This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [Patch] C++ inlining heuristics changed (3.0.1)
- To: manfred dot h at gmx dot net, manfredh at redhat dot com
- Subject: Re: [Patch] C++ inlining heuristics changed (3.0.1)
- From: Kurt Garloff <garloff at suse dot de>
- Date: Thu, 23 Aug 2001 21:43:29 +0200
- Cc: Gerald Pfeifer <pfeifer at dbai dot tuwien dot ac dot at>, gcc at gcc dot gnu dot org, Daniel Berlin <dan at cgsoftware dot com>, gcc-patches at gcc dot gnu dot org, Daniel Morris <danielm at redhat dot com>
- Organization: TU/e(NL), SuSE(DE)
- References: <Pine.BSF.4.33.0108231443440.34493-200000@taygeta.dbai.tuwien.ac.at> <3B855754.4010708@redhat.com>
On Thu, Aug 23, 2001 at 09:19:48PM +0200, Manfred Hollstein wrote:
> Same here; I just re-built gcc-3.0.1 with all languages enabled,
> and I'm amazed! Great work Kurt! The final RPM I built is not only
> smaller than the first one (saved ~3%),
:-)
> but, random tests using the
> new compiler showed improved performance of the generated code by
> up to ~30-40% (and this was plain C code)!
Now, I have to tell you my disbelief!
My 3.0.1 patch (as opposed to the one against 2.95.3) only changes the C++
tree inliner (gcc/cp/optimize.c), you either have some C++ code or you use
at least some C++ libraries ...
Or do I miss the fact that C code now also uses the C++ tree inliner?
> One suggestion though; your patch is changing gcc.info-4, which is
> a generated file. You should be re-generating your patch with changes
> to the original .texi* files instead.
Oops. That was not really clever.
Will submit a new patch with this fixed and some other modifications soon.
Regards,
--
Kurt Garloff <kurt@garloff.de> [Eindhoven, NL]
Physics: Plasma simulations <K.Garloff@Phys.TUE.NL> [TU Eindhoven, NL]
Linux: SCSI, Security <garloff@suse.de> [SuSE Nuernberg, DE]
(See mail header or public key servers for PGP2 and GPG public keys.)
PGP signature