; set_cpu_active(cpu, true); @@ -7352,17 +5581,17 @@ int sched_cpu_activate(unsigned int cpu) * Put the rq online, if not already. This happens: * * 1) In the early boot process, because we build the real domains - * after all cpus have been brought up. + * after all CPUs have been brought up. * * 2) At runtime, if cpuset_cpu_active() fails to rebuild the * domains. */ - raw_spin_lock_irqsave(&rq->lock, flags); + rq_lock_irqsave(rq, &rf); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_online(rq); } - raw_spin_unlock_irqrestore(&rq->lock, flags); + rq_unlock_irqrestore(rq, &rf); update_max_interval(); @@ -7379,15 +5608,9 @@ int sched_cpu_deactivate(unsigned int cpu) * users of this state to go away such that all new such users will * observe it. * - * For CONFIG_PREEMPT we have preemptible RCU and its sync_rcu() might - * not imply sync_sched(), so wait for both. - * * Do sync before park smpboot threads to take care the rcu boost case. */ - if (IS_ENABLED(CONFIG_PREEMPT)) - synchronize_rcu_mult(call_rcu, call_rcu_sched); - else - synchronize_rcu(); + synchronize_rcu_mult(call_rcu, call_rcu_sched); if (!sched_smp_initialized) return 0; @@ -7420,18 +5643,20 @@ int sched_cpu_starting(unsigned int cpu) int sched_cpu_dying(unsigned int cpu) { struct rq *rq = cpu_rq(cpu); - unsigned long flags; + struct rq_flags rf; /* Handle pending wakeups and then migrate everything off */ sched_ttwu_pending(); - raw_spin_lock_irqsave(&rq->lock, flags); + + rq_lock_irqsave(rq, &rf); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_offline(rq); } - migrate_tasks(rq); + migrate_tasks(rq, &rf); BUG_ON(rq->nr_running != 1); - raw_spin_unlock_irqrestore(&rq->lock, flags); + rq_unlock_irqrestore(rq, &rf); + calc_load_migrate(rq); update_max_interval(); nohz_balance_exit_idle(cpu); @@ -7461,17 +5686,16 @@ void __init sched_init_smp(void) cpumask_var_t non_isolated_cpus; alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL); - alloc_cpumask_var(&fallback_doms, GFP_KERNEL); sched_init_numa(); /* * There's no userspace yet to cause hotplug operations; hence all the - * cpu masks are stable and all blatant races in the below code cannot + * CPU masks are stable and all blatant races in the below code cannot * happen. */ mutex_lock(&sched_domains_mutex); - init_sched_domains(cpu_active_mask); + sched_init_domains(cpu_active_mask); cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map); if (cpumask_empty(non_isolated_cpus)) cpumask_set_cpu(smp_processor_id(), non_isolated_cpus); @@ -7527,26 +5751,13 @@ static struct kmem_cache *task_group_cache __read_mostly; DECLARE_PER_CPU(cpumask_var_t, load_balance_mask); DECLARE_PER_CPU(cpumask_var_t, select_idle_mask); -#define WAIT_TABLE_BITS 8 -#define WAIT_TABLE_SIZE (1 << WAIT_TABLE_BITS) -static wait_queue_head_t bit_wait_table[WAIT_TABLE_SIZE] __cacheline_aligned; - -wait_queue_head_t *bit_waitqueue(void *word, int bit) -{ - const int shift = BITS_PER_LONG == 32 ? 5 : 6; - unsigned long val = (unsigned long)word << shift | bit; - - return bit_wait_table + hash_long(val, WAIT_TABLE_BITS); -} -EXPORT_SYMBOL(bit_waitqueue); - void __init sched_init(void) { int i, j; unsigned long alloc_size = 0, ptr; - for (i = 0; i < WAIT_TABLE_SIZE; i++) - init_waitqueue_head(bit_wait_table + i); + sched_clock_init(); + wait_bit_init(); #ifdef CONFIG_FAIR_GROUP_SCHED alloc_size += 2 * nr_cpu_ids * sizeof(void **); @@ -7583,10 +5794,8 @@ void __init sched_init(void) } #endif /* CONFIG_CPUMASK_OFFSTACK */ - init_rt_bandwidth(&def_rt_bandwidth, - global_rt_period(), global_rt_runtime()); - init_dl_bandwidth(&def_dl_bandwidth, - global_rt_period(), global_rt_runtime()); + init_rt_bandwidth(&def_rt_bandwidth, global_rt_period(), global_rt_runtime()); + init_dl_bandwidth(&def_dl_bandwidth, global_rt_period(), global_rt_runtime()); #ifdef CONFIG_SMP init_defrootdomain(); @@ -7622,18 +5831,18 @@ void __init sched_init(void) INIT_LIST_HEAD(&rq->leaf_cfs_rq_list); rq->tmp_alone_branch = &rq->leaf_cfs_rq_list; /* - * How much cpu bandwidth does root_task_group get? + * How much CPU bandwidth does root_task_group get? * * In case of task-groups formed thr' the cgroup filesystem, it - * gets 100% of the cpu resources in the system. This overall - * system cpu resource is divided among the tasks of + * gets 100% of the CPU resources in the system. This overall + * system CPU resource is divided among the tasks of * root_task_group and its child task-groups in a fair manner, * based on each entity's (task or task-group's) weight * (se->load.weight). * * In other words, if root_task_group has 10 tasks of weight * 1024) and two child groups A0 and A1 (of weight 1024 each), - * then A0's share of the cpu resource is: + * then A0's share of the CPU resource is: * * A0's bandwidth = 1024 / (10*1024 + 1024 + 1024) = 8.33% * @@ -7686,7 +5895,7 @@ void __init sched_init(void) /* * The boot idle thread does lazy MMU switching as well: */ - atomic_inc(&init_mm.mm_count); + mmgrab(&init_mm); enter_lazy_tlb(&init_mm, current); /* @@ -7700,7 +5909,6 @@ void __init sched_init(void) calc_load_update = jiffies + LOAD_FREQ; #ifdef CONFIG_SMP - zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT); /* May be allocated at isolcpus cmdline parse time */ if (cpu_isolated_map == NULL) zalloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT); @@ -7742,19 +5950,25 @@ EXPORT_SYMBOL(__might_sleep); void ___might_sleep(const char *file, int line, int preempt_offset) { - static unsigned long prev_jiffy; /* ratelimiting */ + /* Ratelimiting timestamp: */ + static unsigned long prev_jiffy; + unsigned long preempt_disable_ip; - rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */ + /* WARN_ON_ONCE() by default, no rate limit required: */ + rcu_sleep_check(); + if ((preempt_count_equals(preempt_offset) && !irqs_disabled() && !is_idle_task(current)) || - system_state != SYSTEM_RUNNING || oops_in_progress) + system_state == SYSTEM_BOOTING || system_state > SYSTEM_RUNNING || + oops_in_progress) return; + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) return; prev_jiffy = jiffies; - /* Save this before calling printk(), since that will clobber it */ + /* Save this before calling printk(), since that will clobber it: */ preempt_disable_ip = get_preempt_disable_ip(current); printk(KERN_ERR @@ -7833,7 +6047,7 @@ void normalize_rt_tasks(void) */ /** - * curr_task - return the current task for a given cpu. + * curr_task - return the current task for a given CPU. * @cpu: the processor in question. * * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! @@ -7849,13 +6063,13 @@ struct task_struct *curr_task(int cpu) #ifdef CONFIG_IA64 /** - * set_curr_task - set the current task for a given cpu. + * set_curr_task - set the current task for a given CPU. * @cpu: the processor in question. * @p: the task pointer to set. * * Description: This function must only be used when non-maskable interrupts * are serviced on a separate stack. It allows the architecture to switch the - * notion of the current task on a cpu in a non-blocking manner. This function + * notion of the current task on a CPU in a non-blocking manner. This function * must be called with all CPU's synchronized, and interrupts disabled, the * and caller must save the original value of the current task (see * curr_task() above) and restore that value before reenabling interrupts and @@ -7911,7 +6125,8 @@ void sched_online_group(struct task_group *tg, struct task_group *parent) spin_lock_irqsave(&task_group_lock, flags); list_add_rcu(&tg->list, &task_groups); - WARN_ON(!parent); /* root should already exist */ + /* Root should already exist: */ + WARN_ON(!parent); tg->parent = parent; INIT_LIST_HEAD(&tg->children); @@ -7924,13 +6139,13 @@ void sched_online_group(struct task_group *tg, struct task_group *parent) /* rcu callback to free various structures associated with a task group */ static void sched_free_group_rcu(struct rcu_head *rhp) { - /* now it should be safe to free those cfs_rqs */ + /* Now it should be safe to free those cfs_rqs: */ sched_free_group(container_of(rhp, struct task_group, rcu)); } void sched_destroy_group(struct task_group *tg) { - /* wait for possible concurrent references to cfs_rqs complete */ + /* Wait for possible concurrent references to cfs_rqs complete: */ call_rcu(&tg->rcu, sched_free_group_rcu); } @@ -7938,7 +6153,7 @@ void sched_offline_group(struct task_group *tg) { unsigned long flags; - /* end participation in shares distribution */ + /* End participation in shares distribution: */ unregister_fair_sched_group(tg); spin_lock_irqsave(&task_group_lock, flags); @@ -7978,405 +6193,31 @@ static void sched_change_group(struct task_struct *tsk, int type) */ void sched_move_task(struct task_struct *tsk) { - int queued, running; + int queued, running, queue_flags = + DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; struct rq_flags rf; struct rq *rq; rq = task_rq_lock(tsk, &rf); + update_rq_clock(rq); running = task_current(rq, tsk); queued = task_on_rq_queued(tsk); if (queued) - dequeue_task(rq, tsk, DEQUEUE_SAVE | DEQUEUE_MOVE); - if (unlikely(running)) + dequeue_task(rq, tsk, queue_flags); + if (running) put_prev_task(rq, tsk); sched_change_group(tsk, TASK_MOVE_GROUP); if (queued) - enqueue_task(rq, tsk, ENQUEUE_RESTORE | ENQUEUE_MOVE); - if (unlikely(running)) + enqueue_task(rq, tsk, queue_flags); + if (running) set_curr_task(rq, tsk); task_rq_unlock(rq, tsk, &rf); } -#endif /* CONFIG_CGROUP_SCHED */ - -#ifdef CONFIG_RT_GROUP_SCHED -/* - * Ensure that the real time constraints are schedulable. - */ -static DEFINE_MUTEX(rt_constraints_mutex); - -/* Must be called with tasklist_lock held */ -static inline int tg_has_rt_tasks(struct task_group *tg) -{ - struct task_struct *g, *p; - - /* - * Autogroups do not have RT tasks; see autogroup_create(). - */ - if (task_group_is_autogroup(tg)) - return 0; - - for_each_process_thread(g, p) { - if (rt_task(p) && task_group(p) == tg) - return 1; - } - - return 0; -} - -struct rt_schedulable_data { - struct task_group *tg; - u64 rt_period; - u64 rt_runtime; -}; - -static int tg_rt_schedulable(struct task_group *tg, void *data) -{ - struct rt_schedulable_data *d = data; - struct task_group *child; - unsigned long total, sum = 0; - u64 period, runtime; - - period = ktime_to_ns(tg->rt_bandwidth.rt_period); - runtime = tg->rt_bandwidth.rt_runtime; - - if (tg == d->tg) { - period = d->rt_period; - runtime = d->rt_runtime; - } - - /* - * Cannot have more runtime than the period. - */ - if (runtime > period && runtime != RUNTIME_INF) - return -EINVAL; - - /* - * Ensure we don't starve existing RT tasks. - */ - if (rt_bandwidth_enabled() && !runtime && tg_has_rt_tasks(tg)) - return -EBUSY; - - total = to_ratio(period, runtime); - - /* - * Nobody can have more than the global setting allows. - */ - if (total > to_ratio(global_rt_period(), global_rt_runtime())) - return -EINVAL; - - /* - * The sum of our children's runtime should not exceed our own. - */ - list_for_each_entry_rcu(child, &tg->children, siblings) { - period = ktime_to_ns(child->rt_bandwidth.rt_period); - runtime = child->rt_bandwidth.rt_runtime; - - if (child == d->tg) { - period = d->rt_period; - runtime = d->rt_runtime; - } - - sum += to_ratio(period, runtime); - } - - if (sum > total) - return -EINVAL; - - return 0; -} - -static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) -{ - int ret; - - struct rt_schedulable_data data = { - .tg = tg, - .rt_period = period, - .rt_runtime = runtime, - }; - - rcu_read_lock(); - ret = walk_tg_tree(tg_rt_schedulable, tg_nop, &data); - rcu_read_unlock(); - - return ret; -} - -static int tg_set_rt_bandwidth(struct task_group *tg, - u64 rt_period, u64 rt_runtime) -{ - int i, err = 0; - - /* - * Disallowing the root group RT runtime is BAD, it would disallow the - * kernel creating (and or operating) RT threads. - */ - if (tg == &root_task_group && rt_runtime == 0) - return -EINVAL; - - /* No period doesn't make any sense. */ - if (rt_period == 0) - return -EINVAL; - - mutex_lock(&rt_constraints_mutex); - read_lock(&tasklist_lock); - err = __rt_schedulable(tg, rt_period, rt_runtime); - if (err) - goto unlock; - - raw_spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); - tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period); - tg->rt_bandwidth.rt_runtime = rt_runtime; - - for_each_possible_cpu(i) { - struct rt_rq *rt_rq = tg->rt_rq[i]; - - raw_spin_lock(&rt_rq->rt_runtime_lock); - rt_rq->rt_runtime = rt_runtime; - raw_spin_unlock(&rt_rq->rt_runtime_lock); - } - raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); -unlock: - read_unlock(&tasklist_lock); - mutex_unlock(&rt_constraints_mutex); - - return err; -} - -static int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us) -{ - u64 rt_runtime, rt_period; - - rt_period = ktime_to_ns(tg->rt_bandwidth.rt_period); - rt_runtime = (u64)rt_runtime_us * NSEC_PER_USEC; - if (rt_runtime_us < 0) - rt_runtime = RUNTIME_INF; - - return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); -} - -static long sched_group_rt_runtime(struct task_group *tg) -{ - u64 rt_runtime_us; - - if (tg->rt_bandwidth.rt_runtime == RUNTIME_INF) - return -1; - - rt_runtime_us = tg->rt_bandwidth.rt_runtime; - do_div(rt_runtime_us, NSEC_PER_USEC); - return rt_runtime_us; -} - -static int sched_group_set_rt_period(struct task_group *tg, u64 rt_period_us) -{ - u64 rt_runtime, rt_period; - - rt_period = rt_period_us * NSEC_PER_USEC; - rt_runtime = tg->rt_bandwidth.rt_runtime; - - return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); -} - -static long sched_group_rt_period(struct task_group *tg) -{ - u64 rt_period_us; - - rt_period_us = ktime_to_ns(tg->rt_bandwidth.rt_period); - do_div(rt_period_us, NSEC_PER_USEC); - return rt_period_us; -} -#endif /* CONFIG_RT_GROUP_SCHED */ - -#ifdef CONFIG_RT_GROUP_SCHED -static int sched_rt_global_constraints(void) -{ - int ret = 0; - - mutex_lock(&rt_constraints_mutex); - read_lock(&tasklist_lock); - ret = __rt_schedulable(NULL, 0, 0); - read_unlock(&tasklist_lock); - mutex_unlock(&rt_constraints_mutex); - - return ret; -} - -static int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk) -{ - /* Don't accept realtime tasks when there is no way for them to run */ - if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) - return 0; - - return 1; -} - -#else /* !CONFIG_RT_GROUP_SCHED */ -static int sched_rt_global_constraints(void) -{ - unsigned long flags; - int i; - - raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); - for_each_possible_cpu(i) { - struct rt_rq *rt_rq = &cpu_rq(i)->rt; - - raw_spin_lock(&rt_rq->rt_runtime_lock); - rt_rq->rt_runtime = global_rt_runtime(); - raw_spin_unlock(&rt_rq->rt_runtime_lock); - } - raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); - - return 0; -} -#endif /* CONFIG_RT_GROUP_SCHED */ - -static int sched_dl_global_validate(void) -{ - u64 runtime = global_rt_runtime(); - u64 period = global_rt_period(); - u64 new_bw = to_ratio(period, runtime); - struct dl_bw *dl_b; - int cpu, ret = 0; - unsigned long flags; - - /* - * Here we want to check the bandwidth not being set to some - * value smaller than the currently allocated bandwidth in - * any of the root_domains. - * - * FIXME: Cycling on all the CPUs is overdoing, but simpler than - * cycling on root_domains... Discussion on different/better - * solutions is welcome! - */ - for_each_possible_cpu(cpu) { - rcu_read_lock_sched(); - dl_b = dl_bw_of(cpu); - - raw_spin_lock_irqsave(&dl_b->lock, flags); - if (new_bw < dl_b->total_bw) - ret = -EBUSY; - raw_spin_unlock_irqrestore(&dl_b->lock, flags); - - rcu_read_unlock_sched(); - - if (ret) - break; - } - - return ret; -} - -static void sched_dl_do_global(void) -{ - u64 new_bw = -1; - struct dl_bw *dl_b; - int cpu; - unsigned long flags; - - def_dl_bandwidth.dl_period = global_rt_period(); - def_dl_bandwidth.dl_runtime = global_rt_runtime(); - - if (global_rt_runtime() != RUNTIME_INF) - new_bw = to_ratio(global_rt_period(), global_rt_runtime()); - - /* - * FIXME: As above... - */ - for_each_possible_cpu(cpu) { - rcu_read_lock_sched(); - dl_b = dl_bw_of(cpu); - - raw_spin_lock_irqsave(&dl_b->lock, flags); - dl_b->bw = new_bw; - raw_spin_unlock_irqrestore(&dl_b->lock, flags); - - rcu_read_unlock_sched(); - } -} - -static int sched_rt_global_validate(void) -{ - if (sysctl_sched_rt_period <= 0) - return -EINVAL; - - if ((sysctl_sched_rt_runtime != RUNTIME_INF) && - (sysctl_sched_rt_runtime > sysctl_sched_rt_period)) - return -EINVAL; - - return 0; -} - -static void sched_rt_do_global(void) -{ - def_rt_bandwidth.rt_runtime = global_rt_runtime(); - def_rt_bandwidth.rt_period = ns_to_ktime(global_rt_period()); -} - -int sched_rt_handler(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) -{ - int old_period, old_runtime; - static DEFINE_MUTEX(mutex); - int ret; - - mutex_lock(&mutex); - old_period = sysctl_sched_rt_period; - old_runtime = sysctl_sched_rt_runtime; - - ret = proc_dointvec(table, write, buffer, lenp, ppos); - - if (!ret && write) { - ret = sched_rt_global_validate(); - if (ret) - goto undo; - - ret = sched_dl_global_validate(); - if (ret) - goto undo; - - ret = sched_rt_global_constraints(); - if (ret) - goto undo; - - sched_rt_do_global(); - sched_dl_do_global(); - } - if (0) { -undo: - sysctl_sched_rt_period = old_period; - sysctl_sched_rt_runtime = old_runtime; - } - mutex_unlock(&mutex); - - return ret; -} - -int sched_rr_handler(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) -{ - int ret; - static DEFINE_MUTEX(mutex); - - mutex_lock(&mutex); - ret = proc_dointvec(table, write, buffer, lenp, ppos); - /* make sure that internally we keep jiffies */ - /* also, writing zero resets timeslice to default */ - if (!ret && write) { - sched_rr_timeslice = sched_rr_timeslice <= 0 ? - RR_TIMESLICE : msecs_to_jiffies(sched_rr_timeslice); - } - mutex_unlock(&mutex); - return ret; -} - -#ifdef CONFIG_CGROUP_SCHED static inline struct task_group *css_tg(struct cgroup_subsys_state *css) { @@ -8398,11 +6239,20 @@ cpu_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) if (IS_ERR(tg)) return ERR_PTR(-ENOMEM); - sched_online_group(tg, parent); - return &tg->css; } +/* Expose task group only after completing cgroup initialization */ +static int cpu_cgroup_css_online(struct cgroup_subsys_state *css) +{ + struct task_group *tg = css_tg(css); + struct task_group *parent = css_tg(css->parent); + + if (parent) + sched_online_group(tg, parent); + return 0; +} + static void cpu_cgroup_css_released(struct cgroup_subsys_state *css) { struct task_group *tg = css_tg(css); @@ -8431,6 +6281,7 @@ static void cpu_cgroup_fork(struct task_struct *task) rq = task_rq_lock(task, &rf); + update_rq_clock(rq); sched_change_group(task, TASK_SET_GROUP); task_rq_unlock(rq, task, &rf); @@ -8550,22 +6401,25 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) cfs_b->quota = quota; __refill_cfs_bandwidth_runtime(cfs_b); - /* restart the period timer (if active) to handle new period expiry */ + + /* Restart the period timer (if active) to handle new period expiry: */ if (runtime_enabled) start_cfs_bandwidth(cfs_b); + raw_spin_unlock_irq(&cfs_b->lock); for_each_online_cpu(i) { struct cfs_rq *cfs_rq = tg->cfs_rq[i]; struct rq *rq = cfs_rq->rq; + struct rq_flags rf; - raw_spin_lock_irq(&rq->lock); + rq_lock_irq(rq, &rf); cfs_rq->runtime_enabled = runtime_enabled; cfs_rq->runtime_remaining = 0; if (cfs_rq->throttled) unthrottle_cfs_rq(cfs_rq); - raw_spin_unlock_irq(&rq->lock); + rq_unlock_irq(rq, &rf); } if (runtime_was_enabled && !runtime_enabled) cfs_bandwidth_usage_dec(); @@ -8690,8 +6544,8 @@ static int tg_cfs_schedulable_down(struct task_group *tg, void *data) parent_quota = parent_b->hierarchical_quota; /* - * ensure max(child_quota) <= parent_quota, inherit when no - * limit is set + * Ensure max(child_quota) <= parent_quota, inherit when no + * limit is set: */ if (quota == RUNTIME_INF) quota = parent_quota; @@ -8800,11 +6654,12 @@ static struct cftype cpu_files[] = { .write_u64 = cpu_rt_period_write_uint, }, #endif - { } /* terminate */ + { } /* Terminate */ }; struct cgroup_subsys cpu_cgrp_subsys = { .css_alloc = cpu_cgroup_css_alloc, + .css_online = cpu_cgroup_css_online, .css_released = cpu_cgroup_css_released, .css_free = cpu_cgroup_css_free, .fork = cpu_cgroup_fork, diff --git a/kernel/sched/cpuacct.c b/kernel/sched/cpuacct.c index 9add206b5608..f95ab29a45d0 100644 --- a/kernel/sched/cpuacct.c +++ b/kernel/sched/cpuacct.c @@ -297,7 +297,7 @@ static int cpuacct_stats_show(struct seq_file *sf, void *v) for (stat = 0; stat < CPUACCT_STAT_NSTATS; stat++) { seq_printf(sf, "%s %lld\n", cpuacct_stat_desc[stat], - (long long)cputime64_to_clock_t(val[stat])); + (long long)nsec_to_clock_t(val[stat])); } return 0; diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c index e73119013c53..fba235c7d026 100644 --- a/kernel/sched/cpudeadline.c +++ b/kernel/sched/cpudeadline.c @@ -128,10 +128,10 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p, const struct sched_dl_entity *dl_se = &p->dl; if (later_mask && - cpumask_and(later_mask, cp->free_cpus, tsk_cpus_allowed(p))) { + cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed)) { best_cpu = cpumask_any(later_mask); goto out; - } else if (cpumask_test_cpu(cpudl_maximum(cp), tsk_cpus_allowed(p)) && + } else if (cpumask_test_cpu(cpudl_maximum(cp), &p->cpus_allowed) && dl_time_before(dl_se->deadline, cp->elements[0].dl)) { best_cpu = cpudl_maximum(cp); if (later_mask) diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index fd4659313640..29a397067ffa 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -13,6 +13,7 @@ #include <linux/cpufreq.h> #include <linux/kthread.h> +#include <uapi/linux/sched/types.h> #include <linux/slab.h> #include <trace/events/power.h> @@ -35,6 +36,7 @@ struct sugov_policy { u64 last_freq_update_time; s64 freq_update_delay_ns; unsigned int next_freq; + unsigned int cached_raw_freq; /* The next fields are only needed if fast switch cannot be used. */ struct irq_work irq_work; @@ -51,7 +53,6 @@ struct sugov_cpu { struct update_util_data update_util; struct sugov_policy *sg_policy; - unsigned int cached_raw_freq; unsigned long iowait_boost; unsigned long iowait_boost_max; u64 last_update; @@ -60,6 +61,11 @@ struct sugov_cpu { unsigned long util; unsigned long max; unsigned int flags; + + /* The field below is for single-CPU policies only. */ +#ifdef CONFIG_NO_HZ_COMMON + unsigned long saved_idle_calls; +#endif }; static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu); @@ -92,22 +98,20 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time, { struct cpufreq_policy *policy = sg_policy->policy; + if (sg_policy->next_freq == next_freq) + return; + + sg_policy->next_freq = next_freq; sg_policy->last_freq_update_time = time; if (policy->fast_switch_enabled) { - if (sg_policy->next_freq == next_freq) { - trace_cpu_frequency(policy->cur, smp_processor_id()); - return; - } - sg_policy->next_freq = next_freq; next_freq = cpufreq_driver_fast_switch(policy, next_freq); if (next_freq == CPUFREQ_ENTRY_INVALID) return; policy->cur = next_freq; trace_cpu_frequency(next_freq, smp_processor_id()); - } else if (sg_policy->next_freq != next_freq) { - sg_policy->next_freq = next_freq; + } else { sg_policy->work_in_progress = true; irq_work_queue(&sg_policy->irq_work); } @@ -115,7 +119,7 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time, /** * get_next_freq - Compute a new frequency for a given cpufreq policy. - * @sg_cpu: schedutil cpu object to compute the new frequency for. + * @sg_policy: schedutil policy object to compute the new frequency for. * @util: Current CPU utilization. * @max: CPU capacity. * @@ -135,19 +139,18 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time, * next_freq (as calculated above) is returned, subject to policy min/max and * cpufreq driver limitations. */ -static unsigned int get_next_freq(struct sugov_cpu *sg_cpu, unsigned long util, - unsigned long max) +static unsigned int get_next_freq(struct sugov_policy *sg_policy, + unsigned long util, unsigned long max) { - struct sugov_policy *sg_policy = sg_cpu->sg_policy; struct cpufreq_policy *policy = sg_policy->policy; unsigned int freq = arch_scale_freq_invariant() ? policy->cpuinfo.max_freq : policy->cur; freq = (freq + (freq >> 2)) * util / max; - if (freq == sg_cpu->cached_raw_freq && sg_policy->next_freq != UINT_MAX) + if (freq == sg_policy->cached_raw_freq && sg_policy->next_freq != UINT_MAX) return sg_policy->next_freq; - sg_cpu->cached_raw_freq = freq; + sg_policy->cached_raw_freq = freq; return cpufreq_driver_resolve_freq(policy, freq); } @@ -192,6 +195,19 @@ static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, unsigned long *util, sg_cpu->iowait_boost >>= 1; } +#ifdef CONFIG_NO_HZ_COMMON +static bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) +{ + unsigned long idle_calls = tick_nohz_get_idle_calls(); + bool ret = idle_calls == sg_cpu->saved_idle_calls; + + sg_cpu->saved_idle_calls = idle_calls; + return ret; +} +#else +static inline bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) { return false; } +#endif /* CONFIG_NO_HZ_COMMON */ + static void sugov_update_single(struct update_util_data *hook, u64 time, unsigned int flags) { @@ -200,6 +216,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time, struct cpufreq_policy *policy = sg_policy->policy; unsigned long util, max; unsigned int next_f; + bool busy; sugov_set_iowait_boost(sg_cpu, time, flags); sg_cpu->last_update = time; @@ -207,40 +224,36 @@ static void sugov_update_single(struct update_util_data *hook, u64 time, if (!sugov_should_update_freq(sg_policy, time)) return; + busy = sugov_cpu_is_busy(sg_cpu); + if (flags & SCHED_CPUFREQ_RT_DL) { next_f = policy->cpuinfo.max_freq; } else { sugov_get_util(&util, &max); sugov_iowait_boost(sg_cpu, &util, &max); - next_f = get_next_freq(sg_cpu, util, max); + next_f = get_next_freq(sg_policy, util, max); + /* + * Do not reduce the frequency if the CPU has not been idle + * recently, as the reduction is likely to be premature then. + */ + if (busy && next_f < sg_policy->next_freq) + next_f = sg_policy->next_freq; } sugov_update_commit(sg_policy, time, next_f); } -static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, - unsigned long util, unsigned long max, - unsigned int flags) +static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time) { struct sugov_policy *sg_policy = sg_cpu->sg_policy; struct cpufreq_policy *policy = sg_policy->policy; - unsigned int max_f = policy->cpuinfo.max_freq; - u64 last_freq_update_time = sg_policy->last_freq_update_time; + unsigned long util = 0, max = 1; unsigned int j; - if (flags & SCHED_CPUFREQ_RT_DL) - return max_f; - - sugov_iowait_boost(sg_cpu, &util, &max); - for_each_cpu(j, policy->cpus) { - struct sugov_cpu *j_sg_cpu; + struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j); unsigned long j_util, j_max; s64 delta_ns; - if (j == smp_processor_id()) - continue; - - j_sg_cpu = &per_cpu(sugov_cpu, j); /* * If the CPU utilization was last updated before the previous * frequency update and the time elapsed between the last update @@ -248,13 +261,13 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, * enough, don't take the CPU into account as it probably is * idle now (and clear iowait_boost for it). */ - delta_ns = last_freq_update_time - j_sg_cpu->last_update; + delta_ns = time - j_sg_cpu->last_update; if (delta_ns > TICK_NSEC) { j_sg_cpu->iowait_boost = 0; continue; } if (j_sg_cpu->flags & SCHED_CPUFREQ_RT_DL) - return max_f; + return policy->cpuinfo.max_freq; j_util = j_sg_cpu->util; j_max = j_sg_cpu->max; @@ -266,7 +279,7 @@ static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, sugov_iowait_boost(j_sg_cpu, &util, &max); } - return get_next_freq(sg_cpu, util, max); + return get_next_freq(sg_policy, util, max); } static void sugov_update_shared(struct update_util_data *hook, u64 time, @@ -289,7 +302,11 @@ static void sugov_update_shared(struct update_util_data *hook, u64 time, sg_cpu->last_update = time; if (sugov_should_update_freq(sg_policy, time)) { - next_f = sugov_next_freq_shared(sg_cpu, util, max, flags); + if (flags & SCHED_CPUFREQ_RT_DL) + next_f = sg_policy->policy->cpuinfo.max_freq; + else + next_f = sugov_next_freq_shared(sg_cpu, time); + sugov_update_commit(sg_policy, time, next_f); } @@ -473,7 +490,6 @@ static int sugov_init(struct cpufreq_policy *policy) { struct sugov_policy *sg_policy; struct sugov_tunables *tunables; - unsigned int lat; int ret = 0; /* State should be equivalent to EXIT */ @@ -512,10 +528,16 @@ static int sugov_init(struct cpufreq_policy *policy) goto stop_kthread; } - tunables->rate_limit_us = LATENCY_MULTIPLIER; - lat = policy->cpuinfo.transition_latency / NSEC_PER_USEC; - if (lat) - tunables->rate_limit_us *= lat; + if (policy->transition_delay_us) { + tunables->rate_limit_us = policy->transition_delay_us; + } else { + unsigned int lat; + + tunables->rate_limit_us = LATENCY_MULTIPLIER; + lat = policy->cpuinfo.transition_latency / NSEC_PER_USEC; + if (lat) + tunables->rate_limit_us *= lat; + } policy->governor_data = sg_policy; sg_policy->tunables = tunables; @@ -579,25 +601,24 @@ static int sugov_start(struct cpufreq_policy *policy) sg_policy->next_freq = UINT_MAX; sg_policy->work_in_progress = false; sg_policy->need_freq_update = false; + sg_policy->cached_raw_freq = 0; for_each_cpu(cpu, policy->cpus) { struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu); + memset(sg_cpu, 0, sizeof(*sg_cpu)); sg_cpu->sg_policy = sg_policy; - if (policy_is_shared(policy)) { - sg_cpu->util = 0; - sg_cpu->max = 0; - sg_cpu->flags = SCHED_CPUFREQ_RT; - sg_cpu->last_update = 0; - sg_cpu->cached_raw_freq = 0; - sg_cpu->iowait_boost = 0; - sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq; - cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, - sugov_update_shared); - } else { - cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, - sugov_update_single); - } + sg_cpu->flags = SCHED_CPUFREQ_RT; + sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq; + } + + for_each_cpu(cpu, policy->cpus) { + struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu); + + cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util, + policy_is_shared(policy) ? + sugov_update_shared : + sugov_update_single); } return 0; } diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index 11e9705bf937..981fcd7dc394 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -103,11 +103,11 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, if (skip) continue; - if (cpumask_any_and(tsk_cpus_allowed(p), vec->mask) >= nr_cpu_ids) + if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids) continue; if (lowest_mask) { - cpumask_and(lowest_mask, tsk_cpus_allowed(p), vec->mask); + cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask); /* * We have to ensure that we have at least one bit diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 7700a9cba335..14d2dbf97c53 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -4,11 +4,8 @@ #include <linux/kernel_stat.h> #include <linux/static_key.h> #include <linux/context_tracking.h> +#include <linux/sched/cputime.h> #include "sched.h" -#ifdef CONFIG_PARAVIRT -#include <asm/paravirt.h> -#endif - #ifdef CONFIG_IRQ_TIME_ACCOUNTING @@ -37,6 +34,18 @@ void disable_sched_clock_irqtime(void) sched_clock_irqtime = 0; } +static void irqtime_account_delta(struct irqtime *irqtime, u64 delta, + enum cpu_usage_stat idx) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + + u64_stats_update_begin(&irqtime->sync); + cpustat[idx] += delta; + irqtime->total += delta; + irqtime->tick_delta += delta; + u64_stats_update_end(&irqtime->sync); +} + /* * Called before incrementing preempt_count on {soft,}irq_enter * and before decrementing preempt_count on {soft,}irq_exit. @@ -54,7 +63,6 @@ void irqtime_account_irq(struct task_struct *curr) delta = sched_clock_cpu(cpu) - irqtime->irq_start_time; irqtime->irq_start_time += delta; - u64_stats_update_begin(&irqtime->sync); /* * We do not account for softirq time from ksoftirqd here. * We want to continue accounting softirq time to ksoftirqd thread @@ -62,48 +70,28 @@ void irqtime_account_irq(struct task_struct *curr) * that do not consume any time, but still wants to run. */ if (hardirq_count()) - irqtime->hardirq_time += delta; + irqtime_account_delta(irqtime, delta, CPUTIME_IRQ); else if (in_serving_softirq() && curr != this_cpu_ksoftirqd()) - irqtime->softirq_time += delta; - - u64_stats_update_end(&irqtime->sync); + irqtime_account_delta(irqtime, delta, CPUTIME_SOFTIRQ); } EXPORT_SYMBOL_GPL(irqtime_account_irq); -static cputime_t irqtime_account_update(u64 irqtime, int idx, cputime_t maxtime) +static u64 irqtime_tick_accounted(u64 maxtime) { - u64 *cpustat = kcpustat_this_cpu->cpustat; - cputime_t irq_cputime; - - irq_cputime = nsecs_to_cputime64(irqtime) - cpustat[idx]; - irq_cputime = min(irq_cputime, maxtime); - cpustat[idx] += irq_cputime; + struct irqtime *irqtime = this_cpu_ptr(&cpu_irqtime); + u64 delta; - return irq_cputime; -} + delta = min(irqtime->tick_delta, maxtime); + irqtime->tick_delta -= delta; -static cputime_t irqtime_account_hi_update(cputime_t maxtime) -{ - return irqtime_account_update(__this_cpu_read(cpu_irqtime.hardirq_time), - CPUTIME_IRQ, maxtime); -} - -static cputime_t irqtime_account_si_update(cputime_t maxtime) -{ - return irqtime_account_update(__this_cpu_read(cpu_irqtime.softirq_time), - CPUTIME_SOFTIRQ, maxtime); + return delta; } #else /* CONFIG_IRQ_TIME_ACCOUNTING */ #define sched_clock_irqtime (0) -static cputime_t irqtime_account_hi_update(cputime_t dummy) -{ - return 0; -} - -static cputime_t irqtime_account_si_update(cputime_t dummy) +static u64 irqtime_tick_accounted(u64 dummy) { return 0; } @@ -129,7 +117,7 @@ static inline void task_group_account_field(struct task_struct *p, int index, * @p: the process that the cpu time gets accounted to * @cputime: the cpu time spent in user space since the last update */ -void account_user_time(struct task_struct *p, cputime_t cputime) +void account_user_time(struct task_struct *p, u64 cputime) { int index; @@ -140,7 +128,7 @@ void account_user_time(struct task_struct *p, cputime_t cputime) index = (task_nice(p) > 0) ? CPUTIME_NICE : CPUTIME_USER; /* Add user time to cpustat. */ - task_group_account_field(p, index, (__force u64) cputime); + task_group_account_field(p, index, cputime); /* Account for user time used */ acct_account_cputime(p); @@ -151,7 +139,7 @@ void account_user_time(struct task_struct *p, cputime_t cputime) * @p: the process that the cpu time gets accounted to * @cputime: the cpu time spent in virtual machine since the last update */ -static void account_guest_time(struct task_struct *p, cputime_t cputime) +void account_guest_time(struct task_struct *p, u64 cputime) { u64 *cpustat = kcpustat_this_cpu->cpustat; @@ -162,11 +150,11 @@ static void account_guest_time(struct task_struct *p, cputime_t cputime) /* Add guest time to cpustat. */ if (task_nice(p) > 0) { - cpustat[CPUTIME_NICE] += (__force u64) cputime; - cpustat[CPUTIME_GUEST_NICE] += (__force u64) cputime; + cpustat[CPUTIME_NICE] += cputime; + cpustat[CPUTIME_GUEST_NICE] += cputime; } else { - cpustat[CPUTIME_USER] += (__force u64) cputime; - cpustat[CPUTIME_GUEST] += (__force u64) cputime; + cpustat[CPUTIME_USER] += cputime; + cpustat[CPUTIME_GUEST] += cputime; } } @@ -176,15 +164,15 @@ static void account_guest_time(struct task_struct *p, cputime_t cputime) * @cputime: the cpu time spent in kernel space since the last update * @index: pointer to cpustat field that has to be updated */ -static inline -void __account_system_time(struct task_struct *p, cputime_t cputime, int index) +void account_system_index_time(struct task_struct *p, + u64 cputime, enum cpu_usage_stat index) { /* Add system time to process. */ p->stime += cputime; account_group_system_time(p, cputime); /* Add system time to cpustat. */ - task_group_account_field(p, index, (__force u64) cputime); + task_group_account_field(p, index, cputime); /* Account for system time used */ acct_account_cputime(p); @@ -196,8 +184,7 @@ void __account_system_time(struct task_struct *p, cputime_t cputime, int index) * @hardirq_offset: the offset to subtract from hardirq_count() * @cputime: the cpu time spent in kernel space since the last update */ -void account_system_time(struct task_struct *p, int hardirq_offset, - cputime_t cputime) +void account_system_time(struct task_struct *p, int hardirq_offset, u64 cputime) { int index; @@ -213,33 +200,33 @@ void account_system_time(struct task_struct *p, int hardirq_offset, else index = CPUTIME_SYSTEM; - __account_system_time(p, cputime, index); + account_system_index_time(p, cputime, index); } /* * Account for involuntary wait time. * @cputime: the cpu time spent in involuntary wait */ -void account_steal_time(cputime_t cputime) +void account_steal_time(u64 cputime) { u64 *cpustat = kcpustat_this_cpu->cpustat; - cpustat[CPUTIME_STEAL] += (__force u64) cputime; + cpustat[CPUTIME_STEAL] += cputime; } /* * Account for idle time. * @cputime: the cpu time spent in idle wait */ -void account_idle_time(cputime_t cputime) +void account_idle_time(u64 cputime) { u64 *cpustat = kcpustat_this_cpu->cpustat; struct rq *rq = this_rq(); if (atomic_read(&rq->nr_iowait) > 0) - cpustat[CPUTIME_IOWAIT] += (__force u64) cputime; + cpustat[CPUTIME_IOWAIT] += cputime; else - cpustat[CPUTIME_IDLE] += (__force u64) cputime; + cpustat[CPUTIME_IDLE] += cputime; } /* @@ -247,21 +234,19 @@ void account_idle_time(cputime_t cputime) * ticks are not redelivered later. Due to that, this function may on * occasion account more time than the calling functions think elapsed. */ -static __always_inline cputime_t steal_account_process_time(cputime_t maxtime) +static __always_inline u64 steal_account_process_time(u64 maxtime) { #ifdef CONFIG_PARAVIRT if (static_key_false(¶virt_steal_enabled)) { - cputime_t steal_cputime; u64 steal; steal = paravirt_steal_clock(smp_processor_id()); steal -= this_rq()->prev_steal_time; + steal = min(steal, maxtime); + account_steal_time(steal); + this_rq()->prev_steal_time += steal; - steal_cputime = min(nsecs_to_cputime(steal), maxtime); - account_steal_time(steal_cputime); - this_rq()->prev_steal_time += cputime_to_nsecs(steal_cputime); - - return steal_cputime; + return steal; } #endif return 0; @@ -270,9 +255,9 @@ static __always_inline cputime_t steal_account_process_time(cputime_t maxtime) /* * Account how much elapsed time was spent in steal, irq, or softirq time. */ -static inline cputime_t account_other_time(cputime_t max) +static inline u64 account_other_time(u64 max) { - cputime_t accounted; + u64 accounted; /* Shall be converted to a lockdep-enabled lightweight check */ WARN_ON_ONCE(!irqs_disabled()); @@ -280,10 +265,7 @@ static inline cputime_t account_other_time(cputime_t max) accounted = steal_account_process_time(max); if (accounted < max) - accounted += irqtime_account_hi_update(max - accounted); - - if (accounted < max) - accounted += irqtime_account_si_update(max - accounted); + accounted += irqtime_tick_accounted(max - accounted); return accounted; } @@ -315,7 +297,7 @@ static u64 read_sum_exec_runtime(struct task_struct *t) void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) { struct signal_struct *sig = tsk->signal; - cputime_t utime, stime; + u64 utime, stime; struct task_struct *t; unsigned int seq, nextseq; unsigned long flags; @@ -379,8 +361,7 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) static void irqtime_account_process_tick(struct task_struct *p, int user_tick, struct rq *rq, int ticks) { - u64 cputime = (__force u64) cputime_one_jiffy * ticks; - cputime_t other; + u64 other, cputime = TICK_NSEC * ticks; /* * When returning from idle, many ticks can get accounted at @@ -392,6 +373,7 @@ static void irqtime_account_process_tick(struct task_struct *p, int user_tick, other = account_other_time(ULONG_MAX); if (other >= cputime) return; + cputime -= other; if (this_cpu_ksoftirqd() == p) { @@ -400,7 +382,7 @@ static void irqtime_account_process_tick(struct task_struct *p, int user_tick, * So, we have to handle it separately here. * Also, p->stime needs to be updated for ksoftirqd. */ - __account_system_time(p, cputime, CPUTIME_SOFTIRQ); + account_system_index_time(p, cputime, CPUTIME_SOFTIRQ); } else if (user_tick) { account_user_time(p, cputime); } else if (p == rq->idle) { @@ -408,7 +390,7 @@ static void irqtime_account_process_tick(struct task_struct *p, int user_tick, } else if (p->flags & PF_VCPU) { /* System time or guest time */ account_guest_time(p, cputime); } else { - __account_system_time(p, cputime, CPUTIME_SYSTEM); + account_system_index_time(p, cputime, CPUTIME_SYSTEM); } } @@ -437,9 +419,7 @@ void vtime_common_task_switch(struct task_struct *prev) else vtime_account_system(prev); -#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE - vtime_account_user(prev); -#endif + vtime_flush(prev); arch_vtime_task_switch(prev); } #endif @@ -467,14 +447,14 @@ void vtime_account_irq_enter(struct task_struct *tsk) EXPORT_SYMBOL_GPL(vtime_account_irq_enter); #endif /* __ARCH_HAS_VTIME_ACCOUNT */ -void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) +void task_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st) { *ut = p->utime; *st = p->stime; } EXPORT_SYMBOL_GPL(task_cputime_adjusted); -void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) +void thread_group_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st) { struct task_cputime cputime; @@ -491,7 +471,7 @@ void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime */ void account_process_tick(struct task_struct *p, int user_tick) { - cputime_t cputime, steal; + u64 cputime, steal; struct rq *rq = this_rq(); if (vtime_accounting_cpu_enabled()) @@ -502,7 +482,7 @@ void account_process_tick(struct task_struct *p, int user_tick) return; } - cputime = cputime_one_jiffy; + cputime = TICK_NSEC; steal = steal_account_process_time(ULONG_MAX); if (steal >= cputime) @@ -524,14 +504,14 @@ void account_process_tick(struct task_struct *p, int user_tick) */ void account_idle_ticks(unsigned long ticks) { - cputime_t cputime, steal; + u64 cputime, steal; if (sched_clock_irqtime) { irqtime_account_idle_ticks(ticks); return; } - cputime = jiffies_to_cputime(ticks); + cputime = ticks * TICK_NSEC; steal = steal_account_process_time(ULONG_MAX); if (steal >= cputime) @@ -545,7 +525,7 @@ void account_idle_ticks(unsigned long ticks) * Perform (stime * rtime) / total, but avoid multiplication overflow by * loosing precision when the numbers are big. */ -static cputime_t scale_stime(u64 stime, u64 rtime, u64 total) +static u64 scale_stime(u64 stime, u64 rtime, u64 total) { u64 scaled; @@ -582,7 +562,7 @@ drop_precision: * followed by a 64/32->64 divide. */ scaled = div_u64((u64) (u32) stime * (u64) (u32) rtime, (u32)total); - return (__force cputime_t) scaled; + return scaled; } /* @@ -607,14 +587,14 @@ drop_precision: */ static void cputime_adjust(struct task_cputime *curr, struct prev_cputime *prev, - cputime_t *ut, cputime_t *st) + u64 *ut, u64 *st) { - cputime_t rtime, stime, utime; + u64 rtime, stime, utime; unsigned long flags; /* Serialize concurrent callers such that we can honour our guarantees */ raw_spin_lock_irqsave(&prev->lock, flags); - rtime = nsecs_to_cputime(curr->sum_exec_runtime); + rtime = curr->sum_exec_runtime; /* * This is possible under two circumstances: @@ -631,9 +611,9 @@ static void cputime_adjust(struct task_cputime *curr, utime = curr->utime; /* - * If either stime or both stime and utime are 0, assume all runtime is - * userspace. Once a task gets some ticks, the monotonicy code at - * 'update' will ensure things converge to the observed ratio. + * If either stime or utime are 0, assume all runtime is userspace. + * Once a task gets some ticks, the monotonicy code at 'update:' + * will ensure things converge to the observed ratio. */ if (stime == 0) { utime = rtime; @@ -645,8 +625,7 @@ static void cputime_adjust(struct task_cputime *curr, goto update; } - stime = scale_stime((__force u64)stime, (__force u64)rtime, - (__force u64)(stime + utime)); + stime = scale_stime(stime, rtime, stime + utime); update: /* @@ -679,7 +658,7 @@ out: raw_spin_unlock_irqrestore(&prev->lock, flags); } -void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) +void task_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st) { struct task_cputime cputime = { .sum_exec_runtime = p->se.sum_exec_runtime, @@ -690,7 +669,7 @@ void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) } EXPORT_SYMBOL_GPL(task_cputime_adjusted); -void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) +void thread_group_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st) { struct task_cputime cputime; @@ -700,20 +679,21 @@ void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime #endif /* !CONFIG_VIRT_CPU_ACCOUNTING_NATIVE */ #ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN -static cputime_t vtime_delta(struct task_struct *tsk) +static u64 vtime_delta(struct vtime *vtime) { - unsigned long now = READ_ONCE(jiffies); + unsigned long long clock; - if (time_before(now, (unsigned long)tsk->vtime_snap)) + clock = sched_clock(); + if (clock < vtime->starttime) return 0; - return jiffies_to_cputime(now - tsk->vtime_snap); + return clock - vtime->starttime; } -static cputime_t get_vtime_delta(struct task_struct *tsk) +static u64 get_vtime_delta(struct vtime *vtime) { - unsigned long now = READ_ONCE(jiffies); - cputime_t delta, other; + u64 delta = vtime_delta(vtime); + u64 other; /* * Unlike tick based timing, vtime based timing never has lost @@ -722,126 +702,152 @@ static cputime_t get_vtime_delta(struct task_struct *tsk) * elapsed time. Limit account_other_time to prevent rounding * errors from causing elapsed vtime to go negative. */ - delta = jiffies_to_cputime(now - tsk->vtime_snap); other = account_other_time(delta); - WARN_ON_ONCE(tsk->vtime_snap_whence == VTIME_INACTIVE); - tsk->vtime_snap = now; + WARN_ON_ONCE(vtime->state == VTIME_INACTIVE); + vtime->starttime += delta; return delta - other; } -static void __vtime_account_system(struct task_struct *tsk) +static void __vtime_account_system(struct task_struct *tsk, + struct vtime *vtime) { - cputime_t delta_cpu = get_vtime_delta(tsk); + vtime->stime += get_vtime_delta(vtime); + if (vtime->stime >= TICK_NSEC) { + account_system_time(tsk, irq_count(), vtime->stime); + vtime->stime = 0; + } +} - account_system_time(tsk, irq_count(), delta_cpu); +static void vtime_account_guest(struct task_struct *tsk, + struct vtime *vtime) +{ + vtime->gtime += get_vtime_delta(vtime); + if (vtime->gtime >= TICK_NSEC) { + account_guest_time(tsk, vtime->gtime); + vtime->gtime = 0; + } } void vtime_account_system(struct task_struct *tsk) { - if (!vtime_delta(tsk)) + struct vtime *vtime = &tsk->vtime; + + if (!vtime_delta(vtime)) return; - write_seqcount_begin(&tsk->vtime_seqcount); - __vtime_account_system(tsk); - write_seqcount_end(&tsk->vtime_seqcount); + write_seqcount_begin(&vtime->seqcount); + /* We might have scheduled out from guest path */ + if (current->flags & PF_VCPU) + vtime_account_guest(tsk, vtime); + else + __vtime_account_system(tsk, vtime); + write_seqcount_end(&vtime->seqcount); } -void vtime_account_user(struct task_struct *tsk) +void vtime_user_enter(struct task_struct *tsk) { - cputime_t delta_cpu; + struct vtime *vtime = &tsk->vtime; - write_seqcount_begin(&tsk->vtime_seqcount); - tsk->vtime_snap_whence = VTIME_SYS; - if (vtime_delta(tsk)) { - delta_cpu = get_vtime_delta(tsk); - account_user_time(tsk, delta_cpu); - } - write_seqcount_end(&tsk->vtime_seqcount); + write_seqcount_begin(&vtime->seqcount); + __vtime_account_system(tsk, vtime); + vtime->state = VTIME_USER; + write_seqcount_end(&vtime->seqcount); } -void vtime_user_enter(struct task_struct *tsk) +void vtime_user_exit(struct task_struct *tsk) { - write_seqcount_begin(&tsk->vtime_seqcount); - if (vtime_delta(tsk)) - __vtime_account_system(tsk); - tsk->vtime_snap_whence = VTIME_USER; - write_seqcount_end(&tsk->vtime_seqcount); + struct vtime *vtime = &tsk->vtime; + + write_seqcount_begin(&vtime->seqcount); + vtime->utime += get_vtime_delta(vtime); + if (vtime->utime >= TICK_NSEC) { + account_user_time(tsk, vtime->utime); + vtime->utime = 0; + } + vtime->state = VTIME_SYS; + write_seqcount_end(&vtime->seqcount); } void vtime_guest_enter(struct task_struct *tsk) { + struct vtime *vtime = &tsk->vtime; /* * The flags must be updated under the lock with - * the vtime_snap flush and update. + * the vtime_starttime flush and update. * That enforces a right ordering and update sequence * synchronization against the reader (task_gtime()) * that can thus safely catch up with a tickless delta. */ - write_seqcount_begin(&tsk->vtime_seqcount); - if (vtime_delta(tsk)) - __vtime_account_system(tsk); + write_seqcount_begin(&vtime->seqcount); + __vtime_account_system(tsk, vtime); current->flags |= PF_VCPU; - write_seqcount_end(&tsk->vtime_seqcount); + write_seqcount_end(&vtime->seqcount); } EXPORT_SYMBOL_GPL(vtime_guest_enter); void vtime_guest_exit(struct task_struct *tsk) { - write_seqcount_begin(&tsk->vtime_seqcount); - __vtime_account_system(tsk); + struct vtime *vtime = &tsk->vtime; + + write_seqcount_begin(&vtime->seqcount); + vtime_account_guest(tsk, vtime); current->flags &= ~PF_VCPU; - write_seqcount_end(&tsk->vtime_seqcount); + write_seqcount_end(&vtime->seqcount); } EXPORT_SYMBOL_GPL(vtime_guest_exit); void vtime_account_idle(struct task_struct *tsk) { - cputime_t delta_cpu = get_vtime_delta(tsk); - - account_idle_time(delta_cpu); + account_idle_time(get_vtime_delta(&tsk->vtime)); } void arch_vtime_task_switch(struct task_struct *prev) { - write_seqcount_begin(&prev->vtime_seqcount); - prev->vtime_snap_whence = VTIME_INACTIVE; - write_seqcount_end(&prev->vtime_seqcount); + struct vtime *vtime = &prev->vtime; + + write_seqcount_begin(&vtime->seqcount); + vtime->state = VTIME_INACTIVE; + write_seqcount_end(&vtime->seqcount); + + vtime = ¤t->vtime; - write_seqcount_begin(¤t->vtime_seqcount); - current->vtime_snap_whence = VTIME_SYS; - current->vtime_snap = jiffies; - write_seqcount_end(¤t->vtime_seqcount); + write_seqcount_begin(&vtime->seqcount); + vtime->state = VTIME_SYS; + vtime->starttime = sched_clock(); + write_seqcount_end(&vtime->seqcount); } void vtime_init_idle(struct task_struct *t, int cpu) { + struct vtime *vtime = &t->vtime; unsigned long flags; local_irq_save(flags); - write_seqcount_begin(&t->vtime_seqcount); - t->vtime_snap_whence = VTIME_SYS; - t->vtime_snap = jiffies; - write_seqcount_end(&t->vtime_seqcount); + write_seqcount_begin(&vtime->seqcount); + vtime->state = VTIME_SYS; + vtime->starttime = sched_clock(); + write_seqcount_end(&vtime->seqcount); local_irq_restore(flags); } -cputime_t task_gtime(struct task_struct *t) +u64 task_gtime(struct task_struct *t) { + struct vtime *vtime = &t->vtime; unsigned int seq; - cputime_t gtime; + u64 gtime; if (!vtime_accounting_enabled()) return t->gtime; do { - seq = read_seqcount_begin(&t->vtime_seqcount); + seq = read_seqcount_begin(&vtime->seqcount); gtime = t->gtime; - if (t->vtime_snap_whence == VTIME_SYS && t->flags & PF_VCPU) - gtime += vtime_delta(t); + if (vtime->state == VTIME_SYS && t->flags & PF_VCPU) + gtime += vtime->gtime + vtime_delta(vtime); - } while (read_seqcount_retry(&t->vtime_seqcount, seq)); + } while (read_seqcount_retry(&vtime->seqcount, seq)); return gtime; } @@ -851,10 +857,11 @@ cputime_t task_gtime(struct task_struct *t) * add up the pending nohz execution time since the last * cputime snapshot. */ -void task_cputime(struct task_struct *t, cputime_t *utime, cputime_t *stime) +void task_cputime(struct task_struct *t, u64 *utime, u64 *stime) { - cputime_t delta; + struct vtime *vtime = &t->vtime; unsigned int seq; + u64 delta; if (!vtime_accounting_enabled()) { *utime = t->utime; @@ -863,25 +870,25 @@ void task_cputime(struct task_struct *t, cputime_t *utime, cputime_t *stime) } do { - seq = read_seqcount_begin(&t->vtime_seqcount); + seq = read_seqcount_begin(&vtime->seqcount); *utime = t->utime; *stime = t->stime; /* Task is sleeping, nothing to add */ - if (t->vtime_snap_whence == VTIME_INACTIVE || is_idle_task(t)) + if (vtime->state == VTIME_INACTIVE || is_idle_task(t)) continue; - delta = vtime_delta(t); + delta = vtime_delta(vtime); /* * Task runs either in user or kernel space, add pending nohz time to * the right place. */ - if (t->vtime_snap_whence == VTIME_USER || t->flags & PF_VCPU) - *utime += delta; - else if (t->vtime_snap_whence == VTIME_SYS) - *stime += delta; - } while (read_seqcount_retry(&t->vtime_seqcount, seq)); + if (vtime->state == VTIME_USER || t->flags & PF_VCPU) + *utime += vtime->utime + delta; + else if (vtime->state == VTIME_SYS) + *stime += vtime->stime + delta; + } while (read_seqcount_retry(&vtime->seqcount, seq)); } #endif /* CONFIG_VIRT_CPU_ACCOUNTING_GEN */ diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 70ef2b1901e4..755bd3f1a1a9 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -17,6 +17,7 @@ #include "sched.h" #include <linux/slab.h> +#include <uapi/linux/sched/types.h> struct dl_bandwidth def_dl_bandwidth; @@ -43,6 +44,254 @@ static inline int on_dl_rq(struct sched_dl_entity *dl_se) return !RB_EMPTY_NODE(&dl_se->rb_node); } +#ifdef CONFIG_SMP +static inline struct dl_bw *dl_bw_of(int i) +{ + RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), + "sched RCU must be held"); + return &cpu_rq(i)->rd->dl_bw; +} + +static inline int dl_bw_cpus(int i) +{ + struct root_domain *rd = cpu_rq(i)->rd; + int cpus = 0; + + RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), + "sched RCU must be held"); + for_each_cpu_and(i, rd->span, cpu_active_mask) + cpus++; + + return cpus; +} +#else +static inline struct dl_bw *dl_bw_of(int i) +{ + return &cpu_rq(i)->dl.dl_bw; +} + +static inline int dl_bw_cpus(int i) +{ + return 1; +} +#endif + +static inline +void add_running_bw(u64 dl_bw, struct dl_rq *dl_rq) +{ + u64 old = dl_rq->running_bw; + + lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock); + dl_rq->running_bw += dl_bw; + SCHED_WARN_ON(dl_rq->running_bw < old); /* overflow */ + SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw); +} + +static inline +void sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq) +{ + u64 old = dl_rq->running_bw; + + lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock); + dl_rq->running_bw -= dl_bw; + SCHED_WARN_ON(dl_rq->running_bw > old); /* underflow */ + if (dl_rq->running_bw > old) + dl_rq->running_bw = 0; +} + +static inline +void add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq) +{ + u64 old = dl_rq->this_bw; + + lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock); + dl_rq->this_bw += dl_bw; + SCHED_WARN_ON(dl_rq->this_bw < old); /* overflow */ +} + +static inline +void sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq) +{ + u64 old = dl_rq->this_bw; + + lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock); + dl_rq->this_bw -= dl_bw; + SCHED_WARN_ON(dl_rq->this_bw > old); /* underflow */ + if (dl_rq->this_bw > old) + dl_rq->this_bw = 0; + SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw); +} + +void dl_change_utilization(struct task_struct *p, u64 new_bw) +{ + struct rq *rq; + + if (task_on_rq_queued(p)) + return; + + rq = task_rq(p); + if (p->dl.dl_non_contending) { + sub_running_bw(p->dl.dl_bw, &rq->dl); + p->dl.dl_non_contending = 0; + /* + * If the timer handler is currently running and the + * timer cannot be cancelled, inactive_task_timer() + * will see that dl_not_contending is not set, and + * will not touch the rq's active utilization, + * so we are still safe. + */ + if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1) + put_task_struct(p); + } + sub_rq_bw(p->dl.dl_bw, &rq->dl); + add_rq_bw(new_bw, &rq->dl); +} + +/* + * The utilization of a task cannot be immediately removed from + * the rq active utilization (running_bw) when the task blocks. + * Instead, we have to wait for the so called "0-lag time". + * + * If a task blocks before the "0-lag time", a timer (the inactive + * timer) is armed, and running_bw is decreased when the timer + * fires. + * + * If the task wakes up again before the inactive timer fires, + * the timer is cancelled, whereas if the task wakes up after the + * inactive timer fired (and running_bw has been decreased) the + * task's utilization has to be added to running_bw again. + * A flag in the deadline scheduling entity (dl_non_contending) + * is used to avoid race conditions between the inactive timer handler + * and task wakeups. + * + * The following diagram shows how running_bw is updated. A task is + * "ACTIVE" when its utilization contributes to running_bw; an + * "ACTIVE contending" task is in the TASK_RUNNING state, while an + * "ACTIVE non contending" task is a blocked task for which the "0-lag time" + * has not passed yet. An "INACTIVE" task is a task for which the "0-lag" + * time already passed, which does not contribute to running_bw anymore. + * +------------------+ + * wakeup | ACTIVE | + * +------------------>+ contending | + * | add_running_bw | | + * | +----+------+------+ + * | | ^ + * | dequeue | | + * +--------+-------+ | | + * | | t >= 0-lag | | wakeup + * | INACTIVE |<---------------+ | + * | | sub_running_bw | | + * +--------+-------+ | | + * ^ | | + * | t < 0-lag | | + * | | | + * | V | + * | +----+------+------+ + * | sub_running_bw | ACTIVE | + * +-------------------+ | + * inactive timer | non contending | + * fired +------------------+ + * + * The task_non_contending() function is invoked when a task + * blocks, and checks if the 0-lag time already passed or + * not (in the first case, it directly updates running_bw; + * in the second case, it arms the inactive timer). + * + * The task_contending() function is invoked when a task wakes + * up, and checks if the task is still in the "ACTIVE non contending" + * state or not (in the second case, it updates running_bw). + */ +static void task_non_contending(struct task_struct *p) +{ + struct sched_dl_entity *dl_se = &p->dl; + struct hrtimer *timer = &dl_se->inactive_timer; + struct dl_rq *dl_rq = dl_rq_of_se(dl_se); + struct rq *rq = rq_of_dl_rq(dl_rq); + s64 zerolag_time; + + /* + * If this is a non-deadline task that has been boosted, + * do nothing + */ + if (dl_se->dl_runtime == 0) + return; + + WARN_ON(hrtimer_active(&dl_se->inactive_timer)); + WARN_ON(dl_se->dl_non_contending); + + zerolag_time = dl_se->deadline - + div64_long((dl_se->runtime * dl_se->dl_period), + dl_se->dl_runtime); + + /* + * Using relative times instead of the absolute "0-lag time" + * allows to simplify the code + */ + zerolag_time -= rq_clock(rq); + + /* + * If the "0-lag time" already passed, decrease the active + * utilization now, instead of starting a timer + */ + if (zerolag_time < 0) { + if (dl_task(p)) + sub_running_bw(dl_se->dl_bw, dl_rq); + if (!dl_task(p) || p->state == TASK_DEAD) { + struct dl_bw *dl_b = dl_bw_of(task_cpu(p)); + + if (p->state == TASK_DEAD) + sub_rq_bw(p->dl.dl_bw, &rq->dl); + raw_spin_lock(&dl_b->lock); + __dl_clear(dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p))); + __dl_clear_params(p); + raw_spin_unlock(&dl_b->lock); + } + + return; + } + + dl_se->dl_non_contending = 1; + get_task_struct(p); + hrtimer_start(timer, ns_to_ktime(zerolag_time), HRTIMER_MODE_REL); +} + +static void task_contending(struct sched_dl_entity *dl_se, int flags) +{ + struct dl_rq *dl_rq = dl_rq_of_se(dl_se); + + /* + * If this is a non-deadline task that has been boosted, + * do nothing + */ + if (dl_se->dl_runtime == 0) + return; + + if (flags & ENQUEUE_MIGRATED) + add_rq_bw(dl_se->dl_bw, dl_rq); + + if (dl_se->dl_non_contending) { + dl_se->dl_non_contending = 0; + /* + * If the timer handler is currently running and the + * timer cannot be cancelled, inactive_task_timer() + * will see that dl_not_contending is not set, and + * will not touch the rq's active utilization, + * so we are still safe. + */ + if (hrtimer_try_to_cancel(&dl_se->inactive_timer) == 1) + put_task_struct(dl_task_of(dl_se)); + } else { + /* + * Since "dl_non_contending" is not set, the + * task's utilization has already been removed from + * active utilization (either when the task blocked, + * when the "inactive timer" fired). + * So, add it back. + */ + add_running_bw(dl_se->dl_bw, dl_rq); + } +} + static inline int is_leftmost(struct task_struct *p, struct dl_rq *dl_rq) { struct sched_dl_entity *dl_se = &p->dl; @@ -83,6 +332,10 @@ void init_dl_rq(struct dl_rq *dl_rq) #else init_dl_bw(&dl_rq->dl_bw); #endif + + dl_rq->running_bw = 0; + dl_rq->this_bw = 0; + init_dl_rq_bw_ratio(dl_rq); } #ifdef CONFIG_SMP @@ -134,7 +387,7 @@ static void inc_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq) { struct task_struct *p = dl_task_of(dl_se); - if (tsk_nr_cpus_allowed(p) > 1) + if (p->nr_cpus_allowed > 1) dl_rq->dl_nr_migratory++; update_dl_migration(dl_rq); @@ -144,7 +397,7 @@ static void dec_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq) { struct task_struct *p = dl_task_of(dl_se); - if (tsk_nr_cpus_allowed(p) > 1) + if (p->nr_cpus_allowed > 1) dl_rq->dl_nr_migratory--; update_dl_migration(dl_rq); @@ -252,7 +505,7 @@ static struct rq *dl_task_offline_migration(struct rq *rq, struct task_struct *p * If we cannot preempt any rq, fall back to pick any * online cpu. */ - cpu = cpumask_any_and(cpu_active_mask, tsk_cpus_allowed(p)); + cpu = cpumask_any_and(cpu_active_mask, &p->cpus_allowed); if (cpu >= nr_cpu_ids) { /* * Fail to find any suitable cpu. @@ -445,13 +698,13 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se, * * This function returns true if: * - * runtime / (deadline - t) > dl_runtime / dl_period , + * runtime / (deadline - t) > dl_runtime / dl_deadline , * * IOW we can't recycle current parameters. * - * Notice that the bandwidth check is done against the period. For + * Notice that the bandwidth check is done against the deadline. For * task with deadline equal to period this is the same of using - * dl_deadline instead of dl_period in the equation above. + * dl_period instead of dl_deadline in the equation above. */ static bool dl_entity_overflow(struct sched_dl_entity *dl_se, struct sched_dl_entity *pi_se, u64 t) @@ -476,7 +729,7 @@ static bool dl_entity_overflow(struct sched_dl_entity *dl_se, * of anything below microseconds resolution is actually fiction * (but still we want to give the user that illusion >;). */ - left = (pi_se->dl_period >> DL_SCALE) * (dl_se->runtime >> DL_SCALE); + left = (pi_se->dl_deadline >> DL_SCALE) * (dl_se->runtime >> DL_SCALE); right = ((dl_se->deadline - t) >> DL_SCALE) * (pi_se->dl_runtime >> DL_SCALE); @@ -484,13 +737,84 @@ static bool dl_entity_overflow(struct sched_dl_entity *dl_se, } /* - * When a -deadline entity is queued back on the runqueue, its runtime and - * deadline might need updating. + * Revised wakeup rule [1]: For self-suspending tasks, rather then + * re-initializing task's runtime and deadline, the revised wakeup + * rule adjusts the task's runtime to avoid the task to overrun its + * density. + * + * Reasoning: a task may overrun the density if: + * runtime / (deadline - t) > dl_runtime / dl_deadline + * + * Therefore, runtime can be adjusted to: + * runtime = (dl_runtime / dl_deadline) * (deadline - t) + * + * In such way that runtime will be equal to the maximum density + * the task can use without breaking any rule. + * + * [1] Luca Abeni, Giuseppe Lipari, and Juri Lelli. 2015. Constant + * bandwidth server revisited. SIGBED Rev. 11, 4 (January 2015), 19-24. + */ +static void +update_dl_revised_wakeup(struct sched_dl_entity *dl_se, struct rq *rq) +{ + u64 laxity = dl_se->deadline - rq_clock(rq); + + /* + * If the task has deadline < period, and the deadline is in the past, + * it should already be throttled before this check. + * + * See update_dl_entity() comments for further details. + */ + WARN_ON(dl_time_before(dl_se->deadline, rq_clock(rq))); + + dl_se->runtime = (dl_se->dl_density * laxity) >> BW_SHIFT; +} + +/* + * Regarding the deadline, a task with implicit deadline has a relative + * deadline == relative period. A task with constrained deadline has a + * relative deadline <= relative period. + * + * We support constrained deadline tasks. However, there are some restrictions + * applied only for tasks which do not have an implicit deadline. See + * update_dl_entity() to know more about such restrictions. * - * The policy here is that we update the deadline of the entity only if: - * - the current deadline is in the past, - * - using the remaining runtime with the current deadline would make - * the entity exceed its bandwidth. + * The dl_is_implicit() returns true if the task has an implicit deadline. + */ +static inline bool dl_is_implicit(struct sched_dl_entity *dl_se) +{ + return dl_se->dl_deadline == dl_se->dl_period; +} + +/* + * When a deadline entity is placed in the runqueue, its runtime and deadline + * might need to be updated. This is done by a CBS wake up rule. There are two + * different rules: 1) the original CBS; and 2) the Revisited CBS. + * + * When the task is starting a new period, the Original CBS is used. In this + * case, the runtime is replenished and a new absolute deadline is set. + * + * When a task is queued before the begin of the next period, using the + * remaining runtime and deadline could make the entity to overflow, see + * dl_entity_overflow() to find more about runtime overflow. When such case + * is detected, the runtime and deadline need to be updated. + * + * If the task has an implicit deadline, i.e., deadline == period, the Original + * CBS is applied. the runtime is replenished and a new absolute deadline is + * set, as in the previous cases. + * + * However, the Original CBS does not work properly for tasks with + * deadline < period, which are said to have a constrained deadline. By + * applying the Original CBS, a constrained deadline task would be able to run + * runtime/deadline in a period. With deadline < period, the task would + * overrun the runtime/period allowed bandwidth, breaking the admission test. + * + * In order to prevent this misbehave, the Revisited CBS is used for + * constrained deadline tasks when a runtime overflow is detected. In the + * Revisited CBS, rather than replenishing & setting a new absolute deadline, + * the remaining runtime of the task is reduced to avoid runtime overflow. + * Please refer to the comments update_dl_revised_wakeup() function to find + * more about the Revised CBS rule. */ static void update_dl_entity(struct sched_dl_entity *dl_se, struct sched_dl_entity *pi_se) @@ -500,15 +824,28 @@ static void update_dl_entity(struct sched_dl_entity *dl_se, if (dl_time_before(dl_se->deadline, rq_clock(rq)) || dl_entity_overflow(dl_se, pi_se, rq_clock(rq))) { + + if (unlikely(!dl_is_implicit(dl_se) && + !dl_time_before(dl_se->deadline, rq_clock(rq)) && + !dl_se->dl_boosted)){ + update_dl_revised_wakeup(dl_se, rq); + return; + } + dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline; dl_se->runtime = pi_se->dl_runtime; } } +static inline u64 dl_next_period(struct sched_dl_entity *dl_se) +{ + return dl_se->deadline - dl_se->dl_deadline + dl_se->dl_period; +} + /* * If the entity depleted all its runtime, and if we want it to sleep * while waiting for some new execution time to become available, we - * set the bandwidth enforcement timer to the replenishment instant + * set the bandwidth replenishment timer to the replenishment instant * and try to activate it. * * Notice that it is important for the caller to know if the timer @@ -530,7 +867,7 @@ static int start_dl_timer(struct task_struct *p) * that it is actually coming from rq->clock and not from * hrtimer's time base reading. */ - act = ns_to_ktime(dl_se->deadline); + act = ns_to_ktime(dl_next_period(dl_se)); now = hrtimer_cb_get_time(timer); delta = ktime_to_ns(now) - rq_clock(rq); act = ktime_add_ns(act, delta); @@ -588,10 +925,8 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) * The task might have changed its scheduling policy to something * different than SCHED_DEADLINE (through switched_from_dl()). */ - if (!dl_task(p)) { - __dl_clear_params(p); + if (!dl_task(p)) goto unlock; - } /* * The task might have been boosted by someone else and might be in the @@ -638,6 +973,7 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) lockdep_unpin_lock(&rq->lock, rf.cookie); rq = dl_task_offline_migration(rq, p); rf.cookie = lockdep_pin_lock(&rq->lock); + update_rq_clock(rq); /* * Now that the task has been migrated to the new RQ and we @@ -663,9 +999,9 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) * Nothing relies on rq->lock after this, so its safe to drop * rq->lock. */ - lockdep_unpin_lock(&rq->lock, rf.cookie); + rq_unpin_lock(rq, &rf); push_dl_task(rq); - lockdep_repin_lock(&rq->lock, rf.cookie); + rq_repin_lock(rq, &rf); } #endif @@ -689,6 +1025,39 @@ void init_dl_task_timer(struct sched_dl_entity *dl_se) timer->function = dl_task_timer; } +/* + * During the activation, CBS checks if it can reuse the current task's + * runtime and period. If the deadline of the task is in the past, CBS + * cannot use the runtime, and so it replenishes the task. This rule + * works fine for implicit deadline tasks (deadline == period), and the + * CBS was designed for implicit deadline tasks. However, a task with + * constrained deadline (deadine < period) might be awakened after the + * deadline, but before the next period. In this case, replenishing the + * task would allow it to run for runtime / deadline. As in this case + * deadline < period, CBS enables a task to run for more than the + * runtime / period. In a very loaded system, this can cause a domino + * effect, making other tasks miss their deadlines. + * + * To avoid this problem, in the activation of a constrained deadline + * task after the deadline but before the next period, throttle the + * task and set the replenishing timer to the begin of the next period, + * unless it is boosted. + */ +static inline void dl_check_constrained_dl(struct sched_dl_entity *dl_se) +{ + struct task_struct *p = dl_task_of(dl_se); + struct rq *rq = rq_of_dl_rq(dl_rq_of_se(dl_se)); + + if (dl_time_before(dl_se->deadline, rq_clock(rq)) && + dl_time_before(rq_clock(rq), dl_next_period(dl_se))) { + if (unlikely(dl_se->dl_boosted || !start_dl_timer(p))) + return; + dl_se->dl_throttled = 1; + if (dl_se->runtime > 0) + dl_se->runtime = 0; + } +} + static int dl_runtime_exceeded(struct sched_dl_entity *dl_se) { @@ -698,6 +1067,47 @@ int dl_runtime_exceeded(struct sched_dl_entity *dl_se) extern bool sched_rt_bandwidth_account(struct rt_rq *rt_rq); /* + * This function implements the GRUB accounting rule: + * according to the GRUB reclaiming algorithm, the runtime is + * not decreased as "dq = -dt", but as + * "dq = -max{u / Umax, (1 - Uinact - Uextra)} dt", + * where u is the utilization of the task, Umax is the maximum reclaimable + * utilization, Uinact is the (per-runqueue) inactive utilization, computed + * as the difference between the "total runqueue utilization" and the + * runqueue active utilization, and Uextra is the (per runqueue) extra + * reclaimable utilization. + * Since rq->dl.running_bw and rq->dl.this_bw contain utilizations + * multiplied by 2^BW_SHIFT, the result has to be shifted right by + * BW_SHIFT. + * Since rq->dl.bw_ratio contains 1 / Umax multipled by 2^RATIO_SHIFT, + * dl_bw is multiped by rq->dl.bw_ratio and shifted right by RATIO_SHIFT. + * Since delta is a 64 bit variable, to have an overflow its value + * should be larger than 2^(64 - 20 - 8), which is more than 64 seconds. + * So, overflow is not an issue here. + */ +u64 grub_reclaim(u64 delta, struct rq *rq, struct sched_dl_entity *dl_se) +{ + u64 u_inact = rq->dl.this_bw - rq->dl.running_bw; /* Utot - Uact */ + u64 u_act; + u64 u_act_min = (dl_se->dl_bw * rq->dl.bw_ratio) >> RATIO_SHIFT; + + /* + * Instead of computing max{u * bw_ratio, (1 - u_inact - u_extra)}, + * we compare u_inact + rq->dl.extra_bw with + * 1 - (u * rq->dl.bw_ratio >> RATIO_SHIFT), because + * u_inact + rq->dl.extra_bw can be larger than + * 1 * (so, 1 - u_inact - rq->dl.extra_bw would be negative + * leading to wrong results) + */ + if (u_inact + rq->dl.extra_bw > BW_UNIT - u_act_min) + u_act = u_act_min; + else + u_act = BW_UNIT - u_inact - rq->dl.extra_bw; + + return (delta * u_act) >> BW_SHIFT; +} + +/* * Update the current task's runtime statistics (provided it is still * a -deadline task and has not been removed from the dl_rq). */ @@ -739,6 +1149,8 @@ static void update_curr_dl(struct rq *rq) sched_rt_avg_update(rq, delta_exec); + if (unlikely(dl_se->flags & SCHED_FLAG_RECLAIM)) + delta_exec = grub_reclaim(delta_exec, rq, &curr->dl); dl_se->runtime -= delta_exec; throttle: @@ -778,6 +1190,56 @@ throttle: } } +static enum hrtimer_restart inactive_task_timer(struct hrtimer *timer) +{ + struct sched_dl_entity *dl_se = container_of(timer, + struct sched_dl_entity, + inactive_timer); + struct task_struct *p = dl_task_of(dl_se); + struct rq_flags rf; + struct rq *rq; + + rq = task_rq_lock(p, &rf); + + if (!dl_task(p) || p->state == TASK_DEAD) { + struct dl_bw *dl_b = dl_bw_of(task_cpu(p)); + + if (p->state == TASK_DEAD && dl_se->dl_non_contending) { + sub_running_bw(p->dl.dl_bw, dl_rq_of_se(&p->dl)); + sub_rq_bw(p->dl.dl_bw, dl_rq_of_se(&p->dl)); + dl_se->dl_non_contending = 0; + } + + raw_spin_lock(&dl_b->lock); + __dl_clear(dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p))); + raw_spin_unlock(&dl_b->lock); + __dl_clear_params(p); + + goto unlock; + } + if (dl_se->dl_non_contending == 0) + goto unlock; + + sched_clock_tick(); + update_rq_clock(rq); + + sub_running_bw(dl_se->dl_bw, &rq->dl); + dl_se->dl_non_contending = 0; +unlock: + task_rq_unlock(rq, p, &rf); + put_task_struct(p); + + return HRTIMER_NORESTART; +} + +void init_dl_inactive_task_timer(struct sched_dl_entity *dl_se) +{ + struct hrtimer *timer = &dl_se->inactive_timer; + + hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + timer->function = inactive_task_timer; +} + #ifdef CONFIG_SMP static void inc_dl_deadline(struct dl_rq *dl_rq, u64 deadline) @@ -909,10 +1371,12 @@ enqueue_dl_entity(struct sched_dl_entity *dl_se, * parameters of the task might need updating. Otherwise, * we want a replenishment of its runtime. */ - if (flags & ENQUEUE_WAKEUP) + if (flags & ENQUEUE_WAKEUP) { + task_contending(dl_se, flags); update_dl_entity(dl_se, pi_se); - else if (flags & ENQUEUE_REPLENISH) + } else if (flags & ENQUEUE_REPLENISH) { replenish_dl_entity(dl_se, pi_se); + } __enqueue_dl_entity(dl_se); } @@ -928,17 +1392,19 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags) struct sched_dl_entity *pi_se = &p->dl; /* - * Use the scheduling parameters of the top pi-waiter - * task if we have one and its (absolute) deadline is - * smaller than our one... OTW we keep our runtime and - * deadline. + * Use the scheduling parameters of the top pi-waiter task if: + * - we have a top pi-waiter which is a SCHED_DEADLINE task AND + * - our dl_boosted is set (i.e. the pi-waiter's (absolute) deadline is + * smaller than our deadline OR we are a !SCHED_DEADLINE task getting + * boosted due to a SCHED_DEADLINE pi-waiter). + * Otherwise we keep our runtime and deadline. */ - if (pi_task && p->dl.dl_boosted && dl_prio(pi_task->normal_prio)) { + if (pi_task && dl_prio(pi_task->normal_prio) && p->dl.dl_boosted) { pi_se = &pi_task->dl; } else if (!dl_prio(p->normal_prio)) { /* * Special case in which we have a !SCHED_DEADLINE task - * that is going to be deboosted, but exceedes its + * that is going to be deboosted, but exceeds its * runtime while doing so. No point in replenishing * it, as it's going to return back to its original * scheduling class after this. @@ -948,17 +1414,41 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags) } /* - * If p is throttled, we do nothing. In fact, if it exhausted + * Check if a constrained deadline task was activated + * after the deadline but before the next period. + * If that is the case, the task will be throttled and + * the replenishment timer will be set to the next period. + */ + if (!p->dl.dl_throttled && !dl_is_implicit(&p->dl)) + dl_check_constrained_dl(&p->dl); + + if (p->on_rq == TASK_ON_RQ_MIGRATING || flags & ENQUEUE_RESTORE) { + add_rq_bw(p->dl.dl_bw, &rq->dl); + add_running_bw(p->dl.dl_bw, &rq->dl); + } + + /* + * If p is throttled, we do not enqueue it. In fact, if it exhausted * its budget it needs a replenishment and, since it now is on * its rq, the bandwidth timer callback (which clearly has not * run yet) will take care of this. + * However, the active utilization does not depend on the fact + * that the task is on the runqueue or not (but depends on the + * task's state - in GRUB parlance, "inactive" vs "active contending"). + * In other words, even if a task is throttled its utilization must + * be counted in the active utilization; hence, we need to call + * add_running_bw(). */ - if (p->dl.dl_throttled && !(flags & ENQUEUE_REPLENISH)) + if (p->dl.dl_throttled && !(flags & ENQUEUE_REPLENISH)) { + if (flags & ENQUEUE_WAKEUP) + task_contending(&p->dl, flags); + return; + } enqueue_dl_entity(&p->dl, pi_se, flags); - if (!task_current(rq, p) && tsk_nr_cpus_allowed(p) > 1) + if (!task_current(rq, p) && p->nr_cpus_allowed > 1) enqueue_pushable_dl_task(rq, p); } @@ -972,6 +1462,23 @@ static void dequeue_task_dl(struct rq *rq, struct task_struct *p, int flags) { update_curr_dl(rq); __dequeue_task_dl(rq, p, flags); + + if (p->on_rq == TASK_ON_RQ_MIGRATING || flags & DEQUEUE_SAVE) { + sub_running_bw(p->dl.dl_bw, &rq->dl); + sub_rq_bw(p->dl.dl_bw, &rq->dl); + } + + /* + * This check allows to start the inactive timer (or to immediately + * decrease the active utilization, if needed) in two cases: + * when the task blocks and when it is terminating + * (p->state == TASK_DEAD). We can handle the two cases in the same + * way, because from GRUB's point of view the same thing is happening + * (the task moves from "active contending" to "active non contending" + * or "inactive") + */ + if (flags & DEQUEUE_SLEEP) + task_non_contending(p); } /* @@ -1032,9 +1539,9 @@ select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags) * try to make it stay here, it might be important. */ if (unlikely(dl_task(curr)) && - (tsk_nr_cpus_allowed(curr) < 2 || + (curr->nr_cpus_allowed < 2 || !dl_entity_preempt(&p->dl, &curr->dl)) && - (tsk_nr_cpus_allowed(p) > 1)) { + (p->nr_cpus_allowed > 1)) { int target = find_later_rq(p); if (target != -1 && @@ -1049,13 +1556,44 @@ out: return cpu; } +static void migrate_task_rq_dl(struct task_struct *p) +{ + struct rq *rq; + + if (p->state != TASK_WAKING) + return; + + rq = task_rq(p); + /* + * Since p->state == TASK_WAKING, set_task_cpu() has been called + * from try_to_wake_up(). Hence, p->pi_lock is locked, but + * rq->lock is not... So, lock it + */ + raw_spin_lock(&rq->lock); + if (p->dl.dl_non_contending) { + sub_running_bw(p->dl.dl_bw, &rq->dl); + p->dl.dl_non_contending = 0; + /* + * If the timer handler is currently running and the + * timer cannot be cancelled, inactive_task_timer() + * will see that dl_not_contending is not set, and + * will not touch the rq's active utilization, + * so we are still safe. + */ + if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1) + put_task_struct(p); + } + sub_rq_bw(p->dl.dl_bw, &rq->dl); + raw_spin_unlock(&rq->lock); +} + static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p) { /* * Current can't be migrated, useless to reschedule, * let's hope p can move out. */ - if (tsk_nr_cpus_allowed(rq->curr) == 1 || + if (rq->curr->nr_cpus_allowed == 1 || cpudl_find(&rq->rd->cpudl, rq->curr, NULL) == -1) return; @@ -1063,7 +1601,7 @@ static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p) * p is migratable, so let's not schedule it and * see if it is pushed or pulled somewhere else. */ - if (tsk_nr_cpus_allowed(p) != 1 && + if (p->nr_cpus_allowed != 1 && cpudl_find(&rq->rd->cpudl, p, NULL) != -1) return; @@ -1118,7 +1656,7 @@ static struct sched_dl_entity *pick_next_dl_entity(struct rq *rq, } struct task_struct * -pick_next_task_dl(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie) +pick_next_task_dl(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { struct sched_dl_entity *dl_se; struct task_struct *p; @@ -1133,9 +1671,9 @@ pick_next_task_dl(struct rq *rq, struct task_struct *prev, struct pin_cookie coo * disabled avoiding further scheduler activity on it and we're * being very careful to re-start the picking loop. */ - lockdep_unpin_lock(&rq->lock, cookie); + rq_unpin_lock(rq, rf); pull_dl_task(rq); - lockdep_repin_lock(&rq->lock, cookie); + rq_repin_lock(rq, rf); /* * pull_dl_task() can drop (and re-acquire) rq->lock; this * means a stop task can slip in, in which case we need to @@ -1178,7 +1716,7 @@ static void put_prev_task_dl(struct rq *rq, struct task_struct *p) { update_curr_dl(rq); - if (on_dl_rq(&p->dl) && tsk_nr_cpus_allowed(p) > 1) + if (on_dl_rq(&p->dl) && p->nr_cpus_allowed > 1) enqueue_pushable_dl_task(rq, p); } @@ -1204,19 +1742,6 @@ static void task_fork_dl(struct task_struct *p) */ } -static void task_dead_dl(struct task_struct *p) -{ - struct dl_bw *dl_b = dl_bw_of(task_cpu(p)); - - /* - * Since we are TASK_DEAD we won't slip out of the domain! - */ - raw_spin_lock_irq(&dl_b->lock); - /* XXX we should retain the bw until 0-lag */ - dl_b->total_bw -= p->dl.dl_bw; - raw_spin_unlock_irq(&dl_b->lock); -} - static void set_curr_task_dl(struct rq *rq) { struct task_struct *p = rq->curr; @@ -1235,7 +1760,7 @@ static void set_curr_task_dl(struct rq *rq) static int pick_dl_task(struct rq *rq, struct task_struct *p, int cpu) { if (!task_running(rq, p) && - cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) + cpumask_test_cpu(cpu, &p->cpus_allowed)) return 1; return 0; } @@ -1279,7 +1804,7 @@ static int find_later_rq(struct task_struct *task) if (unlikely(!later_mask)) return -1; - if (tsk_nr_cpus_allowed(task) == 1) + if (task->nr_cpus_allowed == 1) return -1; /* @@ -1384,8 +1909,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) /* Retry if something changed. */ if (double_lock_balance(rq, later_rq)) { if (unlikely(task_rq(task) != rq || - !cpumask_test_cpu(later_rq->cpu, - tsk_cpus_allowed(task)) || + !cpumask_test_cpu(later_rq->cpu, &task->cpus_allowed) || task_running(rq, task) || !dl_task(task) || !task_on_rq_queued(task))) { @@ -1425,7 +1949,7 @@ static struct task_struct *pick_next_pushable_dl_task(struct rq *rq) BUG_ON(rq->cpu != task_cpu(p)); BUG_ON(task_current(rq, p)); - BUG_ON(tsk_nr_cpus_allowed(p) <= 1); + BUG_ON(p->nr_cpus_allowed <= 1); BUG_ON(!task_on_rq_queued(p)); BUG_ON(!dl_task(p)); @@ -1464,7 +1988,7 @@ retry: */ if (dl_task(rq->curr) && dl_time_before(next_task->dl.deadline, rq->curr->dl.deadline) && - tsk_nr_cpus_allowed(rq->curr) > 1) { + rq->curr->nr_cpus_allowed > 1) { resched_curr(rq); return 0; } @@ -1483,7 +2007,7 @@ retry: * then possible that next_task has migrated. */ task = pick_next_pushable_dl_task(rq); - if (task_cpu(next_task) == rq->cpu && task == next_task) { + if (task == next_task) { /* * The task is still there. We don't try * again, some other cpu will pull it when ready. @@ -1501,7 +2025,11 @@ retry: } deactivate_task(rq, next_task, 0); + sub_running_bw(next_task->dl.dl_bw, &rq->dl); + sub_rq_bw(next_task->dl.dl_bw, &rq->dl); set_task_cpu(next_task, later_rq->cpu); + add_rq_bw(next_task->dl.dl_bw, &later_rq->dl); + add_running_bw(next_task->dl.dl_bw, &later_rq->dl); activate_task(later_rq, next_task, 0); ret = 1; @@ -1589,7 +2117,11 @@ static void pull_dl_task(struct rq *this_rq) resched = true; deactivate_task(src_rq, p, 0); + sub_running_bw(p->dl.dl_bw, &src_rq->dl); + sub_rq_bw(p->dl.dl_bw, &src_rq->dl); set_task_cpu(p, this_cpu); + add_rq_bw(p->dl.dl_bw, &this_rq->dl); + add_running_bw(p->dl.dl_bw, &this_rq->dl); activate_task(this_rq, p, 0); dmin = p->dl.deadline; @@ -1611,9 +2143,9 @@ static void task_woken_dl(struct rq *rq, struct task_struct *p) { if (!task_running(rq, p) && !test_tsk_need_resched(rq->curr) && - tsk_nr_cpus_allowed(p) > 1 && + p->nr_cpus_allowed > 1 && dl_task(rq->curr) && - (tsk_nr_cpus_allowed(rq->curr) < 2 || + (rq->curr->nr_cpus_allowed < 2 || !dl_entity_preempt(&p->dl, &rq->curr->dl))) { push_dl_tasks(rq); } @@ -1645,7 +2177,7 @@ static void set_cpus_allowed_dl(struct task_struct *p, * until we complete the update. */ raw_spin_lock(&src_dl_b->lock); - __dl_clear(src_dl_b, p->dl.dl_bw); + __dl_clear(src_dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p))); raw_spin_unlock(&src_dl_b->lock); } @@ -1687,13 +2219,26 @@ void __init init_sched_dl_class(void) static void switched_from_dl(struct rq *rq, struct task_struct *p) { /* - * Start the deadline timer; if we switch back to dl before this we'll - * continue consuming our current CBS slice. If we stay outside of - * SCHED_DEADLINE until the deadline passes, the timer will reset the - * task. + * task_non_contending() can start the "inactive timer" (if the 0-lag + * time is in the future). If the task switches back to dl before + * the "inactive timer" fires, it can continue to consume its current + * runtime using its current deadline. If it stays outside of + * SCHED_DEADLINE until the 0-lag time passes, inactive_task_timer() + * will reset the task parameters. */ - if (!start_dl_timer(p)) - __dl_clear_params(p); + if (task_on_rq_queued(p) && p->dl.dl_runtime) + task_non_contending(p); + + if (!task_on_rq_queued(p)) + sub_rq_bw(p->dl.dl_bw, &rq->dl); + + /* + * We cannot use inactive_task_timer() to invoke sub_running_bw() + * at the 0-lag time, because the task could have been migrated + * while SCHED_OTHER in the meanwhile. + */ + if (p->dl.dl_non_contending) + p->dl.dl_non_contending = 0; /* * Since this might be the only -deadline task on the rq, @@ -1712,11 +2257,15 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p) */ static void switched_to_dl(struct rq *rq, struct task_struct *p) { + if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1) + put_task_struct(p); /* If p is not queued we will update its parameters at next wakeup. */ - if (!task_on_rq_queued(p)) - return; + if (!task_on_rq_queued(p)) { + add_rq_bw(p->dl.dl_bw, &rq->dl); + return; + } /* * If p is boosted we already updated its params in * rt_mutex_setprio()->enqueue_task(..., ENQUEUE_REPLENISH), @@ -1727,14 +2276,13 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p) if (rq->curr != p) { #ifdef CONFIG_SMP - if (tsk_nr_cpus_allowed(p) > 1 && rq->dl.overloaded) + if (p->nr_cpus_allowed > 1 && rq->dl.overloaded) queue_push_tasks(rq); -#else +#endif if (dl_task(rq->curr)) check_preempt_curr_dl(rq, p, 0); else resched_curr(rq); -#endif } } @@ -1787,6 +2335,7 @@ const struct sched_class dl_sched_class = { #ifdef CONFIG_SMP .select_task_rq = select_task_rq_dl, + .migrate_task_rq = migrate_task_rq_dl, .set_cpus_allowed = set_cpus_allowed_dl, .rq_online = rq_online_dl, .rq_offline = rq_offline_dl, @@ -1796,7 +2345,6 @@ const struct sched_class dl_sched_class = { .set_curr_task = set_curr_task_dl, .task_tick = task_tick_dl, .task_fork = task_fork_dl, - .task_dead = task_dead_dl, .prio_changed = prio_changed_dl, .switched_from = switched_from_dl, @@ -1805,6 +2353,317 @@ const struct sched_class dl_sched_class = { .update_curr = update_curr_dl, }; +int sched_dl_global_validate(void) +{ + u64 runtime = global_rt_runtime(); + u64 period = global_rt_period(); + u64 new_bw = to_ratio(period, runtime); + struct dl_bw *dl_b; + int cpu, ret = 0; + unsigned long flags; + + /* + * Here we want to check the bandwidth not being set to some + * value smaller than the currently allocated bandwidth in + * any of the root_domains. + * + * FIXME: Cycling on all the CPUs is overdoing, but simpler than + * cycling on root_domains... Discussion on different/better + * solutions is welcome! + */ + for_each_possible_cpu(cpu) { + rcu_read_lock_sched(); + dl_b = dl_bw_of(cpu); + + raw_spin_lock_irqsave(&dl_b->lock, flags); + if (new_bw < dl_b->total_bw) + ret = -EBUSY; + raw_spin_unlock_irqrestore(&dl_b->lock, flags); + + rcu_read_unlock_sched(); + + if (ret) + break; + } + + return ret; +} + +void init_dl_rq_bw_ratio(struct dl_rq *dl_rq) +{ + if (global_rt_runtime() == RUNTIME_INF) { + dl_rq->bw_ratio = 1 << RATIO_SHIFT; + dl_rq->extra_bw = 1 << BW_SHIFT; + } else { + dl_rq->bw_ratio = to_ratio(global_rt_runtime(), + global_rt_period()) >> (BW_SHIFT - RATIO_SHIFT); + dl_rq->extra_bw = to_ratio(global_rt_period(), + global_rt_runtime()); + } +} + +void sched_dl_do_global(void) +{ + u64 new_bw = -1; + struct dl_bw *dl_b; + int cpu; + unsigned long flags; + + def_dl_bandwidth.dl_period = global_rt_period(); + def_dl_bandwidth.dl_runtime = global_rt_runtime(); + + if (global_rt_runtime() != RUNTIME_INF) + new_bw = to_ratio(global_rt_period(), global_rt_runtime()); + + /* + * FIXME: As above... + */ + for_each_possible_cpu(cpu) { + rcu_read_lock_sched(); + dl_b = dl_bw_of(cpu); + + raw_spin_lock_irqsave(&dl_b->lock, flags); + dl_b->bw = new_bw; + raw_spin_unlock_irqrestore(&dl_b->lock, flags); + + rcu_read_unlock_sched(); + init_dl_rq_bw_ratio(&cpu_rq(cpu)->dl); + } +} + +/* + * We must be sure that accepting a new task (or allowing changing the + * parameters of an existing one) is consistent with the bandwidth + * constraints. If yes, this function also accordingly updates the currently + * allocated bandwidth to reflect the new situation. + * + * This function is called while holding p's rq->lock. + */ +int sched_dl_overflow(struct task_struct *p, int policy, + const struct sched_attr *attr) +{ + struct dl_bw *dl_b = dl_bw_of(task_cpu(p)); + u64 period = attr->sched_period ?: attr->sched_deadline; + u64 runtime = attr->sched_runtime; + u64 new_bw = dl_policy(policy) ? to_ratio(period, runtime) : 0; + int cpus, err = -1; + + /* !deadline task may carry old deadline bandwidth */ + if (new_bw == p->dl.dl_bw && task_has_dl_policy(p)) + return 0; + + /* + * Either if a task, enters, leave, or stays -deadline but changes + * its parameters, we may need to update accordingly the total + * allocated bandwidth of the container. + */ + raw_spin_lock(&dl_b->lock); + cpus = dl_bw_cpus(task_cpu(p)); + if (dl_policy(policy) && !task_has_dl_policy(p) && + !