This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Optimize stdarg functions with void * or char * va_list
- From: Roger Sayle <roger at eyesopen dot com>
- To: Richard Guenther <richard dot guenther at gmail dot com>
- Cc: gcc-patches at gcc dot gnu dot org, <rth at redhat dot com>, <jakub at redhat dot com>
- Date: Thu, 30 Sep 2004 20:25:14 -0600 (MDT)
- Subject: Re: [PATCH] Optimize stdarg functions with void * or char * va_list
On Thu, 30 Sep 2004, Richard Guenther wrote:
> I'm confused. Aren't we in stage3? If it is a regression fix, what's
> its PR? I think there is too much questionable (not clearly stage3)
> material going in right now.
It's the gcc 3.3 and gcc 3.4 release branches that are currently
restricted to regression fixes only. My understanding is that
mainline stage3 additionally allows non-regression bug-fixes and
As for Jakub's "Optimize stdarg functions" patch, this is fixing a
valid bug that many targets save and restore too many registers in
their prologues/epilogues, a.k.a. "pessimizes code" and "missed
optimization". A bug is still a bug even if it doesn't have a
bugzilla PR number, and IMHO RTH's approval fits within the
definition of stage3 in http://gcc.gnu.org/develop.html, even
without his discretionary powers as a global write maintainer or
Mark's decision to allow "reasonable" clean-ups in 4.0's stage3.
I won't argue that there may be questionable material being committed
in stage3, but this particular case is probably not a good example.