From: Evan Cheng <evan.cheng@apple.com>
To: Duncan Sands <duncan.sands@math.u-psud.fr>
Cc: gcc@gcc.gnu.org, Vladimir Makarov <vmakarov@redhat.com>,
Chris Lattner <clattner@apple.com>
Subject: Re: New GCC releases comparison and comparison of GCC4.4 and LLVM2.5 on SPEC2000
Date: Wed, 13 May 2009 20:06:00 -0000 [thread overview]
Message-ID: <2B4EE58F-1701-42EF-B763-6F64A6DBF58E@apple.com> (raw)
In-Reply-To: <200905131351.00961.duncan.sands@math.u-psud.fr>
On May 13, 2009, at 4:51 AM, Duncan Sands wrote:
> Hi,
>
>> Sorry, I missed to mention that I used an additional option -mpc64
>> for
>> 32-bit GCC4.4. It is not possible to generate SPECFP2000 expected
>> results by GCC4.4 without this option. LLVM does not support this
>> option. And this option can significantly improve the
>> performance. So
>> 32-bit comparison of SPECFP2000 should be taken with a grain of salt.
>
> what does -mpc64 do exactly? The gcc docs say:
> `-mpc64' rounds the the significands of results of floating-point
> operations to 53 bits (double precision)
> Does this mean that a rounding operation is performed after each fp
> operation, or that optimizations are permitted that don't result in
> accurate extended double precision values as long as they are correct
> to 53 bits, or something else?
>
> The LLVM code generators have an option called -limit-float-precision:
> -limit-float-precision=<uint> - Generate low-precision inline
> sequences for some float libcalls
> I'm not sure what it does exactly, but perhaps it is similar to -
> mpc64?
No, that inline a small set of libcalls into sequences code that
implement low precision math (6, 8, 12 bits).
Evan
>
> Ciao,
>
> Duncan.
next prev parent reply other threads:[~2009-05-13 18:11 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-05-12 16:27 Vladimir Makarov
2009-05-12 18:05 ` Chris Lattner
2009-05-12 18:21 ` Vladimir Makarov
2009-05-12 19:25 ` Chris Lattner
2009-05-12 22:48 ` Rafael Espindola
2009-05-12 19:41 ` Vladimir Makarov
2009-05-13 12:11 ` Duncan Sands
2009-05-13 12:38 ` Richard Guenther
2009-05-13 12:55 ` Joseph S. Myers
2009-05-13 13:51 ` Duncan Sands
2009-05-26 12:27 ` Chris Lattner
2009-05-26 13:48 ` Vincent Lefevre
2009-05-28 23:00 ` Joseph S. Myers
2009-05-13 20:06 ` Evan Cheng [this message]
2009-05-12 18:29 ` Joseph S. Myers
2009-05-12 18:42 ` Vladimir Makarov
2009-05-13 8:44 ` Andi Kleen
2009-05-13 9:56 ` Jakub Jelinek
2009-05-13 11:32 ` Paolo Bonzini
2009-05-13 11:35 ` Paolo Bonzini
2009-05-13 12:07 ` Andi Kleen
2009-05-13 12:23 ` Steven Bosscher
2009-05-13 12:27 ` Dave Korn
2009-05-13 15:14 ` Paolo Bonzini
2009-05-13 16:16 ` Vladimir Makarov
2009-05-13 17:59 ` Jan Hubicka
2009-05-13 18:11 ` Michael Meissner
2009-05-15 20:19 ` Toon Moene
2009-05-13 6:42 ` Steven Bosscher
2009-05-13 7:33 ` Paolo Bonzini
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=2B4EE58F-1701-42EF-B763-6F64A6DBF58E@apple.com \
--to=evan.cheng@apple.com \
--cc=clattner@apple.com \
--cc=duncan.sands@math.u-psud.fr \
--cc=gcc@gcc.gnu.org \
--cc=vmakarov@redhat.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).