__dl_overflow(dl_b, cpus, 0, new_bw)) { + if (hrtimer_active(&p->dl.inactive_timer)) + __dl_clear(dl_b, p->dl.dl_bw, cpus); + __dl_add(dl_b, new_bw, cpus); + err = 0; + } else if (dl_policy(policy) && task_has_dl_policy(p) && + !__dl_overflow(dl_b, cpus, p->dl.dl_bw, new_bw)) { + /* + * XXX this is slightly incorrect: when the task + * utilization decreases, we should delay the total + * utilization change until the task's 0-lag point. + * But this would require to set the task's "inactive + * timer" when the task is not inactive. + */ + __dl_clear(dl_b, p->dl.dl_bw, cpus); + __dl_add(dl_b, new_bw, cpus); + dl_change_utilization(p, new_bw); + err = 0; + } else if (!dl_policy(policy) && task_has_dl_policy(p)) { + /* + * Do not decrease the total deadline utilization here, + * switched_from_dl() will take care to do it at the correct + * (0-lag) time. + */ + err = 0; + } + raw_spin_unlock(&dl_b->lock); + + return err; +} + +/* + * This function initializes the sched_dl_entity of a newly becoming + * SCHED_DEADLINE task. + * + * Only the static values are considered here, the actual runtime and the + * absolute deadline will be properly calculated when the task is enqueued + * for the first time with its new policy. + */ +void __setparam_dl(struct task_struct *p, const struct sched_attr *attr) +{ + struct sched_dl_entity *dl_se = &p->dl; + + dl_se->dl_runtime = attr->sched_runtime; + dl_se->dl_deadline = attr->sched_deadline; + dl_se->dl_period = attr->sched_period ?: dl_se->dl_deadline; + dl_se->flags = attr->sched_flags; + dl_se->dl_bw = to_ratio(dl_se->dl_period, dl_se->dl_runtime); + dl_se->dl_density = to_ratio(dl_se->dl_deadline, dl_se->dl_runtime); +} + +void __getparam_dl(struct task_struct *p, struct sched_attr *attr) +{ + struct sched_dl_entity *dl_se = &p->dl; + + attr->sched_priority = p->rt_priority; + attr->sched_runtime = dl_se->dl_runtime; + attr->sched_deadline = dl_se->dl_deadline; + attr->sched_period = dl_se->dl_period; + attr->sched_flags = dl_se->flags; +} + +/* + * This function validates the new parameters of a -deadline task. + * We ask for the deadline not being zero, and greater or equal + * than the runtime, as well as the period of being zero or + * greater than deadline. Furthermore, we have to be sure that + * user parameters are above the internal resolution of 1us (we + * check sched_runtime only since it is always the smaller one) and + * below 2^63 ns (we have to check both sched_deadline and + * sched_period, as the latter can be zero). + */ +bool __checkparam_dl(const struct sched_attr *attr) +{ + /* deadline != 0 */ + if (attr->sched_deadline == 0) + return false; + + /* + * Since we truncate DL_SCALE bits, make sure we're at least + * that big. + */ + if (attr->sched_runtime < (1ULL << DL_SCALE)) + return false; + + /* + * Since we use the MSB for wrap-around and sign issues, make + * sure it's not set (mind that period can be equal to zero). + */ + if (attr->sched_deadline & (1ULL << 63) || + attr->sched_period & (1ULL << 63)) + return false; + + /* runtime <= deadline <= period (if period != 0) */ + if ((attr->sched_period != 0 && + attr->sched_period < attr->sched_deadline) || + attr->sched_deadline < attr->sched_runtime) + return false; + + return true; +} + +/* + * This function clears the sched_dl_entity static params. + */ +void __dl_clear_params(struct task_struct *p) +{ + struct sched_dl_entity *dl_se = &p->dl; + + dl_se->dl_runtime = 0; + dl_se->dl_deadline = 0; + dl_se->dl_period = 0; + dl_se->flags = 0; + dl_se->dl_bw = 0; + dl_se->dl_density = 0; + + dl_se->dl_throttled = 0; + dl_se->dl_yielded = 0; + dl_se->dl_non_contending = 0; +} + +bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr) +{ + struct sched_dl_entity *dl_se = &p->dl; + + if (dl_se->dl_runtime != attr->sched_runtime || + dl_se->dl_deadline != attr->sched_deadline || + dl_se->dl_period != attr->sched_period || + dl_se->flags != attr->sched_flags) + return true; + + return false; +} + +#ifdef CONFIG_SMP +int dl_task_can_attach(struct task_struct *p, const struct cpumask *cs_cpus_allowed) +{ + unsigned int dest_cpu = cpumask_any_and(cpu_active_mask, + cs_cpus_allowed); + struct dl_bw *dl_b; + bool overflow; + int cpus, ret; + unsigned long flags; + + rcu_read_lock_sched(); + dl_b = dl_bw_of(dest_cpu); + raw_spin_lock_irqsave(&dl_b->lock, flags); + cpus = dl_bw_cpus(dest_cpu); + overflow = __dl_overflow(dl_b, cpus, 0, p->dl.dl_bw); + if (overflow) + ret = -EBUSY; + else { + /* + * We reserve space for this task in the destination + * root_domain, as we can't fail after this point. + * We will free resources in the source root_domain + * later on (see set_cpus_allowed_dl()). + */ + __dl_add(dl_b, p->dl.dl_bw, cpus); + ret = 0; + } + raw_spin_unlock_irqrestore(&dl_b->lock, flags); + rcu_read_unlock_sched(); + return ret; +} + +int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, + const struct cpumask *trial) +{ + int ret = 1, trial_cpus; + struct dl_bw *cur_dl_b; + unsigned long flags; + + rcu_read_lock_sched(); + cur_dl_b = dl_bw_of(cpumask_any(cur)); + trial_cpus = cpumask_weight(trial); + + raw_spin_lock_irqsave(&cur_dl_b->lock, flags); + if (cur_dl_b->bw != -1 && + cur_dl_b->bw * trial_cpus < cur_dl_b->total_bw) + ret = 0; + raw_spin_unlock_irqrestore(&cur_dl_b->lock, flags); + rcu_read_unlock_sched(); + return ret; +} + +bool dl_cpu_busy(unsigned int cpu) +{ + unsigned long flags; + struct dl_bw *dl_b; + bool overflow; + int cpus; + + rcu_read_lock_sched(); + dl_b = dl_bw_of(cpu); + raw_spin_lock_irqsave(&dl_b->lock, flags); + cpus = dl_bw_cpus(cpu); + overflow = __dl_overflow(dl_b, cpus, 0, 0); + raw_spin_unlock_irqrestore(&dl_b->lock, flags); + rcu_read_unlock_sched(); + return overflow; +} +#endif + #ifdef CONFIG_SCHED_DEBUG extern void print_dl_rq(struct seq_file *m, int cpu, struct dl_rq *dl_rq); diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index fa178b62ea79..4fa66de52bd6 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -11,7 +11,8 @@ */ #include <linux/proc_fs.h> -#include <linux/sched.h> +#include <linux/sched/mm.h> +#include <linux/sched/task.h> #include <linux/seq_file.h> #include <linux/kallsyms.h> #include <linux/utsname.h> @@ -551,15 +552,21 @@ void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq) #define P(x) \ SEQ_printf(m, " .%-30s: %Ld\n", #x, (long long)(rt_rq->x)) +#define PU(x) \ + SEQ_printf(m, " .%-30s: %lu\n", #x, (unsigned long)(rt_rq->x)) #define PN(x) \ SEQ_printf(m, " .%-30s: %Ld.%06ld\n", #x, SPLIT_NS(rt_rq->x)) - P(rt_nr_running); + PU(rt_nr_running); +#ifdef CONFIG_SMP + PU(rt_nr_migratory); +#endif P(rt_throttled); PN(rt_time); PN(rt_runtime); #undef PN +#undef PU #undef P } @@ -568,14 +575,21 @@ void print_dl_rq(struct seq_file *m, int cpu, struct dl_rq *dl_rq) struct dl_bw *dl_bw; SEQ_printf(m, "\ndl_rq[%d]:\n", cpu); - SEQ_printf(m, " .%-30s: %ld\n", "dl_nr_running", dl_rq->dl_nr_running); + +#define PU(x) \ + SEQ_printf(m, " .%-30s: %lu\n", #x, (unsigned long)(dl_rq->x)) + + PU(dl_nr_running); #ifdef CONFIG_SMP + PU(dl_nr_migratory); dl_bw = &cpu_rq(cpu)->rd->dl_bw; #else dl_bw = &dl_rq->dl_bw; #endif SEQ_printf(m, " .%-30s: %lld\n", "dl_bw->bw", dl_bw->bw); SEQ_printf(m, " .%-30s: %lld\n", "dl_bw->total_bw", dl_bw->total_bw); + +#undef PU } extern __read_mostly int sched_clock_running; @@ -953,6 +967,10 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) #endif P(policy); P(prio); + if (p->policy == SCHED_DEADLINE) { + P(dl.runtime); + P(dl.deadline); + } #undef PN_SCHEDSTAT #undef PN #undef __PN diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 6559d197e08a..c95880e216f6 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -20,7 +20,9 @@ * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra */ -#include <linux/sched.h> +#include <linux/sched/mm.h> +#include <linux/sched/topology.h> + #include <linux/latencytop.h> #include <linux/cpumask.h> #include <linux/cpuidle.h> @@ -367,8 +369,9 @@ static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) } /* Iterate thr' all leaf cfs_rq's on a runqueue */ -#define for_each_leaf_cfs_rq(rq, cfs_rq) \ - list_for_each_entry_rcu(cfs_rq, &rq->leaf_cfs_rq_list, leaf_cfs_rq_list) +#define for_each_leaf_cfs_rq_safe(rq, cfs_rq, pos) \ + list_for_each_entry_safe(cfs_rq, pos, &rq->leaf_cfs_rq_list, \ + leaf_cfs_rq_list) /* Do the two (enqueued) entities belong to the same group ? */ static inline struct cfs_rq * @@ -461,8 +464,8 @@ static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) { } -#define for_each_leaf_cfs_rq(rq, cfs_rq) \ - for (cfs_rq = &rq->cfs; cfs_rq; cfs_rq = NULL) +#define for_each_leaf_cfs_rq_safe(rq, cfs_rq, pos) \ + for (cfs_rq = &rq->cfs, pos = NULL; cfs_rq; cfs_rq = pos) static inline struct sched_entity *parent_entity(struct sched_entity *se) { @@ -715,18 +718,12 @@ static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se) } #ifdef CONFIG_SMP + +#include "sched-pelt.h" + static int select_idle_sibling(struct task_struct *p, int prev_cpu, int cpu); static unsigned long task_h_load(struct task_struct *p); -/* - * We choose a half-life close to 1 scheduling period. - * Note: The tables runnable_avg_yN_inv and runnable_avg_yN_sum are - * dependent on this value. - */ -#define LOAD_AVG_PERIOD 32 -#define LOAD_AVG_MAX 47742 /* maximum possible load avg */ -#define LOAD_AVG_MAX_N 345 /* number of full periods to produce LOAD_AVG_MAX */ - /* Give new sched_entity start runnable values to heavy its load in infant time */ void init_entity_runnable_average(struct sched_entity *se) { @@ -1385,7 +1382,6 @@ static unsigned long weighted_cpuload(const int cpu); static unsigned long source_load(int cpu, int type); static unsigned long target_load(int cpu, int type); static unsigned long capacity_of(int cpu); -static long effective_load(struct task_group *tg, int cpu, long wl, long wg); /* Cached statistics for all CPUs within a node */ struct numa_stats { @@ -1551,7 +1547,7 @@ static void task_numa_compare(struct task_numa_env *env, */ if (cur) { /* Skip this swap candidate if cannot move to the source cpu */ - if (!cpumask_test_cpu(env->src_cpu, tsk_cpus_allowed(cur))) + if (!cpumask_test_cpu(env->src_cpu, &cur->cpus_allowed)) goto unlock; /* @@ -1661,7 +1657,7 @@ static void task_numa_find_cpu(struct task_numa_env *env, for_each_cpu(cpu, cpumask_of_node(env->dst_nid)) { /* Skip this CPU if the source task cannot migrate */ - if (!cpumask_test_cpu(cpu, tsk_cpus_allowed(env->p))) + if (!cpumask_test_cpu(cpu, &env->p->cpus_allowed)) continue; env->dst_cpu = cpu; @@ -2473,7 +2469,8 @@ void task_numa_work(struct callback_head *work) return; - down_read(&mm->mmap_sem); + if (!down_read_trylock(&mm->mmap_sem)) + return; vma = find_vma(mm, start); if (!vma) { reset_ptenuma_scan(p); @@ -2588,6 +2585,60 @@ void task_tick_numa(struct rq *rq, struct task_struct *curr) } } } + +/* + * Can a task be moved from prev_cpu to this_cpu without causing a load + * imbalance that would trigger the load balancer? + */ +static inline bool numa_wake_affine(struct sched_domain *sd, + struct task_struct *p, int this_cpu, + int prev_cpu, int sync) +{ + struct numa_stats prev_load, this_load; + s64 this_eff_load, prev_eff_load; + + update_numa_stats(&prev_load, cpu_to_node(prev_cpu)); + update_numa_stats(&this_load, cpu_to_node(this_cpu)); + + /* + * If sync wakeup then subtract the (maximum possible) + * effect of the currently running task from the load + * of the current CPU: + */ + if (sync) { + unsigned long current_load = task_h_load(current); + + if (this_load.load > current_load) + this_load.load -= current_load; + else + this_load.load = 0; + } + + /* + * In low-load situations, where this_cpu's node is idle due to the + * sync cause above having dropped this_load.load to 0, move the task. + * Moving to an idle socket will not create a bad imbalance. + * + * Otherwise check if the nodes are near enough in load to allow this + * task to be woken on this_cpu's node. + */ + if (this_load.load > 0) { + unsigned long task_load = task_h_load(p); + + this_eff_load = 100; + this_eff_load *= prev_load.compute_capacity; + + prev_eff_load = 100 + (sd->imbalance_pct - 100) / 2; + prev_eff_load *= this_load.compute_capacity; + + this_eff_load *= this_load.load + task_load; + prev_eff_load *= prev_load.load - task_load; + + return this_eff_load <= prev_eff_load; + } + + return true; +} #else static void task_tick_numa(struct rq *rq, struct task_struct *curr) { @@ -2600,6 +2651,15 @@ static inline void account_numa_enqueue(struct rq *rq, struct task_struct *p) static inline void account_numa_dequeue(struct rq *rq, struct task_struct *p) { } + +#ifdef CONFIG_SMP +static inline bool numa_wake_affine(struct sched_domain *sd, + struct task_struct *p, int this_cpu, + int prev_cpu, int sync) +{ + return true; +} +#endif /* !SMP */ #endif /* CONFIG_NUMA_BALANCING */ static void @@ -2657,6 +2717,18 @@ static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg) if (tg_weight) shares /= tg_weight; + /* + * MIN_SHARES has to be unscaled here to support per-CPU partitioning + * of a group with small tg->shares value. It is a floor value which is + * assigned as a minimum load.weight to the sched_entity representing + * the group on a CPU. + * + * E.g. on 64-bit for a group with tg->shares of scale_load(15)=15*1024 + * on an 8-core system with 8 tasks each runnable on one CPU shares has + * to be 15*1024*1/8=1920 instead of scale_load(MIN_SHARES)=2*1024. In + * case no task is runnable on a CPU MIN_SHARES=2 should be returned + * instead of 0. + */ if (shares < MIN_SHARES) shares = MIN_SHARES; if (shares > tg->shares) @@ -2689,16 +2761,20 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, static inline int throttled_hierarchy(struct cfs_rq *cfs_rq); -static void update_cfs_shares(struct cfs_rq *cfs_rq) +static void update_cfs_shares(struct sched_entity *se) { + struct cfs_rq *cfs_rq = group_cfs_rq(se); struct task_group *tg; - struct sched_entity *se; long shares; - tg = cfs_rq->tg; - se = tg->se[cpu_of(rq_of(cfs_rq))]; - if (!se || throttled_hierarchy(cfs_rq)) + if (!cfs_rq) + return; + + if (throttled_hierarchy(cfs_rq)) return; + + tg = cfs_rq->tg; + #ifndef CONFIG_SMP if (likely(se->load.weight == tg->shares)) return; @@ -2707,54 +2783,23 @@ static void update_cfs_shares(struct cfs_rq *cfs_rq) reweight_entity(cfs_rq_of(se), se, shares); } + #else /* CONFIG_FAIR_GROUP_SCHED */ -static inline void update_cfs_shares(struct cfs_rq *cfs_rq) +static inline void update_cfs_shares(struct sched_entity *se) { } #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_SMP -/* Precomputed fixed inverse multiplies for multiplication by y^n */ -static const u32 runnable_avg_yN_inv[] = { - 0xffffffff, 0xfa83b2da, 0xf5257d14, 0xefe4b99a, 0xeac0c6e6, 0xe5b906e6, - 0xe0ccdeeb, 0xdbfbb796, 0xd744fcc9, 0xd2a81d91, 0xce248c14, 0xc9b9bd85, - 0xc5672a10, 0xc12c4cc9, 0xbd08a39e, 0xb8fbaf46, 0xb504f333, 0xb123f581, - 0xad583ee9, 0xa9a15ab4, 0xa5fed6a9, 0xa2704302, 0x9ef5325f, 0x9b8d39b9, - 0x9837f050, 0x94f4efa8, 0x91c3d373, 0x8ea4398a, 0x8b95c1e3, 0x88980e80, - 0x85aac367, 0x82cd8698, -}; - -/* - * Precomputed \Sum y^k { 1<=k<=n }. These are floor(true_value) to prevent - * over-estimates when re-combining. - */ -static const u32 runnable_avg_yN_sum[] = { - 0, 1002, 1982, 2941, 3880, 4798, 5697, 6576, 7437, 8279, 9103, - 9909,10698,11470,12226,12966,13690,14398,15091,15769,16433,17082, - 17718,18340,18949,19545,20128,20698,21256,21802,22336,22859,23371, -}; - -/* - * Precomputed \Sum y^k { 1<=k<=n, where n%32=0). Values are rolled down to - * lower integers. See Documentation/scheduler/sched-avg.txt how these - * were generated: - */ -static const u32 __accumulated_sum_N32[] = { - 0, 23371, 35056, 40899, 43820, 45281, - 46011, 46376, 46559, 46650, 46696, 46719, -}; - /* * Approximate: * val * y^n, where y^32 ~= 0.5 (~1 scheduling period) */ -static __always_inline u64 decay_load(u64 val, u64 n) +static u64 decay_load(u64 val, u64 n) { unsigned int local_n; - if (!n) - return val; - else if (unlikely(n > LOAD_AVG_PERIOD * 63)) + if (unlikely(n > LOAD_AVG_PERIOD * 63)) return 0; /* after bounds checking we can collapse to 32-bit */ @@ -2776,30 +2821,97 @@ static __always_inline u64 decay_load(u64 val, u64 n) return val; } +static u32 __accumulate_pelt_segments(u64 periods, u32 d1, u32 d3) +{ + u32 c1, c2, c3 = d3; /* y^0 == 1 */ + + /* + * c1 = d1 y^p + */ + c1 = decay_load((u64)d1, periods); + + /* + * p-1 + * c2 = 1024 \Sum y^n + * n=1 + * + * inf inf + * = 1024 ( \Sum y^n - \Sum y^n - y^0 ) + * n=0 n=p + */ + c2 = LOAD_AVG_MAX - decay_load(LOAD_AVG_MAX, periods) - 1024; + + return c1 + c2 + c3; +} + +#define cap_scale(v, s) ((v)*(s) >> SCHED_CAPACITY_SHIFT) + /* - * For updates fully spanning n periods, the contribution to runnable - * average will be: \Sum 1024*y^n + * Accumulate the three separate parts of the sum; d1 the remainder + * of the last (incomplete) period, d2 the span of full periods and d3 + * the remainder of the (incomplete) current period. + * + * d1 d2 d3 + * ^ ^ ^ + * | | | + * |<->|<----------------->|<--->| + * ... |---x---|------| ... |------|-----x (now) + * + * p-1 + * u' = (u + d1) y^p + 1024 \Sum y^n + d3 y^0 + * n=1 * - * We can compute this reasonably efficiently by combining: - * y^PERIOD = 1/2 with precomputed \Sum 1024*y^n {for n <PERIOD} + * = u y^p + (Step 1) + * + * p-1 + * d1 y^p + 1024 \Sum y^n + d3 y^0 (Step 2) + * n=1 */ -static u32 __compute_runnable_contrib(u64 n) +static __always_inline u32 +accumulate_sum(u64 delta, int cpu, struct sched_avg *sa, + unsigned long weight, int running, struct cfs_rq *cfs_rq) { - u32 contrib = 0; + unsigned long scale_freq, scale_cpu; + u32 contrib = (u32)delta; /* p == 0 -> delta < 1024 */ + u64 periods; - if (likely(n <= LOAD_AVG_PERIOD)) - return runnable_avg_yN_sum[n]; - else if (unlikely(n >= LOAD_AVG_MAX_N)) - return LOAD_AVG_MAX; + scale_freq = arch_scale_freq_capacity(NULL, cpu); + scale_cpu = arch_scale_cpu_capacity(NULL, cpu); - /* Since n < LOAD_AVG_MAX_N, n/LOAD_AVG_PERIOD < 11 */ - contrib = __accumulated_sum_N32[n/LOAD_AVG_PERIOD]; - n %= LOAD_AVG_PERIOD; - contrib = decay_load(contrib, n); - return contrib + runnable_avg_yN_sum[n]; -} + delta += sa->period_contrib; + periods = delta / 1024; /* A period is 1024us (~1ms) */ -#define cap_scale(v, s) ((v)*(s) >> SCHED_CAPACITY_SHIFT) + /* + * Step 1: decay old *_sum if we crossed period boundaries. + */ + if (periods) { + sa->load_sum = decay_load(sa->load_sum, periods); + if (cfs_rq) { + cfs_rq->runnable_load_sum = + decay_load(cfs_rq->runnable_load_sum, periods); + } + sa->util_sum = decay_load((u64)(sa->util_sum), periods); + + /* + * Step 2 + */ + delta %= 1024; + contrib = __accumulate_pelt_segments(periods, + 1024 - sa->period_contrib, delta); + } + sa->period_contrib = delta; + + contrib = cap_scale(contrib, scale_freq); + if (weight) { + sa->load_sum += weight * contrib; + if (cfs_rq) + cfs_rq->runnable_load_sum += weight * contrib; + } + if (running) + sa->util_sum += contrib * scale_cpu; + + return periods; +} /* * We can represent the historical contribution to runnable average as the @@ -2830,13 +2942,10 @@ static u32 __compute_runnable_contrib(u64 n) * = u_0 + u_1*y + u_2*y^2 + ... [re-labeling u_i --> u_{i+1}] */ static __always_inline int -__update_load_avg(u64 now, int cpu, struct sched_avg *sa, +___update_load_avg(u64 now, int cpu, struct sched_avg *sa, unsigned long weight, int running, struct cfs_rq *cfs_rq) { - u64 delta, scaled_delta, periods; - u32 contrib; - unsigned int delta_w, scaled_delta_w, decayed = 0; - unsigned long scale_freq, scale_cpu; + u64 delta; delta = now - sa->last_update_time; /* @@ -2855,83 +2964,52 @@ __update_load_avg(u64 now, int cpu, struct sched_avg *sa, delta >>= 10; if (!delta) return 0; - sa->last_update_time = now; - scale_freq = arch_scale_freq_capacity(NULL, cpu); - scale_cpu = arch_scale_cpu_capacity(NULL, cpu); - - /* delta_w is the amount already accumulated against our next period */ - delta_w = sa->period_contrib; - if (delta + delta_w >= 1024) { - decayed = 1; + sa->last_update_time += delta << 10; - /* how much left for next period will start over, we don't know yet */ - sa->period_contrib = 0; - - /* - * Now that we know we're crossing a period boundary, figure - * out how much from delta we need to complete the current - * period and accrue it. - */ - delta_w = 1024 - delta_w; - scaled_delta_w = cap_scale(delta_w, scale_freq); - if (weight) { - sa->load_sum += weight * scaled_delta_w; - if (cfs_rq) { - cfs_rq->runnable_load_sum += - weight * scaled_delta_w; - } - } - if (running) - sa->util_sum += scaled_delta_w * scale_cpu; - - delta -= delta_w; - - /* Figure out how many additional periods this update spans */ - periods = delta / 1024; - delta %= 1024; + /* + * Now we know we crossed measurement unit boundaries. The *_avg + * accrues by two steps: + * + * Step 1: accumulate *_sum since last_update_time. If we haven't + * crossed period boundaries, finish. + */ + if (!accumulate_sum(delta, cpu, sa, weight, running, cfs_rq)) + return 0; - sa->load_sum = decay_load(sa->load_sum, periods + 1); - if (cfs_rq) { - cfs_rq->runnable_load_sum = - decay_load(cfs_rq->runnable_load_sum, periods + 1); - } - sa->util_sum = decay_load((u64)(sa->util_sum), periods + 1); - - /* Efficiently calculate \sum (1..n_period) 1024*y^i */ - contrib = __compute_runnable_contrib(periods); - contrib = cap_scale(contrib, scale_freq); - if (weight) { - sa->load_sum += weight * contrib; - if (cfs_rq) - cfs_rq->runnable_load_sum += weight * contrib; - } - if (running) - sa->util_sum += contrib * scale_cpu; + /* + * Step 2: update *_avg. + */ + sa->load_avg = div_u64(sa->load_sum, LOAD_AVG_MAX - 1024 + sa->period_contrib); + if (cfs_rq) { + cfs_rq->runnable_load_avg = + div_u64(cfs_rq->runnable_load_sum, LOAD_AVG_MAX - 1024 + sa->period_contrib); } + sa->util_avg = sa->util_sum / (LOAD_AVG_MAX - 1024 + sa->period_contrib); - /* Remainder of delta accrued against u_0` */ - scaled_delta = cap_scale(delta, scale_freq); - if (weight) { - sa->load_sum += weight * scaled_delta; - if (cfs_rq) - cfs_rq->runnable_load_sum += weight * scaled_delta; - } - if (running) - sa->util_sum += scaled_delta * scale_cpu; + return 1; +} - sa->period_contrib += delta; +static int +__update_load_avg_blocked_se(u64 now, int cpu, struct sched_entity *se) +{ + return ___update_load_avg(now, cpu, &se->avg, 0, 0, NULL); +} - if (decayed) { - sa->load_avg = div_u64(sa->load_sum, LOAD_AVG_MAX); - if (cfs_rq) { - cfs_rq->runnable_load_avg = - div_u64(cfs_rq->runnable_load_sum, LOAD_AVG_MAX); - } - sa->util_avg = sa->util_sum / LOAD_AVG_MAX; - } +static int +__update_load_avg_se(u64 now, int cpu, struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + return ___update_load_avg(now, cpu, &se->avg, + se->on_rq * scale_load_down(se->load.weight), + cfs_rq->curr == se, NULL); +} - return decayed; +static int +__update_load_avg_cfs_rq(u64 now, int cpu, struct cfs_rq *cfs_rq) +{ + return ___update_load_avg(now, cpu, &cfs_rq->avg, + scale_load_down(cfs_rq->load.weight), + cfs_rq->curr != NULL, cfs_rq); } /* @@ -2968,8 +3046,7 @@ __update_load_avg(u64 now, int cpu, struct sched_avg *sa, * differential update where we store the last value we propagated. This in * turn allows skipping updates if the differential is 'small'. * - * Updating tg's load_avg is necessary before update_cfs_share() (which is - * done) and effective_load() (which is not done because it is too costly). + * Updating tg's load_avg is necessary before update_cfs_share(). */ static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) { @@ -2995,6 +3072,9 @@ static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) void set_task_rq_fair(struct sched_entity *se, struct cfs_rq *prev, struct cfs_rq *next) { + u64 p_last_update_time; + u64 n_last_update_time; + if (!sched_feat(ATTACH_AGE_LOAD)) return; @@ -3005,11 +3085,11 @@ void set_task_rq_fair(struct sched_entity *se, * time. This will result in the wakee task is less decayed, but giving * the wakee more load sounds not bad. */ - if (se->avg.last_update_time && prev) { - u64 p_last_update_time; - u64 n_last_update_time; + if (!(se->avg.last_update_time && prev)) + return; #ifndef CONFIG_64BIT + { u64 p_last_update_time_copy; u64 n_last_update_time_copy; @@ -3024,14 +3104,13 @@ void set_task_rq_fair(struct sched_entity *se, } while (p_last_update_time != p_last_update_time_copy || n_last_update_time != n_last_update_time_copy); + } #else - p_last_update_time = prev->avg.last_update_time; - n_last_update_time = next->avg.last_update_time; + p_last_update_time = prev->avg.last_update_time; + n_last_update_time = next->avg.last_update_time; #endif - __update_load_avg(p_last_update_time, cpu_of(rq_of(prev)), - &se->avg, 0, 0, NULL); - se->avg.last_update_time = n_last_update_time; - } + __update_load_avg_blocked_se(p_last_update_time, cpu_of(rq_of(prev)), se); + se->avg.last_update_time = n_last_update_time; } /* Take into account change of utilization of a child task group */ @@ -3154,6 +3233,36 @@ static inline int propagate_entity_load_avg(struct sched_entity *se) return 1; } +/* + * Check if we need to update the load and the utilization of a blocked + * group_entity: + */ +static inline bool skip_blocked_update(struct sched_entity *se) +{ + struct cfs_rq *gcfs_rq = group_cfs_rq(se); + + /* + * If sched_entity still have not zero load or utilization, we have to + * decay it: + */ + if (se->avg.load_avg || se->avg.util_avg) + return false; + + /* + * If there is a pending propagation, we have to update the load and + * the utilization of the sched_entity: + */ + if (gcfs_rq->propagate_avg) + return false; + + /* + * Otherwise, the load and the utilization of the sched_entity is + * already zero and there is no pending propagation, so it will be a + * waste of time to try to decay it: + */ + return true; +} + #else /* CONFIG_FAIR_GROUP_SCHED */ static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) {} @@ -3246,8 +3355,7 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq) set_tg_cfs_propagate(cfs_rq); } - decayed = __update_load_avg(now, cpu_of(rq_of(cfs_rq)), sa, - scale_load_down(cfs_rq->load.weight), cfs_rq->curr != NULL, cfs_rq); + decayed = __update_load_avg_cfs_rq(now, cpu_of(rq_of(cfs_rq)), cfs_rq); #ifndef CONFIG_64BIT smp_wmb(); @@ -3279,11 +3387,8 @@ static inline void update_load_avg(struct sched_entity *se, int flags) * Track task load average for carrying it to new CPU after migrated, and * track group sched_entity load average for task_h_load calc in migration */ - if (se->avg.last_update_time && !(flags & SKIP_AGE_LOAD)) { - __update_load_avg(now, cpu, &se->avg, - se->on_rq * scale_load_down(se->load.weight), - cfs_rq->curr == se, NULL); - } + if (se->avg.last_update_time && !(flags & SKIP_AGE_LOAD)) + __update_load_avg_se(now, cpu, cfs_rq, se); decayed = update_cfs_rq_load_avg(now, cfs_rq, true); decayed |= propagate_entity_load_avg(se); @@ -3388,7 +3493,7 @@ void sync_entity_load_avg(struct sched_entity *se) u64 last_update_time; last_update_time = cfs_rq_last_update_time(cfs_rq); - __update_load_avg(last_update_time, cpu_of(rq_of(cfs_rq)), &se->avg, 0, 0, NULL); + __update_load_avg_blocked_se(last_update_time, cpu_of(rq_of(cfs_rq)), se); } /* @@ -3424,7 +3529,7 @@ static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq) return cfs_rq->avg.load_avg; } -static int idle_balance(struct rq *this_rq); +static int idle_balance(struct rq *this_rq, struct rq_flags *rf); #else /* CONFIG_SMP */ @@ -3453,7 +3558,7 @@ attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {} static inline void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {} -static inline int idle_balance(struct rq *rq) +static inline int idle_balance(struct rq *rq, struct rq_flags *rf) { return 0; } @@ -3521,7 +3626,7 @@ static inline void check_schedstat_required(void) trace_sched_stat_runtime_enabled()) { printk_deferred_once("Scheduler tracepoints stat_sleep, stat_iowait, " "stat_blocked and stat_runtime require the " - "kernel parameter schedstats=enabled or " + "kernel parameter schedstats=enable or " "kernel.sched_schedstats=1\n"); } #endif @@ -3582,10 +3687,18 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (renorm && !curr) se->vruntime += cfs_rq->min_vruntime; + /* + * When enqueuing a sched_entity, we must: + * - Update loads to have both entity and cfs_rq synced with now. + * - Add its load to cfs_rq->runnable_avg + * - For group_entity, update its weight to reflect the new share of + * its group cfs_rq + * - Add its new weight to cfs_rq->load.weight + */ update_load_avg(se, UPDATE_TG); enqueue_entity_load_avg(cfs_rq, se); + update_cfs_shares(se); account_entity_enqueue(cfs_rq, se); - update_cfs_shares(cfs_rq); if (flags & ENQUEUE_WAKEUP) place_entity(cfs_rq, se, 0); @@ -3657,6 +3770,15 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * Update run-time statistics of the 'current'. */ update_curr(cfs_rq); + + /* + * When dequeuing a sched_entity, we must: + * - Update loads to have both entity and cfs_rq synced with now. + * - Substract its load from the cfs_rq->runnable_avg. + * - Substract its previous weight from cfs_rq->load.weight. + * - For group entity, update its weight to reflect the new share + * of its group cfs_rq. + */ update_load_avg(se, UPDATE_TG); dequeue_entity_load_avg(cfs_rq, se); @@ -3681,7 +3803,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) /* return excess runtime on last dequeue */ return_cfs_rq_runtime(cfs_rq); - update_cfs_shares(cfs_rq); + update_cfs_shares(se); /* * Now advance min_vruntime if @se was the entity holding it back, @@ -3864,7 +3986,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) * Ensure that runnable average is periodically updated. */ update_load_avg(curr, UPDATE_TG); - update_cfs_shares(cfs_rq); + update_cfs_shares(curr); #ifdef CONFIG_SCHED_HRTICK /* @@ -4235,8 +4357,9 @@ static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b, list_for_each_entry_rcu(cfs_rq, &cfs_b->throttled_cfs_rq, throttled_list) { struct rq *rq = rq_of(cfs_rq); + struct rq_flags rf; - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); if (!cfs_rq_throttled(cfs_rq)) goto next; @@ -4253,7 +4376,7 @@ static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b, unthrottle_cfs_rq(cfs_rq); next: - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); if (!remaining) break; @@ -4582,24 +4705,43 @@ static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) hrtimer_cancel(&cfs_b->slack_timer); } +/* + * Both these cpu hotplug callbacks race against unregister_fair_sched_group() + * + * The race is harmless, since modifying bandwidth settings of unhooked group + * bits doesn't do much. + */ + +/* cpu online calback */ static void __maybe_unused update_runtime_enabled(struct rq *rq) { - struct cfs_rq *cfs_rq; + struct task_group *tg; - for_each_leaf_cfs_rq(rq, cfs_rq) { - struct cfs_bandwidth *cfs_b = &cfs_rq->tg->cfs_bandwidth; + lockdep_assert_held(&rq->lock); + + rcu_read_lock(); + list_for_each_entry_rcu(tg, &task_groups, list) { + struct cfs_bandwidth *cfs_b = &tg->cfs_bandwidth; + struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; raw_spin_lock(&cfs_b->lock); cfs_rq->runtime_enabled = cfs_b->quota != RUNTIME_INF; raw_spin_unlock(&cfs_b->lock); } + rcu_read_unlock(); } +/* cpu offline callback */ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) { - struct cfs_rq *cfs_rq; + struct task_group *tg; + + lockdep_assert_held(&rq->lock); + + rcu_read_lock(); + list_for_each_entry_rcu(tg, &task_groups, list) { + struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; - for_each_leaf_cfs_rq(rq, cfs_rq) { if (!cfs_rq->runtime_enabled) continue; @@ -4617,6 +4759,7 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) if (cfs_rq_throttled(cfs_rq)) unthrottle_cfs_rq(cfs_rq); } + rcu_read_unlock(); } #else /* CONFIG_CFS_BANDWIDTH */ @@ -4761,7 +4904,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) break; update_load_avg(se, UPDATE_TG); - update_cfs_shares(cfs_rq); + update_cfs_shares(se); } if (!se) @@ -4820,7 +4963,7 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) break; update_load_avg(se, UPDATE_TG); - update_cfs_shares(cfs_rq); + update_cfs_shares(se); } if (!se) @@ -5061,15 +5204,16 @@ void cpu_load_update_nohz_stop(void) unsigned long curr_jiffies = READ_ONCE(jiffies); struct rq *this_rq = this_rq(); unsigned long load; + struct rq_flags rf; if (curr_jiffies == this_rq->last_load_update_tick) return; load = weighted_cpuload(cpu_of(this_rq)); - raw_spin_lock(&this_rq->lock); + rq_lock(this_rq, &rf); update_rq_clock(this_rq); cpu_load_update_nohz(this_rq, curr_jiffies, load); - raw_spin_unlock(&this_rq->lock); + rq_unlock(this_rq, &rf); } #else /* !CONFIG_NO_HZ_COMMON */ static inline void cpu_load_update_nohz(struct rq *this_rq, @@ -5154,126 +5298,6 @@ static unsigned long cpu_avg_load_per_task(int cpu) return 0; } -#ifdef CONFIG_FAIR_GROUP_SCHED -/* - * effective_load() calculates the load change as seen from the root_task_group - * - * Adding load to a group doesn't make a group heavier, but can cause movement - * of group shares between cpus. Assuming the shares were perfectly aligned one - * can calculate the shift in shares. - * - * Calculate the effective load difference if @wl is added (subtracted) to @tg - * on this @cpu and results in a total addition (subtraction) of @wg to the - * total group weight. - * - * Given a runqueue weight distribution (rw_i) we can compute a shares - * distribution (s_i) using: - * - * s_i = rw_i / \Sum rw_j (1) - * - * Suppose we have 4 CPUs and our @tg is a direct child of the root group and - * has 7 equal weight tasks, distributed as below (rw_i), with the resulting - * shares distribution (s_i): - * - * rw_i = { 2, 4, 1, 0 } - * s_i = { 2/7, 4/7, 1/7, 0 } - * - * As per wake_affine() we're interested in the load of two CPUs (the CPU the - * task used to run on and the CPU the waker is running on), we need to - * compute the effect of waking a task on either CPU and, in case of a sync - * wakeup, compute the effect of the current task going to sleep. - * - * So for a change of @wl to the local @cpu with an overall group weight change - * of @wl we can compute the new shares distribution (s'_i) using: - * - * s'_i = (rw_i + @wl) / (@wg + \Sum rw_j) (2) - * - * Suppose we're interested in CPUs 0 and 1, and want to compute the load - * differences in waking a task to CPU 0. The additional task changes the - * weight and shares distributions like: - * - * rw'_i = { 3, 4, 1, 0 } - * s'_i = { 3/8, 4/8, 1/8, 0 } - * - * We can then compute the difference in effective weight by using: - * - * dw_i = S * (s'_i - s_i) (3) - * - * Where 'S' is the group weight as seen by its parent. - * - * Therefore the effective change in loads on CPU 0 would be 5/56 (3/8 - 2/7) - * times the weight of the group. The effect on CPU 1 would be -4/56 (4/8 - - * 4/7) times the weight of the group. - */ -static long effective_load(struct task_group *tg, int cpu, long wl, long wg) -{ - struct sched_entity *se = tg->se[cpu]; - - if (!tg->parent) /* the trivial, non-cgroup case */ - return wl; - - for_each_sched_entity(se) { - struct cfs_rq *cfs_rq = se->my_q; - long W, w = cfs_rq_load_avg(cfs_rq); - - tg = cfs_rq->tg; - - /* - * W = @wg + \Sum rw_j - */ - W = wg + atomic_long_read(&tg->load_avg); - - /* Ensure \Sum rw_j >= rw_i */ - W -= cfs_rq->tg_load_avg_contrib; - W += w; - - /* - * w = rw_i + @wl - */ - w += wl; - - /* - * wl = S * s'_i; see (2) - */ - if (W > 0 && w < W) - wl = (w * (long)scale_load_down(tg->shares)) / W; - else - wl = scale_load_down(tg->shares); - - /* - * Per the above, wl is the new se->load.weight value; since - * those are clipped to [MIN_SHARES, ...) do so now. See - * calc_cfs_shares(). - */ - if (wl < MIN_SHARES) - wl = MIN_SHARES; - - /* - * wl = dw_i = S * (s'_i - s_i); see (3) - */ - wl -= se->avg.load_avg; - - /* - * Recursively apply this logic to all parent groups to compute - * the final effective load change on the root group. Since - * only the @tg group gets extra weight, all parent groups can - * only redistribute existing shares. @wl is the shift in shares - * resulting from this level per the above. - */ - wg = 0; - } - - return wl; -} -#else - -static long effective_load(struct task_group *tg, int cpu, long wl, long wg) -{ - return wl; -} - -#endif - static void record_wakee(struct task_struct *p) { /* @@ -5324,67 +5348,25 @@ static int wake_wide(struct task_struct *p) static int wake_affine(struct sched_domain *sd, struct task_struct *p, int prev_cpu, int sync) { - s64 this_load, load; - s64 this_eff_load, prev_eff_load; - int idx, this_cpu; - struct task_group *tg; - unsigned long weight; - int balanced; - - idx = sd->wake_idx; - this_cpu = smp_processor_id(); - load = source_load(prev_cpu, idx); - this_load = target_load(this_cpu, idx); + int this_cpu = smp_processor_id(); + bool affine = false; /* - * If sync wakeup then subtract the (maximum possible) - * effect of the currently running task from the load - * of the current CPU: + * Common case: CPUs are in the same socket, and select_idle_sibling() + * will do its thing regardless of what we return: */ - if (sync) { - tg = task_group(current); - weight = current->se.avg.load_avg; - - this_load += effective_load(tg, this_cpu, -weight, -weight); - load += effective_load(tg, prev_cpu, 0, -weight); - } - - tg = task_group(p); - weight = p->se.avg.load_avg; - - /* - * In low-load situations, where prev_cpu is idle and this_cpu is idle - * due to the sync cause above having dropped this_load to 0, we'll - * always have an imbalance, but there's really nothing you can do - * about that, so that's good too. - * - * Otherwise check if either cpus are near enough in load to allow this - * task to be woken on this_cpu. - */ - this_eff_load = 100; - this_eff_load *= capacity_of(prev_cpu); - - prev_eff_load = 100 + (sd->imbalance_pct - 100) / 2; - prev_eff_load *= capacity_of(this_cpu); - - if (this_load > 0) { - this_eff_load *= this_load + - effective_load(tg, this_cpu, weight, weight); - - prev_eff_load *= load + effective_load(tg, prev_cpu, 0, weight); - } - - balanced = this_eff_load <= prev_eff_load; + if (cpus_share_cache(prev_cpu, this_cpu)) + affine = true; + else + affine = numa_wake_affine(sd, p, this_cpu, prev_cpu, sync); schedstat_inc(p->se.statistics.nr_wakeups_affine_attempts); + if (affine) { + schedstat_inc(sd->ttwu_move_affine); + schedstat_inc(p->se.statistics.nr_wakeups_affine); + } - if (!balanced) - return 0; - - schedstat_inc(sd->ttwu_move_affine); - schedstat_inc(p->se.statistics.nr_wakeups_affine); - - return 1; + return affine; } static inline int task_util(struct task_struct *p); @@ -5423,12 +5405,12 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int i; /* Skip over this group if it has no CPUs allowed */ - if (!cpumask_intersects(sched_group_cpus(group), - tsk_cpus_allowed(p))) + if (!cpumask_intersects(sched_group_span(group), + &p->cpus_allowed)) continue; local_group = cpumask_test_cpu(this_cpu, - sched_group_cpus(group)); + sched_group_span(group)); /* * Tally up the load of all CPUs in the group and find @@ -5438,7 +5420,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, runnable_load = 0; max_spare_cap = 0; - for_each_cpu(i, sched_group_cpus(group)) { + for_each_cpu(i, sched_group_span(group)) { /* Bias balancing toward cpus of our domain */ if (local_group) load = source_load(i, load_idx); @@ -5541,10 +5523,10 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) /* Check if we have any choice: */ if (group->group_weight == 1) - return cpumask_first(sched_group_cpus(group)); + return cpumask_first(sched_group_span(group)); /* Traverse only the allowed CPUs */ - for_each_cpu_and(i, sched_group_cpus(group), tsk_cpus_allowed(p)) { + for_each_cpu_and(i, sched_group_span(group), &p->cpus_allowed) { if (idle_cpu(i)) { struct rq *rq = cpu_rq(i); struct cpuidle_state *idle = idle_get_state(rq); @@ -5579,43 +5561,6 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) return shallowest_idle_cpu != -1 ? shallowest_idle_cpu : least_loaded_cpu; } -/* - * Implement a for_each_cpu() variant that starts the scan at a given cpu - * (@start), and wraps around. - * - * This is used to scan for idle CPUs; such that not all CPUs looking for an - * idle CPU find the same CPU. The down-side is that tasks tend to cycle - * through the LLC domain. - * - * Especially tbench is found sensitive to this. - */ - -static int cpumask_next_wrap(int n, const struct cpumask *mask, int start, int *wrapped) -{ - int next; - -again: - next = find_next_bit(cpumask_bits(mask), nr_cpumask_bits, n+1); - - if (*wrapped) { - if (next >= start) - return nr_cpumask_bits; - } else { - if (next >= nr_cpumask_bits) { - *wrapped = 1; - n = -1; - goto again; - } - } - - return next; -} - -#define for_each_cpu_wrap(cpu, mask, start, wrap) \ - for ((wrap) = 0, (cpu) = (start)-1; \ - (cpu) = cpumask_next_wrap((cpu), (mask), (start), &(wrap)), \ - (cpu) < nr_cpumask_bits; ) - #ifdef CONFIG_SCHED_SMT static inline void set_idle_cores(int cpu, int val) @@ -5675,7 +5620,7 @@ unlock: static int select_idle_core(struct task_struct *p, struct sched_domain *sd, int target) { struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask); - int core, cpu, wrap; + int core, cpu; if (!static_branch_likely(&sched_smt_present)) return -1; @@ -5683,9 +5628,9 @@ static int select_idle_core(struct task_struct *p, struct sched_domain *sd, int if (!test_idle_cores(target, false)) return -1; - cpumask_and(cpus, sched_domain_span(sd), tsk_cpus_allowed(p)); + cpumask_and(cpus, sched_domain_span(sd), &p->cpus_allowed); - for_each_cpu_wrap(core, cpus, target, wrap) { + for_each_cpu_wrap(core, cpus, target) { bool idle = true; for_each_cpu(cpu, cpu_smt_mask(core)) { @@ -5717,7 +5662,7 @@ static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int t return -1; for_each_cpu(cpu, cpu_smt_mask(target)) { - if (!cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) + if (!cpumask_test_cpu(cpu, &p->cpus_allowed)) continue; if (idle_cpu(cpu)) return cpu; @@ -5748,28 +5693,39 @@ static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target) { struct sched_domain *this_sd; - u64 avg_cost, avg_idle = this_rq()->avg_idle; + u64 avg_cost, avg_idle; u64 time, cost; s64 delta; - int cpu, wrap; + int cpu, nr = INT_MAX; this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc)); if (!this_sd) return -1; - avg_cost = this_sd->avg_scan_cost; - /* * Due to large variance we need a large fuzz factor; hackbench in * particularly is sensitive here. */ - if ((avg_idle / 512) < avg_cost) + avg_idle = this_rq()->avg_idle / 512; + avg_cost = this_sd->avg_scan_cost + 1; + + if (sched_feat(SIS_AVG_CPU) && avg_idle < avg_cost) return -1; + if (sched_feat(SIS_PROP)) { + u64 span_avg = sd->span_weight * avg_idle; + if (span_avg > 4*avg_cost) + nr = div_u64(span_avg, avg_cost); + else + nr = 4; + } + time = local_clock(); - for_each_cpu_wrap(cpu, sched_domain_span(sd), target, wrap) { - if (!cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) + for_each_cpu_wrap(cpu, sched_domain_span(sd), target) { + if (!--nr) + return -1; + if (!cpumask_test_cpu(cpu, &p->cpus_allowed)) continue; if (idle_cpu(cpu)) break; @@ -5924,7 +5880,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f if (sd_flag & SD_BALANCE_WAKE) { record_wakee(p); want_affine = !wake_wide(p) && !wake_cap(p, cpu, prev_cpu) - && cpumask_test_cpu(cpu, tsk_cpus_allowed(p)); + && cpumask_test_cpu(cpu, &p->cpus_allowed); } rcu_read_lock(); @@ -5950,11 +5906,15 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f if (affine_sd) { sd = NULL; /* Prefer wake_affine over balance flags */ - if (cpu != prev_cpu && wake_affine(affine_sd, p, prev_cpu, sync)) + if (cpu == prev_cpu) + goto pick_cpu; + + if (wake_affine(affine_sd, p, prev_cpu, sync)) new_cpu = cpu; } if (!sd) { + pick_cpu: if (sd_flag & SD_BALANCE_WAKE) /* XXX always ? */ new_cpu = select_idle_sibling(p, prev_cpu, new_cpu); @@ -6107,8 +6067,11 @@ static void set_last_buddy(struct sched_entity *se) if (entity_is_task(se) && unlikely(task_of(se)->policy == SCHED_IDLE)) return; - for_each_sched_entity(se) + for_each_sched_entity(se) { + if (SCHED_WARN_ON(!se->on_rq)) + return; cfs_rq_of(se)->last = se; + } } static void set_next_buddy(struct sched_entity *se) @@ -6116,8 +6079,11 @@ static void set_next_buddy(struct sched_entity *se) if (entity_is_task(se) && unlikely(task_of(se)->policy == SCHED_IDLE)) return; - for_each_sched_entity(se) + for_each_sched_entity(se) { + if (SCHED_WARN_ON(!se->on_rq)) + return; cfs_rq_of(se)->next = se; + } } static void set_skip_buddy(struct sched_entity *se) @@ -6213,7 +6179,7 @@ preempt: } static struct task_struct * -pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie) +pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { struct cfs_rq *cfs_rq = &rq->cfs; struct sched_entity *se; @@ -6320,15 +6286,8 @@ simple: return p; idle: - /* - * This is OK, because current is on_cpu, which avoids it being picked - * for load-balance and preemption/IRQs are still disabled avoiding - * further scheduler activity on it and we're being very careful to - * re-start the picking loop. - */ - lockdep_unpin_lock(&rq->lock, cookie); - new_tasks = idle_balance(rq); - lockdep_repin_lock(&rq->lock, cookie); + new_tasks = idle_balance(rq, rf); + /* * Because idle_balance() releases (and re-acquires) rq->lock, it is * possible for any higher priority task to appear. In that case we @@ -6632,6 +6591,10 @@ static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env) if (dst_nid == p->numa_preferred_nid) return 0; + /* Leaving a core idle is often worse than degrading locality. */ + if (env->idle != CPU_NOT_IDLE) + return -1; + if (numa_group) { src_faults = group_faults(p, src_nid); dst_faults = group_faults(p, dst_nid); @@ -6671,7 +6634,7 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) if (throttled_lb_pair(task_group(p), env->src_cpu, env->dst_cpu)) return 0; - if (!cpumask_test_cpu(env->dst_cpu, tsk_cpus_allowed(p))) { + if (!cpumask_test_cpu(env->dst_cpu, &p->cpus_allowed)) { int cpu; schedstat_inc(p->se.statistics.nr_failed_migrations_affine); @@ -6683,15 +6646,15 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) * our sched_group. We may want to revisit it if we couldn't * meet load balance goals by pulling other tasks on src_cpu. * - * Also avoid computing new_dst_cpu if we have already computed - * one in current iteration. + * Avoid computing new_dst_cpu for NEWLY_IDLE or if we have + * already computed one in current iteration. */ - if (!env->dst_grpmask || (env->flags & LBF_DST_PINNED)) + if (env->idle == CPU_NEWLY_IDLE || (env->flags & LBF_DST_PINNED)) return 0; /* Prevent to re-select dst_cpu via env's cpus */ for_each_cpu_and(cpu, env->dst_grpmask, env->cpus) { - if (cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) { + if (cpumask_test_cpu(cpu, &p->cpus_allowed)) { env->flags |= LBF_DST_PINNED; env->new_dst_cpu = cpu; break; @@ -6740,7 +6703,7 @@ static void detach_task(struct task_struct *p, struct lb_env *env) lockdep_assert_held(&env->src_rq->lock); p->on_rq = TASK_ON_RQ_MIGRATING; - deactivate_task(env->src_rq, p, 0); + deactivate_task(env->src_rq, p, DEQUEUE_NOCLOCK); set_task_cpu(p, env->dst_cpu); } @@ -6873,7 +6836,7 @@ static void attach_task(struct rq *rq, struct task_struct *p) lockdep_assert_held(&rq->lock); BUG_ON(task_rq(p) != rq); - activate_task(rq, p, 0); + activate_task(rq, p, ENQUEUE_NOCLOCK); p->on_rq = TASK_ON_RQ_QUEUED; check_preempt_curr(rq, p, 0); } @@ -6884,9 +6847,12 @@ static void attach_task(struct rq *rq, struct task_struct *p) */ static void attach_one_task(struct rq *rq, struct task_struct *p) { - raw_spin_lock(&rq->lock); + struct rq_flags rf; + + rq_lock(rq, &rf); + update_rq_clock(rq); attach_task(rq, p); - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); } /* @@ -6897,8 +6863,10 @@ static void attach_tasks(struct lb_env *env) { struct list_head *tasks = &env->tasks; struct task_struct *p; + struct rq_flags rf; - raw_spin_lock(&env->dst_rq->lock); + rq_lock(env->dst_rq, &rf); + update_rq_clock(env->dst_rq); while (!list_empty(tasks)) { p = list_first_entry(tasks, struct task_struct, se.group_node); @@ -6907,24 +6875,44 @@ static void attach_tasks(struct lb_env *env) attach_task(env->dst_rq, p); } - raw_spin_unlock(&env->dst_rq->lock); + rq_unlock(env->dst_rq, &rf); } #ifdef CONFIG_FAIR_GROUP_SCHED + +static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) +{ + if (cfs_rq->load.weight) + return false; + + if (cfs_rq->avg.load_sum) + return false; + + if (cfs_rq->avg.util_sum) + return false; + + if (cfs_rq->runnable_load_sum) + return false; + + return true; +} + static void update_blocked_averages(int cpu) { struct rq *rq = cpu_rq(cpu); - struct cfs_rq *cfs_rq; - unsigned long flags; + struct cfs_rq *cfs_rq, *pos; + struct rq_flags rf; - raw_spin_lock_irqsave(&rq->lock, flags); + rq_lock_irqsave(rq, &rf); update_rq_clock(rq); /* * Iterates the task_group tree in a bottom up fashion, see * list_add_leaf_cfs_rq() for details. */ - for_each_leaf_cfs_rq(rq, cfs_rq) { + for_each_leaf_cfs_rq_safe(rq, cfs_rq, pos) { + struct sched_entity *se; + /* throttled entities do not contribute to load */ if (throttled_hierarchy(cfs_rq)) continue; @@ -6932,11 +6920,19 @@ static void update_blocked_averages(int cpu) if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq, true)) update_tg_load_avg(cfs_rq, 0); - /* Propagate pending load changes to the parent */ - if (cfs_rq->tg->se[cpu]) - update_load_avg(cfs_rq->tg->se[cpu], 0); + /* Propagate pending load changes to the parent, if any: */ + se = cfs_rq->tg->se[cpu]; + if (se && !skip_blocked_update(se)) + update_load_avg(se, 0); + + /* + * There can be a lot of idle CPU cgroups. Don't let fully + * decayed cfs_rqs linger on the list. + */ + if (cfs_rq_is_decayed(cfs_rq)) + list_del_leaf_cfs_rq(cfs_rq); } - raw_spin_unlock_irqrestore(&rq->lock, flags); + rq_unlock_irqrestore(rq, &rf); } /* @@ -6990,12 +6986,12 @@ static inline void update_blocked_averages(int cpu) { struct rq *rq = cpu_rq(cpu); struct cfs_rq *cfs_rq = &rq->cfs; - unsigned long flags; + struct rq_flags rf; - raw_spin_lock_irqsave(&rq->lock, flags); + rq_lock_irqsave(rq, &rf); update_rq_clock(rq); update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq, true); - raw_spin_unlock_irqrestore(&rq->lock, flags); + rq_unlock_irqrestore(rq, &rf); } static unsigned long task_h_load(struct task_struct *p) @@ -7167,7 +7163,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) * span the current group. */ - for_each_cpu(cpu, sched_group_cpus(sdg)) { + for_each_cpu(cpu, sched_group_span(sdg)) { struct sched_group_capacity *sgc; struct rq *rq = cpu_rq(cpu); @@ -7225,7 +7221,7 @@ check_cpu_capacity(struct rq *rq, struct sched_domain *sd) /* * Group imbalance indicates (and tries to solve) the problem where balancing - * groups is inadequate due to tsk_cpus_allowed() constraints. + * groups is inadequate due to ->cpus_allowed constraints. * * Imagine a situation of two groups of 4 cpus each and 4 tasks each with a * cpumask covering 1 cpu of the first group and 3 cpus of the second group. @@ -7346,7 +7342,7 @@ static inline void update_sg_lb_stats(struct lb_env *env, memset(sgs, 0, sizeof(*sgs)); - for_each_cpu_and(i, sched_group_cpus(group), env->cpus) { + for_each_cpu_and(i, sched_group_span(group), env->cpus) { struct rq *rq = cpu_rq(i); /* Bias balancing toward cpus of our domain */ @@ -7496,6 +7492,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd { struct sched_domain *child = env->sd->child; struct sched_group *sg = env->sd->groups; + struct sg_lb_stats *local = &sds->local_stat; struct sg_lb_stats tmp_sgs; int load_idx, prefer_sibling = 0; bool overload = false; @@ -7509,10 +7506,10 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd struct sg_lb_stats *sgs = &tmp_sgs; int local_group; - local_group = cpumask_test_cpu(env->dst_cpu, sched_group_cpus(sg)); + local_group = cpumask_test_cpu(env->dst_cpu, sched_group_span(sg)); if (local_group) { sds->local = sg; - sgs = &sds->local_stat; + sgs = local; if (env->idle != CPU_NEWLY_IDLE || time_after_eq(jiffies, sg->sgc->next_update)) @@ -7536,8 +7533,8 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd * the tasks on the system). */ if (prefer_sibling && sds->local && - group_has_capacity(env, &sds->local_stat) && - (sgs->sum_nr_running > 1)) { + group_has_capacity(env, local) && + (sgs->sum_nr_running > local->sum_nr_running + 1)) { sgs->group_no_capacity = 1; sgs->group_type = group_classify(sg, sgs); } @@ -7568,7 +7565,7 @@ next_group: /** * check_asym_packing - Check to see if the group is packed into the - * sched doman. + * sched domain. * * This is primarily intended to used at the sibling level. Some * cores like POWER7 prefer to use lower numbered SMT threads. In the @@ -7864,7 +7861,7 @@ static struct rq *find_busiest_queue(struct lb_env *env, unsigned long busiest_load = 0, busiest_capacity = 1; int i; - for_each_cpu_and(i, sched_group_cpus(group), env->cpus) { + for_each_cpu_and(i, sched_group_span(group), env->cpus) { unsigned long capacity, wl; enum fbq_type rt; @@ -7970,7 +7967,6 @@ static int active_load_balance_cpu_stop(void *data); static int should_we_balance(struct lb_env *env) { struct sched_group *sg = env->sd->groups; - struct cpumask *sg_cpus, *sg_mask; int cpu, balance_cpu = -1; /* @@ -7980,11 +7976,9 @@ static int should_we_balance(struct lb_env *env) if (env->idle == CPU_NEWLY_IDLE) return 1; - sg_cpus = sched_group_cpus(sg); - sg_mask = sched_group_mask(sg); /* Try to find first idle cpu */ - for_each_cpu_and(cpu, sg_cpus, env->cpus) { - if (!cpumask_test_cpu(cpu, sg_mask) || !idle_cpu(cpu)) + for_each_cpu_and(cpu, group_balance_mask(sg), env->cpus) { + if (!idle_cpu(cpu)) continue; balance_cpu = cpu; @@ -8013,14 +8007,14 @@ static int load_balance(int this_cpu, struct rq *this_rq, struct sched_domain *sd_parent = sd->parent; struct sched_group *group; struct rq *busiest; - unsigned long flags; + struct rq_flags rf; struct cpumask *cpus = this_cpu_cpumask_var_ptr(load_balance_mask); struct lb_env env = { .sd = sd, .dst_cpu = this_cpu, .dst_rq = this_rq, - .dst_grpmask = sched_group_cpus(sd->groups), + .dst_grpmask = sched_group_span(sd->groups), .idle = idle, .loop_break = sched_nr_migrate_break, .cpus = cpus, @@ -8028,14 +8022,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, .tasks = LIST_HEAD_INIT(env.tasks), }; - /* - * For NEWLY_IDLE load_balancing, we don't need to consider - * other cpus in our group - */ - if (idle == CPU_NEWLY_IDLE) - env.dst_grpmask = NULL; - - cpumask_copy(cpus, cpu_active_mask); + cpumask_and(cpus, sched_domain_span(sd), cpu_active_mask); schedstat_inc(sd->lb_count[idle]); @@ -8076,7 +8063,8 @@ redo: env.loop_max = min(sysctl_sched_nr_migrate, busiest->nr_running); more_balance: - raw_spin_lock_irqsave(&busiest->lock, flags); + rq_lock_irqsave(busiest, &rf); + update_rq_clock(busiest); /* * cur_ld_moved - load moved in current iteration @@ -8092,14 +8080,14 @@ more_balance: * See task_rq_lock() family for the details. */ - raw_spin_unlock(&busiest->lock); + rq_unlock(busiest, &rf); if (cur_ld_moved) { attach_tasks(&env); ld_moved += cur_ld_moved; } - local_irq_restore(flags); + local_irq_restore(rf.flags); if (env.flags & LBF_NEED_BREAK) { env.flags &= ~LBF_NEED_BREAK; @@ -8156,7 +8144,15 @@ more_balance: /* All tasks on this runqueue were pinned by CPU affinity */ if (unlikely(env.flags & LBF_ALL_PINNED)) { cpumask_clear_cpu(cpu_of(busiest), cpus); - if (!cpumask_empty(cpus)) { + /* + * Attempting to continue load balancing at the current + * sched_domain level only makes sense if there are + * active CPUs remaining as possible busiest CPUs to + * pull load from which are not contained within the + * destination group that is receiving any migrated + * load. + */ + if (!cpumask_subset(cpus, env.dst_grpmask)) { env.loop = 0; env.loop_break = sched_nr_migrate_break; goto redo; @@ -8177,14 +8173,15 @@ more_balance: sd->nr_balance_failed++; if (need_active_balance(&env)) { + unsigned long flags; + raw_spin_lock_irqsave(&busiest->lock, flags); /* don't kick the active_load_balance_cpu_stop, * if the curr task on busiest cpu can't be * moved to this_cpu */ - if (!cpumask_test_cpu(this_cpu, - tsk_cpus_allowed(busiest->curr))) { + if (!cpumask_test_cpu(this_cpu, &busiest->curr->cpus_allowed)) { raw_spin_unlock_irqrestore(&busiest->lock, flags); env.flags |= LBF_ALL_PINNED; @@ -8297,7 +8294,7 @@ update_next_balance(struct sched_domain *sd, unsigned long *next_balance) * idle_balance is called by schedule() if this_cpu is about to become * idle. Attempts to pull tasks from other CPUs. */ -static int idle_balance(struct rq *this_rq) +static int idle_balance(struct rq *this_rq, struct rq_flags *rf) { unsigned long next_balance = jiffies + HZ; int this_cpu = this_rq->cpu; @@ -8311,6 +8308,14 @@ static int idle_balance(struct rq *this_rq) */ this_rq->idle_stamp = rq_clock(this_rq); + /* + * This is OK, because current is on_cpu, which avoids it being picked + * for load-balance and preemption/IRQs are still disabled avoiding + * further scheduler activity on it and we're being very careful to + * re-start the picking loop. + */ + rq_unpin_lock(this_rq, rf); + if (this_rq->avg_idle < sysctl_sched_migration_cost || !this_rq->rd->overload) { rcu_read_lock(); @@ -8388,6 +8393,8 @@ out: if (pulled_task) this_rq->idle_stamp = 0; + rq_repin_lock(this_rq, rf); + return pulled_task; } @@ -8405,8 +8412,9 @@ static int active_load_balance_cpu_stop(void *data) struct rq *target_rq = cpu_rq(target_cpu); struct sched_domain *sd; struct task_struct *p = NULL; + struct rq_flags rf; - raw_spin_lock_irq(&busiest_rq->lock); + rq_lock_irq(busiest_rq, &rf); /* make sure the requested cpu hasn't gone down in the meantime */ if (unlikely(busiest_cpu != smp_processor_id() || @@ -8440,9 +8448,17 @@ static int active_load_balance_cpu_stop(void *data) .src_cpu = busiest_rq->cpu, .src_rq = busiest_rq, .idle = CPU_IDLE, + /* + * can_migrate_task() doesn't need to compute new_dst_cpu + * for active balancing. Since we have CPU_IDLE, but no + * @dst_grpmask we need to make that test go away with lying + * about DST_PINNED. + */ + .flags = LBF_DST_PINNED, }; schedstat_inc(sd->alb_count); + update_rq_clock(busiest_rq); p = detach_one_task(&env); if (p) { @@ -8456,7 +8472,7 @@ static int active_load_balance_cpu_stop(void *data) rcu_read_unlock(); out_unlock: busiest_rq->active_balance = 0; - raw_spin_unlock(&busiest_rq->lock); + rq_unlock(busiest_rq, &rf); if (p) attach_one_task(target_rq, p); @@ -8582,6 +8598,10 @@ void nohz_balance_enter_idle(int cpu) if (!cpu_active(cpu)) return; + /* Spare idle load balancing on CPUs that don't want to be disturbed: */ + if (!is_housekeeping_cpu(cpu)) + return; + if (test_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu))) return; @@ -8754,10 +8774,13 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) * do the balance. */ if (time_after_eq(jiffies, rq->next_balance)) { - raw_spin_lock_irq(&rq->lock); + struct rq_flags rf; + + rq_lock_irq(rq, &rf); update_rq_clock(rq); cpu_load_update_idle(rq); - raw_spin_unlock_irq(&rq->lock); + rq_unlock_irq(rq, &rf); + rebalance_domains(rq, CPU_IDLE); } @@ -8948,8 +8971,9 @@ static void task_fork_fair(struct task_struct *p) struct cfs_rq *cfs_rq; struct sched_entity *se = &p->se, *curr; struct rq *rq = this_rq(); + struct rq_flags rf; - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); update_rq_clock(rq); cfs_rq = task_cfs_rq(current); @@ -8970,7 +8994,7 @@ static void task_fork_fair(struct task_struct *p) } se->vruntime -= cfs_rq->min_vruntime; - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); } /* @@ -9264,6 +9288,7 @@ void online_fair_sched_group(struct task_group *tg) se = tg->se[i]; raw_spin_lock_irq(&rq->lock); + update_rq_clock(rq); attach_entity_cfs_rq(se); sync_throttle(tg, i); raw_spin_unlock_irq(&rq->lock); @@ -9331,7 +9356,6 @@ static DEFINE_MUTEX(shares_mutex); int sched_group_set_shares(struct task_group *tg, unsigned long shares) { int i; - unsigned long flags; /* * We can't change the weight of the root cgroup. @@ -9348,17 +9372,17 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) tg->shares = shares; for_each_possible_cpu(i) { struct rq *rq = cpu_rq(i); - struct sched_entity *se; + struct sched_entity *se = tg->se[i]; + struct rq_flags rf; - se = tg->se[i]; /* Propagate contribution to hierarchy */ - raw_spin_lock_irqsave(&rq->lock, flags); - - /* Possible calls to update_curr() need rq clock */ + rq_lock_irqsave(rq, &rf); update_rq_clock(rq); - for_each_sched_entity(se) - update_cfs_shares(group_cfs_rq(se)); - raw_spin_unlock_irqrestore(&rq->lock, flags); + for_each_sched_entity(se) { + update_load_avg(se, UPDATE_TG); + update_cfs_shares(se); + } + rq_unlock_irqrestore(rq, &rf); } done: @@ -9442,10 +9466,10 @@ const struct sched_class fair_sched_class = { #ifdef CONFIG_SCHED_DEBUG void print_cfs_stats(struct seq_file *m, int cpu) { - struct cfs_rq *cfs_rq; + struct cfs_rq *cfs_rq, *pos; rcu_read_lock(); - for_each_leaf_cfs_rq(cpu_rq(cpu), cfs_rq) + for_each_leaf_cfs_rq_safe(cpu_rq(cpu), cfs_rq, pos) print_cfs_rq(m, cpu, cfs_rq); rcu_read_unlock(); } diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 69631fa46c2f..d3fb15555291 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -51,6 +51,19 @@ SCHED_FEAT(NONTASK_CAPACITY, true) */ SCHED_FEAT(TTWU_QUEUE, true) +/* + * When doing wakeups, attempt to limit superfluous scans of the LLC domain. + */ +SCHED_FEAT(SIS_AVG_CPU, false) +SCHED_FEAT(SIS_PROP, true) + +/* + * Issue a WARN when we do multiple update_rq_clock() calls + * in a single rq->lock section. Default disabled because the + * annotations are not complete. + */ +SCHED_FEAT(WARN_DOUBLE_CLOCK, false) + #ifdef HAVE_RT_PUSH_IPI /* * In order to avoid a thundering herd attack of CPUs that are @@ -64,7 +77,6 @@ SCHED_FEAT(TTWU_QUEUE, true) SCHED_FEAT(RT_PUSH_IPI, true) #endif -SCHED_FEAT(FORCE_SD_OVERLAP, false) SCHED_FEAT(RT_RUNTIME_SHARE, true) SCHED_FEAT(LB_MIN, false) SCHED_FEAT(ATTACH_AGE_LOAD, true) diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 6a4bae0a649d..6c23e30c0e5c 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -2,6 +2,7 @@ * Generic entry point for the idle threads */ #include <linux/sched.h> +#include <linux/sched/idle.h> #include <linux/cpu.h> #include <linux/cpuidle.h> #include <linux/cpuhotplug.h> @@ -9,6 +10,7 @@ #include <linux/mm.h> #include <linux/stackprotector.h> #include <linux/suspend.h> +#include <linux/livepatch.h> #include <asm/tlb.h> @@ -217,6 +219,7 @@ static void do_idle(void) */ __current_set_polling(); + quiet_vmstat(); tick_nohz_idle_enter(); while (!need_resched()) { @@ -263,7 +266,10 @@ static void do_idle(void) smp_mb__after_atomic(); sched_ttwu_pending(); - schedule_preempt_disabled(); + schedule_idle(); + + if (unlikely(klp_patch_pending(current))) + klp_update_patch_state(current); } bool cpu_in_idle(unsigned long pc) diff --git a/kernel/sched/idle_task.c b/kernel/sched/idle_task.c index 5405d3feb112..0c00172db63e 100644 --- a/kernel/sched/idle_task.c +++ b/kernel/sched/idle_task.c @@ -24,7 +24,7 @@ static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int fl } static struct task_struct * -pick_next_task_idle(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie) +pick_next_task_idle(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { put_prev_task(rq, prev); update_idle_core(rq); diff --git a/kernel/sched/loadavg.c b/kernel/sched/loadavg.c index a2d6eb71f06b..f14716a3522f 100644 --- a/kernel/sched/loadavg.c +++ b/kernel/sched/loadavg.c @@ -7,6 +7,7 @@ */ #include <linux/export.h> +#include <linux/sched/loadavg.h> #include "sched.h" @@ -116,7 +117,7 @@ calc_load(unsigned long load, unsigned long exp, unsigned long active) * load-average relies on per-cpu sampling from the tick, it is affected by * NO_HZ. * - * The basic idea is to fold the nr_active delta into a global idle-delta upon + * The basic idea is to fold the nr_active delta into a global NO_HZ-delta upon * entering NO_HZ state such that we can include this as an 'extra' cpu delta * when we read the global state. * @@ -125,7 +126,7 @@ calc_load(unsigned long load, unsigned long exp, unsigned long active) * - When we go NO_HZ idle during the window, we can negate our sample * contribution, causing under-accounting. * - * We avoid this by keeping two idle-delta counters and flipping them + * We avoid this by keeping two NO_HZ-delta counters and flipping them * when the window starts, thus separating old and new NO_HZ load. * * The only trick is the slight shift in index flip for read vs write. @@ -136,22 +137,22 @@ calc_load(unsigned long load, unsigned long exp, unsigned long active) * r:0 0 1 1 0 0 1 1 0 * w:0 1 1 0 0 1 1 0 0 * - * This ensures we'll fold the old idle contribution in this window while + * This ensures we'll fold the old NO_HZ contribution in this window while * accumlating the new one. * - * - When we wake up from NO_HZ idle during the window, we push up our + * - When we wake up from NO_HZ during the window, we push up our * contribution, since we effectively move our sample point to a known * busy state. * * This is solved by pushing the window forward, and thus skipping the - * sample, for this cpu (effectively using the idle-delta for this cpu which + * sample, for this cpu (effectively using the NO_HZ-delta for this cpu which * was in effect at the time the window opened). This also solves the issue - * of having to deal with a cpu having been in NOHZ idle for multiple - * LOAD_FREQ intervals. + * of having to deal with a cpu having been in NO_HZ for multiple LOAD_FREQ + * intervals. * * When making the ILB scale, we should try to pull this in as well. */ -static atomic_long_t calc_load_idle[2]; +static atomic_long_t calc_load_nohz[2]; static int calc_load_idx; static inline int calc_load_write_idx(void) @@ -166,9 +167,9 @@ static inline int calc_load_write_idx(void) /* * If the folding window started, make sure we start writing in the - * next idle-delta. + * next NO_HZ-delta. */ - if (!time_before(jiffies, calc_load_update)) + if (!time_before(jiffies, READ_ONCE(calc_load_update))) idx++; return idx & 1; @@ -179,30 +180,31 @@ static inline int calc_load_read_idx(void) return calc_load_idx & 1; } -void calc_load_enter_idle(void) +void calc_load_nohz_start(void) { struct rq *this_rq = this_rq(); long delta; /* - * We're going into NOHZ mode, if there's any pending delta, fold it - * into the pending idle delta. + * We're going into NO_HZ mode, if there's any pending delta, fold it + * into the pending NO_HZ delta. */ delta = calc_load_fold_active(this_rq, 0); if (delta) { int idx = calc_load_write_idx(); - atomic_long_add(delta, &calc_load_idle[idx]); + atomic_long_add(delta, &calc_load_nohz[idx]); } } -void calc_load_exit_idle(void) +void calc_load_nohz_stop(void) { struct rq *this_rq = this_rq(); /* - * If we're still before the sample window, we're done. + * If we're still before the pending sample window, we're done. */ + this_rq->calc_load_update = READ_ONCE(calc_load_update); if (time_before(jiffies, this_rq->calc_load_update)) return; @@ -211,18 +213,17 @@ void calc_load_exit_idle(void) * accounted through the nohz accounting, so skip the entire deal and * sync up for the next window. */ - this_rq->calc_load_update = calc_load_update; if (time_before(jiffies, this_rq->calc_load_update + 10)) this_rq->calc_load_update += LOAD_FREQ; } -static long calc_load_fold_idle(void) +static long calc_load_nohz_fold(void) { int idx = calc_load_read_idx(); long delta = 0; - if (atomic_long_read(&calc_load_idle[idx])) - delta = atomic_long_xchg(&calc_load_idle[idx], 0); + if (atomic_long_read(&calc_load_nohz[idx])) + delta = atomic_long_xchg(&calc_load_nohz[idx], 0); return delta; } @@ -298,22 +299,24 @@ calc_load_n(unsigned long load, unsigned long exp, /* * NO_HZ can leave us missing all per-cpu ticks calling - * calc_load_account_active(), but since an idle CPU folds its delta into - * calc_load_tasks_idle per calc_load_account_idle(), all we need to do is fold - * in the pending idle delta if our idle period crossed a load cycle boundary. + * calc_load_fold_active(), but since a NO_HZ CPU folds its delta into + * calc_load_nohz per calc_load_nohz_start(), all we need to do is fold + * in the pending NO_HZ delta if our NO_HZ period crossed a load cycle boundary. * * Once we've updated the global active value, we need to apply the exponential * weights adjusted to the number of cycles missed. */ static void calc_global_nohz(void) { + unsigned long sample_window; long delta, active, n; - if (!time_before(jiffies, calc_load_update + 10)) { + sample_window = READ_ONCE(calc_load_update); + if (!time_before(jiffies, sample_window + 10)) { /* * Catch-up, fold however many we are behind still */ - delta = jiffies - calc_load_update - 10; + delta = jiffies - sample_window - 10; n = 1 + (delta / LOAD_FREQ); active = atomic_long_read(&calc_load_tasks); @@ -323,11 +326,11 @@ static void calc_global_nohz(void) avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n); avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n); - calc_load_update += n * LOAD_FREQ; + WRITE_ONCE(calc_load_update, sample_window + n * LOAD_FREQ); } /* - * Flip the idle index... + * Flip the NO_HZ index... * * Make sure we first write the new time then flip the index, so that * calc_load_write_idx() will see the new time when it reads the new @@ -338,7 +341,7 @@ static void calc_global_nohz(void) } #else /* !CONFIG_NO_HZ_COMMON */ -static inline long calc_load_fold_idle(void) { return 0; } +static inline long calc_load_nohz_fold(void) { return 0; } static inline void calc_global_nohz(void) { } #endif /* CONFIG_NO_HZ_COMMON */ @@ -351,15 +354,17 @@ static inline void calc_global_nohz(void) { } */ void calc_global_load(unsigned long ticks) { + unsigned long sample_window; long active, delta; - if (time_before(jiffies, calc_load_update + 10)) + sample_window = READ_ONCE(calc_load_update); + if (time_before(jiffies, sample_window + 10)) return; /* - * Fold the 'old' idle-delta to include all NO_HZ cpus. + * Fold the 'old' NO_HZ-delta to include all NO_HZ cpus. */ - delta = calc_load_fold_idle(); + delta = calc_load_nohz_fold(); if (delta) atomic_long_add(delta, &calc_load_tasks); @@ -370,10 +375,11 @@ void calc_global_load(unsigned long ticks) avenrun[1] = calc_load(avenrun[1], EXP_5, active); avenrun[2] = calc_load(avenrun[2], EXP_15, active); - calc_load_update += LOAD_FREQ; + WRITE_ONCE(calc_load_update, sample_window + LOAD_FREQ); /* - * In case we idled for multiple LOAD_FREQ intervals, catch up in bulk. + * In case we went to NO_HZ for multiple LOAD_FREQ intervals + * catch up in bulk. */ calc_global_nohz(); } diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 2516b8df6dbb..45caf937ef90 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -9,6 +9,7 @@ #include <linux/irq_work.h> int sched_rr_timeslice = RR_TIMESLICE; +int sysctl_sched_rr_timeslice = (MSEC_PER_SEC / HZ) * RR_TIMESLICE; static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun); @@ -334,7 +335,7 @@ static void inc_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) rt_rq = &rq_of_rt_rq(rt_rq)->rt; rt_rq->rt_nr_total++; - if (tsk_nr_cpus_allowed(p) > 1) + if (p->nr_cpus_allowed > 1) rt_rq->rt_nr_migratory++; update_rt_migration(rt_rq); @@ -351,7 +352,7 @@ static void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) rt_rq = &rq_of_rt_rq(rt_rq)->rt; rt_rq->rt_nr_total--; - if (tsk_nr_cpus_allowed(p) > 1) + if (p->nr_cpus_allowed > 1) rt_rq->rt_nr_migratory--; update_rt_migration(rt_rq); @@ -839,6 +840,17 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) int enqueue = 0; struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i); struct rq *rq = rq_of_rt_rq(rt_rq); + int skip; + + /* + * When span == cpu_online_mask, taking each rq->lock + * can be time-consuming. Try to avoid it when possible. + */ + raw_spin_lock(&rt_rq->rt_runtime_lock); + skip = !rt_rq->rt_time && !rt_rq->rt_nr_running; + raw_spin_unlock(&rt_rq->rt_runtime_lock); + if (skip) + continue; raw_spin_lock(&rq->lock); if (rt_rq->rt_time) { @@ -1323,7 +1335,7 @@ enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags) enqueue_rt_entity(rt_se, flags); - if (!task_current(rq, p) && tsk_nr_cpus_allowed(p) > 1) + if (!task_current(rq, p) && p->nr_cpus_allowed > 1) enqueue_pushable_task(rq, p); } @@ -1412,7 +1424,7 @@ select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags) * will have to sort it out. */ if (curr && unlikely(rt_task(curr)) && - (tsk_nr_cpus_allowed(curr) < 2 || + (curr->nr_cpus_allowed < 2 || curr->prio <= p->prio)) { int target = find_lowest_rq(p); @@ -1436,7 +1448,7 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) * Current can't be migrated, useless to reschedule, * let's hope p can move out. */ - if (tsk_nr_cpus_allowed(rq->curr) == 1 || + if (rq->curr->nr_cpus_allowed == 1 || !cpupri_find(&rq->rd->cpupri, rq->curr, NULL)) return; @@ -1444,7 +1456,7 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) * p is migratable, so let's not schedule it and * see if it is pushed or pulled somewhere else. */ - if (tsk_nr_cpus_allowed(p) != 1 + if (p->nr_cpus_allowed != 1 && cpupri_find(&rq->rd->cpupri, p, NULL)) return; @@ -1523,7 +1535,7 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq) } static struct task_struct * -pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie) +pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { struct task_struct *p; struct rt_rq *rt_rq = &rq->rt; @@ -1535,9 +1547,9 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct pin_cookie coo * disabled avoiding further scheduler activity on it and we're * being very careful to re-start the picking loop. */ - lockdep_unpin_lock(&rq->lock, cookie); + rq_unpin_lock(rq, rf); pull_rt_task(rq); - lockdep_repin_lock(&rq->lock, cookie); + rq_repin_lock(rq, rf); /* * pull_rt_task() can drop (and re-acquire) rq->lock; this * means a dl or stop task can slip in, in which case we need @@ -1578,7 +1590,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) * The previous task needs to be made eligible for pushing * if it is still active */ - if (on_rt_rq(&p->rt) && tsk_nr_cpus_allowed(p) > 1) + if (on_rt_rq(&p->rt) && p->nr_cpus_allowed > 1) enqueue_pushable_task(rq, p); } @@ -1590,7 +1602,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu) { if (!task_running(rq, p) && - cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) + cpumask_test_cpu(cpu, &p->cpus_allowed)) return 1; return 0; } @@ -1628,7 +1640,7 @@ static int find_lowest_rq(struct task_struct *task) if (unlikely(!lowest_mask)) return -1; - if (tsk_nr_cpus_allowed(task) == 1) + if (task->nr_cpus_allowed == 1) return -1; /* No other targets possible */ if (!cpupri_find(&task_rq(task)->rd->cpupri, task, lowest_mask)) @@ -1725,8 +1737,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) * Also make sure that it wasn't scheduled on its rq. */ if (unlikely(task_rq(task) != rq || - !cpumask_test_cpu(lowest_rq->cpu, - tsk_cpus_allowed(task)) || + !cpumask_test_cpu(lowest_rq->cpu, &task->cpus_allowed) || task_running(rq, task) || !rt_task(task) || !task_on_rq_queued(task))) { @@ -1761,7 +1772,7 @@ static struct task_struct *pick_next_pushable_task(struct rq *rq) BUG_ON(rq->cpu != task_cpu(p)); BUG_ON(task_current(rq, p)); - BUG_ON(tsk_nr_cpus_allowed(p) <= 1); + BUG_ON(p->nr_cpus_allowed <= 1); BUG_ON(!task_on_rq_queued(p)); BUG_ON(!rt_task(p)); @@ -1819,7 +1830,7 @@ retry: * pushing. */ task = pick_next_pushable_task(rq); - if (task_cpu(next_task) == rq->cpu && task == next_task) { + if (task == next_task) { /* * The task hasn't migrated, and is still the next * eligible task, but we failed to find a run-queue @@ -1927,6 +1938,87 @@ static int find_next_push_cpu(struct rq *rq) #define RT_PUSH_IPI_EXECUTING 1 #define RT_PUSH_IPI_RESTART 2 +/* + * When a high priority task schedules out from a CPU and a lower priority + * task is scheduled in, a check is made to see if there's any RT tasks + * on other CPUs that are waiting to run because a higher priority RT task + * is currently running on its CPU. In this case, the CPU with multiple RT + * tasks queued on it (overloaded) needs to be notified that a CPU has opened + * up that may be able to run one of its non-running queued RT tasks. + * + * On large CPU boxes, there's the case that several CPUs could schedule + * a lower priority task at the same time, in which case it will look for + * any overloaded CPUs that it could pull a task from. To do this, the runqueue + * lock must be taken from that overloaded CPU. Having 10s of CPUs all fighting + * for a single overloaded CPU's runqueue lock can produce a large latency. + * (This has actually been observed on large boxes running cyclictest). + * Instead of taking the runqueue lock of the overloaded CPU, each of the + * CPUs that scheduled a lower priority task simply sends an IPI to the + * overloaded CPU. An IPI is much cheaper than taking an runqueue lock with + * lots of contention. The overloaded CPU will look to push its non-running + * RT task off, and if it does, it can then ignore the other IPIs coming + * in, and just pass those IPIs off to any other overloaded CPU. + * + * When a CPU schedules a lower priority task, it only sends an IPI to + * the "next" CPU that has overloaded RT tasks. This prevents IPI storms, + * as having 10 CPUs scheduling lower priority tasks and 10 CPUs with + * RT overloaded tasks, would cause 100 IPIs to go out at once. + * + * The overloaded RT CPU, when receiving an IPI, will try to push off its + * overloaded RT tasks and then send an IPI to the next CPU that has + * overloaded RT tasks. This stops when all CPUs with overloaded RT tasks + * have completed. Just because a CPU may have pushed off its own overloaded + * RT task does not mean it should stop sending the IPI around to other + * overloaded CPUs. There may be another RT task waiting to run on one of + * those CPUs that are of higher priority than the one that was just + * pushed. + * + * An optimization that could possibly be made is to make a CPU array similar + * to the cpupri array mask of all running RT tasks, but for the overloaded + * case, then the IPI could be sent to only the CPU with the highest priority + * RT task waiting, and that CPU could send off further IPIs to the CPU with + * the next highest waiting task. Since the overloaded case is much less likely + * to happen, the complexity of this implementation may not be worth it. + * Instead, just send an IPI around to all overloaded CPUs. + * + * The rq->rt.push_flags holds the status of the IPI that is going around. + * A run queue can only send out a single IPI at a time. The possible flags + * for rq->rt.push_flags are: + * + * (None or zero): No IPI is going around for the current rq + * RT_PUSH_IPI_EXECUTING: An IPI for the rq is being passed around + * RT_PUSH_IPI_RESTART: The priority of the running task for the rq + * has changed, and the IPI should restart + * circulating the overloaded CPUs again. + * + * rq->rt.push_cpu contains the CPU that is being sent the IPI. It is updated + * before sending to the next CPU. + * + * Instead of having all CPUs that schedule a lower priority task send + * an IPI to the same "first" CPU in the RT overload mask, they send it + * to the next overloaded CPU after their own CPU. This helps distribute + * the work when there's more than one overloaded CPU and multiple CPUs + * scheduling in lower priority tasks. + * + * When a rq schedules a lower priority task than what was currently + * running, the next CPU with overloaded RT tasks is examined first. + * That is, if CPU 1 and 5 are overloaded, and CPU 3 schedules a lower + * priority task, it will send an IPI first to CPU 5, then CPU 5 will + * send to CPU 1 if it is still overloaded. CPU 1 will clear the + * rq->rt.push_flags if RT_PUSH_IPI_RESTART is not set. + * + * The first CPU to notice IPI_RESTART is set, will clear that flag and then + * send an IPI to the next overloaded CPU after the rq->cpu and not the next + * CPU after push_cpu. That is, if CPU 1, 4 and 5 are overloaded when CPU 3 + * schedules a lower priority task, and the IPI_RESTART gets set while the + * handling is being done on CPU 5, it will clear the flag and send it back to + * CPU 4 instead of CPU 1. + * + * Note, the above logic can be disabled by turning off the sched_feature + * RT_PUSH_IPI. Then the rq lock of the overloaded CPU will simply be + * taken by the CPU requesting a pull and the waiting RT task will be pulled + * by that CPU. This may be fine for machines with few CPUs. + */ static void tell_cpu_to_push(struct rq *rq) { int cpu; @@ -2121,9 +2213,9 @@ static void task_woken_rt(struct rq *rq, struct task_struct *p) { if (!task_running(rq, p) && !test_tsk_need_resched(rq->curr) && - tsk_nr_cpus_allowed(p) > 1 && + p->nr_cpus_allowed > 1 && (dl_task(rq->curr) || rt_task(rq->curr)) && - (tsk_nr_cpus_allowed(rq->curr) < 2 || + (rq->curr->nr_cpus_allowed < 2 || rq->curr->prio <= p->prio)) push_rt_tasks(rq); } @@ -2196,12 +2288,11 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p) */ if (task_on_rq_queued(p) && rq->curr != p) { #ifdef CONFIG_SMP - if (tsk_nr_cpus_allowed(p) > 1 && rq->rt.overloaded) + if (p->nr_cpus_allowed > 1 && rq->rt.overloaded) queue_push_tasks(rq); -#else +#endif /* CONFIG_SMP */ if (p->prio < rq->curr->prio) resched_curr(rq); -#endif /* CONFIG_SMP */ } } @@ -2246,6 +2337,7 @@ prio_changed_rt(struct rq *rq, struct task_struct *p, int oldprio) } } +#ifdef CONFIG_POSIX_TIMERS static void watchdog(struct rq *rq, struct task_struct *p) { unsigned long soft, hard; @@ -2267,6 +2359,9 @@ static void watchdog(struct rq *rq, struct task_struct *p) p->cputime_expires.sched_exp = p->se.sum_exec_runtime; } } +#else +static inline void watchdog(struct rq *rq, struct task_struct *p) { } +#endif static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued) { @@ -2354,6 +2449,316 @@ const struct sched_class rt_sched_class = { .update_curr = update_curr_rt, }; +#ifdef CONFIG_RT_GROUP_SCHED +/* + * Ensure that the real time constraints are schedulable. + */ +static DEFINE_MUTEX(rt_constraints_mutex); + +/* Must be called with tasklist_lock held */ +static inline int tg_has_rt_tasks(struct task_group *tg) +{ + struct task_struct *g, *p; + + /* + * Autogroups do not have RT tasks; see autogroup_create(). + */ + if (task_group_is_autogroup(tg)) + return 0; + + for_each_process_thread(g, p) { + if (rt_task(p) && task_group(p) == tg) + return 1; + } + + return 0; +} + +struct rt_schedulable_data { + struct task_group *tg; + u64 rt_period; + u64 rt_runtime; +}; + +static int tg_rt_schedulable(struct task_group *tg, void *data) +{ + struct rt_schedulable_data *d = data; + struct task_group *child; + unsigned long total, sum = 0; + u64 period, runtime; + + period = ktime_to_ns(tg->rt_bandwidth.rt_period); + runtime = tg->rt_bandwidth.rt_runtime; + + if (tg == d->tg) { + period = d->rt_period; + runtime = d->rt_runtime; + } + + /* + * Cannot have more runtime than the period. + */ + if (runtime > period && runtime != RUNTIME_INF) + return -EINVAL; + + /* + * Ensure we don't starve existing RT tasks. + */ + if (rt_bandwidth_enabled() && !runtime && tg_has_rt_tasks(tg)) + return -EBUSY; + + total = to_ratio(period, runtime); + + /* + * Nobody can have more than the global setting allows. + */ + if (total > to_ratio(global_rt_period(), global_rt_runtime())) + return -EINVAL; + + /* + * The sum of our children's runtime should not exceed our own. + */ + list_for_each_entry_rcu(child, &tg->children, siblings) { + period = ktime_to_ns(child->rt_bandwidth.rt_period); + runtime = child->rt_bandwidth.rt_runtime; + + if (child == d->tg) { + period = d->rt_period; + runtime = d->rt_runtime; + } + + sum += to_ratio(period, runtime); + } + + if (sum > total) + return -EINVAL; + + return 0; +} + +static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) +{ + int ret; + + struct rt_schedulable_data data = { + .tg = tg, + .rt_period = period, + .rt_runtime = runtime, + }; + + rcu_read_lock(); + ret = walk_tg_tree(tg_rt_schedulable, tg_nop, &data); + rcu_read_unlock(); + + return ret; +} + +static int tg_set_rt_bandwidth(struct task_group *tg, + u64 rt_period, u64 rt_runtime) +{ + int i, err = 0; + + /* + * Disallowing the root group RT runtime is BAD, it would disallow the + * kernel creating (and or operating) RT threads. + */ + if (tg == &root_task_group && rt_runtime == 0) + return -EINVAL; + + /* No period doesn't make any sense. */ + if (rt_period == 0) + return -EINVAL; + + mutex_lock(&rt_constraints_mutex); + read_lock(&tasklist_lock); + err = __rt_schedulable(tg, rt_period, rt_runtime); + if (err) + goto unlock; + + raw_spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); + tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period); + tg->rt_bandwidth.rt_runtime = rt_runtime; + + for_each_possible_cpu(i) { + struct rt_rq *rt_rq = tg->rt_rq[i]; + + raw_spin_lock(&rt_rq->rt_runtime_lock); + rt_rq->rt_runtime = rt_runtime; + raw_spin_unlock(&rt_rq->rt_runtime_lock); + } + raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); +unlock: + read_unlock(&tasklist_lock); + mutex_unlock(&rt_constraints_mutex); + + return err; +} + +int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us) +{ + u64 rt_runtime, rt_period; + + rt_period = ktime_to_ns(tg->rt_bandwidth.rt_period); + rt_runtime = (u64)rt_runtime_us * NSEC_PER_USEC; + if (rt_runtime_us < 0) + rt_runtime = RUNTIME_INF; + + return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); +} + +long sched_group_rt_runtime(struct task_group *tg) +{ + u64 rt_runtime_us; + + if (tg->rt_bandwidth.rt_runtime == RUNTIME_INF) + return -1; + + rt_runtime_us = tg->rt_bandwidth.rt_runtime; + do_div(rt_runtime_us, NSEC_PER_USEC); + return rt_runtime_us; +} + +int sched_group_set_rt_period(struct task_group *tg, u64 rt_period_us) +{ + u64 rt_runtime, rt_period; + + rt_period = rt_period_us * NSEC_PER_USEC; + rt_runtime = tg->rt_bandwidth.rt_runtime; + + return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); +} + +long sched_group_rt_period(struct task_group *tg) +{ + u64 rt_period_us; + + rt_period_us = ktime_to_ns(tg->rt_bandwidth.rt_period); + do_div(rt_period_us, NSEC_PER_USEC); + return rt_period_us; +} + +static int sched_rt_global_constraints(void) +{ + int ret = 