[torch.compile] register allreduce operations as custom ops (#8526)

This commit is contained in:
youkaichao
2024-09-16 22:57:57 -07:00
committed by GitHub
parent ee2bceaaa6
commit 99aa4eddaf
9 changed files with 137 additions and 50 deletions
-2
View File
@@ -241,8 +241,6 @@ fptr_t init_custom_ar(torch::Tensor& meta, torch::Tensor& rank_data,
const std::vector<std::string>& handles,
const std::vector<int64_t>& offsets, int64_t rank,
bool full_nvlink);
bool should_custom_ar(torch::Tensor& inp, int64_t max_size, int64_t world_size,
bool full_nvlink);
void all_reduce_reg(fptr_t _fa, torch::Tensor& inp, torch::Tensor& out);
void all_reduce_unreg(fptr_t _fa, torch::Tensor& inp, torch::Tensor& reg_buffer,
torch::Tensor& out);