aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rw-r--r--arch/x86/include/asm/msr.h2
-rw-r--r--arch/x86/include/asm/paravirt.h2
-rw-r--r--arch/x86/include/asm/uprobes.h2
-rw-r--r--arch/x86/kernel/cpu/perf_event.c15
-rw-r--r--arch/x86/kernel/cpu/perf_event.h15
-rw-r--r--arch/x86/kernel/cpu/perf_event_intel.c2
-rw-r--r--arch/x86/kernel/cpu/perf_event_intel_ds.c8
-rw-r--r--arch/x86/kernel/uprobes.c3
-rw-r--r--include/linux/perf_event.h1
-rw-r--r--include/linux/sched.h1
-rw-r--r--kernel/events/uprobes.c110
11 files changed, 96 insertions, 65 deletions
diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h
index 084ef95274cd..e489c1475be9 100644
--- a/arch/x86/include/asm/msr.h
+++ b/arch/x86/include/asm/msr.h
@@ -237,6 +237,8 @@ do { \
(high) = (u32)(_l >> 32); \
} while (0)
+#define rdpmcl(counter, val) ((val) = native_read_pmc(counter))
+
#define rdtscp(low, high, aux) \
do { \
unsigned long long _val = native_read_tscp(&(aux)); \
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index 6cbbabf52707..14ce05dfe04e 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -252,6 +252,8 @@ do { \
high = _l >> 32; \
} while (0)
+#define rdpmcl(counter, val) ((val) = paravirt_read_pmc(counter))
+
static inline unsigned long long paravirt_rdtscp(unsigned int *aux)
{
return PVOP_CALL1(u64, pv_cpu_ops.read_tscp, aux);
diff --git a/arch/x86/include/asm/uprobes.h b/arch/x86/include/asm/uprobes.h
index 1e9bed14f7ae..f3971bbcd1de 100644
--- a/arch/x86/include/asm/uprobes.h
+++ b/arch/x86/include/asm/uprobes.h
@@ -48,7 +48,7 @@ struct arch_uprobe_task {
#endif
};
-extern int arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm);
+extern int arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm, unsigned long addr);
extern int arch_uprobe_pre_xol(struct arch_uprobe *aup, struct pt_regs *regs);
extern int arch_uprobe_post_xol(struct arch_uprobe *aup, struct pt_regs *regs);
extern bool arch_uprobe_xol_was_trapped(struct task_struct *tsk);
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index c4706cf9c011..000a4746c7ce 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -35,17 +35,6 @@
#include "perf_event.h"
-#if 0
-#undef wrmsrl
-#define wrmsrl(msr, val) \
-do { \
- trace_printk("wrmsrl(%lx, %lx)\n", (unsigned long)(msr),\
- (unsigned long)(val)); \
- native_write_msr((msr), (u32)((u64)(val)), \
- (u32)((u64)(val) >> 32)); \
-} while (0)
-#endif
-
struct x86_pmu x86_pmu __read_mostly;
DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
@@ -86,7 +75,7 @@ u64 x86_perf_event_update(struct perf_event *event)
*/
again:
prev_raw_count = local64_read(&hwc->prev_count);
- rdmsrl(hwc->event_base, new_raw_count);
+ rdpmcl(hwc->event_base_rdpmc, new_raw_count);
if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
new_raw_count) != prev_raw_count)
@@ -830,9 +819,11 @@ static inline void x86_assign_hw_event(struct perf_event *event,
} else if (hwc->idx >= X86_PMC_IDX_FIXED) {
hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
hwc->event_base = MSR_ARCH_PERFMON_FIXED_CTR0 + (hwc->idx - X86_PMC_IDX_FIXED);
+ hwc->event_base_rdpmc = (hwc->idx - X86_PMC_IDX_FIXED) | 1<<30;
} else {
hwc->config_base = x86_pmu_config_addr(hwc->idx);
hwc->event_base = x86_pmu_event_addr(hwc->idx);
+ hwc->event_base_rdpmc = x86_pmu_addr_offset(hwc->idx);
}
}
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index 7241e2fc3c17..3df3de9452a9 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -14,6 +14,18 @@
#include <linux/perf_event.h>
+#if 0
+#undef wrmsrl
+#define wrmsrl(msr, val) \
+do { \
+ unsigned int _msr = (msr); \
+ u64 _val = (val); \
+ trace_printk("wrmsrl(%x, %Lx)\n", (unsigned int)(_msr), \
+ (unsigned long long)(_val)); \
+ native_write_msr((_msr), (u32)(_val), (u32)(_val >> 32)); \
+} while (0)
+#endif
+
/*
* | NHM/WSM | SNB |
* register -------------------------------
@@ -57,7 +69,7 @@ struct amd_nb {
};
/* The maximal number of PEBS events: */
-#define MAX_PEBS_EVENTS 4
+#define MAX_PEBS_EVENTS 8
/*
* A debug store configuration.
