From 436d34b36202ef724778ded1e9cb10f8c37b32bc Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 2 Apr 2008 17:28:46 +0900 Subject: libata: uninline atapi_cmd_type() Uninline atapi_cmd_type(). It doesn't really have to be inline and more case will be added which need to access unexported libata variable. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- include/linux/libata.h | 22 +--------------------- 1 file changed, 1 insertion(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/libata.h b/include/linux/libata.h index b064bfeb69ee..92c64909ed25 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -849,6 +849,7 @@ extern unsigned int ata_dev_try_classify(struct ata_device *dev, int present, */ extern void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf); extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf); +extern int atapi_cmd_type(u8 opcode); extern void ata_tf_to_fis(const struct ata_taskfile *tf, u8 pmp, int is_cmd, u8 *fis); extern void ata_tf_from_fis(const u8 *fis, struct ata_taskfile *tf); @@ -1379,27 +1380,6 @@ static inline int ata_try_flush_cache(const struct ata_device *dev) ata_id_has_flush_ext(dev->id); } -static inline int atapi_cmd_type(u8 opcode) -{ - switch (opcode) { - case GPCMD_READ_10: - case GPCMD_READ_12: - return ATAPI_READ; - - case GPCMD_WRITE_10: - case GPCMD_WRITE_12: - case GPCMD_WRITE_AND_VERIFY_10: - return ATAPI_WRITE; - - case GPCMD_READ_CD: - case GPCMD_READ_CD_MSF: - return ATAPI_READ_CD; - - default: - return ATAPI_MISC; - } -} - static inline unsigned int ac_err_mask(u8 status) { if (status & (ATA_BUSY | ATA_DRQ)) -- cgit From e52dcc4899cf1b7601379c31542bd91cd2997a64 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 2 Apr 2008 17:35:19 +0900 Subject: libata: ATA_12/16 doesn't fall into ATAPI_MISC SAT passthrus don't really fit into ATAPI_MISC class. SAT passthru commands always transfer multiple of 512 bytes and variable length response is not allowed. This patch creates a separate category - ATAPI_PASS_THRU - for these. This fixes HSM violation on "hdparm -I". Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 5 +++++ include/linux/libata.h | 3 ++- 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index e9b69ba489d2..be95fdb69726 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -353,6 +353,11 @@ int atapi_cmd_type(u8 opcode) case GPCMD_READ_CD_MSF: return ATAPI_READ_CD; + case ATA_16: + case ATA_12: + if (atapi_passthru16) + return ATAPI_PASS_THRU; + /* fall thru */ default: return ATAPI_MISC; } diff --git a/include/linux/libata.h b/include/linux/libata.h index 92c64909ed25..37ee881c42ac 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -350,7 +350,8 @@ enum { ATAPI_READ = 0, /* READs */ ATAPI_WRITE = 1, /* WRITEs */ ATAPI_READ_CD = 2, /* READ CD [MSF] */ - ATAPI_MISC = 3, /* the rest */ + ATAPI_PASS_THRU = 3, /* SAT pass-thru */ + ATAPI_MISC = 4, /* the rest */ }; enum ata_xfer_mask { -- cgit From 5761d64b277c287a7520b868c32d656ef03374b4 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 4 Apr 2008 16:26:10 +0200 Subject: x86: revert assign IRQs to hpet timer The commits: commit 37a47db8d7f0f38dac5acf5a13abbc8f401707fa Author: Balaji Rao Date: Wed Jan 30 13:30:03 2008 +0100 x86: assign IRQs to HPET timers, fix and commit e3f37a54f690d3e64995ea7ecea08c5ab3070faf Author: Balaji Rao Date: Wed Jan 30 13:30:03 2008 +0100 x86: assign IRQs to HPET timers have been identified to cause a regression on some platforms due to the assignement of legacy IRQs which makes the legacy devices connected to those IRQs disfunctional. Revert them. This fixes http://bugzilla.kernel.org/show_bug.cgi?id=10382 Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 9 ++++++--- drivers/char/hpet.c | 51 +++++++------------------------------------------- include/linux/hpet.h | 2 +- 3 files changed, 14 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 235fd6c77504..36652ea1a265 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -133,13 +133,16 @@ static void hpet_reserve_platform_timers(unsigned long id) #ifdef CONFIG_HPET_EMULATE_RTC hpet_reserve_timer(&hd, 1); #endif + hd.hd_irq[0] = HPET_LEGACY_8254; hd.hd_irq[1] = HPET_LEGACY_RTC; - for (i = 2; i < nrtimers; timer++, i++) - hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >> - Tn_INT_ROUTE_CNF_SHIFT; + for (i = 2; i < nrtimers; timer++, i++) + hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >> + Tn_INT_ROUTE_CNF_SHIFT; + hpet_alloc(&hd); + } #else static void hpet_reserve_platform_timers(unsigned long id) { } diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c index 465ad35ed38f..1399971be689 100644 --- a/drivers/char/hpet.c +++ b/drivers/char/hpet.c @@ -731,14 +731,14 @@ static unsigned long hpet_calibrate(struct hpets *hpetp) int hpet_alloc(struct hpet_data *hdp) { - u64 cap, mcfg, hpet_config; + u64 cap, mcfg; struct hpet_dev *devp; - u32 i, ntimer, irq; + u32 i, ntimer; struct hpets *hpetp; size_t siz; struct hpet __iomem *hpet; static struct hpets *last = NULL; - unsigned long period, irq_bitmap; + unsigned long period; unsigned long long temp; /* @@ -765,47 +765,11 @@ int hpet_alloc(struct hpet_data *hdp) hpetp->hp_hpet_phys = hdp->hd_phys_address; hpetp->hp_ntimer = hdp->hd_nirqs; - hpet = hpetp->hp_hpet; - - /* Assign IRQs statically for legacy devices */ - hpetp->hp_dev[0].hd_hdwirq = hdp->hd_irq[0]; - hpetp->hp_dev[1].hd_hdwirq = hdp->hd_irq[1]; - - /* Assign IRQs dynamically for the others */ - for (i = 2, devp = &hpetp->hp_dev[2]; i < hdp->hd_nirqs; i++, devp++) { - struct hpet_timer __iomem *timer; - timer = &hpet->hpet_timers[devp - hpetp->hp_dev]; + for (i = 0; i < hdp->hd_nirqs; i++) + hpetp->hp_dev[i].hd_hdwirq = hdp->hd_irq[i]; - /* Check if there's already an IRQ assigned to the timer */ - if (hdp->hd_irq[i]) { - hpetp->hp_dev[i].hd_hdwirq = hdp->hd_irq[i]; - continue; - } - - hpet_config = readq(&timer->hpet_config); - irq_bitmap = (hpet_config & Tn_INT_ROUTE_CAP_MASK) - >> Tn_INT_ROUTE_CAP_SHIFT; - if (!irq_bitmap) - irq = 0; /* No valid IRQ Assignable */ - else { - irq = find_first_bit(&irq_bitmap, 32); - do { - hpet_config |= irq << Tn_INT_ROUTE_CNF_SHIFT; - writeq(hpet_config, &timer->hpet_config); - - /* - * Verify whether we have written a valid - * IRQ number by reading it back again - */ - hpet_config = readq(&timer->hpet_config); - if (irq == (hpet_config & Tn_INT_ROUTE_CNF_MASK) - >> Tn_INT_ROUTE_CNF_SHIFT) - break; /* Success */ - } while ((irq = (find_next_bit(&irq_bitmap, 32, irq)))); - } - hpetp->hp_dev[i].hd_hdwirq = irq; - } + hpet = hpetp->hp_hpet; cap = readq(&hpet->hpet_cap); @@ -836,8 +800,7 @@ int hpet_alloc(struct hpet_data *hdp) hpetp->hp_which, hdp->hd_phys_address, hpetp->hp_ntimer > 1 ? "s" : ""); for (i = 0; i < hpetp->hp_ntimer; i++) - printk("%s %d", i > 0 ? "," : "", - hpetp->hp_dev[i].hd_hdwirq); + printk("%s %d", i > 0 ? "," : "", hdp->hd_irq[i]); printk("\n"); printk(KERN_INFO "hpet%u: %u %d-bit timers, %Lu Hz\n", diff --git a/include/linux/hpet.h b/include/linux/hpet.h index 9cd94bfd07e5..2dc29ce6c8e4 100644 --- a/include/linux/hpet.h +++ b/include/linux/hpet.h @@ -64,7 +64,7 @@ struct hpet { */ #define Tn_INT_ROUTE_CAP_MASK (0xffffffff00000000ULL) -#define Tn_INT_ROUTE_CAP_SHIFT (32UL) +#define Tn_INI_ROUTE_CAP_SHIFT (32UL) #define Tn_FSB_INT_DELCAP_MASK (0x8000UL) #define Tn_FSB_INT_DELCAP_SHIFT (15) #define Tn_FSB_EN_CNF_MASK (0x4000UL) -- cgit From 8bab8dded67d026c39367bbd5e27d2f6c556c38e Mon Sep 17 00:00:00 2001 From: Paul Menage Date: Fri, 4 Apr 2008 14:29:57 -0700 Subject: cgroups: add cgroup support for enabling controllers at boot time The effects of cgroup_disable=foo are: - foo isn't auto-mounted if you mount all cgroups in a single hierarchy - foo isn't visible as an individually mountable subsystem As a result there will only ever be one call to foo->create(), at init time; all processes will stay in this group, and the group will never be mounted on a visible hierarchy. Any additional effects (e.g. not allocating metadata) are up to the foo subsystem. This doesn't handle early_init subsystems (their "disabled" bit isn't set be, but it could easily be extended to do so if any of the early_init systems wanted it - I think it would just involve some nastier parameter processing since it would occur before the command-line argument parser had been run. Hugh said: Ballpark figures, I'm trying to get this question out rather than processing the exact numbers: CONFIG_CGROUP_MEM_RES_CTLR adds 15% overhead to the affected paths, booting with cgroup_disable=memory cuts that back to 1% overhead (due to slightly bigger struct page). I'm no expert on distros, they may have no interest whatever in CONFIG_CGROUP_MEM_RES_CTLR=y; and the rest of us can easily build with or without it, or apply the cgroup_disable=memory patches. Unix bench's execl test result on x86_64 was == just after boot without mounting any cgroup fs.== mem_cgorup=off : Execl Throughput 43.0 3150.1 732.6 mem_cgroup=on : Execl Throughput 43.0 2932.6 682.0 == [lizf@cn.fujitsu.com: fix boot option parsing] Signed-off-by: Balbir Singh Cc: Paul Menage Cc: Balbir Singh Cc: Pavel Emelyanov Cc: KAMEZAWA Hiroyuki Cc: Hugh Dickins Cc: Sudhir Kumar Cc: YAMAMOTO Takashi Cc: David Rientjes Signed-off-by: Li Zefan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 4 ++++ include/linux/cgroup.h | 1 + kernel/cgroup.c | 42 ++++++++++++++++++++++++++++++++----- 3 files changed, 42 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 4cd1a5da80a4..32e9297ef747 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -375,6 +375,10 @@ and is between 256 and 4096 characters. It is defined in the file ccw_timeout_log [S390] See Documentation/s390/CommonIO for details. + cgroup_disable= [KNL] Disable a particular controller + Format: {name of the controller(s) to disable} + {Currently supported controllers - "memory"} + checkreqprot [SELINUX] Set initial checkreqprot flag value. Format: { "0" | "1" } See security/selinux/Kconfig help text. diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 028ba3b523b1..a6a6035a4e1e 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -256,6 +256,7 @@ struct cgroup_subsys { void (*bind)(struct cgroup_subsys *ss, struct cgroup *root); int subsys_id; int active; + int disabled; int early_init; #define MAX_CGROUP_TYPE_NAMELEN 32 const char *name; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 53d86b4b0ce0..62f1a5231fe9 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -782,7 +782,14 @@ static int parse_cgroupfs_options(char *data, if (!*token) return -EINVAL; if (!strcmp(token, "all")) { - opts->subsys_bits = (1 << CGROUP_SUBSYS_COUNT) - 1; + /* Add all non-disabled subsystems */ + int i; + opts->subsys_bits = 0; + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + if (!ss->disabled) + opts->subsys_bits |= 1ul << i; + } } else if (!strcmp(token, "noprefix")) { set_bit(ROOT_NOPREFIX, &opts->flags); } else if (!strncmp(token, "release_agent=", 14)) { @@ -800,7 +807,8 @@ static int parse_cgroupfs_options(char *data, for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { ss = subsys[i]; if (!strcmp(token, ss->name)) { - set_bit(i, &opts->subsys_bits); + if (!ss->disabled) + set_bit(i, &opts->subsys_bits); break; } } @@ -2600,13 +2608,13 @@ static int proc_cgroupstats_show(struct seq_file *m, void *v) { int i; - seq_puts(m, "#subsys_name\thierarchy\tnum_cgroups\n"); + seq_puts(m, "#subsys_name\thierarchy\tnum_cgroups\tenabled\n"); mutex_lock(&cgroup_mutex); for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; - seq_printf(m, "%s\t%lu\t%d\n", + seq_printf(m, "%s\t%lu\t%d\t%d\n", ss->name, ss->root->subsys_bits, - ss->root->number_of_cgroups); + ss->root->number_of_cgroups, !ss->disabled); } mutex_unlock(&cgroup_mutex); return 0; @@ -3010,3 +3018,27 @@ static void cgroup_release_agent(struct work_struct *work) spin_unlock(&release_list_lock); mutex_unlock(&cgroup_mutex); } + +static int __init cgroup_disable(char *str) +{ + int i; + char *token; + + while ((token = strsep(&str, ",")) != NULL) { + if (!