http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52607
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |UNCONFIRMED
Ever Confirmed|1 |0
--- Comment #19 from Jakub Jelinek <jakub at gcc dot gnu.org> 2012-03-27
16:53:24 UTC ---
(In reply to comment #17)
> By the way, I have trouble understanding this comment:
> /* For d->op0 == d->op1 the only useful vperm2f128 permutation
> is 0x10. */
> Is it really 0x10, or is there a stray 0 at the end and it is really just 1?
Yeah, you are right, perm 1 (which actually is emitted as $33 immediate anyway,
as vperm2f128 insn is used with the same register operand twice.
