From: Jan Hubicka <hubicka@ucw.cz>
To: Marc Espie <espie@nerim.net>
Cc: gcc@gcc.gnu.org
Subject: Re: gcc 3.5 integration branch proposal
Date: Mon, 19 Jan 2004 17:58:00 -0000 [thread overview]
Message-ID: <20040119175745.GA8495@atrey.karlin.mff.cuni.cz> (raw)
In-Reply-To: <20040119132847.GA16721@tetto.gentiane.org>
> On Sun, Jan 18, 2004 at 10:49:30PM -0500, David Edelsohn wrote:
>
> > Did you build head configured with --disable-checking?
>
> Rebuilt with --disable-checking, and double-checked the options. There
> was a bug that meant I was not -O2 in both cases.
>
> Now, the actual options are
> -Wall -Wstrict-prototypes -Wmissing-prototypes
> -Wno-uninitialized -Wno-format -Wno-main -fno-builtin-printf
> -fno-builtin-log -fno-builtin-malloc -O2 -fno-strict-aliasing
> in both cases.
>
> And it's still a bit slower:
>
> gcc 3.3.2:
> 527.98s real 421.02s user 30.36s system
>
> gcc -head with --disable-checking:
> 544.91s real 439.52s user 30.27s system
>
>
> Okay, the slow-down is not as marked as with --enable-checking, but it
> still does exist: 4% is something.
>
> Basically, the code there is just an OpenBSD kernel. It won't fly as a
> testcase unless you have an OpenBSD system available.
>
> It's just plain old C. Nothing really fancy about it.
>
> So, the compiler is not getting faster.
>
> To give you some scope, again. A similar bulk of code used to compile in
> 3 hours on a 68040. This 4% translates to an extra 7 minutes.
Since you apparently do have access to 68040 machine easilly, would be
possible for you to post some comparsion of the speed relative to gcc
2.95?
One of major changes of post 3.0 GCCs is the new x86 backend that has
many advantages for modern CPUs (support for scheduling, SSE, x86-64)
but is whole a lot more complex and memory houngry than the old has
been. I would be very curious how the numbers look like on the machine
where backend didn't changed much.
The profiles would be very welcome too!
I do have one 68040 myself, but it is 120Km away right now :)
Honza
>
> A make build on such a machine takes about two days. This 4% translates to
> 2 hours more.
>
> I'll try doing some profiling in the near future...
next prev parent reply other threads:[~2004-01-19 17:58 UTC|newest]
Thread overview: 185+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-01-10 0:11 Geoffrey Keating
2004-01-10 0:25 ` Phil Edwards
2004-01-10 0:47 ` Geoffrey Keating
2004-01-10 15:41 ` Phil Edwards
2004-01-12 18:47 ` Geoffrey Keating
2004-01-12 19:22 ` Mark Mitchell
2004-01-12 23:42 ` Geoff Keating
2004-01-12 23:48 ` Zack Weinberg
2004-01-12 23:49 ` Mark Mitchell
2004-01-13 0:06 ` Daniel Jacobowitz
2004-01-13 0:16 ` Daniel Jacobowitz
2004-01-13 0:22 ` Steven Bosscher
2004-01-13 19:48 ` Gerald Pfeifer
2004-01-13 0:28 ` Joe Buck
2004-01-13 0:46 ` Daniel Jacobowitz
2004-01-13 1:13 ` Jan Hubicka
2004-01-13 0:11 ` Ziemowit Laski
2004-01-13 0:15 ` Steven Bosscher
2004-01-13 0:23 ` Ziemowit Laski
2004-01-13 0:37 ` Steven Bosscher
2004-01-13 0:45 ` Jan Hubicka
2004-01-13 0:55 ` Ziemowit Laski
