[llvm] r328453 - [X86][MMX] MOVQ2DQ/MOVDQ2Q are better described as WriteVecMove than WriteMove

[Simon Pilgrim via llvm-commits]

Author: rksimon
Date: Sun Mar 25 10:28:06 2018
New Revision: 328453

URL: http://llvm.org/viewvc/llvm-project?rev=328453&view=rev
Log:
[X86][MMX] MOVQ2DQ/MOVDQ2Q are better described as WriteVecMove than WriteMove

Not that it makes a difference to current cost values, but will when we try to better model GPR-SIMD transfer costs

Modified:
    llvm/trunk/lib/Target/X86/X86InstrMMX.td

Modified: llvm/trunk/lib/Target/X86/X86InstrMMX.td
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86InstrMMX.td?rev=328453&r1=328452&r2=328453&view=diff
==============================================================================
--- llvm/trunk/lib/Target/X86/X86InstrMMX.td (original)
+++ llvm/trunk/lib/Target/X86/X86InstrMMX.td Sun Mar 25 10:28:06 2018
@@ -317,7 +317,7 @@ def MMX_MOVQ64mr : MMXI<0x7F, MRMDestMem
                         [(store (x86mmx VR64:$src), addr:$dst)],
                         IIC_MMX_MOVQ_RM>;
 
-let SchedRW = [WriteMove] in {
+let SchedRW = [WriteVecMove] in {
 def MMX_MOVDQ2Qrr : MMXSDIi8<0xD6, MRMSrcReg, (outs VR64:$dst),
                              (ins VR128:$src), "movdq2q\t{$src, $dst|$dst, $src}",
                              [(set VR64:$dst,