LKML Archive on
help / color / mirror / Atom feed
From: Masami Hiramatsu <>
To: Arnaldo Carvalho de Melo <>
Cc: Frederic Weisbecker <>,
	Steven Rostedt <>, Ingo Molnar <>,,
	Andrew Morton <>,
	Thomas Gleixner <>,
	Mathieu Desnoyers <>,
	Lai Jiangshan <>,
	Li Zefan <>, Tom Zanussi <>,
	Peter Zijlstra <>,
Subject: Re: [PATCH 00/14] [GIT PULL][v2.6.39] tracing/filter: More robust filtering
Date: Wed, 16 Feb 2011 22:34:07 +0900	[thread overview]
Message-ID: <> (raw)
In-Reply-To: <>

(2011/02/16 3:35), Arnaldo Carvalho de Melo wrote:
> Em Tue, Feb 15, 2011 at 05:53:06PM +0100, Frederic Weisbecker escreveu:
>> On Tue, Feb 15, 2011 at 11:29:22AM -0500, Steven Rostedt wrote:
>>> On Tue, 2011-02-15 at 08:33 -0500, Steven Rostedt wrote:
>>>> Thanks for letting me waste three days on developing this. I even posted
>>>> an RFC a while back, and no one complained then.
>>> Sorry about being a bit bitchy in my reply here. I need to make a note
>>> not to reply to LKML before my first cup of coffee. ;)
>>> Arnaldo,
>>> Thanks for the post, I'll help you out where you need it. trace-cmd has
>>> some features that reports back to the user on failed filter usage. We
>>> can incorporate that into perf.
> Thanks!

Sounds good :)

>> Cool!
>> That said I agree that we should not block improvements in the generic
>> filtering code because of issues in perf uses of filters.
>> I believe it used to work better in perf by the past, but I saw similar
>> issues lately like those Ingo noticed. So probably something
>> broke and we need to investigate. But until then your patches are
>> still nice improvements: lesser memory usage, lesser kernel stack usage in the
>> fast path, lesser limitation, faster and smarter filter evaluation...
> Yeah, usability of the --filter parameter in perf is a bit
> (understatement) lacking, one has to look at the /format thing in
> /sys/kernel/debug/tracing... and not commit any mistake, else a generic
> invalid 22 message is spit out.

And also, there seems very less comment in perf-record.txt.

        Event filter.

Hmm, I think, at least, there should be a comment where user should
refer to and if possible, how he can use it...

> I tried it and after some back and forth changing hats and scratching my
> head while doing so, it got to work.
> I talked with Steven and the same operation using trace-cmd would
> produce a better error report, stating that the field used in the filter
> expression was not valid.

If possible, please show us what parameters we can use on that event too.
perf list gives us a list of events, but not show us what parameters
we can use on those events. I think we can update perf list for that

perf list --field kvm:kvm_entry
 unsigned short common_type;
 unsigned char common_flags;
 unsigned char common_preempt_count;
 int common_pid;
 int common_lock_depth;
 unsigned int vcpu_id;

> I'll try to get that code from trace-cmd and glue that into
> tools/perf/util/, that eventually will get moved to tools/lib/ or
> something like that, as Borislav has been experimenting with for some
> time already.
> The location in the source tree is not the most important thing at this
> point, usability improvements are, so I'm not rushing to moving code
> around all the time (even doing it more than I would like), lets try to
> improve usability first and then we can move it to tools/lib.
> - Arnaldo

Thank you,

2nd Dept. Linux Technology Center
Hitachi, Ltd., Systems Development Laboratory

  reply	other threads:[~2011-02-16 13:34 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-08  1:56 Steven Rostedt
2011-02-08  1:56 ` [PATCH 01/14] tracing/filter: Have no filter return a match Steven Rostedt
2011-02-08  1:56 ` [PATCH 02/14] tracing/filter: Move OR and AND logic out of fn() method Steven Rostedt
2011-02-08  1:56 ` [PATCH 03/14] tracing/filter: Dynamically allocate preds Steven Rostedt
2011-02-08  1:56 ` [PATCH 04/14] tracing/filter: Call synchronize_sched() just once for system filters Steven Rostedt
2011-02-08  1:56 ` [PATCH 05/14] tracing/filter: Allocate the preds in an array Steven Rostedt
2011-02-08  1:56 ` [PATCH 06/14] tracing/filter: Free pred array on disabling of filter Steven Rostedt
2011-02-08  1:56 ` [PATCH 07/14] tracing/filter: Use a tree instead of stack for filter_match_preds() Steven Rostedt
2011-02-08  1:56 ` [PATCH 08/14] tracing/filter: Optimize short ciruit check Steven Rostedt
2011-02-08  1:56 ` [PATCH 09/14] tracing/filter: Check the created pred tree Steven Rostedt
2011-02-08  1:56 ` [PATCH 10/14] tracing/filter: Optimize filter by folding the tree Steven Rostedt
2011-02-08  1:56 ` [PATCH 11/14] tracing/filter: Move MAX_FILTER_PRED to local tracing directory Steven Rostedt
2011-02-08  1:56 ` [PATCH 12/14] tracing/filter: Increase the max preds to 2^14 Steven Rostedt
2011-02-08  1:56 ` [PATCH 13/14] tracing/filter: Swap entire filter of events Steven Rostedt
2011-02-08  1:56 ` [PATCH 14/14] tracing/filter: Remove synchronize_sched() from __alloc_preds() Steven Rostedt
2011-02-15  4:44 ` [PATCH 00/14] [GIT PULL][v2.6.39] tracing/filter: More robust filtering Ingo Molnar
2011-02-15 13:33   ` Steven Rostedt
2011-02-15 16:29     ` Steven Rostedt
2011-02-15 16:53       ` Frederic Weisbecker
2011-02-15 18:35         ` Arnaldo Carvalho de Melo
2011-02-16 13:34           ` Masami Hiramatsu [this message]
2011-02-16 14:52             ` Arnaldo Carvalho de Melo
2011-02-15 18:42     ` Ingo Molnar
2011-02-15 18:59       ` Steven Rostedt
2011-02-16  9:10         ` Ingo Molnar
2011-02-15 13:44   ` Arnaldo Carvalho de Melo

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:

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

  git send-email \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
    --subject='Re: [PATCH 00/14] [GIT PULL][v2.6.39] tracing/filter: More robust filtering' \

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

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).