This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/48702] [4.6/4.7 Regression] optimization regression with gcc-4.6 on x86_64-unknown-linux-gnu
- From: "rguenth at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 25 May 2011 13:58:51 +0000
- Subject: [Bug tree-optimization/48702] [4.6/4.7 Regression] optimization regression with gcc-4.6 on x86_64-unknown-linux-gnu
- Auto-submitted: auto-generated
- References: <bug-48702-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48702
--- Comment #24 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-05-25 13:58:20 UTC ---
(In reply to comment #22)
> Created attachment 24354 [details]
> alternative patch for 4.6
>
> This is another patch - it avoids creating MEM_REFs for base addresses that
> we do not know whether they are always in bounds of the object it points to.
> Instead we force the use of TARGET_MEM_REFs there and dumb down the alias
> oracle to not assume that base address is within bounds.
>
> Sounds like the most reasonable solution for 4.6 to me.
Requires staging on trunk, there are multiple small fixes that need backporting
and one last cleanup (and fix) that I had pending needs to be finished first.
But I'll continue in this direction.