diff options
| author | Dan Gohman <gohman@apple.com> | 2011-12-17 00:04:22 +0000 |
|---|---|---|
| committer | Dan Gohman <gohman@apple.com> | 2011-12-17 00:04:22 +0000 |
| commit | ce16339930a2b03e53b4e6399ef59c092a7f2cfa (patch) | |
| tree | 8265888014a73aa5764f99d7b6bb0553066f78bc /lib/VMCore/ConstantFold.cpp | |
| parent | 2e1b0c0cd9fab8f1841e1039cf39dcceeba70f73 (diff) | |
The powers that be have decided that LLVM IR should now support 16-bit
"half precision" floating-point with a first-class type.
This patch adds basic IR support (but not codegen support).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@146786 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/VMCore/ConstantFold.cpp')
| -rw-r--r-- | lib/VMCore/ConstantFold.cpp | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/lib/VMCore/ConstantFold.cpp b/lib/VMCore/ConstantFold.cpp index d1a9e7a70f..4d7a857de5 100644 --- a/lib/VMCore/ConstantFold.cpp +++ b/lib/VMCore/ConstantFold.cpp @@ -571,7 +571,8 @@ Constant *llvm::ConstantFoldCastInstruction(unsigned opc, Constant *V, if (ConstantFP *FPC = dyn_cast<ConstantFP>(V)) { bool ignored; APFloat Val = FPC->getValueAPF(); - Val.convert(DestTy->isFloatTy() ? APFloat::IEEEsingle : + Val.convert(DestTy->isHalfTy() ? APFloat::IEEEhalf : + DestTy->isFloatTy() ? APFloat::IEEEsingle : DestTy->isDoubleTy() ? APFloat::IEEEdouble : DestTy->isX86_FP80Ty() ? APFloat::x87DoubleExtended : DestTy->isFP128Ty() ? APFloat::IEEEquad : |