Skip to content

Add ROCm 6.4.3+ support #498

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 9 additions & 2 deletions csrc/cuda/utils.cuh
Original file line number Diff line number Diff line change
Expand Up @@ -6,13 +6,20 @@
AT_ASSERTM(x.device().is_cuda(), #x " must be CUDA tensor")
#define CHECK_INPUT(x) AT_ASSERTM(x, "Input mismatch")

__device__ __inline__ at::Half __shfl_up_sync(const unsigned mask,
// On ROCm, __shfl_*_sync requires a 64-bit mask; on CUDA it's 32-bit.
#ifdef USE_ROCM
using warp_mask_t = unsigned long long;
#else
using warp_mask_t = unsigned int;
#endif

__device__ __inline__ at::Half __shfl_up_sync(const warp_mask_t mask,
const at::Half var,
const unsigned int delta) {
return __shfl_up_sync(mask, var.operator __half(), delta);
}

__device__ __inline__ at::Half __shfl_down_sync(const unsigned mask,
__device__ __inline__ at::Half __shfl_down_sync(const warp_mask_t mask,
const at::Half var,
const unsigned int delta) {
return __shfl_down_sync(mask, var.operator __half(), delta);
Expand Down
Loading