public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "xinliangli at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/35545] tracer pass is run too late
Date: Sun, 28 Sep 2014 18:22:00 -0000 [thread overview]
Message-ID: <bug-35545-4-ptsd1w5c74@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-35545-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=35545
--- Comment #19 from davidxl <xinliangli at gmail dot com> ---
(In reply to Jan Hubicka from comment #18)
> > WE can try some internal benchmarks with this change too.
>
> That would be very welcome. Tracer used to be quite useful pass in old days,
> doing 1.6% on -O3+FDO SPECint (for 1.4% code size cost) that put it very
> close
> to the inliner (1.8%) that however needed 12.9% of code size.
>
> http://www.ucw.cz/~hubicka/papers/amd64/node4.html (see aggressive
> optimization table)
>
> I do not think it was ever resonably returned for GIMPLE. Martin, do you have
> any current scores? Part of benefits came from lack of global optimizers, but
> I think tail duplication has a potential in modern compiler, too.
>
> Perhaps we should try to do that and given that we now have tail merging
> pass,
> consider getting it useful without FDO, too.
>
> Honza
I saw small improvement across most of the benchmarks with this change on
sandybridge machines with FDO. The geomean improvement is about 0.3%. The
baseline compiler is Google gcc-49 compiler, and the test compiler is Google
gcc-49 with the patch.
next prev parent reply other threads:[~2014-09-28 18:22 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <bug-35545-4@http.gcc.gnu.org/bugzilla/>
2013-12-17 16:34 ` [Bug middle-end/35545] virtual call specialization not happening with FDO hubicka at gcc dot gnu.org
2013-12-17 17:29 ` hubicka at gcc dot gnu.org
2013-12-17 18:06 ` rguenther at suse dot de
2013-12-17 18:09 ` hubicka at gcc dot gnu.org
2013-12-17 20:33 ` law at redhat dot com
2013-12-17 20:39 ` hubicka at ucw dot cz
2013-12-17 20:53 ` law at redhat dot com
2013-12-17 21:15 ` hubicka at ucw dot cz
2014-09-25 21:29 ` hubicka at gcc dot gnu.org
2014-09-26 7:44 ` [Bug middle-end/35545] tracer pass is run too late rguenth at gcc dot gnu.org
2014-09-26 15:52 ` law at redhat dot com
2014-09-27 0:04 ` hubicka at gcc dot gnu.org
2014-09-27 0:14 ` hubicka at gcc dot gnu.org
2014-09-27 0:23 ` xinliangli at gmail dot com
2014-09-27 1:04 ` hubicka at ucw dot cz
2014-09-28 18:22 ` xinliangli at gmail dot com [this message]
2014-09-29 8:17 ` rguenther at suse dot de
2014-09-29 10:22 ` rguenther at suse dot de
2014-09-29 16:48 ` hubicka at ucw dot cz
2014-09-29 19:30 ` xinliangli at gmail dot com
2014-10-01 11:30 ` mliska at suse dot cz
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=bug-35545-4-ptsd1w5c74@http.gcc.gnu.org/bugzilla/ \
--to=gcc-bugzilla@gcc.gnu.org \
--cc=gcc-bugs@gcc.gnu.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 read-only IMAP folder(s) and NNTP newsgroup(s).