ftrace, perf: Avoid infinite event generation loop

Vince's perf-trinity fuzzer found yet another 'interesting' problem.

When we sample the irq_work_exit tracepoint with period==1 (or
PERF_SAMPLE_PERIOD) and we add an fasync SIGNAL handler we create an
infinite event generation loop:

  ,-> <IPI>
  |     irq_work_exit() ->
  |       trace_irq_work_exit() ->
  |         ...
  |           __perf_event_overflow() -> (due to fasync)
  |             irq_work_queue() -> (irq_work_list must be empty)
  '---------      arch_irq_work_raise()

Similar things can happen due to regular poll() wakeups if we exceed
the ring-buffer wakeup watermark, or have an event_limit.

To avoid this, dis-allow sampling this particular tracepoint.

In order to achieve this, create a special perf_perm function pointer
for each event and call this (when set) on trying to create a
tracepoint perf event.

[ roasted: use expr... to allow for ',' in your expression ]

Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Tested-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Dave Jones <davej@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/20131114152304.GC5364@laptop.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This commit is contained in:
Peter Zijlstra 2013-11-14 16:23:04 +01:00 committed by Ingo Molnar
parent 801a76050b
commit d5b5f391d4
5 changed files with 44 additions and 0 deletions

View File

@ -71,6 +71,17 @@ DEFINE_IRQ_VECTOR_EVENT(x86_platform_ipi);
*/
DEFINE_IRQ_VECTOR_EVENT(irq_work);
/*
* We must dis-allow sampling irq_work_exit() because perf event sampling
* itself can cause irq_work, which would lead to an infinite loop;
*
* 1) irq_work_exit happens
* 2) generates perf sample
* 3) generates irq_work
* 4) goto 1
*/
TRACE_EVENT_PERF_PERM(irq_work_exit, is_sampling_event(p_event) ? -EPERM : 0);
/*
* call_function - called when entering/exiting a call function interrupt
* vector handler

View File

@ -248,6 +248,9 @@ struct ftrace_event_call {
#ifdef CONFIG_PERF_EVENTS
int perf_refcount;
struct hlist_head __percpu *perf_events;
int (*perf_perm)(struct ftrace_event_call *,
struct perf_event *);
#endif
};
@ -317,6 +320,19 @@ struct ftrace_event_file {
} \
early_initcall(trace_init_flags_##name);
#define __TRACE_EVENT_PERF_PERM(name, expr...) \
static int perf_perm_##name(struct ftrace_event_call *tp_event, \
struct perf_event *p_event) \
{ \
return ({ expr; }); \
} \
static int __init trace_init_perf_perm_##name(void) \
{ \
event_##name.perf_perm = &perf_perm_##name; \
return 0; \
} \
early_initcall(trace_init_perf_perm_##name);
#define PERF_MAX_TRACE_SIZE 2048
#define MAX_FILTER_STR_VAL 256 /* Should handle KSYM_SYMBOL_LEN */

View File

@ -267,6 +267,8 @@ static inline void tracepoint_synchronize_unregister(void)
#define TRACE_EVENT_FLAGS(event, flag)
#define TRACE_EVENT_PERF_PERM(event, expr...)
#endif /* DECLARE_TRACE */
#ifndef TRACE_EVENT
@ -399,4 +401,6 @@ static inline void tracepoint_synchronize_unregister(void)
#define TRACE_EVENT_FLAGS(event, flag)
#define TRACE_EVENT_PERF_PERM(event, expr...)
#endif /* ifdef TRACE_EVENT (see note above) */

View File

@ -90,6 +90,10 @@
#define TRACE_EVENT_FLAGS(name, value) \
__TRACE_EVENT_FLAGS(name, value)
#undef TRACE_EVENT_PERF_PERM
#define TRACE_EVENT_PERF_PERM(name, expr...) \
__TRACE_EVENT_PERF_PERM(name, expr)
#include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
@ -140,6 +144,9 @@
#undef TRACE_EVENT_FLAGS
#define TRACE_EVENT_FLAGS(event, flag)
#undef TRACE_EVENT_PERF_PERM
#define TRACE_EVENT_PERF_PERM(event, expr...)
#include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
/*

View File

@ -24,6 +24,12 @@ static int total_ref_count;
static int perf_trace_event_perm(struct ftrace_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;
}
/* The ftrace function trace is allowed only for root. */
if (ftrace_event_is_function(tp_event) &&
perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN))