From d77c1c4465fda5a7ec5432d4e6644c50aa7a74fe Mon Sep 17 00:00:00 2001
From: Filipe Cabecinhas <me@filcab.net>
Date: Fri, 16 May 2014 22:47:54 +0000
Subject: [PATCH] Implemented special cases for PerformVSELECTCombine.

vselects with constant masks, after legalization, will get turned into
specialized shuffle_vectors so they can be matched to blend+imm
instructions.

Fixed some tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@209044 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/X86/X86ISelLowering.cpp | 62 ++++++++++++++++++++++++++++++++++++++
 test/CodeGen/X86/avx-blend.ll      | 10 +++---
 2 files changed, 67 insertions(+), 5 deletions(-)

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index c6e7730f827..f88ed312490 100644
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -17736,6 +17736,51 @@ matchIntegerMINMAX(SDValue Cond, EVT VT, SDValue LHS, SDValue RHS,
   return std::make_pair(Opc, NeedSplit);
 }
 
+static SDValue
+TransformVSELECTtoBlendVECTOR_SHUFFLE(SDNode *N, SelectionDAG &DAG,
+                                      const X86Subtarget *Subtarget) {
+  SDLoc dl(N);
+  SDValue Cond = N->getOperand(0);
+  SDValue LHS = N->getOperand(1);
+  SDValue RHS = N->getOperand(2);
+
+  if (Cond.getOpcode() == ISD::SIGN_EXTEND) {
+    SDValue CondSrc = Cond->getOperand(0);
+    if (CondSrc->getOpcode() == ISD::SIGN_EXTEND_INREG)
+      Cond = CondSrc->getOperand(0);
+  }
+
+  MVT VT = N->getSimpleValueType(0);
+  MVT EltVT = VT.getVectorElementType();
+  unsigned NumElems = VT.getVectorNumElements();
+  // There is no blend with immediate in AVX-512.
+  if (VT.is512BitVector())
+    return SDValue();
+
+  if (!Subtarget->hasSSE41() || EltVT == MVT::i8)
+    return SDValue();
+  if (!Subtarget->hasInt256() && VT == MVT::v16i16)
+    return SDValue();
+
+  if (!ISD::isBuildVectorOfConstantSDNodes(Cond.getNode()))
+    return SDValue();
+
+  unsigned MaskValue = 0;
+  if (!BUILD_VECTORtoBlendMask(cast<BuildVectorSDNode>(Cond), MaskValue))
+    return SDValue();
+
+  SmallVector<int, 8> ShuffleMask(NumElems, -1);
+  for (unsigned i = 0; i < NumElems; ++i) {
+    // Be sure we emit undef where we can.
+    if (Cond.getOperand(i)->getOpcode() == ISD::UNDEF)
+      ShuffleMask[i] = -1;
+    else
+      ShuffleMask[i] = i + NumElems * ((MaskValue >> i) & 1);
+  }
+
+  return DAG.getVectorShuffle(VT, dl, LHS, RHS, &ShuffleMask[0]);
+}
+
 /// PerformSELECTCombine - Do target-specific dag combines on SELECT and VSELECT
 /// nodes.
 static SDValue PerformSELECTCombine(SDNode *N, SelectionDAG &DAG,
@@ -18277,6 +18322,23 @@ static SDValue PerformSELECTCombine(SDNode *N, SelectionDAG &DAG,
       DCI.CommitTargetLoweringOpt(TLO);
   }
 
+  // We should generate an X86ISD::BLENDI from a vselect if its argument
+  // is a sign_extend_inreg of an any_extend of a BUILD_VECTOR of
+  // constants. This specific pattern gets generated when we split a
+  // selector for a 512 bit vector in a machine without AVX512 (but with
+  // 256-bit vectors), during legalization:
+  //
+  // (vselect (sign_extend (any_extend (BUILD_VECTOR)) i1) LHS RHS)
+  //
+  // Iff we find this pattern and the build_vectors are built from
+  // constants, we translate the vselect into a shuffle_vector that we
+  // know will be matched by LowerVECTOR_SHUFFLEtoBlend.
+  if (N->getOpcode() == ISD::VSELECT && !DCI.isBeforeLegalize()) {
+    SDValue Shuffle = TransformVSELECTtoBlendVECTOR_SHUFFLE(N, DAG, Subtarget);
+    if (Shuffle.getNode())
+      return Shuffle;
+  }
+
   return SDValue();
 }
 
diff --git a/test/CodeGen/X86/avx-blend.ll b/test/CodeGen/X86/avx-blend.ll
index 4757ce0c9fa..8577a616c3c 100644
--- a/test/CodeGen/X86/avx-blend.ll
+++ b/test/CodeGen/X86/avx-blend.ll
@@ -69,7 +69,8 @@ define <8 x i32> @vsel_i328(<8 x i32> %v1, <8 x i32> %v2) {
 }
 
 ;CHECK-LABEL: vsel_double8:
-;CHECK: vblendvpd
+;CHECK: vblendpd    $1
+;CHECK: vblendpd    $1
 ;CHECK: ret
 define <8 x double> @vsel_double8(<8 x double> %v1, <8 x double> %v2) {
   %vsel = select <8 x i1> <i1 true, i1 false, i1 false, i1 false, i1 true, i1 false, i1 false, i1 false>, <8 x double> %v1, <8 x double> %v2
@@ -77,7 +78,8 @@ define <8 x double> @vsel_double8(<8 x double> %v1, <8 x double> %v2) {
 }
 
 ;CHECK-LABEL: vsel_i648:
-;CHECK: vblendvpd
+;CHECK: vblendpd    $1
+;CHECK: vblendpd    $1
 ;CHECK: ret
 define <8 x i64> @vsel_i648(<8 x i64> %v1, <8 x i64> %v2) {
   %vsel = select <8 x i1> <i1 true, i1 false, i1 false, i1 false, i1 true, i1 false, i1 false, i1 false>, <8 x i64> %v1, <8 x i64> %v2
@@ -86,7 +88,7 @@ define <8 x i64> @vsel_i648(<8 x i64> %v1, <8 x i64> %v2) {
 
 ;CHECK-LABEL: vsel_double4:
 ;CHECK-NOT: vinsertf128
-;CHECK: vblendpd    $5
+;CHECK: vshufpd $10
 ;CHECK-NEXT: ret
 define <4 x double> @vsel_double4(<4 x double> %v1, <4 x double> %v2) {
   %vsel = select <4 x i1> <i1 true, i1 false, i1 true, i1 false>, <4 x double> %v1, <4 x double> %v2
@@ -111,5 +113,3 @@ define <2 x double> @testb(<2 x double> %x, <2 x double> %y) {
   %min = select <2 x i1> %min_is_x, <2 x double> %x, <2 x double> %y
   ret <2 x double> %min
 }
-
-
-- 
2.11.0