This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/52607] v4df __builtin_shuffle with {0,2,1,3} or {1,3,0,2}
- From: "marc.glisse at normalesup dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Sat, 17 Mar 2012 22:03:08 +0000
- Subject: [Bug target/52607] v4df __builtin_shuffle with {0,2,1,3} or {1,3,0,2}
- Auto-submitted: auto-generated
- References: <bug-52607-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52607
Marc Glisse <marc.glisse at normalesup dot org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #26908|0 |1
is obsolete| |
--- Comment #4 from Marc Glisse <marc.glisse at normalesup dot org> 2012-03-17 22:03:08 UTC ---
Created attachment 26909
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=26909
patch
Here is a try. Again, I just looked at the generated code on a couple examples,
which isn't very reliable...
expand_vec_perm_vperm2f128_vblend0 is already covered by
expand_vec_perm_vperm2f128_vblend1, but it is confusing to have a 3-instruction
function generate only 2.
I didn't do generic permutations with 4 instructions.
There is probably more that can be done with vshufp[sd].