[DAGCombiner] Added CTTZ vector constant folding support.

llvm-svn: 239293

4 files changed, 192 insertions, 2 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index 307bfa502b7..4b9167723f4 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -4789,7 +4789,7 @@ SDValue DAGCombiner::visitCTTZ(SDNode *N) {
   EVT VT = N->getValueType(0);
 
   // fold (cttz c1) -> c2
-  if (isa<ConstantSDNode>(N0))
+  if (isConstantIntBuildVectorOrConstantInt(N0))
     return DAG.getNode(ISD::CTTZ, SDLoc(N), VT, N0);
   return SDValue();
 }
@@ -4799,7 +4799,7 @@ SDValue DAGCombiner::visitCTTZ_ZERO_UNDEF(SDNode *N) {
   EVT VT = N->getValueType(0);
 
   // fold (cttz_zero_undef c1) -> c2
-  if (isa<ConstantSDNode>(N0))
+  if (isConstantIntBuildVectorOrConstantInt(N0))
     return DAG.getNode(ISD::CTTZ_ZERO_UNDEF, SDLoc(N), VT, N0);
   return SDValue();
 }
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
index f72bfc6ceaa..92b6a0029b4 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
@@ -2911,6 +2911,8 @@ SDValue SelectionDAG::getNode(unsigned Opcode, SDLoc DL,
       case ISD::TRUNCATE:
       case ISD::UINT_TO_FP:
       case ISD::SINT_TO_FP:
+      case ISD::CTTZ:
+      case ISD::CTTZ_ZERO_UNDEF:
       case ISD::CTPOP: {
         EVT SVT = VT.getScalarType();
         EVT InVT = BV->getValueType(0);
diff --git a/llvm/test/CodeGen/X86/vector-tzcnt-128.ll b/llvm/test/CodeGen/X86/vector-tzcnt-128.ll
index bfa398b011c..422fe052d38 100644
--- a/llvm/test/CodeGen/X86/vector-tzcnt-128.ll
+++ b/llvm/test/CodeGen/X86/vector-tzcnt-128.ll
@@ -1666,6 +1666,122 @@ define <16 x i8> @testv16i8u(<16 x i8> %in) {
   ret <16 x i8> %out
 }
 
