[AArch64] Use PerfectShuffle costs in AArch64TTIImpl::getShuffleCost

Given a shuffle with 4 elements size 16 or 32, we can use the costs
directly from the PerfectShuffle tables to get a slightly more accurate
cost for the resulting shuffle.

Differential Revision: https://reviews.llvm.org/D123409

diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
index d5316cd..a5dadd1 100644 (file)
--- a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -11593,23 +11593,8 @@ bool AArch64TargetLowering::isShuffleMaskLegal(ArrayRef<int> M, EVT VT) const {
 
   if (VT.getVectorNumElements() == 4 &&
       (VT.is128BitVector() || VT.is64BitVector())) {
-    unsigned PFIndexes[4];
-    for (unsigned i = 0; i != 4; ++i) {
-      if (M[i] < 0)
-        PFIndexes[i] = 8;
-      else
-        PFIndexes[i] = M[i];
-    }
-
-    // Compute the index in the perfect shuffle table.
-    unsigned PFTableIndex = PFIndexes[0] * 9 * 9 * 9 + PFIndexes[1] * 9 * 9 +
-                            PFIndexes[2] * 9 + PFIndexes[3];
-    unsigned PFEntry = PerfectShuffleTable[PFTableIndex];
-    unsigned Cost = (PFEntry >> 30);
-
-    // The cost tables encode cost 0 or cost 1 shuffles using the value 0 in
-    // the top 2 bits.
-    if (Cost == 0)
+    unsigned Cost = getPerfectShuffleCost(M);
+    if (Cost <= 1)
       return true;
   }
 

diff --git a/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h b/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h
index e58c104..65f77b4 100644 (file)
--- a/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h
+++ b/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h
@@ -14,6 +14,8 @@
 #ifndef LLVM_LIB_TARGET_AARCH64_AARCH64PERFECTSHUFFLE_H
 #define LLVM_LIB_TARGET_AARCH64_AARCH64PERFECTSHUFFLE_H
 
+#include "llvm/ADT/ArrayRef.h"
+
 // 31 entries have cost 0
 // 730 entries have cost 1
 // 3658 entries have cost 2
@@ -6584,4 +6586,36 @@ static const unsigned PerfectShuffleTable[6561 + 1] = {
     835584U,     // <u,u,u,u>: Cost 0 copy LHS
     0};
 
+static unsigned getPerfectShuffleCost(llvm::ArrayRef<int> M) {
+  assert(M.size() == 4 && "Expected a 4 entry perfect shuffle");
+
+  // Special case zero-cost nop copies, from either LHS or RHS.
+  if (llvm::all_of(llvm::enumerate(M), [](auto &E) {
+        return E.value() < 0 || E.value() == (int)E.index();
+      }))
+    return 0;
+  if (llvm::all_of(llvm::enumerate(M), [](auto &E) {
+        return E.value() < 0 || E.value() == (int)E.index() + 4;
+      }))
+    return 0;
+
+  // Get the four mask elementd from the 2 inputs. Perfect shuffles encode undef
+  // elements with value 8.
+  unsigned PFIndexes[4];
+  for (unsigned i = 0; i != 4; ++i) {
+    assert(M[i] < 8 && "Expected a maximum entry of 8 for shuffle mask");
+    if (M[i] < 0)
+      PFIndexes[i] = 8;
+    else
+      PFIndexes[i] = M[i];
+  }
+
+  // Compute the index in the perfect shuffle table.
+  unsigned PFTableIndex = PFIndexes[0] * 9 * 9 * 9 + PFIndexes[1] * 9 * 9 +
+                          PFIndexes[2] * 9 + PFIndexes[3];
+  unsigned PFEntry = PerfectShuffleTable[PFTableIndex];
+  // And extract the cost from the upper bits. The cost is encoded as Cost-1.
+  return (PFEntry >> 30) + 1;
+}
+
 #endif

diff --git a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
index 65f9dbe..2ae1269 100644 (file)
--- a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
+++ b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
@@ -8,6 +8,7 @@
 
