diff options
| author | Farhana Aleen <farhana.aleen@gmail.com> | 2018-05-02 16:24:10 +0000 |
|---|---|---|
| committer | Farhana Aleen <farhana.aleen@gmail.com> | 2018-05-02 16:24:10 +0000 |
| commit | 2f4100f56e605df76d32a5407218e5da32217161 (patch) | |
| tree | 4b3ba9d88562fbe0796507e5116188dcfe8a0ac9 /llvm/lib | |
| parent | a53d3308904714540f3a53458120d6a215433e85 (diff) | |
| download | bcm5719-llvm-2f4100f56e605df76d32a5407218e5da32217161.tar.gz bcm5719-llvm-2f4100f56e605df76d32a5407218e5da32217161.zip | |
[AMDGPU] performAddCombine should run after DAG is legalized.
Summary: performAddCombine should run after DAG is legalized; Otherwise generic optimization
in the DAGCombiner can optimize an addcarry+trunc into an addcarry instruction with
illegal types.
Author: FarhanaAleen
Reviewed By: rampitec
Subscribers: llvm-commits, AMDGPU
Differential Revision: https://reviews.llvm.org/D46337
llvm-svn: 331368
Diffstat (limited to 'llvm/lib')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/SIISelLowering.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp index 1ed6dd963fa..d98c2a272c9 100644 --- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp +++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp @@ -6755,7 +6755,7 @@ SDValue SITargetLowering::performAddCombine(SDNode *N, return SDValue(); } - if (VT != MVT::i32) + if (VT != MVT::i32 || !DCI.isAfterLegalizeDAG()) return SDValue(); // add x, zext (setcc) => addcarry x, 0, setcc |
