[llvm] r313495 - [X86] Synchronize a pattern between SSE1 and AVX/AVX512.

[Craig Topper via llvm-commits]

Author: ctopper
Date: Sun Sep 17 11:59:32 2017
New Revision: 313495

URL: http://llvm.org/viewvc/llvm-project?rev=313495&view=rev
Log:
[X86] Synchronize a pattern between SSE1 and AVX/AVX512.

For some reason the SSE1 pattern expected a X86Movlhps pattern to have a v4f32 type, but AVX and AVX512 expected it to have a v4i32 type.

I'm not even sure this pattern is even reachable post SSE1, but I'm starting with fixing this obvious bug.

Modified:
    llvm/trunk/lib/Target/X86/X86InstrAVX512.td
    llvm/trunk/lib/Target/X86/X86InstrSSE.td

Modified: llvm/trunk/lib/Target/X86/X86InstrAVX512.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86InstrAVX512.td?rev=313495&r1=313494&r2=313495&view=diff
==============================================================================
--- llvm/trunk/lib/Target/X86/X86InstrAVX512.td (original)
+++ llvm/trunk/lib/Target/X86/X86InstrAVX512.td Sun Sep 17 11:59:32 2017
@@ -6088,7 +6088,7 @@ let Predicates = [HasAVX512] in {
                (bc_v4f32 (v2i64 (scalar_to_vector (loadi64 addr:$src2))))),
           (VMOVHPSZ128rm VR128X:$src1, addr:$src2)>;
   def : Pat<(X86Movlhps VR128X:$src1,
-               (bc_v4i32 (v2i64 (X86vzload addr:$src2)))),
+               (bc_v4f32 (v2i64 (X86vzload addr:$src2)))),
           (VMOVHPSZ128rm VR128X:$src1, addr:$src2)>;
   // VMOVHPD patterns
   def : Pat<(v2f64 (X86Unpckl VR128X:$src1,

Modified: llvm/trunk/lib/Target/X86/X86InstrSSE.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86InstrSSE.td?rev=313495&r1=313494&r2=313495&view=diff
==============================================================================
--- llvm/trunk/lib/Target/X86/X86InstrSSE.td (original)
+++ llvm/trunk/lib/Target/X86/X86InstrSSE.td Sun Sep 17 11:59:32 2017
@@ -1034,7 +1034,7 @@ let Predicates = [UseAVX] in {
                  (bc_v4f32 (v2i64 (scalar_to_vector (loadi64 addr:$src2))))),
             (VMOVHPSrm VR128:$src1, addr:$src2)>;
   def : Pat<(X86Movlhps VR128:$src1,
-                 (bc_v4i32 (v2i64 (X86vzload addr:$src2)))),
+                 (bc_v4f32 (v2i64 (X86vzload addr:$src2)))),
             (VMOVHPSrm VR128:$src1, addr:$src2)>;
 
   // FIXME: Instead of X86Unpckl, there should be a X86Movlhpd here, the problem