Skip to content

Migrate linearize cache indices kernels to FBGEMM_LAUNCH_KERNEL #4768

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.

Sign up for GitHub

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: main
Choose a base branch
from
Open

Migrate linearize cache indices kernels to FBGEMM_LAUNCH_KERNEL #4768

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
38 changes: 14 additions & 24 deletions fbgemm_gpu/src/split_embeddings_cache/linearize_cache_indices.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,22 +105,17 @@ DLL_PUBLIC Tensor linearize_cache_indices_cuda(
using offset_t = index_t;
AT_DISPATCH_INDEX_TYPES(
indices.scalar_type(), "linearize_cache_indices_kernel_2", [&] {
#ifdef FBGEMM_GPU_MEMCHECK
const char* func_name = "linearize_cache_indices_kernel";
#endif
linearize_cache_indices_kernel<<<
FBGEMM_LAUNCH_KERNEL(
(linearize_cache_indices_kernel<index_t, offset_t>),
div_round_up(num_indices, kMaxThreads),
kMaxThreads,
0,
at::cuda::getCurrentCUDAStream()>>>(
MAKE_PTA_WITH_NAME(
func_name, cache_hash_size_cumsum, int64_t, 1, 32),
MAKE_PTA_WITH_NAME(func_name, indices, index_t, 1, 32),
MAKE_PTA_WITH_NAME(func_name, table_offsets, offset_t, 1, 32),
MAKE_PTA_WITH_NAME(
func_name, linear_cache_indices, int64_t, 1, 32),
at::cuda::getCurrentCUDAStream(),
PTA_B(cache_hash_size_cumsum, int64_t, 1, 32),
PTA_B(indices, index_t, 1, 32),
PTA_B(table_offsets, offset_t, 1, 32),
PTA_B(linear_cache_indices, int64_t, 1, 32),
indices_base_offset);
C10_CUDA_KERNEL_LAUNCH_CHECK();
});
});
return linear_cache_indices;
Expand Down Expand Up @@ -181,21 +176,16 @@ DLL_PUBLIC Tensor linearize_cache_indices_from_row_idx_cuda(
update_row_indices.scalar_type(),
"linearize_cache_indices_from_row_idx_kernel",
[&] {
#ifdef FBGEMM_GPU_MEMCHECK
const char* func_name = "linearize_cache_indices_from_row_idx_kernel";
#endif
linearize_cache_indices_from_row_idx_kernel<<<
FBGEMM_LAUNCH_KERNEL(
(linearize_cache_indices_from_row_idx_kernel<index_t>),
div_round_up(num_indices, kMaxThreads),
kMaxThreads,
0,
at::cuda::getCurrentCUDAStream()>>>(
MAKE_PTA_WITH_NAME(
func_name, cache_hash_size_cumsum, int64_t, 1, 32),
MAKE_PTA_WITH_NAME(func_name, update_table_indices, index_t, 1, 32),
MAKE_PTA_WITH_NAME(func_name, update_row_indices, index_t, 1, 32),
MAKE_PTA_WITH_NAME(
func_name, linear_cache_indices, index_t, 1, 32));
C10_CUDA_KERNEL_LAUNCH_CHECK();
at::cuda::getCurrentCUDAStream(),
PTA_B(cache_hash_size_cumsum, int64_t, 1, 32),
PTA_B(update_table_indices, index_t, 1, 32),
PTA_B(update_row_indices, index_t, 1, 32),
PTA_B(linear_cache_indices, index_t, 1, 32));
});
return linear_cache_indices;
}
Expand Down
Loading