This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X

From: "Richard Guenther" <richard dot guenther at gmail dot com>
To: "Gabriel Dos Reis" <gdr at integrable-solutions dot net>
Cc: "Ernest L. Williams Jr." <ernesto at ornl dot gov>, "Nickolay Kolchin" <nbkolchin at gmail dot com>, gcc at gcc dot gnu dot org
Date: Sun, 12 Mar 2006 18:18:41 +0100
Subject: Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
References: <2636500f0603112016u1e5990f8vbe8eecde35baa504@mail.gmail.com> <84fc9c000603120243u64a5c571na1e2286871173a67@mail.gmail.com> <2636500f0603120555h39336789tad3f7479053137d@mail.gmail.com> <1142172601.20144.3.camel@bengal> <84fc9c000603120617u24012605kbc0b1e4d99028ba@mail.gmail.com> <1142175491.20144.5.camel@bengal> <84fc9c000603120707m3137bbbax86d9d94fdbd2075e@mail.gmail.com> <m3bqwby4hl.fsf@uniton.integrable-solutions.net> <84fc9c000603120913q58f636e9v2321e676d2f31d90@mail.gmail.com>

On 3/12/06, Richard Guenther <richard.guenther@gmail.com> wrote:
> So, I tried to reproduce the slowdown and on i686 get all
> memcpy/memset inlined on 3.3, 3.4, 4.0 and 4.1.  On ppc I get calls to
> memcpy/memset in all cases.  This might be more a glibc issue I think.

So my suggestion is to file a bugzilla PR about this problem including
_preprocessed_ source to get glibc dependencies out of the way.

Richard.

References:
- 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Nickolay Kolchin
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Richard Guenther
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Nickolay Kolchin
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Ernest L. Williams Jr.
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Richard Guenther
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Ernest L. Williams Jr.
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Richard Guenther
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Gabriel Dos Reis
- Re: 100x perfomance regression between gcc 3.4.5 and gcc 4.X
  - From: Richard Guenther

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]