[X86] Generalize logic blend of (x, -x) combine to match (-x, x).

I suspect this is what let PR26110 lie dormant for so long.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@261024 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 994c493..4a87779 100644 (file)
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -26470,13 +26470,23 @@ static SDValue combineLogicBlendIntoPBLENDV(SDNode *N, SelectionDAG &DAG,
   //   (add (xor X, M), (and M, 1))
   // And further to:
   //   (sub (xor X, M), M)
-  if (Y.getOpcode() == ISD::SUB && Y.getOperand(1) == X &&
-      ISD::isBuildVectorAllZeros(Y.getOperand(0).getNode()) &&
-      X.getValueType() == MaskVT && Y.getValueType() == MaskVT) {
-    assert(EltBits == 8 || EltBits == 16 || EltBits == 32);
-    return DAG.getBitcast(
-        VT, DAG.getNode(ISD::SUB, DL, MaskVT,
-                        DAG.getNode(ISD::XOR, DL, MaskVT, X, Mask), Mask));
+  if (X.getValueType() == MaskVT && Y.getValueType() == MaskVT) {
+    auto IsNegV = [](SDNode *N, SDValue V) {
+      return N->getOpcode() == ISD::SUB && N->getOperand(1) == V &&
+        ISD::isBuildVectorAllZeros(N->getOperand(0).getNode());
+    };
+    SDValue V;
+    if (IsNegV(Y.getNode(), X))
+      V = X;
+    else if (IsNegV(X.getNode(), Y))
+      V = Y;
+
+    if (V) {
+      assert(EltBits == 8 || EltBits == 16 || EltBits == 32);
+      return DAG.getBitcast(
+          VT, DAG.getNode(ISD::SUB, DL, MaskVT,
+                          DAG.getNode(ISD::XOR, DL, MaskVT, V, Mask), Mask));
+    }
   }
 
   // PBLENDVB is only available on SSE 4.1.

diff --git a/test/CodeGen/X86/vector-blend.ll b/test/CodeGen/X86/vector-blend.ll
index 9cb4a3b..6b96f66 100644 (file)
--- a/test/CodeGen/X86/vector-blend.ll
+++ b/test/CodeGen/X86/vector-blend.ll
@@ -1011,11 +1011,8 @@ define <4 x i32> @blend_neg_logic_v4i32_2(<4 x i32> %v, <4 x i32> %c) {
 ; SSE2-NEXT:    psrld $31, %xmm1
 ; SSE2-NEXT:    pslld $31, %xmm1
 ; SSE2-NEXT:    psrad $31, %xmm1
-; SSE2-NEXT:    pxor %xmm2, %xmm2
-; SSE2-NEXT:    psubd %xmm0, %xmm2
-; SSE2-NEXT:    pand %xmm1, %xmm0
-; SSE2-NEXT:    pandn %xmm2, %xmm1
-; SSE2-NEXT:    por %xmm1, %xmm0
+; SSE2-NEXT:    pxor %xmm1, %xmm0
+; SSE2-NEXT:    psubd %xmm1, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: blend_neg_logic_v4i32_2:
@@ -1023,11 +1020,8 @@ define <4 x i32> @blend_neg_logic_v4i32_2(<4 x i32> %v, <4 x i32> %c) {
 ; SSSE3-NEXT:    psrld $31, %xmm1
 ; SSSE3-NEXT:    pslld $31, %xmm1
 ; SSSE3-NEXT:    psrad $31, %xmm1
-; SSSE3-NEXT:    pxor %xmm2, %xmm2
-; SSSE3-NEXT:    psubd %xmm0, %xmm2
-; SSSE3-NEXT:    pand %xmm1, %xmm0
-; SSSE3-NEXT:    pandn %xmm2, %xmm1
-; SSSE3-NEXT:    por %xmm1, %xmm0
+; SSSE3-NEXT:    pxor %xmm1, %xmm0
+; SSSE3-NEXT:    psubd %xmm1, %xmm0
 ; SSSE3-NEXT:    retq
 ;
 ; SSE41-LABEL: blend_neg_logic_v4i32_2: