[X86] Remove the max vector width restriction from combineLoopMAddPattern and rely splitOpsAndApply to handle splitting.

This seems to be a net improvement. There's still an issue under avx512f where we have a 512-bit vpaddd, but not vpmaddwd so we end up doing two 256-bit vpmaddwds and inserting the results before a 512-bit vpaddd. It might be better to do two 512-bits paddds with zeros in the upper half. Same number of instructions, but breaks a dependency.

llvm-svn: 337656

1 files changed, 1 insertions, 7 deletions

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index f17e4bfbf3b..ed34aeaccd8 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -38559,15 +38559,9 @@ static SDValue combineLoopMAddPattern(SDNode *N, SelectionDAG &DAG,
 
   EVT VT = N->getValueType(0);
 
-  unsigned RegSize = 128;
-  if (Subtarget.useBWIRegs())
-    RegSize = 512;
-  else if (Subtarget.hasAVX())
-    RegSize = 256;
-  unsigned VectorSize = VT.getVectorNumElements() * 16;
   // If the vector size is less than 128, or greater than the supported RegSize,
   // do not use PMADD.
-  if (VectorSize < 128 || VectorSize > RegSize)
+  if (VT.getVectorNumElements() < 8)
     return SDValue();
 
   SDLoc DL(N);