0; + + mutex_lock(&rt_constraints_mutex); + read_lock(&tasklist_lock); + ret = __rt_schedulable(NULL, 0, 0); + read_unlock(&tasklist_lock); + mutex_unlock(&rt_constraints_mutex); + + return ret; +} + +int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk) +{ + /* Don't accept realtime tasks when there is no way for them to run */ + if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) + return 0; + + return 1; +} + +#else /* !CONFIG_RT_GROUP_SCHED */ +static int sched_rt_global_constraints(void) +{ + unsigned long flags; + int i; + + raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); + for_each_possible_cpu(i) { + struct rt_rq *rt_rq = &cpu_rq(i)->rt; + + raw_spin_lock(&rt_rq->rt_runtime_lock); + rt_rq->rt_runtime = global_rt_runtime(); + raw_spin_unlock(&rt_rq->rt_runtime_lock); + } + raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); + + return 0; +} +#endif /* CONFIG_RT_GROUP_SCHED */ + +static int sched_rt_global_validate(void) +{ + if (sysctl_sched_rt_period <= 0) + return -EINVAL; + + if ((sysctl_sched_rt_runtime != RUNTIME_INF) && + (sysctl_sched_rt_runtime > sysctl_sched_rt_period)) + return -EINVAL; + + return 0; +} + +static void sched_rt_do_global(void) +{ + def_rt_bandwidth.rt_runtime = global_rt_runtime(); + def_rt_bandwidth.rt_period = ns_to_ktime(global_rt_period()); +} + +int sched_rt_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + int old_period, old_runtime; + static DEFINE_MUTEX(mutex); + int ret; + + mutex_lock(&mutex); + old_period = sysctl_sched_rt_period; + old_runtime = sysctl_sched_rt_runtime; + + ret = proc_dointvec(table, write, buffer, lenp, ppos); + + if (!ret && write) { + ret = sched_rt_global_validate(); + if (ret) + goto undo; + + ret = sched_dl_global_validate(); + if (ret) + goto undo; + + ret = sched_rt_global_constraints(); + if (ret) + goto undo; + + sched_rt_do_global(); + sched_dl_do_global(); + } + if (0) { +undo: + sysctl_sched_rt_period = old_period; + sysctl_sched_rt_runtime = old_runtime; + } + mutex_unlock(&mutex); + + return ret; +} + +int sched_rr_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + int ret; + static DEFINE_MUTEX(mutex); + + mutex_lock(&mutex); + ret = proc_dointvec(table, write, buffer, lenp, ppos); + /* + * Make sure that internally we keep jiffies. + * Also, writing zero resets the timeslice to default: + */ + if (!ret && write) { + sched_rr_timeslice = + sysctl_sched_rr_timeslice <= 0 ? RR_TIMESLICE : + msecs_to_jiffies(sysctl_sched_rr_timeslice); + } + mutex_unlock(&mutex); + return ret; +} + #ifdef CONFIG_SCHED_DEBUG extern void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq); diff --git a/kernel/sched/sched-pelt.h b/kernel/sched/sched-pelt.h new file mode 100644 index 000000000000..cd200d16529e --- /dev/null +++ b/kernel/sched/sched-pelt.h @@ -0,0 +1,13 @@ +/* Generated by Documentation/scheduler/sched-pelt; do not modify. */ + +static const u32 runnable_avg_yN_inv[] = { + 0xffffffff, 0xfa83b2da, 0xf5257d14, 0xefe4b99a, 0xeac0c6e6, 0xe5b906e6, + 0xe0ccdeeb, 0xdbfbb796, 0xd744fcc9, 0xd2a81d91, 0xce248c14, 0xc9b9bd85, + 0xc5672a10, 0xc12c4cc9, 0xbd08a39e, 0xb8fbaf46, 0xb504f333, 0xb123f581, + 0xad583ee9, 0xa9a15ab4, 0xa5fed6a9, 0xa2704302, 0x9ef5325f, 0x9b8d39b9, + 0x9837f050, 0x94f4efa8, 0x91c3d373, 0x8ea4398a, 0x8b95c1e3, 0x88980e80, + 0x85aac367, 0x82cd8698, +}; + +#define LOAD_AVG_PERIOD 32 +#define LOAD_AVG_MAX 47742 diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 7b34c7826ca5..eeef1a3086d1 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1,9 +1,27 @@ #include <linux/sched.h> +#include <linux/sched/autogroup.h> #include <linux/sched/sysctl.h> +#include <linux/sched/topology.h> #include <linux/sched/rt.h> -#include <linux/u64_stats_sync.h> #include <linux/sched/deadline.h> +#include <linux/sched/clock.h> +#include <linux/sched/wake_q.h> +#include <linux/sched/signal.h> +#include <linux/sched/numa_balancing.h> +#include <linux/sched/mm.h> +#include <linux/sched/cpufreq.h> +#include <linux/sched/stat.h> +#include <linux/sched/nohz.h> +#include <linux/sched/debug.h> +#include <linux/sched/hotplug.h> +#include <linux/sched/task.h> +#include <linux/sched/task_stack.h> +#include <linux/sched/cputime.h> +#include <linux/sched/init.h> + +#include <linux/u64_stats_sync.h> +#include <linux/kernel_stat.h> #include <linux/binfmts.h> #include <linux/mutex.h> #include <linux/spinlock.h> @@ -12,14 +30,18 @@ #include <linux/tick.h> #include <linux/slab.h> +#ifdef CONFIG_PARAVIRT +#include <asm/paravirt.h> +#endif + #include "cpupri.h" #include "cpudeadline.h" #include "cpuacct.h" #ifdef CONFIG_SCHED_DEBUG -#define SCHED_WARN_ON(x) WARN_ONCE(x, #x) +# define SCHED_WARN_ON(x) WARN_ONCE(x, #x) #else -#define SCHED_WARN_ON(x) ((void)(x)) +# define SCHED_WARN_ON(x) ({ (void)(x), 0; }) #endif struct rq; @@ -196,23 +218,25 @@ static inline int dl_bandwidth_enabled(void) return sysctl_sched_rt_runtime >= 0; } -extern struct dl_bw *dl_bw_of(int i); - struct dl_bw { raw_spinlock_t lock; u64 bw, total_bw; }; +static inline void __dl_update(struct dl_bw *dl_b, s64 bw); + static inline -void __dl_clear(struct dl_bw *dl_b, u64 tsk_bw) +void __dl_clear(struct dl_bw *dl_b, u64 tsk_bw, int cpus) { dl_b->total_bw -= tsk_bw; + __dl_update(dl_b, (s32)tsk_bw / cpus); } static inline -void __dl_add(struct dl_bw *dl_b, u64 tsk_bw) +void __dl_add(struct dl_bw *dl_b, u64 tsk_bw, int cpus) { dl_b->total_bw += tsk_bw; + __dl_update(dl_b, -((s32)tsk_bw / cpus)); } static inline @@ -222,7 +246,22 @@ bool __dl_overflow(struct dl_bw *dl_b, int cpus, u64 old_bw, u64 new_bw) dl_b->bw * cpus < dl_b->total_bw - old_bw + new_bw; } -extern struct mutex sched_domains_mutex; +void dl_change_utilization(struct task_struct *p, u64 new_bw); +extern void init_dl_bw(struct dl_bw *dl_b); +extern int sched_dl_global_validate(void); +extern void sched_dl_do_global(void); +extern int sched_dl_overflow(struct task_struct *p, int policy, + const struct sched_attr *attr); +extern void __setparam_dl(struct task_struct *p, const struct sched_attr *attr); +extern void __getparam_dl(struct task_struct *p, struct sched_attr *attr); +extern bool __checkparam_dl(const struct sched_attr *attr); +extern void __dl_clear_params(struct task_struct *p); +extern bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr); +extern int dl_task_can_attach(struct task_struct *p, + const struct cpumask *cs_cpus_allowed); +extern int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, + const struct cpumask *trial); +extern bool dl_cpu_busy(unsigned int cpu); #ifdef CONFIG_CGROUP_SCHED @@ -344,6 +383,11 @@ extern int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent extern void init_tg_rt_entry(struct task_group *tg, struct rt_rq *rt_rq, struct sched_rt_entity *rt_se, int cpu, struct sched_rt_entity *parent); +extern int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us); +extern int sched_group_set_rt_period(struct task_group *tg, u64 rt_period_us); +extern long sched_group_rt_runtime(struct task_group *tg); +extern long sched_group_rt_period(struct task_group *tg); +extern int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk); extern struct task_group *sched_create_group(struct task_group *parent); extern void sched_online_group(struct task_group *tg, @@ -536,6 +580,30 @@ struct dl_rq { #else struct dl_bw dl_bw; #endif + /* + * "Active utilization" for this runqueue: increased when a + * task wakes up (becomes TASK_RUNNING) and decreased when a + * task blocks + */ + u64 running_bw; + + /* + * Utilization of the tasks "assigned" to this runqueue (including + * the tasks that are in runqueue and the tasks that executed on this + * CPU and blocked). Increased when a task moves to this runqueue, and + * decreased when the task moves away (migrates, changes scheduling + * policy, or terminates). + * This is needed to compute the "inactive utilization" for the + * runqueue (inactive utilization = this_bw - running_bw). + */ + u64 this_bw; + u64 extra_bw; + + /* + * Inverse of the fraction of CPU utilization that can be reclaimed + * by the GRUB algorithm. + */ + u64 bw_ratio; }; #ifdef CONFIG_SMP @@ -583,6 +651,11 @@ struct root_domain { }; extern struct root_domain def_root_domain; +extern struct mutex sched_domains_mutex; + +extern void init_defrootdomain(void); +extern int sched_init_domains(const struct cpumask *cpu_map); +extern void rq_attach_root(struct rq *rq, struct root_domain *rd); #endif /* CONFIG_SMP */ @@ -644,7 +717,7 @@ struct rq { unsigned long next_balance; struct mm_struct *prev_mm; - unsigned int clock_skip_update; + unsigned int clock_update_flags; u64 clock; u64 clock_task; @@ -768,28 +841,110 @@ static inline u64 __rq_clock_broken(struct rq *rq) return READ_ONCE(rq->clock); } +/* + * rq::clock_update_flags bits + * + * %RQCF_REQ_SKIP - will request skipping of clock update on the next + * call to __schedule(). This is an optimisation to avoid + * neighbouring rq clock updates. + * + * %RQCF_ACT_SKIP - is set from inside of __schedule() when skipping is + * in effect and calls to update_rq_clock() are being ignored. + * + * %RQCF_UPDATED - is a debug flag that indicates whether a call has been + * made to update_rq_clock() since the last time rq::lock was pinned. + * + * If inside of __schedule(), clock_update_flags will have been + * shifted left (a left shift is a cheap operation for the fast path + * to promote %RQCF_REQ_SKIP to %RQCF_ACT_SKIP), so you must use, + * + * if (rq-clock_update_flags >= RQCF_UPDATED) + * + * to check if %RQCF_UPADTED is set. It'll never be shifted more than + * one position though, because the next rq_unpin_lock() will shift it + * back. + */ +#define RQCF_REQ_SKIP 0x01 +#define RQCF_ACT_SKIP 0x02 +#define RQCF_UPDATED 0x04 + +static inline void assert_clock_updated(struct rq *rq) +{ + /* + * The only reason for not seeing a clock update since the + * last rq_pin_lock() is if we're currently skipping updates. + */ + SCHED_WARN_ON(rq->clock_update_flags < RQCF_ACT_SKIP); +} + static inline u64 rq_clock(struct rq *rq) { lockdep_assert_held(&rq->lock); + assert_clock_updated(rq); + return rq->clock; } static inline u64 rq_clock_task(struct rq *rq) { lockdep_assert_held(&rq->lock); + assert_clock_updated(rq); + return rq->clock_task; } -#define RQCF_REQ_SKIP 0x01 -#define RQCF_ACT_SKIP 0x02 - static inline void rq_clock_skip_update(struct rq *rq, bool skip) { lockdep_assert_held(&rq->lock); if (skip) - rq->clock_skip_update |= RQCF_REQ_SKIP; + rq->clock_update_flags |= RQCF_REQ_SKIP; else - rq->clock_skip_update &= ~RQCF_REQ_SKIP; + rq->clock_update_flags &= ~RQCF_REQ_SKIP; +} + +struct rq_flags { + unsigned long flags; + struct pin_cookie cookie; +#ifdef CONFIG_SCHED_DEBUG + /* + * A copy of (rq::clock_update_flags & RQCF_UPDATED) for the + * current pin context is stashed here in case it needs to be + * restored in rq_repin_lock(). + */ + unsigned int clock_update_flags; +#endif +}; + +static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf) +{ + rf->cookie = lockdep_pin_lock(&rq->lock); + +#ifdef CONFIG_SCHED_DEBUG + rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP); + rf->clock_update_flags = 0; +#endif +} + +static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf) +{ +#ifdef CONFIG_SCHED_DEBUG + if (rq->clock_update_flags > RQCF_ACT_SKIP) + rf->clock_update_flags = RQCF_UPDATED; +#endif + + lockdep_unpin_lock(&rq->lock, rf->cookie); +} + +static inline void rq_repin_lock(struct rq *rq, struct rq_flags *rf) +{ + lockdep_repin_lock(&rq->lock, rf->cookie); + +#ifdef CONFIG_SCHED_DEBUG + /* + * Restore the value we stashed in @rf for this pin context. + */ + rq->clock_update_flags |= rf->clock_update_flags; +#endif } #ifdef CONFIG_NUMA @@ -803,6 +958,16 @@ extern int sched_max_numa_distance; extern bool find_numa_distance(int distance); #endif +#ifdef CONFIG_NUMA +extern void sched_init_numa(void); +extern void sched_domains_numa_masks_set(unsigned int cpu); +extern void sched_domains_numa_masks_clear(unsigned int cpu); +#else +static inline void sched_init_numa(void) { } +static inline void sched_domains_numa_masks_set(unsigned int cpu) { } +static inline void sched_domains_numa_masks_clear(unsigned int cpu) { } +#endif + #ifdef CONFIG_NUMA_BALANCING /* The regions in numa_faults array from task_struct */ enum numa_faults_stats { @@ -904,7 +1069,11 @@ struct sched_group_capacity { unsigned long next_update; int imbalance; /* XXX unrelated to capacity but shared group state */ - unsigned long cpumask[0]; /* iteration mask */ +#ifdef CONFIG_SCHED_DEBUG + int id; +#endif + + unsigned long cpumask[0]; /* balance mask */ }; struct sched_group { @@ -925,16 +1094,15 @@ struct sched_group { unsigned long cpumask[0]; }; -static inline struct cpumask *sched_group_cpus(struct sched_group *sg) +static inline struct cpumask *sched_group_span(struct sched_group *sg) { return to_cpumask(sg->cpumask); } /* - * cpumask masking which cpus in the group are allowed to iterate up the domain - * tree. + * See build_balance_mask(). */ -static inline struct cpumask *sched_group_mask(struct sched_group *sg) +static inline struct cpumask *group_balance_mask(struct sched_group *sg) { return to_cpumask(sg->sgc->cpumask); } @@ -945,7 +1113,7 @@ static inline struct cpumask *sched_group_mask(struct sched_group *sg) */ static inline unsigned int group_first_cpu(struct sched_group *group) { - return cpumask_first(sched_group_cpus(group)); + return cpumask_first(sched_group_span(group)); } extern int group_balance_cpu(struct sched_group *sg); @@ -969,7 +1137,7 @@ static inline void sched_ttwu_pending(void) { } #endif /* CONFIG_SMP */ #include "stats.h" -#include "auto_group.h" +#include "autogroup.h" #ifdef CONFIG_CGROUP_SCHED @@ -1210,15 +1378,17 @@ extern const u32 sched_prio_to_wmult[40]; #define DEQUEUE_SLEEP 0x01 #define DEQUEUE_SAVE 0x02 /* matches ENQUEUE_RESTORE */ #define DEQUEUE_MOVE 0x04 /* matches ENQUEUE_MOVE */ +#define DEQUEUE_NOCLOCK 0x08 /* matches ENQUEUE_NOCLOCK */ #define ENQUEUE_WAKEUP 0x01 #define ENQUEUE_RESTORE 0x02 #define ENQUEUE_MOVE 0x04 +#define ENQUEUE_NOCLOCK 0x08 -#define ENQUEUE_HEAD 0x08 -#define ENQUEUE_REPLENISH 0x10 +#define ENQUEUE_HEAD 0x10 +#define ENQUEUE_REPLENISH 0x20 #ifdef CONFIG_SMP -#define ENQUEUE_MIGRATED 0x20 +#define ENQUEUE_MIGRATED 0x40 #else #define ENQUEUE_MIGRATED 0x00 #endif @@ -1245,7 +1415,7 @@ struct sched_class { */ struct task_struct * (*pick_next_task) (struct rq *rq, struct task_struct *prev, - struct pin_cookie cookie); + struct rq_flags *rf); void (*put_prev_task) (struct rq *rq, struct task_struct *p); #ifdef CONFIG_SMP @@ -1299,7 +1469,11 @@ static inline void set_curr_task(struct rq *rq, struct task_struct *curr) curr->sched_class->set_curr_task(rq); } +#ifdef CONFIG_SMP #define sched_class_highest (&stop_sched_class) +#else +#define sched_class_highest (&dl_sched_class) +#endif #define for_each_class(class) \ for (class = sched_class_highest; class; class = class->next) @@ -1344,6 +1518,8 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq) } #endif +extern void schedule_idle(void); + extern void sysrq_sched_debug_show(void); extern void sched_init_granularity(void); extern void update_max_interval(void); @@ -1361,7 +1537,12 @@ extern void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime extern struct dl_bandwidth def_dl_bandwidth; extern void init_dl_bandwidth(struct dl_bandwidth *dl_b, u64 period, u64 runtime); extern void init_dl_task_timer(struct sched_dl_entity *dl_se); +extern void init_dl_inactive_task_timer(struct sched_dl_entity *dl_se); +extern void init_dl_rq_bw_ratio(struct dl_rq *dl_rq); +#define BW_SHIFT 20 +#define BW_UNIT (1 << BW_SHIFT) +#define RATIO_SHIFT 8 unsigned long to_ratio(u64 period, u64 runtime); extern void init_entity_runnable_average(struct sched_entity *se); @@ -1501,13 +1682,9 @@ static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { } static inline void sched_avg_update(struct rq *rq) { } #endif -struct rq_flags { - unsigned long flags; - struct pin_cookie cookie; -}; - struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) __acquires(rq->lock); + struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) __acquires(p->pi_lock) __acquires(rq->lock); @@ -1515,7 +1692,7 @@ struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) static inline void __task_rq_unlock(struct rq *rq, struct rq_flags *rf) __releases(rq->lock) { - lockdep_unpin_lock(&rq->lock, rf->cookie); + rq_unpin_lock(rq, rf); raw_spin_unlock(&rq->lock); } @@ -1524,11 +1701,67 @@ task_rq_unlock(struct rq *rq, struct task_struct *p, struct rq_flags *rf) __releases(rq->lock) __releases(p->pi_lock) { - lockdep_unpin_lock(&rq->lock, rf->cookie); + rq_unpin_lock(rq, rf); raw_spin_unlock(&rq->lock); raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags); } +static inline void +rq_lock_irqsave(struct rq *rq, struct rq_flags *rf) + __acquires(rq->lock) +{ + raw_spin_lock_irqsave(&rq->lock, rf->flags); + rq_pin_lock(rq, rf); +} + +static inline void +rq_lock_irq(struct rq *rq, struct rq_flags *rf) + __acquires(rq->lock) +{ + raw_spin_lock_irq(&rq->lock); + rq_pin_lock(rq, rf); +} + +static inline void +rq_lock(struct rq *rq, struct rq_flags *rf) + __acquires(rq->lock) +{ + raw_spin_lock(&rq->lock); + rq_pin_lock(rq, rf); +} + +static inline void +rq_relock(struct rq *rq, struct rq_flags *rf) + __acquires(rq->lock) +{ + raw_spin_lock(&rq->lock); + rq_repin_lock(rq, rf); +} + +static inline void +rq_unlock_irqrestore(struct rq *rq, struct rq_flags *rf) + __releases(rq->lock) +{ + rq_unpin_lock(rq, rf); + raw_spin_unlock_irqrestore(&rq->lock, rf->flags); +} + +static inline void +rq_unlock_irq(struct rq *rq, struct rq_flags *rf) + __releases(rq->lock) +{ + rq_unpin_lock(rq, rf); + raw_spin_unlock_irq(&rq->lock); +} + +static inline void +rq_unlock(struct rq *rq, struct rq_flags *rf) + __releases(rq->lock) +{ + rq_unpin_lock(rq, rf); + raw_spin_unlock(&rq->lock); +} + #ifdef CONFIG_SMP #ifdef CONFIG_PREEMPT @@ -1674,6 +1907,10 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) __release(rq2->lock); } +extern void set_rq_online (struct rq *rq); +extern void set_rq_offline(struct rq *rq); +extern bool sched_smp_initialized; + #else /* CONFIG_SMP */ /* @@ -1718,7 +1955,6 @@ extern void print_rt_stats(struct seq_file *m, int cpu); extern void print_dl_stats(struct seq_file *m, int cpu); extern void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq); - #ifdef CONFIG_NUMA_BALANCING extern void show_numa_stats(struct task_struct *p, struct seq_file *m); @@ -1748,16 +1984,48 @@ extern void nohz_balance_exit_idle(unsigned int cpu); static inline void nohz_balance_exit_idle(unsigned int cpu) { } #endif + +#ifdef CONFIG_SMP +static inline +void __dl_update(struct dl_bw *dl_b, s64 bw) +{ + struct root_domain *rd = container_of(dl_b, struct root_domain, dl_bw); + int i; + + RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), + "sched RCU must be held"); + for_each_cpu_and(i, rd->span, cpu_active_mask) { + struct rq *rq = cpu_rq(i); + + rq->dl.extra_bw += bw; + } +} +#else +static inline +void __dl_update(struct dl_bw *dl_b, s64 bw) +{ + struct dl_rq *dl = container_of(dl_b, struct dl_rq, dl_bw); + + dl->extra_bw += bw; +} +#endif + + #ifdef CONFIG_IRQ_TIME_ACCOUNTING struct irqtime { - u64 hardirq_time; - u64 softirq_time; + u64 total; + u64 tick_delta; u64 irq_start_time; struct u64_stats_sync sync; }; DECLARE_PER_CPU(struct irqtime, cpu_irqtime); +/* + * Returns the irqtime minus the softirq time computed by ksoftirqd. + * Otherwise ksoftirqd's sum_exec_runtime is substracted its own runtime + * and never move forward. + */ static inline u64 irq_time_read(int cpu) { struct irqtime *irqtime = &per_cpu(cpu_irqtime, cpu); @@ -1766,7 +2034,7 @@ static inline u64 irq_time_read(int cpu) do { seq = __u64_stats_fetch_begin(&irqtime->sync); - total = irqtime->softirq_time + irqtime->hardirq_time; + total = irqtime->total; } while (__u64_stats_fetch_retry(&irqtime->sync, seq)); return total; diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h index 34659a853505..d5710651043b 100644 --- a/kernel/sched/stats.h +++ b/kernel/sched/stats.h @@ -164,106 +164,3 @@ sched_info_switch(struct rq *rq, #define sched_info_arrive(rq, next) do { } while (0) #define sched_info_switch(rq, t, next) do { } while (0) #endif /* CONFIG_SCHED_INFO */ - -/* - * The following are functions that support scheduler-internal time accounting. - * These functions are generally called at the timer tick. None of this depends - * on CONFIG_SCHEDSTATS. - */ - -/** - * cputimer_running - return true if cputimer is running - * - * @tsk: Pointer to target task. - */ -static inline bool cputimer_running(struct task_struct *tsk) - -{ - struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - - /* Check if cputimer isn't running. This is accessed without locking. */ - if (!READ_ONCE(cputimer->running)) - return false; - - /* - * After we flush the task's sum_exec_runtime to sig->sum_sched_runtime - * in __exit_signal(), we won't account to the signal struct further - * cputime consumed by that task, even though the task can still be - * ticking after __exit_signal(). - * - * In order to keep a consistent behaviour between thread group cputime - * and thread group cputimer accounting, lets also ignore the cputime - * elapsing after __exit_signal() in any thread group timer running. - * - * This makes sure that POSIX CPU clocks and timers are synchronized, so - * that a POSIX CPU timer won't expire while the corresponding POSIX CPU - * clock delta is behind the expiring timer value. - */ - if (unlikely(!tsk->sighand)) - return false; - - return true; -} - -/** - * account_group_user_time - Maintain utime for a thread group. - * - * @tsk: Pointer to task structure. - * @cputime: Time value by which to increment the utime field of the - * thread_group_cputime structure. - * - * If thread group time is being maintained, get the structure for the - * running CPU and update the utime field there. - */ -static inline void account_group_user_time(struct task_struct *tsk, - cputime_t cputime) -{ - struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - - if (!cputimer_running(tsk)) - return; - - atomic64_add(cputime, &cputimer->cputime_atomic.utime); -} - -/** - * account_group_system_time - Maintain stime for a thread group. - * - * @tsk: Pointer to task structure. - * @cputime: Time value by which to increment the stime field of the - * thread_group_cputime structure. - * - * If thread group time is being maintained, get the structure for the - * running CPU and update the stime field there. - */ -static inline void account_group_system_time(struct task_struct *tsk, - cputime_t cputime) -{ - struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - - if (!cputimer_running(tsk)) - return; - - atomic64_add(cputime, &cputimer->cputime_atomic.stime); -} - -/** - * account_group_exec_runtime - Maintain exec runtime for a thread group. - * - * @tsk: Pointer to task structure. - * @ns: Time value by which to increment the sum_exec_runtime field - * of the thread_group_cputime structure. - * - * If thread group time is being maintained, get the structure for the - * running CPU and update the sum_exec_runtime field there. - */ -static inline void account_group_exec_runtime(struct task_struct *tsk, - unsigned long long ns) -{ - struct thread_group_cputimer *cputimer = &tsk->signal->cputimer; - - if (!cputimer_running(tsk)) - return; - - atomic64_add(ns, &cputimer->cputime_atomic.sum_exec_runtime); -} diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c index 604297a08b3a..9f69fb630853 100644 --- a/kernel/sched/stop_task.c +++ b/kernel/sched/stop_task.c @@ -24,7 +24,7 @@ check_preempt_curr_stop(struct rq *rq, struct task_struct *p, int flags) } static struct task_struct * -pick_next_task_stop(struct rq *rq, struct task_struct *prev, struct pin_cookie cookie) +pick_next_task_stop(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { struct task_struct *stop = rq->stop; diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c index 82f0dff90030..3d5610dcce11 100644 --- a/kernel/sched/swait.c +++ b/kernel/sched/swait.c @@ -1,4 +1,4 @@ -#include <linux/sched.h> +#include <linux/sched/signal.h> #include <linux/swait.h> void __init_swait_queue_head(struct swait_queue_head *q, const char *name, diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c new file mode 100644 index 000000000000..79895aec281e --- /dev/null +++ b/kernel/sched/topology.c @@ -0,0 +1,1906 @@ +/* + * Scheduler topology setup/handling methods + */ +#include <linux/sched.h> +#include <linux/mutex.h> + +#include "sched.h" + +DEFINE_MUTEX(sched_domains_mutex); + +/* Protected by sched_domains_mutex: */ +cpumask_var_t sched_domains_tmpmask; +cpumask_var_t sched_domains_tmpmask2; + +#ifdef CONFIG_SCHED_DEBUG + +static __read_mostly int sched_debug_enabled; + +static int __init sched_debug_setup(char *str) +{ + sched_debug_enabled = 1; + + return 0; +} +early_param("sched_debug", sched_debug_setup); + +static inline bool sched_debug(void) +{ + return sched_debug_enabled; +} + +static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, + struct cpumask *groupmask) +{ + struct sched_group *group = sd->groups; + + cpumask_clear(groupmask); + + printk(KERN_DEBUG "%*s domain-%d: ", level, "", level); + + if (!(sd->flags & SD_LOAD_BALANCE)) { + printk("does not load-balance\n"); + if (sd->parent) + printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain" + " has parent"); + return -1; + } + + printk(KERN_CONT "span=%*pbl level=%s\n", + cpumask_pr_args(sched_domain_span(sd)), sd->name); + + if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) { + printk(KERN_ERR "ERROR: domain->span does not contain " + "CPU%d\n", cpu); + } + if (!cpumask_test_cpu(cpu, sched_group_span(group))) { + printk(KERN_ERR "ERROR: domain->groups does not contain" + " CPU%d\n", cpu); + } + + printk(KERN_DEBUG "%*s groups:", level + 1, ""); + do { + if (!group) { + printk("\n"); + printk(KERN_ERR "ERROR: group is NULL\n"); + break; + } + + if (!cpumask_weight(sched_group_span(group))) { + printk(KERN_CONT "\n"); + printk(KERN_ERR "ERROR: empty group\n"); + break; + } + + if (!(sd->flags & SD_OVERLAP) && + cpumask_intersects(groupmask, sched_group_span(group))) { + printk(KERN_CONT "\n"); + printk(KERN_ERR "ERROR: repeated CPUs\n"); + break; + } + + cpumask_or(groupmask, groupmask, sched_group_span(group)); + + printk(KERN_CONT " %d:{ span=%*pbl", + group->sgc->id, + cpumask_pr_args(sched_group_span(group))); + + if ((sd->flags & SD_OVERLAP) && + !cpumask_equal(group_balance_mask(group), sched_group_span(group))) { + printk(KERN_CONT " mask=%*pbl", + cpumask_pr_args(group_balance_mask(group))); + } + + if (group->sgc->capacity != SCHED_CAPACITY_SCALE) + printk(KERN_CONT " cap=%lu", group->sgc->capacity); + + if (group == sd->groups && sd->child && + !cpumask_equal(sched_domain_span(sd->child), + sched_group_span(group))) { + printk(KERN_ERR "ERROR: domain->groups does not match domain->child\n"); + } + + printk(KERN_CONT " }"); + + group = group->next; + + if (group != sd->groups) + printk(KERN_CONT ","); + + } while (group != sd->groups); + printk(KERN_CONT "\n"); + + if (!cpumask_equal(sched_domain_span(sd), groupmask)) + printk(KERN_ERR "ERROR: groups don't span domain->span\n"); + + if (sd->parent && + !cpumask_subset(groupmask, sched_domain_span(sd->parent))) + printk(KERN_ERR "ERROR: parent span is not a superset " + "of domain->span\n"); + return 0; +} + +static void sched_domain_debug(struct sched_domain *sd, int cpu) +{ + int level = 0; + + if (!sched_debug_enabled) + return; + + if (!sd) { + printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu); + return; + } + + printk(KERN_DEBUG "CPU%d attaching sched-domain(s):\n", cpu); + + for (;;) { + if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask)) + break; + level++; + sd = sd->parent; + if (!sd) + break; + } +} +#else /* !CONFIG_SCHED_DEBUG */ + +# define sched_debug_enabled 0 +# define sched_domain_debug(sd, cpu) do { } while (0) +static inline bool sched_debug(void) +{ + return false; +} +#endif /* CONFIG_SCHED_DEBUG */ + +static int sd_degenerate(struct sched_domain *sd) +{ + if (cpumask_weight(sched_domain_span(sd)) == 1) + return 1; + + /* Following flags need at least 2 groups */ + if (sd->flags & (SD_LOAD_BALANCE | + SD_BALANCE_NEWIDLE | + SD_BALANCE_FORK | + SD_BALANCE_EXEC | + SD_SHARE_CPUCAPACITY | + SD_ASYM_CPUCAPACITY | + SD_SHARE_PKG_RESOURCES | + SD_SHARE_POWERDOMAIN)) { + if (sd->groups != sd->groups->next) + return 0; + } + + /* Following flags don't use groups */ + if (sd->flags & (SD_WAKE_AFFINE)) + return 0; + + return 1; +} + +static int +sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) +{ + unsigned long cflags = sd->flags, pflags = parent->flags; + + if (sd_degenerate(parent)) + return 1; + + if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent))) + return 0; + + /* Flags needing groups don't count if only 1 group in parent */ + if (parent->groups == parent->groups->next) { + pflags &= ~(SD_LOAD_BALANCE | + SD_BALANCE_NEWIDLE | + SD_BALANCE_FORK | + SD_BALANCE_EXEC | + SD_ASYM_CPUCAPACITY | + SD_SHARE_CPUCAPACITY | + SD_SHARE_PKG_RESOURCES | + SD_PREFER_SIBLING | + SD_SHARE_POWERDOMAIN); + if (nr_node_ids == 1) + pflags &= ~SD_SERIALIZE; + } + if (~cflags & pflags) + return 0; + + return 1; +} + +static void free_rootdomain(struct rcu_head *rcu) +{ + struct root_domain *rd = container_of(rcu, struct root_domain, rcu); + + cpupri_cleanup(&rd->cpupri); + cpudl_cleanup(&rd->cpudl); + free_cpumask_var(rd->dlo_mask); + free_cpumask_var(rd->rto_mask); + free_cpumask_var(rd->online); + free_cpumask_var(rd->span); + kfree(rd); +} + +void rq_attach_root(struct rq *rq, struct root_domain *rd) +{ + struct root_domain *old_rd = NULL; + unsigned long flags; + + raw_spin_lock_irqsave(&rq->lock, flags); + + if (rq->rd) { + old_rd = rq->rd; + + if (cpumask_test_cpu(rq->cpu, old_rd->online)) + set_rq_offline(rq); + + cpumask_clear_cpu(rq->cpu, old_rd->span); + + /* + * If we dont want to free the old_rd yet then + * set old_rd to NULL to skip the freeing later + * in this function: + */ + if (!atomic_dec_and_test(&old_rd->refcount)) + old_rd = NULL; + } + + atomic_inc(&rd->refcount); + rq->rd = rd; + + cpumask_set_cpu(rq->cpu, rd->span); + if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) + set_rq_online(rq); + + raw_spin_unlock_irqrestore(&rq->lock, flags); + + if (old_rd) + call_rcu_sched(&old_rd->rcu, free_rootdomain); +} + +static int init_rootdomain(struct root_domain *rd) +{ + memset(rd, 0, sizeof(*rd)); + + if (!zalloc_cpumask_var(&rd->span, GFP_KERNEL)) + goto out; + if (!zalloc_cpumask_var(&rd->online, GFP_KERNEL)) + goto free_span; + if (!zalloc_cpumask_var(&rd->dlo_mask, GFP_KERNEL)) + goto free_online; + if (!zalloc_cpumask_var(&rd->rto_mask, GFP_KERNEL)) + goto free_dlo_mask; + + init_dl_bw(&rd->dl_bw); + if (cpudl_init(&rd->cpudl) != 0) + goto free_rto_mask; + + if (cpupri_init(&rd->cpupri) != 0) + goto free_cpudl; + return 0; + +free_cpudl: + cpudl_cleanup(&rd->cpudl); +free_rto_mask: + free_cpumask_var(rd->rto_mask); +free_dlo_mask: + free_cpumask_var(rd->dlo_mask); +free_online: + free_cpumask_var(rd->online); +free_span: + free_cpumask_var(rd->span); +out: + return -ENOMEM; +} + +/* + * By default the system creates a single root-domain with all CPUs as + * members (mimicking the global state we have today). + */ +struct root_domain def_root_domain; + +void init_defrootdomain(void) +{ + init_rootdomain(&def_root_domain); + + atomic_set(&def_root_domain.refcount, 1); +} + +static struct root_domain *alloc_rootdomain(void) +{ + struct root_domain *rd; + + rd = kmalloc(sizeof(*rd), GFP_KERNEL); + if (!rd) + return NULL; + + if (init_rootdomain(rd) != 0) { + kfree(rd); + return NULL; + } + + return rd; +} + +static void free_sched_groups(struct sched_group *sg, int free_sgc) +{ + struct sched_group *tmp, *first; + + if (!sg) + return; + + first = sg; + do { + tmp = sg->next; + + if (free_sgc && atomic_dec_and_test(&sg->sgc->ref)) + kfree(sg->sgc); + + kfree(sg); + sg = tmp; + } while (sg != first); +} + +static void destroy_sched_domain(struct sched_domain *sd) +{ + /* + * If its an overlapping domain it has private groups, iterate and + * nuke them all. + */ + if (sd->flags & SD_OVERLAP) { + free_sched_groups(sd->groups, 1); + } else if (atomic_dec_and_test(&sd->groups->ref)) { + kfree(sd->groups->sgc); + kfree(sd->groups); + } + if (sd->shared && atomic_dec_and_test(&sd->shared->ref)) + kfree(sd->shared); + kfree(sd); +} + +static void destroy_sched_domains_rcu(struct rcu_head *rcu) +{ + struct sched_domain *sd = container_of(rcu, struct sched_domain, rcu); + + while (sd) { + struct sched_domain *parent = sd->parent; + destroy_sched_domain(sd); + sd = parent; + } +} + +static void destroy_sched_domains(struct sched_domain *sd) +{ + if (sd) + call_rcu(&sd->rcu, destroy_sched_domains_rcu); +} + +/* + * Keep a special pointer to the highest sched_domain that has + * SD_SHARE_PKG_RESOURCE set (Last Level Cache Domain) for this + * allows us to avoid some pointer chasing select_idle_sibling(). + * + * Also keep a unique ID per domain (we use the first CPU number in + * the cpumask of the domain), this allows us to quickly tell if + * two CPUs are in the same cache domain, see cpus_share_cache(). + */ +DEFINE_PER_CPU(struct sched_domain *, sd_llc); +DEFINE_PER_CPU(int, sd_llc_size); +DEFINE_PER_CPU(int, sd_llc_id); +DEFINE_PER_CPU(struct sched_domain_shared *, sd_llc_shared); +DEFINE_PER_CPU(struct sched_domain *, sd_numa); +DEFINE_PER_CPU(struct sched_domain *, sd_asym); + +static void update_top_cache_domain(int cpu) +{ + struct sched_domain_shared *sds = NULL; + struct sched_domain *sd; + int id = cpu; + int size = 1; + + sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES); + if (sd) { + id = cpumask_first(sched_domain_span(sd)); + size = cpumask_weight(sched_domain_span(sd)); + sds = sd->shared; + } + + rcu_assign_pointer(per_cpu(sd_llc, cpu), sd); + per_cpu(sd_llc_size, cpu) = size; + per_cpu(sd_llc_id, cpu) = id; + rcu_assign_pointer(per_cpu(sd_llc_shared, cpu), sds); + + sd = lowest_flag_domain(cpu, SD_NUMA); + rcu_assign_pointer(per_cpu(sd_numa, cpu), sd); + + sd = highest_flag_domain(cpu, SD_ASYM_PACKING); + rcu_assign_pointer(per_cpu(sd_asym, cpu), sd); +} + +/* + * Attach the domain 'sd' to 'cpu' as its base domain. Callers must + * hold the hotplug lock. + */ +static void +cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) +{ + struct rq *rq = cpu_rq(cpu); + struct sched_domain *tmp; + + /* Remove the sched domains which do not contribute to scheduling. */ + for (tmp = sd; tmp; ) { + struct sched_domain *parent = tmp->parent; + if (!parent) + break; + + if (sd_parent_degenerate(tmp, parent)) { + tmp->parent = parent->parent; + if (parent->parent) + parent->parent->child = tmp; + /* + * Transfer SD_PREFER_SIBLING down in case of a + * degenerate parent; the spans match for this + * so the property transfers. + */ + if (parent->flags & SD_PREFER_SIBLING) + tmp->flags |= SD_PREFER_SIBLING; + destroy_sched_domain(parent); + } else + tmp = tmp->parent; + } + + if (sd && sd_degenerate(sd)) { + tmp = sd; + sd = sd->parent; + destroy_sched_domain(tmp); + if (sd) + sd->child = NULL; + } + + sched_domain_debug(sd, cpu); + + rq_attach_root(rq, rd); + tmp = rq->sd; + rcu_assign_pointer(rq->sd, sd); + destroy_sched_domains(tmp); + + update_top_cache_domain(cpu); +} + +/* Setup the mask of CPUs configured for isolated domains */ +static int __init isolated_cpu_setup(char *str) +{ + int ret; + + alloc_bootmem_cpumask_var(&cpu_isolated_map); + ret = cpulist_parse(str, cpu_isolated_map); + if (ret) { + pr_err("sched: Error, all isolcpus= values must be between 0 and %d\n", nr_cpu_ids); + return 0; + } + return 1; +} +__setup("isolcpus=", isolated_cpu_setup); + +struct s_data { + struct sched_domain ** __percpu sd; + struct root_domain *rd; +}; + +enum s_alloc { + sa_rootdomain, + sa_sd, + sa_sd_storage, + sa_none, +}; + +/* + * Return the canonical balance CPU for this group, this is the first CPU + * of this group that's also in the balance mask. + * + * The balance mask are all those CPUs that could actually end up at this + * group. See build_balance_mask(). + * + * Also see should_we_balance(). + */ +int group_balance_cpu(struct sched_group *sg) +{ + return cpumask_first(group_balance_mask(sg)); +} + + +/* + * NUMA topology (first read the regular topology blurb below) + * + * Given a node-distance table, for example: + * + * node 0 1 2 3 + * 0: 10 20 30 20 + * 1: 20 10 20 30 + * 2: 30 20 10 20 + * 3: 20 30 20 10 + * + * which represents a 4 node ring topology like: + * + * 0 ----- 1 + * | | + * | | + * | | + * 3 ----- 2 + * + * We want to construct domains and groups to represent this. The way we go + * about doing this is to build the domains on 'hops'. For each NUMA level we + * construct the mask of all nodes reachable in @level hops. + * + * For the above NUMA topology that gives 3 levels: + * + * NUMA-2 0-3 0-3 0-3 0-3 + * groups: {0-1,3},{1-3} {0-2},{0,2-3} {1-3},{0-1,3} {0,2-3},{0-2} + * + * NUMA-1 0-1,3 0-2 1-3 0,2-3 + * groups: {0},{1},{3} {0},{1},{2} {1},{2},{3} {0},{2},{3} + * + * NUMA-0 0 1 2 3 + * + * + * As can be seen; things don't nicely line up as with the regular topology. + * When we iterate a domain in child domain chunks some nodes can be + * represented multiple times -- hence the "overlap" naming for this part of + * the topology. + * + * In order to minimize this overlap, we only build enough groups to cover the + * domain. For instance Node-0 NUMA-2 would only get groups: 0-1,3 and 1-3. + * + * Because: + * + * - the first group of each domain is its child domain; this + * gets us the first 0-1,3 + * - the only uncovered node is 2, who's child domain is 1-3. + * + * However, because of the overlap, computing a unique CPU for each group is + * more complicated. Consider for instance the groups of NODE-1 NUMA-2, both + * groups include the CPUs of Node-0, while those CPUs would not in fact ever + * end up at those groups (they would end up in group: 0-1,3). + * + * To correct this we have to introduce the group balance mask. This mask + * will contain those CPUs in the group that can reach this group given the + * (child) domain tree. + * + * With this we can once again compute balance_cpu and sched_group_capacity + * relations. + * + * XXX include words on how balance_cpu is unique and therefore can be + * used for sched_group_capacity links. + * + * + * Another 'interesting' topology is: + * + * node 0 1 2 3 + * 0: 10 20 20 30 + * 1: 20 10 20 20 + * 2: 20 20 10 20 + * 3: 30 20 20 10 + * + * Which looks a little like: + * + * 0 ----- 1 + * | / | + * | / | + * | / | + * 2 ----- 3 + * + * This topology is asymmetric, nodes 1,2 are fully connected, but nodes 0,3 + * are not. + * + * This leads to a few particularly weird cases where the sched_domain's are + * not of the same number for each cpu. Consider: + * + * NUMA-2 0-3 0-3 + * groups: {0-2},{1-3} {1-3},{0-2} + * + * NUMA-1 0-2 0-3 0-3 1-3 + * + * NUMA-0 0 1 2 3 + * + */ + + +/* + * Build the balance mask; it contains only those CPUs that can arrive at this + * group and should be considered to continue balancing. + * + * We do this during the group creation pass, therefore the group information + * isn't complete yet, however since each group represents a (child) domain we + * can fully construct this using the sched_domain bits (which are already + * complete). + */ +static void +build_balance_mask(struct sched_domain *sd, struct sched_group *sg, struct cpumask *mask) +{ + const struct cpumask *sg_span = sched_group_span(sg); + struct sd_data *sdd = sd->private; + struct sched_domain *sibling; + int i; + + cpumask_clear(mask); + + for_each_cpu(i, sg_span) { + sibling = *per_cpu_ptr(sdd->sd, i); + + /* + * Can happen in the asymmetric case, where these siblings are + * unused. The mask will not be empty because those CPUs that + * do have the top domain _should_ span the domain. + */ + if (!sibling->child) + continue; + + /* If we would not end up here, we can't continue from here */ + if (!cpumask_equal(sg_span, sched_domain_span(sibling->child))) + continue; + + cpumask_set_cpu(i, mask); + } + + /* We must not have empty masks here */ + WARN_ON_ONCE(cpumask_empty(mask)); +} + +/* + * XXX: This creates per-node group entries; since the load-balancer will + * immediately access remote memory to construct this group's load-balance + * statistics having the groups node local is of dubious benefit. + */ +static struct sched_group * +build_group_from_child_sched_domain(struct sched_domain *sd, int cpu) +{ + struct sched_group *sg; + struct cpumask *sg_span; + + sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), + GFP_KERNEL, cpu_to_node(cpu)); + + if (!sg) + return NULL; + + sg_span = sched_group_span(sg); + if (sd->child) + cpumask_copy(sg_span, sched_domain_span(sd->child)); + else + cpumask_copy(sg_span, sched_domain_span(sd)); + + return sg; +} + +static void init_overlap_sched_group(struct sched_domain *sd, + struct sched_group *sg) +{ + struct cpumask *mask = sched_domains_tmpmask2; + struct sd_data *sdd = sd->private; + struct cpumask *sg_span; + int cpu; + + build_balance_mask(sd, sg, mask); + cpu = cpumask_first_and(sched_group_span(sg), mask); + + sg->sgc = *per_cpu_ptr(sdd->sgc, cpu); + if (atomic_inc_return(&sg->sgc->ref) == 1) + cpumask_copy(group_balance_mask(sg), mask); + else + WARN_ON_ONCE(!cpumask_equal(group_balance_mask(sg), mask)); + + /* + * Initialize sgc->capacity such that even if we mess up the + * domains and no possible iteration will get us here, we won't + * die on a /0 trap. + */ + sg_span = sched_group_span(sg); + sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sg_span); + sg->sgc->min_capacity = SCHED_CAPACITY_SCALE; +} + +static int +build_overlap_sched_groups(struct sched_domain *sd, int cpu) +{ + struct sched_group *first = NULL, *last = NULL, *sg; + const struct cpumask *span = sched_domain_span(sd); + struct cpumask *covered = sched_domains_tmpmask; + struct sd_data *sdd = sd->private; + struct sched_domain *sibling; + int i; + + cpumask_clear(covered); + + for_each_cpu_wrap(i, span, cpu) { + struct cpumask *sg_span; + + if (cpumask_test_cpu(i, covered)) + continue; + + sibling = *per_cpu_ptr(sdd->sd, i); + + /* + * Asymmetric node setups can result in situations where the + * domain tree is of unequal depth, make sure to skip domains + * that already cover the entire range. + * + * In that case build_sched_domains() will have terminated the + * iteration early and our sibling sd spans will be empty. + * Domains should always include the CPU they're built on, so + * check that. + */ + if (!cpumask_test_cpu(i, sched_domain_span(sibling))) + continue; + + sg = build_group_from_child_sched_domain(sibling, cpu); + if (!sg) + goto fail; + + sg_span = sched_group_span(sg); + cpumask_or(covered, covered, sg_span); + + init_overlap_sched_group(sd, sg); + + if (!first) + first = sg; + if (last) + last->next = sg; + last = sg; + last->next = first; + } + sd->groups = first; + + return 0; + +fail: + free_sched_groups(first, 0); + + return -ENOMEM; +} + + +/* + * Package topology (also see the load-balance blurb in fair.c) + * + * The scheduler builds a tree structure to represent a number of important + * topology features. By default (default_topology[]) these include: + * + * - Simultaneous multithreading (SMT) + * - Multi-Core Cache (MC) + * - Package (DIE) + * + * Where the last one more or less denotes everything up to a NUMA node. + * + * The tree consists of 3 primary data structures: + * + * sched_domain -> sched_group -> sched_group_capacity + * ^ ^ ^ ^ + * `-' `-' + * + * The sched_domains are per-cpu and have a two way link (parent & child) and + * denote the ever growing mask of CPUs belonging to that level of topology. + * + * Each sched_domain has a circular (double) linked list of sched_group's, each + * denoting the domains of the level below (or individual CPUs in case of the + * first domain level). The sched_group linked by a sched_domain includes the + * CPU of that sched_domain [*]. + * + * Take for instance a 2 threaded, 2 core, 2 cache cluster part: + * + * CPU 0 1 2 3 4 5 6 7 + * + * DIE [ ] + * MC [ ] [ ] + * SMT [ ] [ ] [ ] [ ] + * + * - or - + * + * DIE 0-7 0-7 0-7 0-7 0-7 0-7 0-7 0-7 + * MC 0-3 0-3 0-3 0-3 4-7 4-7 4-7 4-7 + * SMT 0-1 0-1 2-3 2-3 4-5 4-5 6-7 6-7 + * + * CPU 0 1 2 3 4 5 6 7 + * + * One way to think about it is: sched_domain moves you up and down among these + * topology levels, while sched_group moves you sideways through it, at child + * domain granularity. + * + * sched_group_capacity ensures each unique sched_group has shared storage. + * + * There are two related construction problems, both require a CPU that + * uniquely identify each group (for a given domain): + * + * - The first is the balance_cpu (see should_we_balance() and the + * load-balance blub in fair.c); for each group we only want 1 CPU to + * continue balancing at a higher domain. + * + * - The second is the sched_group_capacity; we want all identical groups + * to share a single sched_group_capacity. + * + * Since these topologies are exclusive by construction. That is, its + * impossible for an SMT thread to belong to multiple cores, and cores to + * be part of multiple caches. There is a very clear and unique location + * for each CPU in the hierarchy. + * + * Therefore computing a unique CPU for each group is trivial (the iteration + * mask is redundant and set all 1s; all CPUs in a group will end up at _that_ + * group), we can simply pick the first CPU in each group. + * + * + * [*] in other words, the first group of each domain is its child domain. + */ + +static struct sched_group *get_group(int cpu, struct sd_data *sdd) +{ + struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); + struct sched_domain *child = sd->child; + struct sched_group *sg; + + if (child) + cpu = cpumask_first(sched_domain_span(child)); + + sg = *per_cpu_ptr(sdd->sg, cpu); + sg->sgc = *per_cpu_ptr(sdd->sgc, cpu); + + /* For claim_allocations: */ + atomic_inc(&sg->ref); + atomic_inc(&sg->sgc->ref); + + if (child) { + cpumask_copy(sched_group_span(sg), sched_domain_span(child)); + cpumask_copy(group_balance_mask(sg), sched_group_span(sg)); + } else { + cpumask_set_cpu(cpu, sched_group_span(sg)); + cpumask_set_cpu(cpu, group_balance_mask(sg)); + } + + sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sched_group_span(sg)); + sg->sgc->min_capacity = SCHED_CAPACITY_SCALE; + + return sg; +} + +/* + * build_sched_groups will build a circular linked list of the groups + * covered by the given span, and will set each group's ->cpumask correctly, + * and ->cpu_capacity to 0. + * + * Assumes the sched_domain tree is fully constructed + */ +static int +build_sched_groups(struct sched_domain *sd, int cpu) +{ + struct sched_group *first = NULL, *last = NULL; + struct sd_data *sdd = sd->private; + const struct cpumask *span = sched_domain_span(sd); + struct cpumask *covered; + int i; + + lockdep_assert_held(&sched_domains_mutex); + covered = sched_domains_tmpmask; + + cpumask_clear(covered); + + for_each_cpu_wrap(i, span, cpu) { + struct sched_group *sg; + + if (cpumask_test_cpu(i, covered)) + continue; + + sg = get_group(i, sdd); + + cpumask_or(covered, covered, sched_group_span(sg)); + + if (!first) + first = sg; + if (last) + last->next = sg; + last = sg; + } + last->next = first; + sd->groups = first; + + return 0; +} + +/* + * Initialize sched groups cpu_capacity. + * + * cpu_capacity indicates the capacity of sched group, which is used while + * distributing the load between different sched groups in a sched domain. + * Typically cpu_capacity for all the groups in a sched domain will be same + * unless there are asymmetries in the topology. If there are asymmetries, + * group having more cpu_capacity will pickup more load compared to the + * group having less cpu_capacity. + */ +static void init_sched_groups_capacity(int cpu, struct sched_domain *sd) +{ + struct sched_group *sg = sd->groups; + + WARN_ON(!sg); + + do { + int cpu, max_cpu = -1; + + sg->group_weight = cpumask_weight(sched_group_span(sg)); + + if (!(sd->flags & SD_ASYM_PACKING)) + goto next; + + for_each_cpu(cpu, sched_group_span(sg)) { + if (max_cpu < 0) + max_cpu = cpu; + else if (sched_asym_prefer(cpu, max_cpu)) + max_cpu = cpu; + } + sg->asym_prefer_cpu = max_cpu; + +next: + sg = sg->next; + } while (sg != sd->groups); + + if (cpu != group_balance_cpu(sg)) + return; + + update_group_capacity(sd, cpu); +} + +/* + * Initializers for schedule domains + * Non-inlined to reduce accumulated stack pressure in build_sched_domains() + */ + +static int default_relax_domain_level = -1; +int sched_domain_level_max; + +static int __init setup_relax_domain_level(char *str) +{ + if (kstrtoint(str, 0, &default_relax_domain_level)) + pr_warn("Unable to set relax_domain_level\n"); + + return 1; +} +__setup("relax_domain_level=", setup_relax_domain_level); + +static void set_domain_attribute(struct sched_domain *sd, + struct sched_domain_attr *attr) +{ + int request; + + if (!attr || attr->relax_domain_level < 0) { + if (default_relax_domain_level < 0) + return; + else + request = default_relax_domain_level; + } else + request = attr->relax_domain_level; + if (request < sd->level) { + /* Turn off idle balance on this domain: */ + sd->flags &= ~(SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); + } else { + /* Turn on idle balance on this domain: */ + sd->flags |= (SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); + } +} + +static void __sdt_free(const struct cpumask *cpu_map); +static int __sdt_alloc(const struct cpumask *cpu_map); + +static void __free_domain_allocs(struct s_data *d, enum s_alloc what, + const struct cpumask *cpu_map) +{ + switch (what) { + case sa_rootdomain: + if (!atomic_read(&d->rd->refcount)) + free_rootdomain(&d->rd->rcu); + /* Fall through */ + case sa_sd: + free_percpu(d->sd); + /* Fall through */ + case sa_sd_storage: + __sdt_free(cpu_map); + /* Fall through */ + case sa_none: + break; + } +} + +static enum s_alloc +__visit_domain_allocation_hell(struct s_data *d, const struct cpumask *cpu_map) +{ + memset(d, 0, sizeof(*d)); + + if (__sdt_alloc(cpu_map)) + return sa_sd_storage; + d->sd = alloc_percpu(struct sched_domain *); + if (!d->sd) + return sa_sd_storage; + d->rd = alloc_rootdomain(); + if (!d->rd) + return sa_sd; + return sa_rootdomain; +} + +/* + * NULL the sd_data elements we've used to build the sched_domain and + * sched_group structure so that the subsequent __free_domain_allocs() + * will not free the data we're using. + */ +static void claim_allocations(int cpu, struct sched_domain *sd) +{ + struct sd_data *sdd = sd->private; + + WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd); + *per_cpu_ptr(sdd->sd, cpu) = NULL; + + if (atomic_read(&(*per_cpu_ptr(sdd->sds, cpu))->ref)) + *per_cpu_ptr(sdd->sds, cpu) = NULL; + + if (atomic_read(&(*per_cpu_ptr(sdd->sg, cpu))->ref)) + *per_cpu_ptr(sdd->sg, cpu) = NULL; + + if (atomic_read(&(*per_cpu_ptr(sdd->sgc, cpu))->ref)) + *per_cpu_ptr(sdd->sgc, cpu) = NULL; +} + +#ifdef CONFIG_NUMA +static int sched_domains_numa_levels; +enum numa_topology_type sched_numa_topology_type; +static int *sched_domains_numa_distance; +int sched_max_numa_distance; +static struct cpumask ***sched_domains_numa_masks; +static int sched_domains_curr_level; +#endif + +/* + * SD_flags allowed in topology descriptions. + * + * These flags are purely descriptive of the topology and do not prescribe + * behaviour. Behaviour is artificial and mapped in the below sd_init() + * function: + * + * SD_SHARE_CPUCAPACITY - describes SMT topologies + * SD_SHARE_PKG_RESOURCES - describes shared caches + * SD_NUMA - describes NUMA topologies + * SD_SHARE_POWERDOMAIN - describes shared power domain + * SD_ASYM_CPUCAPACITY - describes mixed capacity topologies + * + * Odd one out, which beside describing the topology has a quirk also + * prescribes the desired behaviour that goes along with it: + * + * SD_ASYM_PACKING - describes SMT quirks + */ +#define TOPOLOGY_SD_FLAGS \ + (SD_SHARE_CPUCAPACITY | \ + SD_SHARE_PKG_RESOURCES | \ + SD_NUMA | \ + SD_ASYM_PACKING | \ + SD_ASYM_CPUCAPACITY | \ + SD_SHARE_POWERDOMAIN) + +static struct sched_domain * +sd_init(struct sched_domain_topology_level *tl, + const struct cpumask *cpu_map, + struct sched_domain *child, int cpu) +{ + struct sd_data *sdd = &tl->data; + struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); + int sd_id, sd_weight, sd_flags = 0; + +#ifdef CONFIG_NUMA + /* + * Ugly hack to pass state to sd_numa_mask()... + */ + sched_domains_curr_level = tl->numa_level; +#endif + + sd_weight = cpumask_weight(tl->mask(cpu)); + + if (tl->sd_flags) + sd_flags = (*tl->sd_flags)(); + if (WARN_ONCE(sd_flags & ~TOPOLOGY_SD_FLAGS, + "wrong sd_flags in topology description\n")) + sd_flags &= ~TOPOLOGY_SD_FLAGS; + + *sd = (struct sched_domain){ + .min_interval = sd_weight, + .max_interval = 2*sd_weight, + .busy_factor = 32, + .imbalance_pct = 125, + + .cache_nice_tries = 0, + .busy_idx = 0, + .idle_idx = 0, + .newidle_idx = 0, + .wake_idx = 0, + .forkexec_idx = 0, + + .flags = 1*SD_LOAD_BALANCE + | 1*SD_BALANCE_NEWIDLE + | 1*SD_BALANCE_EXEC + | 1*SD_BALANCE_FORK + | 0*SD_BALANCE_WAKE + | 1*SD_WAKE_AFFINE + | 0*SD_SHARE_CPUCAPACITY + | 0*SD_SHARE_PKG_RESOURCES + | 0*SD_SERIALIZE + | 0*SD_PREFER_SIBLING + | 0*SD_NUMA + | sd_flags + , + + .last_balance = jiffies, + .balance_interval = sd_weight, + .smt_gain = 0, + .max_newidle_lb_cost = 0, + .next_decay_max_lb_cost = jiffies, + .child = child, +#ifdef CONFIG_SCHED_DEBUG + .name = tl->name, +#endif + }; + + cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu)); + sd_id = cpumask_first(sched_domain_span(sd)); + + /* + * Convert topological properties into behaviour. + */ + + if (sd->flags & SD_ASYM_CPUCAPACITY) { + struct sched_domain *t = sd; + + for_each_lower_domain(t) + t->flags |= SD_BALANCE_WAKE; + } + + if (sd->flags & SD_SHARE_CPUCAPACITY) { + sd->flags |= SD_PREFER_SIBLING; + sd->imbalance_pct = 110; + sd->smt_gain = 1178; /* ~15% */ + + } else if (sd->flags & SD_SHARE_PKG_RESOURCES) { + sd->imbalance_pct = 117; + sd->cache_nice_tries = 1; + sd->busy_idx = 2; + +#ifdef CONFIG_NUMA + } else if (sd->flags & SD_NUMA) { + sd->cache_nice_tries = 2; + sd->busy_idx = 3; + sd->idle_idx = 2; + + sd->flags |= SD_SERIALIZE; + if (sched_domains_numa_distance[tl->numa_level] > RECLAIM_DISTANCE) { + sd->flags &= ~(SD_BALANCE_EXEC | + SD_BALANCE_FORK | + SD_WAKE_AFFINE); + } + +#endif + } else { + sd->flags |= SD_PREFER_SIBLING; + sd->cache_nice_tries = 1; + sd->busy_idx = 2; + sd->idle_idx = 1; + } + + /* + * For all levels sharing cache; connect a sched_domain_shared + * instance. + */ + if (sd->flags & SD_SHARE_PKG_RESOURCES) { + sd->shared = *per_cpu_ptr(sdd->sds, sd_id); + atomic_inc(&sd->shared->ref); + atomic_set(&sd->shared->nr_busy_cpus, sd_weight); + } + + sd->private = sdd; + + return sd; +} + +/* + * Topology list, bottom-up. + */ +static struct sched_domain_topology_level default_topology[] = { +#ifdef CONFIG_SCHED_SMT + { cpu_smt_mask, cpu_smt_flags, SD_INIT_NAME(SMT) }, +#endif +#ifdef CONFIG_SCHED_MC + { cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) }, +#endif + { cpu_cpu_mask, SD_INIT_NAME(DIE) }, + { NULL, }, +}; + +static struct sched_domain_topology_level *sched_domain_topology = + default_topology; + +#define for_each_sd_topology(tl) \ + for (tl = sched_domain_topology; tl->mask; tl++) + +void set_sched_topology(struct sched_domain_topology_level *tl) +{ + if (WARN_ON_ONCE(sched_smp_initialized)) + return; + + sched_domain_topology = tl; +} + +#ifdef CONFIG_NUMA + +static const struct cpumask *sd_numa_mask(int cpu) +{ + return sched_domains_numa_masks[sched_domains_curr_level][cpu_to_node(cpu)]; +} + +static void sched_numa_warn(const char *str) +{ + static int done = false; + int i,j; + + if (done) + return; + + done = true; + + printk(KERN_WARNING "ERROR: %s\n\n", str); + + for (i = 0; i < nr_node_ids; i++) { + printk(KERN_WARNING " "); + for (j = 0; j < nr_node_ids; j++) + printk(KERN_CONT "%02d ", node_distance(i,j)); + printk(KERN_CONT "\n"); + } + printk(KERN_WARNING "\n"); +} + +bool find_numa_distance(int distance) +{ + int i; + + if (distance == node_distance(0, 0)) + return true; + + for (i = 0; i < sched_domains_numa_levels; i++) { + if (sched_domains_numa_distance[i] == distance) + return true; + } + + return false; +} + +/* + * A system can have three types of NUMA topology: + * NUMA_DIRECT: all nodes are directly connected, or not a NUMA system + * NUMA_GLUELESS_MESH: some nodes reachable through intermediary nodes + * NUMA_BACKPLANE: nodes can reach other nodes through a backplane + * + * The difference between a glueless mesh topology and a backplane + * topology lies in whether communication between not directly + * connected nodes goes through intermediary nodes (where programs + * could run), or through backplane controllers. This affects + * placement of programs. + * + * The type of topology can be discerned with the following tests: + * - If the maximum distance between any nodes is 1 hop, the system + * is directly connected. + * - If for two nodes A and B, located N > 1 hops away from each other, + * there is an intermediary node C, which is < N hops away from both + * nodes A and B, the system is a glueless mesh. + */ +static void init_numa_topology_type(void) +{ + int a, b, c, n; + + n = sched_max_numa_distance; + + if (sched_domains_numa_levels <= 1) { + sched_numa_topology_type = NUMA_DIRECT; + return; + } + + for_each_online_node(a) { + for_each_online_node(b) { + /* Find two nodes furthest removed from each other. */ + if (node_distance(a, b) < n) + continue; + + /* Is there an intermediary node between a and b? */ + for_each_online_node(c) { + if (node_distance(a, c) < n && + node_distance(b, c) < n) { + sched_numa_topology_type = + NUMA_GLUELESS_MESH; + return; + } + } + + sched_numa_topology_type = NUMA_BACKPLANE; + return; + } + } +} + +void sched_init_numa(void) +{ + int next_distance, curr_distance = node_distance(0, 0); + struct sched_domain_topology_level *tl; + int level = 0; + int i, j, k; + + sched_domains_numa_distance = kzalloc(sizeof(int) * nr_node_ids, GFP_KERNEL); + if (!sched_domains_numa_distance) + return; + + /* + * O(nr_nodes^2) deduplicating selection sort -- in order to find the + * unique distances in the node_distance() table. + * + * Assumes node_distance(0,j) includes all distances in + * node_distance(i,j) in order to avoid cubic time. + */ + next_distance = curr_distance; + for (i = 0; i < nr_node_ids; i++) { + for (j = 0; j < nr_node_ids; j++) { + for (k = 0; k < nr_node_ids; k++) { + int distance = node_distance(i, k); + + if (distance > curr_distance && + (distance < next_distance || + next_distance == curr_distance)) + next_distance = distance; + + /* + * While not a strong assumption it would be nice to know + * about cases where if node A is connected to B, B is not + * equally connected to A. + */ + if (sched_debug() && node_distance(k, i) != distance) + sched_numa_warn("Node-distance not symmetric"); + + if (sched_debug() && i && !find_numa_distance(distance)) + sched_numa_warn("Node-0 not representative"); + } + if (next_distance != curr_distance) { + sched_domains_numa_distance[level++] = next_distance; + sched_domains_numa_levels = level; + curr_distance = next_distance; + } else break; + } + + /* + * In case of sched_debug() we verify the above assumption. + */ + if (!sched_debug()) + break; + } + + if (!level) + return; + + /* + * 'level' contains the number of unique distances, excluding the + * identity distance node_distance(i,i). + * + * The sched_domains_numa_distance[] array includes the actual distance + * numbers. + */ + + /* + * Here, we should temporarily reset sched_domains_numa_levels to 0. + * If it fails to allocate memory for array sched_domains_numa_masks[][], + * the array will contain less then 'level' members. This could be + * dangerous when we use it to iterate array sched_domains_numa_masks[][] + * in other functions. + * + * We reset it to 'level' at the end of this function. + */ + sched_domains_numa_levels = 0; + + sched_domains_numa_masks = kzalloc(sizeof(void *) * level, GFP_KERNEL); + if (!sched_domains_numa_masks) + return; + + /* + * Now for each level, construct a mask per node which contains all + * CPUs of nodes that are that many hops away from us. + */ + for (i = 0; i < level; i++) { + sched_domains_numa_masks[i] = + kzalloc(nr_node_ids * sizeof(void *), GFP_KERNEL); + if (!sched_domains_numa_masks[i]) + return; + + for (j = 0; j < nr_node_ids; j++) { + struct cpumask *mask = kzalloc(cpumask_size(), GFP_KERNEL); + if (!mask) + return; + + sched_domains_numa_masks[i][j] = mask; + + for_each_node(k) { + if (node_distance(j, k) > sched_domains_numa_distance[i]) + continue; + + cpumask_or(mask, mask, cpumask_of_node(k)); + } + } + } + + /* Compute default topology size */ + for (i = 0; sched_domain_topology[i].mask; i++); + + tl = kzalloc((i + level + 1) * + sizeof(struct sched_domain_topology_level), GFP_KERNEL); + if (!tl) + return; + + /* + * Copy the default topology bits.. + */ + for (i = 0; sched_domain_topology[i].mask; i++) + tl[i] = sched_domain_topology[i]; + + /* + * .. and append 'j' levels of NUMA goodness. + */ + for (j = 0; j < level; i++, j++) { + tl[i] = (struct sched_domain_topology_level){ + .mask = sd_numa_mask, + .sd_flags = cpu_numa_flags, + .flags = SDTL_OVERLAP, + .numa_level = j, + SD_INIT_NAME(NUMA) + }; + } + + sched_domain_topology = tl; + + sched_domains_numa_levels = level; + sched_max_numa_distance = sched_domains_numa_distance[level - 1]; + + init_numa_topology_type(); +} + +void sched_domains_numa_masks_set(unsigned int cpu) +{ + int node = cpu_to_node(cpu); + int i, j; + + for (i = 0; i < sched_domains_numa_levels; i++) { + for (j = 0; j < nr_node_ids; j++) { + if (node_distance(j, node) <= sched_domains_numa_distance[i]) + cpumask_set_cpu(cpu, sched_domains_numa_masks[i][j]); + } + } +} + +void sched_domains_numa_masks_clear(unsigned int cpu) +{ + int i, j; + + for (i = 0; i < sched_domains_numa_levels; i++) { + for (j = 0; j < nr_node_ids; j++) + cpumask_clear_cpu(cpu, sched_domains_numa_masks[i][j]); + } +} + +#endif /* CONFIG_NUMA */ + +static int __sdt_alloc(const struct cpumask *cpu_map) +{ + struct sched_domain_topology_level *tl; + int j; + + for_each_sd_topology(tl) { + struct sd_data *sdd = &tl->data; + + sdd->sd = alloc_percpu(struct sched_domain *); + if (!sdd->sd) + return -ENOMEM; + + sdd->sds = alloc_percpu(struct sched_domain_shared *); + if (!sdd->sds) + return -ENOMEM; + + sdd->sg = alloc_percpu(struct sched_group *); + if (!sdd->sg) + return -ENOMEM; + + sdd->sgc = alloc_percpu(struct sched_group_capacity *); + if (!sdd->sgc) + return -ENOMEM; + + for_each_cpu(j, cpu_map) { + struct sched_domain *sd; + struct sched_domain_shared *sds; + struct sched_group *sg; + struct sched_group_capacity *sgc; + + sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(), + GFP_KERNEL, cpu_to_node(j)); + if (!sd) + return -ENOMEM; + + *per_cpu_ptr(sdd->sd, j) = sd; + + sds = kzalloc_node(sizeof(struct sched_domain_shared), + GFP_KERNEL, cpu_to_node(j)); + if (!sds) + return -ENOMEM; + + *per_cpu_ptr(sdd->sds, j) = sds; + + sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), + GFP_KERNEL, cpu_to_node(j)); + if (!sg) + return -ENOMEM; + + sg->next = sg; + + *per_cpu_ptr(sdd->sg, j) = sg; + + sgc = kzalloc_node(sizeof(struct sched_group_capacity) + cpumask_size(), + GFP_KERNEL, cpu_to_node(j)); + if (!sgc) + return -ENOMEM; + +#ifdef CONFIG_SCHED_DEBUG + sgc->id = j; +#endif + + *per_cpu_ptr(sdd->sgc, j) = sgc; + } + } + + return 0; +} + +static void __sdt_free(const struct cpumask *cpu_map) +{ + struct sched_domain_topology_level *tl; + int j; + + for_each_sd_topology(tl) { + struct sd_data *sdd = &tl->data; + + for_each_cpu(j, cpu_map) { + struct sched_domain *sd; + + if (sdd->sd) { + sd = *per_cpu_ptr(sdd->sd, j); + if (sd && (sd->flags & SD_OVERLAP)) + free_sched_groups(sd->groups, 0); + kfree(*per_cpu_ptr(sdd->sd, j)); + } + + if (sdd->sds) + kfree(*per_cpu_ptr(sdd->sds, j)); + if (sdd->sg) + kfree(*per_cpu_ptr(sdd->sg, j)); + if (sdd->sgc) + kfree(*per_cpu_ptr(sdd->sgc, j)); + } + free_percpu(sdd->sd); + sdd->sd = NULL; + free_percpu(sdd->sds); + sdd->sds = NULL; + free_percpu(sdd->sg); + sdd->sg = NULL; + free_percpu(sdd->sgc); + sdd->sgc = NULL; + } +} + +struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl, + const struct cpumask *cpu_map, struct sched_domain_attr *attr, + struct sched_domain *child, int cpu) +{ + struct sched_domain *sd = sd_init(tl, cpu_map, child, cpu); + + if (child) { + sd->level = child->level + 1; + sched_domain_level_max = max(sched_domain_level_max, sd->level); + child->parent = sd; + + if (!cpumask_subset(sched_domain_span(child), + sched_domain_span(sd))) { + pr_err("BUG: arch topology borken\n"); +#ifdef CONFIG_SCHED_DEBUG + pr_err(" the %s domain not a subset of the %s domain\n", + child->name, sd->name); +#endif + /* Fixup, ensure @sd has at least @child cpus. */ + cpumask_or(sched_domain_span(sd), + sched_domain_span(sd), + sched_domain_span(child)); + } + + } + set_domain_attribute(sd, attr); + + return sd; +} + +/* + * Build sched domains for a given set of CPUs and attach the sched domains + * to the individual CPUs + */ +static int +build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *attr) +{ + enum s_alloc alloc_state; + struct sched_domain *sd; + struct s_data d; + struct rq *rq = NULL; + int i, ret = -ENOMEM; + + alloc_state = __visit_domain_allocation_hell(&d, cpu_map); + if (alloc_state != sa_rootdomain) + goto error; + + /* Set up domains for CPUs specified by the cpu_map: */ + for_each_cpu(i, cpu_map) { + struct sched_domain_topology_level *tl; + + sd = NULL; + for_each_sd_topology(tl) { + sd = build_sched_domain(tl, cpu_map, attr, sd, i); + if (tl == sched_domain_topology) + *per_cpu_ptr(d.sd, i) = sd; + if (tl->flags & SDTL_OVERLAP) + sd->flags |= SD_OVERLAP; + if (cpumask_equal(cpu_map, sched_domain_span(sd))) + break; + } + } + + /* Build the groups for the domains */ + for_each_cpu(i, cpu_map) { + for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { + sd->span_weight = cpumask_weight(sched_domain_span(sd)); + if (sd->flags & SD_OVERLAP) { + if (build_overlap_sched_groups(sd, i)) + goto error; + } else { + if (build_sched_groups(sd, i)) + goto error; + } + } + } + + /* Calculate CPU capacity for physical packages and nodes */ + for (i = nr_cpumask_bits-1; i >= 0; i--) { + if (!cpumask_test_cpu(i, cpu_map)) + continue; + + for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { + claim_allocations(i, sd); + init_sched_groups_capacity(i, sd); + } + } + + /* Attach the domains */ + rcu_read_lock(); + for_each_cpu(i, cpu_map) { + rq = cpu_rq(i); + sd = *per_cpu_ptr(d.sd, i); + + /* Use READ_ONCE()/WRITE_ONCE() to avoid load/store tearing: */ + if (rq->cpu_capacity_orig > READ_ONCE(d.rd->max_cpu_capacity)) + WRITE_ONCE(d.rd->max_cpu_capacity, rq->cpu_capacity_orig); + + cpu_attach_domain(sd, d.rd, i); + } + rcu_read_unlock(); + + if (rq && sched_debug_enabled) { + pr_info("span: %*pbl (max cpu_capacity = %lu)\n", + cpumask_pr_args(cpu_map), rq->rd->max_cpu_capacity); + } + + ret = 0; +error: + __free_domain_allocs(&d, alloc_state, cpu_map); + return ret; +} + +/* Current sched domains: */ +static cpumask_var_t *doms_cur; + +/* Number of sched domains in 'doms_cur': */ +static int ndoms_cur; + +/* Attribues of custom domains in 'doms_cur' */ +static struct sched_domain_attr *dattr_cur; + +/* + * Special case: If a kmalloc() of a doms_cur partition (array of + * cpumask) fails, then fallback to a single sched domain, + * as determined by the single cpumask fallback_doms. + */ +static cpumask_var_t fallback_doms; + +/* + * arch_update_cpu_topology lets virtualized architectures update the + * CPU core maps. It is supposed to return 1 if the topology changed + * or 0 if it stayed the same. + */ +int __weak arch_update_cpu_topology(void) +{ + return 0; +} + +cpumask_var_t *alloc_sched_domains(unsigned int ndoms) +{ + int i; + cpumask_var_t *doms; + + doms = kmalloc(sizeof(*doms) * ndoms, GFP_KERNEL); + if (!doms) + return NULL; + for (i = 0; i < ndoms; i++) { + if (!alloc_cpumask_var(&doms[i], GFP_KERNEL)) { + free_sched_domains(doms, i); + return NULL; + } + } + return doms; +} + +void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms) +{ + unsigned int i; + for (i = 0; i < ndoms; i++) + free_cpumask_var(doms[i]); + kfree(doms); +} + +/* + * Set up scheduler domains and groups. Callers must hold the hotplug lock. + * For now this just excludes isolated CPUs, but could be used to + * exclude other special cases in the future. + */ +int sched_init_domains(const struct cpumask *cpu_map) +{ + int err; + + zalloc_cpumask_var(&sched_domains_tmpmask, GFP_KERNEL); + zalloc_cpumask_var(&sched_domains_tmpmask2, GFP_KERNEL); + zalloc_cpumask_var(&fallback_doms, GFP_KERNEL); + + arch_update_cpu_topology(); + ndoms_cur = 1; + doms_cur = alloc_sched_domains(ndoms_cur); + if (!doms_cur) + doms_cur = &fallback_doms; + cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map); + err = build_sched_domains(doms_cur[0], NULL); + register_sched_domain_sysctl(); + + return err; +} + +/* + * Detach sched domains from a group of CPUs specified in cpu_map + * These CPUs will now be attached to the NULL domain + */ +static void detach_destroy_domains(const struct cpumask *cpu_map) +{ + int i; + + rcu_read_lock(); + for_each_cpu(i, cpu_map) + cpu_attach_domain(NULL, &def_root_domain, i); + rcu_read_unlock(); +} + +/* handle null as "default" */ +static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, + struct sched_domain_attr *new, int idx_new) +{ + struct sched_domain_attr tmp; + + /* Fast path: */ + if (!new && !cur) + return 1; + + tmp = SD_ATTR_INIT; + return !memcmp(cur ? (cur + idx_cur) : &tmp, + new ? (new + idx_new) : &tmp, + sizeof(struct sched_domain_attr)); +} + +/* + * Partition sched domains as specified by the 'ndoms_new' + * cpumasks in the array doms_new[] of cpumasks. This compares + * doms_new[] to the current sched domain partitioning, doms_cur[]. + * It destroys each deleted domain and builds each new domain. + * + * 'doms_new' is an array of cpumask_var_t's of length 'ndoms_new'. + * The masks don't intersect (don't overlap.) We should setup one + * sched domain for each mask. CPUs not in any of the cpumasks will + * not be load balanced. If the same cpumask appears both in the + * current 'doms_cur' domains and in the new 'doms_new', we can leave + * it as it is. + * + * The passed in 'doms_new' should be allocated using + * alloc_sched_domains. This routine takes ownership of it and will + * free_sched_domains it when done with it. If the caller failed the + * alloc call, then it can pass in doms_new == NULL && ndoms_new == 1, + * and partition_sched_domains() will fallback to the single partition + * 'fallback_doms', it also forces the domains to be rebuilt. + * + * If doms_new == NULL it will be replaced with cpu_online_mask. + * ndoms_new == 0 is a special case for destroying existing domains, + * and it will not create the default domain. + * + * Call with hotplug lock held + */ +void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], + struct sched_domain_attr *dattr_new) +{ + int i, j, n; + int new_topology; + + mutex_lock(&sched_domains_mutex); + + /* Always unregister in case we don't destroy any domains: */ + unregister_sched_domain_sysctl(); + + /* Let the architecture update CPU core mappings: */ + new_topology = arch_update_cpu_topology(); + + n = doms_new ? ndoms_new : 0; + + /* Destroy deleted domains: */ + for (i = 0; i < ndoms_cur; i++) { + for (j = 0; j < n && !new_topology; j++) { + if (cpumask_equal(doms_cur[i], doms_new[j]) + && dattrs_equal(dattr_cur, i, dattr_new, j)) + goto match1; + } + /* No match - a current sched domain not in new doms_new[] */ + detach_destroy_domains(doms_cur[i]); +match1: + ; + } + + n = ndoms_cur; + if (doms_new == NULL) { + n = 0; + doms_new = &fallback_doms; + cpumask_andnot(doms_new[0], cpu_active_mask, cpu_isolated_map); + WARN_ON_ONCE(dattr_new); + } + + /* Build new domains: */ + for (i = 0; i < ndoms_new; i++) { + for (j = 0; j < n && !new_topology; j++) { + if (cpumask_equal(doms_new[i], doms_cur[j]) + && dattrs_equal(dattr_new, i, dattr_cur, j)) + goto match2; + } + /* No match - add a new doms_new */ + build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL); +match2: + ; + } + + /* Remember the new sched domains: */ + if (doms_cur != &fallback_doms) + free_sched_domains(doms_cur, ndoms_cur); + + kfree(dattr_cur); + doms_cur = doms_new; + dattr_cur = dattr_new; + ndoms_cur = ndoms_new; + + register_sched_domain_sysctl(); + + mutex_unlock(&sched_domains_mutex); +} + diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 9453efe9b25a..17f11c6b0a9f 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -5,50 +5,51 @@ */ #include <linux/init.h> #include <linux/export.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> #include <linux/mm.h> #include <linux/wait.h> #include <linux/hash.h> #include <linux/kthread.h> -void __init_waitqueue_head(wait_queue_head_t *q, const char *name, struct lock_class_key *key) +void __init_waitqueue_head(struct wait_queue_head *wq_head, const char *name, struct lock_class_key *key) { - spin_lock_init(&q->lock); - lockdep_set_class_and_name(&q->lock, key, name); - INIT_LIST_HEAD(&q->task_list); + spin_lock_init(&wq_head->lock); + lockdep_set_class_and_name(&wq_head->lock, key, name); + INIT_LIST_HEAD(&wq_head->head); } EXPORT_SYMBOL(__init_waitqueue_head); -void add_wait_queue(wait_queue_head_t *q, wait_queue_t *wait) +void add_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry) { unsigned long flags; - wait->flags &= ~WQ_FLAG_EXCLUSIVE; - spin_lock_irqsave(&q->lock, flags); - __add_wait_queue(q, wait); - spin_unlock_irqrestore(&q->lock, flags); + wq_entry->flags &= ~WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&wq_head->lock, flags); + __add_wait_queue_entry_tail(wq_head, wq_entry); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL(add_wait_queue); -void add_wait_queue_exclusive(wait_queue_head_t *q, wait_queue_t *wait) +void add_wait_queue_exclusive(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry) { unsigned long flags; - wait->flags |= WQ_FLAG_EXCLUSIVE; - spin_lock_irqsave(&q->lock, flags); - __add_wait_queue_tail(q, wait); - spin_unlock_irqrestore(&q->lock, flags); + wq_entry->flags |= WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&wq_head->lock, flags); + __add_wait_queue_entry_tail(wq_head, wq_entry); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL(add_wait_queue_exclusive); -void remove_wait_queue(wait_queue_head_t *q, wait_queue_t *wait) +void remove_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry) { unsigned long flags; - spin_lock_irqsave(&q->lock, flags); - __remove_wait_queue(q, wait); - spin_unlock_irqrestore(&q->lock, flags); + spin_lock_irqsave(&wq_head->lock, flags); + __remove_wait_queue(wq_head, wq_entry); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL(remove_wait_queue); @@ -62,12 +63,12 @@ EXPORT_SYMBOL(remove_wait_queue); * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns * zero in this (rare) case, and we handle it by continuing to scan the queue. */ -static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, +static void __wake_up_common(struct wait_queue_head *wq_head, unsigned int mode, int nr_exclusive, int wake_flags, void *key) { - wait_queue_t *curr, *next; + wait_queue_entry_t *curr, *next; - list_for_each_entry_safe(curr, next, &q->task_list, task_list) { + list_for_each_entry_safe(curr, next, &wq_head->head, entry) { unsigned flags = curr->flags; if (curr->func(curr, mode, wake_flags, key) && @@ -78,7 +79,7 @@ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, /** * __wake_up - wake up threads blocked on a waitqueue. - * @q: the waitqueue + * @wq_head: the waitqueue * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up * @key: is directly passed to the wakeup function @@ -86,35 +87,35 @@ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, * It may be assumed that this function implies a write memory barrier before * changing the task state if and only if any tasks are woken up. */ -void __wake_up(wait_queue_head_t *q, unsigned int mode, +void __wake_up(struct wait_queue_head *wq_head, unsigned int mode, int nr_exclusive, void *key) { unsigned long flags; - spin_lock_irqsave(&q->lock, flags); - __wake_up_common(q, mode, nr_exclusive, 0, key); - spin_unlock_irqrestore(&q->lock, flags); + spin_lock_irqsave(&wq_head->lock, flags); + __wake_up_common(wq_head, mode, nr_exclusive, 0, key); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL(__wake_up); /* * Same as __wake_up but called with the spinlock in wait_queue_head_t held. */ -void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr) +void __wake_up_locked(struct wait_queue_head *wq_head, unsigned int mode, int nr) { - __wake_up_common(q, mode, nr, 0, NULL); + __wake_up_common(wq_head, mode, nr, 0, NULL); } EXPORT_SYMBOL_GPL(__wake_up_locked); -void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key) +void __wake_up_locked_key(struct wait_queue_head *wq_head, unsigned int mode, void *key) { - __wake_up_common(q, mode, 1, 0, key); + __wake_up_common(wq_head, mode, 1, 0, key); } EXPORT_SYMBOL_GPL(__wake_up_locked_key); /** * __wake_up_sync_key - wake up threads blocked on a waitqueue. - * @q: the waitqueue + * @wq_head: the waitqueue * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up * @key: opaque value to be passed to wakeup targets @@ -129,30 +130,30 @@ EXPORT_SYMBOL_GPL(__wake_up_locked_key); * It may be assumed that this function implies a write memory barrier before * changing the task state if and only if any tasks are woken up. */ -void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, +void __wake_up_sync_key(struct wait_queue_head *wq_head, unsigned int mode, int nr_exclusive, void *key) { unsigned long flags; int wake_flags = 1; /* XXX WF_SYNC */ - if (unlikely(!q)) + if (unlikely(!wq_head)) return; if (unlikely(nr_exclusive != 1)) wake_flags = 0; - spin_lock_irqsave(&q->lock, flags); - __wake_up_common(q, mode, nr_exclusive, wake_flags, key); - spin_unlock_irqrestore(&q->lock, flags); + spin_lock_irqsave(&wq_head->lock, flags); + __wake_up_common(wq_head, mode, nr_exclusive, wake_flags, key); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL_GPL(__wake_up_sync_key); /* * __wake_up_sync - see __wake_up_sync_key() */ -void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive) +void __wake_up_sync(struct wait_queue_head *wq_head, unsigned int mode, int nr_exclusive) { - __wake_up_sync_key(q, mode, nr_exclusive, NULL); + __wake_up_sync_key(wq_head, mode, nr_exclusive, NULL); } EXPORT_SYMBOL_GPL(__wake_up_sync); /* For internal use only */ @@ -169,48 +170,48 @@ EXPORT_SYMBOL_GPL(__wake_up_sync); /* For internal use only */ * loads to move into the critical region). */ void -prepare_to_wait(wait_queue_head_t *q, wait_queue_t *wait, int state) +prepare_to_wait(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state) { unsigned long flags; - wait->flags &= ~WQ_FLAG_EXCLUSIVE; - spin_lock_irqsave(&q->lock, flags); - if (list_empty(&wait->task_list)) - __add_wait_queue(q, wait); + wq_entry->flags &= ~WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&wq_head->lock, flags); + if (list_empty(&wq_entry->entry)) + __add_wait_queue(wq_head, wq_entry); set_current_state(state); - spin_unlock_irqrestore(&q->lock, flags); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL(prepare_to_wait); void -prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state) +prepare_to_wait_exclusive(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state) { unsigned long flags; - wait->flags |= WQ_FLAG_EXCLUSIVE; - spin_lock_irqsave(&q->lock, flags); - if (list_empty(&wait->task_list)) - __add_wait_queue_tail(q, wait); + wq_entry->flags |= WQ_FLAG_EXCLUSIVE; + spin_lock_irqsave(&wq_head->lock, flags); + if (list_empty(&wq_entry->entry)) + __add_wait_queue_entry_tail(wq_head, wq_entry); set_current_state(state); - spin_unlock_irqrestore(&q->lock, flags); + spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL(prepare_to_wait_exclusive); -void init_wait_entry(wait_queue_t *wait, int flags) +void init_wait_entry(struct wait_queue_entry *wq_entry, int flags) { - wait->flags = flags; - wait->private = current; - wait->func = autoremove_wake_function; - INIT_LIST_HEAD(&wait->task_list); + wq_entry->flags = flags; + wq_entry->private = current; + wq_entry->func = autoremove_wake_function; + INIT_LIST_HEAD(&wq_entry->entry); } EXPORT_SYMBOL(init_wait_entry); -long prepare_to_wait_event(wait_queue_head_t *q, wait_queue_t *wait, int state) +long prepare_to_wait_event(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry, int state) { unsigned long flags; long ret = 0; - spin_lock_irqsave(&q->lock, flags); + spin_lock_irqsave(&wq_head->lock, flags); if (unlikely(signal_pending_state(state, current))) { /* * Exclusive waiter must not fail if it was selected by wakeup, @@ -218,39 +219,78 @@ long prepare_to_wait_event(wait_queue_head_t *q, wait_queue_t *wait, int state) * * The caller will recheck the condition and return success if * we were already woken up, we can not miss the event because - * wakeup locks/unlocks the same q->lock. + * wakeup locks/unlocks the same wq_head->lock. * * But we need to ensure that set-condition + wakeup after that * can't see us, it should wake up another exclusive waiter if * we fail. */ - list_del_init(&wait->task_list); + list_del_init(&wq_entry->entry); ret = -ERESTARTSYS; } else { - if (list_empty(&wait->task_list)) { - if (wait->flags & WQ_FLAG_EXCLUSIVE) - __add_wait_queue_tail(q, wait); + if (list_empty(&wq_entry->entry)) { + if (wq_entry->flags & WQ_FLAG_EXCLUSIVE) + __add_wait_queue_entry_tail(wq_head, wq_entry); else - __add_wait_queue(q, wait); + __add_wait_queue(wq_head, wq_entry); } set_current_state(state); } - spin_unlock_irqrestore(&q->lock, flags); + spin_unlock_irqrestore(&wq_head->lock, flags); return ret; } EXPORT_SYMBOL(prepare_to_wait_event); +/* + * Note! These two wait functions are entered with the + * wait-queue lock held (and interrupts off in the _irq + * case), so there is no race with testing the wakeup + * condition in the caller before they add the wait + * entry to the wake queue. + */ +int do_wait_intr(wait_queue_head_t *wq, wait_queue_entry_t *wait) +{ + if (likely(list_empty(&wait->entry))) + __add_wait_queue_entry_tail(wq, wait); + + set_current_state(TASK_INTERRUPTIBLE); + if (signal_pending(current)) + return -ERESTARTSYS; + + spin_unlock(&wq->lock); + schedule(); + spin_lock(&wq->lock); + return 0; +} +EXPORT_SYMBOL(do_wait_intr); + +int do_wait_intr_irq(wait_queue_head_t *wq, wait_queue_entry_t *wait) +{ + if (likely(list_empty(&wait->entry))) + __add_wait_queue_entry_tail(wq, wait); + + set_current_state(TASK_INTERRUPTIBLE); + if (signal_pending(current)) + return -ERESTARTSYS; + + spin_unlock_irq(&wq->lock); + schedule(); + spin_lock_irq(&wq->lock); + return 0; +} +EXPORT_SYMBOL(do_wait_intr_irq); + /** * finish_wait - clean up after waiting in a queue - * @q: waitqueue waited on - * @wait: wait descriptor + * @wq_head: waitqueue waited on + * @wq_entry: wait descriptor * * Sets current thread back to running state and removes * the wait descriptor from the given waitqueue if still * queued. */ -void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) +void finish_wait(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry) { unsigned long flags; @@ -268,20 +308,20 @@ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait) * have _one_ other CPU that looks at or modifies * the list). */ - if (!list_empty_careful(&wait->task_list)) { - spin_lock_irqsave(&q->lock, flags); - list_del_init(&wait->task_list); - spin_unlock_irqrestore(&q->lock, flags); + if (!list_empty_careful(&wq_entry->entry)) { + spin_lock_irqsave(&wq_head->lock, flags); + list_del_init(&wq_entry->entry); + spin_unlock_irqrestore(&wq_head->lock, flags); } } EXPORT_SYMBOL(finish_wait); -int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) +int autoremove_wake_function(struct wait_queue_entry *wq_entry, unsigned mode, int sync, void *key) { - int ret = default_wake_function(wait, mode, sync, key); + int ret = default_wake_function(wq_entry, mode, sync, key); if (ret) - list_del_init(&wait->task_list); + list_del_init(&wq_entry->entry); return ret; } EXPORT_SYMBOL(autoremove_wake_function); @@ -294,24 +334,24 @@ static inline bool is_kthread_should_stop(void) /* * DEFINE_WAIT_FUNC(wait, woken_wake_func); * - * add_wait_queue(&wq, &wait); + * add_wait_queue(&wq_head, &wait); * for (;;) { * if (condition) * break; * * p->state = mode; condition = true; * smp_mb(); // A smp_wmb(); // C - * if (!wait->flags & WQ_FLAG_WOKEN) wait->flags |= WQ_FLAG_WOKEN; + * if (!wq_entry->flags & WQ_FLAG_WOKEN) wq_entry->flags |= WQ_FLAG_WOKEN; * schedule() try_to_wake_up(); * p->state = TASK_RUNNING; ~~~~~~~~~~~~~~~~~~ - * wait->flags &= ~WQ_FLAG_WOKEN; condition = true; + * wq_entry->flags &= ~WQ_FLAG_WOKEN; condition = true; * smp_mb() // B smp_wmb(); // C - * wait->flags |= WQ_FLAG_WOKEN; + * wq_entry->flags |= WQ_FLAG_WOKEN; * } - * remove_wait_queue(&wq, &wait); + * remove_wait_queue(&wq_head, &wait); * */ -long wait_woken(wait_queue_t *wait, unsigned mode, long timeout) +long wait_woken(struct wait_queue_entry *wq_entry, unsigned mode, long timeout) { set_current_state(mode); /* A */ /* @@ -319,7 +359,7 @@ long wait_woken(wait_queue_t *wait, unsigned mode, long timeout) * woken_wake_function() such that if we observe WQ_FLAG_WOKEN we must * also observe all state before the wakeup. */ - if (!(wait->flags & WQ_FLAG_WOKEN) && !is_kthread_should_stop()) + if (!