This is the mail archive of the
mailing list for the GCC project.
Re: 200.sixtrack is miscomparing on x86_64 and i686 (-O2) since Mar 19th
After discussing with David on IRC, it appears that my mail could be
misinterpreted, so I'll try to make it clear:
Thanks for the note. As far as I know, none of the gfortran developers
have access to SPEC (it's google who told me it's from SPEC,
actually), so we'll have to wait until someone reduces this bug.
This is not an excuse for not fixing the regression, this is a call
for help. We don't have any way to find and fix this regression.
Independently of the patch being reversed or not, it would be much
appreciated if someone could help us understand what is going wrong