public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jan Hubicka <jh@suse.cz>
To: Peter Steinmetz <steinmtz@us.ibm.com>
Cc: gcc@gcc.gnu.org, janis187 <janis187@us.ibm.com>,
	girish_vaithees@yahoo.com, hubicka@ucw.cz
Subject: Re: -fprofile-generate and -fprofile-use
Date: Tue, 30 Aug 2005 19:58:00 -0000	[thread overview]
Message-ID: <20050830162401.GB15817@kam.mff.cuni.cz> (raw)
In-Reply-To: <OF05184606.4EBE0844-ON8625706D.0055D24F-8625706D.00573FA5@us.ibm.com>

> 
> There was some discussion a few weeks ago about some apps running slower
> with FDO enabled.
> 
> I've recently investigated a similar situation using mainline.  In my case,
> the fact that the loop_optimize pass is disabled during FDO was the cause
> of the slowdown.  It appears that was recently disabled as part of Jan
> Hubicka's patch to eliminate RTL based profiling.  The commentary indicates
> that the old loop optimizer is incompatible with tree profiling.
> 
> While this doesn't explain all of the degradations discussed (some were
> showing up on older versions of the compiler), it may explain some.

Do you have specific testcase?  It would be interesting to see if new
optimizer can catch up at least on kill-loop branch.

Thanks for investigating!
Honza
> 
> Pete

  parent reply	other threads:[~2005-08-30 16:24 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-08-30 16:24 Peter Steinmetz
2005-08-30 16:24 ` Steven Bosscher
2005-08-30 17:57   ` Peter Steinmetz
2005-08-31 11:44     ` Zdenek Dvorak
2005-08-31 18:16       ` Peter Steinmetz
2005-09-01 19:04         ` Jan Hubicka
2005-09-02  9:01           ` Zdenek Dvorak
2005-08-30 19:58 ` Jan Hubicka [this message]
2005-08-31  1:25   ` girish vaitheeswaran
2005-08-31  4:33   ` Peter Steinmetz
     [not found] <AE7DE2E6-9131-4407-9585-746013E36070@apple.com>
2005-09-01 21:19 ` girish vaitheeswaran
2005-09-01 21:45   ` Steven Bosscher
2005-09-01 22:54     ` Janis Johnson
2005-09-01 23:13       ` Steven Bosscher
2005-09-02 15:29       ` Kai Henningsen
  -- strict thread matches above, loose matches on Subject: below --
2005-07-20 17:45 girish vaitheeswaran
2005-07-20 19:01 ` Janis Johnson
2005-07-20 20:16   ` girish vaitheeswaran
2005-07-20 22:44   ` Jan Hubicka
2005-07-20 23:38     ` girish vaitheeswaran
2005-07-21 13:03       ` Jan Hubicka
2005-07-25 17:39         ` girish vaitheeswaran
2005-07-26  7:00           ` Jan Hubicka
2005-07-26 18:10             ` girish vaitheeswaran
2005-07-30 19:20               ` Jan Hubicka
2005-07-21 18:03       ` Kelley Cook
2005-07-21 21:15         ` girish vaitheeswaran
2005-07-20 16:53 girish vaitheeswaran
2005-07-20 16:59 ` Steven Bosscher

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=20050830162401.GB15817@kam.mff.cuni.cz \
    --to=jh@suse.cz \
    --cc=gcc@gcc.gnu.org \
    --cc=girish_vaithees@yahoo.com \
    --cc=hubicka@ucw.cz \
    --cc=janis187@us.ibm.com \
    --cc=steinmtz@us.ibm.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 read-only IMAP folder(s) and NNTP newsgroup(s).