(wq_entry->flags & WQ_FLAG_WOKEN) && !is_kthread_should_stop()) timeout = schedule_timeout(timeout); __set_current_state(TASK_RUNNING); @@ -329,13 +369,13 @@ long wait_woken(wait_queue_t *wait, unsigned mode, long timeout) * condition being true _OR_ WQ_FLAG_WOKEN such that we will not miss * an event. */ - smp_store_mb(wait->flags, wait->flags & ~WQ_FLAG_WOKEN); /* B */ + smp_store_mb(wq_entry->flags, wq_entry->flags & ~WQ_FLAG_WOKEN); /* B */ return timeout; } EXPORT_SYMBOL(wait_woken); -int woken_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) +int woken_wake_function(struct wait_queue_entry *wq_entry, unsigned mode, int sync, void *key) { /* * Although this function is called under waitqueue lock, LOCK @@ -345,267 +385,8 @@ int woken_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key) * and is paired with smp_store_mb() in wait_woken(). */ smp_wmb(); /* C */ - wait->flags |= WQ_FLAG_WOKEN; + wq_entry->flags |= WQ_FLAG_WOKEN; - return default_wake_function(wait, mode, sync, key); + return default_wake_function(wq_entry, mode, sync, key); } EXPORT_SYMBOL(woken_wake_function); - -int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *arg) -{ - struct wait_bit_key *key = arg; - struct wait_bit_queue *wait_bit - = container_of(wait, struct wait_bit_queue, wait); - - if (wait_bit->key.flags != key->flags || - wait_bit->key.bit_nr != key->bit_nr || - test_bit(key->bit_nr, key->flags)) - return 0; - else - return autoremove_wake_function(wait, mode, sync, key); -} -EXPORT_SYMBOL(wake_bit_function); - -/* - * To allow interruptible waiting and asynchronous (i.e. nonblocking) - * waiting, the actions of __wait_on_bit() and __wait_on_bit_lock() are - * permitted return codes. Nonzero return codes halt waiting and return. - */ -int __sched -__wait_on_bit(wait_queue_head_t *wq, struct wait_bit_queue *q, - wait_bit_action_f *action, unsigned mode) -{ - int ret = 0; - - do { - prepare_to_wait(wq, &q->wait, mode); - if (test_bit(q->key.bit_nr, q->key.flags)) - ret = (*action)(&q->key, mode); - } while (test_bit(q->key.bit_nr, q->key.flags) && !ret); - finish_wait(wq, &q->wait); - return ret; -} -EXPORT_SYMBOL(__wait_on_bit); - -int __sched out_of_line_wait_on_bit(void *word, int bit, - wait_bit_action_f *action, unsigned mode) -{ - wait_queue_head_t *wq = bit_waitqueue(word, bit); - DEFINE_WAIT_BIT(wait, word, bit); - - return __wait_on_bit(wq, &wait, action, mode); -} -EXPORT_SYMBOL(out_of_line_wait_on_bit); - -int __sched out_of_line_wait_on_bit_timeout( - void *word, int bit, wait_bit_action_f *action, - unsigned mode, unsigned long timeout) -{ - wait_queue_head_t *wq = bit_waitqueue(word, bit); - DEFINE_WAIT_BIT(wait, word, bit); - - wait.key.timeout = jiffies + timeout; - return __wait_on_bit(wq, &wait, action, mode); -} -EXPORT_SYMBOL_GPL(out_of_line_wait_on_bit_timeout); - -int __sched -__wait_on_bit_lock(wait_queue_head_t *wq, struct wait_bit_queue *q, - wait_bit_action_f *action, unsigned mode) -{ - int ret = 0; - - for (;;) { - prepare_to_wait_exclusive(wq, &q->wait, mode); - if (test_bit(q->key.bit_nr, q->key.flags)) { - ret = action(&q->key, mode); - /* - * See the comment in prepare_to_wait_event(). - * finish_wait() does not necessarily takes wq->lock, - * but test_and_set_bit() implies mb() which pairs with - * smp_mb__after_atomic() before wake_up_page(). - */ - if (ret) - finish_wait(wq, &q->wait); - } - if (!test_and_set_bit(q->key.bit_nr, q->key.flags)) { - if (!ret) - finish_wait(wq, &q->wait); - return 0; - } else if (ret) { - return ret; - } - } -} -EXPORT_SYMBOL(__wait_on_bit_lock); - -int __sched out_of_line_wait_on_bit_lock(void *word, int bit, - wait_bit_action_f *action, unsigned mode) -{ - wait_queue_head_t *wq = bit_waitqueue(word, bit); - DEFINE_WAIT_BIT(wait, word, bit); - - return __wait_on_bit_lock(wq, &wait, action, mode); -} -EXPORT_SYMBOL(out_of_line_wait_on_bit_lock); - -void __wake_up_bit(wait_queue_head_t *wq, void *word, int bit) -{ - struct wait_bit_key key = __WAIT_BIT_KEY_INITIALIZER(word, bit); - if (waitqueue_active(wq)) - __wake_up(wq, TASK_NORMAL, 1, &key); -} -EXPORT_SYMBOL(__wake_up_bit); - -/** - * wake_up_bit - wake up a waiter on a bit - * @word: the word being waited on, a kernel virtual address - * @bit: the bit of the word being waited on - * - * There is a standard hashed waitqueue table for generic use. This - * is the part of the hashtable's accessor API that wakes up waiters - * on a bit. For instance, if one were to have waiters on a bitflag, - * one would call wake_up_bit() after clearing the bit. - * - * In order for this to function properly, as it uses waitqueue_active() - * internally, some kind of memory barrier must be done prior to calling - * this. Typically, this will be smp_mb__after_atomic(), but in some - * cases where bitflags are manipulated non-atomically under a lock, one - * may need to use a less regular barrier, such fs/inode.c's smp_mb(), - * because spin_unlock() does not guarantee a memory barrier. - */ -void wake_up_bit(void *word, int bit) -{ - __wake_up_bit(bit_waitqueue(word, bit), word, bit); -} -EXPORT_SYMBOL(wake_up_bit); - -/* - * Manipulate the atomic_t address to produce a better bit waitqueue table hash - * index (we're keying off bit -1, but that would produce a horrible hash - * value). - */ -static inline wait_queue_head_t *atomic_t_waitqueue(atomic_t *p) -{ - if (BITS_PER_LONG == 64) { - unsigned long q = (unsigned long)p; - return bit_waitqueue((void *)(q & ~1), q & 1); - } - return bit_waitqueue(p, 0); -} - -static int wake_atomic_t_function(wait_queue_t *wait, unsigned mode, int sync, - void *arg) -{ - struct wait_bit_key *key = arg; - struct wait_bit_queue *wait_bit - = container_of(wait, struct wait_bit_queue, wait); - atomic_t *val = key->flags; - - if (wait_bit->key.flags != key->flags || - wait_bit->key.bit_nr != key->bit_nr || - atomic_read(val) != 0) - return 0; - return autoremove_wake_function(wait, mode, sync, key); -} - -/* - * To allow interruptible waiting and asynchronous (i.e. nonblocking) waiting, - * the actions of __wait_on_atomic_t() are permitted return codes. Nonzero - * return codes halt waiting and return. - */ -static __sched -int __wait_on_atomic_t(wait_queue_head_t *wq, struct wait_bit_queue *q, - int (*action)(atomic_t *), unsigned mode) -{ - atomic_t *val; - int ret = 0; - - do { - prepare_to_wait(wq, &q->wait, mode); - val = q->key.flags; - if (atomic_read(val) == 0) - break; - ret = (*action)(val); - } while (!ret && atomic_read(val) != 0); - finish_wait(wq, &q->wait); - return ret; -} - -#define DEFINE_WAIT_ATOMIC_T(name, p) \ - struct wait_bit_queue name = { \ - .key = __WAIT_ATOMIC_T_KEY_INITIALIZER(p), \ - .wait = { \ - .private = current, \ - .func = wake_atomic_t_function, \ - .task_list = \ - LIST_HEAD_INIT((name).wait.task_list), \ - }, \ - } - -__sched int out_of_line_wait_on_atomic_t(atomic_t *p, int (*action)(atomic_t *), - unsigned mode) -{ - wait_queue_head_t *wq = atomic_t_waitqueue(p); - DEFINE_WAIT_ATOMIC_T(wait, p); - - return __wait_on_atomic_t(wq, &wait, action, mode); -} -EXPORT_SYMBOL(out_of_line_wait_on_atomic_t); - -/** - * wake_up_atomic_t - Wake up a waiter on a atomic_t - * @p: The atomic_t being waited on, a kernel virtual address - * - * Wake up anyone waiting for the atomic_t to go to zero. - * - * Abuse the bit-waker function and its waitqueue hash table set (the atomic_t - * check is done by the waiter's wake function, not the by the waker itself). - */ -void wake_up_atomic_t(atomic_t *p) -{ - __wake_up_bit(atomic_t_waitqueue(p), p, WAIT_ATOMIC_T_BIT_NR); -} -EXPORT_SYMBOL(wake_up_atomic_t); - -__sched int bit_wait(struct wait_bit_key *word, int mode) -{ - schedule(); - if (signal_pending_state(mode, current)) - return -EINTR; - return 0; -} -EXPORT_SYMBOL(bit_wait); - -__sched int bit_wait_io(struct wait_bit_key *word, int mode) -{ - io_schedule(); - if (signal_pending_state(mode, current)) - return -EINTR; - return 0; -} -EXPORT_SYMBOL(bit_wait_io); - -__sched int bit_wait_timeout(struct wait_bit_key *word, int mode) -{ - unsigned long now = READ_ONCE(jiffies); - if (time_after_eq(now, word->timeout)) - return -EAGAIN; - schedule_timeout(word->timeout - now); - if (signal_pending_state(mode, current)) - return -EINTR; - return 0; -} -EXPORT_SYMBOL_GPL(bit_wait_timeout); - -__sched int bit_wait_io_timeout(struct wait_bit_key *word, int mode) -{ - unsigned long now = READ_ONCE(jiffies); - if (time_after_eq(now, word->timeout)) - return -EAGAIN; - io_schedule_timeout(word->timeout - now); - if (signal_pending_state(mode, current)) - return -EINTR; - return 0; -} -EXPORT_SYMBOL_GPL(bit_wait_io_timeout); diff --git a/kernel/sched/wait_bit.c b/kernel/sched/wait_bit.c new file mode 100644 index 000000000000..f8159698aa4d --- /dev/null +++ b/kernel/sched/wait_bit.c @@ -0,0 +1,286 @@ +/* + * The implementation of the wait_bit*() and related waiting APIs: + */ +#include <linux/wait_bit.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> +#include <linux/hash.h> + +#define WAIT_TABLE_BITS 8 +#define WAIT_TABLE_SIZE (1 << WAIT_TABLE_BITS) + +static wait_queue_head_t bit_wait_table[WAIT_TABLE_SIZE] __cacheline_aligned; + +wait_queue_head_t *bit_waitqueue(void *word, int bit) +{ + const int shift = BITS_PER_LONG == 32 ? 5 : 6; + unsigned long val = (unsigned long)word << shift | bit; + + return bit_wait_table + hash_long(val, WAIT_TABLE_BITS); +} +EXPORT_SYMBOL(bit_waitqueue); + +int wake_bit_function(struct wait_queue_entry *wq_entry, unsigned mode, int sync, void *arg) +{ + struct wait_bit_key *key = arg; + struct wait_bit_queue_entry *wait_bit = container_of(wq_entry, struct wait_bit_queue_entry, wq_entry); + + if (wait_bit->key.flags != key->flags || + wait_bit->key.bit_nr != key->bit_nr || + test_bit(key->bit_nr, key->flags)) + return 0; + else + return autoremove_wake_function(wq_entry, mode, sync, key); +} +EXPORT_SYMBOL(wake_bit_function); + +/* + * To allow interruptible waiting and asynchronous (i.e. nonblocking) + * waiting, the actions of __wait_on_bit() and __wait_on_bit_lock() are + * permitted return codes. Nonzero return codes halt waiting and return. + */ +int __sched +__wait_on_bit(struct wait_queue_head *wq_head, struct wait_bit_queue_entry *wbq_entry, + wait_bit_action_f *action, unsigned mode) +{ + int ret = 0; + + do { + prepare_to_wait(wq_head, &wbq_entry->wq_entry, mode); + if (test_bit(wbq_entry->key.bit_nr, wbq_entry->key.flags)) + ret = (*action)(&wbq_entry->key, mode); + } while (test_bit(wbq_entry->key.bit_nr, wbq_entry->key.flags) && !ret); + finish_wait(wq_head, &wbq_entry->wq_entry); + return ret; +} +EXPORT_SYMBOL(__wait_on_bit); + +int __sched out_of_line_wait_on_bit(void *word, int bit, + wait_bit_action_f *action, unsigned mode) +{ + struct wait_queue_head *wq_head = bit_waitqueue(word, bit); + DEFINE_WAIT_BIT(wq_entry, word, bit); + + return __wait_on_bit(wq_head, &wq_entry, action, mode); +} +EXPORT_SYMBOL(out_of_line_wait_on_bit); + +int __sched out_of_line_wait_on_bit_timeout( + void *word, int bit, wait_bit_action_f *action, + unsigned mode, unsigned long timeout) +{ + struct wait_queue_head *wq_head = bit_waitqueue(word, bit); + DEFINE_WAIT_BIT(wq_entry, word, bit); + + wq_entry.key.timeout = jiffies + timeout; + return __wait_on_bit(wq_head, &wq_entry, action, mode); +} +EXPORT_SYMBOL_GPL(out_of_line_wait_on_bit_timeout); + +int __sched +__wait_on_bit_lock(struct wait_queue_head *wq_head, struct wait_bit_queue_entry *wbq_entry, + wait_bit_action_f *action, unsigned mode) +{ + int ret = 0; + + for (;;) { + prepare_to_wait_exclusive(wq_head, &wbq_entry->wq_entry, mode); + if (test_bit(wbq_entry->key.bit_nr, wbq_entry->key.flags)) { + ret = action(&wbq_entry->key, mode); + /* + * See the comment in prepare_to_wait_event(). + * finish_wait() does not necessarily takes wwq_head->lock, + * but test_and_set_bit() implies mb() which pairs with + * smp_mb__after_atomic() before wake_up_page(). + */ + if (ret) + finish_wait(wq_head, &wbq_entry->wq_entry); + } + if (!test_and_set_bit(wbq_entry->key.bit_nr, wbq_entry->key.flags)) { + if (!ret) + finish_wait(wq_head, &wbq_entry->wq_entry); + return 0; + } else if (ret) { + return ret; + } + } +} +EXPORT_SYMBOL(__wait_on_bit_lock); + +int __sched out_of_line_wait_on_bit_lock(void *word, int bit, + wait_bit_action_f *action, unsigned mode) +{ + struct wait_queue_head *wq_head = bit_waitqueue(word, bit); + DEFINE_WAIT_BIT(wq_entry, word, bit); + + return __wait_on_bit_lock(wq_head, &wq_entry, action, mode); +} +EXPORT_SYMBOL(out_of_line_wait_on_bit_lock); + +void __wake_up_bit(struct wait_queue_head *wq_head, void *word, int bit) +{ + struct wait_bit_key key = __WAIT_BIT_KEY_INITIALIZER(word, bit); + if (waitqueue_active(wq_head)) + __wake_up(wq_head, TASK_NORMAL, 1, &key); +} +EXPORT_SYMBOL(__wake_up_bit); + +/** + * wake_up_bit - wake up a waiter on a bit + * @word: the word being waited on, a kernel virtual address + * @bit: the bit of the word being waited on + * + * There is a standard hashed waitqueue table for generic use. This + * is the part of the hashtable's accessor API that wakes up waiters + * on a bit. For instance, if one were to have waiters on a bitflag, + * one would call wake_up_bit() after clearing the bit. + * + * In order for this to function properly, as it uses waitqueue_active() + * internally, some kind of memory barrier must be done prior to calling + * this. Typically, this will be smp_mb__after_atomic(), but in some + * cases where bitflags are manipulated non-atomically under a lock, one + * may need to use a less regular barrier, such fs/inode.c's smp_mb(), + * because spin_unlock() does not guarantee a memory barrier. + */ +void wake_up_bit(void *word, int bit) +{ + __wake_up_bit(bit_waitqueue(word, bit), word, bit); +} +EXPORT_SYMBOL(wake_up_bit); + +/* + * Manipulate the atomic_t address to produce a better bit waitqueue table hash + * index (we're keying off bit -1, but that would produce a horrible hash + * value). + */ +static inline wait_queue_head_t *atomic_t_waitqueue(atomic_t *p) +{ + if (BITS_PER_LONG == 64) { + unsigned long q = (unsigned long)p; + return bit_waitqueue((void *)(q & ~1), q & 1); + } + return bit_waitqueue(p, 0); +} + +static int wake_atomic_t_function(struct wait_queue_entry *wq_entry, unsigned mode, int sync, + void *arg) +{ + struct wait_bit_key *key = arg; + struct wait_bit_queue_entry *wait_bit = container_of(wq_entry, struct wait_bit_queue_entry, wq_entry); + atomic_t *val = key->flags; + + if (wait_bit->key.flags != key->flags || + wait_bit->key.bit_nr != key->bit_nr || + atomic_read(val) != 0) + return 0; + return autoremove_wake_function(wq_entry, mode, sync, key); +} + +/* + * To allow interruptible waiting and asynchronous (i.e. nonblocking) waiting, + * the actions of __wait_on_atomic_t() are permitted return codes. Nonzero + * return codes halt waiting and return. + */ +static __sched +int __wait_on_atomic_t(struct wait_queue_head *wq_head, struct wait_bit_queue_entry *wbq_entry, + int (*action)(atomic_t *), unsigned mode) +{ + atomic_t *val; + int ret = 0; + + do { + prepare_to_wait(wq_head, &wbq_entry->wq_entry, mode); + val = wbq_entry->key.flags; + if (atomic_read(val) == 0) + break; + ret = (*action)(val); + } while (!ret && atomic_read(val) != 0); + finish_wait(wq_head, &wbq_entry->wq_entry); + return ret; +} + +#define DEFINE_WAIT_ATOMIC_T(name, p) \ + struct wait_bit_queue_entry name = { \ + .key = __WAIT_ATOMIC_T_KEY_INITIALIZER(p), \ + .wq_entry = { \ + .private = current, \ + .func = wake_atomic_t_function, \ + .entry = \ + LIST_HEAD_INIT((name).wq_entry.entry), \ + }, \ + } + +__sched int out_of_line_wait_on_atomic_t(atomic_t *p, int (*action)(atomic_t *), + unsigned mode) +{ + struct wait_queue_head *wq_head = atomic_t_waitqueue(p); + DEFINE_WAIT_ATOMIC_T(wq_entry, p); + + return __wait_on_atomic_t(wq_head, &wq_entry, action, mode); +} +EXPORT_SYMBOL(out_of_line_wait_on_atomic_t); + +/** + * wake_up_atomic_t - Wake up a waiter on a atomic_t + * @p: The atomic_t being waited on, a kernel virtual address + * + * Wake up anyone waiting for the atomic_t to go to zero. + * + * Abuse the bit-waker function and its waitqueue hash table set (the atomic_t + * check is done by the waiter's wake function, not the by the waker itself). + */ +void wake_up_atomic_t(atomic_t *p) +{ + __wake_up_bit(atomic_t_waitqueue(p), p, WAIT_ATOMIC_T_BIT_NR); +} +EXPORT_SYMBOL(wake_up_atomic_t); + +__sched int bit_wait(struct wait_bit_key *word, int mode) +{ + schedule(); + if (signal_pending_state(mode, current)) + return -EINTR; + return 0; +} +EXPORT_SYMBOL(bit_wait); + +__sched int bit_wait_io(struct wait_bit_key *word, int mode) +{ + io_schedule(); + if (signal_pending_state(mode, current)) + return -EINTR; + return 0; +} +EXPORT_SYMBOL(bit_wait_io); + +__sched int bit_wait_timeout(struct wait_bit_key *word, int mode) +{ + unsigned long now = READ_ONCE(jiffies); + if (time_after_eq(now, word->timeout)) + return -EAGAIN; + schedule_timeout(word->timeout - now); + if (signal_pending_state(mode, current)) + return -EINTR; + return 0; +} +EXPORT_SYMBOL_GPL(bit_wait_timeout); + +__sched int bit_wait_io_timeout(struct wait_bit_key *word, int mode) +{ + unsigned long now = READ_ONCE(jiffies); + if (time_after_eq(now, word->timeout)) + return -EAGAIN; + io_schedule_timeout(word->timeout - now); + if (signal_pending_state(mode, current)) + return -EINTR; + return 0; +} +EXPORT_SYMBOL_GPL(bit_wait_io_timeout); + +void __init wait_bit_init(void) +{ + int i; + + for (i = 0; i < WAIT_TABLE_SIZE; i++) + init_waitqueue_head(bit_wait_table + i); +} diff --git a/kernel/seccomp.c b/kernel/seccomp.c index f7ce79a46050..98b59b5db90b 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -13,10 +13,12 @@ * of Berkeley Packet Filters/Linux Socket Filters. */ -#include <linux/atomic.h> +#include <linux/refcount.h> #include <linux/audit.h> #include <linux/compat.h> +#include <linux/coredump.h> #include <linux/sched.h> +#include <linux/sched/task_stack.h> #include <linux/seccomp.h> #include <linux/slab.h> #include <linux/syscalls.h> @@ -54,7 +56,7 @@ * to a task_struct (other than @usage). */ struct seccomp_filter { - atomic_t usage; + refcount_t usage; struct seccomp_filter *prev; struct bpf_prog *prog; }; @@ -376,7 +378,7 @@ static struct seccomp_filter *seccomp_prepare_filter(struct sock_fprog *fprog) return ERR_PTR(ret); } - atomic_set(&sfilter->usage, 1); + refcount_set(&sfilter->usage, 1); return sfilter; } @@ -463,7 +465,7 @@ void get_seccomp_filter(struct task_struct *tsk) if (!orig) return; /* Reference count is bounded by the number of total processes. */ - atomic_inc(&orig->usage); + refcount_inc(&orig->usage); } static inline void seccomp_filter_free(struct seccomp_filter *filter) @@ -479,13 +481,24 @@ void put_seccomp_filter(struct task_struct *tsk) { struct seccomp_filter *orig = tsk->seccomp.filter; /* Clean up single-reference branches iteratively. */ - while (orig && atomic_dec_and_test(&orig->usage)) { + while (orig && refcount_dec_and_test(&orig->usage)) { struct seccomp_filter *freeme = orig; orig = orig->prev; seccomp_filter_free(freeme); } } +static void seccomp_init_siginfo(siginfo_t *info, int syscall, int reason) +{ + memset(info, 0, sizeof(*info)); + info->si_signo = SIGSYS; + info->si_code = SYS_SECCOMP; + info->si_call_addr = (void __user *)KSTK_EIP(current); + info->si_errno = reason; + info->si_arch = syscall_get_arch(); + info->si_syscall = syscall; +} + /** * seccomp_send_sigsys - signals the task to allow in-process syscall emulation * @syscall: syscall number to send to userland @@ -496,13 +509,7 @@ void put_seccomp_filter(struct task_struct *tsk) static void seccomp_send_sigsys(int syscall, int reason) { struct siginfo info; - memset(&info, 0, sizeof(info)); - info.si_signo = SIGSYS; - info.si_code = SYS_SECCOMP; - info.si_call_addr = (void __user *)KSTK_EIP(current); - info.si_errno = reason; - info.si_arch = syscall_get_arch(); - info.si_syscall = syscall; + seccomp_init_siginfo(&info, syscall, reason); force_sig_info(SIGSYS, &info, current); } #endif /* CONFIG_SECCOMP_FILTER */ @@ -636,6 +643,16 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd, case SECCOMP_RET_KILL: default: audit_seccomp(this_syscall, SIGSYS, action); + /* Dump core only if this is the last remaining thread. */ + if (get_nr_threads(current) == 1) { + siginfo_t info; + + /* Show the original registers in the dump. */ + syscall_rollback(current, task_pt_regs(current)); + /* Trigger a manual coredump since do_exit skips it. */ + seccomp_init_siginfo(&info, this_syscall, data); + do_coredump(&info); + } do_exit(SIGSYS); } diff --git a/kernel/signal.c b/kernel/signal.c index 3603d93a1968..caed9133ae52 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -13,7 +13,12 @@ #include <linux/slab.h> #include <linux/export.h> #include <linux/init.h> -#include <linux/sched.h> +#include <linux/sched/mm.h> +#include <linux/sched/user.h> +#include <linux/sched/debug.h> +#include <linux/sched/task.h> +#include <linux/sched/task_stack.h> +#include <linux/sched/cputime.h> #include <linux/fs.h> #include <linux/tty.h> #include <linux/binfmts.h> @@ -34,6 +39,7 @@ #include <linux/compat.h> #include <linux/cn_proc.h> #include <linux/compiler.h> +#include <linux/posix-timers.h> #define CREATE_TRACE_POINTS #include <trace/events/signal.h> @@ -505,7 +511,8 @@ int unhandled_signal(struct task_struct *tsk, int sig) return !tsk->ptrace; } -static void collect_signal(int sig, struct sigpending *list, siginfo_t *info) +static void collect_signal(int sig, struct sigpending *list, siginfo_t *info, + bool *resched_timer) { struct sigqueue *q, *first = NULL; @@ -527,6 +534,12 @@ static void collect_signal(int sig, struct sigpending *list, siginfo_t *info) still_pending: list_del_init(&first->list); copy_siginfo(info, &first->info); + + *resched_timer = + (first->flags & SIGQUEUE_PREALLOC) && + (info->si_code == SI_TIMER) && + (info->si_sys_private); + __sigqueue_free(first); } else { /* @@ -543,12 +556,12 @@ still_pending: } static int __dequeue_signal(struct sigpending *pending, sigset_t *mask, - siginfo_t *info) + siginfo_t *info, bool *resched_timer) { int sig = next_signal(pending, mask); if (sig) - collect_signal(sig, pending, info); + collect_signal(sig, pending, info, resched_timer); return sig; } @@ -560,15 +573,16 @@ static int __dequeue_signal(struct sigpending *pending, sigset_t *mask, */ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info) { + bool resched_timer = false; int signr; /* We only dequeue private signals from ourselves, we don't let * signalfd steal them */ - signr = __dequeue_signal(&tsk->pending, mask, info); + signr = __dequeue_signal(&tsk->pending, mask, info, &resched_timer); if (!signr) { signr = __dequeue_signal(&tsk->signal->shared_pending, - mask, info); + mask, info, &resched_timer); #ifdef CONFIG_POSIX_TIMERS /* * itimer signal ? @@ -616,7 +630,7 @@ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info) current->jobctl |= JOBCTL_STOP_DEQUEUED; } #ifdef CONFIG_POSIX_TIMERS - if ((info->si_code & __SI_MASK) == __SI_TIMER && info->si_sys_private) { + if (resched_timer) { /* * Release the siglock to ensure proper locking order * of timer locks outside of siglocks. Note, we leave @@ -624,7 +638,7 @@ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info) * about to disable them again anyway. */ spin_unlock(&tsk->sighand->siglock); - do_schedule_next_timer(info); + posixtimer_rearm(info); spin_lock(&tsk->sighand->siglock); } #endif @@ -1232,7 +1246,7 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk, } /* * This sighand can be already freed and even reused, but - * we rely on SLAB_DESTROY_BY_RCU and sighand_ctor() which + * we rely on SLAB_TYPESAFE_BY_RCU and sighand_ctor() which * initializes ->siglock: this slab can't go away, it has * the same object type, ->siglock can't be reinitialized. * @@ -1313,7 +1327,7 @@ int kill_pid_info(int sig, struct siginfo *info, struct pid *pid) } } -int kill_proc_info(int sig, struct siginfo *info, pid_t pid) +static int kill_proc_info(int sig, struct siginfo *info, pid_t pid) { int error; rcu_read_lock(); @@ -1388,6 +1402,10 @@ static int kill_something_info(int sig, struct siginfo *info, pid_t pid) return ret; } + /* -INT_MIN is undefined. Exclude this case to avoid a UBSAN warning */ + if (pid == INT_MIN) + return -ESRCH; + read_lock(&tasklist_lock); if (pid != -1) { ret = __kill_pgrp_info(sig, info, @@ -1581,7 +1599,7 @@ bool do_notify_parent(struct task_struct *tsk, int sig) unsigned long flags; struct sighand_struct *psig; bool autoreap = false; - cputime_t utime, stime; + u64 utime, stime; BUG_ON(sig == -1); @@ -1620,8 +1638,8 @@ bool do_notify_parent(struct task_struct *tsk, int sig) rcu_read_unlock(); task_cputime(tsk, &utime, &stime); - info.si_utime = cputime_to_clock_t(utime + tsk->signal->utime); - info.si_stime = cputime_to_clock_t(stime + tsk->signal->stime); + info.si_utime = nsec_to_clock_t(utime + tsk->signal->utime); + info.si_stime = nsec_to_clock_t(stime + tsk->signal->stime); info.si_status = tsk->exit_code & 0x7f; if (tsk->exit_code & 0x80) @@ -1685,7 +1703,7 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, unsigned long flags; struct task_struct *parent; struct sighand_struct *sighand; - cputime_t utime, stime; + u64 utime, stime; if (for_ptracer) { parent = tsk->parent; @@ -1705,8 +1723,8 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, rcu_read_unlock(); task_cputime(tsk, &utime, &stime); - info.si_utime = cputime_to_clock_t(utime); - info.si_stime = cputime_to_clock_t(stime); + info.si_utime = nsec_to_clock_t(utime); + info.si_stime = nsec_to_clock_t(stime); info.si_code = why; switch (why) { @@ -2087,7 +2105,6 @@ static void do_jobctl_trap(void) static int ptrace_signal(int signr, siginfo_t *info) { - ptrace_signal_deliver(); /* * We do not check sig_kernel_stop(signr) but set this marker * unconditionally because we do not know whether debugger will @@ -2395,11 +2412,11 @@ void exit_signals(struct task_struct *tsk) * @tsk is about to have PF_EXITING set - lock out users which * expect stable threadgroup. */ - threadgroup_change_begin(tsk); + cgroup_threadgroup_change_begin(tsk); if (thread_group_empty(tsk) || signal_group_exit(tsk->signal)) { tsk->flags |= PF_EXITING; - threadgroup_change_end(tsk); + cgroup_threadgroup_change_end(tsk); return; } @@ -2410,7 +2427,7 @@ void exit_signals(struct task_struct *tsk) */ tsk->flags |= PF_EXITING; - threadgroup_change_end(tsk); + cgroup_threadgroup_change_end(tsk); if (!signal_pending(tsk)) goto out; @@ -2763,7 +2780,7 @@ int copy_siginfo_to_user(siginfo_t __user *to, const siginfo_t *from) * @info: if non-null, the signal's siginfo is returned here * @ts: upper bound on process time suspension */ -int do_sigtimedwait(const sigset_t *which, siginfo_t *info, +static int do_sigtimedwait(const sigset_t *which, siginfo_t *info, const struct timespec *ts) { ktime_t *to = NULL, timeout = KTIME_MAX; @@ -2852,6 +2869,40 @@ SYSCALL_DEFINE4(rt_sigtimedwait, const sigset_t __user *, uthese, return ret; } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(rt_sigtimedwait, compat_sigset_t __user *, uthese, + struct compat_siginfo __user *, uinfo, + struct compat_timespec __user *, uts, compat_size_t, sigsetsize) +{ + compat_sigset_t s32; + sigset_t s; + struct timespec t; + siginfo_t info; + long ret; + + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + + if (copy_from_user(&s32, uthese, sizeof(compat_sigset_t))) + return -EFAULT; + sigset_from_compat(&s, &s32); + + if (uts) { + if (compat_get_timespec(&t, uts)) + return -EFAULT; + } + + ret = do_sigtimedwait(&s, &info, uts ? &t : NULL); + + if (ret > 0 && uinfo) { + if (copy_siginfo_to_user32(uinfo, &info)) + ret = -EFAULT; + } + + return ret; +} +#endif + /** * sys_kill - send a signal to a process * @pid: the PID of the process @@ -3108,78 +3159,68 @@ int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact) } static int -do_sigaltstack (const stack_t __user *uss, stack_t __user *uoss, unsigned long sp) +do_sigaltstack (const stack_t *ss, stack_t *oss, unsigned long sp) { - stack_t oss; - int error; + struct task_struct *t = current; - oss.ss_sp = (void __user *) current->sas_ss_sp; - oss.ss_size = current->sas_ss_size; - oss.ss_flags = sas_ss_flags(sp) | - (current->sas_ss_flags & SS_FLAG_BITS); + if (oss) { + memset(oss, 0, sizeof(stack_t)); + oss->ss_sp = (void __user *) t->sas_ss_sp; + oss->ss_size = t->sas_ss_size; + oss->ss_flags = sas_ss_flags(sp) | + (current->sas_ss_flags & SS_FLAG_BITS); + } - if (uss) { - void __user *ss_sp; - size_t ss_size; - unsigned ss_flags; + if (ss) { + void __user *ss_sp = ss->ss_sp; + size_t ss_size = ss->ss_size; + unsigned ss_flags = ss->ss_flags; int ss_mode; - error = -EFAULT; - if (!access_ok(VERIFY_READ, uss, sizeof(*uss))) - goto out; - error = __get_user(ss_sp, &uss->ss_sp) | - __get_user(ss_flags, &uss->ss_flags) | - __get_user(ss_size, &uss->ss_size); - if (error) - goto out; - - error = -EPERM; - if (on_sig_stack(sp)) - goto out; + if (unlikely(on_sig_stack(sp))) + return -EPERM; ss_mode = ss_flags & ~SS_FLAG_BITS; - error = -EINVAL; - if (ss_mode != SS_DISABLE && ss_mode != SS_ONSTACK && - ss_mode != 0) - goto out; + if (unlikely(ss_mode != SS_DISABLE && ss_mode != SS_ONSTACK && + ss_mode != 0)) + return -EINVAL; if (ss_mode == SS_DISABLE) { ss_size = 0; ss_sp = NULL; } else { - error = -ENOMEM; - if (ss_size < MINSIGSTKSZ) - goto out; + if (unlikely(ss_size < MINSIGSTKSZ)) + return -ENOMEM; } - current->sas_ss_sp = (unsigned long) ss_sp; - current->sas_ss_size = ss_size; - current->sas_ss_flags = ss_flags; - } - - error = 0; - if (uoss) { - error = -EFAULT; - if (!access_ok(VERIFY_WRITE, uoss, sizeof(*uoss))) - goto out; - error = __put_user(oss.ss_sp, &uoss->ss_sp) | - __put_user(oss.ss_size, &uoss->ss_size) | - __put_user(oss.ss_flags, &uoss->ss_flags); + t->sas_ss_sp = (unsigned long) ss_sp; + t->sas_ss_size = ss_size; + t->sas_ss_flags = ss_flags; } - -out: - return error; + return 0; } + SYSCALL_DEFINE2(sigaltstack,const stack_t __user *,uss, stack_t __user *,uoss) { - return do_sigaltstack(uss, uoss, current_user_stack_pointer()); + stack_t new, old; + int err; + if (uss && copy_from_user(&new, uss, sizeof(stack_t))) + return -EFAULT; + err = do_sigaltstack(uss ? &new : NULL, uoss ? &old : NULL, + current_user_stack_pointer()); + if (!err && uoss && copy_to_user(uoss, &old, sizeof(stack_t))) + err = -EFAULT; + return err; } int restore_altstack(const stack_t __user *uss) { - int err = do_sigaltstack(uss, NULL, current_user_stack_pointer()); + stack_t new; + if (copy_from_user(&new, uss, sizeof(stack_t))) + return -EFAULT; + (void)do_sigaltstack(&new, NULL, current_user_stack_pointer()); /* squash all but EFAULT for now */ - return err == -EFAULT ? err : 0; + return 0; } int __save_altstack(stack_t __user *uss, unsigned long sp) @@ -3202,29 +3243,24 @@ COMPAT_SYSCALL_DEFINE2(sigaltstack, { stack_t uss, uoss; int ret; - mm_segment_t seg; if (uss_ptr) { compat_stack_t uss32; - - memset(&uss, 0, sizeof(stack_t)); if (copy_from_user(&uss32, uss_ptr, sizeof(compat_stack_t))) return -EFAULT; uss.ss_sp = compat_ptr(uss32.ss_sp); uss.ss_flags = uss32.ss_flags; uss.ss_size = uss32.ss_size; } - seg = get_fs(); - set_fs(KERNEL_DS); - ret = do_sigaltstack((stack_t __force __user *) (uss_ptr ? &uss : NULL), - (stack_t __force __user *) &uoss, + ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss, compat_user_stack_pointer()); - set_fs(seg); if (ret >= 0 && uoss_ptr) { - if (!access_ok(VERIFY_WRITE, uoss_ptr, sizeof(compat_stack_t)) || - __put_user(ptr_to_compat(uoss.ss_sp), &uoss_ptr->ss_sp) || - __put_user(uoss.ss_flags, &uoss_ptr->ss_flags) || - __put_user(uoss.ss_size, &uoss_ptr->ss_size)) + compat_stack_t old; + memset(&old, 0, sizeof(old)); + old.ss_sp = ptr_to_compat(uoss.ss_sp); + old.ss_flags = uoss.ss_flags; + old.ss_size = uoss.ss_size; + if (copy_to_user(uoss_ptr, &old, sizeof(compat_stack_t))) ret = -EFAULT; } return ret; @@ -3239,10 +3275,17 @@ int compat_restore_altstack(const compat_stack_t __user *uss) int __compat_save_altstack(compat_stack_t __user *uss, unsigned long sp) { + int err; struct task_struct *t = current; - return __put_user(ptr_to_compat((void __user *)t->sas_ss_sp), &uss->ss_sp) | - __put_user(sas_ss_flags(sp), &uss->ss_flags) | + err = __put_user(ptr_to_compat((void __user *)t->sas_ss_sp), + &uss->ss_sp) | + __put_user(t->sas_ss_flags, &uss->ss_flags) | __put_user(t->sas_ss_size, &uss->ss_size); + if (err) + return err; + if (t->sas_ss_flags & SS_AUTODISARM) + sas_ss_reset(t); + return 0; } #endif @@ -3257,6 +3300,18 @@ SYSCALL_DEFINE1(sigpending, old_sigset_t __user *, set) return sys_rt_sigpending((sigset_t __user *)set, sizeof(old_sigset_t)); } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE1(sigpending, compat_old_sigset_t __user *, set32) +{ + sigset_t set; + int err = do_sigpending(&set, sizeof(old_sigset_t)); + if (err == 0) + if (copy_to_user(set32, &set, sizeof(old_sigset_t))) + err = -EFAULT; + return err; +} +#endif + #endif #ifdef __ARCH_WANT_SYS_SIGPROCMASK diff --git a/kernel/smp.c b/kernel/smp.c index 77fcdb9f2775..3061483cb3ad 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -17,6 +17,7 @@ #include <linux/smp.h> #include <linux/cpu.h> #include <linux/sched.h> +#include <linux/sched/idle.h> #include <linux/hypervisor.h> #include "smpboot.h" @@ -29,6 +30,7 @@ enum { struct call_function_data { struct call_single_data __percpu *csd; cpumask_var_t cpumask; + cpumask_var_t cpumask_ipi; }; static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_function_data, cfd_data); @@ -44,9 +46,15 @@ int smpcfd_prepare_cpu(unsigned int cpu) if (!zalloc_cpumask_var_node(&cfd->cpumask, GFP_KERNEL, cpu_to_node(cpu))) return -ENOMEM; + if (!zalloc_cpumask_var_node(&cfd->cpumask_ipi, GFP_KERNEL, + cpu_to_node(cpu))) { + free_cpumask_var(cfd->cpumask); + return -ENOMEM; + } cfd->csd = alloc_percpu(struct call_single_data); if (!cfd->csd) { free_cpumask_var(cfd->cpumask); + free_cpumask_var(cfd->cpumask_ipi); return -ENOMEM; } @@ -58,6 +66,7 @@ int smpcfd_dead_cpu(unsigned int cpu) struct call_function_data *cfd = &per_cpu(cfd_data, cpu); free_cpumask_var(cfd->cpumask); + free_cpumask_var(cfd->cpumask_ipi); free_percpu(cfd->csd); return 0; } @@ -427,12 +436,13 @@ void smp_call_function_many(const struct cpumask *mask, cfd = this_cpu_ptr(&cfd_data); cpumask_and(cfd->cpumask, mask, cpu_online_mask); - cpumask_clear_cpu(this_cpu, cfd->cpumask); + __cpumask_clear_cpu(this_cpu, cfd->cpumask); /* Some callers race with other cpus changing the passed mask */ if (unlikely(!cpumask_weight(cfd->cpumask))) return; + cpumask_clear(cfd->cpumask_ipi); for_each_cpu(cpu, cfd->cpumask) { struct call_single_data *csd = per_cpu_ptr(cfd->csd, cpu); @@ -441,11 +451,12 @@ void smp_call_function_many(const struct cpumask *mask, csd->flags |= CSD_FLAG_SYNCHRONOUS; csd->func = func; csd->info = info; - llist_add(&csd->llist, &per_cpu(call_single_queue, cpu)); + if (llist_add(&csd->llist, &per_cpu(call_single_queue, cpu))) + __cpumask_set_cpu(cpu, cfd->cpumask_ipi); } /* Send a message to all CPUs in the map */ - arch_send_call_function_ipi_mask(cfd->cpumask); + arch_send_call_function_ipi_mask(cfd->cpumask_ipi); if (wait) { for_each_cpu(cpu, cfd->cpumask) { diff --git a/kernel/smpboot.c b/kernel/smpboot.c index 4a5c6e73ecd4..1d71c051a951 100644 --- a/kernel/smpboot.c +++ b/kernel/smpboot.c @@ -9,6 +9,7 @@ #include <linux/list.h> #include <linux/slab.h> #include <linux/sched.h> +#include <linux/sched/task.h> #include <linux/export.h> #include <linux/percpu.h> #include <linux/kthread.h> diff --git a/kernel/softirq.c b/kernel/softirq.c index 744fa611cae0..4e09821f9d9e 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -309,7 +309,7 @@ restart: account_irq_exit_time(current); __local_bh_enable(SOFTIRQ_OFFSET); WARN_ON_ONCE(in_interrupt()); - tsk_restore_flags(current, old_flags, PF_MEMALLOC); + current_restore_flags(old_flags, PF_MEMALLOC); } asmlinkage __visible void do_softirq(void) diff --git a/kernel/stacktrace.c b/kernel/stacktrace.c index b6e4c16377c7..f8edee9c792d 100644 --- a/kernel/stacktrace.c +++ b/kernel/stacktrace.c @@ -18,10 +18,8 @@ void print_stack_trace(struct stack_trace *trace, int spaces) if (WARN_ON(!trace->entries)) return; - for (i = 0; i < trace->nr_entries; i++) { - printk("%*c", 1 + spaces, ' '); - print_ip_sym(trace->entries[i]); - } + for (i = 0; i < trace->nr_entries; i++) + printk("%*c%pS\n", 1 + spaces, ' ', (void *)trace->entries[i]); } EXPORT_SYMBOL_GPL(print_stack_trace); @@ -29,7 +27,6 @@ int snprint_stack_trace(char *buf, size_t size, struct stack_trace *trace, int spaces) { int i; - unsigned long ip; int generated; int total = 0; @@ -37,9 +34,8 @@ int snprint_stack_trace(char *buf, size_t size, return 0; for (i = 0; i < trace->nr_entries; i++) { - ip = trace->entries[i]; - generated = snprintf(buf, size, "%*c[<%p>] %pS\n", - 1 + spaces, ' ', (void *) ip, (void *) ip); + generated = snprintf(buf, size, "%*c%pS\n", 1 + spaces, ' ', + (void *)trace->entries[i]); total += generated; @@ -58,8 +54,8 @@ int snprint_stack_trace(char *buf, size_t size, EXPORT_SYMBOL_GPL(snprint_stack_trace); /* - * Architectures that do not implement save_stack_trace_tsk or - * save_stack_trace_regs get this weak alias and a once-per-bootup warning + * Architectures that do not implement save_stack_trace_*() + * get these weak aliases and once-per-bootup warnings * (whenever this facility is utilized - for example by procfs): */ __weak void @@ -73,3 +69,11 @@ save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace) { WARN_ONCE(1, KERN_INFO "save_stack_trace_regs() not implemented yet.\n"); } + +__weak int +save_stack_trace_tsk_reliable(struct task_struct *tsk, + struct stack_trace *trace) +{ + WARN_ONCE(1, KERN_INFO "save_stack_tsk_reliable() not implemented yet.\n"); + return -ENOSYS; +} diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 1eb82661ecdb..b7591261652d 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -552,7 +552,8 @@ static int __init cpu_stop_init(void) } early_initcall(cpu_stop_init); -static int __stop_machine(cpu_stop_fn_t fn, void *data, const struct cpumask *cpus) +int stop_machine_cpuslocked(cpu_stop_fn_t fn, void *data, + const struct cpumask *cpus) { struct multi_stop_data msdata = { .fn = fn, @@ -561,6 +562,8 @@ static int __stop_machine(cpu_stop_fn_t fn, void *data, const struct cpumask *cp .active_cpus = cpus, }; + lockdep_assert_cpus_held(); + if (!stop_machine_initialized) { /* * Handle the case where stop_machine() is called @@ -590,9 +593,9 @@ int stop_machine(cpu_stop_fn_t fn, void *data, const struct cpumask *cpus) int ret; /* No CPUs can come up or down during this. */ - get_online_cpus(); - ret = __stop_machine(fn, data, cpus); - put_online_cpus(); + cpus_read_lock(); + ret = stop_machine_cpuslocked(fn, data, cpus); + cpus_read_unlock(); return ret; } EXPORT_SYMBOL_GPL(stop_machine); diff --git a/kernel/sys.c b/kernel/sys.c index 842914ef7de4..2855ee73acd0 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -49,6 +49,13 @@ #include <linux/binfmts.h> #include <linux/sched.h> +#include <linux/sched/autogroup.h> +#include <linux/sched/loadavg.h> +#include <linux/sched/stat.h> +#include <linux/sched/mm.h> +#include <linux/sched/coredump.h> +#include <linux/sched/task.h> +#include <linux/sched/cputime.h> #include <linux/rcupdate.h> #include <linux/uidgid.h> #include <linux/cred.h> @@ -879,17 +886,17 @@ SYSCALL_DEFINE0(getegid) return from_kgid_munged(current_user_ns(), current_egid()); } -void do_sys_times(struct tms *tms) +static void do_sys_times(struct tms *tms) { - cputime_t tgutime, tgstime, cutime, cstime; + u64 tgutime, tgstime, cutime, cstime; thread_group_cputime_adjusted(current, &tgutime, &tgstime); cutime = current->signal->cutime; cstime = current->signal->cstime; - tms->tms_utime = cputime_to_clock_t(tgutime); - tms->tms_stime = cputime_to_clock_t(tgstime); - tms->tms_cutime = cputime_to_clock_t(cutime); - tms->tms_cstime = cputime_to_clock_t(cstime); + tms->tms_utime = nsec_to_clock_t(tgutime); + tms->tms_stime = nsec_to_clock_t(tgstime); + tms->tms_cutime = nsec_to_clock_t(cutime); + tms->tms_cstime = nsec_to_clock_t(cstime); } SYSCALL_DEFINE1(times, struct tms __user *, tbuf) @@ -905,6 +912,32 @@ SYSCALL_DEFINE1(times, struct tms __user *, tbuf) return (long) jiffies_64_to_clock_t(get_jiffies_64()); } +#ifdef CONFIG_COMPAT +static compat_clock_t clock_t_to_compat_clock_t(clock_t x) +{ + return compat_jiffies_to_clock_t(clock_t_to_jiffies(x)); +} + +COMPAT_SYSCALL_DEFINE1(times, struct compat_tms __user *, tbuf) +{ + if (tbuf) { + struct tms tms; + struct compat_tms tmp; + + do_sys_times(&tms); + /* Convert our struct tms to the compat version. */ + tmp.tms_utime = clock_t_to_compat_clock_t(tms.tms_utime); + tmp.tms_stime = clock_t_to_compat_clock_t(tms.tms_stime); + tmp.tms_cutime = clock_t_to_compat_clock_t(tms.tms_cutime); + tmp.tms_cstime = clock_t_to_compat_clock_t(tms.tms_cstime); + if (copy_to_user(tbuf, &tmp, sizeof(tmp))) + return -EFAULT; + } + force_successful_syscall_return(); + return compat_jiffies_to_clock_t(jiffies); +} +#endif + /* * This needs some heavy checking ... * I just haven't the stomach for it. I also don't fully @@ -1299,6 +1332,54 @@ SYSCALL_DEFINE2(getrlimit, unsigned int, resource, struct rlimit __user *, rlim) return ret; } +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE2(setrlimit, unsigned int, resource, + struct compat_rlimit __user *, rlim) +{ + struct rlimit r; + struct compat_rlimit r32; + + if (copy_from_user(&r32, rlim, sizeof(struct compat_rlimit))) + return -EFAULT; + + if (r32.rlim_cur == COMPAT_RLIM_INFINITY) + r.rlim_cur = RLIM_INFINITY; + else + r.rlim_cur = r32.rlim_cur; + if (r32.rlim_max == COMPAT_RLIM_INFINITY) + r.rlim_max = RLIM_INFINITY; + else + r.rlim_max = r32.rlim_max; + return do_prlimit(current, resource, &r, NULL); +} + +COMPAT_SYSCALL_DEFINE2(getrlimit, unsigned int, resource, + struct compat_rlimit __user *, rlim) +{ + struct rlimit r; + int ret; + + ret = do_prlimit(current, resource, NULL, &r); + if (!ret) { + struct compat_rlimit r32; + if (r.rlim_cur > COMPAT_RLIM_INFINITY) + r32.rlim_cur = COMPAT_RLIM_INFINITY; + else + r32.rlim_cur = r.rlim_cur; + if (r.rlim_max > COMPAT_RLIM_INFINITY) + r32.rlim_max = COMPAT_RLIM_INFINITY; + else + r32.rlim_max = r.rlim_max; + + if (copy_to_user(rlim, &r32, sizeof(struct compat_rlimit))) + return -EFAULT; + } + return ret; +} + +#endif + #ifdef __ARCH_WANT_SYS_OLD_GETRLIMIT /* @@ -1321,6 +1402,30 @@ SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, return copy_to_user(rlim, &x, sizeof(x)) ? -EFAULT : 0; } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, + struct compat_rlimit __user *, rlim) +{ + struct rlimit r; + + if (resource >= RLIM_NLIMITS) + return -EINVAL; + + task_lock(current->group_leader); + r = current->signal->rlim[resource]; + task_unlock(current->group_leader); + if (r.rlim_cur > 0x7FFFFFFF) + r.rlim_cur = 0x7FFFFFFF; + if (r.rlim_max > 0x7FFFFFFF) + r.rlim_max = 0x7FFFFFFF; + + if (put_user(r.rlim_cur, &rlim->rlim_cur) || + put_user(r.rlim_max, &rlim->rlim_max)) + return -EFAULT; + return 0; +} +#endif + #endif static inline bool rlim64_is_infinity(__u64 rlim64) @@ -1389,8 +1494,7 @@ int do_prlimit(struct task_struct *tsk, unsigned int resource, !capable(CAP_SYS_RESOURCE)) retval = -EPERM; if (!retval) - retval = security_task_setrlimit(tsk->group_leader, - resource, new_rlim); + retval = security_task_setrlimit(tsk, resource, new_rlim); if (resource == RLIMIT_CPU && new_rlim->rlim_cur == 0) { /* * The caller is asking for an immediate RLIMIT_CPU @@ -1425,25 +1529,26 @@ out: } /* rcu lock must be held */ -static int check_prlimit_permission(struct task_struct *task) +static int check_prlimit_permission(struct task_struct *task, + unsigned int flags) { const struct cred *cred = current_cred(), *tcred; + bool id_match; if (current == task) return 0; tcred = __task_cred(task); - if (uid_eq(cred->uid, tcred->euid) && - uid_eq(cred->uid, tcred->suid) && - uid_eq(cred->uid, tcred->uid) && - gid_eq(cred->gid, tcred->egid) && - gid_eq(cred->gid, tcred->sgid) && - gid_eq(cred->gid, tcred->gid)) - return 0; - if (ns_capable(tcred->user_ns, CAP_SYS_RESOURCE)) - return 0; + id_match = (uid_eq(cred->uid, tcred->euid) && + uid_eq(cred->uid, tcred->suid) && + uid_eq(cred->uid, tcred->uid) && + gid_eq(cred->gid, tcred->egid) && + gid_eq(cred->gid, tcred->sgid) && + gid_eq(cred->gid, tcred->gid)); + if (!id_match && !ns_capable(tcred->user_ns, CAP_SYS_RESOURCE)) + return -EPERM; - return -EPERM; + return security_task_prlimit(cred, tcred, flags); } SYSCALL_DEFINE4(prlimit64, pid_t, pid, unsigned int, resource, @@ -1453,12 +1558,17 @@ SYSCALL_DEFINE4(prlimit64, pid_t, pid, unsigned int, resource, struct rlimit64 old64, new64; struct rlimit old, new; struct task_struct *tsk; + unsigned int checkflags = 0; int ret; + if (old_rlim) + checkflags |= LSM_PRLIMIT_READ; + if (new_rlim) { if (copy_from_user(&new64, new_rlim, sizeof(new64))) return -EFAULT; rlim64_to_rlim(&new64, &new); + checkflags |= LSM_PRLIMIT_WRITE; } rcu_read_lock(); @@ -1467,7 +1577,7 @@ SYSCALL_DEFINE4(prlimit64, pid_t, pid, unsigned int, resource, rcu_read_unlock(); return -ESRCH; } - ret = check_prlimit_permission(tsk); + ret = check_prlimit_permission(tsk, checkflags); if (ret) { rcu_read_unlock(); return ret; @@ -1540,11 +1650,11 @@ static void accumulate_thread_rusage(struct task_struct *t, struct rusage *r) r->ru_oublock += task_io_get_oublock(t); } -static void k_getrusage(struct task_struct *p, int who, struct rusage *r) +void getrusage(struct task_struct *p, int who, struct rusage *r) { struct task_struct *t; unsigned long flags; - cputime_t tgutime, tgstime, utime, stime; + u64 tgutime, tgstime, utime, stime; unsigned long maxrss = 0; memset((char *)r, 0, sizeof (*r)); @@ -1600,8 +1710,8 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) unlock_task_sighand(p, &flags); out: - cputime_to_timeval(utime, &r->ru_utime); - cputime_to_timeval(stime, &r->ru_stime); + r->ru_utime = ns_to_timeval(utime); + r->ru_stime = ns_to_timeval(stime); if (who != RUSAGE_CHILDREN) { struct mm_struct *mm = get_task_mm(p); @@ -1614,20 +1724,16 @@ out: r->ru_maxrss = maxrss * (PAGE_SIZE / 1024); /* convert pages to KBs */ } -int getrusage(struct task_struct *p, int who, struct rusage __user *ru) +SYSCALL_DEFINE2(getrusage, int, who, struct rusage __user *, ru) { struct rusage r; - k_getrusage(p, who, &r); - return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0; -} - -SYSCALL_DEFINE2(getrusage, int, who, struct rusage __user *, ru) -{ if (who != RUSAGE_SELF && who != RUSAGE_CHILDREN && who != RUSAGE_THREAD) return -EINVAL; - return getrusage(current, who, ru); + + getrusage(current, who, &r); + return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0; } #ifdef CONFIG_COMPAT @@ -1639,7 +1745,7 @@ COMPAT_SYSCALL_DEFINE2(getrusage, int, who, struct compat_rusage __user *, ru) who != RUSAGE_THREAD) return -EINVAL; - k_getrusage(current, who, &r); + getrusage(current, who, &r); return put_compat_rusage(&r, ru); } #endif @@ -2063,6 +2169,24 @@ static int prctl_get_tid_address(struct task_struct *me, int __user **tid_addr) } #endif +static int propagate_has_child_subreaper(struct task_struct *p, void *data) +{ + /* + * If task has has_child_subreaper - all its decendants + * already have these flag too and new decendants will + * inherit it on fork, skip them. + * + * If we've found child_reaper - skip descendants in + * it's subtree as they will never get out pidns. + */ + if (p->signal->has_child_subreaper || + is_child_reaper(task_pid(p))) + return 0; + + p->signal->has_child_subreaper = 1; + return 1; +} + SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5) { @@ -2214,6 +2338,10 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; case PR_SET_CHILD_SUBREAPER: me->signal->is_child_subreaper = !!arg2; + if (!arg2) + break; + + walk_process_tree(me, propagate_has_child_subreaper, NULL); break; case PR_GET_CHILD_SUBREAPER: error = put_user(me->signal->is_child_subreaper, @@ -2232,7 +2360,7 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_GET_THP_DISABLE: if (arg2 || arg3 || arg4 || arg5) return -EINVAL; - error = !!(me->mm->def_flags & VM_NOHUGEPAGE); + error = !!test_bit(MMF_DISABLE_THP, &me->mm->flags); break; case PR_SET_THP_DISABLE: if (arg3 || arg4 || arg5) @@ -2240,9 +2368,9 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, if (down_write_killable(&me->mm->mmap_sem)) return -EINTR; if (arg2) - me->mm->def_flags |= VM_NOHUGEPAGE; + set_bit(MMF_DISABLE_THP, &me->mm->flags); else - me->mm->def_flags &= ~VM_NOHUGEPAGE; + clear_bit(MMF_DISABLE_THP, &me->mm->flags); up_write(&me->mm->mmap_sem); break; case PR_MPX_ENABLE_MANAGEMENT: diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 1aea594a54db..6648fbbb8157 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -63,6 +63,7 @@ #include <linux/capability.h> #include <linux/binfmts.h> #include <linux/sched/sysctl.h> +#include <linux/sched/coredump.h> #include <linux/kexec.h> #include <linux/bpf.h> #include <linux/mount.h> @@ -173,11 +174,32 @@ extern int no_unaligned_warning; #ifdef CONFIG_PROC_SYSCTL -#define SYSCTL_WRITES_LEGACY -1 -#define SYSCTL_WRITES_WARN 0 -#define SYSCTL_WRITES_STRICT 1 +/** + * enum sysctl_writes_mode - supported sysctl write modes + * + * @SYSCTL_WRITES_LEGACY: each write syscall must fully contain the sysctl value + * to be written, and multiple writes on the same sysctl file descriptor + * will rewrite the sysctl value, regardless of file position. No warning + * is issued when the initial position is not 0. + * @SYSCTL_WRITES_WARN: same as above but warn when the initial file position is + * not 0. + * @SYSCTL_WRITES_STRICT: writes to numeric sysctl entries must always be at + * file position 0 and the value must be fully contained in the buffer + * sent to the write syscall. If dealing with strings respect the file + * position, but restrict this to the max length of the buffer, anything + * passed the max lenght will be ignored. Multiple writes will append + * to the buffer. + * + * These write modes control how current file position affects the behavior of + * updating sysctl values through the proc interface on each write. + */ +enum sysctl_writes_mode { + SYSCTL_WRITES_LEGACY = -1, + SYSCTL_WRITES_WARN = 0, + SYSCTL_WRITES_STRICT = 1, +}; -static int sysctl_writes_strict = SYSCTL_WRITES_STRICT; +static enum sysctl_writes_mode sysctl_writes_strict = SYSCTL_WRITES_STRICT; static int proc_do_cad_pid(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); @@ -416,7 +438,7 @@ static struct ctl_table kern_table[] = { }, { .procname = "sched_rr_timeslice_ms", - .data = &sched_rr_timeslice, + .data = &sysctl_sched_rr_timeslice, .maxlen = sizeof(int), .mode = 0644, .proc_handler = sched_rr_handler, @@ -879,6 +901,14 @@ static struct ctl_table kern_table[] = { #endif }, { + .procname = "watchdog_cpumask", + .data = &watchdog_cpumask_bits, + .maxlen = NR_CPUS, + .mode = 0644, + .proc_handler = proc_watchdog_cpumask, + }, +#ifdef CONFIG_SOFTLOCKUP_DETECTOR + { .procname = "soft_watchdog", .data = &soft_watchdog_enabled, .maxlen = sizeof (int), @@ -888,13 +918,6 @@ static struct ctl_table kern_table[] = { .extra2 = &one, }, { - .procname = "watchdog_cpumask", - .data = &watchdog_cpumask_bits, - .maxlen = NR_CPUS, - .mode = 0644, - .proc_handler = proc_watchdog_cpumask, - }, - { .procname = "softlockup_panic", .data = &softlockup_panic, .maxlen = sizeof(int), @@ -903,27 +926,29 @@ static struct ctl_table kern_table[] = { .extra1 = &zero, .extra2 = &one, }, -#ifdef CONFIG_HARDLOCKUP_DETECTOR +#ifdef CONFIG_SMP { - .procname = "hardlockup_panic", - .data = &hardlockup_panic, + .procname = "softlockup_all_cpu_backtrace", + .data = &sysctl_softlockup_all_cpu_backtrace, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one, }, +#endif /* CONFIG_SMP */ #endif -#ifdef CONFIG_SMP +#ifdef CONFIG_HARDLOCKUP_DETECTOR { - .procname = "softlockup_all_cpu_backtrace", - .data = &sysctl_softlockup_all_cpu_backtrace, + .procname = "hardlockup_panic", + .data = &hardlockup_panic, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, .extra1 = &zero, .extra2 = &one, }, +#ifdef CONFIG_SMP { .procname = "hardlockup_all_cpu_backtrace", .data = &sysctl_hardlockup_all_cpu_backtrace, @@ -935,6 +960,8 @@ static struct ctl_table kern_table[] = { }, #endif /* CONFIG_SMP */ #endif +#endif + #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) { .procname = "unknown_nmi_panic", @@ -1175,6 +1202,8 @@ static struct ctl_table kern_table[] = { .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = timer_migration_handler, + .extra1 = &zero, + .extra2 = &one, }, #endif #ifdef CONFIG_BPF_SYSCALL @@ -1947,6 +1976,32 @@ static void warn_sysctl_write(struct ctl_table *table) } /** + * proc_first_pos_non_zero_ignore - check if firs position is allowed + * @ppos: file position + * @table: the sysctl table + * + * Returns true if the first position is non-zero and the sysctl_writes_strict + * mode indicates this is not allowed for numeric input types. String proc + * hadlers can ignore the return value. + */ +static bool proc_first_pos_non_zero_ignore(loff_t *ppos, + struct ctl_table *table) +{ + if (!*ppos) + return false; + + switch (sysctl_writes_strict) { + case SYSCTL_WRITES_STRICT: + return true; + case SYSCTL_WRITES_WARN: + warn_sysctl_write(table); + return false; + default: + return false; + } +} + +/** * proc_dostring - read a string sysctl * @table: the sysctl table * @write: %TRUE if this is a write to the sysctl file @@ -1966,8 +2021,8 @@ static void warn_sysctl_write(struct ctl_table *table) int proc_dostring(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - if (write && *ppos && sysctl_writes_strict == SYSCTL_WRITES_WARN) - warn_sysctl_write(table); + if (write) + proc_first_pos_non_zero_ignore(ppos, table); return _proc_do_string((char *)(table->data), table->maxlen, write, (char __user *)buffer, lenp, ppos); @@ -2125,12 +2180,14 @@ static int do_proc_dointvec_conv(bool *negp, unsigned long *lvalp, return 0; } -static int do_proc_douintvec_conv(bool *negp, unsigned long *lvalp, - int *valp, - int write, void *data) +static int do_proc_douintvec_conv(unsigned long *lvalp, + unsigned int *valp, + int write, void *data) { if (write) { - if (*negp) + if (*lvalp > UINT_MAX) + return -EINVAL; + if (*lvalp > UINT_MAX) return -EINVAL; *valp = *lvalp; } else { @@ -2166,17 +2223,8 @@ static int __do_proc_dointvec(void *tbl_data, struct ctl_table *table, conv = do_proc_dointvec_conv; if (write) { - if (*ppos) { - switch (sysctl_writes_strict) { - case SYSCTL_WRITES_STRICT: - goto out; - case SYSCTL_WRITES_WARN: - warn_sysctl_write(table); - break; - default: - break; - } - } + if (proc_first_pos_non_zero_ignore(ppos, table)) + goto out; if (left > PAGE_SIZE - 1) left = PAGE_SIZE - 1; @@ -2243,6 +2291,146 @@ static int do_proc_dointvec(struct ctl_table *table, int write, buffer, lenp, ppos, conv, data); } +static int do_proc_douintvec_w(unsigned int *tbl_data, + struct ctl_table *table, + void __user *buffer, + size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + unsigned long lval; + int err = 0; + size_t left; + bool neg; + char *kbuf = NULL, *p; + + left = *lenp; + + if (proc_first_pos_non_zero_ignore(ppos, table)) + goto bail_early; + + if (left > PAGE_SIZE - 1) + left = PAGE_SIZE - 1; + + p = kbuf = memdup_user_nul(buffer, left); + if (IS_ERR(kbuf)) + return -EINVAL; + + left -= proc_skip_spaces(&p); + if (!left) { + err = -EINVAL; + goto out_free; + } + + err = proc_get_long(&p, &left, &lval, &neg, + proc_wspace_sep, + sizeof(proc_wspace_sep), NULL); + if (err || neg) { + err = -EINVAL; + goto out_free; + } + + if (conv(&lval, tbl_data, 1, data)) { + err = -EINVAL; + goto out_free; + } + + if (!err && left) + left -= proc_skip_spaces(&p); + +out_free: + kfree(kbuf); + if (err) + return -EINVAL; + + return 0; + + /* This is in keeping with old __do_proc_dointvec() */ +bail_early: + *ppos += *lenp; + return err; +} + +static int do_proc_douintvec_r(unsigned int *tbl_data, void __user *buffer, + size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + unsigned long lval; + int err = 0; + size_t left; + + left = *lenp; + + if (conv(&lval, tbl_data, 0, data)) { + err = -EINVAL; + goto out; + } + + err = proc_put_long(&buffer, &left, lval, false); + if (err || !left) + goto out; + + err = proc_put_char(&buffer, &left, '\n'); + +out: + *lenp -= left; + *ppos += *lenp; + + return err; +} + +static int __do_proc_douintvec(void *tbl_data, struct ctl_table *table, + int write, void __user *buffer, + size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + unsigned int *i, vleft; + + if (!tbl_data || !table->maxlen || !*lenp || (*ppos && !write)) { + *lenp = 0; + return 0; + } + + i = (unsigned int *) tbl_data; + vleft = table->maxlen / sizeof(*i); + + /* + * Arrays are not supported, keep this simple. *Do not* add + * support for them. + */ + if (vleft != 1) { + *lenp = 0; + return -EINVAL; + } + + if (!conv) + conv = do_proc_douintvec_conv; + + if (write) + return do_proc_douintvec_w(i, table, buffer, lenp, ppos, + conv, data); + return do_proc_douintvec_r(i, buffer, lenp, ppos, conv, data); +} + +static int do_proc_douintvec(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos, + int (*conv)(unsigned long *lvalp, + unsigned int *valp, + int write, void *data), + void *data) +{ + return __do_proc_douintvec(table->data, table, write, + buffer, lenp, ppos, conv, data); +} + /** * proc_dointvec - read a vector of integers * @table: the sysctl table @@ -2278,8 +2466,8 @@ int proc_dointvec(struct ctl_table *table, int write, int proc_douintvec(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - return do_proc_dointvec(table, write, buffer, lenp, ppos, - do_proc_douintvec_conv, NULL); + return do_proc_douintvec(table, write, buffer, lenp, ppos, + do_proc_douintvec_conv, NULL); } /* @@ -2384,6 +2572,65 @@ int proc_dointvec_minmax(struct ctl_table *table, int write, do_proc_dointvec_minmax_conv, ¶m); } +struct do_proc_douintvec_minmax_conv_param { + unsigned int *min; + unsigned int *max; +}; + +static int do_proc_douintvec_minmax_conv(unsigned long *lvalp, + unsigned int *valp, + int write, void *data) +{ + struct do_proc_douintvec_minmax_conv_param *param = data; + + if (write) { + unsigned int val = *lvalp; + + if ((param->min && *param->min > val) || + (param->max && *param->max < val)) + return -ERANGE; + + if (*lvalp > UINT_MAX) + return -EINVAL; + *valp = val; + } else { + unsigned int val = *valp; + *lvalp = (unsigned long) val; + } + + return 0; +} + +/** + * proc_douintvec_minmax - read a vector of unsigned ints with min/max values + * @table: the sysctl table + * @write: %TRUE if this is a write to the sysctl file + * @buffer: the user buffer + * @lenp: the size of the user buffer + * @ppos: file position + * + * Reads/writes up to table->maxlen/sizeof(unsigned int) unsigned integer + * values from/to the user buffer, treated as an ASCII string. Negative + * strings are not allowed. + * + * This routine will ensure the values are within the range specified by + * table->extra1 (min) and table->extra2 (max). There is a final sanity + * check for UINT_MAX to avoid having to support wrap around uses from + * userspace. + * + * Returns 0 on success. + */ +int proc_douintvec_minmax(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + struct do_proc_douintvec_minmax_conv_param param = { + .min = (unsigned int *) table->extra1, + .max = (unsigned int *) table->extra2, + }; + return do_proc_douintvec(table, write, buffer, lenp, ppos, + do_proc_douintvec_minmax_conv, ¶m); +} + static void validate_coredump_safety(void) { #ifdef CONFIG_COREDUMP @@ -2441,17 +2688,8 @@ static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int left = *lenp; if (write) { - if (*ppos) { - switch (sysctl_writes_strict) { - case SYSCTL_WRITES_STRICT: - goto out; - case SYSCTL_WRITES_WARN: - warn_sysctl_write(table); - break; - default: - break; - } - } + if (proc_first_pos_non_zero_ignore(ppos, table)) + goto out; if (left > PAGE_SIZE - 1) left = PAGE_SIZE - 1; @@ -2570,7 +2808,7 @@ static int do_proc_dointvec_jiffies_conv(bool *negp, unsigned long *lvalp, int write, void *data) { if (write) { - if (*lvalp > LONG_MAX / HZ) + if (*lvalp > INT_MAX / HZ) return 1; *valp = *negp ? -(*lvalp*HZ) : (*lvalp*HZ); } else { @@ -2892,6 +3130,12 @@ int proc_dointvec_minmax(struct ctl_table *table, int write, return -ENOSYS; } +int proc_douintvec_minmax(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + return -ENOSYS; +} + int proc_dointvec_jiffies(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { @@ -2934,6 +3178,7 @@ EXPORT_SYMBOL(proc_dointvec); EXPORT_SYMBOL(proc_douintvec); EXPORT_SYMBOL(proc_dointvec_jiffies); EXPORT_SYMBOL(proc_dointvec_minmax); +EXPORT_SYMBOL_GPL(proc_douintvec_minmax); EXPORT_SYMBOL(proc_dointvec_userhz_jiffies); EXPORT_SYMBOL(proc_dointvec_ms_jiffies); EXPORT_SYMBOL(proc_dostring); diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index ece4b177052b..02e1859f2ca8 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -1119,7 +1119,7 @@ static ssize_t bin_uuid(struct file *file, /* Only supports reads */ if (oldval && oldlen) { char buf[UUID_STRING_LEN + 1]; - uuid_be uuid; + uuid_t uuid; result = kernel_read(file, 0, buf, sizeof(buf) - 1); if (result < 0) @@ -1128,7 +1128,7 @@ static ssize_t bin_uuid(struct file *file, buf[result] = '\0'; result = -EIO; - if (uuid_be_to_bin(buf, &uuid)) + if (uuid_parse(buf, &uuid)) goto out; if (oldlen > 16) @@ -1346,7 +1346,7 @@ static void deprecated_sysctl_warning(const int *name, int nlen) * CTL_KERN/KERN_VERSION is used by older glibc and cannot * ever go away. */ - if (name[0] == CTL_KERN && name[1] == KERN_VERSION) + if (nlen >= 2 && name[0] == CTL_KERN && name[1] == KERN_VERSION) return; if (printk_ratelimit()) { diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 8a5e44236f78..4559e914452b 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -30,6 +30,7 @@ #include <linux/pid_namespace.h> #include <net/genetlink.h> #include <linux/atomic.h> +#include <linux/sched/cputime.h> /* * Maximum length of a cpumask that can be specified in @@ -210,6 +211,8 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats) struct task_struct *tsk, *first; unsigned long flags; int rc = -ESRCH; + u64 delta, utime, stime; + u64 start_time; /* * Add additional stats from live tasks except zombie thread group @@ -227,6 +230,7 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats) memset(stats, 0, sizeof(*stats)); tsk = first; + start_time = ktime_get_ns(); do { if (tsk->exit_state) continue; @@ -238,6 +242,16 @@ static int fill_stats_for_tgid(pid_t tgid, struct taskstats *stats) */ delayacct_add_tsk(stats, tsk); + /* calculate task elapsed time in nsec */ + delta = start_time - tsk->start_time; + /* Convert to micro seconds */ + do_div(delta, NSEC_PER_USEC); + stats->ac_etime += delta; + + task_cputime(tsk, &utime, &stime); + stats->ac_utime += div_u64(utime, NSEC_PER_USEC); + stats->ac_stime += div_u64(stime, NSEC_PER_USEC); + stats->nvcsw += tsk->nvcsw; stats->nivcsw += tsk->nivcsw; } while_each_thread(first, tsk); diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index 4008d9f95dd7..ac09bc29eb08 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -126,56 +126,6 @@ config NO_HZ_FULL_ALL Note the boot CPU will still be kept outside the range to handle the timekeeping duty. -config NO_HZ_FULL_SYSIDLE - bool "Detect full-system idle state for full dynticks system" - depends on NO_HZ_FULL - default n - help - At least one CPU must keep the scheduling-clock tick running for - timekeeping purposes whenever there is a non-idle CPU, where - "non-idle" also includes dynticks CPUs as long as they are - running non-idle tasks. Because the underlying adaptive-tick - support cannot distinguish between all CPUs being idle and - all CPUs each running a single task in dynticks mode, the - underlying support simply ensures that there is always a CPU - handling the scheduling-clock tick, whether or not all CPUs - are idle. This Kconfig option enables scalable detection of - the all-CPUs-idle state, thus allowing the scheduling-clock - tick to be disabled when all CPUs are idle. Note that scalable - detection of the all-CPUs-idle state means that larger systems - will be slower to declare the all-CPUs-idle state. - - Say Y if you would like to help debug all-CPUs-idle detection. - - Say N if you are unsure. - -config NO_HZ_FULL_SYSIDLE_SMALL - int "Number of CPUs above which large-system approach is used" - depends on NO_HZ_FULL_SYSIDLE - range 1 NR_CPUS - default 8 - help - The full-system idle detection mechanism takes a lazy approach - on large systems, as is required to attain decent scalability. - However, on smaller systems, scalability is not anywhere near as - large a concern as is energy efficiency. The sysidle subsystem - therefore uses a fast but non-scalable algorithm for small - systems and a lazier but scalable algorithm for large systems. - This Kconfig parameter defines the number of CPUs in the largest - system that will be considered to be "small". - - The default value will be fine in most cases. Battery-powered - systems that (1) enable NO_HZ_FULL_SYSIDLE, (2) have larger - numbers of CPUs, and (3) are suffering from battery-lifetime - problems due to long sysidle latencies might wish to experiment - with larger values for this Kconfig parameter. On the other - hand, they might be even better served by disabling NO_HZ_FULL - entirely, given that NO_HZ_FULL is intended for HPC and - real-time workloads that at present do not tend to be run on - battery-powered systems. - - Take the default if you are unsure. - config NO_HZ bool "Old Idle dynticks config" depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS diff --git a/kernel/time/Makefile b/kernel/time/Makefile index 976840d29a71..938dbf33ef49 100644 --- a/kernel/time/Makefile +++ b/kernel/time/Makefile @@ -15,6 +15,5 @@ ifeq ($(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST),y) endif obj-$(CONFIG_GENERIC_SCHED_CLOCK) += sched_clock.o obj-$(CONFIG_TICK_ONESHOT) += tick-oneshot.o tick-sched.o -obj-$(CONFIG_TIMER_STATS) += timer_stats.o obj-$(CONFIG_DEBUG_FS) += timekeeping_debug.o obj-$(CONFIG_TEST_UDELAY) += test_udelay.o diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index e6dc9a538efa..0b8ff7d257ea 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -19,12 +19,17 @@ #include <linux/hrtimer.h> #include <linux/timerqueue.h> #include <linux/rtc.h> +#include <linux/sched/signal.h> +#include <linux/sched/debug.h> #include <linux/alarmtimer.h> #include <linux/mutex.h> #include <linux/platform_device.h> #include <linux/posix-timers.h> #include <linux/workqueue.h> #include <linux/freezer.h> +#include <linux/compat.h> + +#include "posix-timers.h" #define CREATE_TRACE_POINTS #include <trace/events/alarmtimer.h> @@ -43,11 +48,13 @@ static struct alarm_base { clockid_t base_clockid; } alarm_bases[ALARM_NUMTYPE]; +#if defined(CONFIG_POSIX_TIMERS) || defined(CONFIG_RTC_CLASS) /* freezer information to handle clock_nanosleep triggered wakeups */ static enum alarmtimer_type freezer_alarmtype; static ktime_t freezer_expires; static ktime_t freezer_delta; static DEFINE_SPINLOCK(freezer_delta_lock); +#endif static struct wakeup_source *ws; @@ -305,38 +312,6 @@ static int alarmtimer_resume(struct device *dev) } #endif -static void alarmtimer_freezerset(ktime_t absexp, enum alarmtimer_type type) -{ - struct alarm_base *base; - unsigned long flags; - ktime_t delta; - - switch(type) { - case ALARM_REALTIME: - base = &alarm_bases[ALARM_REALTIME]; - type = ALARM_REALTIME_FREEZER; - break; - case ALARM_BOOTTIME: - base = &alarm_bases[ALARM_BOOTTIME]; - type = ALARM_BOOTTIME_FREEZER; - break; - default: - WARN_ONCE(1, "Invalid alarm type: %d\n", type); - return; - } - - delta = ktime_sub(absexp, base->gettime()); - - spin_lock_irqsave(&freezer_delta_lock, flags); - if (!freezer_delta || (delta < freezer_delta)) { - freezer_delta = delta; - freezer_expires = absexp; - freezer_alarmtype = type; - } - spin_unlock_irqrestore(&freezer_delta_lock, flags); -} - - /** * alarm_init - Initialize an alarm structure * @alarm: ptr to alarm to be initialized @@ -385,7 +360,7 @@ void alarm_start_relative(struct alarm *alarm, ktime_t start) { struct alarm_base *base = &alarm_bases[alarm->type]; - start = ktime_add(start, base->gettime()); + start = ktime_add_safe(start, base->gettime()); alarm_start(alarm, start); } EXPORT_SYMBOL_GPL(alarm_start_relative); @@ -473,7 +448,7 @@ u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval) overrun++; } - alarm->node.expires = ktime_add(alarm->node.expires, interval); + alarm->node.expires = ktime_add_safe(alarm->node.expires, interval); return overrun; } EXPORT_SYMBOL_GPL(alarm_forward); @@ -486,6 +461,38 @@ u64 alarm_forward_now(struct alarm *alarm, ktime_t interval) } EXPORT_SYMBOL_GPL(alarm_forward_now); +#ifdef CONFIG_POSIX_TIMERS + +static void alarmtimer_freezerset(ktime_t absexp, enum alarmtimer_type type) +{ + struct alarm_base *base; + unsigned long flags; + ktime_t delta; + + switch(type) { + case ALARM_REALTIME: + base = &alarm_bases[ALARM_REALTIME]; + type = ALARM_REALTIME_FREEZER; + break; + case ALARM_BOOTTIME: + base = &alarm_bases[ALARM_BOOTTIME]; + type = ALARM_BOOTTIME_FREEZER; + break; + default: + WARN_ONCE(1, "Invalid alarm type: %d\n", type); + return; + } + + delta = ktime_sub(absexp, base->gettime()); + + spin_lock_irqsave(&freezer_delta_lock, flags); + if (!freezer_delta || (delta < freezer_delta)) { + freezer_delta = delta; + freezer_expires = absexp; + freezer_alarmtype = type; + } + spin_unlock_irqrestore(&freezer_delta_lock, flags); +} /** * clock2alarm - helper that converts from clockid to alarmtypes @@ -509,22 +516,26 @@ static enum alarmtimer_type clock2alarm(clockid_t clockid) static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, ktime_t now) { - unsigned long flags; struct k_itimer *ptr = container_of(alarm, struct k_itimer, - it.alarm.alarmtimer); + it.alarm.alarmtimer); enum alarmtimer_restart result = ALARMTIMER_NORESTART; + unsigned long flags; + int si_private = 0; spin_lock_irqsave(&ptr->it_lock, flags); - if ((ptr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) { - if (IS_ENABLED(CONFIG_POSIX_TIMERS) && - posix_timer_event(ptr, 0) != 0) - ptr->it_overrun++; - } - /* Re-add periodic timers */ - if (ptr->it.alarm.interval) { - ptr->it_overrun += alarm_forward(alarm, now, - ptr->it.alarm.interval); + ptr->it_active = 0; + if (ptr->it_interval) + si_private = ++ptr->it_requeue_pending; + + if (posix_timer_event(ptr, si_private) && ptr->it_interval) { + /* + * Handle ignored signals and rearm the timer. This will go + * away once we handle ignored signals proper. + */ + ptr->it_overrun += alarm_forward_now(alarm, ptr->it_interval); + ++ptr->it_requeue_pending; + ptr->it_active = 1; result = ALARMTIMER_RESTART; } spin_unlock_irqrestore(&ptr->it_lock, flags); @@ -533,13 +544,79 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, } /** + * alarm_timer_rearm - Posix timer callback for rearming timer + * @timr: Pointer to the posixtimer data struct + */ +static void alarm_timer_rearm(struct k_itimer *timr) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + + timr->it_overrun += alarm_forward_now(alarm, timr->it_interval); + alarm_start(alarm, alarm->node.expires); +} + +/** + * alarm_timer_forward - Posix timer callback for forwarding timer + * @timr: Pointer to the posixtimer data struct + * @now: Current time to forward the timer against + */ +static int alarm_timer_forward(struct k_itimer *timr, ktime_t now) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + + return (int) alarm_forward(alarm, timr->it_interval, now); +} + +/** + * alarm_timer_remaining - Posix timer callback to retrieve remaining time + * @timr: Pointer to the posixtimer data struct + * @now: Current time to calculate against + */ +static ktime_t alarm_timer_remaining(struct k_itimer *timr, ktime_t now) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + + return ktime_sub(now, alarm->node.expires); +} + +/** + * alarm_timer_try_to_cancel - Posix timer callback to cancel a timer + * @timr: Pointer to the posixtimer data struct + */ +static int alarm_timer_try_to_cancel(struct k_itimer *timr) +{ + return alarm_try_to_cancel(&timr->it.alarm.alarmtimer); +} + +/** + * alarm_timer_arm - Posix timer callback to arm a timer + * @timr: Pointer to the posixtimer data struct + * @expires: The new expiry time + * @absolute: Expiry value is absolute time + * @sigev_none: Posix timer does not deliver signals + */ +static void alarm_timer_arm(struct k_itimer *timr, ktime_t expires, + bool absolute, bool sigev_none) +{ + struct alarm *alarm = &timr->it.alarm.alarmtimer; + struct alarm_base *base = &alarm_bases[alarm->type]; + + if (!absolute) + expires = ktime_add_safe(expires, base->gettime()); + if (sigev_none) + alarm->node.expires = expires; + else + alarm_start(&timr->it.alarm.alarmtimer, expires); +} + +/** * alarm_clock_getres - posix getres interface * @which_clock: clockid * @tp: timespec to fill * * Returns the granularity of underlying alarm base clock */ -static int alarm_clock_getres(const clockid_t which_clock, struct timespec *tp) +static int alarm_clock_getres(const clockid_t which_clock, struct timespec64 *tp) { if (!alarmtimer_get_rtcdev()) return -EINVAL; @@ -556,14 +633,14 @@ static int alarm_clock_getres(const clockid_t which_clock, struct timespec *tp) * * Provides the underlying alarm base time. */ -static int alarm_clock_get(clockid_t which_clock, struct timespec *tp) +static int alarm_clock_get(clockid_t which_clock, struct timespec64 *tp) { struct alarm_base *base = &alarm_bases[clock2alarm(which_clock)]; if (!alarmtimer_get_rtcdev()) return -EINVAL; - *tp = ktime_to_timespec(base->gettime()); + *tp = ktime_to_timespec64(base->gettime()); return 0; } @@ -589,89 +666,6 @@ static int alarm_timer_create(struct k_itimer *new_timer) } /** - * alarm_timer_get - posix timer_get interface - * @new_timer: k_itimer pointer - * @cur_setting: itimerspec data to fill - * - * Copies out the current itimerspec data - */ -static void alarm_timer_get(struct k_itimer *timr, - struct itimerspec *cur_setting) -{ - ktime_t relative_expiry_time = - alarm_expires_remaining(&(timr->it.alarm.alarmtimer)); - - if (ktime_to_ns(relative_expiry_time) > 0) { - cur_setting->it_value = ktime_to_timespec(relative_expiry_time); - } else { - cur_setting->it_value.tv_sec = 0; - cur_setting->it_value.tv_nsec = 0; - } - - cur_setting->it_interval = ktime_to_timespec(timr->it.alarm.interval); -} - -/** - * alarm_timer_del - posix timer_del interface - * @timr: k_itimer pointer to be deleted - * - * Cancels any programmed alarms for the given timer. - */ -static int alarm_timer_del(struct k_itimer *timr) -{ - if (!rtcdev) - return -ENOTSUPP; - - if (alarm_try_to_cancel(&timr->it.alarm.alarmtimer) < 0) - return TIMER_RETRY; - - return 0; -} - -/** - * alarm_timer_set - posix timer_set interface - * @timr: k_itimer pointer to be deleted - * @flags: timer flags - * @new_setting: itimerspec to be used - * @old_setting: itimerspec being replaced - * - * Sets the timer to new_setting, and starts the timer. - */ -static int alarm_timer_set(struct k_itimer *timr, int flags, - struct itimerspec *new_setting, - struct itimerspec *old_setting) -{ - ktime_t exp; - - if (!rtcdev) - return -ENOTSUPP; - - if (flags & ~TIMER_ABSTIME) - return -EINVAL; - - if (old_setting) - alarm_timer_get(timr, old_setting); - - /* If the timer was already set, cancel it */ - if (alarm_try_to_cancel(&timr->it.alarm.alarmtimer) < 0) - return TIMER_RETRY; - - /* start the timer */ - timr->it.alarm.interval = timespec_to_ktime(new_setting->it_interval); - exp = timespec_to_ktime(new_setting->it_value); - /* Convert (if necessary) to absolute time */ - if (flags != TIMER_ABSTIME) { - ktime_t now; - - now = alarm_bases[timr->it.alarm.alarmtimer.type].gettime(); - exp = ktime_add(now, exp); - } - - alarm_start(&timr->it.alarm.alarmtimer, exp); - return 0; -} - -/** * alarmtimer_nsleep_wakeup - Wakeup function for alarm_timer_nsleep * @alarm: ptr to alarm that fired * @@ -695,8 +689,10 @@ static enum alarmtimer_restart alarmtimer_nsleep_wakeup(struct alarm *alarm, * * Sets the alarm timer and sleeps until it is fired or interrupted. */ -static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp) +static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp, + enum alarmtimer_type type) { + struct restart_block *restart; alarm->data = (void *)current; do { set_current_state(TASK_INTERRUPTIBLE); @@ -709,36 +705,25 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp) __set_current_state(TASK_RUNNING); - return (alarm->data == NULL); -} - - -/** - * update_rmtp - Update remaining timespec value - * @exp: expiration time - * @type: timer type - * @rmtp: user pointer to remaining timepsec value - * - * Helper function that fills in rmtp value with time between - * now and the exp value - */ -static int update_rmtp(ktime_t exp, enum alarmtimer_type type, - struct timespec __user *rmtp) -{ - struct timespec rmt; - ktime_t rem; - - rem = ktime_sub(exp, alarm_bases[type].gettime()); - - if (rem <= 0) + if (!alarm->data) return 0; - rmt = ktime_to_timespec(rem); - if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) - return -EFAULT; + if (freezing(current)) + alarmtimer_freezerset(absexp, type); + restart = ¤t->restart_block; + if (restart->nanosleep.type != TT_NONE) { + struct timespec64 rmt; + ktime_t rem; + + rem = ktime_sub(absexp, alarm_bases[type].gettime()); - return 1; + if (rem <= 0) + return 0; + rmt = ktime_to_timespec64(rem); + return nanosleep_copyout(restart, &rmt); + } + return -ERESTART_RESTARTBLOCK; } /** @@ -750,32 +735,12 @@ static int update_rmtp(ktime_t exp, enum alarmtimer_type type, static long __sched alarm_timer_nsleep_restart(struct restart_block *restart) { enum alarmtimer_type type = restart->nanosleep.clockid; - ktime_t exp; - struct timespec __user *rmtp; + ktime_t exp = restart->nanosleep.expires; struct alarm alarm; - int ret = 0; - exp = restart->nanosleep.expires; alarm_init(&alarm, type, alarmtimer_nsleep_wakeup); - if (alarmtimer_do_nsleep(&alarm, exp)) - goto out; - - if (freezing(current)) - alarmtimer_freezerset(exp, type); - - rmtp = restart->nanosleep.rmtp; - if (rmtp) { - ret = update_rmtp(exp, type, rmtp); - if (ret <= 0) - goto out; - } - - - /* The other values in restart are already filled in */ - ret = -ERESTART_RESTARTBLOCK; -out: - return ret; + return alarmtimer_do_nsleep(&alarm, exp, type); } /** @@ -788,13 +753,13 @@ out: * Handles clock_nanosleep calls against _ALARM clockids */ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, - struct timespec *tsreq, struct timespec __user *rmtp) + const struct timespec64 *tsreq) { enum alarmtimer_type type = clock2alarm(which_clock); + struct restart_block *restart = ¤t->restart_block; struct alarm alarm; ktime_t exp; int ret = 0; - struct restart_block *restart; if (!alarmtimer_get_rtcdev()) return -ENOTSUPP; @@ -807,42 +772,43 @@ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, alarm_init(&alarm, type, alarmtimer_nsleep_wakeup); - exp = timespec_to_ktime(*tsreq); + exp = timespec64_to_ktime(*tsreq); /* Convert (if necessary) to absolute time */ if (flags != TIMER_ABSTIME) { ktime_t now = alarm_bases[type].gettime(); exp = ktime_add(now, exp); } - if (alarmtimer_do_nsleep(&alarm, exp)) - goto out; - - if (freezing(current)) - alarmtimer_freezerset(exp, type); + ret = alarmtimer_do_nsleep(&alarm, exp, type); + if (ret != -ERESTART_RESTARTBLOCK) + return ret; /* abs timers don't set remaining time or restart */ - if (flags == TIMER_ABSTIME) { - ret = -ERESTARTNOHAND; - goto out; - } + if (flags == TIMER_ABSTIME) + return -ERESTARTNOHAND; - if (rmtp) { - ret = update_rmtp(exp, type, rmtp); - if (ret <= 0) - goto out; - } - - restart = ¤t->restart_block; restart->fn = alarm_timer_nsleep_restart; restart->nanosleep.clockid = type; restart->nanosleep.expires = exp; - restart->nanosleep.rmtp = rmtp; - ret = -ERESTART_RESTARTBLOCK; - -out: return ret; } +const struct k_clock alarm_clock = { + .clock_getres = alarm_clock_getres, + .clock_get = alarm_clock_get, + .timer_create = alarm_timer_create, + .timer_set = common_timer_set, + .timer_del = common_timer_del, + .timer_get = common_timer_get, + .timer_arm = alarm_timer_arm, + .timer_rearm = alarm_timer_rearm, + .timer_forward = alarm_timer_forward, + .timer_remaining = alarm_timer_remaining, + .timer_try_to_cancel = alarm_timer_try_to_cancel, + .nsleep = alarm_timer_nsleep, +}; +#endif /* CONFIG_POSIX_TIMERS */ + /* Suspend hook structures */ static const struct dev_pm_ops alarmtimer_pm_ops = { @@ -868,23 +834,9 @@ static int __init alarmtimer_init(void) struct platform_device *pdev; int error = 0; int i; - struct k_clock alarm_clock = { - .clock_getres = alarm_clock_getres, - .clock_get = alarm_clock_get, - .timer_create = alarm_timer_create, - .timer_set = alarm_timer_set, - .timer_del = alarm_timer_del, - .timer_get = alarm_timer_get, - .nsleep = alarm_timer_nsleep, - }; alarmtimer_rtc_timer_init(); - if (IS_ENABLED(CONFIG_POSIX_TIMERS)) { - posix_timers_register_clock(CLOCK_REALTIME_ALARM, &alarm_clock); - posix_timers_register_clock(CLOCK_BOOTTIME_ALARM, &alarm_clock); - } - /* Initialize alarm bases */ alarm_bases[ALARM_REALTIME].base_clockid = CLOCK_REALTIME; alarm_bases[ALARM_REALTIME].gettime = &ktime_get_real; diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 97ac0951f164..4237e0744e26 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -468,7 +468,7 @@ void clockevents_register_device(struct clock_event_device *dev) } EXPORT_SYMBOL_GPL(clockevents_register_device); -void clockevents_config(struct clock_event_device *dev, u32 freq) +static void clockevents_config(struct clock_event_device *dev, u32 freq) { u64 sec; diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 665985b0a89a..03918a19cf2d 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -141,6 +141,10 @@ static void __clocksource_unstable(struct clocksource *cs) { cs->flags &= ~(CLOCK_SOURCE_VALID_FOR_HRES | CLOCK_SOURCE_WATCHDOG); cs->flags |= CLOCK_SOURCE_UNSTABLE; + + if (cs->mark_unstable) + cs->mark_unstable(cs); + if (finished_booting) schedule_work(&watchdog_work); } @@ -229,6 +233,9 @@ static void clocksource_watchdog(unsigned long data) continue; } + if (cs == curr_clocksource && cs->tick_stable) + cs->tick_stable(cs); + if (!(cs->flags & CLOCK_SOURCE_VALID_FOR_HRES) && (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS) && (watchdog->flags & CLOCK_SOURCE_IS_CONTINUOUS)) { diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index c6ecedd3b839..88f75f92ef36 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -43,12 +43,15 @@ #include <linux/seq_file.h> #include <linux/err.h> #include <linux/debugobjects.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> #include <linux/sched/sysctl.h> #include <linux/sched/rt.h> #include <linux/sched/deadline.h> +#include <linux/sched/nohz.h> +#include <linux/sched/debug.h> #include <linux/timer.h> #include <linux/freezer.h> +#include <linux/compat.h> #include <linux/uaccess.h> @@ -94,17 +97,15 @@ DEFINE_PER_CPU(struct hrtimer_cpu_base, hrtimer_bases) = }; static const int hrtimer_clock_to_base_table[MAX_CLOCKS] = { + /* Make sure we catch unsupported clockids */ + [0 ... MAX_CLOCKS - 1] = HRTIMER_MAX_CLOCK_BASES, + [CLOCK_REALTIME] = HRTIMER_BASE_REALTIME, [CLOCK_MONOTONIC] = HRTIMER_BASE_MONOTONIC, [CLOCK_BOOTTIME] = HRTIMER_BASE_BOOTTIME, [CLOCK_TAI] = HRTIMER_BASE_TAI, }; -static inline int hrtimer_clockid_to_base(clockid_t clock_id) -{ - return hrtimer_clock_to_base_table[clock_id]; -} - /* * Functions and macros which are different for UP/SMP systems are kept in a * single place @@ -766,34 +767,6 @@ void hrtimers_resume(void) clock_was_set_delayed(); } -static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer) -{ -#ifdef CONFIG_TIMER_STATS - if (timer->start_site) - return; - timer->start_site = __builtin_return_address(0); - memcpy(timer->start_comm, current->comm, TASK_COMM_LEN); - timer->start_pid = current->pid; -#endif -} - -static inline void timer_stats_hrtimer_clear_start_info(struct hrtimer *timer) -{ -#ifdef CONFIG_TIMER_STATS - timer->start_site = NULL; -#endif -} - -static inline void timer_stats_account_hrtimer(struct hrtimer *timer) -{ -#ifdef CONFIG_TIMER_STATS - if (likely(!timer_stats_active)) - return; - timer_stats_update_stats(timer, timer->start_pid, timer->start_site, - timer->function, timer->start_comm, 0); -#endif -} - /* * Counterpart to lock_hrtimer_base above: */ @@ -932,7 +905,6 @@ remove_hrtimer(struct hrtimer *timer, struct hrtimer_clock_base *base, bool rest * rare case and less expensive than a smp call. */ debug_deactivate(timer); - timer_stats_hrtimer_clear_start_info(timer); reprogram = base->cpu_base == this_cpu_ptr(&hrtimer_bases); if (!restart) @@ -990,8 +962,6 @@ void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, /* Switch the timer base, if necessary: */ new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED); - timer_stats_hrtimer_set_start_info(timer); - leftmost = enqueue_hrtimer(timer, new_base); if (!leftmost) goto unlock; @@ -1018,7 +988,7 @@ EXPORT_SYMBOL_GPL(hrtimer_start_range_ns); * Returns: * 0 when the timer was not active * 1 when the timer was active - * -1 when the timer is currently excuting the callback function and + * -1 when the timer is currently executing the callback function and * cannot be stopped */ int hrtimer_try_to_cancel(struct hrtimer *timer) @@ -1112,6 +1082,18 @@ u64 hrtimer_get_next_event(void) } #endif +static inline int hrtimer_clockid_to_base(clockid_t clock_id) +{ + if (likely(clock_id < MAX_CLOCKS)) { + int base = hrtimer_clock_to_base_table[clock_id]; + + if (likely(base != HRTIMER_MAX_CLOCK_BASES)) + return base; + } + WARN(1, "Invalid clockid %d. Using MONOTONIC\n", clock_id); + return HRTIMER_BASE_MONOTONIC; +} + static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, enum hrtimer_mode mode) { @@ -1128,12 +1110,6 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, base = hrtimer_clockid_to_base(clock_id); timer->base = &cpu_base->clock_base[base]; timerqueue_init(&timer->node); - -#ifdef CONFIG_TIMER_STATS - timer->start_site = NULL; - timer->start_pid = -1; - memset(timer->start_comm, 0, TASK_COMM_LEN); -#endif } /** @@ -1217,7 +1193,6 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base, raw_write_seqcount_barrier(&cpu_base->seq); __remove_hrtimer(timer, base, HRTIMER_STATE_INACTIVE, 0); - timer_stats_account_hrtimer(timer); fn = timer->function; /* @@ -1394,10 +1369,7 @@ retry: ktime_to_ns(delta)); } -/* - * local version of hrtimer_peek_ahead_timers() called with interrupts - * disabled. - */ +/* called with interrupts disabled */ static inline void __hrtimer_peek_ahead_timers(void) { struct tick_device *td; @@ -1468,8 +1440,29 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task) } EXPORT_SYMBOL_GPL(hrtimer_init_sleeper); +int nanosleep_copyout(struct restart_block *restart, struct timespec64 *ts) +{ + switch(restart->nanosleep.type) { +#ifdef CONFIG_COMPAT + case TT_COMPAT: + if (compat_put_timespec64(ts, restart->nanosleep.compat_rmtp)) + return -EFAULT; + break; +#endif + case TT_NATIVE: + if (put_timespec64(ts, restart->nanosleep.rmtp)) + return -EFAULT; + break; + default: + BUG(); + } + return -ERESTART_RESTARTBLOCK; +} + static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode) { + struct restart_block *restart; + hrtimer_init_sleeper(t, current); do { @@ -1486,53 +1479,38 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod __set_current_state(TASK_RUNNING); - return t->task == NULL; -} - -static int update_rmtp(struct hrtimer *timer, struct timespec __user *rmtp) -{ - struct timespec rmt; - ktime_t rem; - - rem = hrtimer_expires_remaining(timer); - if (rem <= 0) + if (!t->task) return 0; - rmt = ktime_to_timespec(rem); - if (copy_to_user(rmtp, &rmt, sizeof(*rmtp))) - return -EFAULT; + restart = ¤t->restart_block; + if (restart->nanosleep.type != TT_NONE) { + ktime_t rem = hrtimer_expires_remaining(&t->timer); + struct timespec64 rmt; + + if (rem <= 0) + return 0; + rmt = ktime_to_timespec64(rem); - return 1; + return nanosleep_copyout(restart, &rmt); + } + return -ERESTART_RESTARTBLOCK; } -long __sched hrtimer_nanosleep_restart(struct restart_block *restart) +static long __sched hrtimer_nanosleep_restart(struct restart_block *restart) { struct hrtimer_sleeper t; - struct timespec __user *rmtp; - int ret = 0; + int ret; hrtimer_init_on_stack(&t.timer, restart->nanosleep.clockid, HRTIMER_MODE_ABS); hrtimer_set_expires_tv64(&t.timer, restart->nanosleep.expires); - if (do_nanosleep(&t, HRTIMER_MODE_ABS)) - goto out; - - rmtp = restart->nanosleep.rmtp; - if (rmtp) { - ret = update_rmtp(&t.timer, rmtp); - if (ret <= 0) - goto out; - } - - /* The other values in restart are already filled in */ - ret = -ERESTART_RESTARTBLOCK; -out: + ret = do_nanosleep(&t, HRTIMER_MODE_ABS); destroy_hrtimer_on_stack(&t.timer); return ret; } -long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp, +long hrtimer_nanosleep(const struct timespec64 *rqtp, const enum hrtimer_mode mode, const clockid_t clockid) { struct restart_block *restart; @@ -1545,8 +1523,9 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp, slack = 0; hrtimer_init_on_stack(&t.timer, clockid, mode); - hrtimer_set_expires_range_ns(&t.timer, timespec_to_ktime(*rqtp), slack); - if (do_nanosleep(&t, mode)) + hrtimer_set_expires_range_ns(&t.timer, timespec64_to_ktime(*rqtp), slack); + ret = do_nanosleep(&t, mode); + if (ret != -ERESTART_RESTARTBLOCK) goto out; /* Absolute timers do not update the rmtp value and restart: */ @@ -1555,19 +1534,10 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp, goto out; } - if (rmtp) { - ret = update_rmtp(&t.timer, rmtp); - if (ret <= 0) - goto out; - } - restart = ¤t->restart_block; restart->fn = hrtimer_nanosleep_restart; restart->nanosleep.clockid = t.timer.base->clockid; - restart->nanosleep.rmtp = rmtp; restart->nanosleep.expires = hrtimer_get_expires_tv64(&t.timer); - - ret = -ERESTART_RESTARTBLOCK; out: destroy_hrtimer_on_stack(&t.timer); return ret; @@ -1576,16 +1546,37 @@ out: SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp, struct timespec __user *, rmtp) { - struct timespec tu; + struct timespec64 tu; + + if (get_timespec64(&tu, rqtp)) + return -EFAULT; + + if (!timespec64_valid(&tu)) + return -EINVAL; + + current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; + current->restart_block.nanosleep.rmtp = rmtp; + return hrtimer_nanosleep(&tu, HRTIMER_MODE_REL, CLOCK_MONOTONIC); +} + +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE2(nanosleep, struct compat_timespec __user *, rqtp, + struct compat_timespec __user *, rmtp) +{ + struct timespec64 tu; - if (copy_from_user(&tu, rqtp, sizeof(tu))) + if (compat_get_timespec64(&tu, rqtp)) return -EFAULT; - if (!timespec_valid(&tu)) + if (!timespec64_valid(&tu)) return -EINVAL; - return hrtimer_nanosleep(&tu, rmtp, HRTIMER_MODE_REL, CLOCK_MONOTONIC); + current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; + current->restart_block.nanosleep.compat_rmtp = rmtp; + return hrtimer_nanosleep(&tu, HRTIMER_MODE_REL, CLOCK_MONOTONIC); } +#endif /* * Functions related to boot-time initialization: diff --git a/kernel/time/itimer.c b/kernel/time/itimer.c index 8c89143f9ebf..2ef98a02376a 100644 --- a/kernel/time/itimer.c +++ b/kernel/time/itimer.c @@ -10,9 +10,12 @@ #include <linux/interrupt.h> #include <linux/syscalls.h> #include <linux/time.h> +#include <linux/sched/signal.h> +#include <linux/sched/cputime.h> #include <linux/posix-timers.h> #include <linux/hrtimer.h> #include <trace/events/timer.h> +#include <linux/compat.h> #include <linux/uaccess.h> @@ -45,16 +48,16 @@ static struct timeval itimer_get_remtime(struct hrtimer *timer) static void get_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, struct itimerval *const value) { - cputime_t cval, cinterval; + u64 val, interval; struct cpu_itimer *it = &tsk->signal->it[clock_id]; spin_lock_irq(&tsk->sighand->siglock); - cval = it->expires; - cinterval = it->incr; - if (cval) { + val = it->expires; + interval = it->incr; + if (val) { struct task_cputime cputime; - cputime_t t; + u64 t; thread_group_cputimer(tsk, &cputime); if (clock_id == CPUCLOCK_PROF) @@ -63,17 +66,17 @@ static void get_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, /* CPUCLOCK_VIRT */ t = cputime.utime; - if (cval < t) + if (val < t) /* about to fire */ - cval = cputime_one_jiffy; + val = TICK_NSEC; else - cval = cval - t; + val -= t; } spin_unlock_irq(&tsk->sighand->siglock); - cputime_to_timeval(cval, &value->it_value); - cputime_to_timeval(cinterval, &value->it_interval); + value->it_value = ns_to_timeval(val); + value->it_interval = ns_to_timeval(interval); } int do_getitimer(int which, struct itimerval *value) @@ -114,6 +117,19 @@ SYSCALL_DEFINE2(getitimer, int, which, struct itimerval __user *, value) return error; } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(getitimer, int, which, + struct compat_itimerval __user *, it) +{ + struct itimerval kit; + int error = do_getitimer(which, &kit); + + if (!error && put_compat_itimerval(it, &kit)) + error = -EFAULT; + return error; +} +#endif + /* * The timer is automagically restarted, when interval != 0 @@ -129,55 +145,39 @@ enum hrtimer_restart it_real_fn(struct hrtimer *timer) return HRTIMER_NORESTART; } -static inline u32 cputime_sub_ns(cputime_t ct, s64 real_ns) -{ - struct timespec ts; - s64 cpu_ns; - - cputime_to_timespec(ct, &ts); - cpu_ns = timespec_to_ns(&ts); - - return (cpu_ns <= real_ns) ? 0 : cpu_ns - real_ns; -} - static void set_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, const struct itimerval *const value, struct itimerval *const ovalue) { - cputime_t cval, nval, cinterval, ninterval; - s64 ns_ninterval, ns_nval; - u32 error, incr_error; + u64 oval, nval, ointerval, ninterval; struct cpu_itimer *it = &tsk->signal->it[clock_id]; - nval = timeval_to_cputime(&value->it_value); - ns_nval = timeval_to_ns(&value->it_value); - ninterval = timeval_to_cputime(&value->it_interval); - ns_ninterval = timeval_to_ns(&value->it_interval); - - error = cputime_sub_ns(nval, ns_nval); - incr_error = cputime_sub_ns(ninterval, ns_ninterval); + /* + * Use the to_ktime conversion because that clamps the maximum + * value to KTIME_MAX and avoid multiplication overflows. + */ + nval = ktime_to_ns(timeval_to_ktime(value->it_value)); + ninterval = ktime_to_ns(timeval_to_ktime(value->it_interval)); spin_lock_irq(&tsk->sighand->siglock); - cval = it->expires; - cinterval = it->incr; - if (cval || nval) { + oval = it->expires; + ointerval = it->incr; + if (oval || nval) { if (nval > 0) - nval += cputime_one_jiffy; - set_process_cpu_timer(tsk, clock_id, &nval, &cval); + nval += TICK_NSEC; + set_process_cpu_timer(tsk, clock_id, &nval, &oval); } it->expires = nval; it->incr = ninterval; - it->error = error; - it->incr_error = incr_error; trace_itimer_state(clock_id == CPUCLOCK_VIRT ? ITIMER_VIRTUAL : ITIMER_PROF, value, nval); spin_unlock_irq(&tsk->sighand->siglock); if (ovalue) { - cputime_to_timeval(cval, &ovalue->it_value); - cputime_to_timeval(cinterval, &ovalue->it_interval); + ovalue->it_value = ns_to_timeval(oval); + ovalue->it_interval = ns_to_timeval(ointerval); } } @@ -312,3 +312,27 @@ SYSCALL_DEFINE3(setitimer, int, which, struct itimerval __user *, value, return -EFAULT; return 0; } + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE3(setitimer, int, which, + struct compat_itimerval __user *, in, + struct compat_itimerval __user *, out) +{ + struct itimerval kin, kout; + int error; + + if (in) { + if (get_compat_itimerval(&kin, in)) + return -EFAULT; + } else { + memset(&kin, 0, sizeof(kin)); + } + + error = do_setitimer(which, &kin, out ? &kout : NULL); + if (error || !out) + return error; + if (put_compat_itimerval(out, &kout)) + return -EFAULT; + return 0; +} +#endif diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index a4a0e478e44d..497719127bf9 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -27,19 +27,8 @@ #include "timekeeping.h" -/* The Jiffies based clocksource is the lowest common - * denominator clock source which should function on - * all systems. It has the same coarse resolution as - * the timer interrupt frequency HZ and it suffers - * inaccuracies caused by missed or lost timer - * interrupts and the inability for the timer - * interrupt hardware to accuratly tick at the - * requested HZ value. It is also not recommended - * for "tick-less" systems. - */ -#define NSEC_PER_JIFFY ((NSEC_PER_SEC+HZ/2)/HZ) -/* Since jiffies uses a simple NSEC_PER_JIFFY multiplier +/* Since jiffies uses a simple TICK_NSEC multiplier * conversion, the .shift value could be zero. However * this would make NTP adjustments impossible as they are * in units of 1/2^.shift. Thus we use JIFFIES_SHIFT to @@ -47,8 +36,8 @@ * amount, and give ntp adjustments in units of 1/2^8 * * The value 8 is somewhat carefully chosen, as anything - * larger can result in overflows. NSEC_PER_JIFFY grows as - * HZ shrinks, so values greater than 8 overflow 32bits when + * larger can result in overflows. TICK_NSEC grows as HZ + * shrinks, so values greater than 8 overflow 32bits when * HZ=100. */ #if HZ < 34 @@ -64,12 +53,23 @@ static u64 jiffies_read(struct clocksource *cs) return (u64) jiffies; } +/* + * The Jiffies based clocksource is the lowest common + * denominator clock source which should function on + * all systems. It has the same coarse resolution as + * the timer interrupt frequency HZ and it suffers + * inaccuracies caused by missed or lost timer + * interrupts and the inability for the timer + * interrupt hardware to accuratly tick at the + * requested HZ value. It is also not recommended + * for "tick-less" systems. + */ static struct clocksource clocksource_jiffies = { .name = "jiffies", .rating = 1, /* lowest valid rating*/ .read = jiffies_read, .mask = CLOCKSOURCE_MASK(32), - .mult = NSEC_PER_JIFFY << JIFFIES_SHIFT, /* details above */ + .mult = TICK_NSEC << JIFFIES_SHIFT, /* details above */ .shift = JIFFIES_SHIFT, .max_cycles = 10, }; diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c index 9cff0ab82b63..17cdc554c9fe 100644 --- a/kernel/time/posix-clock.c +++ b/kernel/time/posix-clock.c @@ -25,6 +25,8 @@ #include <linux/syscalls.h> #include <linux/uaccess.h> +#include "posix-timers.h" + static void delete_clock(struct kref *kref); /* @@ -82,38 +84,6 @@ static unsigned int posix_clock_poll(struct file *fp, poll_table *wait) return result; } -static int posix_clock_fasync(int fd, struct file *fp, int on) -{ - struct posix_clock *clk = get_posix_clock(fp); - int err = 0; - - if (!clk) - return -ENODEV; - - if (clk->ops.fasync) - err = clk->ops.fasync(clk, fd, fp, on); - - put_posix_clock(clk); - - return err; -} - -static int posix_clock_mmap(struct file *fp, struct vm_area_struct *vma) -{ - struct posix_clock *clk = get_posix_clock(fp); - int err = -ENODEV; - - if (!clk) - return -ENODEV; - - if (clk->ops.mmap) - err = clk->ops.mmap(clk, vma); - - put_posix_clock(clk); - - return err; -} - static long posix_clock_ioctl(struct file *fp, unsigned int cmd, unsigned long arg) { @@ -199,8 +169,6 @@ static const struct file_operations posix_clock_file_operations = { .unlocked_ioctl = posix_clock_ioctl, .open = posix_clock_open, .release = posix_clock_release, - .fasync = posix_clock_fasync, - .mmap = posix_clock_mmap, #ifdef CONFIG_COMPAT .compat_ioctl = posix_clock_compat_ioctl, #endif @@ -297,7 +265,7 @@ out: return err; } -static int pc_clock_gettime(clockid_t id, struct timespec *ts) +static int pc_clock_gettime(clockid_t id, struct timespec64 *ts) { struct posix_clock_desc cd; int err; @@ -316,7 +284,7 @@ static int pc_clock_gettime(clockid_t id, struct timespec *ts) return err; } -static int pc_clock_getres(clockid_t id, struct timespec *ts) +static int pc_clock_getres(clockid_t id, struct timespec64 *ts) { struct posix_clock_desc cd; int err; @@ -335,7 +303,7 @@ static int pc_clock_getres(clockid_t id, struct timespec *ts) return err; } -static int pc_clock_settime(clockid_t id, const struct timespec *ts) +static int pc_clock_settime(clockid_t id, const struct timespec64 *ts) { struct posix_clock_desc cd; int err; @@ -359,88 +327,9 @@ out: return err; } -static int pc_timer_create(struct k_itimer *kit) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - int err; - - err = get_clock_desc(id, &cd); - if (err) - return err; - - if (cd.clk->ops.timer_create) - err = cd.clk->ops.timer_create(cd.clk, kit); - else - err = -EOPNOTSUPP; - - put_clock_desc(&cd); - - return err; -} - -static int pc_timer_delete(struct k_itimer *kit) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - int err; - - err = get_clock_desc(id, &cd); - if (err) - return err; - - if (cd.clk->ops.timer_delete) - err = cd.clk->ops.timer_delete(cd.clk, kit); - else - err = -EOPNOTSUPP; - - put_clock_desc(&cd); - - return err; -} - -static void pc_timer_gettime(struct k_itimer *kit, struct itimerspec *ts) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - - if (get_clock_desc(id, &cd)) - return; - - if (cd.clk->ops.timer_gettime) - cd.clk->ops.timer_gettime(cd.clk, kit, ts); - - put_clock_desc(&cd); -} - -static int pc_timer_settime(struct k_itimer *kit, int flags, - struct itimerspec *ts, struct itimerspec *old) -{ - clockid_t id = kit->it_clock; - struct posix_clock_desc cd; - int err; - - err = get_clock_desc(id, &cd); - if (err) - return err; - - if (cd.clk->ops.timer_settime) - err = cd.clk->ops.timer_settime(cd.clk, kit, flags, ts, old); - else - err = -EOPNOTSUPP; - - put_clock_desc(&cd); - - return err; -} - -struct k_clock clock_posix_dynamic = { +const struct k_clock clock_posix_dynamic = { .clock_getres = pc_clock_getres, .clock_set = pc_clock_settime, .clock_get = pc_clock_gettime, .clock_adj = pc_clock_adjtime, - .timer_create = pc_timer_create, - .timer_set = pc_timer_settime, - .timer_del = pc_timer_delete, - .timer_get = pc_timer_gettime, }; diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index e9e8c10f0d9a..a3bd5dbe0dc4 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -2,7 +2,8 @@ * Implement CPU time clocks for the POSIX clock interface. */ -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/cputime.h> #include <linux/posix-timers.h> #include <linux/errno.h> #include <linux/math64.h> @@ -11,6 +12,11 @@ #include <trace/events/timer.h> #include <linux/tick.h> #include <linux/workqueue.h> +#include <linux/compat.h> + +#include "posix-timers.h" + +static void posix_cpu_timer_rearm(struct k_itimer *timer); /* * Called after updating RLIMIT_CPU to run cpu timer and update @@ -20,10 +26,10 @@ */ void update_rlimit_cpu(struct task_struct *task, unsigned long rlim_new) { - cputime_t cputime = secs_to_cputime(rlim_new); + u64 nsecs = rlim_new * NSEC_PER_SEC; spin_lock_irq(&task->sighand->siglock); - set_process_cpu_timer(task, CPUCLOCK_PROF, &cputime, NULL); + set_process_cpu_timer(task, CPUCLOCK_PROF, &nsecs, NULL); spin_unlock_irq(&task->sighand->siglock); } @@ -50,39 +56,14 @@ static int check_clock(const clockid_t which_clock) return error; } -static inline unsigned long long -timespec_to_sample(const clockid_t which_clock, const struct timespec *tp) -{ - unsigned long long ret; - - ret = 0; /* high half always zero when .cpu used */ - if (CPUCLOCK_WHICH(which_clock) == CPUCLOCK_SCHED) { - ret = (unsigned long long)tp->tv_sec * NSEC_PER_SEC + tp->tv_nsec; - } else { - ret = cputime_to_expires(timespec_to_cputime(tp)); - } - return ret; -} - -static void sample_to_timespec(const clockid_t which_clock, - unsigned long long expires, - struct timespec *tp) -{ - if (CPUCLOCK_WHICH(which_clock) == CPUCLOCK_SCHED) - *tp = ns_to_timespec(expires); - else - cputime_to_timespec((__force cputime_t)expires, tp); -} - /* * Update expiry time from increment, and increase overrun count, * given the current clock sample. */ -static void bump_cpu_timer(struct k_itimer *timer, - unsigned long long now) +static void bump_cpu_timer(struct k_itimer *timer, u64 now) { int i; - unsigned long long delta, incr; + u64 delta, incr; if (timer->it.cpu.incr == 0) return; @@ -122,25 +103,25 @@ static inline int task_cputime_zero(const struct task_cputime *cputime) return 0; } -static inline unsigned long long prof_ticks(struct task_struct *p) +static inline u64 prof_ticks(struct task_struct *p) { - cputime_t utime, stime; + u64 utime, stime; task_cputime(p, &utime, &stime); - return cputime_to_expires(utime + stime); + return utime + stime; } -static inline unsigned long long virt_ticks(struct task_struct *p) +static inline u64 virt_ticks(struct task_struct *p) { - cputime_t utime, stime; + u64 utime, stime; task_cputime(p, &utime, &stime); - return cputime_to_expires(utime); + return utime; } static int -posix_cpu_clock_getres(const clockid_t which_clock, struct timespec *tp) +posix_cpu_clock_getres(const clockid_t which_clock, struct timespec64 *tp) { int error = check_clock(which_clock); if (!error) { @@ -159,7 +140,7 @@ posix_cpu_clock_getres(const clockid_t which_clock, struct timespec *tp) } static int -posix_cpu_clock_set(const clockid_t which_clock, const struct timespec *tp) +posix_cpu_clock_set(const clockid_t which_clock, const struct timespec64 *tp) { /* * You can never reset a CPU clock, but we check for other errors @@ -176,8 +157,8 @@ posix_cpu_clock_set(const clockid_t which_clock, const struct timespec *tp) /* * Sample a per-thread clock for the given task. */ -static int cpu_clock_sample(const clockid_t which_clock, struct task_struct *p, - unsigned long long *sample) +static int cpu_clock_sample(const clockid_t which_clock, + struct task_struct *p, u64 *sample) { switch (CPUCLOCK_WHICH(which_clock)) { default: @@ -260,7 +241,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) */ static int cpu_clock_sample_group(const clockid_t which_clock, struct task_struct *p, - unsigned long long *sample) + u64 *sample) { struct task_cputime cputime; @@ -269,11 +250,11 @@ static int cpu_clock_sample_group(const clockid_t which_clock, return -EINVAL; case CPUCLOCK_PROF: thread_group_cputime(p, &cputime); - *sample = cputime_to_expires(cputime.utime + cputime.stime); + *sample = cputime.utime + cputime.stime; break; case CPUCLOCK_VIRT: thread_group_cputime(p, &cputime); - *sample = cputime_to_expires(cputime.utime); + *sample = cputime.utime; break; case CPUCLOCK_SCHED: thread_group_cputime(p, &cputime); @@ -285,10 +266,10 @@ static int cpu_clock_sample_group(const clockid_t which_clock, static int posix_cpu_clock_get_task(struct task_struct *tsk, const clockid_t which_clock, - struct timespec *tp) + struct timespec64 *tp) { int err = -EINVAL; - unsigned long long rtn; + u64 rtn; if (CPUCLOCK_PERTHREAD(which_clock)) { if (same_thread_group(tsk, current)) @@ -299,13 +280,13 @@ static int posix_cpu_clock_get_task(struct task_struct *tsk, } if (!err) - sample_to_timespec(which_clock, rtn, tp); + *tp = ns_to_timespec64(rtn); return err; } -static int posix_cpu_clock_get(const clockid_t which_clock, struct timespec *tp) +static int posix_cpu_clock_get(const clockid_t which_clock, struct timespec64 *tp) { const pid_t pid = CPUCLOCK_PID(which_clock); int err = -EINVAL; @@ -346,6 +327,8 @@ static int posix_cpu_timer_create(struct k_itimer *new_timer) if (CPUCLOCK_WHICH(new_timer->it_clock) >= CPUCLOCK_MAX) return -EINVAL; + new_timer->kclock = &clock_posix_cpu; + INIT_LIST_HEAD(&new_timer->it.cpu.entry); rcu_read_lock(); @@ -453,7 +436,7 @@ void posix_cpu_timers_exit_group(struct task_struct *tsk) cleanup_timers(tsk->signal->cpu_timers); } -static inline int expires_gt(cputime_t expires, cputime_t new_exp) +static inline int expires_gt(u64 expires, u64 new_exp) { return expires == 0 || expires > new_exp; } @@ -488,7 +471,7 @@ static void arm_timer(struct k_itimer *timer) list_add(&nt->entry, listpos); if (listpos == head) { - unsigned long long exp = nt->expires; + u64 exp = nt->expires; /* * We are the new earliest-expiring POSIX 1.b timer, hence @@ -499,16 +482,15 @@ static void arm_timer(struct k_itimer *timer) switch (CPUCLOCK_WHICH(timer->it_clock)) { case CPUCLOCK_PROF: - if (expires_gt(cputime_expires->prof_exp, expires_to_cputime(exp))) - cputime_expires->prof_exp = expires_to_cputime(exp); + if (expires_gt(cputime_expires->prof_exp, exp)) + cputime_expires->prof_exp = exp; break; case CPUCLOCK_VIRT: - if (expires_gt(cputime_expires->virt_exp, expires_to_cputime(exp))) - cputime_expires->virt_exp = expires_to_cputime(exp); + if (expires_gt(cputime_expires->virt_exp, exp)) + cputime_expires->virt_exp = exp; break; case CPUCLOCK_SCHED: - if (cputime_expires->sched_exp == 0 || - cputime_expires->sched_exp > exp) + if (expires_gt(cputime_expires->sched_exp, exp)) cputime_expires->sched_exp = exp; break; } @@ -549,7 +531,8 @@ static void cpu_timer_fire(struct k_itimer *timer) * reload the timer. But we need to keep it * ticking in case the signal is deliverable next time. */ - posix_cpu_timer_schedule(timer); + posix_cpu_timer_rearm(timer); + ++timer->it_requeue_pending; } } @@ -559,8 +542,7 @@ static void cpu_timer_fire(struct k_itimer *timer) * traversal. */ static int cpu_timer_sample_group(const clockid_t which_clock, - struct task_struct *p, - unsigned long long *sample) + struct task_struct *p, u64 *sample) { struct task_cputime cputime; @@ -569,10 +551,10 @@ static int cpu_timer_sample_group(const clockid_t which_clock, default: return -EINVAL; case CPUCLOCK_PROF: - *sample = cputime_to_expires(cputime.utime + cputime.stime); + *sample = cputime.utime + cputime.stime; break; case CPUCLOCK_VIRT: - *sample = cputime_to_expires(cputime.utime); + *sample = cputime.utime; break; case CPUCLOCK_SCHED: *sample = cputime.sum_exec_runtime; @@ -588,17 +570,21 @@ static int cpu_timer_sample_group(const clockid_t which_clock, * and try again. (This happens when the timer is in the middle of firing.) */ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags, - struct itimerspec *new, struct itimerspec *old) + struct itimerspec64 *new, struct itimerspec64 *old) { unsigned long flags; struct sighand_struct *sighand; struct task_struct *p = timer->it.cpu.task; - unsigned long long old_expires, new_expires, old_incr, val; + u64 old_expires, new_expires, old_incr, val; int ret; WARN_ON_ONCE(p == NULL); - new_expires = timespec_to_sample(timer->it_clock, &new->it_value); + /* + * Use the to_ktime conversion because that clamps the maximum + * value to KTIME_MAX and avoid multiplication overflows. + */ + new_expires = ktime_to_ns(timespec64_to_ktime(new->it_value)); /* * Protect against sighand release/switch in exit/exec and p->cpu_timers @@ -659,9 +645,7 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags, bump_cpu_timer(timer, val); if (val < timer->it.cpu.expires) { old_expires = timer->it.cpu.expires - val; - sample_to_timespec(timer->it_clock, - old_expires, - &old->it_value); + old->it_value = ns_to_timespec64(old_expires); } else { old->it_value.tv_nsec = 1; old->it_value.tv_sec = 0; @@ -699,8 +683,7 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags, * Install the new reload setting, and * set up the signal and overrun bookkeeping. */ - timer->it.cpu.incr = timespec_to_sample(timer->it_clock, - &new->it_interval); + timer->it.cpu.incr = timespec64_to_ns(&new->it_interval); /* * This acts as a modification timestamp for the timer, @@ -723,17 +706,15 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags, ret = 0; out: - if (old) { - sample_to_timespec(timer->it_clock, - old_incr, &old->it_interval); - } + if (old) + old->it_interval = ns_to_timespec64(old_incr); return ret; } -static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) +static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec64 *itp) { - unsigned long long now; + u64 now; struct task_struct *p = timer->it.cpu.task; WARN_ON_ONCE(p == NULL); @@ -741,13 +722,10 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) /* * Easy part: convert the reload time. */ - sample_to_timespec(timer->it_clock, - timer->it.cpu.incr, &itp->it_interval); + itp->it_interval = ns_to_timespec64(timer->it.cpu.incr); - if (timer->it.cpu.expires == 0) { /* Timer not armed at all. */ - itp->it_value.tv_sec = itp->it_value.tv_nsec = 0; + if (!timer->it.cpu.expires) return; - } /* * Sample the clock to take the difference with the expiry time. @@ -771,8 +749,6 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) * Call the timer disarmed, nothing else to do. */ timer->it.cpu.expires = 0; - sample_to_timespec(timer->it_clock, timer->it.cpu.expires, - &itp->it_value); return; } else { cpu_timer_sample_group(timer->it_clock, p, &now); @@ -781,9 +757,7 @@ static void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp) } if (now < timer->it.cpu.expires) { - sample_to_timespec(timer->it_clock, - timer->it.cpu.expires - now, - &itp->it_value); + itp->it_value = ns_to_timespec64(timer->it.cpu.expires - now); } else { /* * The timer should have expired already, but the firing @@ -827,7 +801,7 @@ static void check_thread_timers(struct task_struct *tsk, struct list_head *timers = tsk->cpu_timers; struct signal_struct *const sig = tsk->signal; struct task_cputime *tsk_expires = &tsk->cputime_expires; - unsigned long long expires; + u64 expires; unsigned long soft; /* @@ -838,10 +812,10 @@ static void check_thread_timers(struct task_struct *tsk, return; expires = check_timers_list(timers, firing, prof_ticks(tsk)); - tsk_expires->prof_exp = expires_to_cputime(expires); + tsk_expires->prof_exp = expires; expires = check_timers_list(++timers, firing, virt_ticks(tsk)); - tsk_expires->virt_exp = expires_to_cputime(expires); + tsk_expires->virt_exp = expires; tsk_expires->sched_exp = check_timers_list(++timers, firing, tsk->se.sum_exec_runtime); @@ -860,6 +834,10 @@ static void check_thread_timers(struct task_struct *tsk, * At the hard limit, we just die. * No need to calculate anything else now. */ + if (print_fatal_signals) { + pr_info("CPU Watchdog Timeout (hard): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk); return; } @@ -871,9 +849,10 @@ static void check_thread_timers(struct task_struct *tsk, soft += USEC_PER_SEC; sig->rlim[RLIMIT_RTTIME].rlim_cur = soft; } - printk(KERN_INFO - "RT Watchdog Timeout: %s[%d]\n", - tsk->comm, task_pid_nr(tsk)); + if (print_fatal_signals) { + pr_info("RT Watchdog Timeout (soft): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk); } } @@ -890,26 +869,17 @@ static inline void stop_process_timers(struct signal_struct *sig) tick_dep_clear_signal(sig, TICK_DEP_BIT_POSIX_TIMER); } -static u32 onecputick; - static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it, - unsigned long long *expires, - unsigned long long cur_time, int signo) + u64 *expires, u64 cur_time, int signo) { if (!it->expires) return; if (cur_time >= it->expires) { - if (it->incr) { + if (it->incr) it->expires += it->incr; - it->error += it->incr_error; - if (it->error >= onecputick) { - it->expires -= cputime_one_jiffy; - it->error -= onecputick; - } - } else { + else it->expires = 0; - } trace_itimer_expire(signo == SIGPROF ? ITIMER_PROF : ITIMER_VIRTUAL, @@ -917,9 +887,8 @@ static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it, __group_send_sig_info(signo, SEND_SIG_PRIV, tsk); } - if (it->expires && (!*expires || it->expires < *expires)) { + if (it->expires && (!*expires || it->expires < *expires)) *expires = it->expires; - } } /* @@ -931,8 +900,8 @@ static void check_process_timers(struct task_struct *tsk, struct list_head *firing) { struct signal_struct *const sig = tsk->signal; - unsigned long long utime, ptime, virt_expires, prof_expires; - unsigned long long sum_sched_runtime, sched_expires; + u64 utime, ptime, virt_expires, prof_expires; + u64 sum_sched_runtime, sched_expires; struct list_head *timers = sig->cpu_timers; struct task_cputime cputime; unsigned long soft; @@ -954,8 +923,8 @@ static void check_process_timers(struct task_struct *tsk, * Collect the current process totals. */ thread_group_cputimer(tsk, &cputime); - utime = cputime_to_expires(cputime.utime); - ptime = utime + cputime_to_expires(cputime.stime); + utime = cputime.utime; + ptime = utime + cputime.stime; sum_sched_runtime = cputime.sum_exec_runtime; prof_expires = check_timers_list(timers, firing, ptime); @@ -971,15 +940,19 @@ static void check_process_timers(struct task_struct *tsk, SIGVTALRM); soft = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_cur); if (soft != RLIM_INFINITY) { - unsigned long psecs = cputime_to_secs(ptime); + unsigned long psecs = div_u64(ptime, NSEC_PER_SEC); unsigned long hard = READ_ONCE(sig->rlim[RLIMIT_CPU].rlim_max); - cputime_t x; + u64 x; if (psecs >= hard) { /* * At the hard limit, we just die. * No need to calculate anything else now. */ + if (print_fatal_signals) { + pr_info("RT Watchdog Timeout (hard): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk); return; } @@ -987,20 +960,23 @@ static void check_process_timers(struct task_struct *tsk, /* * At the soft limit, send a SIGXCPU every second. */ + if (print_fatal_signals) { + pr_info("CPU Watchdog Timeout (soft): %s[%d]\n", + tsk->comm, task_pid_nr(tsk)); + } __group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk); if (soft < hard) { soft++; sig->rlim[RLIMIT_CPU].rlim_cur = soft; } } - x = secs_to_cputime(soft); - if (!prof_expires || x < prof_expires) { + x = soft * NSEC_PER_SEC; + if (!prof_expires || x < prof_expires) prof_expires = x; - } } - sig->cputime_expires.prof_exp = expires_to_cputime(prof_expires); - sig->cputime_expires.virt_exp = expires_to_cputime(virt_expires); + sig->cputime_expires.prof_exp = prof_expires; + sig->cputime_expires.virt_exp = virt_expires; sig->cputime_expires.sched_exp = sched_expires; if (task_cputime_zero(&sig->cputime_expires)) stop_process_timers(sig); @@ -1009,15 +985,15 @@ static void check_process_timers(struct task_struct *tsk, } /* - * This is called from the signal code (via do_schedule_next_timer) + * This is called from the signal code (via posixtimer_rearm) * when the last timer signal was delivered and we have to reload the timer. */ -void posix_cpu_timer_schedule(struct k_itimer *timer) +static void posix_cpu_timer_rearm(struct k_itimer *timer) { struct sighand_struct *sighand; unsigned long flags; struct task_struct *p = timer->it.cpu.task; - unsigned long long now; + u64 now; WARN_ON_ONCE(p == NULL); @@ -1028,12 +1004,12 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) cpu_clock_sample(timer->it_clock, p, &now); bump_cpu_timer(timer, now); if (unlikely(p->exit_state)) - goto out; + return; /* Protect timer list r/w in arm_timer() */ sighand = lock_task_sighand(p, &flags); if (!sighand) - goto out; + return; } else { /* * Protect arm_timer() and timer sampling in case of call to @@ -1046,11 +1022,10 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) * We can't even collect a sample any more. */ timer->it.cpu.expires = 0; - goto out; + return; } else if (unlikely(p->exit_state) && thread_group_empty(p)) { - unlock_task_sighand(p, &flags); - /* Optimizations: if the process is dying, no need to rearm */ - goto out; + /* If the process is dying, no need to rearm */ + goto unlock; } cpu_timer_sample_group(timer->it_clock, p, &now); bump_cpu_timer(timer, now); @@ -1062,12 +1037,8 @@ void posix_cpu_timer_schedule(struct k_itimer *timer) */ WARN_ON_ONCE(!irqs_disabled()); arm_timer(timer); +unlock: unlock_task_sighand(p, &flags); - -out: - timer->it_overrun_last = timer->it_overrun; - timer->it_overrun = -1; - ++timer->it_requeue_pending; } /** @@ -1214,9 +1185,9 @@ void run_posix_cpu_timers(struct task_struct *tsk) * The tsk->sighand->siglock must be held by the caller. */ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, - cputime_t *newval, cputime_t *oldval) + u64 *newval, u64 *oldval) { - unsigned long long now; + u64 now; WARN_ON_ONCE(clock_idx == CPUCLOCK_SCHED); cpu_timer_sample_group(clock_idx, tsk, &now); @@ -1230,7 +1201,7 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, if (*oldval) { if (*oldval <= now) { /* Just about to fire. */ - *oldval = cputime_one_jiffy; + *oldval = TICK_NSEC; } else { *oldval -= now; } @@ -1260,9 +1231,11 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx, } static int do_cpu_nanosleep(const clockid_t which_clock, int flags, - struct timespec *rqtp, struct itimerspec *it) + const struct timespec64 *rqtp) { + struct itimerspec64 it; struct k_itimer timer; + u64 expires; int error; /* @@ -1275,13 +1248,14 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, error = posix_cpu_timer_create(&timer); timer.it_process = current; if (!error) { - static struct itimerspec zero_it; + static struct itimerspec64 zero_it; + struct restart_block *restart; - memset(it, 0, sizeof *it); - it->it_value = *rqtp; + memset(&it, 0, sizeof(it)); + it.it_value = *rqtp; spin_lock_irq(&timer.it_lock); - error = posix_cpu_timer_set(&timer, flags, it, NULL); + error = posix_cpu_timer_set(&timer, flags, &it, NULL); if (error) { spin_unlock_irq(&timer.it_lock); return error; @@ -1310,8 +1284,8 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, /* * We were interrupted by a signal. */ - sample_to_timespec(which_clock, timer.it.cpu.expires, rqtp); - error = posix_cpu_timer_set(&timer, 0, &zero_it, it); + expires = timer.it.cpu.expires; + error = posix_cpu_timer_set(&timer, 0, &zero_it, &it); if (!error) { /* * Timer is now unarmed, deletion can not fail. @@ -1331,7 +1305,7 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, spin_unlock_irq(&timer.it_lock); } - if ((it->it_value.tv_sec | it->it_value.tv_nsec) == 0) { + if ((it.it_value.tv_sec | it.it_value.tv_nsec) == 0) { /* * It actually did fire already. */ @@ -1339,6 +1313,13 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, } error = -ERESTART_RESTARTBLOCK; + /* + * Report back to the user the time still remaining. + */ + restart = ¤t->restart_block; + restart->nanosleep.expires = expires; + if (restart->nanosleep.type != TT_NONE) + error = nanosleep_copyout(restart, &it.it_value); } return error; @@ -1347,10 +1328,9 @@ static int do_cpu_nanosleep(const clockid_t which_clock, int flags, static long posix_cpu_nsleep_restart(struct restart_block *restart_block); static int posix_cpu_nsleep(const clockid_t which_clock, int flags, - struct timespec *rqtp, struct timespec __user *rmtp) + const struct timespec64 *rqtp) { struct restart_block *restart_block = ¤t->restart_block; - struct itimerspec it; int error; /* @@ -1358,25 +1338,18 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags, */ if (CPUCLOCK_PERTHREAD(which_clock) && (CPUCLOCK_PID(which_clock) == 0 || - CPUCLOCK_PID(which_clock) == current->pid)) + CPUCLOCK_PID(which_clock) == task_pid_vnr(current))) return -EINVAL; - error = do_cpu_nanosleep(which_clock, flags, rqtp, &it); + error = do_cpu_nanosleep(which_clock, flags, rqtp); if (error == -ERESTART_RESTARTBLOCK) { if (flags & TIMER_ABSTIME) return -ERESTARTNOHAND; - /* - * Report back to the user the time still remaining. - */ - if (rmtp && copy_to_user(rmtp, &it.it_value, sizeof *rmtp)) - return -EFAULT; restart_block->fn = posix_cpu_nsleep_restart; restart_block->nanosleep.clockid = which_clock; - restart_block->nanosleep.rmtp = rmtp; - restart_block->nanosleep.expires = timespec_to_ns(rqtp); } return error; } @@ -1384,38 +1357,23 @@ static int posix_cpu_nsleep(const clockid_t which_clock, int flags, static long posix_cpu_nsleep_restart(struct restart_block *restart_block) { clockid_t which_clock = restart_block->nanosleep.clockid; - struct timespec t; - struct itimerspec it; - int error; - - t = ns_to_timespec(restart_block->nanosleep.expires); + struct timespec64 t; - error = do_cpu_nanosleep(which_clock, TIMER_ABSTIME, &t, &it); - - if (error == -ERESTART_RESTARTBLOCK) { - struct timespec __user *rmtp = restart_block->nanosleep.rmtp; - /* - * Report back to the user the time still remaining. - */ - if (rmtp && copy_to_user(rmtp, &it.it_value, sizeof *rmtp)) - return -EFAULT; - - restart_block->nanosleep.expires = timespec_to_ns(&t); - } - return error; + t = ns_to_timespec64(restart_block->nanosleep.expires); + return do_cpu_nanosleep(which_clock, TIMER_ABSTIME, &t); } #define PROCESS_CLOCK MAKE_PROCESS_CPUCLOCK(0, CPUCLOCK_SCHED) #define THREAD_CLOCK MAKE_THREAD_CPUCLOCK(0, CPUCLOCK_SCHED) static int process_cpu_clock_getres(const clockid_t which_clock, - struct timespec *tp) + struct timespec64 *tp) { return posix_cpu_clock_getres(PROCESS_CLOCK, tp); } static int process_cpu_clock_get(const clockid_t which_clock, - struct timespec *tp) + struct timespec64 *tp) { return posix_cpu_clock_get(PROCESS_CLOCK, tp); } @@ -1425,22 +1383,17 @@ static int process_cpu_timer_create(struct k_itimer *timer) return posix_cpu_timer_create(timer); } static int process_cpu_nsleep(const clockid_t which_clock, int flags, - struct timespec *rqtp, - struct timespec __user *rmtp) + const struct timespec64 *rqtp) { - return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp, rmtp); -} -static long process_cpu_nsleep_restart(struct restart_block *restart_block) -{ - return -EINVAL; + return posix_cpu_nsleep(PROCESS_CLOCK, flags, rqtp); } static int thread_cpu_clock_getres(const clockid_t which_clock, - struct timespec *tp) + struct timespec64 *tp) { return posix_cpu_clock_getres(THREAD_CLOCK, tp); } static int thread_cpu_clock_get(const clockid_t which_clock, - struct timespec *tp) + struct timespec64 *tp) { return posix_cpu_clock_get(THREAD_CLOCK, tp); } @@ -1450,41 +1403,27 @@ static int thread_cpu_timer_create(struct k_itimer *timer) return posix_cpu_timer_create(timer); } -struct k_clock clock_posix_cpu = { +const struct k_clock clock_posix_cpu = { .clock_getres = posix_cpu_clock_getres, .clock_set = posix_cpu_clock_set, .clock_get = posix_cpu_clock_get, .timer_create = posix_cpu_timer_create, .nsleep = posix_cpu_nsleep, - .nsleep_restart = posix_cpu_nsleep_restart, .timer_set = posix_cpu_timer_set, .timer_del = posix_cpu_timer_del, .timer_get = posix_cpu_timer_get, + .timer_rearm = posix_cpu_timer_rearm, }; -static __init int init_posix_cpu_timers(void) -{ - struct k_clock process = { - .clock_getres = process_cpu_clock_getres, - .clock_get = process_cpu_clock_get, - .timer_create = process_cpu_timer_create, - .nsleep = process_cpu_nsleep, - .nsleep_restart = process_cpu_nsleep_restart, - }; - struct k_clock thread = { - .clock_getres = thread_cpu_clock_getres, - .clock_get = thread_cpu_clock_get, - .timer_create = thread_cpu_timer_create, - }; - struct timespec ts; - - posix_timers_register_clock(CLOCK_PROCESS_CPUTIME_ID, &process); - posix_timers_register_clock(CLOCK_THREAD_CPUTIME_ID, &thread); - - cputime_to_timespec(cputime_one_jiffy, &ts); - onecputick = ts.tv_nsec; - WARN_ON(ts.tv_sec != 0); +const struct k_clock clock_process = { + .clock_getres = process_cpu_clock_getres, + .clock_get = process_cpu_clock_get, + .timer_create = process_cpu_timer_create, + .nsleep = process_cpu_nsleep, +}; - return 0; -} -__initcall(init_posix_cpu_timers); +const struct k_clock clock_thread = { + .clock_getres = thread_cpu_clock_getres, + .clock_get = thread_cpu_clock_get, + .timer_create = thread_cpu_timer_create, +}; diff --git a/kernel/time/posix-stubs.c b/kernel/time/posix-stubs.c index cd6716e115e8..06f34feb635e 100644 --- a/kernel/time/posix-stubs.c +++ b/kernel/time/posix-stubs.c @@ -17,6 +17,7 @@ #include <linux/ktime.h> #include <linux/timekeeping.h> #include <linux/posix-timers.h> +#include <linux/compat.h> asmlinkage long sys_ni_posix_timers(void) { @@ -27,6 +28,7 @@ asmlinkage long sys_ni_posix_timers(void) } #define SYS_NI(name) SYSCALL_ALIAS(sys_##name, sys_ni_posix_timers) +#define COMPAT_SYS_NI(name) SYSCALL_ALIAS(compat_sys_##name, sys_ni_posix_timers) SYS_NI(timer_create); SYS_NI(timer_gettime); @@ -49,34 +51,52 @@ SYS_NI(alarm); SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, const struct timespec __user *, tp) { - struct timespec new_tp; + struct timespec64 new_tp; if (which_clock != CLOCK_REALTIME) return -EINVAL; - if (copy_from_user(&new_tp, tp, sizeof (*tp))) + if (get_timespec64(&new_tp, tp)) return -EFAULT; - return do_sys_settimeofday(&new_tp, NULL); + + return do_sys_settimeofday64(&new_tp, NULL); } -SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, - struct timespec __user *,tp) +int do_clock_gettime(clockid_t which_clock, struct timespec64 *tp) { - struct timespec kernel_tp; - switch (which_clock) { - case CLOCK_REALTIME: ktime_get_real_ts(&kernel_tp); break; - case CLOCK_MONOTONIC: ktime_get_ts(&kernel_tp); break; - case CLOCK_BOOTTIME: get_monotonic_boottime(&kernel_tp); break; - default: return -EINVAL; + case CLOCK_REALTIME: + ktime_get_real_ts64(tp); + break; + case CLOCK_MONOTONIC: + ktime_get_ts64(tp); + break; + case CLOCK_BOOTTIME: + get_monotonic_boottime64(tp); + break; + default: + return -EINVAL; } - if (copy_to_user(tp, &kernel_tp, sizeof (kernel_tp))) + + return 0; +} +SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, + struct timespec __user *, tp) +{ + int ret; + struct timespec64 kernel_tp; + + ret = do_clock_gettime(which_clock, &kernel_tp); + if (ret) + return ret; + + if (put_timespec64(&kernel_tp, tp)) return -EFAULT; return 0; } SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __user *, tp) { - struct timespec rtn_tp = { + struct timespec64 rtn_tp = { .tv_sec = 0, .tv_nsec = hrtimer_resolution, }; @@ -85,7 +105,7 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __us case CLOCK_REALTIME: case CLOCK_MONOTONIC: case CLOCK_BOOTTIME: - if (copy_to_user(tp, &rtn_tp, sizeof(rtn_tp))) + if (put_timespec64(&rtn_tp, tp)) return -EFAULT; return 0; default: @@ -97,27 +117,115 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, const struct timespec __user *, rqtp, struct timespec __user *, rmtp) { + struct timespec64 t64; struct timespec t; switch (which_clock) { case CLOCK_REALTIME: case CLOCK_MONOTONIC: case CLOCK_BOOTTIME: - if (copy_from_user(&t, rqtp, sizeof (struct timespec))) - return -EFAULT; - if (!timespec_valid(&t)) - return -EINVAL; - return hrtimer_nanosleep(&t, rmtp, flags & TIMER_ABSTIME ? - HRTIMER_MODE_ABS : HRTIMER_MODE_REL, - which_clock); + break; default: return -EINVAL; } + + if (copy_from_user(&t, rqtp, sizeof (struct timespec))) + return -EFAULT; + t64 = timespec_to_timespec64(t); + if (!timespec64_valid(&t64)) + return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; + current->restart_block.nanosleep.rmtp = rmtp; + return hrtimer_nanosleep(&t64, flags & TIMER_ABSTIME ? + HRTIMER_MODE_ABS : HRTIMER_MODE_REL, + which_clock); } #ifdef CONFIG_COMPAT -long clock_nanosleep_restart(struct restart_block *restart_block) +COMPAT_SYS_NI(timer_create); +COMPAT_SYS_NI(clock_adjtime); +COMPAT_SYS_NI(timer_settime); +COMPAT_SYS_NI(timer_gettime); +COMPAT_SYS_NI(getitimer); +COMPAT_SYS_NI(setitimer); + +COMPAT_SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + struct timespec64 new_tp; + + if (which_clock != CLOCK_REALTIME) + return -EINVAL; + if (compat_get_timespec64(&new_tp, tp)) + return -EFAULT; + + return do_sys_settimeofday64(&new_tp, NULL); +} + +COMPAT_SYSCALL_DEFINE2(clock_gettime, clockid_t, which_clock, + struct compat_timespec __user *, tp) { - return hrtimer_nanosleep_restart(restart_block); + int ret; + struct timespec64 kernel_tp; + + ret = do_clock_gettime(which_clock, &kernel_tp); + if (ret) + return ret; + + if (compat_put_timespec64(&kernel_tp, tp)) + return -EFAULT; + return 0; +} + +COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + struct timespec64 rtn_tp = { + .tv_sec = 0, + .tv_nsec = hrtimer_resolution, + }; + + switch (which_clock) { + case CLOCK_REALTIME: + case CLOCK_MONOTONIC: + case CLOCK_BOOTTIME: + if (compat_put_timespec64(&rtn_tp, tp)) + return -EFAULT; + return 0; + default: + return -EINVAL; + } +} + +COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, + struct compat_timespec __user *, rqtp, + struct compat_timespec __user *, rmtp) +{ + struct timespec64 t64; + struct timespec t; + + switch (which_clock) { + case CLOCK_REALTIME: + case CLOCK_MONOTONIC: + case CLOCK_BOOTTIME: + break; + default: + return -EINVAL; + } + + if (compat_get_timespec(&t, rqtp)) + return -EFAULT; + t64 = timespec_to_timespec64(t); + if (!timespec64_valid(&t64)) + return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; + current->restart_block.nanosleep.compat_rmtp = rmtp; + return hrtimer_nanosleep(&t64, flags & TIMER_ABSTIME ? + HRTIMER_MODE_ABS : HRTIMER_MODE_REL, + which_clock); } #endif diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 1e6623d76750..13d6881f908b 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -35,6 +35,7 @@ #include <linux/slab.h> #include <linux/time.h> #include <linux/mutex.h> +#include <linux/sched/task.h> #include <linux/uaccess.h> #include <linux/list.h> @@ -48,8 +49,10 @@ #include <linux/workqueue.h> #include <linux/export.h> #include <linux/hashtable.h> +#include <linux/compat.h> #include "timekeeping.h" +#include "posix-timers.h" /* * Management arrays for POSIX timers. Timers are now kept in static hash table @@ -68,6 +71,10 @@ static struct kmem_cache *posix_timers_cache; static DEFINE_HASHTABLE(posix_timers_hashtable, 9); static DEFINE_SPINLOCK(hash_lock); +static const struct k_clock * const posix_clocks[]; +static const struct k_clock *clockid_to_kclock(const clockid_t id); +static const struct k_clock clock_realtime, clock_monotonic; + /* * we assume that the new SIGEV_THREAD_ID shares no bits with the other * SIGEV values. Here we put out an error if this assumption fails. @@ -123,22 +130,6 @@ static DEFINE_SPINLOCK(hash_lock); * have is CLOCK_REALTIME and its high res counter part, both of * which we beg off on and pass to do_sys_settimeofday(). */ - -static struct k_clock posix_clocks[MAX_CLOCKS]; - -/* - * These ones are defined below. - */ -static int common_nsleep(const clockid_t, int flags, struct timespec *t, - struct timespec __user *rmtp); -static int common_timer_create(struct k_itimer *new_timer); -static void common_timer_get(struct k_itimer *, struct itimerspec *); -static int common_timer_set(struct k_itimer *, int, - struct itimerspec *, struct itimerspec *); -static int common_timer_del(struct k_itimer *timer); - -static enum hrtimer_restart posix_timer_fn(struct hrtimer *data); - static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags); #define lock_timer(tid, flags) \ @@ -203,17 +194,17 @@ static inline void unlock_timer(struct k_itimer *timr, unsigned long flags) } /* Get clock_realtime */ -static int posix_clock_realtime_get(clockid_t which_clock, struct timespec *tp) +static int posix_clock_realtime_get(clockid_t which_clock, struct timespec64 *tp) { - ktime_get_real_ts(tp); + ktime_get_real_ts64(tp); return 0; } /* Set clock_realtime */ static int posix_clock_realtime_set(const clockid_t which_clock, - const struct timespec *tp) + const struct timespec64 *tp) { - return do_sys_settimeofday(tp, NULL); + return do_sys_settimeofday64(tp, NULL); } static int posix_clock_realtime_adj(const clockid_t which_clock, @@ -225,54 +216,54 @@ static int posix_clock_realtime_adj(const clockid_t which_clock, /* * Get monotonic time for posix timers */ -static int posix_ktime_get_ts(clockid_t which_clock, struct timespec *tp) +static int posix_ktime_get_ts(clockid_t which_clock, struct timespec64 *tp) { - ktime_get_ts(tp); + ktime_get_ts64(tp); return 0; } /* * Get monotonic-raw time for posix timers */ -static int posix_get_monotonic_raw(clockid_t which_clock, struct timespec *tp) +static int posix_get_monotonic_raw(clockid_t which_clock, struct timespec64 *tp) { - getrawmonotonic(tp); + getrawmonotonic64(tp); return 0; } -static int posix_get_realtime_coarse(clockid_t which_clock, struct timespec *tp) +static int posix_get_realtime_coarse(clockid_t which_clock, struct timespec64 *tp) { - *tp = current_kernel_time(); + *tp = current_kernel_time64(); return 0; } static int posix_get_monotonic_coarse(clockid_t which_clock, - struct timespec *tp) + struct timespec64 *tp) { - *tp = get_monotonic_coarse(); + *tp = get_monotonic_coarse64(); return 0; } -static int posix_get_coarse_res(const clockid_t which_clock, struct timespec *tp) +static int posix_get_coarse_res(const clockid_t which_clock, struct timespec64 *tp) { - *tp = ktime_to_timespec(KTIME_LOW_RES); + *tp = ktime_to_timespec64(KTIME_LOW_RES); return 0; } -static int posix_get_boottime(const clockid_t which_clock, struct timespec *tp) +static int posix_get_boottime(const clockid_t which_clock, struct timespec64 *tp) { - get_monotonic_boottime(tp); + get_monotonic_boottime64(tp); return 0; } -static int posix_get_tai(clockid_t which_clock, struct timespec *tp) +static int posix_get_tai(clockid_t which_clock, struct timespec64 *tp) { - timekeeping_clocktai(tp); + timekeeping_clocktai64(tp); return 0; } -static int posix_get_hrtimer_res(clockid_t which_clock, struct timespec *tp) +static int posix_get_hrtimer_res(clockid_t which_clock, struct timespec64 *tp) { tp->tv_sec = 0; tp->tv_nsec = hrtimer_resolution; @@ -284,91 +275,23 @@ static int posix_get_hrtimer_res(clockid_t which_clock, struct timespec *tp) */ static __init int init_posix_timers(void) { - struct k_clock clock_realtime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_clock_realtime_get, - .clock_set = posix_clock_realtime_set, - .clock_adj = posix_clock_realtime_adj, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - struct k_clock clock_monotonic = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_ktime_get_ts, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - struct k_clock clock_monotonic_raw = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_monotonic_raw, - }; - struct k_clock clock_realtime_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_realtime_coarse, - }; - struct k_clock clock_monotonic_coarse = { - .clock_getres = posix_get_coarse_res, - .clock_get = posix_get_monotonic_coarse, - }; - struct k_clock clock_tai = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_tai, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - struct k_clock clock_boottime = { - .clock_getres = posix_get_hrtimer_res, - .clock_get = posix_get_boottime, - .nsleep = common_nsleep, - .nsleep_restart = hrtimer_nanosleep_restart, - .timer_create = common_timer_create, - .timer_set = common_timer_set, - .timer_get = common_timer_get, - .timer_del = common_timer_del, - }; - - posix_timers_register_clock(CLOCK_REALTIME, &clock_realtime); - posix_timers_register_clock(CLOCK_MONOTONIC, &clock_monotonic); - posix_timers_register_clock(CLOCK_MONOTONIC_RAW, &clock_monotonic_raw); - posix_timers_register_clock(CLOCK_REALTIME_COARSE, &clock_realtime_coarse); - posix_timers_register_clock(CLOCK_MONOTONIC_COARSE, &clock_monotonic_coarse); - posix_timers_register_clock(CLOCK_BOOTTIME, &clock_boottime); - posix_timers_register_clock(CLOCK_TAI, &clock_tai); - posix_timers_cache = kmem_cache_create("posix_timers_cache", sizeof (struct k_itimer), 0, SLAB_PANIC, NULL); return 0; } - __initcall(init_posix_timers); -static void schedule_next_timer(struct k_itimer *timr) +static void common_hrtimer_rearm(struct k_itimer *timr) { struct hrtimer *timer = &timr->it.real.timer; - if (timr->it.real.interval == 0) + if (!timr->it_interval) return; timr->it_overrun += (unsigned int) hrtimer_forward(timer, timer->base->get_time(), - timr->it.real.interval); - - timr->it_overrun_last = timr->it_overrun; - timr->it_overrun = -1; - ++timr->it_requeue_pending; + timr->it_interval); hrtimer_restart(timer); } @@ -383,24 +306,27 @@ static void schedule_next_timer(struct k_itimer *timr) * To protect against the timer going away while the interrupt is queued, * we require that the it_requeue_pending flag be set. */ -void do_schedule_next_timer(struct siginfo *info) +void posixtimer_rearm(struct siginfo *info) { struct k_itimer *timr; unsigned long flags; timr = lock_timer(info->si_tid, &flags); + if (!timr) + return; - if (timr && timr->it_requeue_pending == info->si_sys_private) { - if (timr->it_clock < 0) - posix_cpu_timer_schedule(timr); - else - schedule_next_timer(timr); + if (timr->it_requeue_pending == info->si_sys_private) { + timr->kclock->timer_rearm(timr); + + timr->it_active = 1; + timr->it_overrun_last = timr->it_overrun; + timr->it_overrun = -1; + ++timr->it_requeue_pending; info->si_overrun += timr->it_overrun_last; } - if (timr) - unlock_timer(timr, flags); + unlock_timer(timr, flags); } int posix_timer_event(struct k_itimer *timr, int si_private) @@ -409,12 +335,12 @@ int posix_timer_event(struct k_itimer *timr, int si_private) int shared, ret = -1; /* * FIXME: if ->sigq is queued we can race with - * dequeue_signal()->do_schedule_next_timer(). + * dequeue_signal()->posixtimer_rearm(). * * If dequeue_signal() sees the "right" value of - * si_sys_private it calls do_schedule_next_timer(). + * si_sys_private it calls posixtimer_rearm(). * We re-queue ->sigq and drop ->it_lock(). - * do_schedule_next_timer() locks the timer + * posixtimer_rearm() locks the timer * and re-schedules it while ->sigq is pending. * Not really bad, but not that we want. */ @@ -430,7 +356,6 @@ int posix_timer_event(struct k_itimer *timr, int si_private) /* If we failed to send the signal the timer stops. */ return ret > 0; } -EXPORT_SYMBOL_GPL(posix_timer_event); /* * This function gets called when a POSIX.1b interval timer expires. It @@ -449,7 +374,8 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) timr = container_of(timer, struct k_itimer, it.real.timer); spin_lock_irqsave(&timr->it_lock, flags); - if (timr->it.real.interval != 0) + timr->it_active = 0; + if (timr->it_interval != 0) si_private = ++timr->it_requeue_pending; if (posix_timer_event(timr, si_private)) { @@ -458,7 +384,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) * we will not get a call back to restart it AND * it should be restarted. */ - if (timr->it.real.interval != 0) { + if (timr->it_interval != 0) { ktime_t now = hrtimer_cb_get_time(timer); /* @@ -487,15 +413,16 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer) { ktime_t kj = NSEC_PER_SEC / HZ; - if (timr->it.real.interval < kj) + if (timr->it_interval < kj) now = ktime_add(now, kj); } #endif timr->it_overrun += (unsigned int) hrtimer_forward(timer, now, - timr->it.real.interval); + timr->it_interval); ret = HRTIMER_RESTART; ++timr->it_requeue_pending; + timr->it_active = 1; } } @@ -520,30 +447,6 @@ static struct pid *good_sigevent(sigevent_t * event) return task_pid(rtn); } -void posix_timers_register_clock(const clockid_t clock_id, - struct k_clock *new_clock) -{ - if ((unsigned) clock_id >= MAX_CLOCKS) { - printk(KERN_WARNING "POSIX clock register failed for clock_id %d\n", - clock_id); - return; - } - - if (!new_clock->clock_get) { - printk(KERN_WARNING "POSIX clock id %d lacks clock_get()\n", - clock_id); - return; - } - if (!new_clock->clock_getres) { - printk(KERN_WARNING "POSIX clock id %d lacks clock_getres()\n", - clock_id); - return; - } - - posix_clocks[clock_id] = *new_clock; -} -EXPORT_SYMBOL_GPL(posix_timers_register_clock); - static struct k_itimer * alloc_posix_timer(void) { struct k_itimer *tmr; @@ -580,17 +483,6 @@ static void release_posix_timer(struct k_itimer *tmr, int it_id_set) call_rcu(&tmr->it.rcu, k_itimer_rcu_free); } -static struct k_clock *clockid_to_kclock(const clockid_t id) -{ - if (id < 0) - return (id & CLOCKFD_MASK) == CLOCKFD ? - &clock_posix_dynamic : &clock_posix_cpu; - - if (id >= MAX_CLOCKS || !posix_clocks[id].clock_getres) - return NULL; - return &posix_clocks[id]; -} - static int common_timer_create(struct k_itimer *new_timer) { hrtimer_init(&new_timer->it.real.timer, new_timer->it_clock, 0); @@ -598,15 +490,12 @@ static int common_timer_create(struct k_itimer *new_timer) } /* Create a POSIX.1b interval timer. */ - -SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, - struct sigevent __user *, timer_event_spec, - timer_t __user *, created_timer_id) +static int do_timer_create(clockid_t which_clock, struct sigevent *event, + timer_t __user *created_timer_id) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct k_itimer *new_timer; int error, new_timer_id; - sigevent_t event; int it_id_set = IT_ID_NOT_SET; if (!kc) @@ -628,31 +517,28 @@ SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, it_id_set = IT_ID_SET; new_timer->it_id = (timer_t) new_timer_id; new_timer->it_clock = which_clock; + new_timer->kclock = kc; new_timer->it_overrun = -1; - if (timer_event_spec) { - if (copy_from_user(&event, timer_event_spec, sizeof (event))) { - error = -EFAULT; - goto out; - } + if (event) { rcu_read_lock(); - new_timer->it_pid = get_pid(good_sigevent(&event)); + new_timer->it_pid = get_pid(good_sigevent(event)); rcu_read_unlock(); if (!new_timer->it_pid) { error = -EINVAL; goto out; } + new_timer->it_sigev_notify = event->sigev_notify; + new_timer->sigq->info.si_signo = event->sigev_signo; + new_timer->sigq->info.si_value = event->sigev_value; } else { - memset(&event.sigev_value, 0, sizeof(event.sigev_value)); - event.sigev_notify = SIGEV_SIGNAL; - event.sigev_signo = SIGALRM; - event.sigev_value.sival_int = new_timer->it_id; + new_timer->it_sigev_notify = SIGEV_SIGNAL; + new_timer->sigq->info.si_signo = SIGALRM; + memset(&new_timer->sigq->info.si_value, 0, sizeof(sigval_t)); + new_timer->sigq->info.si_value.sival_int = new_timer->it_id; new_timer->it_pid = get_pid(task_tgid(current)); } - new_timer->it_sigev_notify = event.sigev_notify; - new_timer->sigq->info.si_signo = event.sigev_signo; - new_timer->sigq->info.si_value = event.sigev_value; new_timer->sigq->info.si_tid = new_timer->it_id; new_timer->sigq->info.si_code = SI_TIMER; @@ -683,6 +569,36 @@ out: return error; } +SYSCALL_DEFINE3(timer_create, const clockid_t, which_clock, + struct sigevent __user *, timer_event_spec, + timer_t __user *, created_timer_id) +{ + if (timer_event_spec) { + sigevent_t event; + + if (copy_from_user(&event, timer_event_spec, sizeof (event))) + return -EFAULT; + return do_timer_create(which_clock, &event, created_timer_id); + } + return do_timer_create(which_clock, NULL, created_timer_id); +} + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE3(timer_create, clockid_t, which_clock, + struct compat_sigevent __user *, timer_event_spec, + timer_t __user *, created_timer_id) +{ + if (timer_event_spec) { + sigevent_t event; + + if (get_compat_sigevent(&event, timer_event_spec)) + return -EFAULT; + return do_timer_create(which_clock, &event, created_timer_id); + } + return do_timer_create(which_clock, NULL, created_timer_id); +} +#endif + /* * Locking issues: We need to protect the result of the id look up until * we get the timer locked down so it is not deleted under us. The @@ -716,6 +632,20 @@ static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags) return NULL; } +static ktime_t common_hrtimer_remaining(struct k_itimer *timr, ktime_t now) +{ + struct hrtimer *timer = &timr->it.real.timer; + + return __hrtimer_expires_remaining_adjusted(timer, now); +} + +static int common_hrtimer_forward(struct k_itimer *timr, ktime_t now) +{ + struct hrtimer *timer = &timr->it.real.timer; + + return (int)hrtimer_forward(timer, now, timr->it_interval); +} + /* * Get the time remaining on a POSIX.1b interval timer. This function * is ALWAYS called with spin_lock_irq on the timer, thus it must not @@ -732,54 +662,61 @@ static struct k_itimer *__lock_timer(timer_t timer_id, unsigned long *flags) * it is the same as a requeue pending timer WRT to what we should * report. */ -static void -common_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting) +void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting) { + const struct k_clock *kc = timr->kclock; ktime_t now, remaining, iv; - struct hrtimer *timer = &timr->it.real.timer; + struct timespec64 ts64; + bool sig_none; - memset(cur_setting, 0, sizeof(struct itimerspec)); - - iv = timr->it.real.interval; + sig_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE; + iv = timr->it_interval; /* interval timer ? */ - if (iv) - cur_setting->it_interval = ktime_to_timespec(iv); - else if (!hrtimer_active(timer) && - (timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) - return; + if (iv) { + cur_setting->it_interval = ktime_to_timespec64(iv); + } else if (!timr->it_active) { + /* + * SIGEV_NONE oneshot timers are never queued. Check them + * below. + */ + if (!sig_none) + return; + } - now = timer->base->get_time(); + /* + * The timespec64 based conversion is suboptimal, but it's not + * worth to implement yet another callback. + */ + kc->clock_get(timr->it_clock, &ts64); + now = timespec64_to_ktime(ts64); /* - * When a requeue is pending or this is a SIGEV_NONE - * timer move the expiry time forward by intervals, so - * expiry is > now. + * When a requeue is pending or this is a SIGEV_NONE timer move the + * expiry time forward by intervals, so expiry is > now. */ - if (iv && (timr->it_requeue_pending & REQUEUE_PENDING || - (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE)) - timr->it_overrun += (unsigned int) hrtimer_forward(timer, now, iv); + if (iv && (timr->it_requeue_pending & REQUEUE_PENDING || sig_none)) + timr->it_overrun += kc->timer_forward(timr, now); - remaining = __hrtimer_expires_remaining_adjusted(timer, now); + remaining = kc->timer_remaining(timr, now); /* Return 0 only, when the timer is expired and not pending */ if (remaining <= 0) { /* * A single shot SIGEV_NONE timer must return 0, when * it is expired ! */ - if ((timr->it_sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) + if (!sig_none) cur_setting->it_value.tv_nsec = 1; - } else - cur_setting->it_value = ktime_to_timespec(remaining); + } else { + cur_setting->it_value = ktime_to_timespec64(remaining); + } } /* Get the time remaining on a POSIX.1b interval timer. */ -SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, - struct itimerspec __user *, setting) +static int do_timer_gettime(timer_t timer_id, struct itimerspec64 *setting) { - struct itimerspec cur_setting; struct k_itimer *timr; - struct k_clock *kc; + const struct k_clock *kc; unsigned long flags; int ret = 0; @@ -787,19 +724,45 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, if (!timr) return -EINVAL; - kc = clockid_to_kclock(timr->it_clock); + memset(setting, 0, sizeof(*setting)); + kc = timr->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_get)) ret = -EINVAL; else - kc->timer_get(timr, &cur_setting); + kc->timer_get(timr, setting); unlock_timer(timr, flags); + return ret; +} - if (!ret && copy_to_user(setting, &cur_setting, sizeof (cur_setting))) - return -EFAULT; +/* Get the time remaining on a POSIX.1b interval timer. */ +SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, + struct itimerspec __user *, setting) +{ + struct itimerspec64 cur_setting; + + int ret = do_timer_gettime(timer_id, &cur_setting); + if (!ret) { + if (put_itimerspec64(&cur_setting, setting)) + ret = -EFAULT; + } + return ret; +} + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, + struct compat_itimerspec __user *, setting) +{ + struct itimerspec64 cur_setting; + int ret = do_timer_gettime(timer_id, &cur_setting); + if (!ret) { + if (put_compat_itimerspec64(&cur_setting, setting)) + ret = -EFAULT; + } return ret; } +#endif /* * Get the number of overruns of a POSIX.1b interval timer. This is to @@ -807,7 +770,7 @@ SYSCALL_DEFINE2(timer_gettime, timer_t, timer_id, * accumulating overruns on the next timer. The overrun is frozen when * the signal is delivered, either at the notify time (if the info block * is not queued) or at the actual delivery time (as we are informed by - * the call back to do_schedule_next_timer(). So all we need to do is + * the call back to posixtimer_rearm(). So all we need to do is * to pick up the frozen overrun. */ SYSCALL_DEFINE1(timer_getoverrun, timer_t, timer_id) @@ -826,114 +789,175 @@ SYSCALL_DEFINE1(timer_getoverrun, timer_t, timer_id) return overrun; } -/* Set a POSIX.1b interval timer. */ -/* timr->it_lock is taken. */ -static int -common_timer_set(struct k_itimer *timr, int flags, - struct itimerspec *new_setting, struct itimerspec *old_setting) +static void common_hrtimer_arm(struct k_itimer *timr, ktime_t expires, + bool absolute, bool sigev_none) { struct hrtimer *timer = &timr->it.real.timer; enum hrtimer_mode mode; + mode = absolute ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL; + /* + * Posix magic: Relative CLOCK_REALTIME timers are not affected by + * clock modifications, so they become CLOCK_MONOTONIC based under the + * hood. See hrtimer_init(). Update timr->kclock, so the generic + * functions which use timr->kclock->clock_get() work. + * + * Note: it_clock stays unmodified, because the next timer_set() might + * use ABSTIME, so it needs to switch back. + */ + if (timr->it_clock == CLOCK_REALTIME) + timr->kclock = absolute ? &clock_realtime : &clock_monotonic; + + hrtimer_init(&timr->it.real.timer, timr->it_clock, mode); + timr->it.real.timer.function = posix_timer_fn; + + if (!absolute) + expires = ktime_add_safe(expires, timer->base->get_time()); + hrtimer_set_expires(timer, expires); + + if (!sigev_none) + hrtimer_start_expires(timer, HRTIMER_MODE_ABS); +} + +static int common_hrtimer_try_to_cancel(struct k_itimer *timr) +{ + return hrtimer_try_to_cancel(&timr->it.real.timer); +} + +/* Set a POSIX.1b interval timer. */ +int common_timer_set(struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting) +{ + const struct k_clock *kc = timr->kclock; + bool sigev_none; + ktime_t expires; + if (old_setting) common_timer_get(timr, old_setting); - /* disable the timer */ - timr->it.real.interval = 0; + /* Prevent rearming by clearing the interval */ + timr->it_interval = 0; /* - * careful here. If smp we could be in the "fire" routine which will - * be spinning as we hold the lock. But this is ONLY an SMP issue. + * Careful here. On SMP systems the timer expiry function could be + * active and spinning on timr->it_lock. */ - if (hrtimer_try_to_cancel(timer) < 0) + if (kc->timer_try_to_cancel(timr) < 0) return TIMER_RETRY; - timr->it_requeue_pending = (timr->it_requeue_pending + 2) & + timr->it_active = 0; + timr->it_requeue_pending = (timr->it_requeue_pending + 2) & ~REQUEUE_PENDING; timr->it_overrun_last = 0; - /* switch off the timer when it_value is zero */ + /* Switch off the timer when it_value is zero */ if (!new_setting->it_value.tv_sec && !new_setting->it_value.tv_nsec) return 0; - mode = flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL; - hrtimer_init(&timr->it.real.timer, timr->it_clock, mode); - timr->it.real.timer.function = posix_timer_fn; - - hrtimer_set_expires(timer, timespec_to_ktime(new_setting->it_value)); - - /* Convert interval */ - timr->it.real.interval = timespec_to_ktime(new_setting->it_interval); - - /* SIGEV_NONE timers are not queued ! See common_timer_get */ - if (((timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE)) { - /* Setup correct expiry time for relative timers */ - if (mode == HRTIMER_MODE_REL) { - hrtimer_add_expires(timer, timer->base->get_time()); - } - return 0; - } + timr->it_interval = timespec64_to_ktime(new_setting->it_interval); + expires = timespec64_to_ktime(new_setting->it_value); + sigev_none = (timr->it_sigev_notify & ~SIGEV_THREAD_ID) == SIGEV_NONE; - hrtimer_start_expires(timer, mode); + kc->timer_arm(timr, expires, flags & TIMER_ABSTIME, sigev_none); + timr->it_active = !sigev_none; return 0; } -/* Set a POSIX.1b interval timer */ -SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, - const struct itimerspec __user *, new_setting, - struct itimerspec __user *, old_setting) +static int do_timer_settime(timer_t timer_id, int flags, + struct itimerspec64 *new_spec64, + struct itimerspec64 *old_spec64) { + const struct k_clock *kc; struct k_itimer *timr; - struct itimerspec new_spec, old_spec; - int error = 0; unsigned long flag; - struct itimerspec *rtn = old_setting ? &old_spec : NULL; - struct k_clock *kc; + int error = 0; - if (!new_setting) + if (!timespec64_valid(&new_spec64->it_interval) || + !timespec64_valid(&new_spec64->it_value)) return -EINVAL; - if (copy_from_user(&new_spec, new_setting, sizeof (new_spec))) - return -EFAULT; - - if (!timespec_valid(&new_spec.it_interval) || - !timespec_valid(&new_spec.it_value)) - return -EINVAL; + if (old_spec64) + memset(old_spec64, 0, sizeof(*old_spec64)); retry: timr = lock_timer(timer_id, &flag); if (!timr) return -EINVAL; - kc = clockid_to_kclock(timr->it_clock); + kc = timr->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_set)) error = -EINVAL; else - error = kc->timer_set(timr, flags, &new_spec, rtn); + error = kc->timer_set(timr, flags, new_spec64, old_spec64); unlock_timer(timr, flag); if (error == TIMER_RETRY) { - rtn = NULL; // We already got the old time... + old_spec64 = NULL; // We already got the old time... goto retry; } - if (old_setting && !error && - copy_to_user(old_setting, &old_spec, sizeof (old_spec))) - error = -EFAULT; + return error; +} + +/* Set a POSIX.1b interval timer */ +SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, + const struct itimerspec __user *, new_setting, + struct itimerspec __user *, old_setting) +{ + struct itimerspec64 new_spec, old_spec; + struct itimerspec64 *rtn = old_setting ? &old_spec : NULL; + int error = 0; + + if (!new_setting) + return -EINVAL; + if (get_itimerspec64(&new_spec, new_setting)) + return -EFAULT; + + error = do_timer_settime(timer_id, flags, &new_spec, rtn); + if (!error && old_setting) { + if (put_itimerspec64(&old_spec, old_setting)) + error = -EFAULT; + } + return error; +} + +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags, + struct compat_itimerspec __user *, new, + struct compat_itimerspec __user *, old) +{ + struct itimerspec64 new_spec, old_spec; + struct itimerspec64 *rtn = old ? &old_spec : NULL; + int error = 0; + + if (!new) + return -EINVAL; + if (get_compat_itimerspec64(&new_spec, new)) + return -EFAULT; + + error = do_timer_settime(timer_id, flags, &new_spec, rtn); + if (!error && old) { + if (put_compat_itimerspec64(&old_spec, old)) + error = -EFAULT; + } return error; } +#endif -static int common_timer_del(struct k_itimer *timer) +int common_timer_del(struct k_itimer *timer) { - timer->it.real.interval = 0; + const struct k_clock *kc = timer->kclock; - if (hrtimer_try_to_cancel(&timer->it.real.timer) < 0) + timer->it_interval = 0; + if (kc->timer_try_to_cancel(timer) < 0) return TIMER_RETRY; + timer->it_active = 0; return 0; } static inline int timer_delete_hook(struct k_itimer *timer) { - struct k_clock *kc = clockid_to_kclock(timer->it_clock); + const struct k_clock *kc = timer->kclock; if (WARN_ON_ONCE(!kc || !kc->timer_del)) return -EINVAL; @@ -1012,13 +1036,13 @@ void exit_itimers(struct signal_struct *sig) SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, const struct timespec __user *, tp) { - struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec new_tp; + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 new_tp; if (!kc || !kc->clock_set) return -EINVAL; - if (copy_from_user(&new_tp, tp, sizeof (*tp))) + if (get_timespec64(&new_tp, tp)) return -EFAULT; return kc->clock_set(which_clock, &new_tp); @@ -1027,8 +1051,8 @@ SYSCALL_DEFINE2(clock_settime, const clockid_t, which_clock, SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, struct timespec __user *,tp) { - struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec kernel_tp; + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 kernel_tp; int error; if (!kc) @@ -1036,7 +1060,7 @@ SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, error = kc->clock_get(which_clock, &kernel_tp); - if (!error && copy_to_user(tp, &kernel_tp, sizeof (kernel_tp))) + if (!error && put_timespec64(&kernel_tp, tp)) error = -EFAULT; return error; @@ -1045,7 +1069,7 @@ SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock, SYSCALL_DEFINE2(clock_adjtime, const clockid_t, which_clock, struct timex __user *, utx) { - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); struct timex ktx; int err; @@ -1068,8 +1092,8 @@ SYSCALL_DEFINE2(clock_adjtime, const clockid_t, which_clock, SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, struct timespec __user *, tp) { - struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec rtn_tp; + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 rtn_tp; int error; if (!kc) @@ -1077,19 +1101,97 @@ SYSCALL_DEFINE2(clock_getres, const clockid_t, which_clock, error = kc->clock_getres(which_clock, &rtn_tp); - if (!error && tp && copy_to_user(tp, &rtn_tp, sizeof (rtn_tp))) + if (!error && tp && put_timespec64(&rtn_tp, tp)) error = -EFAULT; return error; } +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE2(clock_settime, clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 ts; + + if (!kc || !kc->clock_set) + return -EINVAL; + + if (compat_get_timespec64(&ts, tp)) + return -EFAULT; + + return kc->clock_set(which_clock, &ts); +} + +COMPAT_SYSCALL_DEFINE2(clock_gettime, clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 ts; + int err; + + if (!kc) + return -EINVAL; + + err = kc->clock_get(which_clock, &ts); + + if (!err && compat_put_timespec64(&ts, tp)) + err = -EFAULT; + + return err; +} + +COMPAT_SYSCALL_DEFINE2(clock_adjtime, clockid_t, which_clock, + struct compat_timex __user *, utp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timex ktx; + int err; + + if (!kc) + return -EINVAL; + if (!kc->clock_adj) + return -EOPNOTSUPP; + + err = compat_get_timex(&ktx, utp); + if (err) + return err; + + err = kc->clock_adj(which_clock, &ktx); + + if (err >= 0) + err = compat_put_timex(utp, &ktx); + + return err; +} + +COMPAT_SYSCALL_DEFINE2(clock_getres, clockid_t, which_clock, + struct compat_timespec __user *, tp) +{ + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 ts; + int err; + + if (!kc) + return -EINVAL; + + err = kc->clock_getres(which_clock, &ts); + if (!err && tp && compat_put_timespec64(&ts, tp)) + return -EFAULT; + + return err; +} + +#endif + /* * nanosleep for monotonic and realtime clocks */ static int common_nsleep(const clockid_t which_clock, int flags, - struct timespec *tsave, struct timespec __user *rmtp) + const struct timespec64 *rqtp) { - return hrtimer_nanosleep(tsave, rmtp, flags & TIMER_ABSTIME ? + return hrtimer_nanosleep(rqtp, flags & TIMER_ABSTIME ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL, which_clock); } @@ -1098,34 +1200,152 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags, const struct timespec __user *, rqtp, struct timespec __user *, rmtp) { - struct k_clock *kc = clockid_to_kclock(which_clock); - struct timespec t; + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 t; if (!kc) return -EINVAL; if (!kc->nsleep) return -ENANOSLEEP_NOTSUP; - if (copy_from_user(&t, rqtp, sizeof (struct timespec))) + if (get_timespec64(&t, rqtp)) return -EFAULT; - if (!timespec_valid(&t)) + if (!timespec64_valid(&t)) return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE; + current->restart_block.nanosleep.rmtp = rmtp; - return kc->nsleep(which_clock, flags, &t, rmtp); + return kc->nsleep(which_clock, flags, &t); } -/* - * This will restart clock_nanosleep. This is required only by - * compat_clock_nanosleep_restart for now. - */ -long clock_nanosleep_restart(struct restart_block *restart_block) +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE4(clock_nanosleep, clockid_t, which_clock, int, flags, + struct compat_timespec __user *, rqtp, + struct compat_timespec __user *, rmtp) { - clockid_t which_clock = restart_block->nanosleep.clockid; - struct k_clock *kc = clockid_to_kclock(which_clock); + const struct k_clock *kc = clockid_to_kclock(which_clock); + struct timespec64 t; + + if (!kc) + return -EINVAL; + if (!kc->nsleep) + return -ENANOSLEEP_NOTSUP; - if (WARN_ON_ONCE(!kc || !kc->nsleep_restart)) + if (compat_get_timespec64(&t, rqtp)) + return -EFAULT; + + if (!timespec64_valid(&t)) return -EINVAL; + if (flags & TIMER_ABSTIME) + rmtp = NULL; + current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE; + current->restart_block.nanosleep.compat_rmtp = rmtp; + + return kc->nsleep(which_clock, flags, &t); +} +#endif - return kc->nsleep_restart(restart_block); +static const struct k_clock clock_realtime = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_clock_realtime_get, + .clock_set = posix_clock_realtime_set, + .clock_adj = posix_clock_realtime_adj, + .nsleep = common_nsleep, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, +}; + +static const struct k_clock clock_monotonic = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_ktime_get_ts, + .nsleep = common_nsleep, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, +}; + +static const struct k_clock clock_monotonic_raw = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_monotonic_raw, +}; + +static const struct k_clock clock_realtime_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_realtime_coarse, +}; + +static const struct k_clock clock_monotonic_coarse = { + .clock_getres = posix_get_coarse_res, + .clock_get = posix_get_monotonic_coarse, +}; + +static const struct k_clock clock_tai = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_tai, + .nsleep = common_nsleep, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, +}; + +static const struct k_clock clock_boottime = { + .clock_getres = posix_get_hrtimer_res, + .clock_get = posix_get_boottime, + .nsleep = common_nsleep, + .timer_create = common_timer_create, + .timer_set = common_timer_set, + .timer_get = common_timer_get, + .timer_del = common_timer_del, + .timer_rearm = common_hrtimer_rearm, + .timer_forward = common_hrtimer_forward, + .timer_remaining = common_hrtimer_remaining, + .timer_try_to_cancel = common_hrtimer_try_to_cancel, + .timer_arm = common_hrtimer_arm, +}; + +static const struct k_clock * const posix_clocks[] = { + [CLOCK_REALTIME] = &clock_realtime, + [CLOCK_MONOTONIC] = &clock_monotonic, + [CLOCK_PROCESS_CPUTIME_ID] = &clock_process, + [CLOCK_THREAD_CPUTIME_ID] = &clock_thread, + [CLOCK_MONOTONIC_RAW] = &clock_monotonic_raw, + [CLOCK_REALTIME_COARSE] = &clock_realtime_coarse, + [CLOCK_MONOTONIC_COARSE] = &clock_monotonic_coarse, + [CLOCK_BOOTTIME] = &clock_boottime, + [CLOCK_REALTIME_ALARM] = &alarm_clock, + [CLOCK_BOOTTIME_ALARM] = &alarm_clock, + [CLOCK_TAI] = &clock_tai, +}; + +static const struct k_clock *clockid_to_kclock(const clockid_t id) +{ + if (id < 0) + return (id & CLOCKFD_MASK) == CLOCKFD ? + &clock_posix_dynamic : &clock_posix_cpu; + + if (id >= ARRAY_SIZE(posix_clocks) || !posix_clocks[id]) + return NULL; + return posix_clocks[id]; } diff --git a/kernel/time/posix-timers.h b/kernel/time/posix-timers.h new file mode 100644 index 000000000000..fb303c3be4d3 --- /dev/null +++ b/kernel/time/posix-timers.h @@ -0,0 +1,40 @@ +#define TIMER_RETRY 1 + +struct k_clock { + int (*clock_getres)(const clockid_t which_clock, + struct timespec64 *tp); + int (*clock_set)(const clockid_t which_clock, + const struct timespec64 *tp); + int (*clock_get)(const clockid_t which_clock, + struct timespec64 *tp); + int (*clock_adj)(const clockid_t which_clock, struct timex *tx); + int (*timer_create)(struct k_itimer *timer); + int (*nsleep)(const clockid_t which_clock, int flags, + const struct timespec64 *); + int (*timer_set)(struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting); + int (*timer_del)(struct k_itimer *timr); + void (*timer_get)(struct k_itimer *timr, + struct itimerspec64 *cur_setting); + void (*timer_rearm)(struct k_itimer *timr); + int (*timer_forward)(struct k_itimer *timr, ktime_t now); + ktime_t (*timer_remaining)(struct k_itimer *timr, ktime_t now); + int (*timer_try_to_cancel)(struct k_itimer *timr); + void (*timer_arm)(struct k_itimer *timr, ktime_t expires, + bool absolute, bool sigev_none); +}; + +extern const struct k_clock clock_posix_cpu; +extern const struct k_clock clock_posix_dynamic; +extern const struct k_clock clock_process; +extern const struct k_clock clock_thread; +extern const struct k_clock alarm_clock; + +int posix_timer_event(struct k_itimer *timr, int si_private); + +void common_timer_get(struct k_itimer *timr, struct itimerspec64 *cur_setting); +int common_timer_set(struct k_itimer *timr, int flags, + struct itimerspec64 *new_setting, + struct itimerspec64 *old_setting); +int common_timer_del(struct k_itimer *timer); diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c index a26036d37a38..2d8f05aad442 100644 --- a/kernel/time/sched_clock.c +++ b/kernel/time/sched_clock.c @@ -13,6 +13,7 @@ #include <linux/kernel.h> #include <linux/moduleparam.h> #include <linux/sched.h> +#include <linux/sched/clock.h> #include <linux/syscore_ops.h> #include <linux/hrtimer.h> #include <linux/sched_clock.h> @@ -205,6 +206,11 @@ sched_clock_register(u64 (*read)(void), int bits, unsigned long rate) update_clock_read_data(&rd); + if (sched_clock_timer.function != NULL) { + /* update timeout for clock wrap */ + hrtimer_start(&sched_clock_timer, cd.wrap_kt, HRTIMER_MODE_REL); + } + r = rate; if (r >= 4000000) { r /= 1000000; diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 3109204c87cc..b398c2ea69b2 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -29,16 +29,19 @@ */ static struct tick_device tick_broadcast_device; -static cpumask_var_t tick_broadcast_mask; -static cpumask_var_t tick_broadcast_on; -static cpumask_var_t tmpmask; -static DEFINE_RAW_SPINLOCK(tick_broadcast_lock); +static cpumask_var_t tick_broadcast_mask __cpumask_var_read_mostly; +static cpumask_var_t tick_broadcast_on __cpumask_var_read_mostly; +static cpumask_var_t tmpmask __cpumask_var_read_mostly; static int tick_broadcast_forced; +static __cacheline_aligned_in_smp DEFINE_RAW_SPINLOCK(tick_broadcast_lock); + #ifdef CONFIG_TICK_ONESHOT +static void tick_broadcast_setup_oneshot(struct clock_event_device *bc); static void tick_broadcast_clear_oneshot(int cpu); static void tick_resume_broadcast_oneshot(struct clock_event_device *bc); #else +static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { BUG(); } static inline void tick_broadcast_clear_oneshot(int cpu) { } static inline void tick_resume_broadcast_oneshot(struct clock_event_device *bc) { } #endif @@ -347,17 +350,16 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) * * Called when the system enters a state where affected tick devices * might stop. Note: TICK_BROADCAST_FORCE cannot be undone. - * - * Called with interrupts disabled, so clockevents_lock is not - * required here because the local clock event device cannot go away - * under us. */ void tick_broadcast_control(enum tick_broadcast_mode mode) { struct clock_event_device *bc, *dev; struct tick_device *td; int cpu, bc_stopped; + unsigned long flags; + /* Protects also the local clockevent device. */ + raw_spin_lock_irqsave(&tick_broadcast_lock, flags); td = this_cpu_ptr(&tick_cpu_device); dev = td->evtdev; @@ -365,12 +367,11 @@ void tick_broadcast_control(enum tick_broadcast_mode mode) * Is the device not affected by the powerstate ? */ if (!dev || !(dev->features & CLOCK_EVT_FEAT_C3STOP)) - return; + goto out; if (!tick_device_is_functional(dev)) - return; + goto out; - raw_spin_lock(&tick_broadcast_lock); cpu = smp_processor_id(); bc = tick_broadcast_device.evtdev; bc_stopped = cpumask_empty(tick_broadcast_mask); @@ -420,7 +421,8 @@ void tick_broadcast_control(enum tick_broadcast_mode mode) tick_broadcast_setup_oneshot(bc); } } - raw_spin_unlock(&tick_broadcast_lock); +out: + raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } EXPORT_SYMBOL_GPL(tick_broadcast_control); @@ -517,9 +519,9 @@ void tick_resume_broadcast(void) #ifdef CONFIG_TICK_ONESHOT -static cpumask_var_t tick_broadcast_oneshot_mask; -static cpumask_var_t tick_broadcast_pending_mask; -static cpumask_var_t tick_broadcast_force_mask; +static cpumask_var_t tick_broadcast_oneshot_mask __cpumask_var_read_mostly; +static cpumask_var_t tick_broadcast_pending_mask __cpumask_var_read_mostly; +static cpumask_var_t tick_broadcast_force_mask __cpumask_var_read_mostly; /* * Exposed for debugging: see timer_list.c @@ -867,7 +869,7 @@ static void tick_broadcast_init_next_event(struct cpumask *mask, /** * tick_broadcast_setup_oneshot - setup the broadcast device */ -void tick_broadcast_setup_oneshot(struct clock_event_device *bc) +static void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { int cpu = smp_processor_id(); diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index f738251000fe..be0ac01f2e12 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -126,7 +126,6 @@ static inline int tick_check_oneshot_change(int allow_nohz) { return 0; } /* Functions related to oneshot broadcasting */ #if defined(CONFIG_GENERIC_CLOCKEVENTS_BROADCAST) && defined(CONFIG_TICK_ONESHOT) -extern void tick_broadcast_setup_oneshot(struct clock_event_device *bc); extern void tick_broadcast_switch_to_oneshot(void); extern void tick_shutdown_broadcast_oneshot(unsigned int cpu); extern int tick_broadcast_oneshot_active(void); @@ -134,7 +133,6 @@ extern void tick_check_oneshot_broadcast_this_cpu(void); bool tick_broadcast_oneshot_available(void); extern struct cpumask *tick_get_broadcast_oneshot_mask(void); #else /* !(BROADCAST && ONESHOT): */ -static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { BUG(); } static inline void tick_broadcast_switch_to_oneshot(void) { } static inline void tick_shutdown_broadcast_oneshot(unsigned int cpu) { } static inline int tick_broadcast_oneshot_active(void) { return 0; } diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 74e0388cc88d..c7a899c5ce64 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -17,8 +17,12 @@ #include <linux/interrupt.h> #include <linux/kernel_stat.h> #include <linux/percpu.h> +#include <linux/nmi.h> #include <linux/profile.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/clock.h> +#include <linux/sched/stat.h> +#include <linux/sched/nohz.h> #include <linux/module.h> #include <linux/irq_work.h> #include <linux/posix-timers.h> @@ -146,6 +150,12 @@ static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs) touch_softlockup_watchdog_sched(); if (is_idle_task(current)) ts->idle_jiffies++; + /* + * In case the current tick fired too early past its expected + * expiration, make sure we don't bypass the next clock reprogramming + * to the same deadline. + */ + ts->next_tick = 0; } #endif update_process_times(user_mode(regs)); @@ -550,7 +560,7 @@ static void tick_nohz_stop_idle(struct tick_sched *ts, ktime_t now) update_ts_time_stats(smp_processor_id(), ts, now, NULL); ts->idle_active = 0; - sched_clock_idle_wakeup_event(0); + sched_clock_idle_wakeup_event(); } static ktime_t tick_nohz_start_idle(struct tick_sched *ts) @@ -656,6 +666,12 @@ static void tick_nohz_restart(struct tick_sched *ts, ktime_t now) hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED); else tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1); + + /* + * Reset to make sure next tick stop doesn't get fooled by past + * cached clock deadline. + */ + ts->next_tick = 0; } static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, @@ -697,8 +713,6 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, */ delta = next_tick - basemono; if (delta <= (u64)TICK_NSEC) { - tick = 0; - /* * Tell the timer code that the base is not idle, i.e. undo * the effect of get_next_timer_interrupt(): @@ -708,23 +722,8 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, * We've not stopped the tick yet, and there's a timer in the * next period, so no point in stopping it either, bail. */ - if (!ts->tick_stopped) - goto out; - - /* - * If, OTOH, we did stop it, but there's a pending (expired) - * timer reprogram the timer hardware to fire now. - * - * We will not restart the tick proper, just prod the timer - * hardware into firing an interrupt to process the pending - * timers. Just like tick_irq_exit() will not restart the tick - * for 'normal' interrupts. - * - * Only once we exit the idle loop will we re-enable the tick, - * see tick_nohz_idle_exit(). - */ - if (delta == 0) { - tick_nohz_restart(ts, now); + if (!ts->tick_stopped) { + tick = 0; goto out; } } @@ -767,8 +766,16 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, tick = expires; /* Skip reprogram of event if its not changed */ - if (ts->tick_stopped && (expires == ts->next_tick)) - goto out; + if (ts->tick_stopped && (expires == ts->next_tick)) { + /* Sanity check: make sure clockevent is actually programmed */ + if (tick == KTIME_MAX || ts->next_tick == hrtimer_get_expires(&ts->sched_timer)) + goto out; + + WARN_ON_ONCE(1); + printk_once("basemono: %llu ts->next_tick: %llu dev->next_event: %llu timer->active: %d timer->expires: %llu\n", + basemono, ts->next_tick, dev->next_event, + hrtimer_active(&ts->sched_timer), hrtimer_get_expires(&ts->sched_timer)); + } /* * nohz_stop_sched_tick can be called several times before @@ -778,8 +785,7 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, * the scheduler tick in nohz_restart_sched_tick. */ if (!ts->tick_stopped) { - nohz_balance_enter_idle(cpu); - calc_load_enter_idle(); + calc_load_nohz_start(); cpu_load_update_nohz_start(); ts->last_tick = hrtimer_get_expires(&ts->sched_timer); @@ -799,8 +805,10 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts, goto out; } + hrtimer_set_expires(&ts->sched_timer, tick); + if (ts->nohz_mode == NOHZ_MODE_HIGHRES) - hrtimer_start(&ts->sched_timer, tick, HRTIMER_MODE_ABS_PINNED); + hrtimer_start_expires(&ts->sched_timer, HRTIMER_MODE_ABS_PINNED); else tick_program_event(tick, 1); out: @@ -824,7 +832,7 @@ static void tick_nohz_restart_sched_tick(struct tick_sched *ts, ktime_t now) */ timer_clear_idle(); - calc_load_exit_idle(); + calc_load_nohz_stop(); touch_softlockup_watchdog_sched(); /* * Cancel the scheduled timer and restore the tick @@ -865,6 +873,11 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts) if (unlikely(!cpu_online(cpu))) { if (cpu == tick_do_timer_cpu) tick_do_timer_cpu = TICK_DO_TIMER_NONE; + /* + * Make sure the CPU doesn't get fooled by obsolete tick + * deadline if it comes back online later. + */ + ts->next_tick = 0; return false; } @@ -924,8 +937,10 @@ static void __tick_nohz_idle_enter(struct tick_sched *ts) ts->idle_expires = expires; } - if (!was_stopped && ts->tick_stopped) + if (!was_stopped && ts->tick_stopped) { ts->idle_jiffies = ts->last_jiffies; + nohz_balance_enter_idle(cpu); + } } } @@ -994,6 +1009,18 @@ ktime_t tick_nohz_get_sleep_length(void) return ts->sleep_length; } +/** + * tick_nohz_get_idle_calls - return the current idle calls counter value + * + * Called from the schedutil frequency scaling governor in scheduler context. + */ +unsigned long tick_nohz_get_idle_calls(void) +{ + struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched); + + return ts->idle_calls; +} + static void tick_nohz_account_idle_ticks(struct tick_sched *ts) { #ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE @@ -1161,6 +1188,8 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer) */ if (regs) tick_sched_handle(ts, regs); + else + ts->next_tick = 0; /* No need to reprogram if we are in idle or full dynticks mode */ if (unlikely(ts->tick_stopped)) diff --git a/kernel/time/time.c b/kernel/time/time.c index a3a9a8a029dc..44a8c1402133 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -39,6 +39,7 @@ #include <linux/ptrace.h> #include <linux/uaccess.h> +#include <linux/compat.h> #include <asm/unistd.h> #include <generated/timeconst.h> @@ -99,6 +100,47 @@ SYSCALL_DEFINE1(stime, time_t __user *, tptr) #endif /* __ARCH_WANT_SYS_TIME */ +#ifdef CONFIG_COMPAT +#ifdef __ARCH_WANT_COMPAT_SYS_TIME + +/* compat_time_t is a 32 bit "long" and needs to get converted. */ +COMPAT_SYSCALL_DEFINE1(time, compat_time_t __user *, tloc) +{ + struct timeval tv; + compat_time_t i; + + do_gettimeofday(&tv); + i = tv.tv_sec; + + if (tloc) { + if (put_user(i,tloc)) + return -EFAULT; + } + force_successful_syscall_return(); + return i; +} + +COMPAT_SYSCALL_DEFINE1(stime, compat_time_t __user *, tptr) +{ + struct timespec tv; + int err; + + if (get_user(tv.tv_sec, tptr)) + return -EFAULT; + + tv.tv_nsec = 0; + + err = security_settime(&tv, NULL); + if (err) + return err; + + do_settimeofday(&tv); + return 0; +} + +#endif /* __ARCH_WANT_COMPAT_SYS_TIME */ +#endif + SYSCALL_DEFINE2(gettimeofday, struct timeval __user *, tv, struct timezone __user *, tz) { @@ -193,8 +235,8 @@ int do_sys_settimeofday64(const struct timespec64 *tv, const struct timezone *tz SYSCALL_DEFINE2(settimeofday, struct timeval __user *, tv, struct timezone __user *, tz) { + struct timespec64 new_ts; struct timeval user_tv; - struct timespec new_ts; struct timezone new_tz; if (tv) { @@ -212,9 +254,50 @@ SYSCALL_DEFINE2(settimeofday, struct timeval __user *, tv, return -EFAULT; } - return do_sys_settimeofday(tv ? &new_ts : NULL, tz ? &new_tz : NULL); + return do_sys_settimeofday64(tv ? &new_ts : NULL, tz ? &new_tz : NULL); } +#ifdef CONFIG_COMPAT +COMPAT_SYSCALL_DEFINE2(gettimeofday, struct compat_timeval __user *, tv, + struct timezone __user *, tz) +{ + if (tv) { + struct timeval ktv; + + do_gettimeofday(&ktv); + if (compat_put_timeval(&ktv, tv)) + return -EFAULT; + } + if (tz) { + if (copy_to_user(tz, &sys_tz, sizeof(sys_tz))) + return -EFAULT; + } + + return 0; +} + +COMPAT_SYSCALL_DEFINE2(settimeofday, struct compat_timeval __user *, tv, + struct timezone __user *, tz) +{ + struct timespec64 new_ts; + struct timeval user_tv; + struct timezone new_tz; + + if (tv) { + if (compat_get_timeval(&user_tv, tv)) + return -EFAULT; + new_ts.tv_sec = user_tv.tv_sec; + new_ts.tv_nsec = user_tv.tv_usec * NSEC_PER_USEC; + } + if (tz) { + if (copy_from_user(&new_tz, tz, sizeof(*tz))) + return -EFAULT; + } + + return do_sys_settimeofday64(tv ? &new_ts : NULL, tz ? &new_tz : NULL); +} +#endif + SYSCALL_DEFINE1(adjtimex, struct timex __user *, txc_p) { struct timex txc; /* Local copy of parameter */ @@ -224,25 +307,32 @@ SYSCALL_DEFINE1(adjtimex, struct timex __user *, txc_p) * structure. But bear in mind that the structures * may change */ - if(copy_from_user(&txc, txc_p, sizeof(struct timex))) + if (copy_from_user(&txc, txc_p, sizeof(struct timex))) return -EFAULT; ret = do_adjtimex(&txc); return copy_to_user(txc_p, &txc, sizeof(struct timex)) ? -EFAULT : ret; } -/** - * current_fs_time - Return FS time - * @sb: Superblock. - * - * Return the current time truncated to the time granularity supported by - * the fs. - */ -struct timespec current_fs_time(struct super_block *sb) +#ifdef CONFIG_COMPAT + +COMPAT_SYSCALL_DEFINE1(adjtimex, struct compat_timex __user *, utp) { - struct timespec now = current_kernel_time(); - return timespec_trunc(now, sb->s_time_gran); + struct timex txc; + int err, ret; + + err = compat_get_timex(&txc, utp); + if (err) + return err; + + ret = do_adjtimex(&txc); + + err = compat_put_timex(utp, &txc); + if (err) + return err; + + return ret; } -EXPORT_SYMBOL(current_fs_time); +#endif /* * Convert jiffies to milliseconds and back. @@ -702,6 +792,16 @@ u64 nsec_to_clock_t(u64 x) #endif } +u64 jiffies64_to_nsecs(u64 j) +{ +#if !