*token) + continue; + + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + + if (!strcmp(token, ss->name)) { + ss->disabled = 1; + printk(KERN_INFO "Disabling %s control group" + " subsystem\n", ss->name); + break; + } + } + } + return 1; +} +__setup("cgroup_disable=", cgroup_disable); -- cgit From 2557a933b795c1988c721ebb871cd735128bb9cb Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 7 Apr 2008 14:30:28 +1000 Subject: virtio: remove overzealous BUG_ON. The 'disable_cb' callback is designed as an optimization to tell the host we don't need callbacks now. As it is not reliable, the debug check is overzealous: it can happen on two CPUs at the same time. Document this. Even if it were reliable, the virtio_net driver doesn't disable callbacks on transmit so the START_USE/END_USE debugging reentrance protection can be easily tripped even on UP. Thanks to Balaji Rao for the bug report and testing. Signed-off-by: Rusty Russell CC: Balaji Rao Signed-off-by: Linus Torvalds --- drivers/virtio/virtio_ring.c | 3 --- include/linux/virtio.h | 5 ++++- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index aa714028641e..c2fa5c630813 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -214,10 +214,7 @@ static void vring_disable_cb(struct virtqueue *_vq) { struct vring_virtqueue *vq = to_vvq(_vq); - START_USE(vq); - BUG_ON(vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT); vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT; - END_USE(vq); } static bool vring_enable_cb(struct virtqueue *_vq) diff --git a/include/linux/virtio.h b/include/linux/virtio.h index 12c18ac1b973..e7d10845b3c1 100644 --- a/include/linux/virtio.h +++ b/include/linux/virtio.h @@ -41,6 +41,8 @@ struct virtqueue * Returns NULL or the "data" token handed to add_buf. * @disable_cb: disable callbacks * vq: the struct virtqueue we're talking about. + * Note that this is not necessarily synchronous, hence unreliable and only + * useful as an optimization. * @enable_cb: restart callbacks after disable_cb. * vq: the struct virtqueue we're talking about. * This re-enables callbacks; it returns "false" if there are pending @@ -48,7 +50,8 @@ struct virtqueue * checking for more work, and enabling callbacks. * * Locking rules are straightforward: the driver is responsible for - * locking. No two operations may be invoked simultaneously. + * locking. No two operations may be invoked simultaneously, with the exception + * of @disable_cb. * * All operations can be called in any context. */ -- cgit From 54a015104136974262afa4b8ddd943ea70dec8a2 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 10 Apr 2008 15:37:38 -0700 Subject: asmlinkage_protect replaces prevent_tail_call The prevent_tail_call() macro works around the problem of the compiler clobbering argument words on the stack, which for asmlinkage functions is the caller's (user's) struct pt_regs. The tail/sibling-call optimization is not the only way that the compiler can decide to use stack argument words as scratch space, which we have to prevent. Other optimizations can do it too. Until we have new compiler support to make "asmlinkage" binding on the compiler's own use of the stack argument frame, we have work around all the manifestations of this issue that crop up. More cases seem to be prevented by also keeping the incoming argument variables live at the end of the function. This makes their original stack slots attractive places to leave those variables, so the compiler tends not clobber them for something else. It's still no guarantee, but it handles some observed cases that prevent_tail_call() did not. Signed-off-by: Roland McGrath Signed-off-by: Linus Torvalds --- arch/x86/kernel/tls.c | 4 ++-- fs/open.c | 8 ++++---- include/asm-x86/linkage.h | 24 +++++++++++++++++++++++- include/linux/linkage.h | 4 ++-- kernel/exit.c | 4 ++-- kernel/uid16.c | 22 +++++++++++----------- 6 files changed, 44 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c index 022bcaa3b42e..ab6bf375a307 100644 --- a/arch/x86/kernel/tls.c +++ b/arch/x86/kernel/tls.c @@ -92,7 +92,7 @@ int do_set_thread_area(struct task_struct *p, int idx, asmlinkage int sys_set_thread_area(struct user_desc __user *u_info) { int ret = do_set_thread_area(current, -1, u_info, 1); - prevent_tail_call(ret); + asmlinkage_protect(1, ret, u_info); return ret; } @@ -142,7 +142,7 @@ int do_get_thread_area(struct task_struct *p, int idx, asmlinkage int sys_get_thread_area(struct user_desc __user *u_info) { int ret = do_get_thread_area(current, -1, u_info); - prevent_tail_call(ret); + asmlinkage_protect(1, ret, u_info); return