[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <tip-b85da9c1e52eb570fc16a42ae241ee5039c9151b@git.kernel.org>
Date: Fri, 4 Dec 2015 03:51:52 -0800
From: tip-bot for Jiri Olsa <tipbot@...or.com>
To: linux-tip-commits@...r.kernel.org
Cc: tglx@...utronix.de, dsahern@...il.com, vincent.weaver@...ne.edu,
acme@...hat.com, eranian@...gle.com, mingo@...nel.org,
linux-kernel@...r.kernel.org, hpa@...or.com, acme@...nel.org,
torvalds@...ux-foundation.org, peterz@...radead.org,
namhyung@...nel.org, jolsa@...hat.com, jolsa@...nel.org
Subject: [tip:perf/core] perf: Do not send exit event twice
Commit-ID: b85da9c1e52eb570fc16a42ae241ee5039c9151b
Gitweb: http://git.kernel.org/tip/b85da9c1e52eb570fc16a42ae241ee5039c9151b
Author: Jiri Olsa <jolsa@...nel.org>
AuthorDate: Wed, 4 Nov 2015 16:00:05 +0100
Committer: Ingo Molnar <mingo@...nel.org>
CommitDate: Fri, 4 Dec 2015 10:08:07 +0100
perf: Do not send exit event twice
In case we monitor events system wide, we get EXIT event
(when configured) twice for each task that exited.
Note doubled lines with same pid/tid in following example:
$ sudo ./perf record -a
^C[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.480 MB perf.data (2518 samples) ]
$ sudo ./perf report -D | grep EXIT
0 60290687567581 0x59910 [0x38]: PERF_RECORD_EXIT(1250:1250):(1250:1250)
0 60290687568354 0x59948 [0x38]: PERF_RECORD_EXIT(1250:1250):(1250:1250)
0 60290687988744 0x59ad8 [0x38]: PERF_RECORD_EXIT(1250:1250):(1250:1250)
0 60290687989198 0x59b10 [0x38]: PERF_RECORD_EXIT(1250:1250):(1250:1250)
1 60290692567895 0x62af0 [0x38]: PERF_RECORD_EXIT(1253:1253):(1253:1253)
1 60290692568322 0x62b28 [0x38]: PERF_RECORD_EXIT(1253:1253):(1253:1253)
2 60290692739276 0x69a18 [0x38]: PERF_RECORD_EXIT(1252:1252):(1252:1252)
2 60290692739910 0x69a50 [0x38]: PERF_RECORD_EXIT(1252:1252):(1252:1252)
The reason is that the cpu contexts are processes each time
we call perf_event_task. I'm changing the perf_event_aux logic
to serve task_ctx and cpu contexts separately, which ensure we
don't get EXIT event generated twice on same cpu context.
This does not affect other auxiliary events, as they don't
use task_ctx at all.
Signed-off-by: Jiri Olsa <jolsa@...nel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Cc: Arnaldo Carvalho de Melo <acme@...nel.org>
Cc: Arnaldo Carvalho de Melo <acme@...hat.com>
Cc: David Ahern <dsahern@...il.com>
Cc: Jiri Olsa <jolsa@...hat.com>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Namhyung Kim <namhyung@...nel.org>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Stephane Eranian <eranian@...gle.com>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Vince Weaver <vincent.weaver@...ne.edu>
Link: http://lkml.kernel.org/r/1446649205-5822-1-git-send-email-jolsa@kernel.org
Signed-off-by: Ingo Molnar <mingo@...nel.org>
---
kernel/events/core.c | 42 +++++++++++++++++++++++++++++++-----------
1 file changed, 31 insertions(+), 11 deletions(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 49a5118..39cf4a4 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5683,6 +5683,17 @@ perf_event_aux_ctx(struct perf_event_context *ctx,
}
static void
+perf_event_aux_task_ctx(perf_event_aux_output_cb output, void *data,
+ struct perf_event_context *task_ctx)
+{
+ rcu_read_lock();
+ preempt_disable();
+ perf_event_aux_ctx(task_ctx, output, data);
+ preempt_enable();
+ rcu_read_unlock();
+}
+
+static void
perf_event_aux(perf_event_aux_output_cb output, void *data,
struct perf_event_context *task_ctx)
{
@@ -5691,14 +5702,23 @@ perf_event_aux(perf_event_aux_output_cb output, void *data,
struct pmu *pmu;
int ctxn;
+ /*
+ * If we have task_ctx != NULL we only notify
+ * the task context itself. The task_ctx is set
+ * only for EXIT events before releasing task
+ * context.
+ */
+ if (task_ctx) {
+ perf_event_aux_task_ctx(output, data, task_ctx);
+ return;
+ }
+
rcu_read_lock();
list_for_each_entry_rcu(pmu, &pmus, entry) {
cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
if (cpuctx->unique_pmu != pmu)
goto next;
perf_event_aux_ctx(&cpuctx->ctx, output, data);
- if (task_ctx)
- goto next;
ctxn = pmu->task_ctx_nr;
if (ctxn < 0)
goto next;
@@ -5708,12 +5728,6 @@ perf_event_aux(perf_event_aux_output_cb output, void *data,
next:
put_cpu_ptr(pmu->pmu_cpu_context);
}
-
- if (task_ctx) {
- preempt_disable();
- perf_event_aux_ctx(task_ctx, output, data);
- preempt_enable();
- }
rcu_read_unlock();
}
@@ -8803,10 +8817,8 @@ static void perf_event_exit_task_context(struct task_struct *child, int ctxn)
struct perf_event_context *child_ctx, *clone_ctx = NULL;
unsigned long flags;
- if (likely(!child->perf_event_ctxp[ctxn])) {
- perf_event_task(child, NULL, 0);
+ if (likely(!child->perf_event_ctxp[ctxn]))
return;
- }
local_irq_save(flags);
/*
@@ -8890,6 +8902,14 @@ void perf_event_exit_task(struct task_struct *child)
for_each_task_context_nr(ctxn)
perf_event_exit_task_context(child, ctxn);
+
+ /*
+ * The perf_event_exit_task_context calls perf_event_task
+ * with child's task_ctx, which generates EXIT events for
+ * child contexts and sets child->perf_event_ctxp[] to NULL.
+ * At this point we need to send EXIT events to cpu contexts.
+ */
+ perf_event_task(child, NULL, 0);
}
static void perf_free_event(struct perf_event *event,
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists