public inbox for dwarves@vger.kernel.org
 help / color / mirror / Atom feed
From: Jiri Olsa <olsajiri@gmail.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: dwarves@vger.kernel.org, "Clark Williams" <williams@redhat.com>,
	"Kate Carcia" <kcarcia@redhat.com>,
	bpf@vger.kernel.org, "Alan Maguire" <alan.maguire@oracle.com>,
	"Kui-Feng Lee" <kuifeng@fb.com>,
	"Thomas Weißschuh" <linux@weissschuh.net>
Subject: Re: [RFC/PATCHES 00/12] pahole: Reproducible parallel DWARF loading/serial BTF encoding
Date: Thu, 4 Apr 2024 11:42:24 +0200	[thread overview]
Message-ID: <Zg52AE0QByIT_07A@krava> (raw)
In-Reply-To: <20240402193945.17327-1-acme@kernel.org>

On Tue, Apr 02, 2024 at 04:39:33PM -0300, Arnaldo Carvalho de Melo wrote:
> Hi,
> 
> 	This allows us to have reproducible builds while keeping the
> DWARF loading phase in parallel, achieving a noticeable speedup as
> showed in the commit log messages:
> 
> On a:
> 
>   model name    : Intel(R) Core(TM) i7-14700K
> 
>   8 performance cores (16 threads), 12 efficiency cores.
> 
> Serial encoding:
> 
>   $ perf stat -e cycles -r5 pahole --btf_encode_detached=vmlinux.btf.serial vmlinux
>              5.18276 +- 0.00952 seconds time elapsed  ( +-  0.18% )
> 
> Parallel, but non-reproducible:
> 
>   $ perf stat -e cycles -r5 pahole -j --btf_encode_detached=vmlinux.btf.parallel vmlinux
>               1.8529 +- 0.0159 seconds time elapsed  ( +-  0.86% )
> 
> reproducible build done using parallel DWARF loading + CUs-ordered-as-in-vmlinux serial BTF encoding:
> 
>   $ perf stat -e cycles -r5 pahole -j --reproducible_build --btf_encode_detached=vmlinux.btf.parallel.reproducible_build vmlinux
>               2.3632 +- 0.0164 seconds time elapsed  ( +-  0.69% )

hm, got it even faster than paralel build ;-) but it's within the
1 second deviation, I guess it shows better on bigger kernels

reproducible_build:

	[root@krava linux-qemu]# perf stat -e cycles -r 3 -- /home/jolsa/kernel/bpf/pahole/build/pahole -j --reproducible_build --btf_encode_detached=btf.2 ./vmlinux

	 Performance counter stats for '/home/jolsa/kernel/bpf/pahole/build/pahole -j --reproducible_build --btf_encode_detached=btf.2 ./vmlinux' (3 runs):

	   295,519,117,258      cycles                                                                  ( +- 19.48% )

		      9.43 +- 1.02 seconds time elapsed  ( +- 10.84% )

paralel build:

	[root@krava linux-qemu]# perf stat -e cycles -r 3 -- /home/jolsa/kernel/bpf/pahole/build/pahole -j  --btf_encode_detached=btf.2 ./vmlinux

	 Performance counter stats for '/home/jolsa/kernel/bpf/pahole/build/pahole -j --btf_encode_detached=btf.2 ./vmlinux' (3 runs):

	   391,320,973,331      cycles                                                                  ( +- 19.19% )

		     9.851 +- 0.695 seconds time elapsed  ( +-  7.06% )

1 cpu:

	[root@krava linux-qemu]# perf stat -e cycles -r 3 -- /home/jolsa/kernel/bpf/pahole/build/pahole --btf_encode_detached=btf.2 ./vmlinux

	 Performance counter stats for '/home/jolsa/kernel/bpf/pahole/build/pahole --btf_encode_detached=btf.2 ./vmlinux' (3 runs):

	   208,492,342,135      cycles                                                                  ( +- 19.43% )

		    16.761 +- 0.644 seconds time elapsed  ( +-  3.84% )

