This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: New loop unroller broken?
- From: David Edelsohn <dje at watson dot ibm dot com>
- To: Andreas Jaeger <aj at suse dot de>
- Cc: Mircea Namolaru <NAMOLARU at il dot ibm dot com>, Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>, Dale Johannesen <dalej at apple dot com>, gcc at gcc dot gnu dot org, Andrew Pinski <pinskia at physics dot uc dot edu>, Ulrich Weigand <weigand at i1 dot informatik dot uni-erlangen dot de>
- Date: Fri, 30 Jan 2004 01:10:23 -0500
- Subject: Re: New loop unroller broken?
- References: <OFFC3CC7AD.F3A9B29E-ON42256E2A.003A892D-42256E2A.004159B0@il.ibm.com> <m3llnq9eks.fsf@gromit.moeb>
>>>>> Andreas Jaeger writes:
> OK to submit for GCC 3.4, after bootstraping and regtesting etc.?
Andreas> This needs to go into GCC mainline as requirement to go into the 3.4
Andreas> release branch (or is already in mainline?). So, please test on both
Andreas> branches and let's make this question:
Andreas> Ok to commit to mainline and 3.4 release branch, after...
Mircea is trying to ask whether something like the proposed patch
would be acceptable for GCC 3.4 to fix the doloop regression. The patch
is part of a larger loop performance patch for the trunk that will be
submitted in stages, building on this patch. Instead of submitting a
large patch and then discussing how to extract a subset to fix GCC 3.4, he
is starting with just the regression fix.
David