================ @@ -0,0 +1,56 @@ +// RUN: %clang_cc1 -triple nvptx64-unknown-unknown -target-cpu sm_61 \ +// RUN: -fcuda-is-device -target-feature +ptx62 \ +// RUN: -emit-llvm -o - -x cuda %s | FileCheck %s +// Make sure the builtins still work with newer GPUs. ---------------- ilya-biryukov wrote:
I agree, the two groups are: - "works on new archs" and - "does not work on old archs". They are now explicit and the comment highlights the use of CHECK in one and expected-error in the other. https://github.com/llvm/llvm-project/pull/233434 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