@@ -366,6 +378,7 @@ struct x86_pmu {
void (*drain_pebs)(struct pt_regs *regs);
struct event_constraint *pebs_constraints;
void (*pebs_aliases)(struct perf_event *event);
+ int max_pebs_events;
/*
* Intel LBR
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 187c294bc658..e23e71f25264 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1800,6 +1800,8 @@ __init int intel_pmu_init(void)
x86_pmu.events_maskl = ebx.full;
x86_pmu.events_mask_len = eax.split.mask_length;
+ x86_pmu.max_pebs_events = min_t(unsigned, MAX_PEBS_EVENTS, x86_pmu.num_counters);
+
/*
* Quirk: v2 perfmon does not report fixed-purpose events, so
* assume at least 3 events:
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 35e2192df9f4..026373edef7f 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -620,7 +620,7 @@ static void intel_pmu_drain_pebs_core(struct pt_regs *iregs)
* Should not happen, we program the threshold at 1 and do not
* set a reset value.
*/
- WARN_ON_ONCE(n > 1);
+ WARN_ONCE(n > 1, "bad leftover pebs %d\n", n);
at += n - 1;
__intel_pmu_pebs_event(event, iregs, at);
@@ -651,10 +651,10 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
* Should not happen, we program the threshold at 1 and do not
* set a reset value.
*/
- WARN_ON_ONCE(n > MAX_PEBS_EVENTS);
+ WARN_ONCE(n > x86_pmu.max_pebs_events, "Unexpected number of pebs records %d\n", n);
for ( ; at < top; at++) {
- for_each_set_bit(bit, (unsigned long *)&at->status, MAX_PEBS_EVENTS) {
+ for_each_set_bit(bit, (unsigned long *)&at->status, x86_pmu.max_pebs_events) {
event = cpuc->events[bit];
if (!test_bit(bit, cpuc->active_mask))
continue;
@@ -670,7 +670,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
break;
}
- if (!event || bit >= MAX_PEBS_EVENTS)
+ if (!event || bit >= x86_pmu.max_pebs_events)
continue;
__intel_pmu_pebs_event(event, iregs, at);
diff --git a/arch/x86/kernel/uprobes.c b/arch/x86/kernel/uprobes.c
index dc4e910a7d96..36fd42091fa7 100644
--- a/arch/x86/kernel/uprobes.c
+++ b/arch/x86/kernel/uprobes.c
@@ -409,9 +409,10 @@ static int validate_insn_bits(struct arch_uprobe *auprobe, struct mm_struct *mm,
* arch_uprobe_analyze_insn - instruction analysis including validity and fixups.
* @mm: the probed address space.
* @arch_uprobe: the probepoint information.
+ * @addr: virtual address at which to install the probepoint
* Return 0 on success or a -ve number on error.
