[libc-commits] [libc] [libc] [gpu] Add Generic and NvSin Throughput Benchmark (PR #101917)
Joseph Huber via libc-commits
libc-commits at lists.llvm.org
Sun Aug 4 19:25:13 PDT 2024
jhuber6 wrote:
> > Ah, I see you added new ones called `latency` as well.
>
> Yeah, I think the original intent of changing the constraints AMDGPU was that I realized using the pointer might be possible instead of having to do that `FORCE_TO_REGISTER()` workaround before - should we just keep that part in?
I'll need to double check the ASM generated there, I forget if capturing the pointer prevented it from carrying the input in a register.
https://github.com/llvm/llvm-project/pull/101917
More information about the libc-commits
mailing list