JonChesterfield added a comment.
I'm happy with this as-is. @jdoerfert is this close enough to what you expected
when we discussed this offline?
Adding it directly to AMDGPU.h was a good suggestion. Makes it easy for other
amdgpu language drivers to pick it up.
That it's a binary on disk also means that more DIY build systems, e.g.
freestanding C++ stuff, can call it directly, in `-march=$(./amdgpu-arch)`
fashion.
================
Comment at: clang/tools/amdgpu-arch/AMDGPUArch.cpp:51
+
+ for (unsigned I = 0; I < GPUs.size(); I++) {
+ printf("%s\n", GPUs[I].c_str());
----------------
This strategy looks good. Briefly considered whether we should print while
iterating, but that will misbehave if there is an error after printing the
first gpu.
unsigned is strictly the wrong type here, though it doesn't matter in practice.
Could go with a range based for loop instead.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D99949/new/
https://reviews.llvm.org/D99949
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
