This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] [asan] Fix for PR asan/56330


On Sat, Feb 16, 2013 at 09:44:51AM +0100, Dodji Seketeli wrote:
> Right.  My first bootstrap actually caught this, I updated the patch
> locally to just modify that test and sent out the wrong one.  Sorry for
> this.  This is the patch that actually passed bootstrap.

Ok.  Actually, please change gcc/testsuite/c-c++-common/asan/pr56330.c
to the same thing, I've just verified it ICEs the same way without the patch
even when it is
int
foo (void)
{
  int d = __builtin_memcmp (s.a, e, 100);
  d += __builtin_memcmp (s.a, e, 200);
  return d;
}

(while with
  __builtin_memcpy (s.a, e, 100);
  __builtin_memcpy (s.a, e, 200);
it doesn't ICE without the patch).
While pr56330.c test look to be redundant, it is at least tortured at all
compilation options, so it tests something that the other tests don't.

No need to rebootstrap/retest.

	Jakub


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]