From: Jeff Law <law@redhat.com>
To: "Martin Liška" <mliska@suse.cz>, "GCC Development" <gcc@gcc.gnu.org>
Cc: Jan Hubicka <hubicka@ucw.cz>
Subject: Re: [RFC] zstd as a compression algorithm for LTO
Date: Wed, 19 Jun 2019 16:03:00 -0000 [thread overview]
Message-ID: <ecf95d2c-bc83-1fef-3d0d-74db34d7f1ea@redhat.com> (raw)
In-Reply-To: <89e98dc0-e766-ffef-da0f-263c3b284e96@suse.cz>
On 6/19/19 3:21 AM, Martin Liška wrote:
> Hi.
>
> I've written a patch draft that replaces zlib with the zstd compression algorithm ([1])
> in LTO. I'm also sending statistics that are collected for couple of quite big C++ source
> files. Observation I did:
>
> - LTO stream compression takes 3-4% of LGEN compile time
> - zstd in default compression level (3) generated slighly smaller LTO elf files
> - zstd compression is 4-8x faster
> - decompression is quite negligible, but for a bigger project (godot) I can
> reduction from 1.37 to 0.53 seconds
> - ZSTD API is much simpler to use
>
> Suggestion based on the observation:
> - I would suggest to make zstd optional (--enable-zstd) and one would
> use #include <zstd> + -lzstd
> - I like the default level as we want to mainly speed up LTO compilation
> - we can provide an option to control algorithm (-flto-compression-algorithm),
> similarly to -flto-compression-level
> - we can discuss possible compression of LTO bytecode that is distributed between WPA
> stage and individual LTRANS phases.
Presumably the reason we're not being more aggressive about switching is
the build/run time dependency on zstd? I wonder if we could default to
zstd and fallback to zlib when zstd isn't available?
jeff
next prev parent reply other threads:[~2019-06-19 16:03 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-19 9:21 Martin Liška
2019-06-19 16:03 ` Jeff Law [this message]
2019-06-19 18:55 ` Richard Biener
2019-06-19 19:26 ` Andrew Pinski
2019-06-19 19:29 ` Jan Hubicka
2019-06-19 19:34 ` Andrew Pinski
2019-06-20 9:08 ` Martin Liška
2019-06-20 10:59 ` Thomas Koenig
2019-06-20 11:42 ` Martin Liška
2019-06-20 12:02 ` Jan Hubicka
2019-06-21 10:20 ` [PATCH] Add .gnu.lto_.meta section Martin Liška
2019-06-21 12:34 ` Richard Biener
2019-06-21 12:49 ` Martin Liška
2019-06-21 12:57 ` Jan Hubicka
2019-06-21 14:01 ` Martin Liška
2019-06-24 12:02 ` Richard Biener
2019-06-24 12:12 ` Martin Liška
2019-06-24 12:44 ` Richard Biener
2019-06-24 13:31 ` Martin Liška
2019-06-24 14:25 ` Iain Sandoe
2019-06-24 18:05 ` Richard Biener
2019-06-25 8:14 ` Martin Liška
2019-06-25 14:15 ` Richard Biener
2019-06-27 12:28 ` [PATCH] Add .gnu.lto_.lto section Martin Liška
2019-07-01 10:59 ` Martin Liška
2019-07-01 11:01 ` [PATCH 2/2] Add zstd support for LTO bytecode compression Martin Liška
2019-07-02 20:50 ` Jeff Law
2019-07-02 20:49 ` [PATCH] Add .gnu.lto_.lto section Jeff Law
2019-06-20 12:12 ` [RFC] zstd as a compression algorithm for LTO Thomas Koenig
2019-06-20 17:02 ` Joseph Myers
2019-06-20 10:46 ` Segher Boessenkool
2019-06-20 11:44 ` Martin Liška
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=ecf95d2c-bc83-1fef-3d0d-74db34d7f1ea@redhat.com \
--to=law@redhat.com \
--cc=gcc@gcc.gnu.org \
--cc=hubicka@ucw.cz \
--cc=mliska@suse.cz \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).