perf: Remove WARN_ON_ONCE() check in __perf_event_enable() for valid scenario
authorJiri Olsa <jolsa@redhat.com>
Tue, 9 Jul 2013 15:44:11 +0000 (17:44 +0200)
committerIngo Molnar <mingo@kernel.org>
Fri, 12 Jul 2013 09:11:01 +0000 (11:11 +0200)
The '!ctx->is_active' check has a valid scenario, so
there's no need for the warning.

The reason is that there's a time window between the
'ctx->is_active' check in the perf_event_enable() function
and the __perf_event_enable() function having:

  - IRQs on
  - ctx->lock unlocked

where the task could be killed and 'ctx' deactivated by
perf_event_exit_task(), ending up with the warning below.

So remove the WARN_ON_ONCE() check and add comments to
explain it all.

This addresses the following warning reported by Vince Weaver:

[  324.983534] ------------[ cut here ]------------
[  324.984420] WARNING: at kernel/events/core.c:1953 __perf_event_enable+0x187/0x190()
[  324.984420] Modules linked in:
[  324.984420] CPU: 19 PID: 2715 Comm: nmi_bug_snb Not tainted 3.10.0+ #246
[  324.984420] Hardware name: Supermicro X8DTN/X8DTN, BIOS 4.6.3 01/08/2010
[  324.984420]  0000000000000009 ffff88043fce3ec8 ffffffff8160ea0b ffff88043fce3f00
[  324.984420]  ffffffff81080ff0 ffff8802314fdc00 ffff880231a8f800 ffff88043fcf7860
[  324.984420]  0000000000000286 ffff880231a8f800 ffff88043fce3f10 ffffffff8108103a
[  324.984420] Call Trace:
[  324.984420]  <IRQ>  [<ffffffff8160ea0b>] dump_stack+0x19/0x1b
[  324.984420]  [<ffffffff81080ff0>] warn_slowpath_common+0x70/0xa0
[  324.984420]  [<ffffffff8108103a>] warn_slowpath_null+0x1a/0x20
[  324.984420]  [<ffffffff81134437>] __perf_event_enable+0x187/0x190
[  324.984420]  [<ffffffff81130030>] remote_function+0x40/0x50
[  324.984420]  [<ffffffff810e51de>] generic_smp_call_function_single_interrupt+0xbe/0x130
[  324.984420]  [<ffffffff81066a47>] smp_call_function_single_interrupt+0x27/0x40
[  324.984420]  [<ffffffff8161fd2f>] call_function_single_interrupt+0x6f/0x80
[  324.984420]  <EOI>  [<ffffffff816161a1>] ? _raw_spin_unlock_irqrestore+0x41/0x70
[  324.984420]  [<ffffffff8113799d>] perf_event_exit_task+0x14d/0x210
[  324.984420]  [<ffffffff810acd04>] ? switch_task_namespaces+0x24/0x60
[  324.984420]  [<ffffffff81086946>] do_exit+0x2b6/0xa40
[  324.984420]  [<ffffffff8161615c>] ? _raw_spin_unlock_irq+0x2c/0x30
[  324.984420]  [<ffffffff81087279>] do_group_exit+0x49/0xc0
[  324.984420]  [<ffffffff81096854>] get_signal_to_deliver+0x254/0x620
[  324.984420]  [<ffffffff81043057>] do_signal+0x57/0x5a0
[  324.984420]  [<ffffffff8161a164>] ? __do_page_fault+0x2a4/0x4e0
[  324.984420]  [<ffffffff8161665c>] ? retint_restore_args+0xe/0xe
[  324.984420]  [<ffffffff816166cd>] ? retint_signal+0x11/0x84
[  324.984420]  [<ffffffff81043605>] do_notify_resume+0x65/0x80
[  324.984420]  [<ffffffff81616702>] retint_signal+0x46/0x84
[  324.984420] ---[ end trace 442ec2f04db3771a ]---

Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Suggested-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1373384651-6109-2-git-send-email-jolsa@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/events/core.c

index 1d1f030e2f1ef8218d939a7167c686fffc1be2a2..ef5e7cc686e31ba07929ccec570fb0380c7b8b47 100644 (file)
@@ -1950,7 +1950,16 @@ static int __perf_event_enable(void *info)
        struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
        int err;
 
-       if (WARN_ON_ONCE(!ctx->is_active))
+       /*
+        * There's a time window between 'ctx->is_active' check
+        * in perf_event_enable function and this place having:
+        *   - IRQs on
+        *   - ctx->lock unlocked
+        *
+        * where the task could be killed and 'ctx' deactivated
+        * by perf_event_exit_task.
+        */
+       if (!ctx->is_active)
                return -EINVAL;
 
        raw_spin_lock(&ctx->lock);