From: Frederic Weisbecker <fweisbec@gmail.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: LKML <linux-kernel@vger.kernel.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Paul Mackerras <paulus@samba.org>
Subject: [PATCH 6/7] perf: Don't print traces when debugging ordering
Date: Thu, 24 Jun 2010 23:40:57 +0200 [thread overview]
Message-ID: <1277415658-10036-7-git-send-regression-fweisbec@gmail.com> (raw)
In-Reply-To: <1277415658-10036-1-git-send-regression-fweisbec@gmail.com>
Errors due to ordering bugs are easily lost in the middle
of traces.
When we are in this mode, don't print the traces so that
we don't miss the debugging messages.
But display a comforting message if we didn't encounter any
ordering problem.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
---
tools/perf/builtin-trace.c | 12 +++++++++++-
1 files changed, 11 insertions(+), 1 deletions(-)
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index dddf3f0..83df8db 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -13,6 +13,7 @@ static char const *script_name;
static char const *generate_script_lang;
static bool debug_ordering;
static u64 last_timestamp;
+static u64 nr_unordered;
static int default_start_script(const char *script __unused,
int argc __unused,
@@ -96,8 +97,10 @@ static int process_sample_event(event_t *event, struct perf_session *session)
pr_err("Samples misordered, previous: %llu "
"this: %llu\n", last_timestamp,
data.time);
+ nr_unordered++;
}
last_timestamp = data.time;
+ return 0;
}
/*
* FIXME: better resolve from pid from the struct trace_entry
@@ -132,9 +135,16 @@ static void sig_handler(int sig __unused)
static int __cmd_trace(struct perf_session *session)
{
+ int ret;
+
signal(SIGINT, sig_handler);
- return perf_session__process_events(session, &event_ops);
+ ret = perf_session__process_events(session, &event_ops);
+
+ if (debug_ordering)
+ pr_err("Misordered timestamps: %llu\n", nr_unordered);
+
+ return ret;
}
struct script_spec {
--
1.6.2.3
next prev parent reply other threads:[~2010-06-24 21:41 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-24 21:40 [GIT PULL] perf events and breakpoints updates Frederic Weisbecker
2010-06-24 21:40 ` [PATCH 1/7] hw_breakpoints: Fix per task breakpoint tracking Frederic Weisbecker
2010-06-24 21:40 ` [PATCH 2/7] perf: Fix argument of perf_arch_fetch_caller_regs Frederic Weisbecker
2010-06-24 21:40 ` [PATCH 3/7] x86: Set resume bit before returning from breakpoint exception Frederic Weisbecker
2010-06-24 21:40 ` [PATCH 4/7] x86: Support for instruction breakpoints Frederic Weisbecker
2010-06-24 21:40 ` [PATCH 5/7] perf: Don't use 4 bytes as a default instruction breakpoint length Frederic Weisbecker
2010-06-24 21:40 ` Frederic Weisbecker [this message]
2010-06-24 21:40 ` [PATCH 7/7] perf: Report lost events in perf trace debug mode Frederic Weisbecker
2010-06-25 9:33 ` [GIT PULL] perf events and breakpoints updates 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=1277415658-10036-7-git-send-regression-fweisbec@gmail.com \
--to=fweisbec@gmail.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.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