(NSEC_PER_SEC % HZ) + return (NSEC_PER_SEC / HZ) * j; +# else + return div_u64(j * HZ_TO_NSEC_NUM, HZ_TO_NSEC_DEN); +#endif +} +EXPORT_SYMBOL(jiffies64_to_nsecs); + /** * nsecs_to_jiffies64 - Convert nsecs in u64 to jiffies64 * @@ -790,3 +890,61 @@ struct timespec64 timespec64_add_safe(const struct timespec64 lhs, return res; } + +int get_timespec64(struct timespec64 *ts, + const struct timespec __user *uts) +{ + struct timespec kts; + int ret; + + ret = copy_from_user(&kts, uts, sizeof(kts)); + if (ret) + return -EFAULT; + + ts->tv_sec = kts.tv_sec; + ts->tv_nsec = kts.tv_nsec; + + return 0; +} +EXPORT_SYMBOL_GPL(get_timespec64); + +int put_timespec64(const struct timespec64 *ts, + struct timespec __user *uts) +{ + struct timespec kts = { + .tv_sec = ts->tv_sec, + .tv_nsec = ts->tv_nsec + }; + return copy_to_user(uts, &kts, sizeof(kts)) ? -EFAULT : 0; +} +EXPORT_SYMBOL_GPL(put_timespec64); + +int get_itimerspec64(struct itimerspec64 *it, + const struct itimerspec __user *uit) +{ + int ret; + + ret = get_timespec64(&it->it_interval, &uit->it_interval); + if (ret) + return ret; + + ret = get_timespec64(&it->it_value, &uit->it_value); + + return ret; +} +EXPORT_SYMBOL_GPL(get_itimerspec64); + +int put_itimerspec64(const struct itimerspec64 *it, + struct itimerspec __user *uit) +{ + int ret; + + ret = put_timespec64(&it->it_interval, &uit->it_interval); + if (ret) + return ret; + + ret = put_timespec64(&it->it_value, &uit->it_value); + + return ret; +} +EXPORT_SYMBOL_GPL(put_itimerspec64); diff --git a/kernel/time/timeconst.bc b/kernel/time/timeconst.bc index c48688904f9f..f83bbb81600b 100644 --- a/kernel/time/timeconst.bc +++ b/kernel/time/timeconst.bc @@ -98,6 +98,12 @@ define timeconst(hz) { print "#define HZ_TO_USEC_DEN\t\t", hz/cd, "\n" print "#define USEC_TO_HZ_NUM\t\t", hz/cd, "\n" print "#define USEC_TO_HZ_DEN\t\t", 1000000/cd, "\n" + + cd=gcd(hz,1000000000) + print "#define HZ_TO_NSEC_NUM\t\t", 1000000000/cd, "\n" + print "#define HZ_TO_NSEC_DEN\t\t", hz/cd, "\n" + print "#define NSEC_TO_HZ_NUM\t\t", hz/cd, "\n" + print "#define NSEC_TO_HZ_DEN\t\t", 1000000000/cd, "\n" print "\n" print "#endif /* KERNEL_TIMECONST_H */\n" diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index db087d7e106d..cedafa008de5 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -14,7 +14,9 @@ #include <linux/percpu.h> #include <linux/init.h> #include <linux/mm.h> +#include <linux/nmi.h> #include <linux/sched.h> +#include <linux/sched/loadavg.h> #include <linux/syscore_ops.h> #include <linux/clocksource.h> #include <linux/jiffies.h> @@ -70,6 +72,10 @@ static inline void tk_normalize_xtime(struct timekeeper *tk) tk->tkr_mono.xtime_nsec -= (u64)NSEC_PER_SEC << tk->tkr_mono.shift; tk->xtime_sec++; } + while (tk->tkr_raw.xtime_nsec >= ((u64)NSEC_PER_SEC << tk->tkr_raw.shift)) { + tk->tkr_raw.xtime_nsec -= (u64)NSEC_PER_SEC << tk->tkr_raw.shift; + tk->raw_sec++; + } } static inline struct timespec64 tk_xtime(struct timekeeper *tk) @@ -116,6 +122,26 @@ static inline void tk_update_sleep_time(struct timekeeper *tk, ktime_t delta) tk->offs_boot = ktime_add(tk->offs_boot, delta); } +/* + * tk_clock_read - atomic clocksource read() helper + * + * This helper is necessary to use in the read paths because, while the + * seqlock ensures we don't return a bad value while structures are updated, + * it doesn't protect from potential crashes. There is the possibility that + * the tkr's clocksource may change between the read reference, and the + * clock reference passed to the read function. This can cause crashes if + * the wrong clocksource is passed to the wrong read function. + * This isn't necessary to use when holding the timekeeper_lock or doing + * a read of the fast-timekeeper tkrs (which is protected by its own locking + * and update logic). + */ +static inline u64 tk_clock_read(struct tk_read_base *tkr) +{ + struct clocksource *clock = READ_ONCE(tkr->clock); + + return clock->read(clock); +} + #ifdef CONFIG_DEBUG_TIMEKEEPING #define WARNING_FREQ (HZ*300) /* 5 minute rate-limiting */ @@ -173,7 +199,7 @@ static inline u64 timekeeping_get_delta(struct tk_read_base *tkr) */ do { seq = read_seqcount_begin(&tk_core.seq); - now = tkr->read(tkr->clock); + now = tk_clock_read(tkr); last = tkr->cycle_last; mask = tkr->mask; max = tkr->clock->max_cycles; @@ -207,7 +233,7 @@ static inline u64 timekeeping_get_delta(struct tk_read_base *tkr) u64 cycle_now, delta; /* read clocksource */ - cycle_now = tkr->read(tkr->clock); + cycle_now = tk_clock_read(tkr); /* calculate the delta since the last update_wall_time */ delta = clocksource_delta(cycle_now, tkr->cycle_last, tkr->mask); @@ -236,12 +262,10 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) ++tk->cs_was_changed_seq; old_clock = tk->tkr_mono.clock; tk->tkr_mono.clock = clock; - tk->tkr_mono.read = clock->read; tk->tkr_mono.mask = clock->mask; - tk->tkr_mono.cycle_last = tk->tkr_mono.read(clock); + tk->tkr_mono.cycle_last = tk_clock_read(&tk->tkr_mono); tk->tkr_raw.clock = clock; - tk->tkr_raw.read = clock->read; tk->tkr_raw.mask = clock->mask; tk->tkr_raw.cycle_last = tk->tkr_mono.cycle_last; @@ -260,17 +284,19 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock) /* Go back from cycles -> shifted ns */ tk->xtime_interval = interval * clock->mult; tk->xtime_remainder = ntpinterval - tk->xtime_interval; - tk->raw_interval = (interval * clock->mult) >> clock->shift; + tk->raw_interval = interval * clock->mult; /* if changing clocks, convert xtime_nsec shift units */ if (old_clock) { int shift_change = clock->shift - old_clock->shift; - if (shift_change < 0) + if (shift_change < 0) { tk->tkr_mono.xtime_nsec >>= -shift_change; - else + tk->tkr_raw.xtime_nsec >>= -shift_change; + } else { tk->tkr_mono.xtime_nsec <<= shift_change; + tk->tkr_raw.xtime_nsec <<= shift_change; + } } - tk->tkr_raw.xtime_nsec = 0; tk->tkr_mono.shift = clock->shift; tk->tkr_raw.shift = clock->shift; @@ -402,7 +428,7 @@ static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf) now += timekeeping_delta_to_ns(tkr, clocksource_delta( - tkr->read(tkr->clock), + tk_clock_read(tkr), tkr->cycle_last, tkr->mask)); } while (read_seqcount_retry(&tkf->seq, seq)); @@ -459,6 +485,10 @@ static u64 dummy_clock_read(struct clocksource *cs) return cycles_at_suspend; } +static struct clocksource dummy_clock = { + .read = dummy_clock_read, +}; + /** * halt_fast_timekeeper - Prevent fast timekeeper from accessing clocksource. * @tk: Timekeeper to snapshot. @@ -475,17 +505,18 @@ static void halt_fast_timekeeper(struct timekeeper *tk) struct tk_read_base *tkr = &tk->tkr_mono; memcpy(&tkr_dummy, tkr, sizeof(tkr_dummy)); - cycles_at_suspend = tkr->read(tkr->clock); - tkr_dummy.read = dummy_clock_read; + cycles_at_suspend = tk_clock_read(tkr); + tkr_dummy.clock = &dummy_clock; update_fast_timekeeper(&tkr_dummy, &tk_fast_mono); tkr = &tk->tkr_raw; memcpy(&tkr_dummy, tkr, sizeof(tkr_dummy)); - tkr_dummy.read = dummy_clock_read; + tkr_dummy.clock = &dummy_clock; update_fast_timekeeper(&tkr_dummy, &tk_fast_raw); } #ifdef CONFIG_GENERIC_TIME_VSYSCALL_OLD +#warning Please contact your maintainers, as GENERIC_TIME_VSYSCALL_OLD compatibity will disappear soon. static inline void update_vsyscall(struct timekeeper *tk) { @@ -595,9 +626,6 @@ static inline void tk_update_ktime_data(struct timekeeper *tk) nsec = (u32) tk->wall_to_monotonic.tv_nsec; tk->tkr_mono.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); - /* Update the monotonic raw base */ - tk->tkr_raw.base = timespec64_to_ktime(tk->raw_time); - /* * The sum of the nanoseconds portions of xtime and * wall_to_monotonic can be greater/equal one second. Take @@ -607,6 +635,11 @@ static inline void tk_update_ktime_data(struct timekeeper *tk) if (nsec >= NSEC_PER_SEC) seconds++; tk->ktime_sec = seconds; + + /* Update the monotonic raw base */ + seconds = tk->raw_sec; + nsec = (u32)(tk->tkr_raw.xtime_nsec >> tk->tkr_raw.shift); + tk->tkr_raw.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); } /* must hold timekeeper_lock */ @@ -647,11 +680,9 @@ static void timekeeping_update(struct timekeeper *tk, unsigned int action) */ static void timekeeping_forward_now(struct timekeeper *tk) { - struct clocksource *clock = tk->tkr_mono.clock; u64 cycle_now, delta; - u64 nsec; - cycle_now = tk->tkr_mono.read(clock); + cycle_now = tk_clock_read(&tk->tkr_mono); delta = clocksource_delta(cycle_now, tk->tkr_mono.cycle_last, tk->tkr_mono.mask); tk->tkr_mono.cycle_last = cycle_now; tk->tkr_raw.cycle_last = cycle_now; @@ -661,10 +692,13 @@ static void timekeeping_forward_now(struct timekeeper *tk) /* If arch requires, add in get_arch_timeoffset() */ tk->tkr_mono.xtime_nsec += (u64)arch_gettimeoffset() << tk->tkr_mono.shift; - tk_normalize_xtime(tk); - nsec = clocksource_cyc2ns(delta, tk->tkr_raw.mult, tk->tkr_raw.shift); - timespec64_add_ns(&tk->raw_time, nsec); + tk->tkr_raw.xtime_nsec += delta * tk->tkr_raw.mult; + + /* If arch requires, add in get_arch_timeoffset() */ + tk->tkr_raw.xtime_nsec += (u64)arch_gettimeoffset() << tk->tkr_raw.shift; + + tk_normalize_xtime(tk); } /** @@ -927,8 +961,7 @@ void ktime_get_snapshot(struct system_time_snapshot *systime_snapshot) do { seq = read_seqcount_begin(&tk_core.seq); - - now = tk->tkr_mono.read(tk->tkr_mono.clock); + now = tk_clock_read(&tk->tkr_mono); systime_snapshot->cs_was_changed_seq = tk->cs_was_changed_seq; systime_snapshot->clock_was_set_seq = tk->clock_was_set_seq; base_real = ktime_add(tk->tkr_mono.base, @@ -994,8 +1027,7 @@ static int adjust_historical_crosststamp(struct system_time_snapshot *history, return 0; /* Interpolate shortest distance from beginning or end of history */ - interp_forward = partial_history_cycles > total_history_cycles/2 ? - true : false; + interp_forward = partial_history_cycles > total_history_cycles / 2; partial_history_cycles = interp_forward ? total_history_cycles - partial_history_cycles : partial_history_cycles; @@ -1107,7 +1139,7 @@ int get_device_system_crosststamp(int (*get_time_fn) * Check whether the system counter value provided by the * device driver is on the current timekeeping interval. */ - now = tk->tkr_mono.read(tk->tkr_mono.clock); + now = tk_clock_read(&tk->tkr_mono); interval_start = tk->tkr_mono.cycle_last; if (!cycle_between(interval_start, cycles, now)) { clock_was_set_seq = tk->clock_was_set_seq; @@ -1275,27 +1307,8 @@ error: /* even if we error out, we forwarded the time, so call update */ } EXPORT_SYMBOL(timekeeping_inject_offset); - -/** - * timekeeping_get_tai_offset - Returns current TAI offset from UTC - * - */ -s32 timekeeping_get_tai_offset(void) -{ - struct timekeeper *tk = &tk_core.timekeeper; - unsigned int seq; - s32 ret; - - do { - seq = read_seqcount_begin(&tk_core.seq); - ret = tk->tai_offset; - } while (read_seqcount_retry(&tk_core.seq, seq)); - - return ret; -} - /** - * __timekeeping_set_tai_offset - Lock free worker function + * __timekeeping_set_tai_offset - Sets the TAI offset from UTC and monotonic * */ static void __timekeeping_set_tai_offset(struct timekeeper *tk, s32 tai_offset) @@ -1305,24 +1318,6 @@ static void __timekeeping_set_tai_offset(struct timekeeper *tk, s32 tai_offset) } /** - * timekeeping_set_tai_offset - Sets the current TAI offset from UTC - * - */ -void timekeeping_set_tai_offset(s32 tai_offset) -{ - struct timekeeper *tk = &tk_core.timekeeper; - unsigned long flags; - - raw_spin_lock_irqsave(&timekeeper_lock, flags); - write_seqcount_begin(&tk_core.seq); - __timekeeping_set_tai_offset(tk, tai_offset); - timekeeping_update(tk, TK_MIRROR | TK_CLOCK_WAS_SET); - write_seqcount_end(&tk_core.seq); - raw_spin_unlock_irqrestore(&timekeeper_lock, flags); - clock_was_set(); -} - -/** * change_clocksource - Swaps clocksources if a new one is available * * Accumulates current time interval and initializes new clocksource @@ -1389,19 +1384,18 @@ int timekeeping_notify(struct clocksource *clock) void getrawmonotonic64(struct timespec64 *ts) { struct timekeeper *tk = &tk_core.timekeeper; - struct timespec64 ts64; unsigned long seq; u64 nsecs; do { seq = read_seqcount_begin(&tk_core.seq); + ts->tv_sec = tk->raw_sec; nsecs = timekeeping_get_ns(&tk->tkr_raw); - ts64 = tk->raw_time; } while (read_seqcount_retry(&tk_core.seq, seq)); - timespec64_add_ns(&ts64, nsecs); - *ts = ts64; + ts->tv_nsec = 0; + timespec64_add_ns(ts, nsecs); } EXPORT_SYMBOL(getrawmonotonic64); @@ -1525,8 +1519,7 @@ void __init timekeeping_init(void) tk_setup_internals(tk, clock); tk_set_xtime(tk, &now); - tk->raw_time.tv_sec = 0; - tk->raw_time.tv_nsec = 0; + tk->raw_sec = 0; if (boot.tv_sec == 0 && boot.tv_nsec == 0) boot = tk_xtime(tk); @@ -1665,7 +1658,7 @@ void timekeeping_resume(void) * The less preferred source will only be tried if there is no better * usable source. The rtc part is handled separately in rtc core code. */ - cycle_now = tk->tkr_mono.read(clock); + cycle_now = tk_clock_read(&tk->tkr_mono); if ((clock->flags & CLOCK_SOURCE_SUSPEND_NONSTOP) && cycle_now > tk->tkr_mono.cycle_last) { u64 nsec, cyc_delta; @@ -2012,7 +2005,7 @@ static u64 logarithmic_accumulation(struct timekeeper *tk, u64 offset, u32 shift, unsigned int *clock_set) { u64 interval = tk->cycle_interval << shift; - u64 raw_nsecs; + u64 snsec_per_sec; /* If the offset is smaller than a shifted interval, do nothing */ if (offset < interval) @@ -2027,14 +2020,12 @@ static u64 logarithmic_accumulation(struct timekeeper *tk, u64 offset, *clock_set |= accumulate_nsecs_to_secs(tk); /* Accumulate raw time */ - raw_nsecs = (u64)tk->raw_interval << shift; - raw_nsecs += tk->raw_time.tv_nsec; - if (raw_nsecs >= NSEC_PER_SEC) { - u64 raw_secs = raw_nsecs; - raw_nsecs = do_div(raw_secs, NSEC_PER_SEC); - tk->raw_time.tv_sec += raw_secs; + tk->tkr_raw.xtime_nsec += tk->raw_interval << shift; + snsec_per_sec = (u64)NSEC_PER_SEC << tk->tkr_raw.shift; + while (tk->tkr_raw.xtime_nsec >= snsec_per_sec) { + tk->tkr_raw.xtime_nsec -= snsec_per_sec; + tk->raw_sec++; } - tk->raw_time.tv_nsec = raw_nsecs; /* Accumulate error between NTP and clock interval */ tk->ntp_error += tk->ntp_tick << shift; @@ -2066,7 +2057,7 @@ void update_wall_time(void) #ifdef CONFIG_ARCH_USES_GETTIMEOFFSET offset = real_tk->cycle_interval; #else - offset = clocksource_delta(tk->tkr_mono.read(tk->tkr_mono.clock), + offset = clocksource_delta(tk_clock_read(&tk->tkr_mono), tk->tkr_mono.cycle_last, tk->tkr_mono.mask); #endif diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h index 704f595ce83f..d0914676d4c5 100644 --- a/kernel/time/timekeeping.h +++ b/kernel/time/timekeeping.h @@ -11,8 +11,6 @@ extern ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, extern int timekeeping_valid_for_hres(void); extern u64 timekeeping_max_deferment(void); extern int timekeeping_inject_offset(struct timespec *ts); -extern s32 timekeeping_get_tai_offset(void); -extern void timekeeping_set_tai_offset(s32 tai_offset); extern int timekeeping_suspend(void); extern void timekeeping_resume(void); diff --git a/kernel/time/timekeeping_debug.c b/kernel/time/timekeeping_debug.c index ca9fb800336b..38bc4d2208e8 100644 --- a/kernel/time/timekeeping_debug.c +++ b/kernel/time/timekeeping_debug.c @@ -75,7 +75,7 @@ void tk_debug_account_sleep_time(struct timespec64 *t) int bin = min(fls(t->tv_sec), NUM_BINS-1); sleep_time_bin[bin]++; - pr_info("Suspended for %lld.%03lu seconds\n", (s64)t->tv_sec, - t->tv_nsec / NSEC_PER_MSEC); + printk_deferred(KERN_INFO "Suspended for %lld.%03lu seconds\n", + (s64)t->tv_sec, t->tv_nsec / NSEC_PER_MSEC); } diff --git a/kernel/time/timer.c b/kernel/time/timer.c index ec33a6933eae..71ce3f4eead3 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -38,8 +38,10 @@ #include <linux/tick.h> #include <linux/kallsyms.h> #include <linux/irq_work.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> #include <linux/sched/sysctl.h> +#include <linux/sched/nohz.h> +#include <linux/sched/debug.h> #include <linux/slab.h> #include <linux/compat.h> @@ -193,7 +195,7 @@ EXPORT_SYMBOL(jiffies_64); #endif struct timer_base { - spinlock_t lock; + raw_spinlock_t lock; struct timer_list *running_timer; unsigned long clk; unsigned long next_expiry; @@ -239,7 +241,7 @@ int timer_migration_handler(struct ctl_table *table, int write, int ret; mutex_lock(&mutex); - ret = proc_dointvec(table, write, buffer, lenp, ppos); + ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (!ret && write) timers_update_migration(false); mutex_unlock(&mutex); @@ -571,38 +573,6 @@ internal_add_timer(struct timer_base *base, struct timer_list *timer) trigger_dyntick_cpu(base, timer); } -#ifdef CONFIG_TIMER_STATS -void __timer_stats_timer_set_start_info(struct timer_list *timer, void *addr) -{ - if (timer->start_site) - return; - - timer->start_site = addr; - memcpy(timer->start_comm, current->comm, TASK_COMM_LEN); - timer->start_pid = current->pid; -} - -static void timer_stats_account_timer(struct timer_list *timer) -{ - void *site; - - /* - * start_site can be concurrently reset by - * timer_stats_timer_clear_start_info() - */ - site = READ_ONCE(timer->start_site); - if (likely(!site)) - return; - - timer_stats_update_stats(timer, timer->start_pid, site, - timer->function, timer->start_comm, - timer->flags); -} - -#else -static void timer_stats_account_timer(struct timer_list *timer) {} -#endif - #ifdef CONFIG_DEBUG_OBJECTS_TIMERS static struct debug_obj_descr timer_debug_descr; @@ -789,11 +759,6 @@ static void do_init_timer(struct timer_list *timer, unsigned int flags, { timer->entry.pprev = NULL; timer->flags = flags | raw_smp_processor_id(); -#ifdef CONFIG_TIMER_STATS - timer->start_site = NULL; - timer->start_pid = -1; - memset(timer->start_comm, 0, TASK_COMM_LEN); -#endif lockdep_init_map(&timer->lockdep_map, name, key, 0); } @@ -948,10 +913,10 @@ static struct timer_base *lock_timer_base(struct timer_list *timer, if (!(tf & TIMER_MIGRATING)) { base = get_timer_base(tf); - spin_lock_irqsave(&base->lock, *flags); + raw_spin_lock_irqsave(&base->lock, *flags); if (timer->flags == tf) return base; - spin_unlock_irqrestore(&base->lock, *flags); + raw_spin_unlock_irqrestore(&base->lock, *flags); } cpu_relax(); } @@ -1001,8 +966,6 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) base = lock_timer_base(timer, &flags); } - timer_stats_timer_set_start_info(timer); - ret = detach_if_pending(timer, base, false); if (!ret && pending_only) goto out_unlock; @@ -1023,9 +986,9 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) /* See the comment in lock_timer_base() */ timer->flags |= TIMER_MIGRATING; - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); base = new_base; - spin_lock(&base->lock); + raw_spin_lock(&base->lock); WRITE_ONCE(timer->flags, (timer->flags & ~TIMER_BASEMASK) | base->cpu); } @@ -1050,7 +1013,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only) } out_unlock: - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); return ret; } @@ -1130,7 +1093,6 @@ void add_timer_on(struct timer_list *timer, int cpu) struct timer_base *new_base, *base; unsigned long flags; - timer_stats_timer_set_start_info(timer); BUG_ON(timer_pending(timer) || !timer->function); new_base = get_timer_cpu_base(timer->flags, cpu); @@ -1144,21 +1106,21 @@ void add_timer_on(struct timer_list *timer, int cpu) if (base != new_base) { timer->flags |= TIMER_MIGRATING; - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); base = new_base; - spin_lock(&base->lock); + raw_spin_lock(&base->lock); WRITE_ONCE(timer->flags, (timer->flags & ~TIMER_BASEMASK) | cpu); } debug_activate(timer, timer->expires); internal_add_timer(base, timer); - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); } EXPORT_SYMBOL_GPL(add_timer_on); /** - * del_timer - deactive a timer. + * del_timer - deactivate a timer. * @timer: the timer to be deactivated * * del_timer() deactivates a timer - this works on both active and inactive @@ -1176,11 +1138,10 @@ int del_timer(struct timer_list *timer) debug_assert_init(timer); - timer_stats_timer_clear_start_info(timer); if (timer_pending(timer)) { base = lock_timer_base(timer, &flags); ret = detach_if_pending(timer, base, true); - spin_unlock_irqrestore(&base->lock, flags); + raw_spin_unlock_irqrestore(&base->lock, flags); } return ret; @@ -1189,7 +1150,7 @@ EXPORT_SYMBOL(del_timer); /** * try_to_del_timer_sync - Try to deactivate a timer - * @timer: timer do del + * @timer: timer to delete * * This function tries to deactivate a timer. Upon successful (ret >= 0) * exit the timer is not queued and the handler is not running on any CPU. @@ -1204,11 +1165,10 @@ int try_to_del_timer_sync(struct timer_list *timer) base = lock_timer_base(timer, &flags); - if (base->running_timer != timer) { - timer_stats_timer_clear_start_info(timer); + if (base->running_timer != timer) ret = detach_if_pending(timer, base, true); - } - spin_unlock_irqrestore(&base->lock, flags); + + raw_spin_unlock_irqrestore(&base->lock, flags); return ret; } @@ -1331,7 +1291,6 @@ static void expire_timers(struct timer_base *base, struct hlist_head *head) unsigned long data; timer = hlist_entry(head->first, struct timer_list, entry); - timer_stats_account_timer(timer); base->running_timer = timer; detach_timer(timer, true); @@ -1340,13 +1299,13 @@ static void expire_timers(struct timer_base *base, struct hlist_head *head) data = timer->data; if (timer->flags & TIMER_IRQSAFE) { - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); call_timer_fn(timer, fn, data); - spin_lock(&base->lock); + raw_spin_lock(&base->lock); } else { - spin_unlock_irq(&base->lock); + raw_spin_unlock_irq(&base->lock); call_timer_fn(timer, fn, data); - spin_lock_irq(&base->lock); + raw_spin_lock_irq(&base->lock); } } } @@ -1515,7 +1474,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) if (cpu_is_offline(smp_processor_id())) return expires; - spin_lock(&base->lock); + raw_spin_lock(&base->lock); nextevt = __next_timer_interrupt(base); is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA); base->next_expiry = nextevt; @@ -1543,7 +1502,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) if ((expires - basem) > TICK_NSEC) base->is_idle = true; } - spin_unlock(&base->lock); + raw_spin_unlock(&base->lock); return cmp_next_hrtimer_event(basem, expires); } @@ -1631,7 +1590,7 @@ static inline void __run_timers(struct timer_base *base) if (!time_after_eq(jiffies, base->clk)) return; - spin_lock_irq(&base->lock); + raw_spin_lock_irq(&base->lock); while (time_after_eq(jiffies, base->clk)) { @@ -1642,7 +1601,7 @@ static inline void __run_timers(struct timer_base *base) expire_timers(base, heads + levels); } base->running_timer = NULL; - spin_unlock_irq(&base->lock); + raw_spin_unlock_irq(&base->lock); } /* @@ -1827,16 +1786,16 @@ int timers_dead_cpu(unsigned int cpu) * The caller is globally serialized and nobody else * takes two locks at once, deadlock is not possible. */ - spin_lock_irq(&new_base->lock); - spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock_irq(&new_base->lock); + raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); BUG_ON(old_base->running_timer); for (i = 0; i < WHEEL_SIZE; i++) migrate_timer_list(new_base, old_base->vectors + i); - spin_unlock(&old_base->lock); - spin_unlock_irq(&new_base->lock); + raw_spin_unlock(&old_base->lock); + raw_spin_unlock_irq(&new_base->lock); put_cpu_ptr(&timer_bases); } return 0; @@ -1852,7 +1811,7 @@ static void __init init_timer_cpu(int cpu) for (i = 0; i < NR_BASES; i++) { base = per_cpu_ptr(&timer_bases[i], cpu); base->cpu = cpu; - spin_lock_init(&base->lock); + raw_spin_lock_init(&base->lock); base->clk = jiffies; } } @@ -1868,7 +1827,6 @@ static void __init init_timer_cpus(void) void __init init_timers(void) { init_timer_cpus(); - init_timer_stats(); open_softirq(TIMER_SOFTIRQ, run_timer_softirq); } diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index afe6cd1944fc..0e7f5428a148 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -16,6 +16,7 @@ #include <linux/sched.h> #include <linux/seq_file.h> #include <linux/kallsyms.h> +#include <linux/nmi.h> #include <linux/uaccess.h> @@ -62,21 +63,11 @@ static void print_timer(struct seq_file *m, struct hrtimer *taddr, struct hrtimer *timer, int idx, u64 now) { -#ifdef CONFIG_TIMER_STATS - char tmp[TASK_COMM_LEN + 1]; -#endif SEQ_printf(m, " #%d: ", idx); print_name_offset(m, taddr); SEQ_printf(m, ", "); print_name_offset(m, timer->function); SEQ_printf(m, ", S:%02x", timer->state); -#ifdef CONFIG_TIMER_STATS - SEQ_printf(m, ", "); - print_name_offset(m, timer->start_site); - memcpy(tmp, timer->start_comm, TASK_COMM_LEN); - tmp[TASK_COMM_LEN] = 0; - SEQ_printf(m, ", %s/%d", tmp, timer->start_pid); -#endif SEQ_printf(m, "\n"); SEQ_printf(m, " # expires at %Lu-%Lu nsecs [in %Ld to %Ld nsecs]\n", (unsigned long long)ktime_to_ns(hrtimer_get_softexpires(timer)), @@ -96,6 +87,9 @@ print_active_timers(struct seq_file *m, struct hrtimer_clock_base *base, next_one: i = 0; + + touch_nmi_watchdog(); + raw_spin_lock_irqsave(&base->cpu_base->lock, flags); curr = timerqueue_getnext(&base->active); @@ -127,7 +121,7 @@ print_base(struct seq_file *m, struct hrtimer_clock_base *base, u64 now) SEQ_printf(m, " .base: %pK\n", base); SEQ_printf(m, " .index: %d\n", base->index); - SEQ_printf(m, " .resolution: %u nsecs\n", (unsigned) hrtimer_resolution); + SEQ_printf(m, " .resolution: %u nsecs\n", hrtimer_resolution); SEQ_printf(m, " .get_time: "); print_name_offset(m, base->get_time); @@ -207,6 +201,8 @@ print_tickdevice(struct seq_file *m, struct tick_device *td, int cpu) { struct clock_event_device *dev = td->evtdev; + touch_nmi_watchdog(); + SEQ_printf(m, "Tick Device: mode: %d\n", td->mode); if (cpu < 0) SEQ_printf(m, "Broadcast device\n"); diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c deleted file mode 100644 index afddded947df..000000000000 --- a/kernel/time/timer_stats.c +++ /dev/null @@ -1,425 +0,0 @@ -/* - * kernel/time/timer_stats.c - * - * Collect timer usage statistics. - * - * Copyright(C) 2006, Red Hat, Inc., Ingo Molnar - * Copyright(C) 2006 Timesys Corp., Thomas Gleixner <[email protected]> - * - * timer_stats is based on timer_top, a similar functionality which was part of - * Con Kolivas dyntick patch set. It was developed by Daniel Petrini at the - * Instituto Nokia de Tecnologia - INdT - Manaus. timer_top's design was based - * on dynamic allocation of the statistics entries and linear search based - * lookup combined with a global lock, rather than the static array, hash - * and per-CPU locking which is used by timer_stats. It was written for the - * pre hrtimer kernel code and therefore did not take hrtimers into account. - * Nevertheless it provided the base for the timer_stats implementation and - * was a helpful source of inspiration. Kudos to Daniel and the Nokia folks - * for this effort. - * - * timer_top.c is - * Copyright (C) 2005 Instituto Nokia de Tecnologia - INdT - Manaus - * Written by Daniel Petrini <[email protected]> - * timer_top.c was released under the GNU General Public License version 2 - * - * We export the addresses and counting of timer functions being called, - * the pid and cmdline from the owner process if applicable. - * - * Start/stop data collection: - * # echo [1|0] >/proc/timer_stats - * - * Display the information collected so far: - * # cat /proc/timer_stats - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/proc_fs.h> -#include <linux/module.h> -#include <linux/spinlock.h> -#include <linux/sched.h> -#include <linux/seq_file.h> -#include <linux/kallsyms.h> - -#include <linux/uaccess.h> - -/* - * This is our basic unit of interest: a timer expiry event identified - * by the timer, its start/expire functions and the PID of the task that - * started the timer. We count the number of times an event happens: - */ -struct entry { - /* - * Hash list: - */ - struct entry *next; - - /* - * Hash keys: - */ - void *timer; - void *start_func; - void *expire_func; - pid_t pid; - - /* - * Number of timeout events: - */ - unsigned long count; - u32 flags; - - /* - * We save the command-line string to preserve - * this information past task exit: - */ - char comm[TASK_COMM_LEN + 1]; - -} ____cacheline_aligned_in_smp; - -/* - * Spinlock protecting the tables - not taken during lookup: - */ -static DEFINE_RAW_SPINLOCK(table_lock); - -/* - * Per-CPU lookup locks for fast hash lookup: - */ -static DEFINE_PER_CPU(raw_spinlock_t, tstats_lookup_lock); - -/* - * Mutex to serialize state changes with show-stats activities: - */ -static DEFINE_MUTEX(show_mutex); - -/* - * Collection status, active/inactive: - */ -int __read_mostly timer_stats_active; - -/* - * Beginning/end timestamps of measurement: - */ -static ktime_t time_start, time_stop; - -/* - * tstat entry structs only get allocated while collection is - * active and never freed during that time - this simplifies - * things quite a bit. - * - * They get freed when a new collection period is started. - */ -#define MAX_ENTRIES_BITS 10 -#define MAX_ENTRIES (1UL << MAX_ENTRIES_BITS) - -static unsigned long nr_entries; -static struct entry entries[MAX_ENTRIES]; - -static atomic_t overflow_count; - -/* - * The entries are in a hash-table, for fast lookup: - */ -#define TSTAT_HASH_BITS (MAX_ENTRIES_BITS - 1) -#define TSTAT_HASH_SIZE (1UL << TSTAT_HASH_BITS) -#define TSTAT_HASH_MASK (TSTAT_HASH_SIZE - 1) - -#define __tstat_hashfn(entry) \ - (((unsigned long)(entry)->timer ^ \ - (unsigned long)(entry)->start_func ^ \ - (unsigned long)(entry)->expire_func ^ \ - (unsigned long)(entry)->pid ) & TSTAT_HASH_MASK) - -#define tstat_hashentry(entry) (tstat_hash_table + __tstat_hashfn(entry)) - -static struct entry *tstat_hash_table[TSTAT_HASH_SIZE] __read_mostly; - -static void reset_entries(void) -{ - nr_entries = 0; - memset(entries, 0, sizeof(entries)); - memset(tstat_hash_table, 0, sizeof(tstat_hash_table)); - atomic_set(&overflow_count, 0); -} - -static struct entry *alloc_entry(void) -{ - if (nr_entries >= MAX_ENTRIES) - return NULL; - - return entries + nr_entries++; -} - -static int match_entries(struct entry *entry1, struct entry *entry2) -{ - return entry1->timer == entry2->timer && - entry1->start_func == entry2->start_func && - entry1->expire_func == entry2->expire_func && - entry1->pid == entry2->pid; -} - -/* - * Look up whether an entry matching this item is present - * in the hash already. Must be called with irqs off and the - * lookup lock held: - */ -static struct entry *tstat_lookup(struct entry *entry, char *comm) -{ - struct entry **head, *curr, *prev; - - head = tstat_hashentry(entry); - curr = *head; - - /* - * The fastpath is when the entry is already hashed, - * we do this with the lookup lock held, but with the - * table lock not held: - */ - while (curr) { - if (match_entries(curr, entry)) - return curr; - - curr = curr->next; - } - /* - * Slowpath: allocate, set up and link a new hash entry: - */ - prev = NULL; - curr = *head; - - raw_spin_lock(&table_lock); - /* - * Make sure we have not raced with another CPU: - */ - while (curr) { - if (match_entries(curr, entry)) - goto out_unlock; - - prev = curr; - curr = curr->next; - } - - curr = alloc_entry(); - if (curr) { - *curr = *entry; - curr->count = 0; - curr->next = NULL; - memcpy(curr->comm, comm, TASK_COMM_LEN); - - smp_mb(); /* Ensure that curr is initialized before insert */ - - if (prev) - prev->next = curr; - else - *head = curr; - } - out_unlock: - raw_spin_unlock(&table_lock); - - return curr; -} - -/** - * timer_stats_update_stats - Update the statistics for a timer. - * @timer: pointer to either a timer_list or a hrtimer - * @pid: the pid of the task which set up the timer - * @startf: pointer to the function which did the timer setup - * @timerf: pointer to the timer callback function of the timer - * @comm: name of the process which set up the timer - * @tflags: The flags field of the timer - * - * When the timer is already registered, then the event counter is - * incremented. Otherwise the timer is registered in a free slot. - */ -void timer_stats_update_stats(void *timer, pid_t pid, void *startf, - void *timerf, char *comm, u32 tflags) -{ - /* - * It doesn't matter which lock we take: - */ - raw_spinlock_t *lock; - struct entry *entry, input; - unsigned long flags; - - if (likely(!timer_stats_active)) - return; - - lock = &per_cpu(tstats_lookup_lock, raw_smp_processor_id()); - - input.timer = timer; - input.start_func = startf; - input.expire_func = timerf; - input.pid = pid; - input.flags = tflags; - - raw_spin_lock_irqsave(lock, flags); - if (!timer_stats_active) - goto out_unlock; - - entry = tstat_lookup(&input, comm); - if (likely(entry)) - entry->count++; - else - atomic_inc(&overflow_count); - - out_unlock: - raw_spin_unlock_irqrestore(lock, flags); -} - -static void print_name_offset(struct seq_file *m, unsigned long addr) -{ - char symname[KSYM_NAME_LEN]; - - if (lookup_symbol_name(addr, symname) < 0) - seq_printf(m, "<%p>", (void *)addr); - else - seq_printf(m, "%s", symname); -} - -static int tstats_show(struct seq_file *m, void *v) -{ - struct timespec64 period; - struct entry *entry; - unsigned long ms; - long events = 0; - ktime_t time; - int i; - - mutex_lock(&show_mutex); - /* - * If still active then calculate up to now: - */ - if (timer_stats_active) - time_stop = ktime_get(); - - time = ktime_sub(time_stop, time_start); - - period = ktime_to_timespec64(time); - ms = period.tv_nsec / 1000000; - - seq_puts(m, "Timer Stats Version: v0.3\n"); - seq_printf(m, "Sample period: %ld.%03ld s\n", (long)period.tv_sec, ms); - if (atomic_read(&overflow_count)) - seq_printf(m, "Overflow: %d entries\n", atomic_read(&overflow_count)); - seq_printf(m, "Collection: %s\n", timer_stats_active ? "active" : "inactive"); - - for (i = 0; i < nr_entries; i++) { - entry = entries + i; - if (entry->flags & TIMER_DEFERRABLE) { - seq_printf(m, "%4luD, %5d %-16s ", - entry->count, entry->pid, entry->comm); - } else { - seq_printf(m, " %4lu, %5d %-16s ", - entry->count, entry->pid, entry->comm); - } - - print_name_offset(m, (unsigned long)entry->start_func); - seq_puts(m, " ("); - print_name_offset(m, (unsigned long)entry->expire_func); - seq_puts(m, ")\n"); - - events += entry->count; - } - - ms += period.tv_sec * 1000; - if (!ms) - ms = 1; - - if (events && period.tv_sec) - seq_printf(m, "%ld total events, %ld.%03ld events/sec\n", - events, events * 1000 / ms, - (events * 1000000 / ms) % 1000); - else - seq_printf(m, "%ld total events\n", events); - - mutex_unlock(&show_mutex); - - return 0; -} - -/* - * After a state change, make sure all concurrent lookup/update - * activities have stopped: - */ -static void sync_access(void) -{ - unsigned long flags; - int cpu; - - for_each_online_cpu(cpu) { - raw_spinlock_t *lock = &per_cpu(tstats_lookup_lock, cpu); - - raw_spin_lock_irqsave(lock, flags); - /* nothing */ - raw_spin_unlock_irqrestore(lock, flags); - } -} - -static ssize_t tstats_write(struct file *file, const char __user *buf, - size_t count, loff_t *offs) -{ - char ctl[2]; - - if (count != 2 || *offs) - return -EINVAL; - - if (copy_from_user(ctl, buf, count)) - return -EFAULT; - - mutex_lock(&show_mutex); - switch (ctl[0]) { - case '0': - if (timer_stats_active) { - timer_stats_active = 0; - time_stop = ktime_get(); - sync_access(); - } - break; - case '1': - if (!timer_stats_active) { - reset_entries(); - time_start = ktime_get(); - smp_mb(); - timer_stats_active = 1; - } - break; - default: - count = -EINVAL; - } - mutex_unlock(&show_mutex); - - return count; -} - -static int tstats_open(struct inode *inode, struct file *filp) -{ - return single_open(filp, tstats_show, NULL); -} - -static const struct file_operations tstats_fops = { - .open = tstats_open, - .read = seq_read, - .write = tstats_write, - .llseek = seq_lseek, - .release = single_release, -}; - -void __init init_timer_stats(void) -{ - int cpu; - - for_each_possible_cpu(cpu) - raw_spin_lock_init(&per_cpu(tstats_lookup_lock, cpu)); -} - -static int __init init_tstats_procfs(void) -{ - struct proc_dir_entry *pe; - - pe = proc_create("timer_stats", 0644, NULL, &tstats_fops); - if (!pe) - return -ENOMEM; - return 0; -} -__initcall(init_tstats_procfs); diff --git a/kernel/torture.c b/kernel/torture.c index 0d887eb62856..55de96529287 100644 --- a/kernel/torture.c +++ b/kernel/torture.c @@ -30,6 +30,7 @@ #include <linux/smp.h> #include <linux/interrupt.h> #include <linux/sched.h> +#include <linux/sched/clock.h> #include <linux/atomic.h> #include <linux/bitops.h> #include <linux/completion.h> @@ -311,7 +312,7 @@ EXPORT_SYMBOL_GPL(torture_random); /* * Variables for shuffling. The idea is to ensure that each CPU stays * idle for an extended period to test interactions with dyntick idle, - * as well as interactions with any per-CPU varibles. + * as well as interactions with any per-CPU variables. */ struct shuffle_task { struct list_head st_l; diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index d5038005eb5d..434c840e2d82 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -134,7 +134,8 @@ config FUNCTION_TRACER select KALLSYMS select GENERIC_TRACER select CONTEXT_SWITCH_TRACER - select GLOB + select GLOB + select TASKS_RCU if PREEMPT help Enable the kernel to trace every kernel function. This is done by using a compiler feature to insert a small, 5-byte No-Operation @@ -429,7 +430,7 @@ config BLK_DEV_IO_TRACE If unsure, say N. -config KPROBE_EVENT +config KPROBE_EVENTS depends on KPROBES depends on HAVE_REGS_AND_STACK_ACCESS_API bool "Enable kprobes-based dynamic events" @@ -447,7 +448,7 @@ config KPROBE_EVENT This option is also required by perf-probe subcommand of perf tools. If you want to use perf tools, this option is strongly recommended. -config UPROBE_EVENT +config UPROBE_EVENTS bool "Enable uprobes-based dynamic events" depends on ARCH_SUPPORTS_UPROBES depends on MMU @@ -455,7 +456,7 @@ config UPROBE_EVENT select UPROBES select PROBE_EVENTS select TRACING - default n + default y help This allows the user to add tracing events on top of userspace dynamic events (similar to tracepoints) on the fly via the trace @@ -466,7 +467,7 @@ config UPROBE_EVENT config BPF_EVENTS depends on BPF_SYSCALL - depends on (KPROBE_EVENT || UPROBE_EVENT) && PERF_EVENTS + depends on (KPROBE_EVENTS || UPROBE_EVENTS) && PERF_EVENTS bool default y help @@ -666,30 +667,30 @@ config RING_BUFFER_STARTUP_TEST If unsure, say N -config TRACE_ENUM_MAP_FILE - bool "Show enum mappings for trace events" +config TRACE_EVAL_MAP_FILE + bool "Show eval mappings for trace events" depends on TRACING help - The "print fmt" of the trace events will show the enum names instead - of their values. This can cause problems for user space tools that - use this string to parse the raw data as user space does not know + The "print fmt" of the trace events will show the enum/sizeof names + instead of their values. This can cause problems for user space tools + that use this string to parse the raw data as user space does not know how to convert the string to its value. To fix this, there's a special macro in the kernel that can be used - to convert the enum into its value. If this macro is used, then the - print fmt strings will have the enums converted to their values. + to convert an enum/sizeof into its value. If this macro is used, then + the print fmt strings will be converted to their values. If something does not get converted properly, this option can be - used to show what enums the kernel tried to convert. + used to show what enums/sizeof the kernel tried to convert. - This option is for debugging the enum conversions. A file is created - in the tracing directory called "enum_map" that will show the enum + This option is for debugging the conversions. A file is created + in the tracing directory called "eval_map" that will show the names matched with their values and what trace event system they belong too. Normally, the mapping of the strings to values will be freed after boot up or module load. With this option, they will not be freed, as - they are needed for the "enum_map" file. Enabling this option will + they are needed for the "eval_map" file. Enabling this option will increase the memory footprint of the running kernel. If unsure, say N diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index e57980845549..90f2701d92a7 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -57,7 +57,7 @@ obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_EVENT_TRACING) += trace_events_trigger.o obj-$(CONFIG_HIST_TRIGGERS) += trace_events_hist.o obj-$(CONFIG_BPF_EVENTS) += bpf_trace.o -obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o +obj-$(CONFIG_KPROBE_EVENTS) += trace_kprobe.o obj-$(CONFIG_TRACEPOINTS) += power-traces.o ifeq ($(CONFIG_PM),y) obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o @@ -66,7 +66,7 @@ ifeq ($(CONFIG_TRACING),y) obj-$(CONFIG_KGDB_KDB) += trace_kdb.o endif obj-$(CONFIG_PROBE_EVENTS) += trace_probe.o -obj-$(CONFIG_UPROBE_EVENT) += trace_uprobe.o +obj-$(CONFIG_UPROBE_EVENTS) += trace_uprobe.o obj-$(CONFIG_TRACEPOINT_BENCHMARK) += trace_benchmark.o diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 95cecbf67f5c..bc364f86100a 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -28,6 +28,8 @@ #include <linux/uaccess.h> #include <linux/list.h> +#include "../../block/blk.h" + #include <trace/events/block.h> #include "trace_output.h" @@ -292,9 +294,6 @@ record_it: local_irq_restore(flags); } -static struct dentry *blk_tree_root; -static DEFINE_MUTEX(blk_tree_mutex); - static void blk_trace_free(struct blk_trace *bt) { debugfs_remove(bt->msg_file); @@ -433,9 +432,9 @@ static void blk_trace_setup_lba(struct blk_trace *bt, /* * Setup everything required to start tracing */ -int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, - struct block_device *bdev, - struct blk_user_trace_setup *buts) +static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, + struct block_device *bdev, + struct blk_user_trace_setup *buts) { struct blk_trace *bt = NULL; struct dentry *dir = NULL; @@ -468,22 +467,15 @@ int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, ret = -ENOENT; - mutex_lock(&blk_tree_mutex); - if (!blk_tree_root) { - blk_tree_root = debugfs_create_dir("block", NULL); - if (!blk_tree_root) { - mutex_unlock(&blk_tree_mutex); - goto err; - } - } - mutex_unlock(&blk_tree_mutex); - - dir = debugfs_create_dir(buts->name, blk_tree_root); + if (!blk_debugfs_root) + goto err; + dir = debugfs_lookup(buts->name, blk_debugfs_root); + if (!dir) + bt->dir = dir = debugfs_create_dir(buts->name, blk_debugfs_root); if (!dir) goto err; - bt->dir = dir; bt->dev = dev; atomic_set(&bt->dropped, 0); INIT_LIST_HEAD(&bt->running_list); @@ -525,9 +517,12 @@ int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, if (atomic_inc_return(&blk_probes_ref) == 1) blk_register_tracepoints(); - return 0; + ret = 0; err: - blk_trace_free(bt); + if (dir && !bt->dir) + dput(dir); + if (ret) + blk_trace_free(bt); return ret; } @@ -695,8 +690,8 @@ void blk_trace_shutdown(struct request_queue *q) /** * blk_add_trace_rq - Add a trace for a request oriented action - * @q: queue the io is for * @rq: the source request + * @error: return status to log * @nr_bytes: number of completed bytes * @what: the action * @@ -704,56 +699,46 @@ void blk_trace_shutdown(struct request_queue *q) * Records an action against a request. Will log the bio offset + size. * **/ -static void blk_add_trace_rq(struct request_queue *q, struct request *rq, +static void blk_add_trace_rq(struct request *rq, int error, unsigned int nr_bytes, u32 what) { - struct blk_trace *bt = q->blk_trace; + struct blk_trace *bt = rq->q->blk_trace; if (likely(!bt)) return; - if (rq->cmd_type == REQ_TYPE_BLOCK_PC) { + if (blk_rq_is_passthrough(rq)) what |= BLK_TC_ACT(BLK_TC_PC); - __blk_add_trace(bt, 0, nr_bytes, req_op(rq), rq->cmd_flags, - what, rq->errors, rq->cmd_len, rq->cmd); - } else { + else what |= BLK_TC_ACT(BLK_TC_FS); - __blk_add_trace(bt, blk_rq_pos(rq), nr_bytes, req_op(rq), - rq->cmd_flags, what, rq->errors, 0, NULL); - } -} -static void blk_add_trace_rq_abort(void *ignore, - struct request_queue *q, struct request *rq) -{ - blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_ABORT); + __blk_add_trace(bt, blk_rq_trace_sector(rq), nr_bytes, req_op(rq), + rq->cmd_flags, what, error, 0, NULL); } static void blk_add_trace_rq_insert(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_INSERT); + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_INSERT); } static void blk_add_trace_rq_issue(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_ISSUE); + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_ISSUE); } static void blk_add_trace_rq_requeue(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_REQUEUE); + blk_add_trace_rq(rq, 0, blk_rq_bytes(rq), BLK_TA_REQUEUE); } -static void blk_add_trace_rq_complete(void *ignore, - struct request_queue *q, - struct request *rq, - unsigned int nr_bytes) +static void blk_add_trace_rq_complete(void *ignore, struct request *rq, + int error, unsigned int nr_bytes) { - blk_add_trace_rq(q, rq, nr_bytes, BLK_TA_COMPLETE); + blk_add_trace_rq(rq, error, nr_bytes, BLK_TA_COMPLETE); } /** @@ -882,7 +867,7 @@ static void blk_add_trace_split(void *ignore, __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, bio_op(bio), bio->bi_opf, - BLK_TA_SPLIT, bio->bi_error, sizeof(rpdu), + BLK_TA_SPLIT, bio->bi_status, sizeof(rpdu), &rpdu); } } @@ -915,7 +900,7 @@ static void blk_add_trace_bio_remap(void *ignore, r.sector_from = cpu_to_be64(from); __blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size, - bio_op(bio), bio->bi_opf, BLK_TA_REMAP, bio->bi_error, + bio_op(bio), bio->bi_opf, BLK_TA_REMAP, bio->bi_status, sizeof(r), &r); } @@ -948,7 +933,7 @@ static void blk_add_trace_rq_remap(void *ignore, r.sector_from = cpu_to_be64(from); __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), - rq_data_dir(rq), 0, BLK_TA_REMAP, !!rq->errors, + rq_data_dir(rq), 0, BLK_TA_REMAP, 0, sizeof(r), &r); } @@ -972,12 +957,8 @@ void blk_add_driver_data(struct request_queue *q, if (likely(!bt)) return; - if (rq->cmd_type == REQ_TYPE_BLOCK_PC) - __blk_add_trace(bt, 0, blk_rq_bytes(rq), 0, 0, - BLK_TA_DRV_DATA, rq->errors, len, data); - else - __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), 0, 0, - BLK_TA_DRV_DATA, rq->errors, len, data); + __blk_add_trace(bt, blk_rq_trace_sector(rq), blk_rq_bytes(rq), 0, 0, + BLK_TA_DRV_DATA, 0, len, data); } EXPORT_SYMBOL_GPL(blk_add_driver_data); @@ -985,8 +966,6 @@ static void blk_register_tracepoints(void) { int ret; - ret = register_trace_block_rq_abort(blk_add_trace_rq_abort, NULL); - WARN_ON(ret); ret = register_trace_block_rq_insert(blk_add_trace_rq_insert, NULL); WARN_ON(ret); ret = register_trace_block_rq_issue(blk_add_trace_rq_issue, NULL); @@ -1039,7 +1018,6 @@ static void blk_unregister_tracepoints(void) unregister_trace_block_rq_requeue(blk_add_trace_rq_requeue, NULL); unregister_trace_block_rq_issue(blk_add_trace_rq_issue, NULL); unregister_trace_block_rq_insert(blk_add_trace_rq_insert, NULL); - unregister_trace_block_rq_abort(blk_add_trace_rq_abort, NULL); tracepoint_synchronize_unregister(); } @@ -1684,14 +1662,14 @@ static ssize_t sysfs_blk_trace_attr_store(struct device *dev, goto out; if (attr == &dev_attr_act_mask) { - if (sscanf(buf, "%llx", &value) != 1) { + if (kstrtoull(buf, 0, &value)) { /* Assume it is a list of trace category names */ ret = blk_trace_str2mask(buf); if (ret < 0) goto out; value = ret; } - } else if (sscanf(buf, "%llu", &value) != 1) + } else if (kstrtoull(buf, 0, &value)) goto out; ret = -ENXIO; @@ -1752,31 +1730,6 @@ void blk_trace_remove_sysfs(struct device *dev) #ifdef CONFIG_EVENT_TRACING -void blk_dump_cmd(char *buf, struct request *rq) -{ - int i, end; - int len = rq->cmd_len; - unsigned char *cmd = rq->cmd; - - if (rq->cmd_type != REQ_TYPE_BLOCK_PC) { - buf[0] = '\0'; - return; - } - - for (end = len - 1; end >= 0; end--) - if (cmd[end]) - break; - end++; - - for (i = 0; i < len; i++) { - buf += sprintf(buf, "%s%02x", i == 0 ? "" : " ", cmd[i]); - if (i == end && end != len - 1) { - sprintf(buf, " .."); - break; - } - } -} - void blk_fill_rwbs(char *rwbs, unsigned int op, int bytes) { int i = 0; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index fa77311dadb2..37385193a608 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -76,8 +76,8 @@ static const struct bpf_func_proto bpf_probe_read_proto = { .func = bpf_probe_read, .gpl_only = true, .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_RAW_STACK, - .arg2_type = ARG_CONST_STACK_SIZE, + .arg1_type = ARG_PTR_TO_UNINIT_MEM, + .arg2_type = ARG_CONST_SIZE, .arg3_type = ARG_ANYTHING, }; @@ -96,7 +96,7 @@ BPF_CALL_3(bpf_probe_write_user, void *, unsafe_ptr, const void *, src, if (unlikely(in_interrupt() || current->flags & (PF_KTHREAD | PF_EXITING))) return -EPERM; - if (unlikely(segment_eq(get_fs(), KERNEL_DS))) + if (unlikely(uaccess_kernel())) return -EPERM; if (!access_ok(VERIFY_WRITE, unsafe_ptr, size)) return -EPERM; @@ -109,8 +109,8 @@ static const struct bpf_func_proto bpf_probe_write_user_proto = { .gpl_only = true, .ret_type = RET_INTEGER, .arg1_type = ARG_ANYTHING, - .arg2_type = ARG_PTR_TO_STACK, - .arg3_type = ARG_CONST_STACK_SIZE, + .arg2_type = ARG_PTR_TO_MEM, + .arg3_type = ARG_CONST_SIZE, }; static const struct bpf_func_proto *bpf_get_probe_write_proto(void) @@ -122,8 +122,8 @@ static const struct bpf_func_proto *bpf_get_probe_write_proto(void) } /* - * limited trace_printk() - * only %d %u %x %ld %lu %lx %lld %llu %llx %p %s conversion specifiers allowed + * Only limited trace_printk() conversion specifiers allowed: + * %d %i %u %x %ld %li %lu %lx %lld %lli %llu %llx %p %s */ BPF_CALL_5(bpf_trace_printk, char *, fmt, u32, fmt_size, u64, arg1, u64, arg2, u64, arg3) @@ -198,7 +198,8 @@ BPF_CALL_5(bpf_trace_printk, char *, fmt, u32, fmt_size, u64, arg1, i++; } - if (fmt[i] != 'd' && fmt[i] != 'u' && fmt[i] != 'x') + if (fmt[i] != 'i' && fmt[i] != 'd' && + fmt[i] != 'u' && fmt[i] != 'x') return -EINVAL; fmt_cnt++; } @@ -213,8 +214,8 @@ static const struct bpf_func_proto bpf_trace_printk_proto = { .func = bpf_trace_printk, .gpl_only = true, .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_STACK, - .arg2_type = ARG_CONST_STACK_SIZE, + .arg1_type = ARG_PTR_TO_MEM, + .arg2_type = ARG_CONST_SIZE, }; const struct bpf_func_proto *bpf_get_trace_printk_proto(void) @@ -234,7 +235,8 @@ BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) unsigned int cpu = smp_processor_id(); u64 index = flags & BPF_F_INDEX_MASK; struct bpf_event_entry *ee; - struct perf_event *event; + u64 value = 0; + int err; if (unlikely(flags & ~(BPF_F_INDEX_MASK))) return -EINVAL; @@ -247,21 +249,14 @@ BPF_CALL_2(bpf_perf_event_read, struct bpf_map *, map, u64, flags) if (!ee) return -ENOENT; - event = ee->event; - if (unlikely(event->attr.type != PERF_TYPE_HARDWARE && - event->attr.type != PERF_TYPE_RAW)) - return -EINVAL; - - /* make sure event is local and doesn't have pmu::count */ - if (unlikely(event->oncpu != cpu || event->pmu->count)) - return -EINVAL; - + err = perf_event_read_local(ee->event, &value); /* - * we don't know if the function is run successfully by the - * return value. It can be judged in other places, such as - * eBPF programs. + * this api is ugly since we miss [-22..-2] range of valid + * counter values, but that's uapi */ - return perf_event_read_local(event); + if (err) + return err; + return value; } static const struct bpf_func_proto bpf_perf_event_read_proto = { @@ -272,14 +267,16 @@ static const struct bpf_func_proto bpf_perf_event_read_proto = { .arg2_type = ARG_ANYTHING, }; +static DEFINE_PER_CPU(struct perf_sample_data, bpf_sd); + static __always_inline u64 __bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map, u64 flags, struct perf_raw_record *raw) { struct bpf_array *array = container_of(map, struct bpf_array, map); + struct perf_sample_data *sd = this_cpu_ptr(&bpf_sd); unsigned int cpu = smp_processor_id(); u64 index = flags & BPF_F_INDEX_MASK; - struct perf_sample_data sample_data; struct bpf_event_entry *ee; struct perf_event *event; @@ -300,9 +297,9 @@ __bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map, if (unlikely(event->oncpu != cpu)) return -EOPNOTSUPP; - perf_sample_data_init(&sample_data, 0, 0); - sample_data.raw = raw; - perf_event_output(event, &sample_data, regs); + perf_sample_data_init(sd, 0, 0); + sd->raw = raw; + perf_event_output(event, sd, regs); return 0; } @@ -329,8 +326,8 @@ static const struct bpf_func_proto bpf_perf_event_output_proto = { .arg1_type = ARG_PTR_TO_CTX, .arg2_type = ARG_CONST_MAP_PTR, .arg3_type = ARG_ANYTHING, - .arg4_type = ARG_PTR_TO_STACK, - .arg5_type = ARG_CONST_STACK_SIZE, + .arg4_type = ARG_PTR_TO_MEM, + .arg5_type = ARG_CONST_SIZE, }; static DEFINE_PER_CPU(struct pt_regs, bpf_pt_regs); @@ -395,6 +392,36 @@ static const struct bpf_func_proto bpf_current_task_under_cgroup_proto = { .arg2_type = ARG_ANYTHING, }; +BPF_CALL_3(bpf_probe_read_str, void *, dst, u32, size, + const void *, unsafe_ptr) +{ + int ret; + + /* + * The strncpy_from_unsafe() call will likely not fill the entire + * buffer, but that's okay in this circumstance as we're probing + * arbitrary memory anyway similar to bpf_probe_read() and might + * as well probe the stack. Thus, memory is explicitly cleared + * only in error case, so that improper users ignoring return + * code altogether don't copy garbage; otherwise length of string + * is returned that can be used for bpf_perf_event_output() et al. + */ + ret = strncpy_from_unsafe(dst, unsafe_ptr, size); + if (unlikely(ret < 0)) + memset(dst, 0, size); + + return ret; +} + +static const struct bpf_func_proto bpf_probe_read_str_proto = { + .func = bpf_probe_read_str, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_UNINIT_MEM, + .arg2_type = ARG_CONST_SIZE, + .arg3_type = ARG_ANYTHING, +}; + static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) { switch (func_id) { @@ -432,6 +459,8 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) return &bpf_current_task_under_cgroup_proto; case BPF_FUNC_get_prandom_u32: return &bpf_get_prandom_u32_proto; + case BPF_FUNC_probe_read_str: + return &bpf_probe_read_str_proto; default: return NULL; } @@ -451,7 +480,7 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func /* bpf+kprobe programs can access fields of 'struct pt_regs' */ static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type type, - enum bpf_reg_type *reg_type) + struct bpf_insn_access_aux *info) { if (off < 0 || off >= sizeof(struct pt_regs)) return false; @@ -459,19 +488,21 @@ static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type return false; if (off % size != 0) return false; + /* + * Assertion for 32 bit to make sure last 8 byte access + * (BPF_DW) to the last 4 byte member is disallowed. + */ + if (off + size > sizeof(struct pt_regs)) + return false; + return true; } -static const struct bpf_verifier_ops kprobe_prog_ops = { +const struct bpf_verifier_ops kprobe_prog_ops = { .get_func_proto = kprobe_prog_func_proto, .is_valid_access = kprobe_prog_is_valid_access, }; -static struct bpf_prog_type_list kprobe_tl = { - .ops = &kprobe_prog_ops, - .type = BPF_PROG_TYPE_KPROBE, -}; - BPF_CALL_5(bpf_perf_event_output_tp, void *, tp_buff, struct bpf_map *, map, u64, flags, void *, data, u64, size) { @@ -492,8 +523,8 @@ static const struct bpf_func_proto bpf_perf_event_output_proto_tp = { .arg1_type = ARG_PTR_TO_CTX, .arg2_type = ARG_CONST_MAP_PTR, .arg3_type = ARG_ANYTHING, - .arg4_type = ARG_PTR_TO_STACK, - .arg5_type = ARG_CONST_STACK_SIZE, + .arg4_type = ARG_PTR_TO_MEM, + .arg5_type = ARG_CONST_SIZE, }; BPF_CALL_3(bpf_get_stackid_tp, void *, tp_buff, struct bpf_map *, map, @@ -532,7 +563,7 @@ static const struct bpf_func_proto *tp_prog_func_proto(enum bpf_func_id func_id) } static bool tp_prog_is_valid_access(int off, int size, enum bpf_access_type type, - enum bpf_reg_type *reg_type) + struct bpf_insn_access_aux *info) { if (off < sizeof(void *) || off >= PERF_MAX_TRACE_SIZE) return false; @@ -540,82 +571,73 @@ static bool tp_prog_is_valid_access(int off, int size, enum bpf_access_type type return false; if (off % size != 0) return false; + + BUILD_BUG_ON(PERF_MAX_TRACE_SIZE % sizeof(__u64)); return true; } -static const struct bpf_verifier_ops tracepoint_prog_ops = { +const struct bpf_verifier_ops tracepoint_prog_ops = { .get_func_proto = tp_prog_func_proto, .is_valid_access = tp_prog_is_valid_access, }; -static struct bpf_prog_type_list tracepoint_tl = { - .ops = &tracepoint_prog_ops, - .type = BPF_PROG_TYPE_TRACEPOINT, -}; - static bool pe_prog_is_valid_access(int off, int size, enum bpf_access_type type, - enum bpf_reg_type *reg_type) + struct bpf_insn_access_aux *info) { + const int size_sp = FIELD_SIZEOF(struct bpf_perf_event_data, + sample_period); + if (off < 0 || off >= sizeof(struct bpf_perf_event_data)) return false; if (type != BPF_READ) return false; if (off % size != 0) return false; - if (off == offsetof(struct bpf_perf_event_data, sample_period)) { - if (size != sizeof(u64)) + + switch (off) { + case bpf_ctx_range(struct bpf_perf_event_data, sample_period): + bpf_ctx_record_field_size(info, size_sp); + if (!bpf_ctx_narrow_access_ok(off, size, size_sp)) return false; - } else { + break; + default: if (size != sizeof(long)) return false; } + return true; } -static u32 pe_prog_convert_ctx_access(enum bpf_access_type type, int dst_reg, - int src_reg, int ctx_off, +static u32 pe_prog_convert_ctx_access(enum bpf_access_type type, + const struct bpf_insn *si, struct bpf_insn *insn_buf, - struct bpf_prog *prog) + struct bpf_prog *prog, u32 *target_size) { struct bpf_insn *insn = insn_buf; - switch (ctx_off) { + switch (si->off) { case offsetof(struct bpf_perf_event_data, sample_period): - BUILD_BUG_ON(FIELD_SIZEOF(struct perf_sample_data, period) != sizeof(u64)); - *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_perf_event_data_kern, - data), dst_reg, src_reg, + data), si->dst_reg, si->src_reg, offsetof(struct bpf_perf_event_data_kern, data)); - *insn++ = BPF_LDX_MEM(BPF_DW, dst_reg, dst_reg, - offsetof(struct perf_sample_data, period)); + *insn++ = BPF_LDX_MEM(BPF_DW, si->dst_reg, si->dst_reg, + bpf_target_off(struct perf_sample_data, period, 8, + target_size)); break; default: *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_perf_event_data_kern, - regs), dst_reg, src_reg, + regs), si->dst_reg, si->src_reg, offsetof(struct bpf_perf_event_data_kern, regs)); - *insn++ = BPF_LDX_MEM(BPF_SIZEOF(long), dst_reg, dst_reg, ctx_off); + *insn++ = BPF_LDX_MEM(BPF_SIZEOF(long), si->dst_reg, si->dst_reg, + si->off); break; } return insn - insn_buf; } -static const struct bpf_verifier_ops perf_event_prog_ops = { +const struct bpf_verifier_ops perf_event_prog_ops = { .get_func_proto = tp_prog_func_proto, .is_valid_access = pe_prog_is_valid_access, .convert_ctx_access = pe_prog_convert_ctx_access, }; - -static struct bpf_prog_type_list perf_event_tl = { - .ops = &perf_event_prog_ops, - .type = BPF_PROG_TYPE_PERF_EVENT, -}; - -static int __init register_kprobe_prog_ops(void) -{ - bpf_register_prog_type(&kprobe_tl); - bpf_register_prog_type(&tracepoint_tl); - bpf_register_prog_type(&perf_event_tl); - return 0; -} -late_initcall(register_kprobe_prog_ops); diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index eb230f06ba41..02004ae91860 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -15,6 +15,7 @@ #include <linux/stop_machine.h> #include <linux/clocksource.h> +#include <linux/sched/task.h> #include <linux/kallsyms.h> #include <linux/seq_file.h> #include <linux/suspend.h> @@ -35,6 +36,7 @@ #include <trace/events/sched.h> +#include <asm/sections.h> #include <asm/setup.h> #include "trace_output.h" @@ -111,7 +113,7 @@ static int ftrace_disabled __read_mostly; static DEFINE_MUTEX(ftrace_lock); -static struct ftrace_ops *ftrace_ops_list __read_mostly = &ftrace_list_end; +static struct ftrace_ops __rcu *ftrace_ops_list __read_mostly = &ftrace_list_end; ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; static struct ftrace_ops global_ops; @@ -167,8 +169,11 @@ int ftrace_nr_registered_ops(void) mutex_lock(&ftrace_lock); - for (ops = ftrace_ops_list; - ops != &ftrace_list_end; ops = ops->next) + for (ops = rcu_dereference_protected(ftrace_ops_list, + lockdep_is_held(&ftrace_lock)); + ops != &ftrace_list_end; + ops = rcu_dereference_protected(ops->next, + lockdep_is_held(&ftrace_lock))) cnt++; mutex_unlock(&ftrace_lock); @@ -273,10 +278,11 @@ static void update_ftrace_function(void) * If there's only one ftrace_ops registered, the ftrace_ops_list * will point to the ops we want. */ - set_function_trace_op = ftrace_ops_list; + set_function_trace_op = rcu_dereference_protected(ftrace_ops_list, + lockdep_is_held(&ftrace_lock)); /* If there's no ftrace_ops registered, just call the stub function */ - if (ftrace_ops_list == &ftrace_list_end) { + if (set_function_trace_op == &ftrace_list_end) { func = ftrace_stub; /* @@ -284,7 +290,8 @@ static void update_ftrace_function(void) * recursion safe and not dynamic and the arch supports passing ops, * then have the mcount trampoline call the function directly. */ - } else if (ftrace_ops_list->next == &ftrace_list_end) { + } else if (rcu_dereference_protected(ftrace_ops_list->next, + lockdep_is_held(&ftrace_lock)) == &ftrace_list_end) { func = ftrace_ops_get_list_func(ftrace_ops_list); } else { @@ -346,9 +353,11 @@ int using_ftrace_ops_list_func(void) return ftrace_trace_function == ftrace_ops_list_func; } -static void add_ftrace_ops(struct ftrace_ops **list, struct ftrace_ops *ops) +static void add_ftrace_ops(struct ftrace_ops __rcu **list, + struct ftrace_ops *ops) { - ops->next = *list; + rcu_assign_pointer(ops->next, *list); + /* * We are entering ops into the list but another * CPU might be walking that list. We need to make sure @@ -358,7 +367,8 @@ static void add_ftrace_ops(struct ftrace_ops **list, struct ftrace_ops *ops) rcu_assign_pointer(*list, ops); } -static int remove_ftrace_ops(struct ftrace_ops **list, struct ftrace_ops *ops) +static int remove_ftrace_ops(struct ftrace_ops __rcu **list, + struct ftrace_ops *ops) { struct ftrace_ops **p; @@ -366,7 +376,10 @@ static int remove_ftrace_ops(struct ftrace_ops **list, struct ftrace_ops *ops) * If we are removing the last function, then simply point * to the ftrace_stub. */ - if (*list == ops && ops->next == &ftrace_list_end) { + if (rcu_dereference_protected(*list, + lockdep_is_held(&ftrace_lock)) == ops && + rcu_dereference_protected(ops->next, + lockdep_is_held(&ftrace_lock)) == &ftrace_list_end) { *list = &ftrace_list_end; return 0; } @@ -1094,27 +1107,18 @@ static bool update_all_ops; # error Dynamic ftrace depends on MCOUNT_RECORD #endif -static struct hlist_head ftrace_func_hash[FTRACE_FUNC_HASHSIZE] __read_mostly; - -struct ftrace_func_probe { - struct hlist_node node; - struct ftrace_probe_ops *ops; - unsigned long flags; - unsigned long ip; - void *data; - struct list_head free_list; -}; - struct ftrace_func_entry { struct hlist_node hlist; unsigned long ip; }; -struct ftrace_hash { - unsigned long size_bits; - struct hlist_head *buckets; - unsigned long count; - struct rcu_head rcu; +struct ftrace_func_probe { + struct ftrace_probe_ops *probe_ops; + struct ftrace_ops ops; + struct trace_array *tr; + struct list_head list; + void *data; + int ref; }; /* @@ -1192,26 +1196,24 @@ struct ftrace_page { static struct ftrace_page *ftrace_pages_start; static struct ftrace_page *ftrace_pages; -static bool __always_inline ftrace_hash_empty(struct ftrace_hash *hash) +static __always_inline unsigned long +ftrace_hash_key(struct ftrace_hash *hash, unsigned long ip) { - return !hash || !hash->count; + if (hash->size_bits > 0) + return hash_long(ip, hash->size_bits); + + return 0; } -static struct ftrace_func_entry * -ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) +/* Only use this function if ftrace_hash_empty() has already been tested */ +static __always_inline struct ftrace_func_entry * +__ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) { unsigned long key; struct ftrace_func_entry *entry; struct hlist_head *hhd; - if (ftrace_hash_empty(hash)) - return NULL; - - if (hash->size_bits > 0) - key = hash_long(ip, hash->size_bits); - else - key = 0; - + key = ftrace_hash_key(hash, ip); hhd = &hash->buckets[key]; hlist_for_each_entry_rcu_notrace(entry, hhd, hlist) { @@ -1221,17 +1223,32 @@ ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) return NULL; } +/** + * ftrace_lookup_ip - Test to see if an ip exists in an ftrace_hash + * @hash: The hash to look at + * @ip: The instruction pointer to test + * + * Search a given @hash to see if a given instruction pointer (@ip) + * exists in it. + * + * Returns the entry that holds the @ip if found. NULL otherwise. + */ +struct ftrace_func_entry * +ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip) +{ + if (ftrace_hash_empty(hash)) + return NULL; + + return __ftrace_lookup_ip(hash, ip); +} + static void __add_hash_entry(struct ftrace_hash *hash, struct ftrace_func_entry *entry) { struct hlist_head *hhd; unsigned long key; - if (hash->size_bits) - key = hash_long(entry->ip, hash->size_bits); - else - key = 0; - + key = ftrace_hash_key(hash, entry->ip); hhd = &hash->buckets[key]; hlist_add_head(&entry->hlist, hhd); hash->count++; @@ -1264,7 +1281,7 @@ static void remove_hash_entry(struct ftrace_hash *hash, struct ftrace_func_entry *entry) { - hlist_del(&entry->hlist); + hlist_del_rcu(&entry->hlist); hash->count--; } @@ -1287,6 +1304,28 @@ static void ftrace_hash_clear(struct ftrace_hash *hash) FTRACE_WARN_ON(hash->count); } +static void free_ftrace_mod(struct ftrace_mod_load *ftrace_mod) +{ + list_del(&ftrace_mod->list); + kfree(ftrace_mod->module); + kfree(ftrace_mod->func); + kfree(ftrace_mod); +} + +static void clear_ftrace_mod_list(struct list_head *head) +{ + struct ftrace_mod_load *p, *n; + + /* stack tracer isn't supported yet */ + if (!head) + return; + + mutex_lock(&ftrace_lock); + list_for_each_entry_safe(p, n, head, list) + free_ftrace_mod(p); + mutex_unlock(&ftrace_lock); +} + static void free_ftrace_hash(struct ftrace_hash *hash) { if (!hash || hash == EMPTY_HASH) @@ -1340,6 +1379,35 @@ static struct ftrace_hash *alloc_ftrace_hash(int size_bits) return hash; } + +static int ftrace_add_mod(struct trace_array *tr, + const char *func, const char *module, + int enable) +{ + struct ftrace_mod_load *ftrace_mod; + struct list_head *mod_head = enable ? &tr->mod_trace : &tr->mod_notrace; + + ftrace_mod = kzalloc(sizeof(*ftrace_mod), GFP_KERNEL); + if (!ftrace_mod) + return -ENOMEM; + + ftrace_mod->func = kstrdup(func, GFP_KERNEL); + ftrace_mod->module = kstrdup(module, GFP_KERNEL); + ftrace_mod->enable = enable; + + if (!ftrace_mod->func || !ftrace_mod->module) + goto out_free; + + list_add(&ftrace_mod->list, mod_head); + + return 0; + + out_free: + free_ftrace_mod(ftrace_mod); + + return -ENOMEM; +} + static struct ftrace_hash * alloc_and_copy_ftrace_hash(int size_bits, struct ftrace_hash *hash) { @@ -1353,6 +1421,9 @@ alloc_and_copy_ftrace_hash(int size_bits, struct ftrace_hash *hash) if (!new_hash) return NULL; + if (hash) + new_hash->flags = hash->flags; + /* Empty hash? */ if (ftrace_hash_empty(hash)) return new_hash; @@ -1383,9 +1454,8 @@ ftrace_hash_rec_enable_modify(struct ftrace_ops *ops, int filter_hash); static int ftrace_hash_ipmodify_update(struct ftrace_ops *ops, struct ftrace_hash *new_hash); -static int -ftrace_hash_move(struct ftrace_ops *ops, int enable, - struct ftrace_hash **dst, struct ftrace_hash *src) +static struct ftrace_hash * +__ftrace_hash_move(struct ftrace_hash *src) { struct ftrace_func_entry *entry; struct hlist_node *tn; @@ -1393,21 +1463,13 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, struct ftrace_hash *new_hash; int size = src->count; int bits = 0; - int ret; int i; - /* Reject setting notrace hash on IPMODIFY ftrace_ops */ - if (ops->flags & FTRACE_OPS_FL_IPMODIFY && !enable) - return -EINVAL; - /* - * If the new source is empty, just free dst and assign it - * the empty_hash. + * If the new source is empty, just return the empty_hash. */ - if (!src->count) { - new_hash = EMPTY_HASH; - goto update; - } + if (ftrace_hash_empty(src)) + return EMPTY_HASH; /* * Make the hash size about 1/2 the # found @@ -1421,7 +1483,9 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, new_hash = alloc_ftrace_hash(bits); if (!new_hash) - return -ENOMEM; + return NULL; + + new_hash->flags = src->flags; size = 1 << src->size_bits; for (i = 0; i < size; i++) { @@ -1432,7 +1496,24 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, } } -update: + return new_hash; +} + +static int +ftrace_hash_move(struct ftrace_ops *ops, int enable, + struct ftrace_hash **dst, struct ftrace_hash *src) +{ + struct ftrace_hash *new_hash; + int ret; + + /* Reject setting notrace hash on IPMODIFY ftrace_ops */ + if (ops->flags & FTRACE_OPS_FL_IPMODIFY && !enable) + return -EINVAL; + + new_hash = __ftrace_hash_move(src); + if (!new_hash) + return -ENOMEM; + /* Make sure this can be applied if it is IPMODIFY ftrace_ops */ if (enable) { /* IPMODIFY should be updated only when filter_hash updating */ @@ -1466,9 +1547,9 @@ static bool hash_contains_ip(unsigned long ip, * notrace hash is considered not in the notrace hash. */ return (ftrace_hash_empty(hash->filter_hash) || - ftrace_lookup_ip(hash->filter_hash, ip)) && + __ftrace_lookup_ip(hash->filter_hash, ip)) && (ftrace_hash_empty(hash->notrace_hash) || - !ftrace_lookup_ip(hash->notrace_hash, ip)); + !__ftrace_lookup_ip(hash->notrace_hash, ip)); } /* @@ -1499,8 +1580,8 @@ ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) return 0; #endif - hash.filter_hash = rcu_dereference_raw_notrace(ops->func_hash->filter_hash); - hash.notrace_hash = rcu_dereference_raw_notrace(ops->func_hash->notrace_hash); + rcu_assign_pointer(hash.filter_hash, ops->func_hash->filter_hash); + rcu_assign_pointer(hash.notrace_hash, ops->func_hash->notrace_hash); if (hash_contains_ip(ip, &hash)) ret = 1; @@ -1636,7 +1717,7 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, struct dyn_ftrace *rec; bool update = false; int count = 0; - int all = 0; + int all = false; /* Only update if the ops has been registered */ if (!(ops->flags & FTRACE_OPS_FL_ENABLED)) @@ -1657,7 +1738,7 @@ static bool __ftrace_hash_rec_update(struct ftrace_ops *ops, hash = ops->func_hash->filter_hash; other_hash = ops->func_hash->notrace_hash; if (ftrace_hash_empty(hash)) - all = 1; + all = true; } else { inc = !inc; hash = ops->func_hash->notrace_hash; @@ -2770,7 +2851,8 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) * If there's no more ops registered with ftrace, run a * sanity check to make sure all rec flags are cleared. */ - if (ftrace_ops_list == &ftrace_list_end) { + if (rcu_dereference_protected(ftrace_ops_list, + lockdep_is_held(&ftrace_lock)) == &ftrace_list_end) { struct ftrace_page *pg; struct dyn_ftrace *rec; @@ -2792,18 +2874,28 @@ static int ftrace_shutdown(struct ftrace_ops *ops, int command) * callers are done before leaving this function. * The same goes for freeing the per_cpu data of the per_cpu * ops. - * - * Again, normal synchronize_sched() is not good enough. - * We need to do a hard force of sched synchronization. - * This is because we use preempt_disable() to do RCU, but - * the function tracers can be called where RCU is not watching - * (like before user_exit()). We can not rely on the RCU - * infrastructure to do the synchronization, thus we must do it - * ourselves. */ if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_PER_CPU)) { + /* + * We need to do a hard force of sched synchronization. + * This is because we use preempt_disable() to do RCU, but + * the function tracers can be called where RCU is not watching + * (like before user_exit()). We can not rely on the RCU + * infrastructure to do the synchronization, thus we must do it + * ourselves. + */ schedule_on_each_cpu(ftrace_sync); + /* + * When the kernel is preeptive, tasks can be preempted + * while on a ftrace trampoline. Just scheduling a task on + * a CPU is not good enough to flush them. Calling + * synchornize_rcu_tasks() will wait for those tasks to + * execute and either schedule voluntarily or enter user space. + */ + if (IS_ENABLED(CONFIG_PREEMPT)) + synchronize_rcu_tasks(); + arch_ftrace_trampoline_free(ops); if (ops->flags & FTRACE_OPS_FL_PER_CPU) @@ -2880,7 +2972,7 @@ ops_references_rec(struct ftrace_ops *ops, struct dyn_ftrace *rec) /* The function must be in the filter */ if (!ftrace_hash_empty(ops->func_hash->filter_hash) && - !ftrace_lookup_ip(ops->func_hash->filter_hash, rec->ip)) + !__ftrace_lookup_ip(ops->func_hash->filter_hash, rec->ip)) return 0; /* If in notrace hash, we ignore it too */ @@ -3037,37 +3129,69 @@ ftrace_allocate_pages(unsigned long num_to_init) struct ftrace_iterator { loff_t pos; loff_t func_pos; + loff_t mod_pos; struct ftrace_page *pg; struct dyn_ftrace *func; struct ftrace_func_probe *probe; + struct ftrace_func_entry *probe_entry; struct trace_parser parser; struct ftrace_hash *hash; struct ftrace_ops *ops; - int hidx; + struct trace_array *tr; + struct list_head *mod_list; + int pidx; int idx; unsigned flags; }; static void * -t_hash_next(struct seq_file *m, loff_t *pos) +t_probe_next(struct seq_file *m, loff_t *pos) { struct ftrace_iterator *iter = m->private; + struct trace_array *tr = iter->ops->private; + struct list_head *func_probes; + struct ftrace_hash *hash; + struct list_head *next; struct hlist_node *hnd = NULL; struct hlist_head *hhd; + int size; (*pos)++; iter->pos = *pos; - if (iter->probe) - hnd = &iter->probe->node; - retry: - if (iter->hidx >= FTRACE_FUNC_HASHSIZE) + if (!tr) return NULL; - hhd = &ftrace_func_hash[iter->hidx]; + func_probes = &tr->func_probes; + if (list_empty(func_probes)) + return NULL; + + if (!iter->probe) { + next = func_probes->next; + iter->probe = list_entry(next, struct ftrace_func_probe, list); + } + + if (iter->probe_entry) + hnd = &iter->probe_entry->hlist; + + hash = iter->probe->ops.func_hash->filter_hash; + size = 1 << hash->size_bits; + + retry: + if (iter->pidx >= size) { + if (iter->probe->list.next == func_probes) + return NULL; + next = iter->probe->list.next; + iter->probe = list_entry(next, struct ftrace_func_probe, list); + hash = iter->probe->ops.func_hash->filter_hash; + size = 1 << hash->size_bits; + iter->pidx = 0; + } + + hhd = &hash->buckets[iter->pidx]; if (hlist_empty(hhd)) { - iter->hidx++; + iter->pidx++; hnd = NULL; goto retry; } @@ -3077,7 +3201,7 @@ t_hash_next(struct seq_file *m, loff_t *pos) else { hnd = hnd->next; if (!hnd) { - iter->hidx++; + iter->pidx++; goto retry; } } @@ -3085,26 +3209,28 @@ t_hash_next(struct seq_file *m, loff_t *pos) if (WARN_ON_ONCE(!hnd)) return NULL; - iter->probe = hlist_entry(hnd, struct ftrace_func_probe, node); + iter->probe_entry = hlist_entry(hnd, struct ftrace_func_entry, hlist); return iter; } -static void *t_hash_start(struct seq_file *m, loff_t *pos) +static void *t_probe_start(struct seq_file *m, loff_t *pos) { struct ftrace_iterator *iter = m->private; void *p = NULL; loff_t l; - if (!(iter->flags & FTRACE_ITER_DO_HASH)) + if (!(iter->flags & FTRACE_ITER_DO_PROBES)) return NULL; - if (iter->func_pos > *pos) + if (iter->mod_pos > *pos) return NULL; - iter->hidx = 0; - for (l = 0; l <= (*pos - iter->func_pos); ) { - p = t_hash_next(m, &l); + iter->probe = NULL; + iter->probe_entry = NULL; + iter->pidx = 0; + for (l = 0; l <= (*pos - iter->mod_pos); ) { + p = t_probe_next(m, &l); if (!p) break; } @@ -3112,50 +3238,118 @@ static void *t_hash_start(struct seq_file *m, loff_t *pos) return NULL; /* Only set this if we have an item */ - iter->flags |= FTRACE_ITER_HASH; + iter->flags |= FTRACE_ITER_PROBE; return iter; } static int -t_hash_show(struct seq_file *m, struct ftrace_iterator *iter) +t_probe_show(struct seq_file *m, struct ftrace_iterator *iter) { - struct ftrace_func_probe *rec; + struct ftrace_func_entry *probe_entry; + struct ftrace_probe_ops *probe_ops; + struct ftrace_func_probe *probe; - rec = iter->probe; - if (WARN_ON_ONCE(!rec)) + probe = iter->probe; + probe_entry = iter->probe_entry; + + if (WARN_ON_ONCE(!probe || !probe_entry)) return -EIO; - if (rec->ops->print) - return rec->ops->print(m, rec->ip, rec->ops, rec->data); + probe_ops = probe->probe_ops; - seq_printf(m, "%ps:%ps", (void *)rec->ip, (void *)rec->ops->func); + if (probe_ops->print) + return probe_ops->print(m, probe_entry->ip, probe_ops, probe->data); - if (rec->data) - seq_printf(m, ":%p", rec->data); - seq_putc(m, '\n'); + seq_printf(m, "%ps:%ps\n", (void *)probe_entry->ip, + (void *)probe_ops->func); return 0; } static void * -t_next(struct seq_file *m, void *v, loff_t *pos) +t_mod_next(struct seq_file *m, loff_t *pos) { struct ftrace_iterator *iter = m->private; - struct ftrace_ops *ops = iter->ops; - struct dyn_ftrace *rec = NULL; + struct trace_array *tr = iter->tr; - if (unlikely(ftrace_disabled)) + (*pos)++; + iter->pos = *pos; + + iter->mod_list = iter->mod_list->next; + + if (iter->mod_list == &tr->mod_trace || + iter->mod_list == &tr->mod_notrace) { + iter->flags &= ~FTRACE_ITER_MOD; return NULL; + } - if (iter->flags & FTRACE_ITER_HASH) - return t_hash_next(m, pos); + iter->mod_pos = *pos; - (*pos)++; - iter->pos = iter->func_pos = *pos; + return iter; +} + +static void *t_mod_start(struct seq_file *m, loff_t *pos) +{ + struct ftrace_iterator *iter = m->private; + void *p = NULL; + loff_t l; + + if (iter->func_pos > *pos) + return NULL; + + iter->mod_pos = iter->func_pos; + + /* probes are only available if tr is set */ + if (!iter->tr) + return NULL; + + for (l = 0; l <= (*pos - iter->func_pos); ) { + p = t_mod_next(m, &l); + if (!p) + break; + } + if (!p) { + iter->flags &= ~FTRACE_ITER_MOD; + return t_probe_start(m, pos); + } + + /* Only set this if we have an item */ + iter->flags |= FTRACE_ITER_MOD; + + return iter; +} + +static int +t_mod_show(struct seq_file *m, struct ftrace_iterator *iter) +{ + struct ftrace_mod_load *ftrace_mod; + struct trace_array *tr = iter->tr; + + if (WARN_ON_ONCE(!iter->mod_list) || + iter->mod_list == &tr->mod_trace || + iter->mod_list == &tr->mod_notrace) + return -EIO; - if (iter->flags & FTRACE_ITER_PRINTALL) - return t_hash_start(m, pos); + ftrace_mod = list_entry(iter->mod_list, struct ftrace_mod_load, list); + + if (ftrace_mod->func) + seq_printf(m, "%s", ftrace_mod->func); + else + seq_putc(m, '*'); + + seq_printf(m, ":mod:%s\n", ftrace_mod->module); + + return 0; +} + +static void * +t_func_next(struct seq_file *m, loff_t *pos) +{ + struct ftrace_iterator *iter = m->private; + struct dyn_ftrace *rec = NULL; + + (*pos)++; retry: if (iter->idx >= iter->pg->index) { @@ -3166,11 +3360,8 @@ t_next(struct seq_file *m, void *v, loff_t *pos) } } else { rec = &iter->pg->records[iter->idx++]; - if (((iter->flags & FTRACE_ITER_FILTER) && - !(ftrace_lookup_ip(ops->func_hash->filter_hash, rec->ip))) || - - ((iter->flags & FTRACE_ITER_NOTRACE) && - !ftrace_lookup_ip(ops->func_hash->notrace_hash, rec->ip)) || + if (((iter->flags & (FTRACE_ITER_FILTER | FTRACE_ITER_NOTRACE)) && + !ftrace_lookup_ip(iter->hash, rec->ip)) || ((iter->flags & FTRACE_ITER_ENABLED) && !(rec->flags & FTRACE_FL_ENABLED))) { @@ -3181,24 +3372,54 @@ t_next(struct seq_file *m, void *v, loff_t *pos) } if (!rec) - return t_hash_start(m, pos); + return NULL; + iter->pos = iter->func_pos = *pos; iter->func = rec; return iter; } +static void * +t_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct ftrace_iterator *iter = m->private; + loff_t l = *pos; /* t_probe_start() must use original pos */ + void *ret; + + if (unlikely(ftrace_disabled)) + return NULL; + + if (iter->flags & FTRACE_ITER_PROBE) + return t_probe_next(m, pos); + + if (iter->flags & FTRACE_ITER_MOD) + return t_mod_next(m, pos); + + if (iter->flags & FTRACE_ITER_PRINTALL) { + /* next must increment pos, and t_probe_start does not */ + (*pos)++; + return t_mod_start(m, &l); + } + + ret = t_func_next(m, pos); + + if (!ret) + return t_mod_start(m, &l); + + return ret; +} + static void reset_iter_read(struct ftrace_iterator *iter) { iter->pos = 0; iter->func_pos = 0; - iter->flags &= ~(FTRACE_ITER_PRINTALL | FTRACE_ITER_HASH); + iter->flags &= ~(FTRACE_ITER_PRINTALL | FTRACE_ITER_PROBE | FTRACE_ITER_MOD); } static void *t_start(struct seq_file *m, loff_t *pos) { struct ftrace_iterator *iter = m->private; - struct ftrace_ops *ops = iter->ops; void *p = NULL; loff_t l; @@ -3218,20 +3439,19 @@ static void *t_start(struct seq_file *m, loff_t *pos) * off, we can short cut and just print out that all * functions are enabled. */ - if ((iter->flags & FTRACE_ITER_FILTER && - ftrace_hash_empty(ops->func_hash->filter_hash)) || - (iter->flags & FTRACE_ITER_NOTRACE && - ftrace_hash_empty(ops->func_hash->notrace_hash))) { + if ((iter->flags & (FTRACE_ITER_FILTER | FTRACE_ITER_NOTRACE)) && + ftrace_hash_empty(iter->hash)) { + iter->func_pos = 1; /* Account for the message */ if (*pos > 0) - return t_hash_start(m, pos); + return t_mod_start(m, pos); iter->flags |= FTRACE_ITER_PRINTALL; /* reset in case of seek/pread */ - iter->flags &= ~FTRACE_ITER_HASH; + iter->flags &= ~FTRACE_ITER_PROBE; return iter; } - if (iter->flags & FTRACE_ITER_HASH) - return t_hash_start(m, pos); + if (iter->flags & FTRACE_ITER_MOD) + return t_mod_start(m, pos); /* * Unfortunately, we need to restart at ftrace_pages_start @@ -3241,13 +3461,13 @@ static void *t_start(struct seq_file *m, loff_t *pos) iter->pg = ftrace_pages_start; iter->idx = 0; for (l = 0; l <= *pos; ) { - p = t_next(m, p, &l); + p = t_func_next(m, &l); if (!p) break; } if (!p) - return t_hash_start(m, pos); + return t_mod_start(m, pos); return iter; } @@ -3278,8 +3498,11 @@ static int t_show(struct seq_file *m, void *v) struct ftrace_iterator *iter = m->private; struct dyn_ftrace *rec; - if (iter->flags & FTRACE_ITER_HASH) - return t_hash_show(m, iter); + if (iter->flags & FTRACE_ITER_PROBE) + return t_probe_show(m, iter); + + if (iter->flags & FTRACE_ITER_MOD) + return t_mod_show(m, iter); if (iter->flags & FTRACE_ITER_PRINTALL) { if (iter->flags & FTRACE_ITER_NOTRACE) @@ -3340,12 +3563,13 @@ ftrace_avail_open(struct inode *inode, struct file *file) return -ENODEV; iter = __seq_open_private(file, &show_ftrace_seq_ops, sizeof(*iter)); - if (iter) { - iter->pg = ftrace_pages_start; - iter->ops = &global_ops; - } + if (!iter) + return -ENOMEM; + + iter->pg = ftrace_pages_start; + iter->ops = &global_ops; - return iter ? 0 : -ENOMEM; + return 0; } static int @@ -3354,13 +3578,14 @@ ftrace_enabled_open(struct inode *inode, struct file *file) struct ftrace_iterator *iter; iter = __seq_open_private(file, &show_ftrace_seq_ops, sizeof(*iter)); - if (iter) { - iter->pg = ftrace_pages_start; - iter->flags = FTRACE_ITER_ENABLED; - iter->ops = &global_ops; - } + if (!iter) + return -ENOMEM; + + iter->pg = ftrace_pages_start; + iter->flags = FTRACE_ITER_ENABLED; + iter->ops = &global_ops; - return iter ? 