[Mlir-commits] [mlir] [mlir][x86vector] AVX Convert/Broadcast BF16 to F32 instructions (PR #135143)

[Renato Golin]

rengolin wrote:

> Interesting. There is a CMake cache that the failing bots are using in `offload/cmake/caches/AMDGPUBot.cmake`. If you have info on which headers are missing / which problems you ran into, I'd be quite curious as the whole point is to have that build config be reproducible. Dockerfiles for the containers we use are in https://github.com/ROCm/aomp/tree/aomp-dev/upstream-buildbots

I think they managed to find the issue. Thanks!

https://github.com/llvm/llvm-project/pull/135143