MrSidims wrote: > This is the lowering to the target.
My glossary might not be lacking some definitions, but what I really meant by lowering is: https://github.com/llvm/llvm-project/blob/main/llvm/lib/Target/AMDGPU or https://github.com/llvm/llvm-project/tree/main/llvm/lib/Target/SPIRV . I definitely see a value of having this resolved in one place instead of multiple places, and clang is a good candidate for it. On the other hand: a. some targets don't have such restriction; 2. clang is not a single frontend, especially now - in MLIR world, does it mean, what every frontend should make such adjustment when lowering to AMDGPU or SPIR-V? Just a note/disclaimer, this change makes sense and it doesn't bring (known to me, the reason I'm here is a surprise to see [this](https://godbolt.org/z/zW84fdace) during my experiments) regressions in some downsteam and adjusting tests in the translator is totally fine for me. What I'm trying to understand if it's really should be done in clang for OpenCL for every target. https://github.com/llvm/llvm-project/pull/115821 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
