linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Namhyung Kim <namhyung@kernel.org>
To: Ian Rogers <irogers@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@kernel.org>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Kan Liang <kan.liang@linux.intel.com>,
	Masahiro Yamada <masahiroy@kernel.org>,
	Arnd Bergmann <arnd@arndb.de>, Jann Horn <jannh@google.com>,
	Colin Ian King <colin.i.king@gmail.com>,
	Casey Chen <cachen@purestorage.com>,
	Athira Rajeev <atrajeev@linux.vnet.ibm.com>,
	Chaitanya S Prakash <chaitanyas.prakash@arm.com>,
	James Clark <james.clark@linaro.org>,
	Ze Gao <zegao2021@gmail.com>,
	Yang Jihong <yangjihong1@huawei.com>,
	Yunseong Kim <yskelg@gmail.com>,
	Weilin Wang <weilin.wang@intel.com>,
	Dominique Martinet <asmadeus@codewreck.org>,
	Anne Macedo <retpolanne@posteo.net>,
	Sun Haiyong <sunhaiyong@loongson.cn>,
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1 00/13] perf inject improvements
Date: Mon, 2 Sep 2024 11:27:04 -0700	[thread overview]
Message-ID: <ZtYDeE5kTy4KrRC5@google.com> (raw)
In-Reply-To: <20240817064442.2152089-1-irogers@google.com>

On Fri, Aug 16, 2024 at 11:44:29PM -0700, Ian Rogers wrote:
> Fix the existing build id injection by adding sample IDs on to the
> synthesized events. This correctly orders the events and addresses
> issues such as a profiled executable being replaced during its
> execution.
> 
> Add a new --mmap2-buildid-all option that rewrites all mmap events as
> mmap2 events containing build IDs. This removes the need for build_id
> events.
> 
> Add a new -B option that like --mmap2-buildid-all synthesizes mmap2
> with build id events. With -B the behavior is to do it lazily, so only
> when a sample references the particular map. With system wide
> profiling that synthesizes mmap events for all running processes the
> perf.data file savings can be greater than 50%.
> 
> Reduce the memory footprint of perf inject by avoiding creating
> symbols in the callchain, the symbols aren't used during perf inject
> and necessitate the loading of dsos.
> 
> Ian Rogers (13):
>   perf synthetic-events: Avoid unnecessary memset
>   perf map: API clean up
>   perf jit: Constify filename argument
>   perf dso: Constify dso_id
>   perf evsel: Constify evsel__id_hdr_size argument
>   perf test: Expand pipe/inject test
>   perf inject: Combine build_ids and build_id_all into enum
>   perf inject: Combine different mmap and mmap2 functions
>   perf inject: Combine mmap and mmap2 handling
>   perf inject: Fix build ID injection
>   perf inject: Add new mmap2-buildid-all option
>   perf inject: Lazy build-id mmap2 event insertion
>   perf callchain: Allow symbols to be optional when resolving a
>     callchain

For the remaining bits,

Acked-by: Namhyung Kim <namhyung@kernel.org>

Thanks,
Namhyung

> 
>  tools/perf/builtin-inject.c         | 532 ++++++++++++++++++----------
>  tools/perf/builtin-top.c            |   2 +-
>  tools/perf/tests/shell/pipe_test.sh | 103 ++++--
>  tools/perf/tests/vmlinux-kallsyms.c |   4 +-
>  tools/perf/util/build-id.c          |   6 +-
>  tools/perf/util/callchain.c         |   8 +-
>  tools/perf/util/callchain.h         |   2 +-
>  tools/perf/util/dso.c               |   4 +-
>  tools/perf/util/dso.h               |   4 +-
>  tools/perf/util/dsos.c              |  12 +-
>  tools/perf/util/dsos.h              |   2 +-
>  tools/perf/util/evsel.c             |   2 +-
>  tools/perf/util/evsel.h             |   2 +-
>  tools/perf/util/jit.h               |   3 +-
>  tools/perf/util/jitdump.c           |   6 +-
>  tools/perf/util/machine.c           |  95 ++---
>  tools/perf/util/machine.h           |  36 +-
>  tools/perf/util/map.c               |  25 +-
>  tools/perf/util/map.h               |  22 +-
>  tools/perf/util/synthetic-events.c  | 103 +++++-
>  tools/perf/util/synthetic-events.h  |  21 +-
>  21 files changed, 682 insertions(+), 312 deletions(-)
> 
> -- 
> 2.46.0.184.g6999bdac58-goog
> 

      parent reply	other threads:[~2024-09-02 18:27 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-17  6:44 [PATCH v1 00/13] perf inject improvements Ian Rogers
2024-08-17  6:44 ` [PATCH v1 01/13] perf synthetic-events: Avoid unnecessary memset Ian Rogers
2024-08-17  6:44 ` [PATCH v1 02/13] perf map: API clean up Ian Rogers
2024-08-17  6:44 ` [PATCH v1 03/13] perf jit: Constify filename argument Ian Rogers
2024-08-17  6:44 ` [PATCH v1 04/13] perf dso: Constify dso_id Ian Rogers
2024-08-17  6:44 ` [PATCH v1 05/13] perf evsel: Constify evsel__id_hdr_size argument Ian Rogers
2024-08-17  6:44 ` [PATCH v1 06/13] perf test: Expand pipe/inject test Ian Rogers
2024-08-17  6:44 ` [PATCH v1 07/13] perf inject: Combine build_ids and build_id_all into enum Ian Rogers
2024-08-17  6:44 ` [PATCH v1 08/13] perf inject: Combine different mmap and mmap2 functions Ian Rogers
2024-08-17  6:44 ` [PATCH v1 09/13] perf inject: Combine mmap and mmap2 handling Ian Rogers
2024-08-17  6:44 ` [PATCH v1 10/13] perf inject: Fix build ID injection Ian Rogers
2024-08-19 18:01   ` Arnaldo Carvalho de Melo
2024-08-19 19:54     ` Ian Rogers
2024-08-28 15:15       ` Ian Rogers
2024-09-02 18:26         ` Namhyung Kim
2024-09-03 18:34           ` Arnaldo Carvalho de Melo
2024-08-17  6:44 ` [PATCH v1 11/13] perf inject: Add new mmap2-buildid-all option Ian Rogers
2024-08-17  6:44 ` [PATCH v1 12/13] perf inject: Lazy build-id mmap2 event insertion Ian Rogers
2024-08-17  6:44 ` [PATCH v1 13/13] perf callchain: Allow symbols to be optional when resolving a callchain Ian Rogers
2024-09-02 18:27 ` Namhyung Kim [this message]

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=ZtYDeE5kTy4KrRC5@google.com \
    --to=namhyung@kernel.org \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=arnd@arndb.de \
    --cc=asmadeus@codewreck.org \
    --cc=atrajeev@linux.vnet.ibm.com \
    --cc=cachen@purestorage.com \
    --cc=chaitanyas.prakash@arm.com \
    --cc=colin.i.king@gmail.com \
    --cc=irogers@google.com \
    --cc=james.clark@linaro.org \
    --cc=jannh@google.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=masahiroy@kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=retpolanne@posteo.net \
    --cc=sunhaiyong@loongson.cn \
    --cc=weilin.wang@intel.com \
    --cc=yangjihong1@huawei.com \
    --cc=yskelg@gmail.com \
    --cc=zegao2021@gmail.com \
    /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).