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: [RFA] [PATCH] Make VRP understand bitwise OR and AND (better than now)


On Wed, 2008-08-13 at 16:15 -0400, Jakub Jelinek wrote:
> On Tue, Aug 12, 2008 at 02:47:28PM +0200, Denys Vlasenko wrote:
> > On Tue, 2008-08-12 at 04:20 -0400, Jakub Jelinek wrote:
> > > On Tue, Aug 12, 2008 at 10:09:13AM +0200, Richard Guenther wrote:
> > > > > Please find it attached. I did run testsuite, twice, once with
> > > > > deliberately broken test and one with test as it is in the patch.
> > > > > (Boy, that was SLOW).
> > > > 
> > > > The patch is ok for mainline if it passed a bootstrap as well.
> > 
> > It passed bootstrap on my machine.
> 
> It doesn't bootstrap on x86_64-linux.

Mine is also x86-64. Might be triggered by different configute and
CFLAGS. Can you post options you used?

Meanwhile I am retesting bootstrap with different options.

> ../../gcc/reload1.c: In function `reload':
> ../../gcc/reload1.c:702: internal compiler error: in set_value_range, at tree-vrp.c:392
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <http://gcc.gnu.org/bugs.html> for instructions.
> make[3]: *** [reload1.o] Error 1

I was seeing these things until I added the code which avoids doing this
optimization when msb bit is set.

> This needs to be fixed before this is committed.

No question about that.
--
vda



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