[X86][SSE] combineSubToSubus - begun generalizing to work with any type sizes with...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@326030 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 732503f..9ad5c18 100644 (file)
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -37639,7 +37639,7 @@ static SDValue combineSubToSubus(SDNode *N, SelectionDAG &DAG,
   // for v8i32 requires umin, which appears in SSE41.
   if (!(Subtarget.hasSSE2() && (VT == MVT::v16i8 || VT == MVT::v8i16)) &&
       !(Subtarget.hasSSE41() && (VT == MVT::v8i32)) &&
-      !(Subtarget.hasAVX2() && (VT == MVT::v32i8 || VT == MVT::v16i16)) &&
+      !(Subtarget.hasAVX() && (VT == MVT::v32i8 || VT == MVT::v16i16)) &&
       !(Subtarget.useBWIRegs() && (VT == MVT::v64i8 || VT == MVT::v32i16 ||
                                    VT == MVT::v16i32 || VT == MVT::v8i64)))
     return SDValue();
@@ -37671,10 +37671,16 @@ static SDValue combineSubToSubus(SDNode *N, SelectionDAG &DAG,
   } else
     return SDValue();
 
+  auto SUBUSBuilder = [](SelectionDAG &DAG, const SDLoc &DL, SDValue Op0,
+                         SDValue Op1) {
+    return DAG.getNode(X86ISD::SUBUS, DL, Op0.getValueType(), Op0, Op1);
+  };
+
   // PSUBUS doesn't support v8i32/v8i64/v16i32, but it can be enabled with
   // special preprocessing in some cases.
   if (VT != MVT::v8i32 && VT != MVT::v16i32 && VT != MVT::v8i64)
-    return DAG.getNode(X86ISD::SUBUS, SDLoc(N), VT, SubusLHS, SubusRHS);
+    return SplitBinaryOpsAndApply(DAG, Subtarget, SDLoc(N), VT, SubusLHS,
+                                  SubusRHS, SUBUSBuilder);
 
   // Special preprocessing case can be only applied
   // if the value was zero extended from 16 bit,
@@ -37704,8 +37710,9 @@ static SDValue combineSubToSubus(SDNode *N, SelectionDAG &DAG,
   SDValue NewSubusLHS =
       DAG.getZExtOrTrunc(SubusLHS, SDLoc(SubusLHS), ShrinkedType);
   SDValue NewSubusRHS = DAG.getZExtOrTrunc(UMin, SDLoc(SubusRHS), ShrinkedType);
-  SDValue Psubus = DAG.getNode(X86ISD::SUBUS, SDLoc(N), ShrinkedType,
-                               NewSubusLHS, NewSubusRHS);
+  SDValue Psubus =
+      SplitBinaryOpsAndApply(DAG, Subtarget, SDLoc(N), ShrinkedType,
+                             NewSubusLHS, NewSubusRHS, SUBUSBuilder);
   // Zero extend the result, it may be used somewhere as 32 bit,
   // if not zext and following trunc will shrink.
   return DAG.getZExtOrTrunc(Psubus, SDLoc(N), ExtType);

diff --git a/test/CodeGen/X86/psubus.ll b/test/CodeGen/X86/psubus.ll
index e3903f2..8de1ae4 100644 (file)
--- a/test/CodeGen/X86/psubus.ll
+++ b/test/CodeGen/X86/psubus.ll
@@ -1202,16 +1202,16 @@ define <32 x i16> @psubus_32i16_max(<32 x i16> %x, <32 x i16> %y) nounwind {
 ;
 ; AVX1-LABEL: psubus_32i16_max:
 ; AVX1:       # %bb.0: # %vector.ph
-; AVX1-NEXT:    vextractf128 $1, %ymm3, %xmm4
-; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm5
+; AVX1-NEXT:    vextractf128 $1, %ymm2, %xmm4
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm5
 ; AVX1-NEXT:    vpsubusw %xmm4, %xmm5, %xmm4
-; AVX1-NEXT:    vextractf128 $1, %ymm2, %xmm5
-; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm6
-; AVX1-NEXT:    vpsubusw %xmm5, %xmm6, %xmm5
 ; AVX1-NEXT:    vpsubusw %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vinsertf128 $1, %xmm5, %ymm0, %ymm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm0, %ymm0
+; AVX1-NEXT:    vextractf128 $1, %ymm3, %xmm2
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm4
+; AVX1-NEXT:    vpsubusw %xmm2, %xmm4, %xmm2
 ; AVX1-NEXT:    vpsubusw %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm1, %ymm1
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm1
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: psubus_32i16_max:
@@ -1242,16 +1242,16 @@ define <64 x i8> @psubus_64i8_max(<64 x i8> %x, <64 x i8> %y) nounwind {
 ;
 ; AVX1-LABEL: psubus_64i8_max:
 ; AVX1:       # %bb.0: # %vector.ph
-; AVX1-NEXT:    vextractf128 $1, %ymm3, %xmm4
-; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm5
+; AVX1-NEXT:    vextractf128 $1, %ymm2, %xmm4
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm5
 ; AVX1-NEXT:    vpsubusb %xmm4, %xmm5, %xmm4
-; AVX1-NEXT:    vextractf128 $1, %ymm2, %xmm5
-; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm6
-; AVX1-NEXT:    vpsubusb %xmm5, %xmm6, %xmm5
 ; AVX1-NEXT:    vpsubusb %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vinsertf128 $1, %xmm5, %ymm0, %ymm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm0, %ymm0
+; AVX1-NEXT:    vextractf128 $1, %ymm3, %xmm2
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm4
+; AVX1-NEXT:    vpsubusb %xmm2, %xmm4, %xmm2
 ; AVX1-NEXT:    vpsubusb %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm1, %ymm1
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm1
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: psubus_64i8_max: