jhuber6 wrote: > > This is not being handled for AMDGPU Targets. > > I'm assuming this is an artifact of passing all arguments both the host > target and the offload target? @jhuber6 what's the correct way of filtering > out irrelevant codegen options?
I don't know what the desired behavior is. The `Triple` being used here should match the one from the ToolChain. So, it should be `amdgpu` when doing the device side compilation. If we need to change behavior based off of offloading language, you check the JobAction for its `OffloadKind`. https://github.com/llvm/llvm-project/pull/70262 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits