linux/kernel/trace/trace_event_perf.c

516 lines
12 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* trace event based perf event profiling/tracing
*
* Copyright (C) 2009 Red Hat Inc, Peter Zijlstra
perf: Take a hot regs snapshot for trace events We are taking a wrong regs snapshot when a trace event triggers. Either we use get_irq_regs(), which gives us the interrupted registers if we are in an interrupt, or we use task_pt_regs() which gives us the state before we entered the kernel, assuming we are lucky enough to be no kernel thread, in which case task_pt_regs() returns the initial set of regs when the kernel thread was started. What we want is different. We need a hot snapshot of the regs, so that we can get the instruction pointer to record in the sample, the frame pointer for the callchain, and some other things. Let's use the new perf_fetch_caller_regs() for that. Comparison with perf record -e lock: -R -a -f -g Before: perf [kernel] [k] __do_softirq | --- __do_softirq | |--55.16%-- __open | --44.84%-- __write_nocancel After: perf [kernel] [k] perf_tp_event | --- perf_tp_event | |--41.07%-- lock_acquire | | | |--39.36%-- _raw_spin_lock | | | | | |--7.81%-- hrtimer_interrupt | | | smp_apic_timer_interrupt | | | apic_timer_interrupt The old case was producing unreliable callchains. Now having right frame and instruction pointers, we have the trace we want. Also syscalls and kprobe events already have the right regs, let's use them instead of wasting a retrieval. v2: Follow the rename perf_save_regs() -> perf_fetch_caller_regs() Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Jason Baron <jbaron@redhat.com> Cc: Archs <linux-arch@vger.kernel.org>
2010-03-03 14:16:16 +08:00
* Copyright (C) 2009-2010 Frederic Weisbecker <fweisbec@gmail.com>
*/
#include <linux/module.h>
#include <linux/kprobes.h>
#include "trace.h"
#include "trace_probe.h"
static char __percpu *perf_trace_buf[PERF_NR_CONTEXTS];
/*
* Force it to be aligned to unsigned long to avoid misaligned accesses
* suprises
*/
typedef typeof(unsigned long [PERF_MAX_TRACE_SIZE / sizeof(unsigned long)])
perf_trace_t;
/* Count the events in use (per event id, not per instance) */
static int total_ref_count;
static int perf_trace_event_perm(struct trace_event_call *tp_event,
struct perf_event *p_event)
{
if (tp_event->perf_perm) {
int ret = tp_event->perf_perm(tp_event, p_event);
if (ret)
return ret;
}
/*
* We checked and allowed to create parent,
* allow children without checking.
*/
if (p_event->parent)
return 0;
/*
* It's ok to check current process (owner) permissions in here,
* because code below is called only via perf_event_open syscall.
*/
/* The ftrace function trace is allowed only for root. */
if (ftrace_event_is_function(tp_event)) {
if (perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN))
return -EPERM;
if (!is_sampling_event(p_event))
return 0;
/*
* We don't allow user space callchains for function trace
* event, due to issues with page faults while tracing page
* fault handler and its overall trickiness nature.
*/
if (!p_event->attr.exclude_callchain_user)
return -EINVAL;
/*
* Same reason to disable user stack dump as for user space
* callchains above.
*/
if (p_event->attr.sample_type & PERF_SAMPLE_STACK_USER)
return -EINVAL;
}
/* No tracing, just counting, so no obvious leak */
if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW))
return 0;
/* Some events are ok to be traced by non-root users... */
if (p_event->attach_state == PERF_ATTACH_TASK) {
if (tp_event->flags & TRACE_EVENT_FL_CAP_ANY)
return 0;
}
/*
* ...otherwise raw tracepoint data can be a severe data leak,
* only allow root to have these.
