PR 8146

[Joe Buck]

> > > PR opt/7591 (which has nothing to do with optimization btw) is a
> > > regression from gcc 3.0.x, so should really be marked as high-priority
> > > and taken care of on the branch. It's a i686 specific problem (doesn't
> > > occur on i586).

I wrote:
> > Is this the same bug as c/5351?  I haven't checked, but they seem very
> > similar.  c/5351 is also marked high priority.

HJ wrote:

> I got
> 
> # /usr/gcc-3.2/bin/gcc -march=i686 s.c
> # ./a.out
> ifName: this is a test string
> # /usr/gcc-3.2/bin/gcc s.c
> # ./a.out
> ifName: this is a test string

OK, this seems promising (and suggests that 5351 and 7591 are dups).

You referred to two patches, but we need to get them approved one at a
time.  Can you point me to the one patch that you believe fixes this
problem?

I think that RTH possibly rejected the patch as causing no change because
he was looking at the trunk; c/5351 does not occur on the trunk.  Still,
your patch might be the best way of fixing it on the branch.  If I can
verify it, I'll ask RTH and/or Mark to reconsider.