Skip to content

fix deepep init / finalize #7

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
Eddie-Wang1120 wants to merge 2 commits into
base: paddle
Choose a base branch
from
+30 −2
Open

fix deepep init / finalize #7

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
91935c5
fix deepep init / finalize
Eddie-Wang1120 Jan 27, 2026
f33d8a3
fix
Eddie-Wang1120 Feb 2, 2026
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
32 changes: 30 additions & 2 deletions csrc/kernels/runtime.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,26 @@ namespace internode {
nvshmem_team_t cpu_rdma_team = NVSHMEM_TEAM_INVALID;
nvshmem_team_config_t cpu_rdma_team_config;

#ifndef GLOBALS_H
#define GLOBALS_H

class GlobalState {
public:
static GlobalState& instance() {
static GlobalState inst;
return inst;
}

int64_t counter;

private:
GlobalState() : counter(0) {}
GlobalState(const GlobalState&) = delete;
GlobalState& operator=(const GlobalState&) = delete;
};

#endif

std::vector<uint8_t> get_unique_id() {
nvshmemx_uniqueid_t unique_id;
nvshmemx_get_uniqueid(&unique_id);
Expand All @@ -53,6 +73,9 @@ int init(const std::vector<uint8_t>& root_unique_id_val, int rank, int num_ranks
nvshmemx_set_attr_uniqueid_args(rank, num_ranks, &root_unique_id, &attr);
nvshmemx_init_attr(NVSHMEMX_INIT_WITH_UNIQUEID, &attr);

// Initialize before nvshmem_team_split_strided
nvshmem_barrier_all();

// Create sub-RDMA teams
// NOTES: if `num_ranks <= NUM_MAX_NVL_PEERS` then only low-latency kernels are used
if (low_latency_mode and num_ranks > NUM_MAX_NVL_PEERS) {
Expand All @@ -68,7 +91,7 @@ int init(const std::vector<uint8_t>& root_unique_id_val, int rank, int num_ranks
EP_HOST_ASSERT(cpu_rdma_team != NVSHMEM_TEAM_INVALID);
}

nvshmem_barrier_all();
// nvshmem_barrier_all();
return nvshmem_my_pe();
}

Expand All @@ -89,7 +112,12 @@ void finalize() {
nvshmem_team_destroy(cpu_rdma_team);
cpu_rdma_team = NVSHMEM_TEAM_INVALID;
}
nvshmem_finalize();

GlobalState::instance().counter++;
if (GlobalState::instance().counter > 1) {
nvshmem_finalize();
}
// nvshmem_finalize();
}
#endif

Expand Down