+define <2 x i64> @foldv2i64() {
+; SSE-LABEL: foldv2i64:
+; SSE:       # BB#0:
+; SSE-NEXT:    movl $8, %eax
+; SSE-NEXT:    movd %rax, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv2i64:
+; AVX:       # BB#0:
+; AVX-NEXT:    movl $8, %eax
+; AVX-NEXT:    vmovq %rax, %xmm0
+; AVX-NEXT:    retq
+  %out = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> <i64 256, i64 -1>, i1 0)
+  ret <2 x i64> %out
+}
+
+define <2 x i64> @foldv2i64u() {
+; SSE-LABEL: foldv2i64u:
+; SSE:       # BB#0:
+; SSE-NEXT:    movl $8, %eax
+; SSE-NEXT:    movd %rax, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv2i64u:
+; AVX:       # BB#0:
+; AVX-NEXT:    movl $8, %eax
+; AVX-NEXT:    vmovq %rax, %xmm0
+; AVX-NEXT:    retq
+  %out = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> <i64 256, i64 -1>, i1 -1)
+  ret <2 x i64> %out
+}
+
+define <4 x i32> @foldv4i32() {
+; SSE-LABEL: foldv4i32:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [8,0,32,0]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv4i32:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = [8,0,32,0]
+; AVX-NEXT:    retq
+  %out = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> <i32 256, i32 -1, i32 0, i32 255>, i1 0)
+  ret <4 x i32> %out
+}
+
+define <4 x i32> @foldv4i32u() {
+; SSE-LABEL: foldv4i32u:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [8,0,32,0]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv4i32u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = [8,0,32,0]
+; AVX-NEXT:    retq
+  %out = call <4 x i32> @llvm.cttz.v4i32(<4 x i32> <i32 256, i32 -1, i32 0, i32 255>, i1 -1)
+  ret <4 x i32> %out
+}
+
+define <8 x i16> @foldv8i16() {
+; SSE-LABEL: foldv8i16:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [8,0,16,0,16,0,3,3]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv8i16:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = [8,0,16,0,16,0,3,3]
+; AVX-NEXT:    retq
+  %out = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> <i16 256, i16 -1, i16 0, i16 255, i16 -65536, i16 7, i16 24, i16 88>, i1 0)
+  ret <8 x i16> %out
+}
+
+define <8 x i16> @foldv8i16u() {
+; SSE-LABEL: foldv8i16u:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [8,0,16,0,16,0,3,3]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv8i16u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = [8,0,16,0,16,0,3,3]
+; AVX-NEXT:    retq
+  %out = call <8 x i16> @llvm.cttz.v8i16(<8 x i16> <i16 256, i16 -1, i16 0, i16 255, i16 -65536, i16 7, i16 24, i16 88>, i1 -1)
+  ret <8 x i16> %out
+}
+
+define <16 x i8> @foldv16i8() {
+; SSE-LABEL: foldv16i8:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [8,0,8,0,8,0,3,3,1,1,0,1,2,3,4,5]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv16i8:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = [8,0,8,0,8,0,3,3,1,1,0,1,2,3,4,5]
+; AVX-NEXT:    retq
+  %out = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> <i8 256, i8 -1, i8 0, i8 255, i8 -65536, i8 7, i8 24, i8 88, i8 -2, i8 254, i8 1, i8 2, i8 4, i8 8, i8 16, i8 32>, i1 0)
+  ret <16 x i8> %out
+}
+
+define <16 x i8> @foldv16i8u() {
+; SSE-LABEL: foldv16i8u:
+; SSE:       # BB#0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [8,0,8,0,8,0,3,3,1,1,0,1,2,3,4,5]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: foldv16i8u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} xmm0 = [8,0,8,0,8,0,3,3,1,1,0,1,2,3,4,5]
+; AVX-NEXT:    retq
+  %out = call <16 x i8> @llvm.cttz.v16i8(<16 x i8> <i8 256, i8 -1, i8 0, i8 255, i8 -65536, i8 7, i8 24, i8 88, i8 -2, i8 254, i8 1, i8 2, i8 4, i8 8, i8 16, i8 32>, i1 -1)
+  ret <16 x i8> %out
+}
+
 declare <2 x i64> @llvm.cttz.v2i64(<2 x i64>, i1)
 declare <4 x i32> @llvm.cttz.v4i32(<4 x i32>, i1)
 declare <8 x i16> @llvm.cttz.v8i16(<8 x i16>, i1)
diff --git a/llvm/test/CodeGen/X86/vector-tzcnt-256.ll b/llvm/test/CodeGen/X86/vector-tzcnt-256.ll
index 32f7bcb00c5..8f744f79f85 100644
--- a/llvm/test/CodeGen/X86/vector-tzcnt-256.ll
+++ b/llvm/test/CodeGen/X86/vector-tzcnt-256.ll
@@ -1117,6 +1117,78 @@ define <32 x i8> @testv32i8u(<32 x i8> %in) {
   ret <32 x i8> %out
 }
 
