diff options
Diffstat (limited to 'llvm/lib/Transforms')
| -rw-r--r-- | llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp | 19 |
1 files changed, 15 insertions, 4 deletions
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp index 600c8c36392..a7fb7467404 100644 --- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp +++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp @@ -960,7 +960,20 @@ Instruction *InstCombiner::visitCallInst(CallInst &CI) { // This optimization is convoluted because the intrinsic is defined as // getting a vector of floats or doubles for the ps and pd versions. // FIXME: That should be changed. + + Value *Op0 = II->getArgOperand(0); + Value *Op1 = II->getArgOperand(1); Value *Mask = II->getArgOperand(2); + + // fold (blend A, A, Mask) -> A + if (Op0 == Op1) + return ReplaceInstUsesWith(CI, Op0); + + // Zero Mask - select 1st argument. + if (auto C = dyn_cast<ConstantAggregateZero>(Mask)) + return ReplaceInstUsesWith(CI, Op0); + + // Constant Mask - select 1st/2nd argument lane based on top bit of mask. if (auto C = dyn_cast<ConstantDataVector>(Mask)) { auto Tyi1 = Builder->getInt1Ty(); auto SelectorType = cast<VectorType>(Mask->getType()); @@ -983,11 +996,9 @@ Instruction *InstCombiner::visitCallInst(CallInst &CI) { Selectors.push_back(ConstantInt::get(Tyi1, Selector >> (BitWidth - 1))); } auto NewSelector = ConstantVector::get(Selectors); - return SelectInst::Create(NewSelector, II->getArgOperand(1), - II->getArgOperand(0), "blendv"); - } else { - break; + return SelectInst::Create(NewSelector, Op1, Op0, "blendv"); } + break; } case Intrinsic::x86_avx_vpermilvar_ps: |