*/
if (perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN))
return -EPERM;
return 0;
}
static int perf_trace_event_reg(struct trace_event_call *tp_event,
struct perf_event *p_event)
{
struct hlist_head __percpu *list;
int ret = -ENOMEM;
int cpu;
p_event->tp_event = tp_event;
if (tp_event->perf_refcount++ > 0)
return 0;
list = alloc_percpu(struct hlist_head);
if (!list)
goto fail;
for_each_possible_cpu(cpu)
INIT_HLIST_HEAD(per_cpu_ptr(list, cpu));
tp_event->perf_events = list;
if (!total_ref_count) {
char __percpu *buf;
int i;
for (i = 0; i < PERF_NR_CONTEXTS; i++) {
buf = (char __percpu *)alloc_percpu(perf_trace_t);
if (!buf)
goto fail;
perf_trace_buf[i] = buf;
}
}
ret = tp_event->class->reg(tp_event, TRACE_REG_PERF_REGISTER, NULL);
if (ret)
goto fail;
total_ref_count++;
return 0;
fail:
if (!total_ref_count) {
int i;
for (i = 0; i < PERF_NR_CONTEXTS; i++) {
free_percpu(perf_trace_buf[i]);
perf_trace_buf[i] = NULL;
}
}
if (!--tp_event->perf_refcount) {
free_percpu(tp_event->perf_events);
tp_event->perf_events = NULL;
}
return ret;
}
static void perf_trace_event_unreg(struct perf_event *p_event)
{
struct trace_event_call *tp_event = p_event->tp_event;
int i;
if (--tp_event->perf_refcount > 0)
goto out;
tp_event->class->reg(tp_event, TRACE_REG_PERF_UNREGISTER, NULL);
/*
* Ensure our callback won't be called anymore. The buffers
* will be freed after that.
*/
tracepoint_synchronize_unregister();
free_percpu(tp_event->perf_events);
tp_event->perf_events = NULL;
if (!--total_ref_count) {
for (i = 0; i < PERF_NR_CONTEXTS; i++) {
free_percpu(perf_trace_buf[i]);
perf_trace_buf[i] = NULL;
}
}
out:
module_put(tp_event->mod);
}
static int perf_trace_event_open(struct perf_event *p_event)
{
struct trace_event_call *tp_event = p_event->tp_event;
return tp_event->class->reg(tp_event, TRACE_REG_PERF_OPEN, p_event);
}
static void perf_trace_event_close(struct perf_event *p_event)
{
struct trace_event_call *tp_event = p_event->tp_event;
tp_event->class->reg(tp_event, TRACE_REG_PERF_CLOSE, p_event);
}
static int perf_trace_event_init(struct trace_event_call *tp_event,
struct perf_event *p_event)
{
int ret;
ret = perf_trace_event_perm(tp_event, p_event);
if (ret)
return ret;
ret = perf_trace_event_reg(tp_event, p_event);
if (ret)
return ret;
ret = perf_trace_event_open(p_event);
if (ret) {
perf_trace_event_unreg(p_event);
return ret;
}
return 0;
}
int perf_trace_init(struct perf_event *p_event)
{
struct trace_event_call *tp_event;
u64 event_id = p_event->attr.config;
int ret = -EINVAL;
mutex_lock(&event_mutex);
list_for_each_entry(tp_event, &ftrace_events, list) {
if (tp_event->event.type == event_id &&
tp_event->class && tp_event->class->reg &&
try_module_get(tp_event->mod)) {
ret = perf_trace_event_init(tp_event, p_event);
if (ret)
module_put(tp_event->mod);
break;
}
}
mutex_unlock(&event_mutex);
return ret;
}
void perf_trace_destroy(struct perf_event *p_event)
{
mutex_lock(&event_mutex);
perf_trace_event_close(p_event);
perf_trace_event_unreg(p_event);
mutex_unlock(&event_mutex);
}
#ifdef CONFIG_KPROBE_EVENTS
int perf_kprobe_init(struct perf_event *p_event, bool is_retprobe)
{
int ret;
char *func = NULL;
struct trace_event_call *tp_event;
