[CostModel] Add SK_ExtractSubvector handling to getInstructionThroughput (PR39368)

Add ShuffleVectorInst::isExtractSubvectorMask helper to match shuffle masks.

llvm-svn: 346510

4 files changed, 92 insertions, 38 deletions

diff --git a/llvm/include/llvm/IR/Instructions.h b/llvm/include/llvm/IR/Instructions.h
index 7b2c13c5328..01d83df90ae 100644
--- a/llvm/include/llvm/IR/Instructions.h
+++ b/llvm/include/llvm/IR/Instructions.h
@@ -2648,6 +2648,25 @@ public:
     return !changesLength() && isTransposeMask(getMask());
   }
 
+  /// Return true if this shuffle mask is an extract subvector mask.
+  /// A valid extract subvector mask returns a smaller vector from a single
+  /// source operand. The base extraction index is returned as well.
+  static bool isExtractSubvectorMask(ArrayRef<int> Mask, int NumSrcElts,
+                                     int &Index);
+  static bool isExtractSubvectorMask(const Constant *Mask, int NumSrcElts,
+                                     int &Index) {
+    assert(Mask->getType()->isVectorTy() && "Shuffle needs vector constant.");
+    SmallVector<int, 16> MaskAsInts;
+    getShuffleMask(Mask, MaskAsInts);
+    return isExtractSubvectorMask(MaskAsInts, NumSrcElts, Index);
+  }
+
+  /// Return true if this shuffle mask is an extract subvector mask.
+  bool isExtractSubvectorMask(int &Index) const {
+    int NumSrcElts = Op<0>()->getType()->getVectorNumElements();
+    return isExtractSubvectorMask(getMask(), NumSrcElts, Index);
+  }
+
   /// Change values in a shuffle permute mask assuming the two vector operands
   /// of length InVecNumElts have swapped position.
   static void commuteShuffleMask(MutableArrayRef<int> Mask,
diff --git a/llvm/lib/Analysis/TargetTransformInfo.cpp b/llvm/lib/Analysis/TargetTransformInfo.cpp
index 6e4eb8ff0cd..a6a1a872b65 100644
--- a/llvm/lib/Analysis/TargetTransformInfo.cpp
+++ b/llvm/lib/Analysis/TargetTransformInfo.cpp
@@ -1108,14 +1108,20 @@ int TargetTransformInfo::getInstructionThroughput(const Instruction *I) const {
   }
   case Instruction::ShuffleVector: {
     const ShuffleVectorInst *Shuffle = cast<ShuffleVectorInst>(I);
-    // TODO: Identify and add costs for insert/extract subvector, etc.
+    Type *Ty = Shuffle->getType();
+    Type *SrcTy = Shuffle->getOperand(0)->getType();
+
+    // TODO: Identify and add costs for insert subvector, etc.
+    int SubIndex;
+    if (Shuffle->isExtractSubvectorMask(SubIndex))
+      return TTIImpl->getShuffleCost(SK_ExtractSubvector, Ty, SubIndex, SrcTy);
+
     if (Shuffle->changesLength())
       return -1;
 
     if (Shuffle->isIdentity())
       return 0;
 
-    Type *Ty = Shuffle->getType();
     if (Shuffle->isReverse())
       return TTIImpl->getShuffleCost(SK_Reverse, Ty, 0, nullptr);
 
diff --git a/llvm/lib/IR/Instructions.cpp b/llvm/lib/IR/Instructions.cpp
index 7d4b6df18d9..66fdb462e82 100644
--- a/llvm/lib/IR/Instructions.cpp
+++ b/llvm/lib/IR/Instructions.cpp
@@ -1798,6 +1798,35 @@ bool ShuffleVectorInst::isTransposeMask(ArrayRef<int> Mask) {
   return true;
 }
 
+bool ShuffleVectorInst::isExtractSubvectorMask(ArrayRef<int> Mask,
+                                               int NumSrcElts, int &Index) {
+  // Must extract from a single source.
+  if (!isSingleSourceMaskImpl(Mask, NumSrcElts))
+    return false;
+
+  // Must be smaller (else this is an Identity shuffle).
+  if (NumSrcElts <= Mask.size())
+    return false;
+
+  // Find start of extraction, accounting that we may start with an UNDEF.
+  int SubIndex = -1;
+  for (int i = 0, e = Mask.size(); i != e; ++i) {
+    int M = Mask[i];
+    if (M < 0)
+      continue;
+    int Offset = (M % NumSrcElts) - i;
+    if (0 <= SubIndex && SubIndex != Offset)
+      return false;
+    SubIndex = Offset;
+  }
+
+  if (0 <= SubIndex) {
+    Index = SubIndex;
+    return true;
+  }
+  return false;
+}
+
 bool ShuffleVectorInst::isIdentityWithPadding() const {
   int NumOpElts = Op<0>()->getType()->getVectorNumElements();
   int NumMaskElts = getType()->getVectorNumElements();
diff --git a/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll b/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll
index 5bb2e1a756d..0ae25fa8f27 100644
--- a/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll
+++ b/llvm/test/Analysis/CostModel/X86/shuffle-extract_subvector.ll
@@ -18,27 +18,27 @@
 
 define void @test_vXf64(<4 x double> %src256, <8 x double> %src512) {
 ; CHECK-LABEL: 'test_vXf64'
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V256_01 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 0, i32 1>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V256_23 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_01 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 0, i32 1>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_23 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_45 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 4, i32 5>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_67 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 6, i32 7>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_0123 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_2345 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_4567 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_01 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_23 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_01 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_23 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_45 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 4, i32 5>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_67 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 6, i32 7>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_0123 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_2345 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_4567 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; BTVER2-LABEL: 'test_vXf64'
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V256_01 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 0, i32 1>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V256_23 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 2, i32 3>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_01 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 0, i32 1>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_23 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 2, i32 3>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_45 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 4, i32 5>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_67 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 6, i32 7>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_0123 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_2345 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_4567 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_01 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 0, i32 1>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_23 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 2, i32 3>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_01 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 0, i32 1>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_23 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 2, i32 3>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_45 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 4, i32 5>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_67 = shufflevector <8 x double> %src512, <8 x double> undef, <2 x i32> <i32 6, i32 7>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_0123 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_2345 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_4567 = shufflevector <8 x double> %src512, <8 x double> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %V256_01 = shufflevector <4 x double> %src256, <4 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -55,27 +55,27 @@ define void @test_vXf64(<4 x double> %src256, <8 x double> %src512) {
 
 define void @test_vXfi64(<4 x i64> %src256, <8 x i64> %src512) {
 ; CHECK-LABEL: 'test_vXfi64'
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V256_01 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 0, i32 1>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V256_23 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_01 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 0, i32 1>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_23 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_45 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 4, i32 5>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_67 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 6, i32 7>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_0123 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_2345 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
-; CHECK-NEXT:  Cost Model: Unknown cost for instruction: %V512_4567 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_01 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_23 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_01 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_23 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_45 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 4, i32 5>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_67 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 6, i32 7>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_0123 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_2345 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_4567 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; BTVER2-LABEL: 'test_vXfi64'
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V256_01 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 0, i32 1>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V256_23 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 2, i32 3>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_01 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 0, i32 1>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_23 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 2, i32 3>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_45 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 4, i32 5>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_67 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 6, i32 7>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_0123 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_2345 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
-; BTVER2-NEXT:  Cost Model: Unknown cost for instruction: %V512_4567 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_01 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 0, i32 1>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V256_23 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 2, i32 3>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_01 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 0, i32 1>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_23 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 2, i32 3>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_45 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 4, i32 5>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_67 = shufflevector <8 x i64> %src512, <8 x i64> undef, <2 x i32> <i32 6, i32 7>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_0123 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_2345 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V512_4567 = shufflevector <8 x i64> %src512, <8 x i64> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %V256_01 = shufflevector <4 x i64> %src256, <4 x i64> undef, <2 x i32> <i32 0, i32 1>