This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix tree-ssa/loop-14.c test case failure, take 2
- From: Ulrich Weigand <uweigand at de dot ibm dot com>
- To: rakdver at atrey dot karlin dot mff dot cuni dot cz (Zdenek Dvorak)
- Cc: uweigand at de dot ibm dot com (Ulrich Weigand), gcc-patches at gcc dot gnu dot org
- Date: Thu, 19 Jan 2006 18:54:45 +0100 (CET)
- Subject: Re: [PATCH] Fix tree-ssa/loop-14.c test case failure, take 2
Zdenek Dvorak wrote:
> > I can bootstrap on i386, but I'm not really set up to do actual
> > benchmarks on non-s390 platforms ...
>
> OK, I will test the patch for you on x86_64 and ia64.
Thank you very much for that!
> > In this instance, scev_const_prop tries to compute the cost of the
> > expression 'j + 100' (not just the constant 100).
>
> in mainline, the code got changed recently, so now it should test only
> 100. I think the patch is reasonable (although I cannot approve it, of
> course).
Ah, that would explain why the test case stopped failing in mainline
a couple of days ago ...
Do you think it would be less risk for 4.1 to backport this part of
your recent patch only, instead of changing the cost computation?
Bye,
Ulrich
--
Dr. Ulrich Weigand
Linux on zSeries Development
Ulrich.Weigand@de.ibm.com