+define <4 x i64> @foldv4i64() {
+; AVX-LABEL: foldv4i64:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,64,0]
+; AVX-NEXT:    retq
+  %out = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> <i64 256, i64 -1, i64 0, i64 255>, i1 0)
+  ret <4 x i64> %out
+}
+
+define <4 x i64> @foldv4i64u() {
+; AVX-LABEL: foldv4i64u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,64,0]
+; AVX-NEXT:    retq
+  %out = call <4 x i64> @llvm.cttz.v4i64(<4 x i64> <i64 256, i64 -1, i64 0, i64 255>, i1 -1)
+  ret <4 x i64> %out
+}
+
+define <8 x i32> @foldv8i32() {
+; AVX-LABEL: foldv8i32:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,32,0,16,0,3,3]
+; AVX-NEXT:    retq
+  %out = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> <i32 256, i32 -1, i32 0, i32 255, i32 -65536, i32 7, i32 24, i32 88>, i1 0)
+  ret <8 x i32> %out
+}
+
+define <8 x i32> @foldv8i32u() {
+; AVX-LABEL: foldv8i32u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,32,0,16,0,3,3]
+; AVX-NEXT:    retq
+  %out = call <8 x i32> @llvm.cttz.v8i32(<8 x i32> <i32 256, i32 -1, i32 0, i32 255, i32 -65536, i32 7, i32 24, i32 88>, i1 -1)
+  ret <8 x i32> %out
+}
+
+define <16 x i16> @foldv16i16() {
+; AVX-LABEL: foldv16i16:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,16,0,16,0,3,3,1,1,0,1,2,3,4,5]
+; AVX-NEXT:    retq
+  %out = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> <i16 256, i16 -1, i16 0, i16 255, i16 -65536, i16 7, i16 24, i16 88, i16 -2, i16 254, i16 1, i16 2, i16 4, i16 8, i16 16, i16 32>, i1 0)
+  ret <16 x i16> %out
+}
+
+define <16 x i16> @foldv16i16u() {
+; AVX-LABEL: foldv16i16u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,16,0,16,0,3,3,1,1,0,1,2,3,4,5]
+; AVX-NEXT:    retq
+  %out = call <16 x i16> @llvm.cttz.v16i16(<16 x i16> <i16 256, i16 -1, i16 0, i16 255, i16 -65536, i16 7, i16 24, i16 88, i16 -2, i16 254, i16 1, i16 2, i16 4, i16 8, i16 16, i16 32>, i1 -1)
+  ret <16 x i16> %out
+}
+
+define <32 x i8> @foldv32i8() {
+; AVX-LABEL: foldv32i8:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,8,0,8,0,3,3,1,1,0,1,2,3,4,5,6,7,8,8,7,6,5,4,3,2,1,0,0,0,0,0]
+; AVX-NEXT:    retq
+  %out = call <32 x i8> @llvm.cttz.v32i8(<32 x i8> <i8 256, i8 -1, i8 0, i8 255, i8 -65536, i8 7, i8 24, i8 88, i8 -2, i8 254, i8 1, i8 2, i8 4, i8 8, i8 16, i8 32, i8 64, i8 128, i8 256, i8 -256, i8 -128, i8 -64, i8 -32, i8 -16, i8 -8, i8 -4, i8 -2, i8 -1, i8 3, i8 5, i8 7, i8 127>, i1 0)
+  ret <32 x i8> %out
+}
+
+define <32 x i8> @foldv32i8u() {
+; AVX-LABEL: foldv32i8u:
+; AVX:       # BB#0:
+; AVX-NEXT:    vmovaps {{.*#+}} ymm0 = [8,0,8,0,8,0,3,3,1,1,0,1,2,3,4,5,6,7,8,8,7,6,5,4,3,2,1,0,0,0,0,0]
+; AVX-NEXT:    retq
+  %out = call <32 x i8> @llvm.cttz.v32i8(<32 x i8> <i8 256, i8 -1, i8 0, i8 255, i8 -65536, i8 7, i8 24, i8 88, i8 -2, i8 254, i8 1, i8 2, i8 4, i8 8, i8 16, i8 32, i8 64, i8 128, i8 256, i8 -256, i8 -128, i8 -64, i8 -32, i8 -16, i8 -8, i8 -4, i8 -2, i8 -1, i8 3, i8 5, i8 7, i8 127>, i1 -1)
+  ret <32 x i8> %out
+}
+
 declare <4 x i64> @llvm.cttz.v4i64(<4 x i64>, i1)
 declare <8 x i32> @llvm.cttz.v8i32(<8 x i32>, i1)
 declare <16 x i16> @llvm.cttz.v16i16(<16 x i16>, i1)