public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "hubicka at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug ipa/65076] [5 Regression] 16% tramp3d-v4.cpp compile time regression
Date: Sat, 21 Mar 2015 05:32:00 -0000	[thread overview]
Message-ID: <bug-65076-4-3TZjqXOfIh@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-65076-4@http.gcc.gnu.org/bugzilla/>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65076

--- Comment #9 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Concerning Comment #7, I do not think the sreal refactoring screwed things up.
sreals are not high on profile and the code generated is not worse (performance
wise). It is not better, but it is not a surprise - we traditionally keep
inline limits high enough to get tramp3d performance good.

GCC 5 currently generates 11% bigger code than GCC 4.9. Not a good thing, but
from POV of the inliner heuristics it is - it is trying to do as much as
posisble of code duplication before it hits the limits.  New badness metrics
performs better on tramp3d - it does all useful inlining at 10% of unit growth,
hile old one needed 25%. It prioritizes more cases where inliner knows DCE will
happen because of propagation across arguments and thus it manages to get later
optimizers more busy.

Still plan to look into this more, but I think it is kind of non-bug (just
showing the fact that inliner is stupid to believe that all inlining is a good
idea, but there is not much to do except getting realistic static program
profiles that it out of current state of art).


  parent reply	other threads:[~2015-03-20 23:00 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-16 13:00 [Bug ipa/65076] New: " trippels at gcc dot gnu.org
2015-02-16 13:14 ` [Bug ipa/65076] " rguenth at gcc dot gnu.org
2015-02-16 13:22 ` trippels at gcc dot gnu.org
2015-02-16 18:31 ` hubicka at gcc dot gnu.org
2015-02-16 19:07 ` trippels at gcc dot gnu.org
2015-02-16 19:15 ` trippels at gcc dot gnu.org
2015-02-17 10:49 ` rguenth at gcc dot gnu.org
2015-03-04  9:10 ` rguenth at gcc dot gnu.org
2015-03-18 12:56 ` rguenth at gcc dot gnu.org
2015-03-21  5:32 ` hubicka at gcc dot gnu.org [this message]
2015-03-21 10:25 ` hubicka at gcc dot gnu.org
2015-03-21 10:48 ` hubicka at gcc dot gnu.org
2015-03-21 11:32 ` hubicka at gcc dot gnu.org
2015-03-24 14:56 ` rguenth at gcc dot gnu.org
2015-03-24 17:21 ` jakub at gcc dot gnu.org
2015-03-25  8:42 ` hubicka at gcc dot gnu.org
2015-03-25  8:42 ` hubicka at ucw dot cz
2015-03-25  8:50 ` rguenther at suse dot de
2015-03-25 21:35 ` hubicka at gcc dot gnu.org
2015-03-26  3:23 ` hubicka at gcc dot gnu.org
2015-03-27  4:03 ` hubicka at gcc dot gnu.org
2015-03-27  6:21 ` hubicka at gcc dot gnu.org
2015-03-27  9:34 ` rguenth at gcc dot gnu.org
2015-03-28 22:17 ` hubicka at gcc dot gnu.org
2015-03-29 14:29 ` trippels at gcc dot gnu.org
2015-03-30  6:03 ` hubicka at gcc dot gnu.org
2015-03-30  6:15 ` hubicka at gcc dot gnu.org
2015-03-31 13:03 ` rguenth at gcc dot gnu.org
2015-03-31 13:09 ` trippels at gcc dot gnu.org
2015-03-31 14:10 ` rguenther at suse dot de
2015-03-31 14:13 ` trippels at gcc dot gnu.org
2015-03-31 14:47 ` trippels at gcc dot gnu.org
2015-03-31 15:08 ` trippels at gcc dot gnu.org
2015-03-31 15:10 ` evstupac at gmail dot com
2015-03-31 15:29 ` rguenth at gcc dot gnu.org
2015-03-31 16:06 ` hubicka at ucw dot cz
2015-03-31 16:25 ` hubicka at gcc dot gnu.org
2015-03-31 17:36 ` hubicka at gcc dot gnu.org
2015-03-31 17:53 ` hubicka at gcc dot gnu.org
2015-03-31 17:54 ` hubicka at gcc dot gnu.org
2015-03-31 20:31 ` trippels at gcc dot gnu.org
2015-04-01  8:02 ` rguenther at suse dot de
2015-04-01  8:05 ` rguenther at suse dot de
2015-04-01  8:28 ` hubicka at gcc dot gnu.org
2015-04-01  8:34 ` trippels at gcc dot gnu.org
2015-04-02  5:18 ` hubicka at gcc dot gnu.org
2015-04-02  7:07 ` hubicka at gcc dot gnu.org
2015-04-02 23:44 ` hubicka at gcc dot gnu.org
2015-04-03 18:09 ` hubicka at gcc dot gnu.org
2015-04-07  9:32 ` rguenth at gcc dot gnu.org
2015-04-22  1:32 ` [Bug ipa/65076] [5/6 " hubicka at gcc dot gnu.org
2015-04-22 12:06 ` jakub at gcc dot gnu.org
2015-07-16  9:19 ` rguenth at gcc dot gnu.org

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-65076-4-3TZjqXOfIh@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).