ret; } diff --git a/fs/open.c b/fs/open.c index a4b12022edaa..3fa4e4ffce4c 100644 --- a/fs/open.c +++ b/fs/open.c @@ -335,7 +335,7 @@ asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length) { long ret = do_sys_ftruncate(fd, length, 1); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } @@ -350,7 +350,7 @@ asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length) { long ret = do_sys_ftruncate(fd, length, 0); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } #endif @@ -1067,7 +1067,7 @@ asmlinkage long sys_open(const char __user *filename, int flags, int mode) ret = do_sys_open(AT_FDCWD, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, flags, mode); return ret; } @@ -1081,7 +1081,7 @@ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, ret = do_sys_open(dfd, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, dfd, filename, flags, mode); return ret; } diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index 31739c7d66a9..d605eeba0f70 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -8,12 +8,34 @@ #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) -#define prevent_tail_call(ret) __asm__ ("" : "=r" (ret) : "0" (ret)) /* * For 32-bit UML - mark functions implemented in assembly that use * regparm input parameters: */ #define asmregparm __attribute__((regparm(3))) + +#define asmlinkage_protect(n, ret, args...) \ + __asmlinkage_protect##n(ret, ##args) +#define __asmlinkage_protect_n(ret, args...) \ + __asm__ __volatile__ ("" : "=r" (ret) : "0" (ret), ##args) +#define __asmlinkage_protect0(ret) \ + __asmlinkage_protect_n(ret) +#define __asmlinkage_protect1(ret, arg1) \ + __asmlinkage_protect_n(ret, "g" (arg1)) +#define __asmlinkage_protect2(ret, arg1, arg2) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2)) +#define __asmlinkage_protect3(ret, arg1, arg2, arg3) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3)) +#define __asmlinkage_protect4(ret, arg1, arg2, arg3, arg4) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4)) +#define __asmlinkage_protect5(ret, arg1, arg2, arg3, arg4, arg5) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4), "g" (arg5)) +#define __asmlinkage_protect6(ret, arg1, arg2, arg3, arg4, arg5, arg6) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4), "g" (arg5), "g" (arg6)) + #endif #ifdef CONFIG_X86_ALIGNMENT_16 diff --git a/include/linux/linkage.h b/include/linux/linkage.h index 0592936344c4..fe2a39c489b6 100644 --- a/include/linux/linkage.h +++ b/include/linux/linkage.h @@ -17,8 +17,8 @@ # define asmregparm #endif -#ifndef prevent_tail_call -# define prevent_tail_call(ret) do { } while (0) +#ifndef asmlinkage_protect +# define asmlinkage_protect(n, ret, args...) do { } while (0) #endif #ifndef __ALIGN diff --git a/kernel/exit.c b/kernel/exit.c index 53872bf993fa..073005b1cfb2 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -1608,7 +1608,7 @@ asmlinkage long sys_waitid(int which, pid_t upid, put_pid(pid); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(5, ret, which, upid, infop, options, ru); return ret; } @@ -1640,7 +1640,7 @@ asmlinkage long sys_wait4(pid_t upid, int __user *stat_addr, put_pid(pid); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, upid, stat_addr, options, ru); return ret; } diff --git a/kernel/uid16.c b/kernel/uid16.c index dd308ba4e03b..3e41c1673e2f 100644 --- a/kernel/uid16.c +++ b/kernel/uid16.c @@ -21,7 +21,7 @@ asmlinkage long sys_chown16(const char __user * filename, old_uid_t user, old_gi { long ret = sys_chown(filename, low2highuid(user), low2highgid(group)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, user, group); return ret; } @@ -29,7 +29,7 @@ asmlinkage long sys_lchown16(const char __user * filename, old_uid_t user, old_g { long ret = sys_lchown(filename, low2highuid(user), low2highgid(group)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, user, group); return ret; } @@ -37,7 +37,7 @@ asmlinkage long sys_fchown16(unsigned int fd, old_uid_t user, old_gid_t group) { long ret = sys_fchown(fd, low2highuid(user), low2highgid(group)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, fd, user, group); return