Branch: refs/heads/main
Revision: 268935d53f2d87358c9ecaffb3f05cd21d92b8b9
Author: jabraham17
Link: Prefix computed GPU version macros by jabraham17 · Pull Request #28824 · chapel-lang/chapel · GitHub
Log Message:
Prefix computed GPU version macros (#28824)
Adds a CHPL_ prefix to computed GPU sdk version variables, to avoid
conflicting with the ones predefined by the library
Another solution would be to just use the builtin version variables, but
this is more consistent with what we have to do for ROCM (which doesn't
have builtin version variables like CUDA does)
Followup to Fix deprecation warnings with cuda 12.9 by jabraham17 · Pull Request #28795 · chapel-lang/chapel · GitHub
[Reviewed by @arifthpe]
Diff:
M runtime/include/gpu/nvidia/chpl-gpu-dev-reduce.h
M runtime/src/gpu/amd/gpu-amd.c
M util/chplenv/chpl_gpu.py
https://github.com/chapel-lang/chapel/pull/28824.diff