 #include "AArch64TargetTransformInfo.h"
 #include "AArch64ExpandImm.h"
+#include "AArch64PerfectShuffle.h"
 #include "MCTargetDesc/AArch64AddressingModes.h"
 #include "llvm/Analysis/IVDescriptors.h"
 #include "llvm/Analysis/LoopInfo.h"
@@ -2597,19 +2598,27 @@ InstructionCost AArch64TTIImpl::getShuffleCost(TTI::ShuffleKind Kind,
                                                ArrayRef<const Value *> Args) {
   Kind = improveShuffleKindFromMask(Kind, Mask);
   std::pair<InstructionCost, MVT> LT = TLI->getTypeLegalizationCost(DL, Tp);
+
+  // Check for broadcast loads.
+  if (Kind == TTI::SK_Broadcast) {
+    bool IsLoad = !Args.empty() && isa<LoadInst>(Args[0]);
+    if (IsLoad && LT.second.isVector() &&
+        isLegalBroadcastLoad(Tp->getElementType(),
+                              LT.second.getVectorElementCount()))
+      return 0; // broadcast is handled by ld1r
+  }
+
+  // If we have 4 elements for the shuffle and a Mask, get the cost straight
+  // from the perfect shuffle tables.
+  if (Mask.size() == 4 && Tp->getElementCount() == ElementCount::getFixed(4) &&
+      (Tp->getScalarSizeInBits() == 16 || Tp->getScalarSizeInBits() == 32) &&
+      all_of(Mask, [](int E) { return E < 8; }))
+    return getPerfectShuffleCost(Mask);
+
   if (Kind == TTI::SK_Broadcast || Kind == TTI::SK_Transpose ||
       Kind == TTI::SK_Select || Kind == TTI::SK_PermuteSingleSrc ||
       Kind == TTI::SK_Reverse) {
 
-    // Check for broadcast loads.
-    if (Kind == TTI::SK_Broadcast) {
-      bool IsLoad = !Args.empty() && isa<LoadInst>(Args[0]);
-      if (IsLoad && LT.second.isVector() &&
-          isLegalBroadcastLoad(Tp->getElementType(),
-                               LT.second.getVectorElementCount()))
-        return 0; // broadcast is handled by ld1r
-    }
-
     static const CostTblEntry ShuffleTbl[] = {
       // Broadcast shuffle kinds can be performed with 'dup'.
       { TTI::SK_Broadcast, MVT::v8i8,  1 },

diff --git a/llvm/test/Analysis/CostModel/AArch64/shuffle-other.ll b/llvm/test/Analysis/CostModel/AArch64/shuffle-other.ll
index 80be01b..8d07557 100644 (file)
--- a/llvm/test/Analysis/CostModel/AArch64/shuffle-other.ll
+++ b/llvm/test/Analysis/CostModel/AArch64/shuffle-other.ll
@@ -8,12 +8,12 @@ define void @shuffle() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v9 = shufflevector <8 x i8> undef, <8 x i8> undef, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v10 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> <i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v11 = shufflevector <2 x i16> undef, <2 x i16> undef, <2 x i32> <i32 1, i32 0>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v12 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v12 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %v13 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 0, i32 1>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v14 = shufflevector <2 x i32> undef, <2 x i32> undef, <2 x i32> <i32 1, i32 0>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v15 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v15 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16 = shufflevector <2 x float> undef, <2 x float> undef, <2 x i32> <i32 1, i32 0>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v17 = shufflevector <4 x float> undef, <4 x float> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v17 = shufflevector <4 x float> undef, <4 x float> undef, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %v7 = shufflevector <2 x i8> undef, <2 x i8> undef, <2 x i32> <i32 1, i32 0>
@@ -39,16 +39,16 @@ define void @concat() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i8 = shufflevector <2 x i8> undef, <2 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i8 = shufflevector <4 x i8> undef, <4 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i8 = shufflevector <8 x i8> undef, <8 x i8> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16 = shufflevector <2 x i16> undef, <2 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4i16 = shufflevector <2 x i16> undef, <2 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i16 = shufflevector <4 x i16> undef, <4 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i16 = shufflevector <8 x i16> undef, <8 x i16> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32 = shufflevector <2 x i32> undef, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4i32 = shufflevector <2 x i32> undef, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32 = shufflevector <4 x i32> undef, <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i64 = shufflevector <2 x i64> undef, <2 x i64> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v4f16 = shufflevector <2 x half> undef, <2 x half> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4f16 = shufflevector <2 x half> undef, <2 x half> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8f16 = shufflevector <4 x half> undef, <4 x half> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16f16 = shufflevector <8 x half> undef, <8 x half> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f32 = shufflevector <2 x float> undef, <2 x float> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4f32 = shufflevector <2 x float> undef, <2 x float> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8f32 = shufflevector <4 x float> undef, <4 x float> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4f64 = shufflevector <2 x double> undef, <2 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
@@ -92,8 +92,8 @@ define void @insert_subvec() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i8_4_2 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 12, i32 13, i32 14, i32 15>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i8_4_3 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %v16i8_4_05 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> <i32 0, i32 1, i32 16, i32 17, i32 18, i32 19, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %v4i16_2_0 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i16_2_1 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 4, i32 5, i32 0, i32 1>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i16_2_0 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i16_2_1 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> <i32 4, i32 5, i32 0, i32 1>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 42 for instruction: %v8i16_2_0 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> <i32 8, i32 9, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i16_2_1 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> <i32 0, i32 1, i32 8, i32 9, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i16_2_2 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 6, i32 7>
@@ -105,7 +105,7 @@ define void @insert_subvec() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16i16_4_3 = shufflevector <16 x i16> undef, <16 x i16> undef, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %v16i16_4_05 = shufflevector <16 x i16> undef, <16 x i16> undef, <16 x i32> <i32 0, i32 1, i32 16, i32 17, i32 18, i32 19, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_2_0 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4i32_2_1 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> <i32 4, i32 5, i32 0, i32 1>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4i32_2_1 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> <i32 4, i32 5, i32 0, i32 1>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %v8i32_2_0 = shufflevector <8 x i32> undef, <8 x i32> undef, <8 x i32> <i32 8, i32 9, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32_2_1 = shufflevector <8 x i32> undef, <8 x i32> undef, <8 x i32> <i32 0, i32 1, i32 8, i32 9, i32 4, i32 5, i32 6, i32 7>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8i32_2_2 = shufflevector <8 x i32> undef, <8 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 6, i32 7>

