On 12/05/2017 07:49 AM, Jakub Jelinek wrote:
On Mon, Dec 04, 2017 at 11:52:01AM +0100, Jakub Jelinek wrote:
On Wed, Nov 29, 2017 at 05:21:22PM -0500, Vladimir Makarov wrote:
The following patch fixes
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80818
The patch was successfully tested and bootstrapped on x86_64. The patch
has no test because it is hard to check the problem. I checked manually
This changed fixed PR83252 which has a reasonably small testcase.
I've further reduced it using creduce (with -O0 -W{,maybe-}uninitialized
and/or -fsanitize=undefined checking, plus test that it succeeds with
r255258 and fails with r255257). Can you please double check if the
testcase represents the same issue you were working on or if your change
merely made the bug latent again?
Unfortunately, it broke again with r255377. Vladimir, could you please have
a look at this PR?
Yes, I'll work on it.
The more I look at the problem discussed in PR80818, PR83252, and
PR83245, the more I think it is actually a LRA rematerialziation problem.
