This is the mail archive of the
mailing list for the GCC project.
Re: gcc compile-time performance
- From: Stan Shebs <shebs at apple dot com>
- To: Robert Dewar <dewar at gnat dot com>
- Cc: gdr at codesourcery dot com, mark at codesourcery dot com, gcc at gcc dot gnu dot org, scott at coyotegulch dot com
- Date: Sat, 18 May 2002 10:36:07 -0700
- Subject: Re: gcc compile-time performance
- References: <20020518171850.8973EF2A59@nile.gnat.com>
Robert Dewar wrote:
> > In order to support those uses, and to be competitive with other
> > compilers, we do need to improve the speed of the compiler. I've heard
> > this from too many real world customers with spending authority to
> > believe any different.
> The question is can you find ways to make substantial increases in speed.
> I still don't see it being worthwhile to flail away looking at profiles and
> tingering with conditionals. You are only going to get 10-30% that way
> and it is not going to be enough to be significant.
What's your basis for this claim, anyway? We have a local patch
that all by itself achieves a 20% speedup - in special cases that
happen to be common in Mac programming, and wouldn't really affect
anybody else. (Yes yes, we should contribute it even so.)
Precompiled headers are known to be a huge win in another set of
common cases, which is why we have a mechanism and why Red Hat
It's going to take some pretty strong evidence to convince me that
GCC can't be sped up significantly; I refuse to believe that the
GCC developers aren't as capable as those at Metrowerks et al.
If you want to point at specific pieces of code, well then, those
are the right place to start work!