ret; } @@ -45,7 +45,7 @@ asmlinkage long sys_setregid16(old_gid_t rgid, old_gid_t egid) { long ret = sys_setregid(low2highgid(rgid), low2highgid(egid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, rgid, egid); return ret; } @@ -53,7 +53,7 @@ asmlinkage long sys_setgid16(old_gid_t gid) { long ret = sys_setgid(low2highgid(gid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, gid); return ret; } @@ -61,7 +61,7 @@ asmlinkage long sys_setreuid16(old_uid_t ruid, old_uid_t euid) { long ret = sys_setreuid(low2highuid(ruid), low2highuid(euid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, ruid, euid); return ret; } @@ -69,7 +69,7 @@ asmlinkage long sys_setuid16(old_uid_t uid) { long ret = sys_setuid(low2highuid(uid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, uid); return ret; } @@ -78,7 +78,7 @@ asmlinkage long sys_setresuid16(old_uid_t ruid, old_uid_t euid, old_uid_t suid) long ret = sys_setresuid(low2highuid(ruid), low2highuid(euid), low2highuid(suid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, ruid, euid, suid); return ret; } @@ -98,7 +98,7 @@ asmlinkage long sys_setresgid16(old_gid_t rgid, old_gid_t egid, old_gid_t sgid) long ret = sys_setresgid(low2highgid(rgid), low2highgid(egid), low2highgid(sgid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, rgid, egid, sgid); return ret; } @@ -117,7 +117,7 @@ asmlinkage long sys_setfsuid16(old_uid_t uid) { long ret = sys_setfsuid(low2highuid(uid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, uid); return ret; } @@ -125,7 +125,7 @@ asmlinkage long sys_setfsgid16(old_gid_t gid) { long ret = sys_setfsgid(low2highgid(gid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, gid); return ret; } -- cgit From d10d89ec78114f925f63c5126a2b2490f501a462 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 10 Apr 2008 17:35:23 -0700 Subject: Add commentary about the new "asmlinkage_protect()" macro It's really a pretty ugly thing to need, and some day it will hopefully be obviated by teaching gcc about the magic calling conventions for the low-level system call code, but in the meantime we can at least add big honking comments about why we need these insane and strange macros. I took my comments from my version of the macro, but I ended up deciding to just pick Roland's version of the actual code instead (with his prettier syntax that uses vararg macros). Thus the previous two commits that actually implement it. Signed-off-by: Linus Torvalds --- include/asm-x86/linkage.h | 11 +++++++++++ include/linux/linkage.h | 13 +++++++++++++ 2 files changed, 24 insertions(+) (limited to 'include/linux') diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index d605eeba0f70..c048353f4b85 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -14,6 +14,17 @@ */ #define asmregparm __attribute__((regparm(3))) +/* + * Make sure the compiler doesn't do anything stupid with the + * arguments on the stack - they are owned by the *caller*, not + * the callee. This just fools gcc into not spilling into them, + * and keeps it from doing tailcall recursion and/or using the + * stack slots for temporaries, since they are live and "used" + * all the way to the end of the function. + * + * NOTE! On x86-64, all the arguments are in registers, so this + * only matters on a 32-bit kernel. + */ #define asmlinkage_protect(n, ret, args...) \ __asmlinkage_protect##n(ret, ##args) #define __asmlinkage_protect_n(ret, args...) \ diff --git a/include/linux/linkage.h b/include/linux/linkage.h index fe2a39c489b6..b163c5c40dbc 100644 --- a/include/linux/linkage.h +++ b/include/linux/linkage.h @@ -17,6 +17,19 @@ # define asmregparm #endif +/* + * This is used by architectures to keep arguments on the stack + * untouched by the compiler by keeping them live until the end. + * The argument stack may be owned by the assembly-language + * caller, not the callee, and gcc doesn't always understand + * that. + * + * We have the return value, and a maximum of six arguments. + * + * This should always be followed by a "return ret" for the + * protection to work (ie no more work that the compiler might + * end