Skip to content

Fix CUDA all-reduce planning for large inputs #3603

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
sofinvalery wants to merge 1 commit into
base: main
Choose a base branch
from
+6 −6
Open

Fix CUDA all-reduce planning for large inputs #3603

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
12 changes: 6 additions & 6 deletions mlx/backend/cuda/reduce/all_reduce.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,12 +68,12 @@ void all_reduce(

out.set_data(cu::malloc_async(out.nbytes(), encoder));

auto get_args = [](int size, int N) {
int threads = std::min(512, (size + N - 1) / N);
auto get_args = [](size_t size, size_t N) {
int threads =
static_cast<int>(std::min(size_t{512}, cuda::ceil_div(size, N)));
threads = ((threads + WARP_SIZE - 1) / WARP_SIZE) * WARP_SIZE;
int reductions_per_step = threads * N;
size_t steps_needed =
(size + reductions_per_step - 1) / reductions_per_step;
size_t reductions_per_step = threads * N;
size_t steps_needed = cuda::ceil_div(size, reductions_per_step);

int blocks;
if (steps_needed < 32) {
Expand All @@ -88,7 +88,7 @@ void all_reduce(
blocks = 1024;
}

size_t steps_per_block = (steps_needed + blocks - 1) / blocks;
size_t steps_per_block = cuda::ceil_div(steps_needed, blocks);
size_t block_step = steps_per_block * reductions_per_step;

return std::make_tuple(blocks, threads, block_step);
Expand Down
Loading