Artem-B wrote: > > LGTM in principle, but it could use some tests. The change is surprisingly > > nicely compact. Thank you for filling in one of the long-standing gaps in > > clang's cuda support story. > > I might need some hints on where to start. How would you go about testing > this, or are there any tests I can start from? I cribbed heavily from your > texture code, which only has a compile test of the headers.
Ideally it would be great to see that each builtin call generates correct surface instruction, and that those instructions are accepted by ptxas. The catch is that `ptxas` is part of CUDA SDK and we can't use it for testing unconditionally. We've plumbed it through as an optional testing tool on the LLVM side, but I do not think we have it on clang side at the moment. If the header is compileable without CUDA SDK (maybe with some stub headers in tests Inputs), then a source file with a lot of builtin calls and [autogenerated checks](https://github.com/llvm/llvm-project/blob/d724bab8064685c98cdded88157b6e2245e0d7bc/llvm/utils/update_cc_test_checks.py#L2) to verify produced instructions would do. https://github.com/llvm/llvm-project/pull/132883 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
