[compiler-rt] [compiler-rt] Initial support for builtins on GPU targets (PR #95304)
Joseph Huber via llvm-commits
llvm-commits at lists.llvm.org
Thu Jun 13 09:10:44 PDT 2024
jhuber6 wrote:
> > @compnerd is there a canonical way to build for multiple architectures at the same time? I.e. given a standard CMake invocation I'd like to be able to build output for x64, amdgcn, and nvptx64 if they are all supported by the compiler.
>
> Invoke CMake multiple times :) CMake is designed to build for a single host at a time, so you would need to run a build per host.
Okay, that tracks with the runtimes option I'm planning on using. The case that @yxsamliu is describing can probably be done with `-DCOMPILER_RT_DEFAULT_TARGET=amdgcn-amd-amdhsa` instead of a dedicated flat, correct?
https://github.com/llvm/llvm-project/pull/95304
More information about the llvm-commits
mailing list