linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@infradead.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org, David Ahern <dsahern@gmail.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Jiri Olsa <jolsa@redhat.com>, Mike Galbraith <efault@gmx.de>,
	Namhyung Kim <namhyung@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Stephane Eranian <eranian@google.com>,
	Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 07/21] perf tools: Sample after exit loses thread correlation
Date: Wed, 28 Aug 2013 11:59:23 -0300	[thread overview]
Message-ID: <1377701977-2085-8-git-send-email-acme@infradead.org> (raw)
In-Reply-To: <1377701977-2085-1-git-send-email-acme@infradead.org>

From: David Ahern <dsahern@gmail.com>

Occassionally events (e.g., context-switch, sched tracepoints) are losing
the conversion of sample data associated with a thread. For example:

$ perf record -e sched:sched_switch -c 1 -a -- sleep 5
$ perf script
<selected events shown>
    ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
    ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
:30482 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ...

The last line lost the conversion from tid to comm. If you look at the events
(perf script -D) you see why - a SAMPLE event is generated after the EXIT:

0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482)
0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0
... thread: :30482:30482

When perf processes the EXIT event the thread is moved to the dead_threads
list. When the SAMPLE event is processed no thread exists for the pid so a new
one is created by machine__findnew_thread.

This patch address the problem by delaying the move to the dead_threads list
until the tid is re-used (per Adrian's suggestion).

With this patch we get the previous example shows:

  ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
  ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
  ls 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ...

and

  0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482)
  0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0
  ... thread: ls:30482

v4: per Arnaldo's request add dead flag to thread struct and set when task exits

v3: re-do from a time based check to a delayed move to dead_threads list

v2: Rebased to latest perf/core branch. Changed time comparison to use
    a macro which explicitly shows the time basis

Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1376491767-84171-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/machine.c | 34 ++++++++++++++++++++--------------
 tools/perf/util/thread.h  |  5 +++++
 2 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 4514e7e..574feb7 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1031,11 +1031,27 @@ out_problem:
 	return 0;
 }
 
+static void machine__remove_thread(struct machine *machine, struct thread *th)
+{
+	machine->last_match = NULL;
+	rb_erase(&th->rb_node, &machine->threads);
+	/*
+	 * We may have references to this thread, for instance in some hist_entry
+	 * instances, so just move them to a separate list.
+	 */
+	list_add_tail(&th->node, &machine->dead_threads);
+}
+
 int machine__process_fork_event(struct machine *machine, union perf_event *event)
 {
-	struct thread *thread = machine__findnew_thread(machine, event->fork.tid);
+	struct thread *thread = machine__find_thread(machine, event->fork.tid);
 	struct thread *parent = machine__findnew_thread(machine, event->fork.ptid);
 
+	/* if a thread currently exists for the thread id remove it */
+	if (thread != NULL)
+		machine__remove_thread(machine, thread);
+
+	thread = machine__findnew_thread(machine, event->fork.tid);
 	if (dump_trace)
 		perf_event__fprintf_task(event, stdout);
 
@@ -1048,18 +1064,8 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event
 	return 0;
 }
 
-static void machine__remove_thread(struct machine *machine, struct thread *th)
-{
-	machine->last_match = NULL;
-	rb_erase(&th->rb_node, &machine->threads);
-	/*
-	 * We may have references to this thread, for instance in some hist_entry
-	 * instances, so just move them to a separate list.
-	 */
-	list_add_tail(&th->node, &machine->dead_threads);
-}
-
-int machine__process_exit_event(struct machine *machine, union perf_event *event)
+int machine__process_exit_event(struct machine *machine __maybe_unused,
+				union perf_event *event)
 {
 	struct thread *thread = machine__find_thread(machine, event->fork.tid);
 
@@ -1067,7 +1073,7 @@ int machine__process_exit_event(struct machine *machine, union perf_event *event
 		perf_event__fprintf_task(event, stdout);
 
 	if (thread != NULL)
-		machine__remove_thread(machine, thread);
+		thread__exited(thread);
 
 	return 0;
 }
diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h
index 13c62c9..32d0601 100644
--- a/tools/perf/util/thread.h
+++ b/tools/perf/util/thread.h
@@ -16,6 +16,7 @@ struct thread {
 	pid_t			ppid;
 	char			shortname[3];
 	bool			comm_set;
+	bool			dead; /* if set thread has exited */
 	char			*comm;
 	int			comm_len;
 
@@ -26,6 +27,10 @@ struct machine;
 
 struct thread *thread__new(pid_t tid);
 void thread__delete(struct thread *self);
+static inline void thread__exited(struct thread *thread)
+{
+	thread->dead = true;
+}
 
 int thread__set_comm(struct thread *self, const char *comm);
 int thread__comm_len(struct thread *self);
-- 
1.8.1.4


  parent reply	other threads:[~2013-08-28 15:01 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-28 14:59 [GIT PULL 00/21] perf/core improvements and fixes Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 01/21] perf kvm: Remove force option to cmd_record Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 02/21] perf tools: Try to increase the file descriptor limits on EMFILE Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 03/21] perf tools: Re-implement debug print function for linking python/perf.so Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 04/21] perf tools: Add debug prints Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 05/21] perf trace: Implement -o/--output filename Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 06/21] perf trace: Make command line arguments consistent with perf-record Arnaldo Carvalho de Melo
2013-08-28 14:59 ` Arnaldo Carvalho de Melo [this message]
2013-08-28 14:59 ` [PATCH 08/21] perf tools: Don't install scripting files files when disabled Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 09/21] perf trace: Support ! in -e expressions Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 10/21] perf trace: Add --verbose option Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 11/21] perf trace: Hide sys_exit messages about syscall id = -1 Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 12/21] perf trace: Introduce syscall arg formatters Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 13/21] perf trace: Simplify sys_exit return printing Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 14/21] perf trace: Allow printing syscall return values in hex Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 15/21] perf trace: Add aliases to remaining syscalls of the sys_enter_newfoo Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 16/21] perf trace: Allow overiding the formatting of syscall fields Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 17/21] perf trace: Add beautifier for mmap prot parm Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 18/21] perf trace: Add beautifier for mmap flags parm Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 19/21] perf trace: Add beautifier for madvise behaviour/advice parm Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 20/21] perf tools: Add pid to struct thread Arnaldo Carvalho de Melo
2013-08-28 14:59 ` [PATCH 21/21] tools lib traceevent: Fixup jobserver setup Arnaldo Carvalho de Melo
2013-08-29 10:04 ` [GIT PULL 00/21] perf/core improvements and fixes Ingo Molnar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1377701977-2085-8-git-send-email-acme@infradead.org \
    --to=acme@infradead.org \
    --cc=acme@redhat.com \
    --cc=adrian.hunter@intel.com \
    --cc=dsahern@gmail.com \
    --cc=efault@gmx.de \
    --cc=eranian@google.com \
    --cc=fweisbec@gmail.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).