diff --git a/llvm/test/Analysis/CostModel/AArch64/shuffle-select.ll b/llvm/test/Analysis/CostModel/AArch64/shuffle-select.ll
index 9d430ec..afe5f38 100644 (file)
--- a/llvm/test/Analysis/CostModel/AArch64/shuffle-select.ll
+++ b/llvm/test/Analysis/CostModel/AArch64/shuffle-select.ll
@@ -20,7 +20,7 @@ define <16 x i8> @sel.v16i8(<16 x i8> %v0, <16 x i8> %v1) {
 }
 
 ; COST-LABEL: sel.v4i16
-; COST:       Found an estimated cost of 18 for instruction: %tmp0 = shufflevector <4 x i16> %v0, <4 x i16> %v1, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; COST:       Found an estimated cost of 2 for instruction: %tmp0 = shufflevector <4 x i16> %v0, <4 x i16> %v1, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
 ; CODE-LABEL: sel.v4i16
 ; CODE:       rev32 v0.4h, v0.4h
 ; CODE:       trn2 v0.4h, v0.4h, v1.4h

diff --git a/llvm/test/Transforms/PhaseOrdering/AArch64/matrix-extract-insert.ll b/llvm/test/Transforms/PhaseOrdering/AArch64/matrix-extract-insert.ll
index eee888d..f56c4e2 100644 (file)
--- a/llvm/test/Transforms/PhaseOrdering/AArch64/matrix-extract-insert.ll
+++ b/llvm/test/Transforms/PhaseOrdering/AArch64/matrix-extract-insert.ll
@@ -308,16 +308,10 @@ declare void @llvm.lifetime.end.p0i8(i64 immarg, i8* nocapture) #1
 
 define <4 x float> @reverse_hadd_v4f32(<4 x float> %a, <4 x float> %b) {
 ; CHECK-LABEL: @reverse_hadd_v4f32(
-; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[A:%.*]], <4 x float> undef, <2 x i32> <i32 2, i32 0>
-; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[A]], <4 x float> undef, <2 x i32> <i32 3, i32 1>
-; CHECK-NEXT:    [[TMP3:%.*]] = fadd <2 x float> [[TMP1]], [[TMP2]]
-; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <2 x float> [[TMP3]], <2 x float> poison, <4 x i32> <i32 undef, i32 undef, i32 0, i32 1>
-; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <4 x float> [[B:%.*]], <4 x float> undef, <2 x i32> <i32 2, i32 0>
-; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x float> [[B]], <4 x float> undef, <2 x i32> <i32 3, i32 1>
-; CHECK-NEXT:    [[TMP7:%.*]] = fadd <2 x float> [[TMP5]], [[TMP6]]
-; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <2 x float> [[TMP7]], <2 x float> poison, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <4 x float> [[TMP8]], <4 x float> [[TMP4]], <4 x i32> <i32 0, i32 1, i32 6, i32 7>
-; CHECK-NEXT:    ret <4 x float> [[TMP9]]
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[B:%.*]], <4 x float> [[A:%.*]], <4 x i32> <i32 2, i32 0, i32 6, i32 4>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[B]], <4 x float> [[A]], <4 x i32> <i32 3, i32 1, i32 7, i32 5>
+; CHECK-NEXT:    [[TMP3:%.*]] = fadd <4 x float> [[TMP1]], [[TMP2]]
+; CHECK-NEXT:    ret <4 x float> [[TMP3]]
 ;
   %vecext = extractelement <4 x float> %a, i32 0
   %vecext1 = extractelement <4 x float> %a, i32 1