This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/62128] Use vpalignr for AVX2 rotation
- From: "jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Fri, 03 Oct 2014 07:28:22 +0000
- Subject: [Bug target/62128] Use vpalignr for AVX2 rotation
- Auto-submitted: auto-generated
- References: <bug-62128-4 at http dot gcc dot gnu dot org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62128
--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Fri Oct 3 07:27:50 2014
New Revision: 215834
URL: https://gcc.gnu.org/viewcvs?rev=215834&root=gcc&view=rev
Log:
PR target/62128
* config/i386/i386.c (expand_vec_perm_palignr): If op1, op0 order
of palignr arguments can't be used due to min 0 or max - min
too high, try also op0, op1 order of palignr arguments.
* gcc.dg/torture/vshuf-16.inc (TESTS): Add 2 new permutations.
* gcc.dg/torture/vshuf-32.inc (TESTS): Add 5 new permutations.
Modified:
trunk/gcc/ChangeLog
trunk/gcc/config/i386/i386.c
trunk/gcc/testsuite/ChangeLog
trunk/gcc/testsuite/gcc.dg/torture/vshuf-16.inc
trunk/gcc/testsuite/gcc.dg/torture/vshuf-32.inc