*/
-int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm)
+int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long addr)
{
int ret;
struct insn insn;
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 45db49f64bb4..1ce887abcc5c 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -677,6 +677,7 @@ struct hw_perf_event {
u64 last_tag;
unsigned long config_base;
unsigned long event_base;
+ int event_base_rdpmc;
int idx;
int last_cpu;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4059c0f33f07..c7cfa6996db4 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1581,7 +1581,6 @@ struct task_struct {
#endif
#ifdef CONFIG_UPROBES
struct uprobe_task *utask;
- int uprobe_srcu_id;
#endif
};
diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 985be4d80fe8..b52376d02332 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -38,7 +38,6 @@
#define UINSNS_PER_PAGE (PAGE_SIZE/UPROBE_XOL_SLOT_BYTES)
#define MAX_UPROBE_XOL_SLOTS UINSNS_PER_PAGE
-static struct srcu_struct uprobes_srcu;
static struct rb_root uprobes_tree = RB_ROOT;
static DEFINE_SPINLOCK(uprobes_treelock); /* serialize rbtree access */
@@ -312,7 +311,7 @@ static int read_opcode(struct mm_struct *mm, unsigned long vaddr, uprobe_opcode_
void *vaddr_new;
int ret;
- ret = get_user_pages(NULL, mm, vaddr, 1, 0, 0, &page, NULL);
+ ret = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
if (ret <= 0)
return ret;
@@ -333,10 +332,20 @@ static int is_swbp_at_addr(struct mm_struct *mm, unsigned long vaddr)
uprobe_opcode_t opcode;
int result;
+ if (current->mm == mm) {
+ pagefault_disable();
+ result = __copy_from_user_inatomic(&opcode, (void __user*)vaddr,
+ sizeof(opcode));
+ pagefault_enable();
+
+ if (likely(result == 0))
+ goto out;
+ }
+
result = read_opcode(mm, vaddr, &opcode);
if (result)
return result;
-
+out:
if (is_swbp_insn(&opcode))
return 1;
@@ -697,7 +706,7 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm,
if (is_swbp_insn((uprobe_opcode_t *)uprobe->arch.insn))
return -EEXIST;
- ret = arch_uprobe_analyze_insn(&uprobe->arch, mm);
+ ret = arch_uprobe_analyze_insn(&uprobe->arch, mm, addr);
if (ret)
return ret;
@@ -728,20 +737,14 @@ remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, loff_t vaddr)
}
/*
- * There could be threads that have hit the breakpoint and are entering the
- * notifier code and trying to acquire the uprobes_treelock. The thread
- * calling delete_uprobe() that is removing the uprobe from the rb_tree can
- * race with these threads and might acquire the uprobes_treelock compared
- * to some of the breakpoint hit threads. In such a case, the breakpoint
- * hit threads will not find the uprobe. The current unregistering thread
- * waits till all other threads have hit a breakpoint, to acquire the
- * uprobes_treelock before the uprobe is removed from the rbtree.
+ * There could be threads that have already hit the breakpoint. They
+ * will recheck the current insn and restart if find_uprobe() fails.
+ * See find_active_uprobe().
*/
static void delete_uprobe(struct uprobe *uprobe)
{
unsigned long flags;
- synchronize_srcu(&uprobes_srcu);
spin_lock_irqsave(&uprobes_treelock, flags);
rb_erase(&uprobe->rb_node, &uprobes_tree);
spin_unlock_irqrestore(&uprobes_treelock, flags);
@@ -843,12 +846,12 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
}
mm = vi->mm;
- down_read(&mm->mmap_sem);
+ down_write(&mm->mmap_sem);
vma = find_vma(mm, (unsigned long)vi->vaddr);
if (!vma || !valid_vma(vma, is_register)) {
list_del(&vi->probe_list);
kfree(vi);
- up_read(&mm->mmap_sem);
+ up_write(&mm->mmap_sem);
mmput(mm);
continue;