if (p_event->attr.kprobe_func) {
func = kzalloc(KSYM_NAME_LEN, GFP_KERNEL);
if (!func)
return -ENOMEM;
ret = strncpy_from_user(
func, u64_to_user_ptr(p_event->attr.kprobe_func),
KSYM_NAME_LEN);
perf/core: Fix perf_kprobe_init() Fix error handling in perf_kprobe_init(): ================================================================== BUG: KASAN: slab-out-of-bounds in strlen+0x8e/0xa0 lib/string.c:482 Read of size 1 at addr ffff88003f9cc5c0 by task syz-executor2/23095 CPU: 0 PID: 23095 Comm: syz-executor2 Not tainted 4.16.0+ #24 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0xca/0x13e lib/dump_stack.c:113 print_address_description+0x6e/0x2c0 mm/kasan/report.c:256 kasan_report_error mm/kasan/report.c:354 [inline] kasan_report+0x256/0x380 mm/kasan/report.c:412 strlen+0x8e/0xa0 lib/string.c:482 kstrdup+0x21/0x70 mm/util.c:55 alloc_trace_kprobe+0xc8/0x930 kernel/trace/trace_kprobe.c:325 create_local_trace_kprobe+0x4f/0x3a0 kernel/trace/trace_kprobe.c:1438 perf_kprobe_init+0x149/0x1f0 kernel/trace/trace_event_perf.c:264 perf_kprobe_event_init+0xa8/0x120 kernel/events/core.c:8407 perf_try_init_event+0xcb/0x2a0 kernel/events/core.c:9719 perf_init_event kernel/events/core.c:9750 [inline] perf_event_alloc+0x1367/0x1e20 kernel/events/core.c:10022 SYSC_perf_event_open+0x242/0x2330 kernel/events/core.c:10477 do_syscall_64+0x198/0x640 arch/x86/entry/common.c:287 entry_SYSCALL_64_after_hwframe+0x42/0xb7 Reported-by: 范龙飞 <long7573@126.com> Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <songliubraving@fb.com> Cc: Thomas Gleixner <tglx@linutronix.de> Fixes: e12f03d7031a ("perf/core: Implement the 'perf_kprobe' PMU") Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-04-09 20:16:54 +08:00
if (ret == KSYM_NAME_LEN)
ret = -E2BIG;
if (ret < 0)
goto out;
if (func[0] == '\0') {
kfree(func);
func = NULL;
}
}
tp_event = create_local_trace_kprobe(
func, (void *)(unsigned long)(p_event->attr.kprobe_addr),
p_event->attr.probe_offset, is_retprobe);
if (IS_ERR(tp_event)) {
ret = PTR_ERR(tp_event);
goto out;
}
ret = perf_trace_event_init(tp_event, p_event);
if (ret)
destroy_local_trace_kprobe(tp_event);
out:
kfree(func);
return ret;
}
void perf_kprobe_destroy(struct perf_event *p_event)
{
perf_trace_event_close(p_event);
perf_trace_event_unreg(p_event);
destroy_local_trace_kprobe(p_event->tp_event);
}
#endif /* CONFIG_KPROBE_EVENTS */
#ifdef CONFIG_UPROBE_EVENTS
int perf_uprobe_init(struct perf_event *p_event,
unsigned long ref_ctr_offset, bool is_retprobe)
{
int ret;
char *path = NULL;
struct trace_event_call *tp_event;
if (!p_event->attr.uprobe_path)
return -EINVAL;
path = kzalloc(PATH_MAX, GFP_KERNEL);
if (!path)
return -ENOMEM;
ret = strncpy_from_user(
path, u64_to_user_ptr(p_event->attr.uprobe_path), PATH_MAX);
if (ret == PATH_MAX)
return -E2BIG;
if (ret < 0)
goto out;
if (path[0] == '\0') {
ret = -EINVAL;
goto out;
}
tp_event = create_local_trace_uprobe(path, p_event->attr.probe_offset,
ref_ctr_offset, is_retprobe);
if (IS_ERR(tp_event)) {
ret = PTR_ERR(tp_event);
goto out;
}
/*
* local trace_uprobe need to hold event_mutex to call
* uprobe_buffer_enable() and uprobe_buffer_disable().
* event_mutex is not required for local trace_kprobes.
