From: Andrew Vagin <avagin@openvz.org>
To: unlisted-recipients:; (no To-header on input)
Cc: Arun Sharma <asharma@fb.com>, Oleg Strikov <OSTRIKOV@nvidia.com>,
Steven Rostedt <rostedt@goodmis.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
linux-kernel@vger.kernel.org
Subject: [PATCH 3/5] sched: save a previous state on task_struct
Date: Mon, 4 Jun 2012 12:09:40 +0400 [thread overview]
Message-ID: <1338797382-287275-4-git-send-email-avagin@openvz.org> (raw)
In-Reply-To: <1338797382-287275-1-git-send-email-avagin@openvz.org>
prev_state is a state of task, when it stops executing on cpu.
It's needed for filtering sched:sched_switch_finish.
For example If we want to track when a task is sleeping:
./perf record -e sched:sched_switch -g --filter 'prev_state == 1' \
-e sched:sched_switch_finish --filter 'state == 1' -P foo
The previous state is set in sched:sched_switch, so it works when
both events sched:sched_switch and sched:sched_switch_finish are
enabled.
Signed-off-by: Andrew Vagin <avagin@openvz.org>
---
include/linux/sched.h | 3 +++
include/trace/events/sched.h | 12 +++++++++++-
2 files changed, 14 insertions(+), 1 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index f45c0b2..172bbe3 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1576,6 +1576,9 @@ struct task_struct {
struct uprobe_task *utask;
int uprobe_srcu_id;
#endif
+#ifdef CONFIG_EVENT_TRACING
+ volatile long prev_state;
+#endif
};
/* Future-safe accessor for struct task_struct's cpus_allowed. */
diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h
index b2219d8..c1b70b1 100644
--- a/include/trace/events/sched.h
+++ b/include/trace/events/sched.h
@@ -133,6 +133,7 @@ TRACE_EVENT(sched_switch,
__entry->prev_pid = prev->pid;
__entry->prev_prio = prev->prio;
__entry->prev_state = __trace_sched_switch_state(prev);
+ prev->prev_state = __trace_sched_switch_state(prev);
memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
__entry->next_pid = next->pid;
__entry->next_prio = next->prio;
@@ -158,14 +159,23 @@ TRACE_EVENT(sched_switch_finish,
TP_STRUCT__entry(
__array( char, comm, TASK_COMM_LEN )
__field( pid_t, pid )
+ __field( long, state )
),
TP_fast_assign(
__entry->pid = p->pid;
+ __entry->state = p->prev_state;
memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
),
- TP_printk("comm=%s pid=%d", __entry->comm, __entry->pid)
+ TP_printk("comm=%s pid=%d state=%s%s",
+ __entry->comm, __entry->pid,
+ __entry->state & (TASK_STATE_MAX-1) ?
+ __print_flags(__entry->state & (TASK_STATE_MAX-1), "|",
+ { 1, "S"} , { 2, "D" }, { 4, "T" }, { 8, "t" },
+ { 16, "Z" }, { 32, "X" }, { 64, "x" },
+ { 128, "W" }) : "R",
+ __entry->state & TASK_STATE_MAX ? "+" : "")
);
/*
--
1.7.1
next prev parent reply other threads:[~2012-06-04 8:10 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-04 8:09 [RFC] [PATCH 0/5] Teach perf tool to profile sleep times (V4) Andrew Vagin
2012-06-04 8:09 ` [PATCH 1/5] sched: event: add trace events when a task starts executing on a cpu Andrew Vagin
2012-06-04 8:09 ` [PATCH 2/5] sched: send the event sched_switch before perf_event_task_sched_out Andrew Vagin
2012-06-04 8:09 ` Andrew Vagin [this message]
2012-06-04 8:09 ` [PATCH 4/5] perf: teach "perf inject" to work with files Andrew Vagin
2012-06-04 8:09 ` [PATCH 5/5] perf: teach perf inject to merge sched_switch* events for profiling sleep times Andrew Vagin
2012-06-04 12:40 ` [RFC] [PATCH 0/5] Teach perf tool to profile sleep times (V4) Peter Zijlstra
2012-06-04 12:50 ` Peter Zijlstra
2012-06-04 13:46 ` Steven Rostedt
2012-06-04 17:01 ` Arun Sharma
2012-06-04 20:52 ` Peter Zijlstra
2012-06-06 8:32 ` Ingo Molnar
2012-06-12 13:15 ` Andrew Wagin
2012-06-12 17:46 ` Peter Zijlstra
2012-06-04 17:03 ` Arun Sharma
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=1338797382-287275-4-git-send-email-avagin@openvz.org \
--to=avagin@openvz.org \
--cc=OSTRIKOV@nvidia.com \
--cc=a.p.zijlstra@chello.nl \
--cc=asharma@fb.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=rostedt@goodmis.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
all inboxes | Powered by JetHome®