AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update Barrier and Sync APIs (#73)

Browse Source 
* Add thread, wavefront, and workgroup-level `barrier` APIs in IPC and RO conduits; remove collectives on default context
 - Implemented `barrier` APIs for thread, wavefront, and workgroup scopes
 - Added support into both IPC and RO conduits
 - Added functional tests to cover all `barrier` APIs
 - Removed collective operations on default context

* Add thread, wavefront, and workgroup-level `sync` APIs in IPC and RO conduits.
  - Implemented `sync` APIs for thread, wavefront, and workgroup scopes
  - Added support into both IPC and RO conduits
  - Added functional tests to cover all `sync` APIs

* update naming convention for context-based `barrier` APIs
This commit is contained in:
Avinash Kethineedi
2025-04-08 11:25:31 -05:00
committed by GitHub
parent c652f58cef
commit dc61bca066
16 changed files with 347 additions and 67 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/ipc/context_ipc_device.hpp
+8
View File
@@ -67,12 +67,20 @@ class IPCContext : public Context {
__device__ void barrier(rocshmem_team_t team);
__device__ void barrier_wave(rocshmem_team_t team);
__device__ void barrier_wg(rocshmem_team_t team);
__device__ void sync_all();
__device__ void sync_all_wave();
__device__ void sync_all_wg();
__device__ void sync(rocshmem_team_t team);
__device__ void sync_wave(rocshmem_team_t team);
__device__ void sync_wg(rocshmem_team_t team);
template <typename T>