*/
mutex_lock(&event_mutex);
ret = perf_trace_event_init(tp_event, p_event);
if (ret)
destroy_local_trace_uprobe(tp_event);
mutex_unlock(&event_mutex);
out:
kfree(path);
return ret;
}
void perf_uprobe_destroy(struct perf_event *p_event)
{
mutex_lock(&event_mutex);
perf_trace_event_close(p_event);
perf_trace_event_unreg(p_event);
mutex_unlock(&event_mutex);
destroy_local_trace_uprobe(p_event->tp_event);
}
#endif /* CONFIG_UPROBE_EVENTS */
2010-06-16 20:37:10 +08:00
int perf_trace_add(struct perf_event *p_event, int flags)
{
struct trace_event_call *tp_event = p_event->tp_event;
2010-06-16 20:37:10 +08:00
if (!(flags & PERF_EF_START))
p_event->hw.state = PERF_HES_STOPPED;
/*
* If TRACE_REG_PERF_ADD returns false; no custom action was performed
* and we need to take the default action of enqueueing our event on
* the right per-cpu hlist.
*/
if (!tp_event->class->reg(tp_event, TRACE_REG_PERF_ADD, p_event)) {
struct hlist_head __percpu *pcpu_list;
struct hlist_head *list;
pcpu_list = tp_event->perf_events;
if (WARN_ON_ONCE(!pcpu_list))
return -EINVAL;
list = this_cpu_ptr(pcpu_list);
hlist_add_head_rcu(&p_event->hlist_entry, list);
}
return 0;
}
2010-06-16 20:37:10 +08:00
void perf_trace_del(struct perf_event *p_event, int flags)
{
struct trace_event_call *tp_event = p_event->tp_event;
/*
* If TRACE_REG_PERF_DEL returns false; no custom action was performed
* and we need to take the default action of dequeueing our event from
* the right per-cpu hlist.
*/
if (!tp_event->class->reg(tp_event, TRACE_REG_PERF_DEL, p_event))
hlist_del_rcu(&p_event->hlist_entry);
}
void *perf_trace_buf_alloc(int size, struct pt_regs **regs, int *rctxp)
{
char *raw_data;
int rctx;
BUILD_BUG_ON(PERF_MAX_TRACE_SIZE % sizeof(unsigned long));
if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
"perf buffer not large enough"))
return NULL;
*rctxp = rctx = perf_swevent_get_recursion_context();
if (rctx < 0)
return NULL;
perf: Avoid horrible stack usage Both Linus (most recent) and Steve (a while ago) reported that perf related callbacks have massive stack bloat. The problem is that software events need a pt_regs in order to properly report the event location and unwind stack. And because we could not assume one was present we allocated one on stack and filled it with minimal bits required for operation. Now, pt_regs is quite large, so this is undesirable. Furthermore it turns out that most sites actually have a pt_regs pointer available, making this even more onerous, as the stack space is pointless waste. This patch addresses the problem by observing that software events have well defined nesting semantics, therefore we can use static per-cpu storage instead of on-stack. Linus made the further observation that all but the scheduler callers of perf_sw_event() have a pt_regs available, so we change the regular perf_sw_event() to require a valid pt_regs (where it used to be optional) and add perf_sw_event_sched() for the scheduler. We have a scheduler specific call instead of a more generic _noregs() like construct because we can assume non-recursion from the scheduler and thereby simplify the code further (_noregs would have to put the recursion context call inline in order to assertain which __perf_regs element to use). One last note on the implementation of perf_trace_buf_prepare(); we allow .regs = NULL for those cases where we already have a pt_regs pointer available and do not need another. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Reported-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Javi Merino <javi.merino@arm.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Petr Mladek <pmladek@suse.cz> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: Vaibhav Nagarnaik <vnagarnaik@google.com> Link: http://lkml.kernel.org/r/20141216115041.GW3337@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar <mingo@kernel.org>
2014-12-16 19:47:34 +08:00
if (regs)
*regs = this_cpu_ptr(&__perf_regs[rctx]);
raw_data = this_cpu_ptr(perf_trace_buf[rctx]);
/* zero the dead bytes from align to not leak stack to user */
memset(&raw_data[size - sizeof(u64)], 0, sizeof(u64));
return raw_data;
}
EXPORT_SYMBOL_GPL(perf_trace_buf_alloc);
NOKPROBE_SYMBOL(perf_trace_buf_alloc);
void perf_trace_buf_update(void *record, u16 type)
{
struct trace_entry *entry = record;
int pc = preempt_count();
unsigned long flags;
local_save_flags(flags);
tracing_generic_entry_update(entry, flags, pc);
entry->type = type;
}
NOKPROBE_SYMBOL(perf_trace_buf_update);
#ifdef CONFIG_FUNCTION_TRACER
static void
perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip,
struct ftrace_ops *ops, struct pt_regs *pt_regs)
{
struct ftrace_entry *entry;
struct perf_event *event;
struct hlist_head head;
struct pt_regs regs;
int rctx;
if ((unsigned long)ops->private != smp_processor_id())
return;
event = container_of(ops, struct perf_event, ftrace_ops);
/*
* @event->hlist entry is NULL (per INIT_HLIST_NODE), and all
* the perf code does is hlist_for_each_entry_rcu(), so we can
* get away with simply setting the @head.first pointer in order
* to create a singular list.
