This is the mail archive of the
mailing list for the GCC project.
Re: PATCH: [darwin] Always turn on FP_SAVE_INLINE at -O3
- From: Stan Shebs <shebs at apple dot com>
- To: Caroline Tice <ctice at apple dot com>
- Cc: "gcc-patches at gcc dot gnu dot org Patches" <gcc-patches at gcc dot gnu dot org>, geoffrey Keating <geoffk at apple dot com>
- Date: Wed, 05 May 2004 15:47:24 -0700
- Subject: Re: PATCH: [darwin] Always turn on FP_SAVE_INLINE at -O3
- References: <B1CCD8A7-9EE0-11D8-8483-000393BB90B6@apple.com>
Caroline Tice wrote:
The following is a small optimization in the Apple compiler we would
like to put into mainline.
It is Darwin-specific. By always making FP_SAVE_INLINE be true at -O3
or higher we
gain a small performance improvement.
This has been tested on an Apple G4 running apple-darwin. It
bootstraps and passes
DejaGnu. Is this okay to commit to mainline?
This is a little unusual, a quick scan of targets doesn't show any
for which -O3 has different behavior (although there are several
that test -O1 vs -O2), but I can't think of any reason why not, so
yes, let's give it a try.
In the interest of propagating rumors via this mailing list :-), what
is the general order of magnitude of "small improvement" here?