[llvm] r337387 - [X86][SSE] Remove BLENDPD canonicalization from combineTargetShuffle

Simon Pilgrim via llvm-commits llvm-commits at lists.llvm.org
Wed Jul 18 06:01:20 PDT 2018


Author: rksimon
Date: Wed Jul 18 06:01:20 2018
New Revision: 337387

URL: http://llvm.org/viewvc/llvm-project?rev=337387&view=rev
Log:
[X86][SSE] Remove BLENDPD canonicalization from combineTargetShuffle

When rL336971 removed the scalar-fp isel patterns, we lost the need for this canonicalization - commutation/folding can handle everything else.

Modified:
    llvm/trunk/lib/Target/X86/X86ISelLowering.cpp

Modified: llvm/trunk/lib/Target/X86/X86ISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86ISelLowering.cpp?rev=337387&r1=337386&r2=337387&view=diff
==============================================================================
--- llvm/trunk/lib/Target/X86/X86ISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/X86/X86ISelLowering.cpp Wed Jul 18 06:01:20 2018
@@ -30696,31 +30696,6 @@ static SDValue combineTargetShuffle(SDVa
     }
     return SDValue();
   }
-  case X86ISD::BLENDI: {
-    SDValue V0 = N->getOperand(0);
-    SDValue V1 = N->getOperand(1);
-    assert(VT == V0.getSimpleValueType() && VT == V1.getSimpleValueType() &&
-           "Unexpected input vector types");
-
-    // Canonicalize a v2f64 blend with a mask of 2 by swapping the vector
-    // operands and changing the mask to 1. This saves us a bunch of
-    // pattern-matching possibilities related to scalar math ops in SSE/AVX.
-    // x86InstrInfo knows how to commute this back after instruction selection
-    // if it would help register allocation.
-
-    // TODO: If optimizing for size or a processor that doesn't suffer from
-    // partial register update stalls, this should be transformed into a MOVSD
-    // instruction because a MOVSD is 1-2 bytes smaller than a BLENDPD.
-
-    if (VT == MVT::v2f64)
-      if (auto *Mask = dyn_cast<ConstantSDNode>(N->getOperand(2)))
-        if (Mask->getZExtValue() == 2 && !isShuffleFoldableLoad(V0)) {
-          SDValue NewMask = DAG.getConstant(1, DL, MVT::i8);
-          return DAG.getNode(X86ISD::BLENDI, DL, VT, V1, V0, NewMask);
-        }
-
-    return SDValue();
-  }
   case X86ISD::MOVSD:
   case X86ISD::MOVSS: {
     SDValue V0 = peekThroughBitcasts(N->getOperand(0));




More information about the llvm-commits mailing list