*/
head.first = &event->hlist_entry;
#define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \
sizeof(u64)) - sizeof(u32))
BUILD_BUG_ON(ENTRY_SIZE > PERF_MAX_TRACE_SIZE);
memset(&regs, 0, sizeof(regs));
perf_fetch_caller_regs(&regs);
entry = perf_trace_buf_alloc(ENTRY_SIZE, NULL, &rctx);
if (!entry)
return;
entry->ip = ip;
entry->parent_ip = parent_ip;
perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, TRACE_FN,
1, &regs, &head, NULL);
#undef ENTRY_SIZE
}
static int perf_ftrace_function_register(struct perf_event *event)
{
struct ftrace_ops *ops = &event->ftrace_ops;
ops->flags = FTRACE_OPS_FL_RCU;
ops->func = perf_ftrace_function_call;
ops->private = (void *)(unsigned long)nr_cpu_ids;
return register_ftrace_function(ops);
}
static int perf_ftrace_function_unregister(struct perf_event *event)
{
struct ftrace_ops *ops = &event->ftrace_ops;
ftrace, perf: Add filter support for function trace event Adding support to filter function trace event via perf interface. It is now possible to use filter interface in the perf tool like: perf record -e ftrace:function --filter="(ip == mm_*)" ls The filter syntax is restricted to the the 'ip' field only, and following operators are accepted '==' '!=' '||', ending up with the filter strings like: ip == f1[, ]f2 ... || ip != f3[, ]f4 ... with comma ',' or space ' ' as a function separator. If the space ' ' is used as a separator, the right side of the assignment needs to be enclosed in double quotes '"', e.g.: perf record -e ftrace:function --filter '(ip == do_execve,sys_*,ext*)' ls perf record -e ftrace:function --filter '(ip == "do_execve,sys_*,ext*")' ls perf record -e ftrace:function --filter '(ip == "do_execve sys_* ext*")' ls The '==' operator adds trace filter with same effect as would be added via set_ftrace_filter file. The '!=' operator adds trace filter with same effect as would be added via set_ftrace_notrace file. The right side of the '!=', '==' operators is list of functions or regexp. to be added to filter separated by space. The '||' operator is used for connecting multiple filter definitions together. It is possible to have more than one '==' and '!=' operators within one filter string. Link: http://lkml.kernel.org/r/1329317514-8131-8-git-send-email-jolsa@redhat.com Signed-off-by: Jiri Olsa <jolsa@redhat.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2012-02-15 22:51:54 +08:00
int ret = unregister_ftrace_function(ops);
ftrace_free_filter(ops);
return ret;
}
int perf_ftrace_event_register(struct trace_event_call *call,
enum trace_reg type, void *data)
{
struct perf_event *event = data;
switch (type) {
case TRACE_REG_REGISTER:
case TRACE_REG_UNREGISTER:
break;
case TRACE_REG_PERF_REGISTER:
case TRACE_REG_PERF_UNREGISTER:
return 0;
case TRACE_REG_PERF_OPEN:
return perf_ftrace_function_register(data);
case TRACE_REG_PERF_CLOSE:
return perf_ftrace_function_unregister(data);
case TRACE_REG_PERF_ADD:
event->ftrace_ops.private = (void *)(unsigned long)smp_processor_id();
return 1;
case TRACE_REG_PERF_DEL:
event->ftrace_ops.private = (void *)(unsigned long)nr_cpu_ids;
return 1;
}
return -EINVAL;
}
#endif /* CONFIG_FUNCTION_TRACER */