[SelectionDAG] Add support for vector demandedelts in XOR opcodes

llvm-svn: 286075

diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
index 430f39d..09f3a00 100644 (file)
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
@@ -2153,8 +2153,10 @@ void SelectionDAG::computeKnownBits(SDValue Op, APInt &KnownZero,
     KnownOne |= KnownOne2;
     break;
   case ISD::XOR: {
-    computeKnownBits(Op.getOperand(1), KnownZero, KnownOne, Depth+1);
-    computeKnownBits(Op.getOperand(0), KnownZero2, KnownOne2, Depth+1);
+    computeKnownBits(Op.getOperand(1), KnownZero, KnownOne, DemandedElts,
+                     Depth + 1);
+    computeKnownBits(Op.getOperand(0), KnownZero2, KnownOne2, DemandedElts,
+                     Depth + 1);
 
     // Output known-0 bits are known if clear or set in both the LHS & RHS.
     APInt KnownZeroOut = (KnownZero & KnownZero2) | (KnownOne & KnownOne2);

diff --git a/llvm/test/CodeGen/X86/known-bits-vector.ll b/llvm/test/CodeGen/X86/known-bits-vector.ll
index 0fd2899..0c21eb1 100644 (file)
--- a/llvm/test/CodeGen/X86/known-bits-vector.ll
+++ b/llvm/test/CodeGen/X86/known-bits-vector.ll
@@ -119,11 +119,7 @@ define <4 x float> @knownbits_mask_xor_shuffle_uitofp(<4 x i32> %a0) nounwind {
 ; X32-NEXT:    vpand {{\.LCPI.*}}, %xmm0, %xmm0
 ; X32-NEXT:    vpxor {{\.LCPI.*}}, %xmm0, %xmm0
 ; X32-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
-; X32-NEXT:    vpblendw {{.*#+}} xmm1 = xmm0[0],mem[1],xmm0[2],mem[3],xmm0[4],mem[5],xmm0[6],mem[7]
-; X32-NEXT:    vpsrld $16, %xmm0, %xmm0
-; X32-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],mem[1],xmm0[2],mem[3],xmm0[4],mem[5],xmm0[6],mem[7]
-; X32-NEXT:    vaddps {{\.LCPI.*}}, %xmm0, %xmm0
-; X32-NEXT:    vaddps %xmm0, %xmm1, %xmm0
+; X32-NEXT:    vcvtdq2ps %xmm0, %xmm0
 ; X32-NEXT:    retl
 ;
 ; X64-LABEL: knownbits_mask_xor_shuffle_uitofp:
@@ -131,11 +127,7 @@ define <4 x float> @knownbits_mask_xor_shuffle_uitofp(<4 x i32> %a0) nounwind {
 ; X64-NEXT:    vpand {{.*}}(%rip), %xmm0, %xmm0
 ; X64-NEXT:    vpxor {{.*}}(%rip), %xmm0, %xmm0
 ; X64-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
-; X64-NEXT:    vpblendw {{.*#+}} xmm1 = xmm0[0],mem[1],xmm0[2],mem[3],xmm0[4],mem[5],xmm0[6],mem[7]
-; X64-NEXT:    vpsrld $16, %xmm0, %xmm0
-; X64-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],mem[1],xmm0[2],mem[3],xmm0[4],mem[5],xmm0[6],mem[7]
-; X64-NEXT:    vaddps {{.*}}(%rip), %xmm0, %xmm0
-; X64-NEXT:    vaddps %xmm0, %xmm1, %xmm0
+; X64-NEXT:    vcvtdq2ps %xmm0, %xmm0
 ; X64-NEXT:    retq
   %1 = and <4 x i32> %a0, <i32 -1, i32 -1, i32 255, i32 4085>
   %2 = xor <4 x i32> %1, <i32 65535, i32 65535, i32 65535, i32 65535>