Skip to content
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

Jump to bottom

Enable OMP multithreading in lookup_table_v2 #45249

Merged
merged 9 commits into from
Aug 25, 2022
Merged

Enable OMP multithreading in lookup_table_v2 #45249

Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
f4e4f8d
Add omp parallel for directives
piotrekobi Aug 16, 2022
ed2bf76
Revert "Add omp parallel for directives"
piotrekobi Aug 18, 2022
acb1d88
Add #pragma omp parallel for to correct file
piotrekobi Aug 18, 2022
c05f3ca
Add check for _OPENMP definition
piotrekobi Aug 19, 2022
c1b4ee8
Disable omp on gpu
piotrekobi Aug 19, 2022
2073816
Trigger CI
piotrekobi Aug 19, 2022
8218038
Readd check for _OPENMP definition
piotrekobi Aug 19, 2022
465aadf
Change macro disabling changes on GPU
piotrekobi Aug 22, 2022
99392d5
Improve macro readability
piotrekobi Aug 22, 2022
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
4 changes: 4 additions & 0 deletions paddle/phi/kernels/cpu/embedding_kernel.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,10 @@ struct EmbeddingCPUFunctor {
dev_ctx_.template Alloc<T>(out_);
auto* output = out_->data<T>();

#if defined(_OPENMP) && !defined(PADDLE_WITH_CUDA)
#pragma omp parallel for
#endif

for (int64_t i = 0; i < ids_numel; ++i) {
if (padding_idx_ != kNoPadding && ids[i] == padding_idx_) {
memset(output + i * row_width, 0, row_width * sizeof(T));
Expand Down