From: David Ahern <dsahern@gmail.com>
To: Adrian Hunter <adrian.hunter@intel.com>,
Stephane Eranian <eranian@google.com>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
linux-kernel@vger.kernel.org,
Frederic Weisbecker <fweisbec@gmail.com>,
Jiri Olsa <jolsa@redhat.com>, Mike Galbraith <efault@gmx.de>,
Namhyung Kim <namhyung@gmail.com>,
Paul Mackerras <paulus@samba.org>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@kernel.org>
Subject: Re: [PATCH V5 09/12] perf: make events stream always parsable
Date: Thu, 11 Jul 2013 11:16:48 -0600 [thread overview]
Message-ID: <51DEE880.6070908@gmail.com> (raw)
In-Reply-To: <51DED2AA.6030307@gmail.com>
On 7/11/13 9:43 AM, David Ahern wrote:
> On 7/11/13 7:12 AM, Adrian Hunter wrote:
>> The event stream is not always parsable because the format of a sample
>> is dependent on the sample_type of the selected event. When there
>> is more than one selected event and the sample_types are not the
>> same then parsing becomes problematic. A sample can be matched to its
>> selected event using the ID that is allocated when the event is opened.
>> Unfortunately, to get the ID from the sample means first parsing it.
>
> Here's an alternative suggestion -- one that does not involve changing
> the kernel API or requiring a common denominator in sample_type options.
>
> perf handles event streams through an mmap which can be directly tied to
> an evsel (a single event) when the mmap is created. ie., when events are
> read we know exactly which evsel they correspond to. (See
> perf_evlist__mmap_per_cpu and perf_evlist__mmap_per_thread and add
> struct perf_evsel *evsel entry to struct perf_mmap).
Read that code a bit too quickly. All events for a thread are dropped
into the same buffer so as currently organized there is not an easy
correlation.
David
>
> Commands like perf-record can inject a user event into the stream and
> hence the data file every time the evsel changes while walking all of
> the mmap's reading events -- very similar to the way finished round is
> done. The event would only contain a perf_event_header which is 8 bytes
> so this does not add a lot to a data file. As an optimization the evsel
> event could only be injected if the sample_types differ.
>
> Live commands would just use the evsel connected to the mmap -- no
> lookups needed which would simplify things a bit processing the events.
>
> In short, the information to associate event streams to an event (evsel)
> is currently available -- it's just being discarded in the many layers.
>
> I'll try to whip up some code that implements this in the next few days.
>
> David
>
next prev parent reply other threads:[~2013-07-11 17:16 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-11 13:12 [PATCH V5 00/12] perf tools: some fixes and tweaks Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 01/12] perf tools: add debug prints Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 02/12] perf tools: allow non-matching sample types Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 03/12] perf tools: add pid to struct thread Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 04/12] perf tools: change machine__findnew_thread() to set thread pid Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 05/12] perf tools: tidy up sample parsing overflow checking Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 06/12] perf tools: remove unnecessary callchain validation Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 07/12] perf tools: remove references to struct ip_event Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 08/12] perf tools: move " Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 09/12] perf: make events stream always parsable Adrian Hunter
2013-07-11 15:43 ` David Ahern
2013-07-11 17:16 ` David Ahern [this message]
2013-07-12 6:42 ` Adrian Hunter
2013-07-12 9:56 ` Peter Zijlstra
2013-07-12 12:56 ` Adrian Hunter
2013-07-12 14:55 ` Peter Zijlstra
2013-07-15 6:14 ` Adrian Hunter
2013-07-15 11:53 ` Stephane Eranian
2013-07-15 12:09 ` Adrian Hunter
2013-07-16 6:49 ` Adrian Hunter
2013-07-16 15:09 ` Peter Zijlstra
2013-07-17 4:10 ` Stephane Eranian
2013-07-17 12:44 ` Adrian Hunter
2013-07-24 3:55 ` [tip:perf/core] perf: Update perf_event_type documentation tip-bot for Peter Zijlstra
2013-07-24 17:54 ` Vince Weaver
2013-07-25 6:22 ` Adrian Hunter
2013-07-25 12:34 ` Vince Weaver
2013-07-25 16:27 ` Peter Zijlstra
2013-07-26 3:24 ` Vince Weaver
2013-07-26 8:55 ` Peter Zijlstra
2013-07-26 3:20 ` Vince Weaver
2013-07-26 8:57 ` Peter Zijlstra
2013-07-27 3:20 ` Vince Weaver
2013-07-25 10:04 ` Peter Zijlstra
2013-07-25 12:31 ` Vince Weaver
2013-07-25 12:39 ` David Ahern
2013-09-13 21:31 ` Vince Weaver
2013-09-13 21:42 ` Peter Zijlstra
2013-07-11 13:12 ` [PATCH V5 10/12] perf tools: add support for PERF_SAMPLE_IDENTFIER Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 11/12] perf tools: expand perf_event__synthesize_sample() Adrian Hunter
2013-07-11 13:12 ` [PATCH V5 12/12] perf tools: add a sample parsing test Adrian Hunter
2013-07-16 12:48 ` Jiri Olsa
2013-07-17 11:02 ` Adrian Hunter
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=51DEE880.6070908@gmail.com \
--to=dsahern@gmail.com \
--cc=acme@ghostprotocols.net \
--cc=adrian.hunter@intel.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@gmail.com \
--cc=paulus@samba.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).