2004-01-13 0:54 ` Ziemowit Laski
2004-01-13 1:01 ` Mark Mitchell
2004-01-13 1:16 ` Ziemowit Laski
2004-01-13 1:26 ` Mark Mitchell
2004-01-13 1:33 ` Gabriel Dos Reis
2004-01-13 1:19 ` Gabriel Dos Reis
2004-01-13 0:47 ` Daniel Jacobowitz
2004-01-13 0:59 ` Altivec (Re: gcc 3.5 integration branch proposal) Matt Austern
2004-01-13 0:16 ` gcc 3.5 integration branch proposal Mark Mitchell
2004-01-13 0:54 ` Gabriel Dos Reis
2004-01-13 21:31 ` Geoff Keating
2004-01-13 22:41 ` David Edelsohn
2004-01-13 23:13 ` Daniel Berlin
2004-01-13 23:24 ` Paul Koning
2004-01-14 0:01 ` Geoff Keating
2004-01-14 0:26 ` Gabriel Dos Reis
2004-01-14 8:35 ` Laurent GUERBY
2004-01-14 14:58 ` Paul Koning
2004-01-13 23:29 ` Mike Stump
2004-01-14 0:47 ` Andrew Pinski
2004-01-14 1:05 ` Andrew Pinski
2004-01-14 1:36 ` Joe Buck
2004-01-14 19:47 ` Mike Stump
2004-01-13 23:51 ` Eric Christopher
2004-01-13 23:53 ` Daniel Berlin
2004-01-14 6:13 ` Mark Mitchell
2004-01-14 8:39 ` Laurent GUERBY
2004-01-14 8:48 ` Steven Bosscher
2004-01-14 20:59 ` Laurent GUERBY
2004-01-14 21:02 ` Andrew Haley
2004-01-19 1:31 ` Marc Espie
2004-01-19 1:51 ` Gabriel Dos Reis
2004-01-19 10:40 ` Nick Burrett
2004-01-19 13:55 ` Robert Dewar
2004-01-19 15:49 ` Gabriel Dos Reis
2004-01-19 16:09 ` Scott Robert Ladd
2004-01-19 16:24 ` Marc Espie
2004-01-19 17:07 ` Scott Robert Ladd
2004-01-19 17:54 ` Robert Dewar
2004-01-19 18:03 ` Gabriel Dos Reis
2004-01-19 17:41 ` Robert Dewar
2004-01-19 18:08 ` Scott Robert Ladd
2004-01-19 18:09 ` Scott Robert Ladd
2004-01-19 18:34 ` Marc Espie
2004-01-19 19:06 ` Robert Dewar
2004-01-19 20:22 ` Eric Botcazou
2004-01-19 20:38 ` Robert Dewar
2004-01-19 21:09 ` Eric Botcazou
2004-01-19 21:18 ` Eric Christopher
2004-01-19 21:46 ` Eric Botcazou
2004-01-19 22:09 ` Laurent GUERBY
2004-01-19 22:29 ` Marc Espie
2004-01-19 23:04 ` Laurent GUERBY
2004-01-20 3:45 ` Robert Dewar
2004-01-20 7:46 ` Marc Espie
2004-01-20 23:22 ` Gerald Pfeifer
2004-01-20 23:28 ` Zack Winkles
2004-01-19 23:53 ` Mark Hahn
2004-01-20 0:17 ` Gabriel Dos Reis
2004-01-20 14:27 ` Scott Robert Ladd
2004-01-22 14:50 ` Geoffrey Furnish
2004-01-20 6:27 ` gcc compilation speed Matt Austern
2004-01-20 7:23 ` Steven Bosscher
2004-01-20 7:40 ` Karel Gardas
2004-01-20 8:25 ` gcc 3.5 integration branch proposal Karel Gardas
2004-01-20 12:07 ` Laurent GUERBY
2004-01-20 14:41 ` Karel Gardas
2004-01-20 14:52 ` Daniel Berlin
2004-01-20 14:55 ` Karel Gardas
2004-01-20 15:16 ` Daniel Jacobowitz
2004-01-20 15:21 ` Karel Gardas
2004-01-20 16:16 ` Comparison of compilation speed of GCC 3.4.0 040114, ICC 8.0 and COMO 4.3.3 on MICO ORB core sources (C++) Karel Gardas
2004-01-20 15:08 ` gcc 3.5 integration branch proposal Paul Koning
2004-01-19 21:20 ` Robert Dewar
2004-01-20 2:05 ` Mike Stump
2004-01-20 14:25 ` Scott Robert Ladd
2004-01-19 21:34 ` Geoff Keating
2004-01-19 22:03 ` Eric Botcazou
2004-01-19 22:22 ` Geoff Keating
2004-01-19 22:38 ` Eric Botcazou
2004-01-20 3:53 ` Robert Dewar
2004-01-20 4:52 ` Eric Botcazou
2004-01-20 2:44 ` Alexandre Oliva