0 : -ENOMEM; + return 0; } /** @@ -3385,6 +3610,8 @@ ftrace_regex_open(struct ftrace_ops *ops, int flag, { struct ftrace_iterator *iter; struct ftrace_hash *hash; + struct list_head *mod_head; + struct trace_array *tr = ops->private; int ret = 0; ftrace_ops_init(ops); @@ -3403,21 +3630,29 @@ ftrace_regex_open(struct ftrace_ops *ops, int flag, iter->ops = ops; iter->flags = flag; + iter->tr = tr; mutex_lock(&ops->func_hash->regex_lock); - if (flag & FTRACE_ITER_NOTRACE) + if (flag & FTRACE_ITER_NOTRACE) { hash = ops->func_hash->notrace_hash; - else + mod_head = tr ? &tr->mod_notrace : NULL; + } else { hash = ops->func_hash->filter_hash; + mod_head = tr ? &tr->mod_trace : NULL; + } + + iter->mod_list = mod_head; if (file->f_mode & FMODE_WRITE) { const int size_bits = FTRACE_HASH_DEFAULT_BITS; - if (file->f_flags & O_TRUNC) + if (file->f_flags & O_TRUNC) { iter->hash = alloc_ftrace_hash(size_bits); - else + clear_ftrace_mod_list(mod_head); + } else { iter->hash = alloc_and_copy_ftrace_hash(size_bits, hash); + } if (!iter->hash) { trace_parser_put(&iter->parser); @@ -3425,7 +3660,8 @@ ftrace_regex_open(struct ftrace_ops *ops, int flag, ret = -ENOMEM; goto out_unlock; } - } + } else + iter->hash = hash; if (file->f_mode & FMODE_READ) { iter->pg = ftrace_pages_start; @@ -3455,7 +3691,7 @@ ftrace_filter_open(struct inode *inode, struct file *file) struct ftrace_ops *ops = inode->i_private; return ftrace_regex_open(ops, - FTRACE_ITER_FILTER | FTRACE_ITER_DO_HASH, + FTRACE_ITER_FILTER | FTRACE_ITER_DO_PROBES, inode, file); } @@ -3558,22 +3794,20 @@ ftrace_match_record(struct dyn_ftrace *rec, struct ftrace_glob *func_g, /* blank module name to match all modules */ if (!mod_g->len) { /* blank module globbing: modname xor exclude_mod */ - if ((!exclude_mod) != (!modname)) + if (!exclude_mod != !modname) goto func_match; return 0; } - /* not matching the module */ - if (!modname || !mod_matches) { - if (exclude_mod) - goto func_match; - else - return 0; - } - - if (mod_matches && exclude_mod) + /* + * exclude_mod is set to trace everything but the given + * module. If it is set and the module matches, then + * return 0. If it is not set, and the module doesn't match + * also return 0. Otherwise, check the function to see if + * that matches. + */ + if (!mod_matches == !exclude_mod) return 0; - func_match: /* blank search means to match all funcs in the mod */ if (!func_g->len) @@ -3594,7 +3828,7 @@ match_records(struct ftrace_hash *hash, char *func, int len, char *mod) int exclude_mod = 0; int found = 0; int ret; - int clear_filter; + int clear_filter = 0; if (func) { func_g.type = filter_parse_regex(func, len, &func_g.search, @@ -3639,6 +3873,215 @@ ftrace_match_records(struct ftrace_hash *hash, char *buff, int len) return match_records(hash, buff, len, NULL); } +static void ftrace_ops_update_code(struct ftrace_ops *ops, + struct ftrace_ops_hash *old_hash) +{ + struct ftrace_ops *op; + + if (!ftrace_enabled) + return; + + if (ops->flags & FTRACE_OPS_FL_ENABLED) { + ftrace_run_modify_code(ops, FTRACE_UPDATE_CALLS, old_hash); + return; + } + + /* + * If this is the shared global_ops filter, then we need to + * check if there is another ops that shares it, is enabled. + * If so, we still need to run the modify code. + */ + if (ops->func_hash != &global_ops.local_hash) + return; + + do_for_each_ftrace_op(op, ftrace_ops_list) { + if (op->func_hash == &global_ops.local_hash && + op->flags & FTRACE_OPS_FL_ENABLED) { + ftrace_run_modify_code(op, FTRACE_UPDATE_CALLS, old_hash); + /* Only need to do this once */ + return; + } + } while_for_each_ftrace_op(op); +} + +static int ftrace_hash_move_and_update_ops(struct ftrace_ops *ops, + struct ftrace_hash **orig_hash, + struct ftrace_hash *hash, + int enable) +{ + struct ftrace_ops_hash old_hash_ops; + struct ftrace_hash *old_hash; + int ret; + + old_hash = *orig_hash; + old_hash_ops.filter_hash = ops->func_hash->filter_hash; + old_hash_ops.notrace_hash = ops->func_hash->notrace_hash; + ret = ftrace_hash_move(ops, enable, orig_hash, hash); + if (!ret) { + ftrace_ops_update_code(ops, &old_hash_ops); + free_ftrace_hash_rcu(old_hash); + } + return ret; +} + +static bool module_exists(const char *module) +{ + /* All modules have the symbol __this_module */ + const char this_mod[] = "__this_module"; + const int modname_size = MAX_PARAM_PREFIX_LEN + sizeof(this_mod) + 1; + char modname[modname_size + 1]; + unsigned long val; + int n; + + n = snprintf(modname, modname_size + 1, "%s:%s", module, this_mod); + + if (n > modname_size) + return false; + + val = module_kallsyms_lookup_name(modname); + return val != 0; +} + +static int cache_mod(struct trace_array *tr, + const char *func, char *module, int enable) +{ + struct ftrace_mod_load *ftrace_mod, *n; + struct list_head *head = enable ? &tr->mod_trace : &tr->mod_notrace; + int ret; + + mutex_lock(&ftrace_lock); + + /* We do not cache inverse filters */ + if (func[0] == '!') { + func++; + ret = -EINVAL; + + /* Look to remove this hash */ + list_for_each_entry_safe(ftrace_mod, n, head, list) { + if (strcmp(ftrace_mod->module, module) != 0) + continue; + + /* no func matches all */ + if (strcmp(func, "*") == 0 || + (ftrace_mod->func && + strcmp(ftrace_mod->func, func) == 0)) { + ret = 0; + free_ftrace_mod(ftrace_mod); + continue; + } + } + goto out; + } + + ret = -EINVAL; + /* We only care about modules that have not been loaded yet */ + if (module_exists(module)) + goto out; + + /* Save this string off, and execute it when the module is loaded */ + ret = ftrace_add_mod(tr, func, module, enable); + out: + mutex_unlock(&ftrace_lock); + + return ret; +} + +static int +ftrace_set_regex(struct ftrace_ops *ops, unsigned char *buf, int len, + int reset, int enable); + +#ifdef CONFIG_MODULES +static void process_mod_list(struct list_head *head, struct ftrace_ops *ops, + char *mod, bool enable) +{ + struct ftrace_mod_load *ftrace_mod, *n; + struct ftrace_hash **orig_hash, *new_hash; + LIST_HEAD(process_mods); + char *func; + int ret; + + mutex_lock(&ops->func_hash->regex_lock); + + if (enable) + orig_hash = &ops->func_hash->filter_hash; + else + orig_hash = &ops->func_hash->notrace_hash; + + new_hash = alloc_and_copy_ftrace_hash(FTRACE_HASH_DEFAULT_BITS, + *orig_hash); + if (!new_hash) + goto out; /* warn? */ + + mutex_lock(&ftrace_lock); + + list_for_each_entry_safe(ftrace_mod, n, head, list) { + + if (strcmp(ftrace_mod->module, mod) != 0) + continue; + + if (ftrace_mod->func) + func = kstrdup(ftrace_mod->func, GFP_KERNEL); + else + func = kstrdup("*", GFP_KERNEL); + + if (!func) /* warn? */ + continue; + + list_del(&ftrace_mod->list); + list_add(&ftrace_mod->list, &process_mods); + + /* Use the newly allocated func, as it may be "*" */ + kfree(ftrace_mod->func); + ftrace_mod->func = func; + } + + mutex_unlock(&ftrace_lock); + + list_for_each_entry_safe(ftrace_mod, n, &process_mods, list) { + + func = ftrace_mod->func; + + /* Grabs ftrace_lock, which is why we have this extra step */ + match_records(new_hash, func, strlen(func), mod); + free_ftrace_mod(ftrace_mod); + } + + if (enable && list_empty(head)) + new_hash->flags &= ~FTRACE_HASH_FL_MOD; + + mutex_lock(&ftrace_lock); + + ret = ftrace_hash_move_and_update_ops(ops, orig_hash, + new_hash, enable); + mutex_unlock(&ftrace_lock); + + out: + mutex_unlock(&ops->func_hash->regex_lock); + + free_ftrace_hash(new_hash); +} + +static void process_cached_mods(const char *mod_name) +{ + struct trace_array *tr; + char *mod; + + mod = kstrdup(mod_name, GFP_KERNEL); + if (!mod) + return; + + mutex_lock(&trace_types_lock); + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (!list_empty(&tr->mod_trace)) + process_mod_list(&tr->mod_trace, tr->ops, mod, true); + if (!list_empty(&tr->mod_notrace)) + process_mod_list(&tr->mod_notrace, tr->ops, mod, false); + } + mutex_unlock(&trace_types_lock); + + kfree(mod); +} +#endif /* * We register the module command as a template to show others how @@ -3646,11 +4089,17 @@ ftrace_match_records(struct ftrace_hash *hash, char *buff, int len) */ static int -ftrace_mod_callback(struct ftrace_hash *hash, - char *func, char *cmd, char *module, int enable) +ftrace_mod_callback(struct trace_array *tr, struct ftrace_hash *hash, + char *func_orig, char *cmd, char *module, int enable) { + char *func; int ret; + /* match_records() modifies func, and we need the original */ + func = kstrdup(func_orig, GFP_KERNEL); + if (!func) + return -ENOMEM; + /* * cmd == 'mod' because we only registered this func * for the 'mod' ftrace_func_command. @@ -3659,8 +4108,10 @@ ftrace_mod_callback(struct ftrace_hash *hash, * parameter. */ ret = match_records(hash, func, strlen(func), module); + kfree(func); + if (!ret) - return -EINVAL; + return cache_mod(tr, func_orig, module, enable); if (ret < 0) return ret; return 0; @@ -3680,16 +4131,11 @@ core_initcall(ftrace_mod_cmd_init); static void function_trace_probe_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct pt_regs *pt_regs) { - struct ftrace_func_probe *entry; - struct hlist_head *hhd; - unsigned long key; + struct ftrace_probe_ops *probe_ops; + struct ftrace_func_probe *probe; - key = hash_long(ip, FTRACE_HASH_BITS); - - hhd = &ftrace_func_hash[key]; - - if (hlist_empty(hhd)) - return; + probe = container_of(op, struct ftrace_func_probe, ops); + probe_ops = probe->probe_ops; /* * Disable preemption for these calls to prevent a RCU grace @@ -3697,210 +4143,340 @@ static void function_trace_probe_call(unsigned long ip, unsigned long parent_ip, * on the hash. rcu_read_lock is too dangerous here. */ preempt_disable_notrace(); - hlist_for_each_entry_rcu_notrace(entry, hhd, node) { - if (entry->ip == ip) - entry->ops->func(ip, parent_ip, &entry->data); - } + probe_ops->func(ip, parent_ip, probe->tr, probe_ops, probe->data); preempt_enable_notrace(); } -static struct ftrace_ops trace_probe_ops __read_mostly = -{ - .func = function_trace_probe_call, - .flags = FTRACE_OPS_FL_INITIALIZED, - INIT_OPS_HASH(trace_probe_ops) +struct ftrace_func_map { + struct ftrace_func_entry entry; + void *data; }; -static int ftrace_probe_registered; +struct ftrace_func_mapper { + struct ftrace_hash hash; +}; -static void __enable_ftrace_function_probe(struct ftrace_ops_hash *old_hash) +/** + * allocate_ftrace_func_mapper - allocate a new ftrace_func_mapper + * + * Returns a ftrace_func_mapper descriptor that can be used to map ips to data. + */ +struct ftrace_func_mapper *allocate_ftrace_func_mapper(void) { - int ret; - int i; + struct ftrace_hash *hash; - if (ftrace_probe_registered) { - /* still need to update the function call sites */ - if (ftrace_enabled) - ftrace_run_modify_code(&trace_probe_ops, FTRACE_UPDATE_CALLS, - old_hash); - return; - } + /* + * The mapper is simply a ftrace_hash, but since the entries + * in the hash are not ftrace_func_entry type, we define it + * as a separate structure. + */ + hash = alloc_ftrace_hash(FTRACE_HASH_DEFAULT_BITS); + return (struct ftrace_func_mapper *)hash; +} - for (i = 0; i < FTRACE_FUNC_HASHSIZE; i++) { - struct hlist_head *hhd = &ftrace_func_hash[i]; - if (hhd->first) - break; - } - /* Nothing registered? */ - if (i == FTRACE_FUNC_HASHSIZE) - return; +/** + * ftrace_func_mapper_find_ip - Find some data mapped to an ip + * @mapper: The mapper that has the ip maps + * @ip: the instruction pointer to find the data for + * + * Returns the data mapped to @ip if found otherwise NULL. The return + * is actually the address of the mapper data pointer. The address is + * returned for use cases where the data is no bigger than a long, and + * the user can use the data pointer as its data instead of having to + * allocate more memory for the reference. + */ +void **ftrace_func_mapper_find_ip(struct ftrace_func_mapper *mapper, + unsigned long ip) +{ + struct ftrace_func_entry *entry; + struct ftrace_func_map *map; - ret = ftrace_startup(&trace_probe_ops, 0); + entry = ftrace_lookup_ip(&mapper->hash, ip); + if (!entry) + return NULL; - ftrace_probe_registered = 1; + map = (struct ftrace_func_map *)entry; + return &map->data; } -static void __disable_ftrace_function_probe(void) +/** + * ftrace_func_mapper_add_ip - Map some data to an ip + * @mapper: The mapper that has the ip maps + * @ip: The instruction pointer address to map @data to + * @data: The data to map to @ip + * + * Returns 0 on succes otherwise an error. + */ +int ftrace_func_mapper_add_ip(struct ftrace_func_mapper *mapper, + unsigned long ip, void *data) { - int i; + struct ftrace_func_entry *entry; + struct ftrace_func_map *map; - if (!ftrace_probe_registered) - return; + entry = ftrace_lookup_ip(&mapper->hash, ip); + if (entry) + return -EBUSY; - for (i = 0; i < FTRACE_FUNC_HASHSIZE; i++) { - struct hlist_head *hhd = &ftrace_func_hash[i]; - if (hhd->first) - return; - } + map = kmalloc(sizeof(*map), GFP_KERNEL); + if (!map) + return -ENOMEM; - /* no more funcs left */ - ftrace_shutdown(&trace_probe_ops, 0); + map->entry.ip = ip; + map->data = data; - ftrace_probe_registered = 0; -} + __add_hash_entry(&mapper->hash, &map->entry); + return 0; +} -static void ftrace_free_entry(struct ftrace_func_probe *entry) +/** + * ftrace_func_mapper_remove_ip - Remove an ip from the mapping + * @mapper: The mapper that has the ip maps + * @ip: The instruction pointer address to remove the data from + * + * Returns the data if it is found, otherwise NULL. + * Note, if the data pointer is used as the data itself, (see + * ftrace_func_mapper_find_ip(), then the return value may be meaningless, + * if the data pointer was set to zero. + */ +void *ftrace_func_mapper_remove_ip(struct ftrace_func_mapper *mapper, + unsigned long ip) { - if (entry->ops->free) - entry->ops->free(entry->ops, entry->ip, &entry->data); + struct ftrace_func_entry *entry; + struct ftrace_func_map *map; + void *data; + + entry = ftrace_lookup_ip(&mapper->hash, ip); + if (!entry) + return NULL; + + map = (struct ftrace_func_map *)entry; + data = map->data; + + remove_hash_entry(&mapper->hash, entry); kfree(entry); + + return data; +} + +/** + * free_ftrace_func_mapper - free a mapping of ips and data + * @mapper: The mapper that has the ip maps + * @free_func: A function to be called on each data item. + * + * This is used to free the function mapper. The @free_func is optional + * and can be used if the data needs to be freed as well. + */ +void free_ftrace_func_mapper(struct ftrace_func_mapper *mapper, + ftrace_mapper_func free_func) +{ + struct ftrace_func_entry *entry; + struct ftrace_func_map *map; + struct hlist_head *hhd; + int size = 1 << mapper->hash.size_bits; + int i; + + if (free_func && mapper->hash.count) { + for (i = 0; i < size; i++) { + hhd = &mapper->hash.buckets[i]; + hlist_for_each_entry(entry, hhd, hlist) { + map = (struct ftrace_func_map *)entry; + free_func(map); + } + } + } + free_ftrace_hash(&mapper->hash); +} + +static void release_probe(struct ftrace_func_probe *probe) +{ + struct ftrace_probe_ops *probe_ops; + + mutex_lock(&ftrace_lock); + + WARN_ON(probe->ref <= 0); + + /* Subtract the ref that was used to protect this instance */ + probe->ref--; + + if (!probe->ref) { + probe_ops = probe->probe_ops; + /* + * Sending zero as ip tells probe_ops to free + * the probe->data itself + */ + if (probe_ops->free) + probe_ops->free(probe_ops, probe->tr, 0, probe->data); + list_del(&probe->list); + kfree(probe); + } + mutex_unlock(&ftrace_lock); +} + +static void acquire_probe_locked(struct ftrace_func_probe *probe) +{ + /* + * Add one ref to keep it from being freed when releasing the + * ftrace_lock mutex. + */ + probe->ref++; } int -register_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, - void *data) +register_ftrace_function_probe(char *glob, struct trace_array *tr, + struct ftrace_probe_ops *probe_ops, + void *data) { - struct ftrace_ops_hash old_hash_ops; - struct ftrace_func_probe *entry; - struct ftrace_glob func_g; - struct ftrace_hash **orig_hash = &trace_probe_ops.func_hash->filter_hash; - struct ftrace_hash *old_hash = *orig_hash; + struct ftrace_func_entry *entry; + struct ftrace_func_probe *probe; + struct ftrace_hash **orig_hash; + struct ftrace_hash *old_hash; struct ftrace_hash *hash; - struct ftrace_page *pg; - struct dyn_ftrace *rec; - int not; - unsigned long key; int count = 0; + int size; int ret; + int i; - func_g.type = filter_parse_regex(glob, strlen(glob), - &func_g.search, ¬); - func_g.len = strlen(func_g.search); - - /* we do not support '!' for function probes */ - if (WARN_ON(not)) + if (WARN_ON(!tr)) return -EINVAL; - mutex_lock(&trace_probe_ops.func_hash->regex_lock); + /* We do not support '!' for function probes */ + if (WARN_ON(glob[0] == '!')) + return -EINVAL; - old_hash_ops.filter_hash = old_hash; - /* Probes only have filters */ - old_hash_ops.notrace_hash = NULL; - hash = alloc_and_copy_ftrace_hash(FTRACE_HASH_DEFAULT_BITS, old_hash); - if (!hash) { - count = -ENOMEM; - goto out; + mutex_lock(&ftrace_lock); + /* Check if the probe_ops is already registered */ + list_for_each_entry(probe, &tr->func_probes, list) { + if (probe->probe_ops == probe_ops) + break; } - - if (unlikely(ftrace_disabled)) { - count = -ENODEV; - goto out; + if (&probe->list == &tr->func_probes) { + probe = kzalloc(sizeof(*probe), GFP_KERNEL); + if (!probe) { + mutex_unlock(&ftrace_lock); + return -ENOMEM; + } + probe->probe_ops = probe_ops; + probe->ops.func = function_trace_probe_call; + probe->tr = tr; + ftrace_ops_init(&probe->ops); + list_add(&probe->list, &tr->func_probes); } - mutex_lock(&ftrace_lock); + acquire_probe_locked(probe); - do_for_each_ftrace_rec(pg, rec) { + mutex_unlock(&ftrace_lock); - if (rec->flags & FTRACE_FL_DISABLED) - continue; + mutex_lock(&probe->ops.func_hash->regex_lock); - if (!ftrace_match_record(rec, &func_g, NULL, 0)) - continue; + orig_hash = &probe->ops.func_hash->filter_hash; + old_hash = *orig_hash; + hash = alloc_and_copy_ftrace_hash(FTRACE_HASH_DEFAULT_BITS, old_hash); - entry = kmalloc(sizeof(*entry), GFP_KERNEL); - if (!entry) { - /* If we did not process any, then return error */ - if (!count) - count = -ENOMEM; - goto out_unlock; - } + ret = ftrace_match_records(hash, glob, strlen(glob)); - count++; + /* Nothing found? */ + if (!ret) + ret = -EINVAL; - entry->data = data; + if (ret < 0) + goto out; - /* - * The caller might want to do something special - * for each function we find. We call the callback - * to give the caller an opportunity to do so. - */ - if (ops->init) { - if (ops->init(ops, rec->ip, &entry->data) < 0) { - /* caller does not like this func */ - kfree(entry); + size = 1 << hash->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &hash->buckets[i], hlist) { + if (ftrace_lookup_ip(old_hash, entry->ip)) continue; + /* + * The caller might want to do something special + * for each function we find. We call the callback + * to give the caller an opportunity to do so. + */ + if (probe_ops->init) { + ret = probe_ops->init(probe_ops, tr, + entry->ip, data, + &probe->data); + if (ret < 0) { + if (probe_ops->free && count) + probe_ops->free(probe_ops, tr, + 0, probe->data); + probe->data = NULL; + goto out; + } } + count++; } + } - ret = enter_record(hash, rec, 0); - if (ret < 0) { - kfree(entry); - count = ret; - goto out_unlock; - } - - entry->ops = ops; - entry->ip = rec->ip; - - key = hash_long(entry->ip, FTRACE_HASH_BITS); - hlist_add_head_rcu(&entry->node, &ftrace_func_hash[key]); + mutex_lock(&ftrace_lock); - } while_for_each_ftrace_rec(); + if (!count) { + /* Nothing was added? */ + ret = -EINVAL; + goto out_unlock; + } - ret = ftrace_hash_move(&trace_probe_ops, 1, orig_hash, hash); + ret = ftrace_hash_move_and_update_ops(&probe->ops, orig_hash, + hash, 1); + if (ret < 0) + goto err_unlock; - __enable_ftrace_function_probe(&old_hash_ops); + /* One ref for each new function traced */ + probe->ref += count; - if (!ret) - free_ftrace_hash_rcu(old_hash); - else - count = ret; + if (!(probe->ops.flags & FTRACE_OPS_FL_ENABLED)) + ret = ftrace_startup(&probe->ops, 0); out_unlock: mutex_unlock(&ftrace_lock); + + if (!ret) + ret = count; out: - mutex_unlock(&trace_probe_ops.func_hash->regex_lock); + mutex_unlock(&probe->ops.func_hash->regex_lock); free_ftrace_hash(hash); - return count; -} + release_probe(probe); -enum { - PROBE_TEST_FUNC = 1, - PROBE_TEST_DATA = 2 -}; + return ret; -static void -__unregister_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, - void *data, int flags) + err_unlock: + if (!probe_ops->free || !count) + goto out_unlock; + + /* Failed to do the move, need to call the free functions */ + for (i = 0; i < size; i++) { + hlist_for_each_entry(entry, &hash->buckets[i], hlist) { + if (ftrace_lookup_ip(old_hash, entry->ip)) + continue; + probe_ops->free(probe_ops, tr, entry->ip, probe->data); + } + } + goto out_unlock; +} + +int +unregister_ftrace_function_probe_func(char *glob, struct trace_array *tr, + struct ftrace_probe_ops *probe_ops) { - struct ftrace_func_entry *rec_entry; - struct ftrace_func_probe *entry; - struct ftrace_func_probe *p; + struct ftrace_ops_hash old_hash_ops; + struct ftrace_func_entry *entry; + struct ftrace_func_probe *probe; struct ftrace_glob func_g; - struct ftrace_hash **orig_hash = &trace_probe_ops.func_hash->filter_hash; - struct ftrace_hash *old_hash = *orig_hash; - struct list_head free_list; - struct ftrace_hash *hash; + struct ftrace_hash **orig_hash; + struct ftrace_hash *old_hash; + struct ftrace_hash *hash = NULL; struct hlist_node *tmp; + struct hlist_head hhd; char str[KSYM_SYMBOL_LEN]; - int i, ret; + int count = 0; + int i, ret = -ENODEV; + int size; - if (glob && (strcmp(glob, "*") == 0 || !strlen(glob))) + if (!glob || !strlen(glob) || !strcmp(glob, "*")) func_g.search = NULL; - else if (glob) { + else { int not; func_g.type = filter_parse_regex(glob, strlen(glob), @@ -3910,86 +4486,112 @@ __unregister_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, /* we do not support '!' for function probes */ if (WARN_ON(not)) - return; + return -EINVAL; } - mutex_lock(&trace_probe_ops.func_hash->regex_lock); + mutex_lock(&ftrace_lock); + /* Check if the probe_ops is already registered */ + list_for_each_entry(probe, &tr->func_probes, list) { + if (probe->probe_ops == probe_ops) + break; + } + if (&probe->list == &tr->func_probes) + goto err_unlock_ftrace; - hash = alloc_and_copy_ftrace_hash(FTRACE_HASH_DEFAULT_BITS, *orig_hash); - if (!hash) - /* Hmm, should report this somehow */ - goto out_unlock; + ret = -EINVAL; + if (!(probe->ops.flags & FTRACE_OPS_FL_INITIALIZED)) + goto err_unlock_ftrace; - INIT_LIST_HEAD(&free_list); + acquire_probe_locked(probe); - for (i = 0; i < FTRACE_FUNC_HASHSIZE; i++) { - struct hlist_head *hhd = &ftrace_func_hash[i]; + mutex_unlock(&ftrace_lock); - hlist_for_each_entry_safe(entry, tmp, hhd, node) { + mutex_lock(&probe->ops.func_hash->regex_lock); - /* break up if statements for readability */ - if ((flags & PROBE_TEST_FUNC) && entry->ops != ops) - continue; + orig_hash = &probe->ops.func_hash->filter_hash; + old_hash = *orig_hash; - if ((flags & PROBE_TEST_DATA) && entry->data != data) - continue; + if (ftrace_hash_empty(old_hash)) + goto out_unlock; + + old_hash_ops.filter_hash = old_hash; + /* Probes only have filters */ + old_hash_ops.notrace_hash = NULL; + + ret = -ENOMEM; + hash = alloc_and_copy_ftrace_hash(FTRACE_HASH_DEFAULT_BITS, old_hash); + if (!hash) + goto out_unlock; + + INIT_HLIST_HEAD(&hhd); + + size = 1 << hash->size_bits; + for (i = 0; i < size; i++) { + hlist_for_each_entry_safe(entry, tmp, &hash->buckets[i], hlist) { - /* do this last, since it is the most expensive */ if (func_g.search) { kallsyms_lookup(entry->ip, NULL, NULL, NULL, str); if (!ftrace_match(str, &func_g)) continue; } - - rec_entry = ftrace_lookup_ip(hash, entry->ip); - /* It is possible more than one entry had this ip */ - if (rec_entry) - free_hash_entry(hash, rec_entry); - - hlist_del_rcu(&entry->node); - list_add(&entry->free_list, &free_list); + count++; + remove_hash_entry(hash, entry); + hlist_add_head(&entry->hlist, &hhd); } } + + /* Nothing found? */ + if (!count) { + ret = -EINVAL; + goto out_unlock; + } + mutex_lock(&ftrace_lock); - __disable_ftrace_function_probe(); - /* - * Remove after the disable is called. Otherwise, if the last - * probe is removed, a null hash means *all enabled*. - */ - ret = ftrace_hash_move(&trace_probe_ops, 1, orig_hash, hash); + + WARN_ON(probe->ref < count); + + probe->ref -= count; + + if (ftrace_hash_empty(hash)) + ftrace_shutdown(&probe->ops, 0); + + ret = ftrace_hash_move_and_update_ops(&probe->ops, orig_hash, + hash, 1); + + /* still need to update the function call sites */ + if (ftrace_enabled && !ftrace_hash_empty(hash)) + ftrace_run_modify_code(&probe->ops, FTRACE_UPDATE_CALLS, + &old_hash_ops); synchronize_sched(); - if (!ret) - free_ftrace_hash_rcu(old_hash); - list_for_each_entry_safe(entry, p, &free_list, free_list) { - list_del(&entry->free_list); - ftrace_free_entry(entry); + hlist_for_each_entry_safe(entry, tmp, &hhd, hlist) { + hlist_del(&entry->hlist); + if (probe_ops->free) + probe_ops->free(probe_ops, tr, entry->ip, probe->data); + kfree(entry); } mutex_unlock(&ftrace_lock); out_unlock: - mutex_unlock(&trace_probe_ops.func_hash->regex_lock); + mutex_unlock(&probe->ops.func_hash->regex_lock); free_ftrace_hash(hash); -} -void -unregister_ftrace_function_probe(char *glob, struct ftrace_probe_ops *ops, - void *data) -{ - __unregister_ftrace_function_probe(glob, ops, data, - PROBE_TEST_FUNC | PROBE_TEST_DATA); -} + release_probe(probe); -void -unregister_ftrace_function_probe_func(char *glob, struct ftrace_probe_ops *ops) -{ - __unregister_ftrace_function_probe(glob, ops, NULL, PROBE_TEST_FUNC); + return ret; + + err_unlock_ftrace: + mutex_unlock(&ftrace_lock); + return ret; } -void unregister_ftrace_function_probe_all(char *glob) +void clear_ftrace_function_probes(struct trace_array *tr) { - __unregister_ftrace_function_probe(glob, NULL, NULL, 0); + struct ftrace_func_probe *probe, *n; + + list_for_each_entry_safe(probe, n, &tr->func_probes, list) + unregister_ftrace_function_probe_func(NULL, tr, probe->probe_ops); } static LIST_HEAD(ftrace_commands); @@ -4041,9 +4643,11 @@ __init int unregister_ftrace_command(struct ftrace_func_command *cmd) return ret; } -static int ftrace_process_regex(struct ftrace_hash *hash, +static int ftrace_process_regex(struct ftrace_iterator *iter, char *buff, int len, int enable) { + struct ftrace_hash *hash = iter->hash; + struct trace_array *tr = iter->ops->private; char *func, *command, *next = buff; struct ftrace_func_command *p; int ret = -EINVAL; @@ -4066,7 +4670,7 @@ static int ftrace_process_regex(struct ftrace_hash *hash, mutex_lock(&ftrace_cmd_mutex); list_for_each_entry(p, &ftrace_commands, list) { if (strcmp(p->name, command) == 0) { - ret = p->func(hash, func, command, next, enable); + ret = p->func(tr, hash, func, command, next, enable); goto out_unlock; } } @@ -4103,7 +4707,7 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, if (read >= 0 && trace_parser_loaded(parser) && !trace_parser_cont(parser)) { - ret = ftrace_process_regex(iter->hash, parser->buffer, + ret = ftrace_process_regex(iter, parser->buffer, parser->idx, enable); trace_parser_clear(parser); if (ret < 0) @@ -4148,44 +4752,11 @@ ftrace_match_addr(struct ftrace_hash *hash, unsigned long ip, int remove) return add_hash_entry(hash, ip); } -static void ftrace_ops_update_code(struct ftrace_ops *ops, - struct ftrace_ops_hash *old_hash) -{ - struct ftrace_ops *op; - - if (!ftrace_enabled) - return; - - if (ops->flags & FTRACE_OPS_FL_ENABLED) { - ftrace_run_modify_code(ops, FTRACE_UPDATE_CALLS, old_hash); - return; - } - - /* - * If this is the shared global_ops filter, then we need to - * check if there is another ops that shares it, is enabled. - * If so, we still need to run the modify code. - */ - if (ops->func_hash != &global_ops.local_hash) - return; - - do_for_each_ftrace_op(op, ftrace_ops_list) { - if (op->func_hash == &global_ops.local_hash && - op->flags & FTRACE_OPS_FL_ENABLED) { - ftrace_run_modify_code(op, FTRACE_UPDATE_CALLS, old_hash); - /* Only need to do this once */ - return; - } - } while_for_each_ftrace_op(op); -} - static int ftrace_set_hash(struct ftrace_ops *ops, unsigned char *buf, int len, unsigned long ip, int remove, int reset, int enable) { struct ftrace_hash **orig_hash; - struct ftrace_ops_hash old_hash_ops; - struct ftrace_hash *old_hash; struct ftrace_hash *hash; int ret; @@ -4220,14 +4791,7 @@ ftrace_set_hash(struct ftrace_ops *ops, unsigned char *buf, int len, } mutex_lock(&ftrace_lock); - old_hash = *orig_hash; - old_hash_ops.filter_hash = ops->func_hash->filter_hash; - old_hash_ops.notrace_hash = ops->func_hash->notrace_hash; - ret = ftrace_hash_move(ops, enable, orig_hash, hash); - if (!ret) { - ftrace_ops_update_code(ops, &old_hash_ops); - free_ftrace_hash_rcu(old_hash); - } + ret = ftrace_hash_move_and_update_ops(ops, orig_hash, hash, enable); mutex_unlock(&ftrace_lock); out_regex_unlock: @@ -4382,7 +4946,7 @@ __setup("ftrace_filter=", set_ftrace_filter); #ifdef CONFIG_FUNCTION_GRAPH_TRACER static char ftrace_graph_buf[FTRACE_FILTER_SIZE] __initdata; static char ftrace_graph_notrace_buf[FTRACE_FILTER_SIZE] __initdata; -static int ftrace_set_func(unsigned long *array, int *idx, int size, char *buffer); +static int ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer); static unsigned long save_global_trampoline; static unsigned long save_global_flags; @@ -4401,26 +4965,38 @@ static int __init set_graph_notrace_function(char *str) } __setup("ftrace_graph_notrace=", set_graph_notrace_function); +static int __init set_graph_max_depth_function(char *str) +{ + if (!str) + return 0; + fgraph_max_depth = simple_strtoul(str, NULL, 0); + return 1; +} +__setup("ftrace_graph_max_depth=", set_graph_max_depth_function); + static void __init set_ftrace_early_graph(char *buf, int enable) { int ret; char *func; - unsigned long *table = ftrace_graph_funcs; - int *count = &ftrace_graph_count; + struct ftrace_hash *hash; - if (!enable) { - table = ftrace_graph_notrace_funcs; - count = &ftrace_graph_notrace_count; - } + hash = alloc_ftrace_hash(FTRACE_HASH_DEFAULT_BITS); + if (WARN_ON(!hash)) + return; while (buf) { func = strsep(&buf, ","); /* we allow only one expression at a time */ - ret = ftrace_set_func(table, count, FTRACE_GRAPH_MAX_FUNCS, func); + ret = ftrace_graph_set_hash(hash, func); if (ret) printk(KERN_DEBUG "ftrace: function %s not " "traceable\n", func); } + + if (enable) + ftrace_graph_hash = hash; + else + ftrace_graph_notrace_hash = hash; } #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ @@ -4454,10 +5030,8 @@ static void __init set_ftrace_early_filters(void) int ftrace_regex_release(struct inode *inode, struct file *file) { struct seq_file *m = (struct seq_file *)file->private_data; - struct ftrace_ops_hash old_hash_ops; struct ftrace_iterator *iter; struct ftrace_hash **orig_hash; - struct ftrace_hash *old_hash; struct trace_parser *parser; int filter_hash; int ret; @@ -4481,22 +5055,20 @@ int ftrace_regex_release(struct inode *inode, struct file *file) if (file->f_mode & FMODE_WRITE) { filter_hash = !!(iter->flags & FTRACE_ITER_FILTER); - if (filter_hash) + if (filter_hash) { orig_hash = &iter->ops->func_hash->filter_hash; - else + if (iter->tr && !list_empty(&iter->tr->mod_trace)) + iter->hash->flags |= FTRACE_HASH_FL_MOD; + } else orig_hash = &iter->ops->func_hash->notrace_hash; mutex_lock(&ftrace_lock); - old_hash = *orig_hash; - old_hash_ops.filter_hash = iter->ops->func_hash->filter_hash; - old_hash_ops.notrace_hash = iter->ops->func_hash->notrace_hash; - ret = ftrace_hash_move(iter->ops, filter_hash, - orig_hash, iter->hash); - if (!ret) { - ftrace_ops_update_code(iter->ops, &old_hash_ops); - free_ftrace_hash_rcu(old_hash); - } + ret = ftrace_hash_move_and_update_ops(iter->ops, orig_hash, + iter->hash, filter_hash); mutex_unlock(&ftrace_lock); + } else { + /* For read only, the hash is the ops hash */ + iter->hash = NULL; } mutex_unlock(&iter->ops->func_hash->regex_lock); @@ -4540,26 +5112,55 @@ static const struct file_operations ftrace_notrace_fops = { static DEFINE_MUTEX(graph_lock); -int ftrace_graph_count; -int ftrace_graph_notrace_count; -unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly; -unsigned long ftrace_graph_notrace_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly; +struct ftrace_hash *ftrace_graph_hash = EMPTY_HASH; +struct ftrace_hash *ftrace_graph_notrace_hash = EMPTY_HASH; + +enum graph_filter_type { + GRAPH_FILTER_NOTRACE = 0, + GRAPH_FILTER_FUNCTION, +}; + +#define FTRACE_GRAPH_EMPTY ((void *)1) struct ftrace_graph_data { - unsigned long *table; - size_t size; - int *count; - const struct seq_operations *seq_ops; + struct ftrace_hash *hash; + struct ftrace_func_entry *entry; + int idx; /* for hash table iteration */ + enum graph_filter_type type; + struct ftrace_hash *new_hash; + const struct seq_operations *seq_ops; + struct trace_parser parser; }; static void * __g_next(struct seq_file *m, loff_t *pos) { struct ftrace_graph_data *fgd = m->private; + struct ftrace_func_entry *entry = fgd->entry; + struct hlist_head *head; + int i, idx = fgd->idx; - if (*pos >= *fgd->count) + if (*pos >= fgd->hash->count) return NULL; - return &fgd->table[*pos]; + + if (entry) { + hlist_for_each_entry_continue(entry, hlist) { + fgd->entry = entry; + return entry; + } + + idx++; + } + + for (i = idx; i < 1 << fgd->hash->size_bits; i++) { + head = &fgd->hash->buckets[i]; + hlist_for_each_entry(entry, head, hlist) { + fgd->entry = entry; + fgd->idx = i; + return entry; + } + } + return NULL; } static void * @@ -4575,10 +5176,19 @@ static void *g_start(struct seq_file *m, loff_t *pos) mutex_lock(&graph_lock); + if (fgd->type == GRAPH_FILTER_FUNCTION) + fgd->hash = rcu_dereference_protected(ftrace_graph_hash, + lockdep_is_held(&graph_lock)); + else + fgd->hash = rcu_dereference_protected(ftrace_graph_notrace_hash, + lockdep_is_held(&graph_lock)); + /* Nothing, tell g_show to print all functions are enabled */ - if (!*fgd->count && !*pos) - return (void *)1; + if (ftrace_hash_empty(fgd->hash) && !*pos) + return FTRACE_GRAPH_EMPTY; + fgd->idx = 0; + fgd->entry = NULL; return __g_next(m, pos); } @@ -4589,22 +5199,22 @@ static void g_stop(struct seq_file *m, void *p) static int g_show(struct seq_file *m, void *v) { - unsigned long *ptr = v; + struct ftrace_func_entry *entry = v; - if (!ptr) + if (!entry) return 0; - if (ptr == (unsigned long *)1) { + if (entry == FTRACE_GRAPH_EMPTY) { struct ftrace_graph_data *fgd = m->private; - if (fgd->table == ftrace_graph_funcs) + if (fgd->type == GRAPH_FILTER_FUNCTION) seq_puts(m, "#### all functions enabled ####\n"); else seq_puts(m, "#### no functions disabled ####\n"); return 0; } - seq_printf(m, "%ps\n", (void *)*ptr); + seq_printf(m, "%ps\n", (void *)entry->ip); return 0; } @@ -4621,24 +5231,51 @@ __ftrace_graph_open(struct inode *inode, struct file *file, struct ftrace_graph_data *fgd) { int ret = 0; + struct ftrace_hash *new_hash = NULL; - mutex_lock(&graph_lock); - if ((file->f_mode & FMODE_WRITE) && - (file->f_flags & O_TRUNC)) { - *fgd->count = 0; - memset(fgd->table, 0, fgd->size * sizeof(*fgd->table)); + if (file->f_mode & FMODE_WRITE) { + const int size_bits = FTRACE_HASH_DEFAULT_BITS; + + if (trace_parser_get_init(&fgd->parser, FTRACE_BUFF_MAX)) + return -ENOMEM; + + if (file->f_flags & O_TRUNC) + new_hash = alloc_ftrace_hash(size_bits); + else + new_hash = alloc_and_copy_ftrace_hash(size_bits, + fgd->hash); + if (!new_hash) { + ret = -ENOMEM; + goto out; + } } - mutex_unlock(&graph_lock); if (file->f_mode & FMODE_READ) { - ret = seq_open(file, fgd->seq_ops); + ret = seq_open(file, &ftrace_graph_seq_ops); if (!ret) { struct seq_file *m = file->private_data; m->private = fgd; + } else { + /* Failed */ + free_ftrace_hash(new_hash); + new_hash = NULL; } } else file->private_data = fgd; +out: + if (ret < 0 && file->f_mode & FMODE_WRITE) + trace_parser_put(&fgd->parser); + + fgd->new_hash = new_hash; + + /* + * All uses of fgd->hash must be taken with the graph_lock + * held. The graph_lock is going to be released, so force + * fgd->hash to be reinitialized when it is taken again. + */ + fgd->hash = NULL; + return ret; } @@ -4646,6 +5283,7 @@ static int ftrace_graph_open(struct inode *inode, struct file *file) { struct ftrace_graph_data *fgd; + int ret; if (unlikely(ftrace_disabled)) return -ENODEV; @@ -4654,18 +5292,26 @@ ftrace_graph_open(struct inode *inode, struct file *file) if (fgd == NULL) return -ENOMEM; - fgd->table = ftrace_graph_funcs; - fgd->size = FTRACE_GRAPH_MAX_FUNCS; - fgd->count = &ftrace_graph_count; + mutex_lock(&graph_lock); + + fgd->hash = rcu_dereference_protected(ftrace_graph_hash, + lockdep_is_held(&graph_lock)); + fgd->type = GRAPH_FILTER_FUNCTION; fgd->seq_ops = &ftrace_graph_seq_ops; - return __ftrace_graph_open(inode, file, fgd); + ret = __ftrace_graph_open(inode, file, fgd); + if (ret < 0) + kfree(fgd); + + mutex_unlock(&graph_lock); + return ret; } static int ftrace_graph_notrace_open(struct inode *inode, struct file *file) { struct ftrace_graph_data *fgd; + int ret; if (unlikely(ftrace_disabled)) return -ENODEV; @@ -4674,45 +5320,97 @@ ftrace_graph_notrace_open(struct inode *inode, struct file *file) if (fgd == NULL) return -ENOMEM; - fgd->table = ftrace_graph_notrace_funcs; - fgd->size = FTRACE_GRAPH_MAX_FUNCS; - fgd->count = &ftrace_graph_notrace_count; + mutex_lock(&graph_lock); + + fgd->hash = rcu_dereference_protected(ftrace_graph_notrace_hash, + lockdep_is_held(&graph_lock)); + fgd->type = GRAPH_FILTER_NOTRACE; fgd->seq_ops = &ftrace_graph_seq_ops; - return __ftrace_graph_open(inode, file, fgd); + ret = __ftrace_graph_open(inode, file, fgd); + if (ret < 0) + kfree(fgd); + + mutex_unlock(&graph_lock); + return ret; } static int ftrace_graph_release(struct inode *inode, struct file *file) { + struct ftrace_graph_data *fgd; + struct ftrace_hash *old_hash, *new_hash; + struct trace_parser *parser; + int ret = 0; + if (file->f_mode & FMODE_READ) { struct seq_file *m = file->private_data; - kfree(m->private); + fgd = m->private; seq_release(inode, file); } else { - kfree(file->private_data); + fgd = file->private_data; } - return 0; + + if (file->f_mode & FMODE_WRITE) { + + parser = &fgd->parser; + + if (trace_parser_loaded((parser))) { + parser->buffer[parser->idx] = 0; + ret = ftrace_graph_set_hash(fgd->new_hash, + parser->buffer); + } + + trace_parser_put(parser); + + new_hash = __ftrace_hash_move(fgd->new_hash); + if (!new_hash) { + ret = -ENOMEM; + goto out; + } + + mutex_lock(&graph_lock); + + if (fgd->type == GRAPH_FILTER_FUNCTION) { + old_hash = rcu_dereference_protected(ftrace_graph_hash, + lockdep_is_held(&graph_lock)); + rcu_assign_pointer(ftrace_graph_hash, new_hash); + } else { + old_hash = rcu_dereference_protected(ftrace_graph_notrace_hash, + lockdep_is_held(&graph_lock)); + rcu_assign_pointer(ftrace_graph_notrace_hash, new_hash); + } + + mutex_unlock(&graph_lock); + + /* Wait till all users are no longer using the old hash */ + synchronize_sched(); + + free_ftrace_hash(old_hash); + } + + out: + free_ftrace_hash(fgd->new_hash); + kfree(fgd); + + return ret; } static int -ftrace_set_func(unsigned long *array, int *idx, int size, char *buffer) +ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer) { struct ftrace_glob func_g; struct dyn_ftrace *rec; struct ftrace_page *pg; + struct ftrace_func_entry *entry; int fail = 1; int not; - bool exists; - int i; /* decode regex */ func_g.type = filter_parse_regex(buffer, strlen(buffer), &func_g.search, ¬); - if (!not && *idx >= size) - return -EBUSY; func_g.len = strlen(func_g.search); @@ -4729,26 +5427,18 @@ ftrace_set_func(unsigned long *array, int *idx, int size, char *buffer) continue; if (ftrace_match_record(rec, &func_g, NULL, 0)) { - /* if it is in the array */ - exists = false; - for (i = 0; i < *idx; i++) { - if (array[i] == rec->ip) { - exists = true; - break; - } - } + entry = ftrace_lookup_ip(hash, rec->ip); if (!not) { fail = 0; - if (!exists) { - array[(*idx)++] = rec->ip; - if (*idx >= size) - goto out; - } + + if (entry) + continue; + if (add_hash_entry(hash, rec->ip) < 0) + goto out; } else { - if (exists) { - array[i] = array[--(*idx)]; - array[*idx] = 0; + if (entry) { + free_hash_entry(hash, entry); fail = 0; } } @@ -4767,35 +5457,34 @@ static ssize_t ftrace_graph_write(struct file *file, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct trace_parser parser; ssize_t read, ret = 0; struct ftrace_graph_data *fgd = file->private_data; + struct trace_parser *parser; if (!cnt) return 0; - if (trace_parser_get_init(&parser, FTRACE_BUFF_MAX)) - return -ENOMEM; - - read = trace_get_user(&parser, ubuf, cnt, ppos); + /* Read mode uses seq functions */ + if (file->f_mode & FMODE_READ) { + struct seq_file *m = file->private_data; + fgd = m->private; + } - if (read >= 0 && trace_parser_loaded((&parser))) { - parser.buffer[parser.idx] = 0; + parser = &fgd->parser; - mutex_lock(&graph_lock); + read = trace_get_user(parser, ubuf, cnt, ppos); - /* we allow only one expression at a time */ - ret = ftrace_set_func(fgd->table, fgd->count, fgd->size, - parser.buffer); + if (read >= 0 && trace_parser_loaded(parser) && + !trace_parser_cont(parser)) { - mutex_unlock(&graph_lock); + ret = ftrace_graph_set_hash(fgd->new_hash, + parser->buffer); + trace_parser_clear(parser); } if (!ret) ret = read; - trace_parser_put(&parser); - return ret; } @@ -5028,6 +5717,7 @@ void ftrace_release_mod(struct module *mod) if (pg == ftrace_pages) ftrace_pages = next_to_ftrace_page(last_pg); + ftrace_update_tot_cnt -= pg->index; *last_pg = pg->next; order = get_count_order(pg->size / ENTRIES_PER_PAGE); free_pages((unsigned long)pg->records, order); @@ -5106,6 +5796,8 @@ void ftrace_module_enable(struct module *mod) out_unlock: mutex_unlock(&ftrace_lock); + + process_cached_mods(mod->name); } void ftrace_module_init(struct module *mod) @@ -5118,6 +5810,51 @@ void ftrace_module_init(struct module *mod) } #endif /* CONFIG_MODULES */ +void __init ftrace_free_init_mem(void) +{ + unsigned long start = (unsigned long)(&__init_begin); + unsigned long end = (unsigned long)(&__init_end); + struct ftrace_page **last_pg = &ftrace_pages_start; + struct ftrace_page *pg; + struct dyn_ftrace *rec; + struct dyn_ftrace key; + int order; + + key.ip = start; + key.flags = end; /* overload flags, as it is unsigned long */ + + mutex_lock(&ftrace_lock); + + for (pg = ftrace_pages_start; pg; last_pg = &pg->next, pg = *last_pg) { + if (end < pg->records[0].ip || + start >= (pg->records[pg->index - 1].ip + MCOUNT_INSN_SIZE)) + continue; + again: + rec = bsearch(&key, pg->records, pg->index, + sizeof(struct dyn_ftrace), + ftrace_cmp_recs); + if (!rec) + continue; + pg->index--; + ftrace_update_tot_cnt--; + if (!pg->index) { + *last_pg = pg->next; + order = get_count_order(pg->size / ENTRIES_PER_PAGE); + free_pages((unsigned long)pg->records, order); + kfree(pg); + pg = container_of(last_pg, struct ftrace_page, next); + if (!(*last_pg)) + ftrace_pages = pg; + continue; + } + memmove(rec, rec + 1, + (pg->index - (rec - pg->records)) * sizeof(*rec)); + /* More than one function may be in this block */ + goto again; + } + mutex_unlock(&ftrace_lock); +} + void __init ftrace_init(void) { extern unsigned long __start_mcount_loc[]; @@ -5160,25 +5897,15 @@ void __weak arch_ftrace_update_trampoline(struct ftrace_ops *ops) static void ftrace_update_trampoline(struct ftrace_ops *ops) { - -/* - * Currently there's no safe way to free a trampoline when the kernel - * is configured with PREEMPT. That is because a task could be preempted - * when it jumped to the trampoline, it may be preempted for a long time - * depending on the system load, and currently there's no way to know - * when it will be off the trampoline. If the trampoline is freed - * too early, when the task runs again, it will be executing on freed - * memory and crash. - */ -#ifdef CONFIG_PREEMPT - /* Currently, only non dynamic ops can have a trampoline */ - if (ops->flags & FTRACE_OPS_FL_DYNAMIC) - return; -#endif - arch_ftrace_update_trampoline(ops); } +void ftrace_init_trace_array(struct trace_array *tr) +{ + INIT_LIST_HEAD(&tr->func_probes); + INIT_LIST_HEAD(&tr->mod_trace); + INIT_LIST_HEAD(&tr->mod_notrace); +} #else static struct ftrace_ops global_ops = { @@ -5233,6 +5960,7 @@ __init void ftrace_init_global_array_ops(struct trace_array *tr) { tr->ops = &global_ops; tr->ops->private = tr; + ftrace_init_trace_array(tr); } void ftrace_init_array_ops(struct trace_array *tr, ftrace_func_t func) @@ -5357,7 +6085,7 @@ static void ftrace_ops_assist_func(unsigned long ip, unsigned long parent_ip, * Normally the mcount trampoline will call the ops->func, but there * are times that it should not. For example, if the ops does not * have its own recursion protection, then it should call the - * ftrace_ops_recurs_func() instead. + * ftrace_ops_assist_func() instead. * * Returns the function that the trampoline should call for @ops. */ @@ -5387,6 +6115,43 @@ ftrace_filter_pid_sched_switch_probe(void *data, bool preempt, trace_ignore_this_task(pid_list, next)); } +static void +ftrace_pid_follow_sched_process_fork(void *data, + struct task_struct *self, + struct task_struct *task) +{ + struct trace_pid_list *pid_list; + struct trace_array *tr = data; + + pid_list = rcu_dereference_sched(tr->function_pids); + trace_filter_add_remove_task(pid_list, self, task); +} + +static void +ftrace_pid_follow_sched_process_exit(void *data, struct task_struct *task) +{ + struct trace_pid_list *pid_list; + struct trace_array *tr = data; + + pid_list = rcu_dereference_sched(tr->function_pids); + trace_filter_add_remove_task(pid_list, NULL, task); +} + +void ftrace_pid_follow_fork(struct trace_array *tr, bool enable) +{ + if (enable) { + register_trace_sched_process_fork(ftrace_pid_follow_sched_process_fork, + tr); + register_trace_sched_process_exit(ftrace_pid_follow_sched_process_exit, + tr); + } else { + unregister_trace_sched_process_fork(ftrace_pid_follow_sched_process_fork, + tr); + unregister_trace_sched_process_exit(ftrace_pid_follow_sched_process_exit, + tr); + } +} + static void clear_ftrace_pids(struct trace_array *tr) { struct trace_pid_list *pid_list; @@ -5410,6 +6175,15 @@ static void clear_ftrace_pids(struct trace_array *tr) trace_free_pid_list(pid_list); } +void ftrace_clear_pids(struct trace_array *tr) +{ + mutex_lock(&ftrace_lock); + + clear_ftrace_pids(tr); + + mutex_unlock(&ftrace_lock); +} + static void ftrace_pid_reset(struct trace_array *tr) { mutex_lock(&ftrace_lock); @@ -5691,7 +6465,8 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, if (ftrace_enabled) { /* we are starting ftrace again */ - if (ftrace_ops_list != &ftrace_list_end) + if (rcu_dereference_protected(ftrace_ops_list, + lockdep_is_held(&ftrace_lock)) != &ftrace_list_end) update_ftrace_function(); ftrace_startup_sysctl(); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index a85739efcc30..529cc50d7243 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -6,6 +6,7 @@ #include <linux/trace_events.h> #include <linux/ring_buffer.h> #include <linux/trace_clock.h> +#include <linux/sched/clock.h> #include <linux/trace_seq.h> #include <linux/spinlock.h> #include <linux/irq_work.h> @@ -437,6 +438,7 @@ struct ring_buffer_per_cpu { raw_spinlock_t reader_lock; /* serialize readers */ arch_spinlock_t lock; struct lock_class_key lock_key; + struct buffer_data_page *free_page; unsigned long nr_pages; unsigned int current_context; struct list_head *pages; @@ -1134,12 +1136,12 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu) for (i = 0; i < nr_pages; i++) { struct page *page; /* - * __GFP_NORETRY flag makes sure that the allocation fails - * gracefully without invoking oom-killer and the system is - * not destabilized. + * __GFP_RETRY_MAYFAIL flag makes sure that the allocation fails + * gracefully without invoking oom-killer and the system is not + * destabilized. */ bpage = kzalloc_node(ALIGN(sizeof(*bpage), cache_line_size()), - GFP_KERNEL | __GFP_NORETRY, + GFP_KERNEL | __GFP_RETRY_MAYFAIL, cpu_to_node(cpu)); if (!bpage) goto free_pages; @@ -1147,7 +1149,7 @@ static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int cpu) list_add(&bpage->list, pages); page = alloc_pages_node(cpu_to_node(cpu), - GFP_KERNEL | __GFP_NORETRY, 0); + GFP_KERNEL | __GFP_RETRY_MAYFAIL, 0); if (!page) goto free_pages; bpage->page = page_address(page); @@ -3404,11 +3406,23 @@ EXPORT_SYMBOL_GPL(ring_buffer_iter_reset); int ring_buffer_iter_empty(struct ring_buffer_iter *iter) { struct ring_buffer_per_cpu *cpu_buffer; + struct buffer_page *reader; + struct buffer_page *head_page; + struct buffer_page *commit_page; + unsigned commit; cpu_buffer = iter->cpu_buffer; - return iter->head_page == cpu_buffer->commit_page && - iter->head == rb_commit_index(cpu_buffer); + /* Remember, trace recording is off when iterator is in use */ + reader = cpu_buffer->reader_page; + head_page = cpu_buffer->head_page; + commit_page = cpu_buffer->commit_page; + commit = rb_page_commit(commit_page); + + return ((iter->head_page == commit_page && iter->head == commit) || + (iter->head_page == reader && commit_page == head_page && + head_page->read == commit && + iter->head == rb_page_commit(cpu_buffer->reader_page))); } EXPORT_SYMBOL_GPL(ring_buffer_iter_empty); @@ -4376,9 +4390,25 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); */ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) { - struct buffer_data_page *bpage; + struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; + struct buffer_data_page *bpage = NULL; + unsigned long flags; struct page *page; + local_irq_save(flags); + arch_spin_lock(&cpu_buffer->lock); + + if (cpu_buffer->free_page) { + bpage = cpu_buffer->free_page; + cpu_buffer->free_page = NULL; + } + + arch_spin_unlock(&cpu_buffer->lock); + local_irq_restore(flags); + + if (bpage) + goto out; + page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); if (!page) @@ -4386,6 +4416,7 @@ void *ring_buffer_alloc_read_page(struct ring_buffer *buffer, int cpu) bpage = page_address(page); + out: rb_init_page(bpage); return bpage; @@ -4395,13 +4426,29 @@ EXPORT_SYMBOL_GPL(ring_buffer_alloc_read_page); /** * ring_buffer_free_read_page - free an allocated read page * @buffer: the buffer the page was allocate for + * @cpu: the cpu buffer the page came from * @data: the page to free * * Free a page allocated from ring_buffer_alloc_read_page. */ -void ring_buffer_free_read_page(struct ring_buffer *buffer, void *data) +void ring_buffer_free_read_page(struct ring_buffer *buffer, int cpu, void *data) { - free_page((unsigned long)data); + struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; + struct buffer_data_page *bpage = data; + unsigned long flags; + + local_irq_save(flags); + arch_spin_lock(&cpu_buffer->lock); + + if (!cpu_buffer->free_page) { + cpu_buffer->free_page = bpage; + bpage = NULL; + } + + arch_spin_unlock(&cpu_buffer->lock); + local_irq_restore(flags); + + free_page((unsigned long)bpage); } EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); @@ -4825,9 +4872,9 @@ static __init int test_ringbuffer(void) rb_data[cpu].cnt = cpu; rb_threads[cpu] = kthread_create(rb_test, &rb_data[cpu], "rbtester/%d", cpu); - if (WARN_ON(!rb_threads[cpu])) { + if (WARN_ON(IS_ERR(rb_threads[cpu]))) { pr_cont("FAILED\n"); - ret = -1; + ret = PTR_ERR(rb_threads[cpu]); goto out_free; } @@ -4837,9 +4884,9 @@ static __init int test_ringbuffer(void) /* Now create the rb hammer! */ rb_hammer = kthread_run(rb_hammer_test, NULL, "rbhammer"); - if (WARN_ON(!rb_hammer)) { + if (WARN_ON(IS_ERR(rb_hammer))) { pr_cont("FAILED\n"); - ret = -1; + ret = PTR_ERR(rb_hammer); goto out_free; } diff --git a/kernel/trace/ring_buffer_benchmark.c b/kernel/trace/ring_buffer_benchmark.c index 6df9a83e20d7..9fbcaf567886 100644 --- a/kernel/trace/ring_buffer_benchmark.c +++ b/kernel/trace/ring_buffer_benchmark.c @@ -6,6 +6,7 @@ #include <linux/ring_buffer.h> #include <linux/completion.h> #include <linux/kthread.h> +#include <uapi/linux/sched/types.h> #include <linux/module.h> #include <linux/ktime.h> #include <asm/local.h> @@ -170,7 +171,7 @@ static enum event_status read_page(int cpu) } } } - ring_buffer_free_read_page(buffer, bpage); + ring_buffer_free_read_page(buffer, cpu, bpage); if (ret < 0) return EVENT_DROPPED; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d7449783987a..42b9355033d4 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -87,7 +87,7 @@ dummy_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set) * tracing is active, only save the comm when a trace event * occurred. */ -static DEFINE_PER_CPU(bool, trace_cmdline_save); +static DEFINE_PER_CPU(bool, trace_taskinfo_save); /* * Kill all tracing for good (never come back). @@ -120,41 +120,41 @@ enum ftrace_dump_mode ftrace_dump_on_oops; /* When set, tracing will stop when a WARN*() is hit */ int __disable_trace_on_warning; -#ifdef CONFIG_TRACE_ENUM_MAP_FILE -/* Map of enums to their values, for "enum_map" file */ -struct trace_enum_map_head { +#ifdef CONFIG_TRACE_EVAL_MAP_FILE +/* Map of enums to their values, for "eval_map" file */ +struct trace_eval_map_head { struct module *mod; unsigned long length; }; -union trace_enum_map_item; +union trace_eval_map_item; -struct trace_enum_map_tail { +struct trace_eval_map_tail { /* * "end" is first and points to NULL as it must be different - * than "mod" or "enum_string" + * than "mod" or "eval_string" */ - union trace_enum_map_item *next; + union trace_eval_map_item *next; const char *end; /* points to NULL */ }; -static DEFINE_MUTEX(trace_enum_mutex); +static DEFINE_MUTEX(trace_eval_mutex); /* - * The trace_enum_maps are saved in an array with two extra elements, + * The trace_eval_maps are saved in an array with two extra elements, * one at the beginning, and one at the end. The beginning item contains * the count of the saved maps (head.length), and the module they * belong to if not built in (head.mod). The ending item contains a - * pointer to the next array of saved enum_map items. + * pointer to the next array of saved eval_map items. */ -union trace_enum_map_item { - struct trace_enum_map map; - struct trace_enum_map_head head; - struct trace_enum_map_tail tail; +union trace_eval_map_item { + struct trace_eval_map map; + struct trace_eval_map_head head; + struct trace_eval_map_tail tail; }; -static union trace_enum_map_item *trace_enum_maps; -#endif /* CONFIG_TRACE_ENUM_MAP_FILE */ +static union trace_eval_map_item *trace_eval_maps; +#endif /* CONFIG_TRACE_EVAL_MAP_FILE */ static int tracing_set_tracer(struct trace_array *tr, const char *buf); @@ -257,19 +257,11 @@ unsigned long long ns2usecs(u64 nsec) /* trace_flags that are default zero for instances */ #define ZEROED_TRACE_FLAGS \ - TRACE_ITER_EVENT_FORK + (TRACE_ITER_EVENT_FORK | TRACE_ITER_FUNC_FORK) /* - * The global_trace is the descriptor that holds the tracing - * buffers for the live tracing. For each CPU, it contains - * a link list of pages that will store trace entries. The - * page descriptor of the pages in the memory is used to hold - * the link list by linking the lru item in the page descriptor - * to each of the pages in the buffer per CPU. - * - * For each active CPU there is a data field that holds the - * pages for the buffer for that CPU. Each CPU has the same number - * of pages allocated for its buffer. + * The global_trace is the descriptor that holds the top-level tracing + * buffers for the live tracing. */ static struct trace_array global_trace = { .trace_flags = TRACE_DEFAULT_FLAGS, @@ -765,7 +757,7 @@ __trace_buffer_lock_reserve(struct ring_buffer *buffer, return event; } -static void tracer_tracing_on(struct trace_array *tr) +void tracer_tracing_on(struct trace_array *tr) { if (tr->trace_buffer.buffer) ring_buffer_record_on(tr->trace_buffer.buffer); @@ -798,7 +790,7 @@ EXPORT_SYMBOL_GPL(tracing_on); static __always_inline void __buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *event) { - __this_cpu_write(trace_cmdline_save, true); + __this_cpu_write(trace_taskinfo_save, true); /* If this is the temp buffer, we need to commit fully */ if (this_cpu_read(trace_buffered_event) == event) { @@ -902,23 +894,8 @@ int __trace_bputs(unsigned long ip, const char *str) EXPORT_SYMBOL_GPL(__trace_bputs); #ifdef CONFIG_TRACER_SNAPSHOT -/** - * trace_snapshot - take a snapshot of the current buffer. - * - * This causes a swap between the snapshot buffer and the current live - * tracing buffer. You can use this to take snapshots of the live - * trace when some condition is triggered, but continue to trace. - * - * Note, make sure to allocate the snapshot with either - * a tracing_snapshot_alloc(), or by doing it manually - * with: echo 1 > /sys/kernel/debug/tracing/snapshot - * - * If the snapshot buffer is not allocated, it will stop tracing. - * Basically making a permanent snapshot. - */ -void tracing_snapshot(void) +static void tracing_snapshot_instance(struct trace_array *tr) { - struct trace_array *tr = &global_trace; struct tracer *tracer = tr->current_trace; unsigned long flags; @@ -946,6 +923,27 @@ void tracing_snapshot(void) update_max_tr(tr, current, smp_processor_id()); local_irq_restore(flags); } + +/** + * trace_snapshot - take a snapshot of the current buffer. + * + * This causes a swap between the snapshot buffer and the current live + * tracing buffer. You can use this to take snapshots of the live + * trace when some condition is triggered, but continue to trace. + * + * Note, make sure to allocate the snapshot with either + * a tracing_snapshot_alloc(), or by doing it manually + * with: echo 1 > /sys/kernel/debug/tracing/snapshot + * + * If the snapshot buffer is not allocated, it will stop tracing. + * Basically making a permanent snapshot. + */ +void tracing_snapshot(void) +{ + struct trace_array *tr = &global_trace; + + tracing_snapshot_instance(tr); +} EXPORT_SYMBOL_GPL(tracing_snapshot); static int resize_buffer_duplicate_size(struct trace_buffer *trace_buf, @@ -1047,7 +1045,7 @@ void tracing_snapshot_alloc(void) EXPORT_SYMBOL_GPL(tracing_snapshot_alloc); #endif /* CONFIG_TRACER_SNAPSHOT */ -static void tracer_tracing_off(struct trace_array *tr) +void tracer_tracing_off(struct trace_array *tr) { if (tr->trace_buffer.buffer) ring_buffer_record_off(tr->trace_buffer.buffer); @@ -1143,9 +1141,9 @@ unsigned long nsecs_to_usecs(unsigned long nsecs) /* * TRACE_FLAGS is defined as a tuple matching bit masks with strings. - * It uses C(a, b) where 'a' is the enum name and 'b' is the string that + * It uses C(a, b) where 'a' is the eval (enum) name and 'b' is the string that * matches it. By defining "C(a, b) b", TRACE_FLAGS becomes a list - * of strings in the order that the enums were defined. + * of strings in the order that the evals (enum) were defined. */ #undef C #define C(a, b) b @@ -1193,6 +1191,7 @@ int trace_parser_get_init(struct trace_parser *parser, int size) void trace_parser_put(struct trace_parser *parser) { kfree(parser->buffer); + parser->buffer = NULL; } /* @@ -1431,6 +1430,28 @@ static int wait_on_pipe(struct trace_iterator *iter, bool full) } #ifdef CONFIG_FTRACE_STARTUP_TEST +static bool selftests_can_run; + +struct trace_selftests { + struct list_head list; + struct tracer *type; +}; + +static LIST_HEAD(postponed_selftests); + +static int save_selftest(struct tracer *type) +{ + struct trace_selftests *selftest; + + selftest = kmalloc(sizeof(*selftest), GFP_KERNEL); + if (!selftest) + return -ENOMEM; + + selftest->type = type; + list_add(&selftest->list, &postponed_selftests); + return 0; +} + static int run_tracer_selftest(struct tracer *type) { struct trace_array *tr = &global_trace; @@ -1441,6 +1462,14 @@ static int run_tracer_selftest(struct tracer *type) return 0; /* + * If a tracer registers early in boot up (before scheduling is + * initialized and such), then do not run its selftests yet. + * Instead, run it a little later in the boot process. + */ + if (!selftests_can_run) + return save_selftest(type); + + /* * Run a selftest on this tracer. * Here we reset the trace buffer, and set the current * tracer to be this tracer. The tracer can then run some @@ -1489,6 +1518,47 @@ static int run_tracer_selftest(struct tracer *type) printk(KERN_CONT "PASSED\n"); return 0; } + +static __init int init_trace_selftests(void) +{ + struct trace_selftests *p, *n; + struct tracer *t, **last; + int ret; + + selftests_can_run = true; + + mutex_lock(&trace_types_lock); + + if (list_empty(&postponed_selftests)) + goto out; + + pr_info("Running postponed tracer tests:\n"); + + list_for_each_entry_safe(p, n, &postponed_selftests, list) { + ret = run_tracer_selftest(p->type); + /* If the test fails, then warn and remove from available_tracers */ + if (ret < 0) { + WARN(1, "tracer: %s failed selftest, disabling\n", + p->type->name); + last = &trace_types; + for (t = trace_types; t; t = t->next) { + if (t == p->type) { + *last = t->next; + break; + } + last = &t->next; + } + } + list_del(&p->list); + kfree(p); + } + + out: + mutex_unlock(&trace_types_lock); + + return 0; +} +core_initcall(init_trace_selftests); #else static inline int run_tracer_selftest(struct tracer *type) { @@ -1639,6 +1709,8 @@ void tracing_reset_all_online_cpus(void) } } +static int *tgid_map; + #define SAVED_CMDLINES_DEFAULT 128 #define NO_CMDLINE_MAP UINT_MAX static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; @@ -1652,7 +1724,7 @@ struct saved_cmdlines_buffer { static struct saved_cmdlines_buffer *savedcmd; /* temporary disable recording */ -static atomic_t trace_record_cmdline_disabled __read_mostly; +static atomic_t trace_record_taskinfo_disabled __read_mostly; static inline char *get_saved_cmdlines(int idx) { @@ -1840,13 +1912,15 @@ static void tracing_stop_tr(struct trace_array *tr) raw_spin_unlock_irqrestore(&tr->start_lock, flags); } -void trace_stop_cmdline_recording(void); - static int trace_save_cmdline(struct task_struct *tsk) { unsigned pid, idx; - if (!tsk->pid || unlikely(tsk->pid > PID_MAX_DEFAULT)) + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + if (unlikely(tsk->pid > PID_MAX_DEFAULT)) return 0; /* @@ -1906,7 +1980,7 @@ static void __trace_find_cmdline(int pid, char comm[]) map = savedcmd->map_pid_to_cmdline[pid]; if (map != NO_CMDLINE_MAP) - strcpy(comm, get_saved_cmdlines(map)); + strlcpy(comm, get_saved_cmdlines(map), TASK_COMM_LEN); else strcpy(comm, "<...>"); } @@ -1922,18 +1996,121 @@ void trace_find_cmdline(int pid, char comm[]) preempt_enable(); } -void tracing_record_cmdline(struct task_struct *tsk) +int trace_find_tgid(int pid) +{ + if (unlikely(!tgid_map || !pid || pid > PID_MAX_DEFAULT)) + return 0; + + return tgid_map[pid]; +} + +static int trace_save_tgid(struct task_struct *tsk) +{ + /* treat recording of idle task as a success */ + if (!tsk->pid) + return 1; + + if (unlikely(!tgid_map || tsk->pid > PID_MAX_DEFAULT)) + return 0; + + tgid_map[tsk->pid] = tsk->tgid; + return 1; +} + +static bool tracing_record_taskinfo_skip(int flags) +{ + if (unlikely(!(flags & (TRACE_RECORD_CMDLINE | TRACE_RECORD_TGID)))) + return true; + if (atomic_read(&trace_record_taskinfo_disabled) || !tracing_is_on()) + return true; + if (!__this_cpu_read(trace_taskinfo_save)) + return true; + return false; +} + +/** + * tracing_record_taskinfo - record the task info of a task + * + * @task - task to record + * @flags - TRACE_RECORD_CMDLINE for recording comm + * - TRACE_RECORD_TGID for recording tgid + */ +void tracing_record_taskinfo(struct task_struct *task, int flags) +{ + bool done; + + if (tracing_record_taskinfo_skip(flags)) + return; + + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(task); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(task); + + /* If recording any information failed, retry again soon. */ + if (!done) + return; + + __this_cpu_write(trace_taskinfo_save, false); +} + +/** + * tracing_record_taskinfo_sched_switch - record task info for sched_switch + * + * @prev - previous task during sched_switch + * @next - next task during sched_switch + * @flags - TRACE_RECORD_CMDLINE for recording comm + * TRACE_RECORD_TGID for recording tgid + */ +void tracing_record_taskinfo_sched_switch(struct task_struct *prev, + struct task_struct *next, int flags) { - if (atomic_read(&trace_record_cmdline_disabled) || !tracing_is_on()) + bool done; + + if (tracing_record_taskinfo_skip(flags)) return; - if (!__this_cpu_read(trace_cmdline_save)) + /* + * Record as much task information as possible. If some fail, continue + * to try to record the others. + */ + done = !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(prev); + done &= !(flags & TRACE_RECORD_CMDLINE) || trace_save_cmdline(next); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(prev); + done &= !(flags & TRACE_RECORD_TGID) || trace_save_tgid(next); + + /* If recording any information failed, retry again soon. */ + if (!done) return; - if (trace_save_cmdline(tsk)) - __this_cpu_write(trace_cmdline_save, false); + __this_cpu_write(trace_taskinfo_save, false); +} + +/* Helpers to record a specific task information */ +void tracing_record_cmdline(struct task_struct *task) +{ + tracing_record_taskinfo(task, TRACE_RECORD_CMDLINE); } +void tracing_record_tgid(struct task_struct *task) +{ + tracing_record_taskinfo(task, TRACE_RECORD_TGID); +} + +/* + * Several functions return TRACE_TYPE_PARTIAL_LINE if the trace_seq + * overflowed, and TRACE_TYPE_HANDLED otherwise. This helper function + * simplifies those functions and keeps them in sync. + */ +enum print_line_t trace_handle_return(struct trace_seq *s) +{ + return trace_seq_has_overflowed(s) ? + TRACE_TYPE_PARTIAL_LINE : TRACE_TYPE_HANDLED; +} +EXPORT_SYMBOL_GPL(trace_handle_return); + void tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags, int pc) @@ -2486,7 +2663,36 @@ static inline void ftrace_trace_stack(struct trace_array *tr, void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, int pc) { - __ftrace_trace_stack(tr->trace_buffer.buffer, flags, skip, pc, NULL); + struct ring_buffer *buffer = tr->trace_buffer.buffer; + + if (rcu_is_watching()) { + __ftrace_trace_stack(buffer, flags, skip, pc, NULL); + return; + } + + /* + * When an NMI triggers, RCU is enabled via rcu_nmi_enter(), + * but if the above rcu_is_watching() failed, then the NMI + * triggered someplace critical, and rcu_irq_enter() should + * not be called from NMI. + */ + if (unlikely(in_nmi())) + return; + + /* + * It is possible that a function is being traced in a + * location that RCU is not watching. A call to + * rcu_irq_enter() will make sure that it is, but there's + * a few internal rcu functions that could be traced + * where that wont work either. In those cases, we just + * do nothing. + */ + if (unlikely(rcu_irq_enter_disabled())) + return; + + rcu_irq_enter_irqson(); + __ftrace_trace_stack(buffer, flags, skip, pc, NULL); + rcu_irq_exit_irqson(); } /** @@ -3035,7 +3241,7 @@ static void *s_start(struct seq_file *m, loff_t *pos) #endif if (!iter->snapshot) - atomic_inc(&trace_record_cmdline_disabled); + atomic_inc(&trace_record_taskinfo_disabled); if (*pos != iter->pos) { iter->ent = NULL; @@ -3080,7 +3286,7 @@ static void s_stop(struct seq_file *m, void *p) #endif if (!iter->snapshot) - atomic_dec(&trace_record_cmdline_disabled); + atomic_dec(&trace_record_taskinfo_disabled); trace_access_unlock(iter->cpu_file); trace_event_read_unlock(); @@ -3137,23 +3343,38 @@ static void print_event_info(struct trace_buffer *buf, struct seq_file *m) seq_puts(m, "#\n"); } -static void print_func_help_header(struct trace_buffer *buf, struct seq_file *m) +static void print_func_help_header(struct trace_buffer *buf, struct seq_file *m, + unsigned int flags) { + bool tgid = flags & TRACE_ITER_RECORD_TGID; + print_event_info(buf, m); - seq_puts(m, "# TASK-PID CPU# TIMESTAMP FUNCTION\n" - "# | | | | |\n"); + + seq_printf(m, "# TASK-PID CPU# %s TIMESTAMP FUNCTION\n", tgid ? "TGID " : ""); + seq_printf(m, "# | | | %s | |\n", tgid ? " | " : ""); } -static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file *m) +static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file *m, + unsigned int flags) { - print_event_info(buf, m); - seq_puts(m, "# _-----=> irqs-off\n" - "# / _----=> need-resched\n" - "# | / _---=> hardirq/softirq\n" - "# || / _--=> preempt-depth\n" - "# ||| / delay\n" - "# TASK-PID CPU# |||| TIMESTAMP FUNCTION\n" - "# | | | |||| | |\n"); + bool tgid = flags & TRACE_ITER_RECORD_TGID; + const char tgid_space[] = " "; + const char space[] = " "; + + seq_printf(m, "# %s _-----=> irqs-off\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s / _----=> need-resched\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s| / _---=> hardirq/softirq\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s|| / _--=> preempt-depth\n", + tgid ? tgid_space : space); + seq_printf(m, "# %s||| / delay\n", + tgid ? tgid_space : space); + seq_printf(m, "# TASK-PID CPU#%s|||| TIMESTAMP FUNCTION\n", + tgid ? " TGID " : space); + seq_printf(m, "# | | | %s|||| | |\n", + tgid ? " | " : space); } void @@ -3229,13 +3450,14 @@ static void test_cpu_buff_start(struct trace_iterator *iter) if (!(iter->iter_flags & TRACE_FILE_ANNOTATE)) return; - if (iter->started && cpumask_test_cpu(iter->cpu, iter->started)) + if (cpumask_available(iter->started) && + cpumask_test_cpu(iter->cpu, iter->started)) return; if (per_cpu_ptr(iter->trace_buffer->data, iter->cpu)->skipped_entries) return; - if (iter->started) + if (cpumask_available(iter->started)) cpumask_set_cpu(iter->cpu, iter->started); /* Don't print started cpu buffer for the first entry of the trace */ @@ -3468,9 +3690,11 @@ void trace_default_header(struct seq_file *m) } else { if (!(trace_flags & TRACE_ITER_VERBOSE)) { if (trace_flags & TRACE_ITER_IRQ_INFO) - print_func_help_header_irq(iter->trace_buffer, m); + print_func_help_header_irq(iter->trace_buffer, + m, trace_flags); else - print_func_help_header(iter->trace_buffer, m); + print_func_help_header(iter->trace_buffer, m, + trace_flags); } } } @@ -4126,9 +4350,24 @@ int set_tracer_flag(struct trace_array *tr, unsigned int mask, int enabled) if (mask == TRACE_ITER_RECORD_CMD) trace_event_enable_cmd_record(enabled); + if (mask == TRACE_ITER_RECORD_TGID) { + if (!tgid_map) + tgid_map = kzalloc((PID_MAX_DEFAULT + 1) * sizeof(*tgid_map), + GFP_KERNEL); + if (!tgid_map) { + tr->trace_flags &= ~TRACE_ITER_RECORD_TGID; + return -ENOMEM; + } + + trace_event_enable_tgid_record(enabled); + } + if (mask == TRACE_ITER_EVENT_FORK) trace_event_follow_fork(tr, enabled); + if (mask == TRACE_ITER_FUNC_FORK) + ftrace_pid_follow_fork(tr, enabled); + if (mask == TRACE_ITER_OVERWRITE) { ring_buffer_change_overwrite(tr->trace_buffer.buffer, enabled); #ifdef CONFIG_TRACER_MAX_TRACE @@ -4348,22 +4587,24 @@ static const char readme_msg[] = "\t\t\t traces\n" #endif #endif /* CONFIG_STACK_TRACER */ -#ifdef CONFIG_KPROBE_EVENT +#ifdef CONFIG_KPROBE_EVENTS " kprobe_events\t\t- Add/remove/show the kernel dynamic events\n" "\t\t\t Write into this file to define/undefine new trace events.\n" #endif -#ifdef CONFIG_UPROBE_EVENT +#ifdef CONFIG_UPROBE_EVENTS " uprobe_events\t\t- Add/remove/show the userspace dynamic events\n" "\t\t\t Write into this file to define/undefine new trace events.\n" #endif -#if defined(CONFIG_KPROBE_EVENT) || defined(CONFIG_UPROBE_EVENT) +#if defined(CONFIG_KPROBE_EVENTS) || defined(CONFIG_UPROBE_EVENTS) "\t accepts: event-definitions (one definition per line)\n" - "\t Format: p|r[:[<group>/]<event>] <place> [<args>]\n" + "\t Format: p[:[<group>/]<event>] <place> [<args>]\n" + "\t r[maxactive][:[<group>/]<event>] <place> [<args>]\n" "\t -:[<group>/]<event>\n" -#ifdef CONFIG_KPROBE_EVENT +#ifdef CONFIG_KPROBE_EVENTS "\t place: [<module>:]<symbol>[+<offset>]|<memaddr>\n" + "place (kretprobe): [<module>:]<symbol>[+<offset>]|<memaddr>\n" #endif -#ifdef CONFIG_UPROBE_EVENT +#ifdef CONFIG_UPROBE_EVENTS "\t place: <path>:<offset>\n" #endif "\t args: <name>=fetcharg[:type]\n" @@ -4481,6 +4722,76 @@ static const struct file_operations tracing_readme_fops = { .llseek = generic_file_llseek, }; +static void *saved_tgids_next(struct seq_file *m, void *v, loff_t *pos) +{ + int *ptr = v; + + if (*pos || m->count) + ptr++; + + (*pos)++; + + for (; ptr <= &tgid_map[PID_MAX_DEFAULT]; ptr++) { + if (trace_find_tgid(*ptr)) + return ptr; + } + + return NULL; +} + +static void *saved_tgids_start(struct seq_file *m, loff_t *pos) +{ + void *v; + loff_t l = 0; + + if (!tgid_map) + return NULL; + + v = &tgid_map[0]; + while (l <= *pos) { + v = saved_tgids_next(m, v, &l); + if (!v) + return NULL; + } + + return v; +} + +static void saved_tgids_stop(struct seq_file *m, void *v) +{ +} + +static int saved_tgids_show(struct seq_file *m, void *v) +{ + int pid = (int *)v - tgid_map; + + seq_printf(m, "%d %d\n", pid, trace_find_tgid(pid)); + return 0; +} + +static const struct seq_operations tracing_saved_tgids_seq_ops = { + .start = saved_tgids_start, + .stop = saved_tgids_stop, + .next = saved_tgids_next, + .show = saved_tgids_show, +}; + +static int tracing_saved_tgids_open(struct inode *inode, struct file *filp) +{ + if (tracing_disabled) + return -ENODEV; + + return seq_open(filp, &tracing_saved_tgids_seq_ops); +} + + +static const struct file_operations tracing_saved_tgids_fops = { + .open = tracing_saved_tgids_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + static void *saved_cmdlines_next(struct seq_file *m, void *v, loff_t *pos) { unsigned int *ptr = v; @@ -4630,11 +4941,11 @@ static const struct file_operations tracing_saved_cmdlines_size_fops = { .write = tracing_saved_cmdlines_size_write, }; -#ifdef CONFIG_TRACE_ENUM_MAP_FILE -static union trace_enum_map_item * -update_enum_map(union trace_enum_map_item *ptr) +#ifdef CONFIG_TRACE_EVAL_MAP_FILE +static union trace_eval_map_item * +update_eval_map(union trace_eval_map_item *ptr) { - if (!ptr->map.enum_string) { + if (!ptr->map.eval_string) { if (ptr->tail.next) { ptr = ptr->tail.next; /* Set ptr to the next real item (skip head) */ @@ -4645,15 +4956,15 @@ update_enum_map(union trace_enum_map_item *ptr) return ptr; } -static void *enum_map_next(struct seq_file *m, void *v, loff_t *pos) +static void *eval_map_next(struct seq_file *m, void *v, loff_t *pos) { - union trace_enum_map_item *ptr = v; + union trace_eval_map_item *ptr = v; /* * Paranoid! If ptr points to end, we don't want to increment past it. * This really should never happen. */ - ptr = update_enum_map(ptr); + ptr = update_eval_map(ptr); if (WARN_ON_ONCE(!ptr)) return NULL; @@ -4661,104 +4972,104 @@ static void *enum_map_next(struct seq_file *m, void *v, loff_t *pos) (*pos)++; - ptr = update_enum_map(ptr); + ptr = update_eval_map(ptr); return ptr; } -static void *enum_map_start(struct seq_file *m, loff_t *pos) +static void *eval_map_start(struct seq_file *m, loff_t *pos) { - union trace_enum_map_item *v; + union trace_eval_map_item *v; loff_t l = 0; - mutex_lock(&trace_enum_mutex); + mutex_lock(&trace_eval_mutex); - v = trace_enum_maps; + v = trace_eval_maps; if (v) v++; while (v && l < *pos) { - v = enum_map_next(m, v, &l); + v = eval_map_next(m, v, &l); } return v; } -static void enum_map_stop(struct seq_file *m, void *v) +static void eval_map_stop(struct seq_file *m, void *v) { - mutex_unlock(&trace_enum_mutex); + mutex_unlock(&trace_eval_mutex); } -static int enum_map_show(struct seq_file *m, void *v) +static int eval_map_show(struct seq_file *m, void *v) { - union trace_enum_map_item *ptr = v; + union trace_eval_map_item *ptr = v; seq_printf(m, "%s %ld (%s)\n", - ptr->map.enum_string, ptr->map.enum_value, + ptr->map.eval_string, ptr->map.eval_value, ptr->map.system); return 0; } -static const struct seq_operations tracing_enum_map_seq_ops = { - .start = enum_map_start, - .next = enum_map_next, - .stop = enum_map_stop, - .show = enum_map_show, +static const struct seq_operations tracing_eval_map_seq_ops = { + .start = eval_map_start, + .next = eval_map_next, + .stop = eval_map_stop, + .show = eval_map_show, }; -static int tracing_enum_map_open(struct inode *inode, struct file *filp) +static int tracing_eval_map_open(struct inode *inode, struct file *filp) { if (tracing_disabled) return -ENODEV; - return seq_open(filp, &tracing_enum_map_seq_ops); + return seq_open(filp, &tracing_eval_map_seq_ops); } -static const struct file_operations tracing_enum_map_fops = { - .open = tracing_enum_map_open, +static const struct file_operations tracing_eval_map_fops = { + .open = tracing_eval_map_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, }; -static inline union trace_enum_map_item * -trace_enum_jmp_to_tail(union trace_enum_map_item *ptr) +static inline union trace_eval_map_item * +trace_eval_jmp_to_tail(union trace_eval_map_item *ptr) { /* Return tail of array given the head */ return ptr + ptr->head.length + 1; } static void -trace_insert_enum_map_file(struct module *mod, struct trace_enum_map **start, +trace_insert_eval_map_file(struct module *mod, struct trace_eval_map **start, int len) { - struct trace_enum_map **stop; - struct trace_enum_map **map; - union trace_enum_map_item *map_array; - union trace_enum_map_item *ptr; + struct trace_eval_map **stop; + struct trace_eval_map **map; + union trace_eval_map_item *map_array; + union trace_eval_map_item *ptr; stop = start + len; /* - * The trace_enum_maps contains the map plus a head and tail item, + * The trace_eval_maps contains the map plus a head and tail item, * where the head holds the module and length of array, and the * tail holds a pointer to the next list. */ map_array = kmalloc(sizeof(*map_array) * (len + 2), GFP_KERNEL); if (!map_array) { - pr_warn("Unable to allocate trace enum mapping\n"); + pr_warn("Unable to allocate trace eval mapping\n"); return; } - mutex_lock(&trace_enum_mutex); + mutex_lock(&trace_eval_mutex); - if (!trace_enum_maps) - trace_enum_maps = map_array; + if (!trace_eval_maps) + trace_eval_maps = map_array; else { - ptr = trace_enum_maps; + ptr = trace_eval_maps; for (;;) { - ptr = trace_enum_jmp_to_tail(ptr); + ptr = trace_eval_jmp_to_tail(ptr); if (!ptr->tail.next) break; ptr = ptr->tail.next; @@ -4776,34 +5087,34 @@ trace_insert_enum_map_file(struct module *mod, struct trace_enum_map **start, } memset(map_array, 0, sizeof(*map_array)); - mutex_unlock(&trace_enum_mutex); + mutex_unlock(&trace_eval_mutex); } -static void trace_create_enum_file(struct dentry *d_tracer) +static void trace_create_eval_file(struct dentry *d_tracer) { - trace_create_file("enum_map", 0444, d_tracer, - NULL, &tracing_enum_map_fops); + trace_create_file("eval_map", 0444, d_tracer, + NULL, &tracing_eval_map_fops); } -#else /* CONFIG_TRACE_ENUM_MAP_FILE */ -static inline void trace_create_enum_file(struct dentry *d_tracer) { } -static inline void trace_insert_enum_map_file(struct module *mod, - struct trace_enum_map **start, int len) { } -#endif /* !CONFIG_TRACE_ENUM_MAP_FILE */ +#else /* CONFIG_TRACE_EVAL_MAP_FILE */ +static inline void trace_create_eval_file(struct dentry *d_tracer) { } +static inline void trace_insert_eval_map_file(struct module *mod, + struct trace_eval_map **start, int len) { } +#endif /* !CONFIG_TRACE_EVAL_MAP_FILE */ -static void trace_insert_enum_map(struct module *mod, - struct trace_enum_map **start, int len) +static void trace_insert_eval_map(struct module *mod, + struct trace_eval_map **start, int len) { - struct trace_enum_map **map; + struct trace_eval_map **map; if (len <= 0) return; map = start; - trace_event_enum_update(map, len); + trace_event_eval_update(map, len); - trace_insert_enum_map_file(mod, start, len); + trace_insert_eval_map_file(mod, start, len); } static ssize_t @@ -5536,7 +5847,6 @@ static ssize_t tracing_splice_read_pipe(struct file *filp, .partial = partial_def, .nr_pages = 0, /* This gets updated below. */ .nr_pages_max = PIPE_DEF_BUFFERS, - .flags = flags, .ops = &tracing_pipe_buf_ops, .spd_release = tracing_spd_release_pipe, }; @@ -5969,6 +6279,7 @@ static int tracing_clock_open(struct inode *inode, struct file *file) struct ftrace_buffer_info { struct trace_iterator iter; void *spare; + unsigned int spare_cpu; unsigned int read; }; @@ -6298,9 +6609,11 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, return -EBUSY; #endif - if (!info->spare) + if (!info->spare) { info->spare = ring_buffer_alloc_read_page(iter->trace_buffer->buffer, iter->cpu_file); + info->spare_cpu = iter->cpu_file; + } if (!info->spare) return -ENOMEM; @@ -6360,7 +6673,8 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) __trace_array_put(iter->tr); if (info->spare) - ring_buffer_free_read_page(iter->trace_buffer->buffer, info->spare); + ring_buffer_free_read_page(iter->trace_buffer->buffer, + info->spare_cpu, info->spare); kfree(info); mutex_unlock(&trace_types_lock); @@ -6371,6 +6685,7 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) struct buffer_ref { struct ring_buffer *buffer; void *page; + int cpu; int ref; }; @@ -6382,7 +6697,7 @@ static void buffer_pipe_buf_release(struct pipe_inode_info *pipe, if (--ref->ref) return; - ring_buffer_free_read_page(ref->buffer, ref->page); + ring_buffer_free_read_page(ref->buffer, ref->cpu, ref->page); kfree(ref); buf->private = 0; } @@ -6416,7 +6731,7 @@ static void buffer_spd_release(struct splice_pipe_desc *spd, unsigned int i) if (--ref->ref) return; - ring_buffer_free_read_page(ref->buffer, ref->page); + ring_buffer_free_read_page(ref->buffer, ref->cpu, ref->page); kfree(ref); spd->partial[i].private = 0; } @@ -6434,7 +6749,6 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, .pages = pages_def, .partial = partial_def, .nr_pages_max = PIPE_DEF_BUFFERS, - .flags = flags, .ops = &buffer_pipe_buf_ops, .spd_release = buffer_spd_release, }; @@ -6481,11 +6795,13 @@ tracing_buffers_splice_read(struct file *file, loff_t *ppos, kfree(ref); break; } + ref->cpu = iter->cpu_file; r = ring_buffer_read_page(ref->buffer, &ref->page, len, iter->cpu_file, 1); if (r < 0) { - ring_buffer_free_read_page(ref->buffer, ref->page); + ring_buffer_free_read_page(ref->buffer, ref->cpu, + ref->page); kfree(ref); break; } @@ -6618,33 +6934,18 @@ static const struct file_operations tracing_stats_fops = { #ifdef CONFIG_DYNAMIC_FTRACE -int __weak ftrace_arch_read_dyn_info(char *buf, int size) -{ - return 0; -} - static ssize_t tracing_read_dyn_info(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - static char ftrace_dyn_info_buffer[1024]; - static DEFINE_MUTEX(dyn_info_mutex); unsigned long *p = filp->private_data; - char *buf = ftrace_dyn_info_buffer; - int size = ARRAY_SIZE(ftrace_dyn_info_buffer); + char buf[64]; /* Not too big for a shallow stack */ int r; - mutex_lock(&dyn_info_mutex); - r = sprintf(buf, "%ld ", *p); - - r += ftrace_arch_read_dyn_info(buf+r, (size-1)-r); + r = scnprintf(buf, 63, "%ld", *p); buf[r++] = '\n'; - r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); - - mutex_unlock(&dyn_info_mutex); - - return r; + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } static const struct file_operations tracing_dyn_info_fops = { @@ -6656,43 +6957,89 @@ static const struct file_operations tracing_dyn_info_fops = { #if defined(CONFIG_TRACER_SNAPSHOT) && defined(CONFIG_DYNAMIC_FTRACE) static void -ftrace_snapshot(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_snapshot(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - tracing_snapshot(); + tracing_snapshot_instance(tr); } static void -ftrace_count_snapshot(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_count_snapshot(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - unsigned long *count = (long *)data; + struct ftrace_func_mapper *mapper = data; + long *count = NULL; - if (!*count) - return; + if (mapper) + count = (long *)ftrace_func_mapper_find_ip(mapper, ip); + + if (count) { + + if (*count <= 0) + return; - if (*count != -1) (*count)--; + } - tracing_snapshot(); + tracing_snapshot_instance(tr); } static int ftrace_snapshot_print(struct seq_file *m, unsigned long ip, struct ftrace_probe_ops *ops, void *data) { - long count = (long)data; + struct ftrace_func_mapper *mapper = data; + long *count = NULL; seq_printf(m, "%ps:", (void *)ip); seq_puts(m, "snapshot"); - if (count == -1) - seq_puts(m, ":unlimited\n"); + if (mapper) + count = (long *)ftrace_func_mapper_find_ip(mapper, ip); + + if (count) + seq_printf(m, ":count=%ld\n", *count); else - seq_printf(m, ":count=%ld\n", count); + seq_puts(m, ":unlimited\n"); return 0; } +static int +ftrace_snapshot_init(struct ftrace_probe_ops *ops, struct trace_array *tr, + unsigned long ip, void *init_data, void **data) +{ + struct ftrace_func_mapper *mapper = *data; + + if (!mapper) { + mapper = allocate_ftrace_func_mapper(); + if (!mapper) + return -ENOMEM; + *data = mapper; + } + + return ftrace_func_mapper_add_ip(mapper, ip, init_data); +} + +static void +ftrace_snapshot_free(struct ftrace_probe_ops *ops, struct trace_array *tr, + unsigned long ip, void *data) +{ + struct ftrace_func_mapper *mapper = data; + + if (!ip) { + if (!mapper) + return; + free_ftrace_func_mapper(mapper, NULL); + return; + } + + ftrace_func_mapper_remove_ip(mapper, ip); +} + static struct ftrace_probe_ops snapshot_probe_ops = { .func = ftrace_snapshot, .print = ftrace_snapshot_print, @@ -6701,10 +7048,12 @@ static struct ftrace_probe_ops snapshot_probe_ops = { static struct ftrace_probe_ops snapshot_count_probe_ops = { .func = ftrace_count_snapshot, .print = ftrace_snapshot_print, + .init = ftrace_snapshot_init, + .free = ftrace_snapshot_free, }; static int -ftrace_trace_snapshot_callback(struct ftrace_hash *hash, +ftrace_trace_snapshot_callback(struct trace_array *tr, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enable) { struct ftrace_probe_ops *ops; @@ -6712,16 +7061,17 @@ ftrace_trace_snapshot_callback(struct ftrace_hash *hash, char *number; int ret; + if (!tr) + return -ENODEV; + /* hash funcs only work with set_ftrace_filter */ if (!enable) return -EINVAL; ops = param ? &snapshot_count_probe_ops : &snapshot_probe_ops; - if (glob[0] == '!') { - unregister_ftrace_function_probe_func(glob+1, ops); - return 0; - } + if (glob[0] == '!') + return unregister_ftrace_function_probe_func(glob+1, tr, ops); if (!param) goto out_reg; @@ -6740,11 +7090,13 @@ ftrace_trace_snapshot_callback(struct ftrace_hash *hash, return ret; out_reg: - ret = register_ftrace_function_probe(glob, ops, count); + ret = alloc_snapshot(tr); + if (ret < 0) + goto out; - if (ret >= 0) - alloc_snapshot(&global_trace); + ret = register_ftrace_function_probe(glob, tr, ops, count); + out: return ret < 0 ? ret : 0; } @@ -7353,6 +7705,8 @@ static int instance_mkdir(const char *name) goto out_free_tr; } + ftrace_init_trace_array(tr); + init_tracer_tracefs(tr, tr->dir); init_trace_flags_index(tr); __update_tracer_options(tr); @@ -7408,7 +7762,9 @@ static int instance_rmdir(const char *name) } tracing_set_nop(tr); + clear_ftrace_function_probes(tr); event_trace_del_tracer(tr); + ftrace_clear_pids(tr); ftrace_destroy_function_files(tr); tracefs_remove_recursive(tr->dir); free_trace_buffers(tr); @@ -7418,6 +7774,7 @@ static int instance_rmdir(const char *name) } kfree(tr->topts); + free_cpumask_var(tr->tracing_cpumask); kfree(tr->name); kfree(tr); @@ -7503,7 +7860,7 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer) ftrace_init_tracefs(tr, d_tracer); } -static struct vfsmount *trace_automount(void *ingore) +static struct vfsmount *trace_automount(struct dentry *mntpt, void *ingore) { struct vfsmount *mnt; struct file_system_type *type; @@ -7516,7 +7873,7 @@ static struct vfsmount *trace_automount(void *ingore) type = get_fs_type("tracefs"); if (!type) return NULL; - mnt = vfs_kern_mount(type, 0, "tracefs", NULL); + mnt = vfs_submount(mntpt, type, "tracefs", NULL); put_filesystem(type); if (IS_ERR(mnt)) return NULL; @@ -7561,21 +7918,21 @@ struct dentry *tracing_init_dentry(void) return NULL; } -extern struct trace_enum_map *__start_ftrace_enum_maps[]; -extern struct trace_enum_map *__stop_ftrace_enum_maps[]; +extern struct trace_eval_map *__start_ftrace_eval_maps[]; +extern struct trace_eval_map *__stop_ftrace_eval_maps[]; -static void __init trace_enum_init(void) +static void __init trace_eval_init(void) { int len; - len = __stop_ftrace_enum_maps - __start_ftrace_enum_maps; - trace_insert_enum_map(NULL, __start_ftrace_enum_maps, len); + len = __stop_ftrace_eval_maps - __start_ftrace_eval_maps; + trace_insert_eval_map(NULL, __start_ftrace_eval_maps, len); } #ifdef CONFIG_MODULES -static void trace_module_add_enums(struct module *mod) +static void trace_module_add_evals(struct module *mod) { - if (!mod->num_trace_enums) + if (!mod->num_trace_evals) return; /* @@ -7585,40 +7942,40 @@ static void trace_module_add_enums(struct module *mod) if (trace_module_has_bad_taint(mod)) return; - trace_insert_enum_map(mod, mod->trace_enums, mod->num_trace_enums); + trace_insert_eval_map(mod, mod->trace_evals, mod->num_trace_evals); } -#ifdef CONFIG_TRACE_ENUM_MAP_FILE -static void trace_module_remove_enums(struct module *mod) +#ifdef CONFIG_TRACE_EVAL_MAP_FILE +static void trace_module_remove_evals(struct module *mod) { - union trace_enum_map_item *map; - union trace_enum_map_item **last = &trace_enum_maps; + union trace_eval_map_item *map; + union trace_eval_map_item **last = &trace_eval_maps; - if (!mod->num_trace_enums) + if (!mod->num_trace_evals) return; - mutex_lock(&trace_enum_mutex); + mutex_lock(&trace_eval_mutex); - map = trace_enum_maps; + map = trace_eval_maps; while (map) { if (map->head.mod == mod) break; - map = trace_enum_jmp_to_tail(map); + map = trace_eval_jmp_to_tail(map); last = &map->tail.next; map = map->tail.next; } if (!map) goto out; - *last = trace_enum_jmp_to_tail(map)->tail.next; + *last = trace_eval_jmp_to_tail(map)->tail.next; kfree(map); out: - mutex_unlock(&trace_enum_mutex); + mutex_unlock(&trace_eval_mutex); } #else -static inline void trace_module_remove_enums(struct module *mod) { } -#endif /* CONFIG_TRACE_ENUM_MAP_FILE */ +static inline void trace_module_remove_evals(struct module *mod) { } +#endif /* CONFIG_TRACE_EVAL_MAP_FILE */ static int trace_module_notify(struct notifier_block *self, unsigned long val, void *data) @@ -7627,10 +7984,10 @@ static int trace_module_notify(struct notifier_block *self, switch (val) { case MODULE_STATE_COMING: - trace_module_add_enums(mod); + trace_module_add_evals(mod); break; case MODULE_STATE_GOING: - trace_module_remove_enums(mod); + trace_module_remove_evals(mod); break; } @@ -7668,9 +8025,12 @@ static __init int tracer_init_tracefs(void) trace_create_file("saved_cmdlines_size", 0644, d_tracer, NULL, &tracing_saved_cmdlines_size_fops); - trace_enum_init(); + trace_create_file("saved_tgids", 0444, d_tracer, + NULL, &tracing_saved_tgids_fops); - trace_create_enum_file(d_tracer); + trace_eval_init(); + + trace_create_eval_file(d_tracer); #ifdef CONFIG_MODULES register_module_notifier(&trace_module_nb); @@ -7972,6 +8332,9 @@ __init static int tracer_alloc_buffers(void) register_tracer(&nop_trace); + /* Function tracing may start here (via kernel command line) */ + init_function_trace(); + /* All seems OK, enable tracing */ tracing_disabled = 0; @@ -8006,7 +8369,7 @@ out: return ret; } -void __init trace_init(void) +void __init early_trace_init(void) { if (tracepoint_printk) { tracepoint_print_iter = @@ -8017,6 +8380,10 @@ void __init trace_init(void) static_key_enable(&tracepoint_printk_key.key); } tracer_alloc_buffers(); +} + +void __init trace_init(void) +{ trace_event_init(); } diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 1ea51ab53edf..490ba229931d 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -262,6 +262,12 @@ struct trace_array { #ifdef CONFIG_FUNCTION_TRACER struct ftrace_ops *ops; struct trace_pid_list __rcu *function_pids; +#ifdef CONFIG_DYNAMIC_FTRACE + /* All of these are protected by the ftrace_lock */ + struct list_head func_probes; + struct list_head mod_trace; + struct list_head mod_notrace; +#endif /* function tracing enabled */ int function_enabled; #endif @@ -579,6 +585,8 @@ void tracing_reset_all_online_cpus(void); int tracing_open_generic(struct inode *inode, struct file *filp); bool tracing_is_disabled(void); int tracer_tracing_is_on(struct trace_array *tr); +void tracer_tracing_on(struct trace_array *tr); +void tracer_tracing_off(struct trace_array *tr); struct dentry *trace_create_file(const char *name, umode_t mode, struct dentry *parent, @@ -632,6 +640,9 @@ void set_graph_array(struct trace_array *tr); void tracing_start_cmdline_record(void); void tracing_stop_cmdline_record(void); +void tracing_start_tgid_record(void); +void tracing_stop_tgid_record(void); + int register_tracer(struct tracer *type); int is_tracing_stopped(void); @@ -692,10 +703,14 @@ static inline void __trace_stack(struct trace_array *tr, unsigned long flags, extern u64 ftrace_now(int cpu); extern void trace_find_cmdline(int pid, char comm[]); +extern int trace_find_tgid(int pid); extern void trace_event_follow_fork(struct trace_array *tr, bool enable); #ifdef CONFIG_DYNAMIC_FTRACE extern unsigned long ftrace_update_tot_cnt; +void ftrace_init_trace_array(struct trace_array *tr); +#else +static inline void ftrace_init_trace_array(struct trace_array *tr) { } #endif #define DYN_FTRACE_TEST_NAME trace_selftest_dynamic_test_func extern int DYN_FTRACE_TEST_NAME(void); @@ -753,6 +768,35 @@ enum print_line_t print_trace_line(struct trace_iterator *iter); extern char trace_find_mark(unsigned long long duration); +struct ftrace_hash; + +struct ftrace_mod_load { + struct list_head list; + char *func; + char *module; + int enable; +}; + +enum { + FTRACE_HASH_FL_MOD = (1 << 0), +}; + +struct ftrace_hash { + unsigned long size_bits; + struct hlist_head *buckets; + unsigned long count; + unsigned long flags; + struct rcu_head rcu; +}; + +struct ftrace_func_entry * +ftrace_lookup_ip(struct ftrace_hash *hash, unsigned long ip); + +static __always_inline bool ftrace_hash_empty(struct ftrace_hash *hash) +{ + return !hash || !