diff options
| author | River Riddle <riverriddle@google.com> | 2019-12-22 21:59:55 -0800 |
|---|---|---|
| committer | A. Unique TensorFlower <gardener@tensorflow.org> | 2019-12-22 22:00:23 -0800 |
| commit | 35807bc4c5c9d8abc31ba0b2f955a82abf276e12 (patch) | |
| tree | d083d37d993a774239081509a50e3e6c65366421 /mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp | |
| parent | 22954a0e408afde1d8686dffb3a3dcab107a2cd3 (diff) | |
| download | bcm5719-llvm-35807bc4c5c9d8abc31ba0b2f955a82abf276e12.tar.gz bcm5719-llvm-35807bc4c5c9d8abc31ba0b2f955a82abf276e12.zip | |
NFC: Introduce new ValuePtr/ValueRef typedefs to simplify the transition to Value being value-typed.
This is an initial step to refactoring the representation of OpResult as proposed in: https://groups.google.com/a/tensorflow.org/g/mlir/c/XXzzKhqqF_0/m/v6bKb08WCgAJ
This change will make it much simpler to incrementally transition all of the existing code to use value-typed semantics.
PiperOrigin-RevId: 286844725
Diffstat (limited to 'mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp')
| -rw-r--r-- | mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp b/mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp index 21abc3cf99b..63836883512 100644 --- a/mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp +++ b/mlir/lib/Conversion/LoopsToGPU/LoopsToGPUPass.cpp @@ -98,7 +98,7 @@ struct ImperfectlyNestedForLoopMapper // pass is only used for testing. FuncOp funcOp = getFunction(); OpBuilder builder(funcOp.getOperation()->getRegion(0)); - SmallVector<Value *, 3> numWorkGroupsVal, workGroupSizeVal; + SmallVector<ValuePtr, 3> numWorkGroupsVal, workGroupSizeVal; for (auto val : numWorkGroups) { auto constOp = builder.create<ConstantOp>( funcOp.getLoc(), builder.getIntegerAttr(builder.getIndexType(), val)); |
