[PATCH][InstCombine][X86] Teach how to fold calls to SSE2/AVX2 psrl intrinsics.
bigcheesegs at gmail.com
Fri Apr 25 12:59:49 PDT 2014
On Fri, Apr 25, 2014 at 9:43 AM, Andrea Di Biagio
<andrea.dibiagio at gmail.com> wrote:
> this patch extends the fix committed at revision 20758
> (http://llvm.org/viewvc/llvm-project?view=revision&revision=207058) to
> also teach InstCombine how to fold calls to SSE2/AVX2 packed logical
> shift right intrinsics.
> A packed logical shift right with a shift count bigger than or equal
> to the element size always produces a zero vector. In all other cases,
> it can be safely replaced by a 'lshr' instruction.
> Please let me know if ok to submit.
> Andrea Di Biagio
> SN Systems - Sony Computer Entertainment Group.
- Michael Spencer
More information about the llvm-commits