[X86] Teach method 'isVectorClearMaskLegal' how to check for legal blend masks.
This patch improves the folding of vector AND nodes into blend operations for
targets that feature SSE4.1. A vector AND node where one of the operands is
a constant build_vector with elements that are either zero or all-ones can be
converted into a blend.
This allows for example to simplify the following code:
define <4 x i32> @test(<4 x i32> %A, <4 x i32> %B) {
%1 = and <4 x i32> %A, <i32 0, i32 0, i32 0, i32 -1>
%2 = and <4 x i32> %B, <i32 -1, i32 -1, i32 -1, i32 0>
%3 = or <4 x i32> %1, %2
ret <4 x i32> %3
}
Before this patch llc (-mcpu=corei7) generated:
andps LCPI1_0(%rip), %xmm0, %xmm0
andps LCPI1_1(%rip), %xmm1, %xmm1
orps %xmm1, %xmm0, %xmm0
retq
With this patch we generate a single 'vpblendw'.
llvm-svn: 221343
diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index 107f5a1..33b2527 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -11293,7 +11293,7 @@
if (cast<ConstantSDNode>(Elt)->isAllOnesValue())
Indices.push_back(i);
else if (cast<ConstantSDNode>(Elt)->isNullValue())
- Indices.push_back(NumElts);
+ Indices.push_back(NumElts+i);
else
return SDValue();
}
diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index aa77a3e..bde9481 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -19383,7 +19383,9 @@
return (isMOVLMask(Mask, SVT) ||
isCommutedMOVLMask(Mask, SVT, true) ||
isSHUFPMask(Mask, SVT) ||
- isSHUFPMask(Mask, SVT, /* Commuted */ true));
+ isSHUFPMask(Mask, SVT, /* Commuted */ true) ||
+ isBlendMask(Mask, SVT, Subtarget->hasSSE41(),
+ Subtarget->hasInt256()));
}
return false;
}