jhuber6 wrote: This is a draft, I'm trying to get this to map to either `s_memtime` or `s_sendmsg_rtn(0x83)` depending on the target for AMDGPU. Currently, it will recognize the new intrinsic but just lower it to `i64 0`. I'm not overly familiar with the backend, so any suggestions would be greatly appreciated.
https://github.com/llvm/llvm-project/pull/81331 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