jirka

> 
> Please take a look, its in the 'next' branch at:
> 
>   https://git.kernel.org/pub/scm/devel/pahole/pahole.git
>   https://git.kernel.org/pub/scm/devel/pahole/pahole.git/log/?h=next
> 
> There is a new tool to do regression testing on this feature:
> 
>   https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?h=next&id=c751214c19bf8591bf8e4abdc677cbadee08f630
>   
> And here a more detailed set of tests using it:
> 
>   https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?h=next&id=4451467ca16a6e31834f6f98661c63587ce556f7
> 
> Working on libbpf to allow for parallel reproducible BTF encoding is the
> next step.
> 
> Thanks a lot,
> 
> - Arnaldo
> 
> Arnaldo Carvalho de Melo (12):
>   core: Allow asking for a reproducible build
>   pahole: Disable BTF multithreaded encoded when doing reproducible builds
>   dwarf_loader: Separate creating the cu/dcu pair from processing it
>   dwarf_loader: Introduce dwarf_cus__process_cu()
>   dwarf_loader: Create the cu/dcu pair in dwarf_cus__nextcu()
>   dwarf_loader: Remove unused 'thr_data' arg from dwarf_cus__create_and_process_cu()
>   core: Add unlocked cus__add() variant
>   core: Add cus__remove(), counterpart of cus__add()
>   dwarf_loader: Add the cu to the cus list early, remove on LSK_DELETE
>   core/dwarf_loader: Add functions to set state of CU processing
>   pahole: Encode BTF serially in a reproducible build
>   tests: Add a BTF reproducible generation test
> 
>  dwarf_loader.c              | 73 +++++++++++++++++++++++---------
>  dwarves.c                   | 58 ++++++++++++++++++++++++-
>  dwarves.h                   | 17 ++++++++
>  pahole.c                    | 84 +++++++++++++++++++++++++++++++++++--
>  tests/reproducible_build.sh | 56 +++++++++++++++++++++++++
>  5 files changed, 264 insertions(+), 24 deletions(-)
>  create mode 100755 tests/reproducible_build.sh
> 
> -- 
> 2.44.0
> 

      parent reply	other threads:[~2024-04-04  9:42 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-02 19:39 [RFC/PATCHES 00/12] pahole: Reproducible parallel DWARF loading/serial BTF encoding Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 01/12] core: Allow asking for a reproducible build Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 02/12] pahole: Disable BTF multithreaded encoded when doing reproducible builds Arnaldo Carvalho de Melo
2024-04-03 18:19   ` Andrii Nakryiko
2024-04-03 21:38     ` Arnaldo Carvalho de Melo
2024-04-03 21:43       ` Andrii Nakryiko
2024-04-04  9:42   ` Jiri Olsa
2024-04-02 19:39 ` [PATCH 03/12] dwarf_loader: Separate creating the cu/dcu pair from processing it Arnaldo Carvalho de Melo
2024-04-04  9:42   ` Jiri Olsa
2024-04-02 19:39 ` [PATCH 04/12] dwarf_loader: Introduce dwarf_cus__process_cu() Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 05/12] dwarf_loader: Create the cu/dcu pair in dwarf_cus__nextcu() Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 06/12] dwarf_loader: Remove unused 'thr_data' arg from dwarf_cus__create_and_process_cu() Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 07/12] core: Add unlocked cus__add() variant Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 08/12] core: Add cus__remove(), counterpart of cus__add() Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 09/12] dwarf_loader: Add the cu to the cus list early, remove on LSK_DELETE Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 10/12] core/dwarf_loader: Add functions to set state of CU processing Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 11/12] pahole: Encode BTF serially in a reproducible build Arnaldo Carvalho de Melo
2024-04-02 19:39 ` [PATCH 12/12] tests: Add a BTF reproducible generation test Arnaldo Carvalho de Melo
2024-04-04  0:08 ` [RFC/PATCHES 00/12] pahole: Reproducible parallel DWARF loading/serial BTF encoding Eduard Zingerman
2024-04-04  8:05   ` Alan Maguire
2024-04-09 14:34     ` Eduard Zingerman
2024-04-09 14:56       ` Alexei Starovoitov
2024-04-09 15:01         ` Eduard Zingerman
2024-04-09 18:45           ` Arnaldo Carvalho de Melo
2024-04-09 19:29             ` Eduard Zingerman
2024-04-09 19:34               ` Alexei Starovoitov
2024-04-09 19:57               ` Arnaldo Carvalho de Melo
2024-04-12 20:37       ` Arnaldo Carvalho de Melo
2024-04-12 20:40         ` Eduard Zingerman
2024-04-12 21:09           ` Arnaldo Carvalho de Melo
2024-04-12 21:10             ` Eduard Zingerman
2024-04-04  8:58 ` Alan Maguire
2024-04-08 12:00   ` Alan Maguire
2024-04-08 14:39     ` Arnaldo Carvalho de Melo
2024-04-12 20:36       ` Arnaldo Carvalho de Melo
2024-04-04  9:42 ` Jiri Olsa [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=Zg52AE0QByIT_07A@krava \
    --to=olsajiri@gmail.com \
    --cc=acme@kernel.org \
    --cc=alan.maguire@oracle.com \
    --cc=bpf@vger.kernel.org \
    --cc=dwarves@vger.kernel.org \
    --cc=kcarcia@redhat.com \
    --cc=kuifeng@fb.com \
    --cc=linux@weissschuh.net \
    --cc=williams@redhat.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