up needing stack temporaries for). + */ #ifndef asmlinkage_protect # define asmlinkage_protect(n, ret, args...) do { } while (0) #endif -- cgit From 544451a1a36b06f43fc67112ede3b92e6203eb9d Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 10 Apr 2008 21:29:28 -0700 Subject: pnp: increase number of devices supported per protocol Increase the PNP "number of devices" limit. We currently use an unsigned char, which limits us to 256 devices per protocol. This patch changes that to an unsigned int. Not all backends can take advantage of this: we limit ISAPNP to 10 devices in isapnp_cfg_begin(), and PNPBIOS is limited to 256 devices because the BIOS interfaces use a one-byte device node number. But there is no limit on the number of PNPACPI devices we may have. Large HP Integrity machines have more than 256, which causes the current "unsigned char number" to wrap around. This causes errors like this: pnp: PnP ACPI init kobject_add failed for 00:00 with -EEXIST, don't try to register things with the same name in the same directory. Call Trace: [] show_stack+0x40/0xa0 [] dump_stack+0x30/0x60 [] kobject_add+0x290/0x2c0 [] device_add+0x160/0x860 [] device_register+0x30/0x60 [] __pnp_add_device+0x130/0x180 [] pnp_add_device+0xb0/0xe0 [] pnpacpi_add_device+0x510/0x5a0 [] pnpacpi_add_device_handler+0x50/0x80 This patch increases the limit to fix this PNPACPI problem. It should not have any adverse effect on ISAPNP or PNPBIOS because their limits are still enforced in the backends. Signed-off-by: Bjorn Helgaas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pnp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pnp.h b/include/linux/pnp.h index 29dd55838e84..b2f05c230f4b 100644 --- a/include/linux/pnp.h +++ b/include/linux/pnp.h @@ -175,7 +175,7 @@ static inline void pnp_set_card_drvdata(struct pnp_card_link *pcard, void *data) struct pnp_dev { struct device dev; /* Driver Model device interface */ u64 dma_mask; - unsigned char number; /* used as an index, must be unique */ + unsigned int number; /* used as an index, must be unique */ int status; struct list_head global_list; /* node in global list of devices */ -- cgit From b0fac02370cffad956ff3de5e8ed4df7e7b875d7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 11 Apr 2008 13:46:54 +0200 Subject: Fix "$(AS) -traditional" compile breakage caused by asmlinkage_protect git commit 54a015104136974262afa4b8ddd943ea70dec8a2 ("asmlinkage_protect replaces prevent_tail_call") causes this build failure on s390: AS arch/s390/kernel/entry64.o In file included from arch/s390/kernel/entry64.S:14: include/linux/linkage.h:34: error: syntax error in macro parameter list make[1]: *** [arch/s390/kernel/entry64.o] Error 1 make: *** [arch/s390/kernel] Error 2 and some other architectures. The reason is that some architectures add the "-traditional" flag to the invocation of $(AS), which disables variadic macro argument support. So just surround the new define with an #ifndef __ASSEMBLY__ to prevent any side effects on asm code. Cc: Roland McGrath Cc: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Linus Torvalds --- include/linux/linkage.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/linkage.h b/include/linux/linkage.h index b163c5c40dbc..2119610b24f8 100644 --- a/include/linux/linkage.h +++ b/include/linux/linkage.h @@ -30,9 +30,12 @@ * protection to work (ie no more work that the compiler might * end up needing stack temporaries for). */ +/* Assembly files may be compiled with -traditional .. */ +#ifndef __ASSEMBLY__ #ifndef asmlinkage_protect # define asmlinkage_protect(n, ret, args...) do { } while (0) #endif +#endif #ifndef __ALIGN #define __ALIGN .align 4,0x90 -- cgit From dc07e721a26ec7e0adb66340f1763d220cfbbd0c Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 7 Apr 2008 15:59:05 -0400 Subject: Spell out behavior of atomic_dec_and_lock() in kerneldoc A little more detail here wouldn't hurt. Signed-off-by: J. Bruce Fields Signed-off-by: Jonathan Corbet --- include/linux/spinlock.