2004-01-20 3:04 ` Mike Stump
2004-01-20 19:03 ` Alexandre Oliva
2004-01-20 20:41 ` Mike Stump
2004-01-20 22:06 ` Alexandre Oliva
2004-01-20 22:59 ` Mike Stump
2004-01-20 18:52 ` Geoffrey Keating
2004-01-20 22:12 ` Alexandre Oliva
2004-01-20 22:30 ` Geoffrey Keating
2004-01-21 0:02 ` Alexandre Oliva
2004-01-21 12:12 ` Richard Earnshaw
2004-01-22 6:37 ` Segher Boessenkool
2004-01-22 10:38 ` Richard Earnshaw
2004-01-22 6:27 ` Segher Boessenkool
2004-01-26 18:55 ` compile speed [was: gcc 3.5 integration branch proposal] Per Bothner
2004-01-26 19:34 ` Alexandre Oliva
2004-01-20 0:01 ` GCC's hardware requirements - Where is the freedom? [was: " Matthias Benkmann
2004-01-20 3:13 ` Andrew Pinski
2004-01-20 3:17 ` Andrew Pinski
2004-01-20 9:23 ` Bernd Jendrissek
2004-01-20 21:38 ` Mike Stump
2004-01-20 22:50 ` Toon Moene
2004-01-20 8:12 ` Laurent GUERBY
2004-01-20 9:09 ` Eric Botcazou
2004-01-19 17:27 ` gcc 3.5 integration branch proposal Robert Dewar
2004-01-19 16:26 ` Per Bothner
2004-01-20 2:14 ` Mike Stump
2004-01-20 3:49 ` Zack Weinberg
2004-01-20 10:08 ` Jan Hubicka
2004-01-20 15:13 ` Daniel Jacobowitz
2004-01-20 18:04 ` Zack Weinberg
2004-01-20 18:48 ` Geoffrey Keating
2004-01-20 19:01 ` Zack Weinberg
2004-01-20 22:04 ` Geoff Keating
2004-01-21 2:23 ` Zack Weinberg
2004-01-22 6:27 ` Segher Boessenkool
2004-01-19 20:51 ` Dale Johannesen
2004-01-19 23:01 ` Gabriel Dos Reis
2004-01-19 3:42 ` Marc Espie
2004-01-19 3:47 ` Robert McNulty Junior
2004-01-19 3:49 ` David Edelsohn
2004-01-19 13:28 ` Marc Espie
2004-01-19 13:35 ` Jan Hubicka
2004-01-19 19:10 ` Marc Espie
2004-01-19 19:15 ` Marc Espie
2004-01-19 20:49 ` Diego Novillo
2004-01-19 19:39 ` Jan Hubicka
2004-01-19 13:35 ` Diego Novillo
2004-01-19 13:39 ` Jan Hubicka
2004-01-19 17:58 ` Jan Hubicka [this message]
2004-01-19 3:59 ` Zack Weinberg
2004-01-19 8:19 ` Steven Bosscher
2004-01-19 11:29 ` Jan Hubicka
2004-01-19 13:18 ` Jan Hubicka
2004-01-19 14:13 ` Robert Dewar
2004-01-19 14:18 ` Jan Hubicka
2004-01-19 14:48 ` Vladimir Makarov
2004-01-19 17:04 ` Jan Hubicka
2004-01-20 2:46 ` Alexandre Oliva
2004-01-19 4:37 ` Giovanni Bajo
2004-01-19 4:46 ` Gabriel Dos Reis
2004-01-19 7:04 ` Eric Botcazou
2004-01-19 7:10 ` Andreas Jaeger
2004-01-19 8:12 ` Steven Bosscher
2004-01-19 8:38 ` Gabriel Dos Reis
2004-01-19 8:45 ` Steven Bosscher
2004-01-19 9:35 ` Gabriel Dos Reis
2004-01-12 22:40 ` Mike Stump
2004-01-13 23:31 Michael Elizabeth Chastain
2004-01-19 16:24 Paul Koning
2004-01-19 17:46 ` Robert Dewar
2004-01-19 17:51 ` Gabriel Dos Reis
2004-01-19 18:12 ` Scott Robert Ladd
2004-01-19 18:20 ` Gabriel Dos Reis
2004-01-19 18:29 ` Giovanni Bajo
2004-01-19 18:22 ` Robert Dewar
2004-01-19 18:30 ` Paul Koning
2004-01-20 10:46 ` Russ Allbery
2004-01-20 3:00 D. Starner
2004-01-20 3:14 ` Andrew Pinski
2004-01-20 5:19 D. Starner
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=20040119175745.GA8495@atrey.karlin.mff.cuni.cz \
--to=hubicka@ucw.cz \
--cc=espie@nerim.net \
--cc=gcc@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).