[Mlir-commits] [mlir] [mlir] Execute same operand name constraints before user constraints. (PR #162526)

[Chenguang Wang]

wecing wrote:

> LLVM Buildbot has detected a new failure on builder `mlir-nvidia-gcc7` running on `mlir-nvidia` while building `mlir` at step 7 "test-build-check-mlir-build-only-check-mlir".
> 
> Full details are available at: https://lab.llvm.org/buildbot/#/builders/116/builds/19516
> 
> Here is the relevant piece of the build log for the reference

I am having some troubles reproducing the error locally.

```
$ cmake -G Ninja ../llvm \
          -DLLVM_ENABLE_PROJECTS=mlir \
          -DLLVM_BUILD_EXAMPLES=ON \
          -DLLVM_TARGETS_TO_BUILD="Native;NVPTX;AMDGPU" \
          -DCMAKE_BUILD_TYPE=Release -DMLIR_ENABLE_CUDA_RUNNER=1 -DMLIR_INCLUDE_INTEGRATION_TESTS=ON \
          -DLLVM_ENABLE_ASSERTIONS=ON
$ cmake --build . --target check-mlir
$ bin/llvm-lit -v tools/mlir/test/Integration/GPU/CUDA/async.mlir
...
# executed command: /usr/local/google/home/chenguangwang/Code/llvm-project/build/bin/mlir-opt -gpu-async-region -gpu-to-llvm -reconcile-unrealized-casts -gpu-module-to-binary=format=fatbin
# .---command stderr------------                                                                                                                                                      
# | <stdin>:35:3: error: LibDevice path: /usr/nvvm/libdevice/libdevice.10.bc does not exist or is not a file.
```

@chsigg, are you still the owner of this test? Do we need special devices/setups for it?

https://github.com/llvm/llvm-project/pull/162526