Re: [PATCH][4.1] Fix PRs 27795, 27639, 26719 - wrong code with scev/loop

[Richard Guenther <rguenther at suse dot de>]

On Mon, 24 Jul 2006, Richard Guenther wrote:

> On Mon, 24 Jul 2006, Zdenek Dvorak wrote:
> 
> > Hello,
> > 
> > > > >> This is a backport of a patch from mainline that has been sitting there
> > > > >> since May 24.
> > > > >>
> > > > >> Bootstrapped and tested on x86_64-unknown-linux-gnu for all langs + ada.
> > > > > 
> > > > > if this patch is approved for 4.1, http://gcc.gnu.org/ml/gcc-patches/2006-07/msg00229.html
> > > > > should probably go there as well.
> > > > 
> > > > In general, it is fine to backport patches from mainline to the branch
> > > > to fix regressions.  I don't need to sign off on that; if it's already
> > > > in mainline, and it fixes the PR, and you're confident in the backport
> > > > it's OK.  Please do take care, though; if you feel a backport might be
> > > > risky, please ask for a review (from me, or any other qualified reviewer).
> > > 
> > > I have applied the original patch to the branch - does the other patch
> > > fix a regression?  It didn't come with a PR number or a testcase.
> > 
> > Eric said that it is for PR 27937; I cannot verify that it really fixes
> > that PR, though, as there is no testcase available there.
> 
> Ok, I'll bootstrap and test the patch on the 4.1 branch.

Bootstrapped and tested on x86_64-unknown-linux-gnu.  I'll apply to the
branch in a sec.

Richard.

--
Richard Guenther <rguenther@suse.de>
Novell / SUSE Labs