mirror of
https://github.com/deepseek-ai/DeepEP
synced 2025-06-26 18:28:11 +00:00
Update deep_ep.cpp
This commit is contained in:
@@ -614,6 +614,8 @@ Buffer::internode_dispatch(const torch::Tensor& x, const std::optional<torch::Te
|
|||||||
const std::optional<torch::Tensor>& cached_rdma_channel_prefix_matrix, const std::optional<torch::Tensor>& cached_recv_rdma_rank_prefix_sum,
|
const std::optional<torch::Tensor>& cached_rdma_channel_prefix_matrix, const std::optional<torch::Tensor>& cached_recv_rdma_rank_prefix_sum,
|
||||||
const std::optional<torch::Tensor>& cached_gbl_channel_prefix_matrix, const std::optional<torch::Tensor>& cached_recv_gbl_rank_prefix_sum,
|
const std::optional<torch::Tensor>& cached_gbl_channel_prefix_matrix, const std::optional<torch::Tensor>& cached_recv_gbl_rank_prefix_sum,
|
||||||
int expert_alignment, const Config& config, std::optional<EventHandle>& previous_event, bool async, bool allocate_on_comm_stream) {
|
int expert_alignment, const Config& config, std::optional<EventHandle>& previous_event, bool async, bool allocate_on_comm_stream) {
|
||||||
|
pybind11::gil_scoped_release release;
|
||||||
|
|
||||||
const int num_channels = config.num_sms / 2;
|
const int num_channels = config.num_sms / 2;
|
||||||
EP_HOST_ASSERT(config.num_sms % 2 == 0);
|
EP_HOST_ASSERT(config.num_sms % 2 == 0);
|
||||||
EP_HOST_ASSERT(0 < get_num_rdma_ranks() and get_num_rdma_ranks() <= NUM_MAX_RDMA_PEERS);
|
EP_HOST_ASSERT(0 < get_num_rdma_ranks() and get_num_rdma_ranks() <= NUM_MAX_RDMA_PEERS);
|
||||||
|
|||||||
Reference in New Issue
Block a user