PR 11319 analyzed (Re: Patch fixing 3.3 bug PR 9745 and PR 10021)

[Mark Mitchell]

On Tue, 2003-07-15 at 15:29, Jim Wilson wrote:
> On Tue, 2003-07-15 at 13:31, Mark Mitchell wrote:
> > I thought that Jim's patch might have already fixed this kind of
> > issue.   The problem here is that the aliasing code thinks that because
> > there is a different "base register" for the two MEMs there must be no
> > conflict.
> 
> There are multiple problems and patches here.
> 
> My first patch was an alias.c patch which I think is conservatively
> correct.  David Edelsohn objected to the performance regressions caused
> by it, but did not provide any specifics.  This patch is in limbo until
> I have more time to work on this problem.

Please go ahead and install your patch on the 3.3 branch.

I see no choice but to take the performance hit.  To ship a compiler
that frequently and silently miscompiles real code is just not a good
idea.

I hope that you will continue to work with everyone to try to find ways
to mitigate that performance hit.  You suggested some ideas that might
mitigate the hit in your email, and it would be great if you could
follow up on those ideas.

I would recommend that we also apply the patch on the mainline, but I
think we should try to reach consensus with David about that.

Thank you for working hard on this issue.

-- 
Mark Mitchell <mark@codesourcery.com>
CodeSourcery, LLC