Skip to content

[ROCm] Limit number of values per thread for reductions on three dimensions #2460

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

Merged
merged 1 commit into from
Aug 6, 2025
Merged

[ROCm] Limit number of values per thread for reductions on three dimensions #2460

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
15 changes: 14 additions & 1 deletion aten/src/ATen/native/cuda/Reduce.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -209,6 +209,10 @@ struct ReduceConfig {
int values_per_thread() const {
return div_up(num_inputs, step_input);
}

int mock_values_per_thread(int parallelism) {
return div_up(num_inputs, step_input * parallelism);
}
};

std::ostream& operator<<(std::ostream& out, const ReduceConfig& config);
Expand Down Expand Up @@ -1166,8 +1170,17 @@ ReduceConfig setReduceConfig(const TensorIterator& iter){
else if (config.ctas_per_output < 16)
config.ctas_per_output = 1;
bool is_channel_last = iter.tensor_base(1).is_contiguous(at::MemoryFormat::ChannelsLast);
if (iter.ndim() == 3 && !reduction_on_fastest_striding_dimension && !is_channel_last)
if (iter.ndim() == 3 && !reduction_on_fastest_striding_dimension && !is_channel_last) {
config.ctas_per_output = 4;
int vpt = config.values_per_thread();
// Capping the number of values per thread to 2048 for now
// based on known use cases.
while (vpt >= 2048) {
config.ctas_per_output *= 2;
// Computes the new values per thread without side effects
vpt = config.mock_values_per_thread(config.ctas_per_output);
}
}
#endif
if (config.ctas_per_output > 1) {
config.input_mult[2] = config.split_input(config.ctas_per_output);
Expand Down