}
@@ -857,7 +860,7 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
vaddr != vi->vaddr) {
list_del(&vi->probe_list);
kfree(vi);
- up_read(&mm->mmap_sem);
+ up_write(&mm->mmap_sem);
mmput(mm);
continue;
}
@@ -867,7 +870,7 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register)
else
remove_breakpoint(uprobe, mm, vi->vaddr);
- up_read(&mm->mmap_sem);
+ up_write(&mm->mmap_sem);
mmput(mm);
if (is_register) {
if (ret && ret == -EEXIST)
@@ -1378,9 +1381,6 @@ void uprobe_free_utask(struct task_struct *t)
{
struct uprobe_task *utask = t->utask;
- if (t->uprobe_srcu_id != -1)
- srcu_read_unlock_raw(&uprobes_srcu, t->uprobe_srcu_id);
-
if (!utask)
return;
@@ -1398,7 +1398,6 @@ void uprobe_free_utask(struct task_struct *t)
void uprobe_copy_process(struct task_struct *t)
{
t->utask = NULL;
- t->uprobe_srcu_id = -1;
}
/*
@@ -1479,41 +1478,64 @@ static bool can_skip_sstep(struct uprobe *uprobe, struct pt_regs *regs)
return false;
}
+static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp)
+{
+ struct mm_struct *mm = current->mm;
+ struct uprobe *uprobe = NULL;
+ struct vm_area_struct *vma;
+
+ down_read(&mm->mmap_sem);
+ vma = find_vma(mm, bp_vaddr);
+ if (vma && vma->vm_start <= bp_vaddr) {
+ if (valid_vma(vma, false)) {
+ struct inode *inode;
+ loff_t offset;
+
+ inode = vma->vm_file->f_mapping->host;
+ offset = bp_vaddr - vma->vm_start;
+ offset += (vma->vm_pgoff << PAGE_SHIFT);
+ uprobe = find_uprobe(inode, offset);
+ }
+
+ if (!uprobe)
+ *is_swbp = is_swbp_at_addr(mm, bp_vaddr);
+ } else {
+ *is_swbp = -EFAULT;
+ }
+ up_read(&mm->mmap_sem);
+
+ return uprobe;
+}
+
/*
* Run handler and ask thread to singlestep.
* Ensure all non-fatal signals cannot interrupt thread while it singlesteps.
*/
static void handle_swbp(struct pt_regs *regs)
{
- struct vm_area_struct *vma;
struct uprobe_task *utask;
struct uprobe *uprobe;
- struct mm_struct *mm;
unsigned long bp_vaddr;
+ int uninitialized_var(is_swbp);
- uprobe = NULL;
bp_vaddr = uprobe_get_swbp_addr(regs);
- mm = current->mm;
- down_read(&mm->mmap_sem);
- vma = find_vma(mm, bp_vaddr);
-
- if (vma && vma->vm_start <= bp_vaddr && valid_vma(vma, false)) {
- struct inode *inode;
- loff_t offset;
-
- inode = vma->vm_file->f_mapping->host;
- offset = bp_vaddr - vma->vm_start;
- offset += (vma->vm_pgoff << PAGE_SHIFT);
- uprobe = find_uprobe(inode, offset);
- }
-
- srcu_read_unlock_raw(&uprobes_srcu, current->uprobe_srcu_id);
- current->uprobe_srcu_id = -1;
- up_read(&mm->mmap_sem);
+ uprobe = find_active_uprobe(bp_vaddr, &is_swbp);
if (!uprobe) {
- /* No matching uprobe; signal SIGTRAP. */
- send_sig(SIGTRAP, current, 0);
+ if (is_swbp > 0) {
+ /* No matching uprobe; signal SIGTRAP. */
+ send_sig(SIGTRAP, current, 0);
+ } else {
+ /*
+ * Either we raced with uprobe_unregister() or we can't
+ * access this memory. The latter is only possible if
+ * another thread plays with our ->mm. In both cases
+ * we can simply restart. If this vma was unmapped we
+ * can pretend this insn was not executed yet and get
+ * the (correct) SIGSEGV after restart.
+ */
+ instruction_pointer_set(regs, bp_vaddr);
+ }
return;
}
@@ -1620,7 +1642,6 @@ int uprobe_pre_sstep_notifier(struct pt_regs *regs)
utask->state = UTASK_BP_HIT;
set_thread_flag(TIF_UPROBE);
- current->uprobe_srcu_id = srcu_read_lock_raw(&uprobes_srcu);
return 1;
}
@@ -1655,7 +1676,6 @@ static int __init init_uprobes(void)
mutex_init(&uprobes_mutex[i]);
mutex_init(&uprobes_mmap_mutex[i]);
}
- init_srcu_struct(&uprobes_srcu);
return register_die_notifier(&uprobe_exception_nb);
}
Powered by cgit, Gruvboxed by Blaster4385.