(hash->count || (hash->flags & FTRACE_HASH_FL_MOD)); +} + /* Standard output formatting function used for function return traces */ #ifdef CONFIG_FUNCTION_GRAPH_TRACER @@ -787,53 +831,50 @@ extern void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, unsigned long flags, int pc); - #ifdef CONFIG_DYNAMIC_FTRACE -/* TODO: make this variable */ -#define FTRACE_GRAPH_MAX_FUNCS 32 -extern int ftrace_graph_count; -extern unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS]; -extern int ftrace_graph_notrace_count; -extern unsigned long ftrace_graph_notrace_funcs[FTRACE_GRAPH_MAX_FUNCS]; +extern struct ftrace_hash *ftrace_graph_hash; +extern struct ftrace_hash *ftrace_graph_notrace_hash; static inline int ftrace_graph_addr(unsigned long addr) { - int i; - - if (!ftrace_graph_count) - return 1; - - for (i = 0; i < ftrace_graph_count; i++) { - if (addr == ftrace_graph_funcs[i]) { - /* - * If no irqs are to be traced, but a set_graph_function - * is set, and called by an interrupt handler, we still - * want to trace it. - */ - if (in_irq()) - trace_recursion_set(TRACE_IRQ_BIT); - else - trace_recursion_clear(TRACE_IRQ_BIT); - return 1; - } + int ret = 0; + + preempt_disable_notrace(); + + if (ftrace_hash_empty(ftrace_graph_hash)) { + ret = 1; + goto out; } - return 0; + if (ftrace_lookup_ip(ftrace_graph_hash, addr)) { + /* + * If no irqs are to be traced, but a set_graph_function + * is set, and called by an interrupt handler, we still + * want to trace it. + */ + if (in_irq()) + trace_recursion_set(TRACE_IRQ_BIT); + else + trace_recursion_clear(TRACE_IRQ_BIT); + ret = 1; + } + +out: + preempt_enable_notrace(); + return ret; } static inline int ftrace_graph_notrace_addr(unsigned long addr) { - int i; + int ret = 0; - if (!ftrace_graph_notrace_count) - return 0; + preempt_disable_notrace(); - for (i = 0; i < ftrace_graph_notrace_count; i++) { - if (addr == ftrace_graph_notrace_funcs[i]) - return 1; - } + if (ftrace_lookup_ip(ftrace_graph_notrace_hash, addr)) + ret = 1; - return 0; + preempt_enable_notrace(); + return ret; } #else static inline int ftrace_graph_addr(unsigned long addr) @@ -868,6 +909,14 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags) extern struct list_head ftrace_pids; #ifdef CONFIG_FUNCTION_TRACER +struct ftrace_func_command { + struct list_head list; + char *name; + int (*func)(struct trace_array *tr, + struct ftrace_hash *hash, + char *func, char *cmd, + char *params, int enable); +}; extern bool ftrace_filter_param __initdata; static inline int ftrace_trace_task(struct trace_array *tr) { @@ -884,6 +933,9 @@ int using_ftrace_ops_list_func(void); void ftrace_init_tracefs(struct trace_array *tr, struct dentry *d_tracer); void ftrace_init_tracefs_toplevel(struct trace_array *tr, struct dentry *d_tracer); +void ftrace_clear_pids(struct trace_array *tr); +int init_function_trace(void); +void ftrace_pid_follow_fork(struct trace_array *tr, bool enable); #else static inline int ftrace_trace_task(struct trace_array *tr) { @@ -902,15 +954,76 @@ ftrace_init_global_array_ops(struct trace_array *tr) { } static inline void ftrace_reset_array_ops(struct trace_array *tr) { } static inline void ftrace_init_tracefs(struct trace_array *tr, struct dentry *d) { } static inline void ftrace_init_tracefs_toplevel(struct trace_array *tr, struct dentry *d) { } +static inline void ftrace_clear_pids(struct trace_array *tr) { } +static inline int init_function_trace(void) { return 0; } +static inline void ftrace_pid_follow_fork(struct trace_array *tr, bool enable) { } /* ftace_func_t type is not defined, use macro instead of static inline */ #define ftrace_init_array_ops(tr, func) do { } while (0) #endif /* CONFIG_FUNCTION_TRACER */ #if defined(CONFIG_FUNCTION_TRACER) && defined(CONFIG_DYNAMIC_FTRACE) + +struct ftrace_probe_ops { + void (*func)(unsigned long ip, + unsigned long parent_ip, + struct trace_array *tr, + struct ftrace_probe_ops *ops, + void *data); + int (*init)(struct ftrace_probe_ops *ops, + struct trace_array *tr, + unsigned long ip, void *init_data, + void **data); + void (*free)(struct ftrace_probe_ops *ops, + struct trace_array *tr, + unsigned long ip, void *data); + int (*print)(struct seq_file *m, + unsigned long ip, + struct ftrace_probe_ops *ops, + void *data); +}; + +struct ftrace_func_mapper; +typedef int (*ftrace_mapper_func)(void *data); + +struct ftrace_func_mapper *allocate_ftrace_func_mapper(void); +void **ftrace_func_mapper_find_ip(struct ftrace_func_mapper *mapper, + unsigned long ip); +int ftrace_func_mapper_add_ip(struct ftrace_func_mapper *mapper, + unsigned long ip, void *data); +void *ftrace_func_mapper_remove_ip(struct ftrace_func_mapper *mapper, + unsigned long ip); +void free_ftrace_func_mapper(struct ftrace_func_mapper *mapper, + ftrace_mapper_func free_func); + +extern int +register_ftrace_function_probe(char *glob, struct trace_array *tr, + struct ftrace_probe_ops *ops, void *data); +extern int +unregister_ftrace_function_probe_func(char *glob, struct trace_array *tr, + struct ftrace_probe_ops *ops); +extern void clear_ftrace_function_probes(struct trace_array *tr); + +int register_ftrace_command(struct ftrace_func_command *cmd); +int unregister_ftrace_command(struct ftrace_func_command *cmd); + void ftrace_create_filter_files(struct ftrace_ops *ops, struct dentry *parent); void ftrace_destroy_filter_files(struct ftrace_ops *ops); #else +struct ftrace_func_command; + +static inline __init int register_ftrace_command(struct ftrace_func_command *cmd) +{ + return -EINVAL; +} +static inline __init int unregister_ftrace_command(char *cmd_name) +{ + return -EINVAL; +} +static inline void clear_ftrace_function_probes(struct trace_array *tr) +{ +} + /* * The ops parameter passed in is usually undefined. * This must be a macro. @@ -975,11 +1088,13 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, #ifdef CONFIG_FUNCTION_TRACER # define FUNCTION_FLAGS \ - C(FUNCTION, "function-trace"), + C(FUNCTION, "function-trace"), \ + C(FUNC_FORK, "function-fork"), # define FUNCTION_DEFAULT_FLAGS TRACE_ITER_FUNCTION #else # define FUNCTION_FLAGS # define FUNCTION_DEFAULT_FLAGS 0UL +# define TRACE_ITER_FUNC_FORK 0UL #endif #ifdef CONFIG_STACKTRACE @@ -1013,6 +1128,7 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, C(CONTEXT_INFO, "context-info"), /* Print pid/cpu/time */ \ C(LATENCY_FMT, "latency-format"), \ C(RECORD_CMD, "record-cmd"), \ + C(RECORD_TGID, "record-tgid"), \ C(OVERWRITE, "overwrite"), \ C(STOP_ON_FREE, "disable_on_free"), \ C(IRQ_INFO, "irq-info"), \ @@ -1094,9 +1210,9 @@ struct ftrace_event_field { struct event_filter { int n_preds; /* Number assigned */ int a_preds; /* allocated */ - struct filter_pred *preds; - struct filter_pred *root; - char *filter_string; + struct filter_pred __rcu *preds; + struct filter_pred __rcu *root; + char *filter_string; }; struct event_subsystem { @@ -1300,7 +1416,8 @@ static inline bool is_string_field(struct ftrace_event_field *field) { return field->filter_type == FILTER_DYN_STRING || field->filter_type == FILTER_STATIC_STRING || - field->filter_type == FILTER_PTR_STRING; + field->filter_type == FILTER_PTR_STRING || + field->filter_type == FILTER_COMM; } static inline bool is_function_field(struct ftrace_event_field *field) @@ -1328,6 +1445,8 @@ struct ftrace_event_field * trace_find_event_field(struct trace_event_call *call, char *name); extern void trace_event_enable_cmd_record(bool enable); +extern void trace_event_enable_tgid_record(bool enable); + extern int event_trace_add_tracer(struct dentry *parent, struct trace_array *tr); extern int event_trace_del_tracer(struct trace_array *tr); @@ -1678,10 +1797,10 @@ static inline const char *get_syscall_name(int syscall) #ifdef CONFIG_EVENT_TRACING void trace_event_init(void); -void trace_event_enum_update(struct trace_enum_map **map, int len); +void trace_event_eval_update(struct trace_eval_map **map, int len); #else static inline void __init trace_event_init(void) { } -static inline void trace_event_enum_update(struct trace_enum_map **map, int len) { } +static inline void trace_event_eval_update(struct trace_eval_map **map, int len) { } #endif extern struct trace_iterator *tracepoint_print_iter; diff --git a/kernel/trace/trace_benchmark.c b/kernel/trace/trace_benchmark.c index e3b488825ae3..16a8cf02eee9 100644 --- a/kernel/trace/trace_benchmark.c +++ b/kernel/trace/trace_benchmark.c @@ -153,10 +153,18 @@ static int benchmark_event_kthread(void *arg) trace_do_benchmark(); /* - * We don't go to sleep, but let others - * run as well. + * We don't go to sleep, but let others run as well. + * This is bascially a "yield()" to let any task that + * wants to run, schedule in, but if the CPU is idle, + * we'll keep burning cycles. + * + * Note the _rcu_qs() version of cond_resched() will + * notify synchronize_rcu_tasks() that this thread has + * passed a quiescent state for rcu_tasks. Otherwise + * this thread will never voluntarily schedule which would + * block synchronize_rcu_tasks() indefinitely. */ - cond_resched(); + cond_resched_rcu_qs(); } return 0; @@ -175,9 +183,9 @@ int trace_benchmark_reg(void) bm_event_thread = kthread_run(benchmark_event_kthread, NULL, "event_benchmark"); - if (!bm_event_thread) { + if (IS_ERR(bm_event_thread)) { pr_warning("trace benchmark failed to create kernel thread\n"); - return -ENOMEM; + return PTR_ERR(bm_event_thread); } return 0; diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index 75489de546b6..4d8fdf3184dc 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -27,7 +27,7 @@ static DEFINE_MUTEX(branch_tracing_mutex); static struct trace_array *branch_tracer; static void -probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) +probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) { struct trace_event_call *call = &event_branch; struct trace_array *tr = branch_tracer; @@ -68,16 +68,17 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) entry = ring_buffer_event_data(event); /* Strip off the path, only save the file */ - p = f->file + strlen(f->file); - while (p >= f->file && *p != '/') + p = f->data.file + strlen(f->data.file); + while (p >= f->data.file && *p != '/') p--; p++; - strncpy(entry->func, f->func, TRACE_FUNC_SIZE); + strncpy(entry->func, f->data.func, TRACE_FUNC_SIZE); strncpy(entry->file, p, TRACE_FILE_SIZE); entry->func[TRACE_FUNC_SIZE] = 0; entry->file[TRACE_FILE_SIZE] = 0; - entry->line = f->line; + entry->constant = f->constant; + entry->line = f->data.line; entry->correct = val == expect; if (!call_filter_check_discard(call, entry, buffer, event)) @@ -89,7 +90,7 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) } static inline -void trace_likely_condition(struct ftrace_branch_data *f, int val, int expect) +void trace_likely_condition(struct ftrace_likely_data *f, int val, int expect) { if (!branch_tracing_enabled) return; @@ -195,13 +196,19 @@ core_initcall(init_branch_tracer); #else static inline -void trace_likely_condition(struct ftrace_branch_data *f, int val, int expect) +void trace_likely_condition(struct ftrace_likely_data *f, int val, int expect) { } #endif /* CONFIG_BRANCH_TRACER */ -void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect) +void ftrace_likely_update(struct ftrace_likely_data *f, int val, + int expect, int is_constant) { + /* A constant is always correct */ + if (is_constant) { + f->constant++; + val = expect; + } /* * I would love to have a trace point here instead, but the * trace point code is so inundated with unlikely and likely @@ -212,9 +219,9 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect) /* FIXME: Make this atomic! */ if (val == expect) - f->correct++; + f->data.correct++; else - f->incorrect++; + f->data.incorrect++; } EXPORT_SYMBOL(ftrace_likely_update); @@ -245,29 +252,60 @@ static inline long get_incorrect_percent(struct ftrace_branch_data *p) return percent; } -static int branch_stat_show(struct seq_file *m, void *v) +static const char *branch_stat_process_file(struct ftrace_branch_data *p) { - struct ftrace_branch_data *p = v; const char *f; - long percent; /* Only print the file, not the path */ f = p->file + strlen(p->file); while (f >= p->file && *f != '/') f--; - f++; + return ++f; +} + +static void branch_stat_show(struct seq_file *m, + struct ftrace_branch_data *p, const char *f) +{ + long percent; /* * The miss is overlayed on correct, and hit on incorrect. */ percent = get_incorrect_percent(p); - seq_printf(m, "%8lu %8lu ", p->correct, p->incorrect); if (percent < 0) seq_puts(m, " X "); else seq_printf(m, "%3ld ", percent); + seq_printf(m, "%-30.30s %-20.20s %d\n", p->func, f, p->line); +} + +static int branch_stat_show_normal(struct seq_file *m, + struct ftrace_branch_data *p, const char *f) +{ + seq_printf(m, "%8lu %8lu ", p->correct, p->incorrect); + branch_stat_show(m, p, f); + return 0; +} + +static int annotate_branch_stat_show(struct seq_file *m, void *v) +{ + struct ftrace_likely_data *p = v; + const char *f; + int l; + + f = branch_stat_process_file(&p->data); + + if (!p->constant) + return branch_stat_show_normal(m, &p->data, f); + + l = snprintf(NULL, 0, "/%lu", p->constant); + l = l > 8 ? 0 : 8 - l; + + seq_printf(m, "%8lu/%lu %*lu ", + p->data.correct, p->constant, l, p->data.incorrect); + branch_stat_show(m, &p->data, f); return 0; } @@ -279,7 +317,7 @@ static void *annotated_branch_stat_start(struct tracer_stat *trace) static void * annotated_branch_stat_next(void *v, int idx) { - struct ftrace_branch_data *p = v; + struct ftrace_likely_data *p = v; ++p; @@ -328,7 +366,7 @@ static struct tracer_stat annotated_branch_stats = { .stat_next = annotated_branch_stat_next, .stat_cmp = annotated_branch_stat_cmp, .stat_headers = annotated_branch_stat_headers, - .stat_show = branch_stat_show + .stat_show = annotate_branch_stat_show }; __init static int init_annotated_branch_stats(void) @@ -379,12 +417,21 @@ all_branch_stat_next(void *v, int idx) return p; } +static int all_branch_stat_show(struct seq_file *m, void *v) +{ + struct ftrace_branch_data *p = v; + const char *f; + + f = branch_stat_process_file(p); + return branch_stat_show_normal(m, p, f); +} + static struct tracer_stat all_branch_stats = { .name = "branch_all", .stat_start = all_branch_stat_start, .stat_next = all_branch_stat_next, .stat_headers = all_branch_stat_headers, - .stat_show = branch_stat_show + .stat_show = all_branch_stat_show }; __init static int all_annotated_branch_stats(void) diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 0f06532a755b..5fdc779f411d 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -18,6 +18,7 @@ #include <linux/module.h> #include <linux/percpu.h> #include <linux/sched.h> +#include <linux/sched/clock.h> #include <linux/ktime.h> #include <linux/trace_clock.h> diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h index eb7396b7e7c3..adcdbbeae010 100644 --- a/kernel/trace/trace_entries.h +++ b/kernel/trace/trace_entries.h @@ -328,11 +328,13 @@ FTRACE_ENTRY(branch, trace_branch, __array( char, func, TRACE_FUNC_SIZE+1 ) __array( char, file, TRACE_FILE_SIZE+1 ) __field( char, correct ) + __field( char, constant ) ), - F_printk("%u:%s:%s (%u)", + F_printk("%u:%s:%s (%u)%s", __entry->line, - __entry->func, __entry->file, __entry->correct), + __entry->func, __entry->file, __entry->correct, + __entry->constant ? " CONSTANT" : ""), FILTER_OTHER ); @@ -346,14 +348,14 @@ FTRACE_ENTRY(hwlat, hwlat_entry, __field( u64, duration ) __field( u64, outer_duration ) __field( u64, nmi_total_ts ) - __field_struct( struct timespec, timestamp ) - __field_desc( long, timestamp, tv_sec ) + __field_struct( struct timespec64, timestamp ) + __field_desc( s64, timestamp, tv_sec ) __field_desc( long, timestamp, tv_nsec ) __field( unsigned int, nmi_count ) __field( unsigned int, seqnum ) ), - F_printk("cnt:%u\tts:%010lu.%010lu\tinner:%llu\touter:%llunmi-ts:%llu\tnmi-count:%u\n", + F_printk("cnt:%u\tts:%010llu.%010lu\tinner:%llu\touter:%llunmi-ts:%llu\tnmi-count:%u\n", __entry->seqnum, __entry->tv_sec, __entry->tv_nsec, diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 93116549a284..36132f9280e6 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -343,6 +343,28 @@ void trace_event_enable_cmd_record(bool enable) mutex_unlock(&event_mutex); } +void trace_event_enable_tgid_record(bool enable) +{ + struct trace_event_file *file; + struct trace_array *tr; + + mutex_lock(&event_mutex); + do_for_each_event_file(tr, file) { + if (!(file->flags & EVENT_FILE_FL_ENABLED)) + continue; + + if (enable) { + tracing_start_tgid_record(); + set_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, &file->flags); + } else { + tracing_stop_tgid_record(); + clear_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, + &file->flags); + } + } while_for_each_event_file(); + mutex_unlock(&event_mutex); +} + static int __ftrace_event_enable_disable(struct trace_event_file *file, int enable, int soft_disable) { @@ -381,6 +403,12 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, tracing_stop_cmdline_record(); clear_bit(EVENT_FILE_FL_RECORDED_CMD_BIT, &file->flags); } + + if (file->flags & EVENT_FILE_FL_RECORDED_TGID) { + tracing_stop_tgid_record(); + clear_bit(EVENT_FILE_FL_RECORDED_CMD_BIT, &file->flags); + } + call->class->reg(call, TRACE_REG_UNREGISTER, file); } /* If in SOFT_MODE, just set the SOFT_DISABLE_BIT, else clear it */ @@ -407,18 +435,30 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, } if (!(file->flags & EVENT_FILE_FL_ENABLED)) { + bool cmd = false, tgid = false; /* Keep the event disabled, when going to SOFT_MODE. */ if (soft_disable) set_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &file->flags); if (tr->trace_flags & TRACE_ITER_RECORD_CMD) { + cmd = true; tracing_start_cmdline_record(); set_bit(EVENT_FILE_FL_RECORDED_CMD_BIT, &file->flags); } + + if (tr->trace_flags & TRACE_ITER_RECORD_TGID) { + tgid = true; + tracing_start_tgid_record(); + set_bit(EVENT_FILE_FL_RECORDED_TGID_BIT, &file->flags); + } + ret = call->class->reg(call, TRACE_REG_REGISTER, file); if (ret) { - tracing_stop_cmdline_record(); + if (cmd) + tracing_stop_cmdline_record(); + if (tgid) + tracing_stop_tgid_record(); pr_info("event trace: Could not enable event " "%s\n", trace_event_name(call)); break; @@ -2067,18 +2107,18 @@ __register_event(struct trace_event_call *call, struct module *mod) return 0; } -static char *enum_replace(char *ptr, struct trace_enum_map *map, int len) +static char *eval_replace(char *ptr, struct trace_eval_map *map, int len) { int rlen; int elen; - /* Find the length of the enum value as a string */ - elen = snprintf(ptr, 0, "%ld", map->enum_value); + /* Find the length of the eval value as a string */ + elen = snprintf(ptr, 0, "%ld", map->eval_value); /* Make sure there's enough room to replace the string with the value */ if (len < elen) return NULL; - snprintf(ptr, elen + 1, "%ld", map->enum_value); + snprintf(ptr, elen + 1, "%ld", map->eval_value); /* Get the rest of the string of ptr */ rlen = strlen(ptr + len); @@ -2090,11 +2130,11 @@ static char *enum_replace(char *ptr, struct trace_enum_map *map, int len) } static void update_event_printk(struct trace_event_call *call, - struct trace_enum_map *map) + struct trace_eval_map *map) { char *ptr; int quote = 0; - int len = strlen(map->enum_string); + int len = strlen(map->eval_string); for (ptr = call->print_fmt; *ptr; ptr++) { if (*ptr == '\\') { @@ -2125,16 +2165,16 @@ static void update_event_printk(struct trace_event_call *call, continue; } if (isalpha(*ptr) || *ptr == '_') { - if (strncmp(map->enum_string, ptr, len) == 0 && + if (strncmp(map->eval_string, ptr, len) == 0 && !isalnum(ptr[len]) && ptr[len] != '_') { - ptr = enum_replace(ptr, map, len); - /* Hmm, enum string smaller than value */ + ptr = eval_replace(ptr, map, len); + /* enum/sizeof string smaller than value */ if (WARN_ON_ONCE(!ptr)) return; /* - * No need to decrement here, as enum_replace() + * No need to decrement here, as eval_replace() * returns the pointer to the character passed - * the enum, and two enums can not be placed + * the eval, and two evals can not be placed * back to back without something in between. * We can skip that something in between. */ @@ -2165,7 +2205,7 @@ static void update_event_printk(struct trace_event_call *call, } } -void trace_event_enum_update(struct trace_enum_map **map, int len) +void trace_event_eval_update(struct trace_eval_map **map, int len) { struct trace_event_call *call, *p; const char *last_system = NULL; @@ -2460,15 +2500,8 @@ struct event_probe_data { bool enable; }; -static void -event_enable_probe(unsigned long ip, unsigned long parent_ip, void **_data) +static void update_event_probe(struct event_probe_data *data) { - struct event_probe_data **pdata = (struct event_probe_data **)_data; - struct event_probe_data *data = *pdata; - - if (!data) - return; - if (data->enable) clear_bit(EVENT_FILE_FL_SOFT_DISABLED_BIT, &data->file->flags); else @@ -2476,77 +2509,141 @@ event_enable_probe(unsigned long ip, unsigned long parent_ip, void **_data) } static void -event_enable_count_probe(unsigned long ip, unsigned long parent_ip, void **_data) +event_enable_probe(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - struct event_probe_data **pdata = (struct event_probe_data **)_data; - struct event_probe_data *data = *pdata; + struct ftrace_func_mapper *mapper = data; + struct event_probe_data *edata; + void **pdata; - if (!data) + pdata = ftrace_func_mapper_find_ip(mapper, ip); + if (!pdata || !*pdata) return; - if (!data->count) + edata = *pdata; + update_event_probe(edata); +} + +static void +event_enable_count_probe(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) +{ + struct ftrace_func_mapper *mapper = data; + struct event_probe_data *edata; + void **pdata; + + pdata = ftrace_func_mapper_find_ip(mapper, ip); + if (!pdata || !*pdata) + return; + + edata = *pdata; + + if (!edata->count) return; /* Skip if the event is in a state we want to switch to */ - if (data->enable == !(data->file->flags & EVENT_FILE_FL_SOFT_DISABLED)) + if (edata->enable == !(edata->file->flags & EVENT_FILE_FL_SOFT_DISABLED)) return; - if (data->count != -1) - (data->count)--; + if (edata->count != -1) + (edata->count)--; - event_enable_probe(ip, parent_ip, _data); + update_event_probe(edata); } static int event_enable_print(struct seq_file *m, unsigned long ip, - struct ftrace_probe_ops *ops, void *_data) + struct ftrace_probe_ops *ops, void *data) { - struct event_probe_data *data = _data; + struct ftrace_func_mapper *mapper = data; + struct event_probe_data *edata; + void **pdata; + + pdata = ftrace_func_mapper_find_ip(mapper, ip); + + if (WARN_ON_ONCE(!pdata || !*pdata)) + return 0; + + edata = *pdata; seq_printf(m, "%ps:", (void *)ip); seq_printf(m, "%s:%s:%s", - data->enable ? ENABLE_EVENT_STR : DISABLE_EVENT_STR, - data->file->event_call->class->system, - trace_event_name(data->file->event_call)); + edata->enable ? ENABLE_EVENT_STR : DISABLE_EVENT_STR, + edata->file->event_call->class->system, + trace_event_name(edata->file->event_call)); - if (data->count == -1) + if (edata->count == -1) seq_puts(m, ":unlimited\n"); else - seq_printf(m, ":count=%ld\n", data->count); + seq_printf(m, ":count=%ld\n", edata->count); return 0; } static int -event_enable_init(struct ftrace_probe_ops *ops, unsigned long ip, - void **_data) +event_enable_init(struct ftrace_probe_ops *ops, struct trace_array *tr, + unsigned long ip, void *init_data, void **data) { - struct event_probe_data **pdata = (struct event_probe_data **)_data; - struct event_probe_data *data = *pdata; + struct ftrace_func_mapper *mapper = *data; + struct event_probe_data *edata = init_data; + int ret; + + if (!mapper) { + mapper = allocate_ftrace_func_mapper(); + if (!mapper) + return -ENODEV; + *data = mapper; + } + + ret = ftrace_func_mapper_add_ip(mapper, ip, edata); + if (ret < 0) + return ret; + + edata->ref++; - data->ref++; + return 0; +} + +static int free_probe_data(void *data) +{ + struct event_probe_data *edata = data; + + edata->ref--; + if (!edata->ref) { + /* Remove the SOFT_MODE flag */ + __ftrace_event_enable_disable(edata->file, 0, 1); + module_put(edata->file->event_call->mod); + kfree(edata); + } return 0; } static void -event_enable_free(struct ftrace_probe_ops *ops, unsigned long ip, - void **_data) +event_enable_free(struct ftrace_probe_ops *ops, struct trace_array *tr, + unsigned long ip, void *data) { - struct event_probe_data **pdata = (struct event_probe_data **)_data; - struct event_probe_data *data = *pdata; + struct ftrace_func_mapper *mapper = data; + struct event_probe_data *edata; - if (WARN_ON_ONCE(data->ref <= 0)) + if (!ip) { + if (!mapper) + return; + free_ftrace_func_mapper(mapper, free_probe_data); return; - - data->ref--; - if (!data->ref) { - /* Remove the SOFT_MODE flag */ - __ftrace_event_enable_disable(data->file, 0, 1); - module_put(data->file->event_call->mod); - kfree(data); } - *pdata = NULL; + + edata = ftrace_func_mapper_remove_ip(mapper, ip); + + if (WARN_ON_ONCE(!edata)) + return; + + if (WARN_ON_ONCE(edata->ref <= 0)) + return; + + free_probe_data(edata); } static struct ftrace_probe_ops event_enable_probe_ops = { @@ -2578,10 +2675,9 @@ static struct ftrace_probe_ops event_disable_count_probe_ops = { }; static int -event_enable_func(struct ftrace_hash *hash, +event_enable_func(struct trace_array *tr, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enabled) { - struct trace_array *tr = top_trace_array(); struct trace_event_file *file; struct ftrace_probe_ops *ops; struct event_probe_data *data; @@ -2619,12 +2715,12 @@ event_enable_func(struct ftrace_hash *hash, ops = param ? &event_disable_count_probe_ops : &event_disable_probe_ops; if (glob[0] == '!') { - unregister_ftrace_function_probe_func(glob+1, ops); - ret = 0; + ret = unregister_ftrace_function_probe_func(glob+1, tr, ops); goto out; } ret = -ENOMEM; + data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) goto out; @@ -2661,7 +2757,8 @@ event_enable_func(struct ftrace_hash *hash, ret = __ftrace_event_enable_disable(file, 1, 1); if (ret < 0) goto out_put; - ret = register_ftrace_function_probe(glob, ops, data); + + ret = register_ftrace_function_probe(glob, tr, ops, data); /* * The above returns on success the # of functions enabled, * but if it didn't find any functions it returns zero. diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index f3a960ed75a1..1c21d0e2a145 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -19,6 +19,7 @@ #include <linux/mutex.h> #include <linux/slab.h> #include <linux/stacktrace.h> +#include <linux/rculist.h> #include "tracing_map.h" #include "trace.h" diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index 6721a1e89f39..f2ac9d44f6c4 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -22,6 +22,7 @@ #include <linux/ctype.h> #include <linux/mutex.h> #include <linux/slab.h> +#include <linux/rculist.h> #include "trace.h" diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 0efa00d80623..a0910c0cdf2e 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -267,10 +267,14 @@ static struct tracer function_trace __tracer_data = }; #ifdef CONFIG_DYNAMIC_FTRACE -static void update_traceon_count(void **data, bool on) +static void update_traceon_count(struct ftrace_probe_ops *ops, + unsigned long ip, + struct trace_array *tr, bool on, + void *data) { - long *count = (long *)data; - long old_count = *count; + struct ftrace_func_mapper *mapper = data; + long *count; + long old_count; /* * Tracing gets disabled (or enabled) once per count. @@ -301,23 +305,22 @@ static void update_traceon_count(void **data, bool on) * setting the tracing_on file. But we currently don't care * about that. */ - if (!old_count) + count = (long *)ftrace_func_mapper_find_ip(mapper, ip); + old_count = *count; + + if (old_count <= 0) return; /* Make sure we see count before checking tracing state */ smp_rmb(); - if (on == !!tracing_is_on()) + if (on == !!tracer_tracing_is_on(tr)) return; if (on) - tracing_on(); + tracer_tracing_on(tr); else - tracing_off(); - - /* unlimited? */ - if (old_count == -1) - return; + tracer_tracing_off(tr); /* Make sure tracing state is visible before updating count */ smp_wmb(); @@ -326,33 +329,41 @@ static void update_traceon_count(void **data, bool on) } static void -ftrace_traceon_count(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_traceon_count(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - update_traceon_count(data, 1); + update_traceon_count(ops, ip, tr, 1, data); } static void -ftrace_traceoff_count(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_traceoff_count(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - update_traceon_count(data, 0); + update_traceon_count(ops, ip, tr, 0, data); } static void -ftrace_traceon(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_traceon(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - if (tracing_is_on()) + if (tracer_tracing_is_on(tr)) return; - tracing_on(); + tracer_tracing_on(tr); } static void -ftrace_traceoff(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_traceoff(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - if (!tracing_is_on()) + if (!tracer_tracing_is_on(tr)) return; - tracing_off(); + tracer_tracing_off(tr); } /* @@ -364,144 +375,218 @@ ftrace_traceoff(unsigned long ip, unsigned long parent_ip, void **data) */ #define STACK_SKIP 4 +static __always_inline void trace_stack(struct trace_array *tr) +{ + unsigned long flags; + int pc; + + local_save_flags(flags); + pc = preempt_count(); + + __trace_stack(tr, flags, STACK_SKIP, pc); +} + static void -ftrace_stacktrace(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_stacktrace(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - trace_dump_stack(STACK_SKIP); + trace_stack(tr); } static void -ftrace_stacktrace_count(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_stacktrace_count(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - long *count = (long *)data; + struct ftrace_func_mapper *mapper = data; + long *count; long old_count; long new_count; + if (!tracing_is_on()) + return; + + /* unlimited? */ + if (!mapper) { + trace_stack(tr); + return; + } + + count = (long *)ftrace_func_mapper_find_ip(mapper, ip); + /* * Stack traces should only execute the number of times the * user specified in the counter. */ do { - - if (!tracing_is_on()) - return; - old_count = *count; if (!old_count) return; - /* unlimited? */ - if (old_count == -1) { - trace_dump_stack(STACK_SKIP); - return; - } - new_count = old_count - 1; new_count = cmpxchg(count, old_count, new_count); if (new_count == old_count) - trace_dump_stack(STACK_SKIP); + trace_stack(tr); + + if (!tracing_is_on()) + return; } while (new_count != old_count); } -static int update_count(void **data) +static int update_count(struct ftrace_probe_ops *ops, unsigned long ip, + void *data) { - unsigned long *count = (long *)data; + struct ftrace_func_mapper *mapper = data; + long *count = NULL; - if (!*count) - return 0; + if (mapper) + count = (long *)ftrace_func_mapper_find_ip(mapper, ip); - if (*count != -1) + if (count) { + if (*count <= 0) + return 0; (*count)--; + } return 1; } static void -ftrace_dump_probe(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_dump_probe(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - if (update_count(data)) + if (update_count(ops, ip, data)) ftrace_dump(DUMP_ALL); } /* Only dump the current CPU buffer. */ static void -ftrace_cpudump_probe(unsigned long ip, unsigned long parent_ip, void **data) +ftrace_cpudump_probe(unsigned long ip, unsigned long parent_ip, + struct trace_array *tr, struct ftrace_probe_ops *ops, + void *data) { - if (update_count(data)) + if (update_count(ops, ip, data)) ftrace_dump(DUMP_ORIG); } static int ftrace_probe_print(const char *name, struct seq_file *m, - unsigned long ip, void *data) + unsigned long ip, struct ftrace_probe_ops *ops, + void *data) { - long count = (long)data; + struct ftrace_func_mapper *mapper = data; + long *count = NULL; seq_printf(m, "%ps:%s", (void *)ip, name); - if (count == -1) - seq_puts(m, ":unlimited\n"); + if (mapper) + count = (long *)ftrace_func_mapper_find_ip(mapper, ip); + + if (count) + seq_printf(m, ":count=%ld\n", *count); else - seq_printf(m, ":count=%ld\n", count); + seq_puts(m, ":unlimited\n"); return 0; } static int ftrace_traceon_print(struct seq_file *m, unsigned long ip, - struct ftrace_probe_ops *ops, void *data) + struct ftrace_probe_ops *ops, + void *data) { - return ftrace_probe_print("traceon", m, ip, data); + return ftrace_probe_print("traceon", m, ip, ops, data); } static int ftrace_traceoff_print(struct seq_file *m, unsigned long ip, struct ftrace_probe_ops *ops, void *data) { - return ftrace_probe_print("traceoff", m, ip, data); + return ftrace_probe_print("traceoff", m, ip, ops, data); } static int ftrace_stacktrace_print(struct seq_file *m, unsigned long ip, struct ftrace_probe_ops *ops, void *data) { - return ftrace_probe_print("stacktrace", m, ip, data); + return ftrace_probe_print("stacktrace", m, ip, ops, data); } static int ftrace_dump_print(struct seq_file *m, unsigned long ip, struct ftrace_probe_ops *ops, void *data) { - return ftrace_probe_print("dump", m, ip, data); + return ftrace_probe_print("dump", m, ip, ops, data); } static int ftrace_cpudump_print(struct seq_file *m, unsigned long ip, struct ftrace_probe_ops *ops, void *data) { - return ftrace_probe_print("cpudump", m, ip, data); + return ftrace_probe_print("cpudump", m, ip, ops, data); +} + + +static int +ftrace_count_init(struct ftrace_probe_ops *ops, struct trace_array *tr, + unsigned long ip, void *init_data, void **data) +{ + struct ftrace_func_mapper *mapper = *data; + + if (!mapper) { + mapper = allocate_ftrace_func_mapper(); + if (!mapper) + return -ENOMEM; + *data = mapper; + } + + return ftrace_func_mapper_add_ip(mapper, ip, init_data); +} + +static void +ftrace_count_free(struct ftrace_probe_ops *ops, struct trace_array *tr, + unsigned long ip, void *data) +{ + struct ftrace_func_mapper *mapper = data; + + if (!ip) { + free_ftrace_func_mapper(mapper, NULL); + return; + } + + ftrace_func_mapper_remove_ip(mapper, ip); } static struct ftrace_probe_ops traceon_count_probe_ops = { .func = ftrace_traceon_count, .print = ftrace_traceon_print, + .init = ftrace_count_init, + .free = ftrace_count_free, }; static struct ftrace_probe_ops traceoff_count_probe_ops = { .func = ftrace_traceoff_count, .print = ftrace_traceoff_print, + .init = ftrace_count_init, + .free = ftrace_count_free, }; static struct ftrace_probe_ops stacktrace_count_probe_ops = { .func = ftrace_stacktrace_count, .print = ftrace_stacktrace_print, + .init = ftrace_count_init, + .free = ftrace_count_free, }; static struct ftrace_probe_ops dump_probe_ops = { .func = ftrace_dump_probe, .print = ftrace_dump_print, + .init = ftrace_count_init, + .free = ftrace_count_free, }; static struct ftrace_probe_ops cpudump_probe_ops = { @@ -525,7 +610,8 @@ static struct ftrace_probe_ops stacktrace_probe_ops = { }; static int -ftrace_trace_probe_callback(struct ftrace_probe_ops *ops, +ftrace_trace_probe_callback(struct trace_array *tr, + struct ftrace_probe_ops *ops, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enable) { @@ -537,10 +623,8 @@ ftrace_trace_probe_callback(struct ftrace_probe_ops *ops, if (!enable) return -EINVAL; - if (glob[0] == '!') { - unregister_ftrace_function_probe_func(glob+1, ops); - return 0; - } + if (glob[0] == '!') + return unregister_ftrace_function_probe_func(glob+1, tr, ops); if (!param) goto out_reg; @@ -559,62 +643,74 @@ ftrace_trace_probe_callback(struct ftrace_probe_ops *ops, return ret; out_reg: - ret = register_ftrace_function_probe(glob, ops, count); + ret = register_ftrace_function_probe(glob, tr, ops, count); return ret < 0 ? ret : 0; } static int -ftrace_trace_onoff_callback(struct ftrace_hash *hash, +ftrace_trace_onoff_callback(struct trace_array *tr, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enable) { struct ftrace_probe_ops *ops; + if (!tr) + return -ENODEV; + /* we register both traceon and traceoff to this callback */ if (strcmp(cmd, "traceon") == 0) ops = param ? &traceon_count_probe_ops : &traceon_probe_ops; else ops = param ? &traceoff_count_probe_ops : &traceoff_probe_ops; - return ftrace_trace_probe_callback(ops, hash, glob, cmd, + return ftrace_trace_probe_callback(tr, ops, hash, glob, cmd, param, enable); } static int -ftrace_stacktrace_callback(struct ftrace_hash *hash, +ftrace_stacktrace_callback(struct trace_array *tr, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enable) { struct ftrace_probe_ops *ops; + if (!tr) + return -ENODEV; + ops = param ? &stacktrace_count_probe_ops : &stacktrace_probe_ops; - return ftrace_trace_probe_callback(ops, hash, glob, cmd, + return ftrace_trace_probe_callback(tr, ops, hash, glob, cmd, param, enable); } static int -ftrace_dump_callback(struct ftrace_hash *hash, +ftrace_dump_callback(struct trace_array *tr, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enable) { struct ftrace_probe_ops *ops; + if (!tr) + return -ENODEV; + ops = &dump_probe_ops; /* Only dump once. */ - return ftrace_trace_probe_callback(ops, hash, glob, cmd, + return ftrace_trace_probe_callback(tr, ops, hash, glob, cmd, "1", enable); } static int -ftrace_cpudump_callback(struct ftrace_hash *hash, +ftrace_cpudump_callback(struct trace_array *tr, struct ftrace_hash *hash, char *glob, char *cmd, char *param, int enable) { struct ftrace_probe_ops *ops; + if (!tr) + return -ENODEV; + ops = &cpudump_probe_ops; /* Only dump once. */ - return ftrace_trace_probe_callback(ops, hash, glob, cmd, + return ftrace_trace_probe_callback(tr, ops, hash, glob, cmd, "1", enable); } @@ -687,9 +783,8 @@ static inline int init_func_cmd_traceon(void) } #endif /* CONFIG_DYNAMIC_FTRACE */ -static __init int init_function_trace(void) +__init int init_function_trace(void) { init_func_cmd_traceon(); return register_tracer(&function_trace); } -core_initcall(init_function_trace); diff --git a/kernel/trace/trace_hwlat.c b/kernel/trace/trace_hwlat.c index af344a1bf0d0..d7c8e4ec3d9d 100644 --- a/kernel/trace/trace_hwlat.c +++ b/kernel/trace/trace_hwlat.c @@ -44,6 +44,7 @@ #include <linux/uaccess.h> #include <linux/cpumask.h> #include <linux/delay.h> +#include <linux/sched/clock.h> #include "trace.h" static struct trace_array *hwlat_trace; @@ -78,12 +79,12 @@ static u64 last_tracing_thresh = DEFAULT_LAT_THRESHOLD * NSEC_PER_USEC; /* Individual latency samples are stored here when detected. */ struct hwlat_sample { - u64 seqnum; /* unique sequence */ - u64 duration; /* delta */ - u64 outer_duration; /* delta (outer loop) */ - u64 nmi_total_ts; /* Total time spent in NMIs */ - struct timespec timestamp; /* wall time */ - int nmi_count; /* # NMIs during this sample */ + u64 seqnum; /* unique sequence */ + u64 duration; /* delta */ + u64 outer_duration; /* delta (outer loop) */ + u64 nmi_total_ts; /* Total time spent in NMIs */ + struct timespec64 timestamp; /* wall time */ + int nmi_count; /* # NMIs during this sample */ }; /* keep the global state somewhere. */ @@ -249,7 +250,7 @@ static int get_sample(void) s.seqnum = hwlat_data.count; s.duration = sample; s.outer_duration = outer_sample; - s.timestamp = CURRENT_TIME; + ktime_get_real_ts64(&s.timestamp); s.nmi_total_ts = nmi_total_ts; s.nmi_count = nmi_count; trace_hwlat_sample(&s); @@ -266,24 +267,13 @@ out: static struct cpumask save_cpumask; static bool disable_migrate; -static void move_to_next_cpu(bool initmask) +static void move_to_next_cpu(void) { - static struct cpumask *current_mask; + struct cpumask *current_mask = &save_cpumask; int next_cpu; if (disable_migrate) return; - - /* Just pick the first CPU on first iteration */ - if (initmask) { - current_mask = &save_cpumask; - get_online_cpus(); - cpumask_and(current_mask, cpu_online_mask, tracing_buffer_mask); - put_online_cpus(); - next_cpu = cpumask_first(current_mask); - goto set_affinity; - } - /* * If for some reason the user modifies the CPU affinity * of this thread, than stop migrating for the duration @@ -300,7 +290,6 @@ static void move_to_next_cpu(bool initmask) if (next_cpu >= nr_cpu_ids) next_cpu = cpumask_first(current_mask); - set_affinity: if (next_cpu >= nr_cpu_ids) /* Shouldn't happen! */ goto disable; @@ -322,20 +311,15 @@ static void move_to_next_cpu(bool initmask) * need to ensure nothing else might be running (and thus preempting). * Obviously this should never be used in production environments. * - * Currently this runs on which ever CPU it was scheduled on, but most - * real-world hardware latency situations occur across several CPUs, - * but we might later generalize this if we find there are any actualy - * systems with alternate SMI delivery or other hardware latencies. + * Executes one loop interaction on each CPU in tracing_cpumask sysfs file. */ static int kthread_fn(void *data) { u64 interval; - bool initmask = true; while (!kthread_should_stop()) { - move_to_next_cpu(initmask); - initmask = false; + move_to_next_cpu(); local_irq_disable(); get_sample(); @@ -366,13 +350,27 @@ static int kthread_fn(void *data) */ static int start_kthread(struct trace_array *tr) { + struct cpumask *current_mask = &save_cpumask; struct task_struct *kthread; + int next_cpu; + + /* Just pick the first CPU on first iteration */ + current_mask = &save_cpumask; + get_online_cpus(); + cpumask_and(current_mask, cpu_online_mask, tracing_buffer_mask); + put_online_cpus(); + next_cpu = cpumask_first(current_mask); kthread = kthread_create(kthread_fn, NULL, "hwlatd"); if (IS_ERR(kthread)) { pr_err(BANNER "could not start sampling thread\n"); return -ENOMEM; } + + cpumask_clear(current_mask); + cpumask_set_cpu(next_cpu, current_mask); + sched_setaffinity(kthread->pid, current_mask); + hwlat_kthread = kthread; wake_up_process(kthread); diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 7ad9e53ad174..c9b5aa10fbf9 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -16,13 +16,16 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ +#define pr_fmt(fmt) "trace_kprobe: " fmt #include <linux/module.h> #include <linux/uaccess.h> +#include <linux/rculist.h> #include "trace_probe.h" #define KPROBE_EVENT_SYSTEM "kprobes" +#define KRETPROBE_MAXACTIVE_MAX 4096 /** * Kprobe event core functions @@ -280,6 +283,7 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group, void *addr, const char *symbol, unsigned long offs, + int maxactive, int nargs, bool is_return) { struct trace_kprobe *tk; @@ -307,6 +311,8 @@ static struct trace_kprobe *alloc_trace_kprobe(const char *group, else tk->rp.kp.pre_handler = kprobe_dispatcher; + tk->rp.maxactive = maxactive; + if (!event || !is_good_name(event)) { ret = -EINVAL; goto error; @@ -592,12 +598,22 @@ static struct notifier_block trace_kprobe_module_nb = { .priority = 1 /* Invoked after kprobe module callback */ }; +/* Convert certain expected symbols into '_' when generating event names */ +static inline void sanitize_event_name(char *name) +{ + while (*name++ != '\0') + if (*name == ':' || *name == '.') + *name = '_'; +} + static int create_trace_kprobe(int argc, char **argv) { /* * Argument syntax: - * - Add kprobe: p[:[GRP/]EVENT] [MOD:]KSYM[+OFFS]|KADDR [FETCHARGS] - * - Add kretprobe: r[:[GRP/]EVENT] [MOD:]KSYM[+0] [FETCHARGS] + * - Add kprobe: + * p[:[GRP/]EVENT] [MOD:]KSYM[+OFFS]|KADDR [FETCHARGS] + * - Add kretprobe: + * r[MAXACTIVE][:[GRP/]EVENT] [MOD:]KSYM[+0] [FETCHARGS] * Fetch args: * $retval : fetch return value * $stack : fetch stack address @@ -617,6 +633,7 @@ static int create_trace_kprobe(int argc, char **argv) int i, ret = 0; bool is_return = false, is_delete = false; char *symbol = NULL, *event = NULL, *group = NULL; + int maxactive = 0; char *arg; unsigned long offset = 0; void *addr = NULL; @@ -635,8 +652,28 @@ static int create_trace_kprobe(int argc, char **argv) return -EINVAL; } - if (argv[0][1] == ':') { - event = &argv[0][2]; + event = strchr(&argv[0][1], ':'); + if (event) { + event[0] = '\0'; + event++; + } + if (is_return && isdigit(argv[0][1])) { + ret = kstrtouint(&argv[0][1], 0, &maxactive); + if (ret) { + pr_info("Failed to parse maxactive.\n"); + return ret; + } + /* kretprobes instances are iterated over via a list. The + * maximum should stay reasonable. + */ + if (maxactive > KRETPROBE_MAXACTIVE_MAX) { + pr_info("Maxactive is too big (%d > %d).\n", + maxactive, KRETPROBE_MAXACTIVE_MAX); + return -E2BIG; + } + } + + if (event) { if (strchr(event, '/')) { group = event; event = strchr(group, '/') + 1; @@ -678,28 +715,21 @@ static int create_trace_kprobe(int argc, char **argv) pr_info("Probe point is not specified.\n"); return -EINVAL; } - if (isdigit(argv[1][0])) { - if (is_return) { - pr_info("Return probe point must be a symbol.\n"); - return -EINVAL; - } - /* an address specified */ - ret = kstrtoul(&argv[1][0], 0, (unsigned long *)&addr); - if (ret) { - pr_info("Failed to parse address.\n"); - return ret; - } - } else { + + /* try to parse an address. if that fails, try to read the + * input as a symbol. */ + if (kstrtoul(argv[1], 0, (unsigned long *)&addr)) { /* a symbol specified */ symbol = argv[1]; /* TODO: support .init module functions */ ret = traceprobe_split_symbol_offset(symbol, &offset); if (ret) { - pr_info("Failed to parse symbol.\n"); + pr_info("Failed to parse either an address or a symbol.\n"); return ret; } - if (offset && is_return) { - pr_info("Return probe must be used without offset.\n"); + if (offset && is_return && + !kprobe_on_func_entry(NULL, symbol, offset)) { + pr_info("Given offset is not valid for return probe.\n"); return -EINVAL; } } @@ -714,10 +744,11 @@ static int create_trace_kprobe(int argc, char **argv) else snprintf(buf, MAX_EVENT_NAME_LEN, "%c_0x%p", is_return ? 'r' : 'p', addr); + sanitize_event_name(buf); event = buf; } - tk = alloc_trace_kprobe(group, event, addr, symbol, offset, argc, - is_return); + tk = alloc_trace_kprobe(group, event, addr, symbol, offset, maxactive, + argc, is_return); if (IS_ERR(tk)) { pr_info("Failed to allocate trace_probe.(%d)\n", (int)PTR_ERR(tk)); @@ -1509,6 +1540,11 @@ static __init int kprobe_trace_self_tests_init(void) end: release_all_trace_kprobes(); + /* + * Wait for the optimizer work to finish. Otherwise it might fiddle + * with probes in already freed __init text. + */ + wait_for_kprobe_optimizer(); if (warn) pr_cont("NG: Some tests are failed. Please check them.\n"); else diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index 5d33a7352919..bac629af2285 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -4,10 +4,11 @@ * Copyright (C) 2008 Red Hat Inc, Steven Rostedt <[email protected]> * */ - #include <linux/module.h> #include <linux/mutex.h> #include <linux/ftrace.h> +#include <linux/sched/clock.h> +#include <linux/sched/mm.h> #include "trace_output.h" @@ -124,6 +125,44 @@ EXPORT_SYMBOL(trace_print_symbols_seq); #if BITS_PER_LONG == 32 const char * +trace_print_flags_seq_u64(struct trace_seq *p, const char *delim, + unsigned long long flags, + const struct trace_print_flags_u64 *flag_array) +{ + unsigned long long mask; + const char *str; + const char *ret = trace_seq_buffer_ptr(p); + int i, first = 1; + + for (i = 0; flag_array[i].name && flags; i++) { + + mask = flag_array[i].mask; + if ((flags & mask) != mask) + continue; + + str = flag_array[i].name; + flags &= ~mask; + if (!first && delim) + trace_seq_puts(p, delim); + else + first = 0; + trace_seq_puts(p, str); + } + + /* check for left over flags */ + if (flags) { + if (!first && delim) + trace_seq_puts(p, delim); + trace_seq_printf(p, "0x%llx", flags); + } + + trace_seq_putc(p, 0); + + return ret; +} +EXPORT_SYMBOL(trace_print_flags_seq_u64); + +const char * trace_print_symbols_seq_u64(struct trace_seq *p, unsigned long long val, const struct trace_print_flags_u64 *symbol_array) { @@ -162,15 +201,27 @@ trace_print_bitmask_seq(struct trace_seq *p, void *bitmask_ptr, } EXPORT_SYMBOL_GPL(trace_print_bitmask_seq); +/** + * trace_print_hex_seq - print buffer as hex sequence + * @p: trace seq struct to write to + * @buf: The buffer to print + * @buf_len: Length of @buf in bytes + * @concatenate: Print @buf as single hex string or with spacing + * + * Prints the passed buffer as a hex sequence either as a whole, + * single hex string if @concatenate is true or with spacing after + * each byte in case @concatenate is false. + */ const char * -trace_print_hex_seq(struct trace_seq *p, const unsigned char *buf, int buf_len) +trace_print_hex_seq(struct trace_seq *p, const unsigned char *buf, int buf_len, + bool concatenate) { int i; const char *ret = trace_seq_buffer_ptr(p); for (i = 0; i < buf_len; i++) - trace_seq_printf(p, "%s%2.2x", i == 0 ? "" : " ", buf[i]); - + trace_seq_printf(p, "%s%2.2x", concatenate || i == 0 ? "" : " ", + buf[i]); trace_seq_putc(p, 0); return ret; @@ -289,31 +340,41 @@ static inline const char *kretprobed(const char *name) static void seq_print_sym_short(struct trace_seq *s, const char *fmt, unsigned long address) { -#ifdef CONFIG_KALLSYMS char str[KSYM_SYMBOL_LEN]; +#ifdef CONFIG_KALLSYMS const char *name; kallsyms_lookup(address, NULL, NULL, NULL, str); name = kretprobed(str); - trace_seq_printf(s, fmt, name); + if (name && strlen(name)) { + trace_seq_printf(s, fmt, name); + return; + } #endif + snprintf(str, KSYM_SYMBOL_LEN, "0x%08lx", address); + trace_seq_printf(s, fmt, str); } static void seq_print_sym_offset(struct trace_seq *s, const char *fmt, unsigned long address) { -#ifdef CONFIG_KALLSYMS char str[KSYM_SYMBOL_LEN]; +#ifdef CONFIG_KALLSYMS const char *name; sprint_symbol(str, address); name = kretprobed(str); - trace_seq_printf(s, fmt, name); + if (name && strlen(name)) { + trace_seq_printf(s, fmt, name); + return; + } #endif + snprintf(str, KSYM_SYMBOL_LEN, "0x%08lx", address); + trace_seq_printf(s, fmt, str); } #ifndef CONFIG_64BIT @@ -536,6 +597,15 @@ int trace_print_context(struct trace_iterator *iter) trace_seq_printf(s, "%16s-%-5d [%03d] ", comm, entry->pid, iter->cpu); + if (tr->trace_flags & TRACE_ITER_RECORD_TGID) { + unsigned int tgid = trace_find_tgid(entry->pid); + + if (!tgid) + trace_seq_printf(s, "(-----) "); + else + trace_seq_printf(s, "(%5d) ", tgid); + } + if (tr->trace_flags & TRACE_ITER_IRQ_INFO) trace_print_lat_fmt(s, entry); @@ -1109,11 +1179,11 @@ trace_hwlat_print(struct trace_iterator *iter, int flags, trace_assign_type(field, entry); - trace_seq_printf(s, "#%-5u inner/outer(us): %4llu/%-5llu ts:%ld.%09ld", + trace_seq_printf(s, "#%-5u inner/outer(us): %4llu/%-5llu ts:%lld.%09ld", field->seqnum, field->duration, field->outer_duration, - field->timestamp.tv_sec, + (long long)field->timestamp.tv_sec, field->timestamp.tv_nsec); if (field->nmi_count) { @@ -1143,10 +1213,10 @@ trace_hwlat_raw(struct trace_iterator *iter, int flags, trace_assign_type(field, iter->ent); - trace_seq_printf(s, "%llu %lld %ld %09ld %u\n", + trace_seq_printf(s, "%llu %lld %lld %09ld %u\n", field->duration, field->outer_duration, - field->timestamp.tv_sec, + (long long)field->timestamp.tv_sec, field->timestamp.tv_nsec, field->seqnum); diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 8c0553d9afd3..52478f033f88 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -21,6 +21,7 @@ * Copyright (C) IBM Corporation, 2010-2011 * Author: Srikar Dronamraju */ +#define pr_fmt(fmt) "trace_probe: " fmt #include "trace_probe.h" @@ -647,7 +648,7 @@ ssize_t traceprobe_probes_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos, int (*createfn)(int, char **)) { - char *kbuf, *tmp; + char *kbuf, *buf, *tmp; int ret = 0; size_t done = 0; size_t size; @@ -667,27 +668,38 @@ ssize_t traceprobe_probes_write(struct file *file, const char __user *buffer, goto out; } kbuf[size] = '\0'; - tmp = strchr(kbuf, '\n'); + buf = kbuf; + do { + tmp = strchr(buf, '\n'); + if (tmp) { + *tmp = '\0'; + size = tmp - buf + 1; + } else { + size = strlen(buf); + if (done + size < count) { + if (buf != kbuf) + break; + /* This can accept WRITE_BUFSIZE - 2 ('\n' + '\0') */ + pr_warn("Line length is too long: Should be less than %d\n", + WRITE_BUFSIZE - 2); + ret = -EINVAL; + goto out; + } + } + done += size; - if (tmp) { - *tmp = '\0'; - size = tmp - kbuf + 1; - } else if (done + size < count) { - pr_warn("Line length is too long: Should be less than %d\n", - WRITE_BUFSIZE); - ret = -EINVAL; - goto out; - } - done += size; - /* Remove comments */ - tmp = strchr(kbuf, '#'); + /* Remove comments */ + tmp = strchr(buf, '#'); - if (tmp) - *tmp = '\0'; + if (tmp) + *tmp = '\0'; - ret = traceprobe_command(kbuf, createfn); - if (ret) - goto out; + ret = traceprobe_command(buf, createfn); + if (ret) + goto out; + buf += size; + + } while (done < count); } ret = done; diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index 0c0ae54d44c6..903273c93e61 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -248,7 +248,7 @@ ASSIGN_FETCH_FUNC(file_offset, ftype), \ #define FETCH_TYPE_STRING 0 #define FETCH_TYPE_STRSIZE 1 -#ifdef CONFIG_KPROBE_EVENT +#ifdef CONFIG_KPROBE_EVENTS struct symbol_cache; unsigned long update_symbol_cache(struct symbol_cache *sc); void free_symbol_cache(struct symbol_cache *sc); @@ -278,7 +278,7 @@ alloc_symbol_cache(const char *sym, long offset) { return NULL; } -#endif /* CONFIG_KPROBE_EVENT */ +#endif /* CONFIG_KPROBE_EVENTS */ struct probe_arg { struct fetch_param fetch; diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index 4c896a0101bd..b341c02730be 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -12,27 +12,38 @@ #include "trace.h" -static int sched_ref; +#define RECORD_CMDLINE 1 +#define RECORD_TGID 2 + +static int sched_cmdline_ref; +static int sched_tgid_ref; static DEFINE_MUTEX(sched_register_mutex); static void probe_sched_switch(void *ignore, bool preempt, struct task_struct *prev, struct task_struct *next) { - if (unlikely(!sched_ref)) - return; + int flags; + + flags = (RECORD_TGID * !!sched_tgid_ref) + + (RECORD_CMDLINE * !!sched_cmdline_ref); - tracing_record_cmdline(prev); - tracing_record_cmdline(next); + if (!flags) + return; + tracing_record_taskinfo_sched_switch(prev, next, flags); } static void probe_sched_wakeup(void *ignore, struct task_struct *wakee) { - if (unlikely(!sched_ref)) - return; + int flags; + + flags = (RECORD_TGID * !!sched_tgid_ref) + + (RECORD_CMDLINE * !!sched_cmdline_ref); - tracing_record_cmdline(current); + if (!flags) + return; + tracing_record_taskinfo(current, flags); } static int tracing_sched_register(void) @@ -75,28 +86,61 @@ static void tracing_sched_unregister(void) unregister_trace_sched_wakeup(probe_sched_wakeup, NULL); } -static void tracing_start_sched_switch(void) +static void tracing_start_sched_switch(int ops) { + bool sched_register = (!sched_cmdline_ref && !sched_tgid_ref); mutex_lock(&sched_register_mutex); - if (!(sched_ref++)) + + switch (ops) { + case RECORD_CMDLINE: + sched_cmdline_ref++; + break; + + case RECORD_TGID: + sched_tgid_ref++; + break; + } + + if (sched_register && (sched_cmdline_ref || sched_tgid_ref)) tracing_sched_register(); mutex_unlock(&sched_register_mutex); } -static void tracing_stop_sched_switch(void) +static void tracing_stop_sched_switch(int ops) { mutex_lock(&sched_register_mutex); - if (!(--sched_ref)) + + switch (ops) { + case RECORD_CMDLINE: + sched_cmdline_ref--; + break; + + case RECORD_TGID: + sched_tgid_ref--; + break; + } + + if (!sched_cmdline_ref && !sched_tgid_ref) tracing_sched_unregister(); mutex_unlock(&sched_register_mutex); } void tracing_start_cmdline_record(void) { - tracing_start_sched_switch(); + tracing_start_sched_switch(RECORD_CMDLINE); } void tracing_stop_cmdline_record(void) { - tracing_stop_sched_switch(); + tracing_stop_sched_switch(RECORD_CMDLINE); +} + +void tracing_start_tgid_record(void) +{ + tracing_start_sched_switch(RECORD_TGID); +} + +void tracing_stop_tgid_record(void) +{ + tracing_stop_sched_switch(RECORD_TGID); } diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index b0f86ea77881..cb917cebae29 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -1,5 +1,6 @@ /* Include in trace.c */ +#include <uapi/linux/sched/types.h> #include <linux/stringify.h> #include <linux/kthread.h> #include <linux/delay.h> diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 2a1abbaca10e..a4df67cbc711 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -2,6 +2,7 @@ * Copyright (C) 2008 Steven Rostedt <[email protected]> * */ +#include <linux/sched/task_stack.h> #include <linux/stacktrace.h> #include <linux/kallsyms.h> #include <linux/seq_file.h> @@ -34,7 +35,7 @@ unsigned long stack_trace_max_size; arch_spinlock_t stack_trace_max_lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; -static DEFINE_PER_CPU(int, trace_active); +DEFINE_PER_CPU(int, disable_stack_tracer); static DEFINE_MUTEX(stack_sysctl_mutex); int stack_tracer_enabled; @@ -64,7 +65,7 @@ void stack_trace_print(void) } /* - * When arch-specific code overides this function, the following + * When arch-specific code overrides this function, the following * data should be filled up, assuming stack_trace_max_lock is held to * prevent concurrent updates. * stack_trace_index[] @@ -95,6 +96,14 @@ check_stack(unsigned long ip, unsigned long *stack) if (in_nmi()) return; + /* + * There's a slight chance that we are tracing inside the + * RCU infrastructure, and rcu_irq_enter() will not work + * as expected. + */ + if (unlikely(rcu_irq_enter_disabled())) + return; + local_irq_save(flags); arch_spin_lock(&stack_trace_max_lock); @@ -206,13 +215,12 @@ stack_trace_call(unsigned long ip, unsigned long parent_ip, struct ftrace_ops *op, struct pt_regs *pt_regs) { unsigned long stack; - int cpu; preempt_disable_notrace(); - cpu = raw_smp_processor_id(); /* no atomic needed, we only modify this variable by this cpu */ - if (per_cpu(trace_active, cpu)++ != 0) + __this_cpu_inc(disable_stack_tracer); + if (__this_cpu_read(disable_stack_tracer) != 1) goto out; ip += MCOUNT_INSN_SIZE; @@ -220,7 +228,7 @@ stack_trace_call(unsigned long ip, unsigned long parent_ip, check_stack(ip, &stack); out: - per_cpu(trace_active, cpu)--; + __this_cpu_dec(disable_stack_tracer); /* prevent recursion in schedule */ preempt_enable_notrace(); } @@ -252,7 +260,6 @@ stack_max_size_write(struct file *filp, const char __user *ubuf, long *ptr = filp->private_data; unsigned long val, flags; int ret; - int cpu; ret = kstrtoul_from_user(ubuf, count, 10, &val); if (ret) @@ -263,16 +270,15 @@ stack_max_size_write(struct file *filp, const char __user *ubuf, /* * In case we trace inside arch_spin_lock() or after (NMI), * we will cause circular lock, so we also need to increase - * the percpu trace_active here. + * the percpu disable_stack_tracer here. */ - cpu = smp_processor_id(); - per_cpu(trace_active, cpu)++; + __this_cpu_inc(disable_stack_tracer); arch_spin_lock(&stack_trace_max_lock); *ptr = val; arch_spin_unlock(&stack_trace_max_lock); - per_cpu(trace_active, cpu)--; + __this_cpu_dec(disable_stack_tracer); local_irq_restore(flags); return count; @@ -306,12 +312,9 @@ t_next(struct seq_file *m, void *v, loff_t *pos) static void *t_start(struct seq_file *m, loff_t *pos) { - int cpu; - local_irq_disable(); - cpu = smp_processor_id(); - per_cpu(trace_active, cpu)++; + __this_cpu_inc(disable_stack_tracer); arch_spin_lock(&stack_trace_max_lock); @@ -323,12 +326,9 @@ static void *t_start(struct seq_file *m, loff_t *pos) static void t_stop(struct seq_file *m, void *p) { - int cpu; - arch_spin_unlock(&stack_trace_max_lock); - cpu = smp_processor_id(); - per_cpu(trace_active, cpu)--; + __this_cpu_dec(disable_stack_tracer); local_irq_enable(); } @@ -406,10 +406,14 @@ static const struct file_operations stack_trace_fops = { .release = seq_release, }; +#ifdef CONFIG_DYNAMIC_FTRACE + static int stack_trace_filter_open(struct inode *inode, struct file *file) { - return ftrace_regex_open(&trace_ops, FTRACE_ITER_FILTER, + struct ftrace_ops *ops = inode->i_private; + + return ftrace_regex_open(ops, FTRACE_ITER_FILTER, inode, file); } @@ -421,6 +425,8 @@ static const struct file_operations stack_trace_filter_fops = { .release = ftrace_regex_release, }; +#endif /* CONFIG_DYNAMIC_FTRACE */ + int stack_trace_sysctl(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, @@ -475,8 +481,10 @@ static __init int stack_trace_init(void) trace_create_file("stack_trace", 0444, d_tracer, NULL, &stack_trace_fops); +#ifdef CONFIG_DYNAMIC_FTRACE trace_create_file("stack_trace_filter", 0444, d_tracer, - NULL, &stack_trace_filter_fops); + &trace_ops, &stack_trace_filter_fops); +#endif if (stack_trace_filter_buf[0]) ftrace_set_early_filter(&trace_ops, stack_trace_filter_buf, 1); diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 0913693caf6e..a7581fec9681 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -17,12 +17,14 @@ * Copyright (C) IBM Corporation, 2010-2012 * Author: Srikar Dronamraju <[email protected]> */ +#define pr_fmt(fmt) "trace_kprobe: " fmt #include <linux/module.h> #include <linux/uaccess.h> #include <linux/uprobes.h> #include <linux/namei.h> #include <linux/string.h> +#include <linux/rculist.h> #include "trace_probe.h" @@ -431,7 +433,8 @@ static int create_trace_uprobe(int argc, char **argv) pr_info("Probe point is not specified.\n"); return -EINVAL; } - arg = strchr(argv[1], ':'); + /* Find the last occurrence, in case the path contains ':' too. */ + arg = strrchr(argv[1], ':'); if (!arg) { ret = -EINVAL; goto fail_address_parse; diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 1f9a31f934a4..685c50ae6300 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -24,7 +24,8 @@ #include <linux/tracepoint.h> #include <linux/err.h> #include <linux/slab.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/task.h> #include <linux/static_key.h> extern struct tracepoint * const __start___tracepoints_ptrs[]; diff --git a/kernel/tsacct.c b/kernel/tsacct.c index f8e26ab963ed..370724b45391 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -17,7 +17,9 @@ */ #include <linux/kernel.h> -#include <linux/sched.h> +#include <linux/sched/signal.h> +#include <linux/sched/mm.h> +#include <linux/sched/cputime.h> #include <linux/tsacct_kern.h> #include <linux/acct.h> #include <linux/jiffies.h> @@ -31,7 +33,7 @@ void bacct_add_tsk(struct user_namespace *user_ns, struct taskstats *stats, struct task_struct *tsk) { const struct cred *tcred; - cputime_t utime, stime, utimescaled, stimescaled; + u64 utime, stime, utimescaled, stimescaled; u64 delta; BUILD_BUG_ON(TS_COMM_LEN < TASK_COMM_LEN); @@ -67,12 +69,12 @@ void bacct_add_tsk(struct user_namespace *user_ns, rcu_read_unlock(); task_cputime(tsk, &utime, &stime); - stats->ac_utime = cputime_to_usecs(utime); - stats->ac_stime = cputime_to_usecs(stime); + stats->ac_utime = div_u64(utime, NSEC_PER_USEC); + stats->ac_stime = div_u64(stime, NSEC_PER_USEC); task_cputime_scaled(tsk, &utimescaled, &stimescaled); - stats->ac_utimescaled = cputime_to_usecs(utimescaled); - stats->ac_stimescaled = cputime_to_usecs(stimescaled); + stats->ac_utimescaled = div_u64(utimescaled, NSEC_PER_USEC); + stats->ac_stimescaled = div_u64(stimescaled, NSEC_PER_USEC); stats->ac_minflt = tsk->min_flt; stats->ac_majflt = tsk->maj_flt; @@ -123,18 +125,15 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) #undef MB static void __acct_update_integrals(struct task_struct *tsk, - cputime_t utime, cputime_t stime) + u64 utime, u64 stime) { - cputime_t time, dtime; - u64 delta; + u64 time, delta; if (!likely(tsk->mm)) return; time = stime + utime; - dtime = time - tsk->acct_timexpd; - /* Avoid division: cputime_t is often in nanoseconds already. */ - delta = cputime_to_nsecs(dtime); + delta = time - tsk->acct_timexpd; if (delta < TICK_NSEC) return; @@ -155,7 +154,7 @@ static void __acct_update_integrals(struct task_struct *tsk, */ void acct_update_integrals(struct task_struct *tsk) { - cputime_t utime, stime; + u64 utime, stime; unsigned long flags; local_irq_save(flags); diff --git a/kernel/ucount.c b/kernel/ucount.c index 95c6336fc2b3..b4eeee03934f 100644 --- a/kernel/ucount.c +++ b/kernel/ucount.c @@ -8,6 +8,7 @@ #include <linux/stat.h> #include <linux/sysctl.h> #include <linux/slab.h> +#include <linux/cred.h> #include <linux/hash.h> #include <linux/user_namespace.h> @@ -57,7 +58,7 @@ static struct ctl_table_root set_root = { static int zero = 0; static int int_max = INT_MAX; -#define UCOUNT_ENTRY(name) \ +#define UCOUNT_ENTRY(name) \ { \ .procname = name, \ .maxlen = sizeof(int), \ @@ -74,6 +75,10 @@ static struct ctl_table user_table[] = { UCOUNT_ENTRY("max_net_namespaces"), UCOUNT_ENTRY("max_mnt_namespaces"), UCOUNT_ENTRY("max_cgroup_namespaces"), +#ifdef CONFIG_INOTIFY_USER + UCOUNT_ENTRY("max_inotify_instances"), + UCOUNT_ENTRY("max_inotify_watches"), +#endif { } }; #endif /* CONFIG_SYSCTL */ @@ -139,7 +144,7 @@ static struct ucounts *get_ucounts(struct user_namespace *ns, kuid_t uid) new->ns = ns; new->uid = uid; - atomic_set(&new->count, 0); + new->count = 0; spin_lock_irq(&ucounts_lock); ucounts = find_ucounts(ns, uid, hashent); @@ -150,8 +155,10 @@ static struct ucounts *get_ucounts(struct user_namespace *ns, kuid_t uid) ucounts = new; } } - if (!atomic_add_unless(&ucounts->count, 1, INT_MAX)) + if (ucounts->count == INT_MAX) ucounts = NULL; + else + ucounts->count += 1; spin_unlock_irq(&ucounts_lock); return ucounts; } @@ -160,13 +167,15 @@ static void put_ucounts(struct ucounts *ucounts) { unsigned long flags; - if (atomic_dec_and_test(&ucounts->count)) { - spin_lock_irqsave(&ucounts_lock, flags); + spin_lock_irqsave(&ucounts_lock, flags); + ucounts->count -= 1; + if (!ucounts->count) hlist_del_init(&ucounts->node); - spin_unlock_irqrestore(&ucounts_lock, flags); + else + ucounts = NULL; + spin_unlock_irqrestore(&ucounts_lock, flags); - kfree(ucounts); - } + kfree(ucounts); } static inline bool atomic_inc_below(atomic_t *v, int u) diff --git a/kernel/uid16.c b/kernel/uid16.c index 71645ae9303a..5c2dc5b2bf4f 100644 --- a/kernel/uid16.c +++ b/kernel/uid16.c @@ -12,6 +12,7 @@ #include <linux/init.h> #include <linux/highuid.h> #include <linux/security.h> +#include <linux/cred.h> #include <linux/syscalls.h> #include <linux/uaccess.h> diff --git a/kernel/user.c b/kernel/user.c index b069ccbfb0b0..00281add65b2 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -13,6 +13,7 @@ #include <linux/slab.h> #include <linux/bitops.h> #include <linux/key.h> +#include <linux/sched/user.h> #include <linux/interrupt.h> #include <linux/export.h> #include <linux/user_namespace.h> diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index 86b7854fec8e..2f735cbe05e8 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -8,6 +8,7 @@ #include <linux/export.h> #include <linux/nsproxy.h> #include <linux/slab.h> +#include <linux/sched/signal.h> #include <linux/user_namespace.h> #include <linux/proc_ns.h> #include <linux/highuid.h> diff --git a/kernel/utsname.c b/kernel/utsname.c index 6976cd47dcf6..913fe4336d2b 100644 --- a/kernel/utsname.c +++ b/kernel/utsname.c @@ -14,8 +14,10 @@ #include <linux/utsname.h> #include <linux/err.h> #include <linux/slab.h> +#include <linux/cred.h> #include <linux/user_namespace.h> #include <linux/proc_ns.h> +#include <linux/sched/task.h> static struct ucounts *inc_uts_namespaces(struct user_namespace *ns) { diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c index c8eac43267e9..233cd8fc6910 100644 --- a/kernel/utsname_sysctl.c +++ b/kernel/utsname_sysctl.c @@ -14,6 +14,7 @@ #include <linux/utsname.h> #include <linux/sysctl.h> #include <linux/wait.h> +#include <linux/rwsem.h> #ifdef CONFIG_PROC_SYSCTL diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 63177be0159e..06d3389bca0d 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -9,7 +9,7 @@ * to those contributors as well. */ -#define pr_fmt(fmt) "NMI watchdog: " fmt +#define pr_fmt(fmt) "watchdog: " fmt #include <linux/mm.h> #include <linux/cpu.h> @@ -19,22 +19,68 @@ #include <linux/sysctl.h> #include <linux/smpboot.h> #include <linux/sched/rt.h> +#include <uapi/linux/sched/types.h> #include <linux/tick.h> #include <linux/workqueue.h> +#include <linux/sched/clock.h> +#include <linux/sched/debug.h> #include <asm/irq_regs.h> #include <linux/kvm_para.h> #include <linux/kthread.h> +/* Watchdog configuration */ static DEFINE_MUTEX(watchdog_proc_mutex); -#if defined(CONFIG_HAVE_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR) -unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED|NMI_WATCHDOG_ENABLED; +int __read_mostly nmi_watchdog_enabled; + +#if defined(CONFIG_HARDLOCKUP_DETECTOR) || defined(CONFIG_HAVE_NMI_WATCHDOG) +unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED | + NMI_WATCHDOG_ENABLED; #else unsigned long __read_mostly watchdog_enabled = SOFT_WATCHDOG_ENABLED; #endif -int __read_mostly nmi_watchdog_enabled; + +#ifdef CONFIG_HARDLOCKUP_DETECTOR +/* boot commands */ +/* + * Should we panic when a soft-lockup or hard-lockup occurs: + */ +unsigned int __read_mostly hardlockup_panic = + CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE; +/* + * We may not want to enable hard lockup detection by default in all cases, + * for example when running the kernel as a guest on a hypervisor. In these + * cases this function can be called to disable hard lockup detection. This + * function should only be executed once by the boot processor before the + * kernel command line parameters are parsed, because otherwise it is not + * possible to override this in hardlockup_panic_setup(). + */ +void hardlockup_detector_disable(void) +{ + watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; +} + +static int __init hardlockup_panic_setup(char *str) +{ + if (!strncmp(str, "panic", 5)) + hardlockup_panic = 1; + else if (!strncmp(str, "nopanic", 7)) + hardlockup_panic = 0; + else if (!strncmp(str, "0", 1)) + watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; + else if (!strncmp(str, "1", 1)) + watchdog_enabled |= NMI_WATCHDOG_ENABLED; + return 1; +} +__setup("nmi_watchdog=", hardlockup_panic_setup); + +#endif + +#ifdef CONFIG_SOFTLOCKUP_DETECTOR int __read_mostly soft_watchdog_enabled; +#endif + int __read_mostly watchdog_user_enabled; int __read_mostly watchdog_thresh = 10; @@ -42,15 +88,9 @@ int __read_mostly watchdog_thresh = 10; int __read_mostly sysctl_softlockup_all_cpu_backtrace; int __read_mostly sysctl_hardlockup_all_cpu_backtrace; #endif -static struct cpumask watchdog_cpumask __read_mostly; +struct cpumask watchdog_cpumask __read_mostly; unsigned long *watchdog_cpumask_bits = cpumask_bits(&watchdog_cpumask); -/* Helper for online, unparked cpus. */ -#define for_each_watchdog_cpu(cpu) \ - for_each_cpu_and((cpu), cpu_online_mask, &watchdog_cpumask) - -atomic_t watchdog_park_in_progress = ATOMIC_INIT(0); - /* * The 'watchdog_running' variable is set to 1 when the watchdog threads * are registered/started and is set to 0 when the watchdog threads are @@ -69,7 +109,47 @@ static int __read_mostly watchdog_running; * of 'watchdog_running' cannot change while the watchdog is deactivated * temporarily (see related code in 'proc' handlers). */ -static int __read_mostly watchdog_suspended; +int __read_mostly watchdog_suspended; + +/* + * These functions can be overridden if an architecture implements its + * own hardlockup detector. + * + * watchdog_nmi_enable/disable can be implemented to start and stop when + * softlockup watchdog threads start and stop. The arch must select the + * SOFTLOCKUP_DETECTOR Kconfig. + */ +int __weak watchdog_nmi_enable(unsigned int cpu) +{ + return 0; +} +void __weak watchdog_nmi_disable(unsigned int cpu) +{ +} + +/* + * watchdog_nmi_reconfigure can be implemented to be notified after any + * watchdog configuration change. The arch hardlockup watchdog should + * respond to the following variables: + * - nmi_watchdog_enabled + * - watchdog_thresh + * - watchdog_cpumask + * - sysctl_hardlockup_all_cpu_backtrace + * - hardlockup_panic + * - watchdog_suspended + */ +void __weak watchdog_nmi_reconfigure(void) +{ +} + + +#ifdef CONFIG_SOFTLOCKUP_DETECTOR + +/* Helper for online, unparked cpus. */ +#define for_each_watchdog_cpu(cpu) \ + for_each_cpu_and((cpu), cpu_online_mask, &watchdog_cpumask) + +atomic_t watchdog_park_in_progress = ATOMIC_INIT(0); static u64 __read_mostly sample_period; @@ -117,6 +197,7 @@ static int __init softlockup_all_cpu_backtrace_setup(char *str) return 1; } __setup("softlockup_all_cpu_backtrace=", softlockup_all_cpu_backtrace_setup); +#ifdef CONFIG_HARDLOCKUP_DETECTOR static int __init hardlockup_all_cpu_backtrace_setup(char *str) { sysctl_hardlockup_all_cpu_backtrace = @@ -125,6 +206,7 @@ static int __init hardlockup_all_cpu_backtrace_setup(char *str) } __setup("hardlockup_all_cpu_backtrace=", hardlockup_all_cpu_backtrace_setup); #endif +#endif /* * Hard-lockup warnings should be triggered after just a few seconds. Soft- @@ -210,18 +292,6 @@ void touch_softlockup_watchdog_sync(void) __this_cpu_write(watchdog_touch_ts, 0); } -/* watchdog detector functions */ -bool is_hardlockup(void) -{ - unsigned long hrint = __this_cpu_read(hrtimer_interrupts); - - if (__this_cpu_read(hrtimer_interrupts_saved) == hrint) - return true; - - __this_cpu_write(hrtimer_interrupts_saved, hrint); - return false; -} - static int is_softlockup(unsigned long touch_ts) { unsigned long now = get_timestamp(); @@ -234,21 +304,21 @@ static int is_softlockup(unsigned long touch_ts) return 0; } -static void watchdog_interrupt_count(void) +/* watchdog detector functions */ +bool is_hardlockup(void) { - __this_cpu_inc(hrtimer_interrupts); -} + unsigned long hrint = __this_cpu_read(hrtimer_interrupts); -/* - * These two functions are mostly architecture specific - * defining them as weak here. - */ -int __weak watchdog_nmi_enable(unsigned int cpu) -{ - return 0; + if (__this_cpu_read(hrtimer_interrupts_saved) == hrint) + return true; + + __this_cpu_write(hrtimer_interrupts_saved, hrint); + return false; } -void __weak watchdog_nmi_disable(unsigned int cpu) + +static void watchdog_interrupt_count(void) { + __this_cpu_inc(hrtimer_interrupts); } static int watchdog_enable_all_cpus(void); @@ -499,57 +569,6 @@ static void watchdog_unpark_threads(void) kthread_unpark(per_cpu(softlockup_watchdog, cpu)); } -/* - * Suspend the hard and soft lockup detector by parking the watchdog threads. - */ -int lockup_detector_suspend(void) -{ - int ret = 0; - - get_online_cpus(); - mutex_lock(&watchdog_proc_mutex); - /* - * Multiple suspend requests can be active in parallel (counted by - * the 'watchdog_suspended' variable). If the watchdog threads are - * running, the first caller takes care that they will be parked. - * The state of 'watchdog_running' cannot change while a suspend - * request is active (see related code in 'proc' handlers). - */ - if (watchdog_running && !watchdog_suspended) - ret = watchdog_park_threads(); - - if (ret == 0) - watchdog_suspended++; - else { - watchdog_disable_all_cpus(); - pr_err("Failed to suspend lockup detectors, disabled\n"); - watchdog_enabled = 0; - } - - mutex_unlock(&watchdog_proc_mutex); - - return ret; -} - -/* - * Resume the hard and soft lockup detector by unparking the watchdog threads. - */ -void lockup_detector_resume(void) -{ - mutex_lock(&watchdog_proc_mutex); - - watchdog_suspended--; - /* - * The watchdog threads are unparked if they were previously running - * and if there is no more active suspend request. - */ - if (watchdog_running && !watchdog_suspended) - watchdog_unpark_threads(); - - mutex_unlock(&watchdog_proc_mutex); - put_online_cpus(); -} - static int update_watchdog_all_cpus(void) { int ret; @@ -602,6 +621,100 @@ static void watchdog_disable_all_cpus(void) } #ifdef CONFIG_SYSCTL +static int watchdog_update_cpus(void) +{ + return smpboot_update_cpumask_percpu_thread( + &watchdog_threads, &watchdog_cpumask); +} +#endif + +#else /* SOFTLOCKUP */ +static int watchdog_park_threads(void) +{ + return 0; +} + +static void watchdog_unpark_threads(void) +{ +} + +static int watchdog_enable_all_cpus(void) +{ + return 0; +} + +static void watchdog_disable_all_cpus(void) +{ +} + +#ifdef CONFIG_SYSCTL +static int watchdog_update_cpus(void) +{ + return 0; +} +#endif + +static void set_sample_period(void) +{ +} +#endif /* SOFTLOCKUP */ + +/* + * Suspend the hard and soft lockup detector by parking the watchdog threads. + */ +int lockup_detector_suspend(void) +{ + int ret = 0; + + get_online_cpus(); + mutex_lock(&watchdog_proc_mutex); + /* + * Multiple suspend requests can be active in parallel (counted by + * the 'watchdog_suspended' variable). If the watchdog threads are + * running, the first caller takes care that they will be parked. + * The state of 'watchdog_running' cannot change while a suspend + * request is active (see related code in 'proc' handlers). + */ + if (watchdog_running && !watchdog_suspended) + ret = watchdog_park_threads(); + + if (ret == 0) + watchdog_suspended++; + else { + watchdog_disable_all_cpus(); + pr_err("Failed to suspend lockup detectors, disabled\n"); + watchdog_enabled = 0; + } + + watchdog_nmi_reconfigure(); + + mutex_unlock(&watchdog_proc_mutex); + + return ret; +} + +/* + * Resume the hard and soft lockup detector by unparking the watchdog threads. + */ +void lockup_detector_resume(void) +{ + mutex_lock(&watchdog_proc_mutex); + + watchdog_suspended--; + /* + * The watchdog threads are unparked if they were previously running + * and if there is no more active suspend request. + */ + if (watchdog_running && !watchdog_suspended) + watchdog_unpark_threads(); + + watchdog_nmi_reconfigure(); + + mutex_unlock(&watchdog_proc_mutex); + put_online_cpus(); +} + +#ifdef CONFIG_SYSCTL /* * Update the run state of the lockup detectors. @@ -622,6 +735,8 @@ static int proc_watchdog_update(void) else watchdog_disable_all_cpus(); + watchdog_nmi_reconfigure(); + return err; } @@ -807,10 +922,11 @@ int proc_watchdog_cpumask(struct ctl_table *table, int write, * a temporary cpumask, so we are likely not in a * position to do much else to make things better. */ - if (smpboot_update_cpumask_percpu_thread( - &watchdog_threads, &watchdog_cpumask) != 0) + if (watchdog_update_cpus() != 0) pr_err("cpumask update failed\n"); } + + watchdog_nmi_reconfigure(); } out: mutex_unlock(&watchdog_proc_mutex); diff --git a/kernel/watchdog_hld.c b/kernel/watchdog_hld.c index 12b8dd640786..295a0d84934c 100644 --- a/kernel/watchdog_hld.c +++ b/kernel/watchdog_hld.c @@ -13,6 +13,8 @@ #include <linux/nmi.h> #include <linux/module.h> +#include <linux/sched/debug.h> + #include <asm/irq_regs.h> #include <linux/perf_event.h> @@ -20,41 +22,9 @@ static DEFINE_PER_CPU(bool, hard_watchdog_warn); static DEFINE_PER_CPU(bool, watchdog_nmi_touch); static DEFINE_PER_CPU(struct perf_event *, watchdog_ev); -/* boot commands */ -/* - * Should we panic when a soft-lockup or hard-lockup occurs: - */ -unsigned int __read_mostly hardlockup_panic = - CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE; static unsigned long hardlockup_allcpu_dumped; -/* - * We may not want to enable hard lockup detection by default in all cases, - * for example when running the kernel as a guest on a hypervisor. In these - * cases this function can be called to disable hard lockup detection. This - * function should only be executed once by the boot processor before the - * kernel command line parameters are parsed, because otherwise it is not - * possible to override this in hardlockup_panic_setup(). - */ -void hardlockup_detector_disable(void) -{ - watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; -} -static int __init hardlockup_panic_setup(char *str) -{ - if (!strncmp(str, "panic", 5)) - hardlockup_panic = 1; - else if (!strncmp(str, "nopanic", 7)) - hardlockup_panic = 0; - else if (!strncmp(str, "0", 1)) - watchdog_enabled &= ~NMI_WATCHDOG_ENABLED; - else if (!strncmp(str, "1", 1)) - watchdog_enabled |= NMI_WATCHDOG_ENABLED; - return 1; -} -__setup("nmi_watchdog=", hardlockup_panic_setup); - -void touch_nmi_watchdog(void) +void arch_touch_nmi_watchdog(void) { /* * Using __raw here because some code paths have @@ -64,9 +34,8 @@ void touch_nmi_watchdog(void) * going off. */ raw_cpu_write(watchdog_nmi_touch, true); - touch_softlockup_watchdog(); } -EXPORT_SYMBOL(touch_nmi_watchdog); +EXPORT_SYMBOL(arch_touch_nmi_watchdog); static struct perf_event_attr wd_hw_attr = { .type = PERF_TYPE_HARDWARE, @@ -137,12 +106,14 @@ static void watchdog_overflow_callback(struct perf_event *event, * Reduce the watchdog noise by only printing messages * that are different from what cpu0 displayed. */ -static unsigned long cpu0_err; +static unsigned long firstcpu_err; +static atomic_t watchdog_cpus; int watchdog_nmi_enable(unsigned int cpu) { struct perf_event_attr *wd_attr; struct perf_event *event = per_cpu(watchdog_ev, cpu); + int firstcpu = 0; /* nothing to do if the hard lockup detector is disabled */ if (!(watchdog_enabled & NMI_WATCHDOG_ENABLED)) @@ -156,19 +127,22 @@ int watchdog_nmi_enable(unsigned int cpu) if (event != NULL) goto out_enable; + if (atomic_inc_return(&watchdog_cpus) == 1) + firstcpu = 1; + wd_attr = &wd_hw_attr; wd_attr->sample_period = hw_nmi_get_sample_period(watchdog_thresh); /* Try to register using hardware perf events */ event = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback, NULL); - /* save cpu0 error for future comparision */ - if (cpu == 0 && IS_ERR(event)) - cpu0_err = PTR_ERR(event); + /* save the first cpu's error for future comparision */ + if (firstcpu && IS_ERR(event)) + firstcpu_err = PTR_ERR(event); if (!IS_ERR(event)) { - /* only print for cpu0 or different than cpu0 */ - if (cpu == 0 || cpu0_err) + /* only print for the first cpu initialized */ + if (firstcpu || firstcpu_err) pr_info("enabled on all CPUs, permanently consumes one hw-PMU counter.\n"); goto out_save; } @@ -186,7 +160,7 @@ int watchdog_nmi_enable(unsigned int cpu) smp_mb__after_atomic(); /* skip displaying the same error again */ - if (cpu > 0 && (PTR_ERR(event) == cpu0_err)) + if (!firstcpu && (PTR_ERR(event) == firstcpu_err)) return PTR_ERR(event); /* vary the KERN level based on the returned errno */ @@ -222,9 +196,9 @@ void watchdog_nmi_disable(unsigned int cpu) /* should be in cleanup, but blocks oprofile */ perf_event_release_kernel(event); - } - if (cpu == 0) { + /* watchdog_nmi_enable() expects this to be zero initially. */ - cpu0_err = 0; + if (atomic_dec_and_test(&watchdog_cpus)) + firstcpu_err = 0; } } diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1d9fb6543a66..a86688fabc55 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1507,6 +1507,7 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq, struct timer_list *timer = &dwork->timer; struct work_struct *work = &dwork->work; + WARN_ON_ONCE(!wq); WARN_ON_ONCE(timer->function != delayed_work_timer_fn || timer->data != (unsigned long)dwork); WARN_ON_ONCE(timer_pending(timer)); @@ -1523,8 +1524,6 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq, return; } - timer_stats_timer_set_start_info(&dwork->timer); - dwork->wq = wq; dwork->cpu = cpu; timer->expires = jiffies + delay; @@ -2865,11 +2864,11 @@ bool flush_work(struct work_struct *work) EXPORT_SYMBOL_GPL(flush_work); struct cwt_wait { - wait_queue_t wait; + wait_queue_entry_t wait; struct work_struct *work; }; -static int cwt_wakefn(wait_queue_t *wait, unsigned mode, int sync, void *key) +static int cwt_wakefn(wait_queue_entry_t *wait, unsigned mode, int sync, void *key) { struct cwt_wait *cwait = container_of(wait, struct cwt_wait, wait); @@ -3210,9 +3209,8 @@ static int init_worker_pool(struct worker_pool *pool) INIT_LIST_HEAD(&pool->idle_list); hash_init(pool->busy_hash); - init_timer_deferrable(&pool->idle_timer); - pool->idle_timer.function = idle_worker_timeout; - pool->idle_timer.data = (unsigned long)pool; + setup_deferrable_timer(&pool->idle_timer, idle_worker_timeout, + (unsigned long)pool); setup_timer(&pool->mayday_timer, pool_mayday_timeout, (unsigned long)pool); @@ -4736,6 +4734,29 @@ long work_on_cpu(int cpu, long (*fn)(void *), void *arg) return wfc.ret; } EXPORT_SYMBOL_GPL(work_on_cpu); + +/** + * work_on_cpu_safe - run a function in thread context on a particular cpu + * @cpu: the cpu to run on + * @fn: the function to run + * @arg: the function argument + * + * Disables CPU hotplug and calls work_on_cpu(). The caller must not hold + * any locks which would prevent @fn from completing. + * + * Return: The value @fn returns. + */ +long work_on_cpu_safe(int cpu, long (*fn)(void *), void *arg) +{ + long ret = -ENODEV; + + get_online_cpus(); + if (cpu_online(cpu)) + ret = work_on_cpu(cpu, fn, arg); + put_online_cpus(); + return ret; +} +EXPORT_SYMBOL_GPL(work_on_cpu_safe); #endif /* CONFIG_SMP */ #ifdef CONFIG_FREEZER |