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 576a5f77d3bd..1129ee0a7180 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -341,6 +341,9 @@ static inline void double_spin_unlock(spinlock_t *l1, spinlock_t *l2, * atomic_dec_and_lock - lock on reaching reference count zero * @atomic: the atomic counter * @lock: the spinlock in question + * + * Decrements @atomic by 1. If the result is 0, returns true and locks + * @lock. Returns false for all other cases. */ extern int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock); #define atomic_dec_and_lock(atomic, lock) \ -- cgit From 14daa02139dcb3193b2b0250c0720a23ef610c49 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 12 Apr 2008 18:48:58 -0700 Subject: net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell Acked-by: Max Krasnyanskiy Signed-off-by: David S. Miller --- drivers/net/tun.c | 33 +++++++++++++++++++++++++++++++++ include/linux/Kbuild | 2 +- include/linux/if_tun.h | 39 --------------------------------------- 3 files changed, 34 insertions(+), 40 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 7b816a032957..970ec4793442 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -67,9 +67,42 @@ #include #include +/* Uncomment to enable debugging */ +/* #define TUN_DEBUG 1 */ + #ifdef TUN_DEBUG static int debug; + +#define DBG if(tun->debug)printk +#define DBG1 if(debug==2)printk +#else +#define DBG( a... ) +#define DBG1( a... ) +#endif + +struct tun_struct { + struct list_head list; + unsigned long flags; + int attached; + uid_t owner; + gid_t group; + + wait_queue_head_t read_wait; + struct sk_buff_head readq; + + struct net_device *dev; + + struct fasync_struct *fasync; + + unsigned long if_flags; + u8 dev_addr[ETH_ALEN]; + u32 chr_filter[2]; + u32 net_filter[2]; + +#ifdef TUN_DEBUG + int debug; #endif +}; /* Network device part of the driver */ diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 9cdd12a9e843..cedbbd806bf6 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -86,6 +86,7 @@ header-y += if_plip.h header-y += if_ppp.h header-y += if_slip.h header-y += if_strip.h +header-y += if_tun.h header-y += if_tunnel.h header-y += in6.h header-y += in_route.h @@ -229,7 +230,6 @@ unifdef-y += if_link.h unifdef-y += if_pppol2tp.h unifdef-y += if_pppox.h unifdef-y += if_tr.h -unifdef-y += if_tun.h unifdef-y += if_vlan.h unifdef-y += if_wanpipe.h unifdef-y += igmp.h diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h index 72f1c5f47be3..8c71fe2fb1f5 100644 --- a/include/linux/if_tun.h +++ b/include/linux/if_tun.h @@ -18,47 +18,8 @@ #ifndef __IF_TUN_H #define __IF_TUN_H -/* Uncomment to enable debugging */ -/* #define TUN_DEBUG 1 */ - #include -#ifdef __KERNEL__ - -#ifdef TUN_DEBUG -#define DBG if(tun->debug)printk -#define DBG1 if(debug==2)printk -#else -#define DBG( a... ) -#define DBG1( a... ) -#endif - -struct tun_struct { - struct list_head list; - unsigned long flags; - int attached; - uid_t owner; - gid_t group; - - wait_queue_head_t read_wait; - struct sk_buff_head readq; - - struct net_device *dev; - - struct fasync_struct *fasync; - - unsigned long if_flags; - u8 dev_addr[ETH_ALEN]; - u32 chr_filter[2]; - u32 net_filter[2]; - -#ifdef TUN_DEBUG - int debug; -#endif -}; - -#endif /* __KERNEL__ */ - /* Read queue size */ #define TUN_READQ_SIZE 500 -- cgit From 7cd636fe9ce5de0051c11283911baca4291868c8 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sun, 13 Apr 2008 23:33:52 -0700 Subject: [IPV6]: IPv6 extension header structures need to be packed. struct ipv6_opt_hdr is the common structure for IPv6 extension headers, and it is common to increment the pointer to get the real content. On the other hand, since the structure consists only of 1-byte next-header field and 1-byte length field, size of that structure depends on architecture; 2 or 4. Add "packed" attribute to get 2. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/linux/ipv6.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 4aaefc349a4b..134c8e5cf07c 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -53,7 +53,7 @@ struct ipv6_opt_hdr { /* * TLV encoded option data follows. */ -}; +} __attribute__ ((packed)); /* required for some archs */ #define ipv6_destopt_hdr ipv6_opt_hdr #define ipv6_hopopt_hdr ipv6_opt_hdr -- cgit