Releases: ROCm/rocSHMEM
Releases · ROCm/rocSHMEM
rocm-7.2.2
ROCm release v7.2.2
rocm-7.2.1
ROCm release v7.2.1
ROCm release v7.2.0 (rocSHMEM v3.2.0)
rocSHMEM 3.2.0 for ROCm 7.2.0
Added
- Added the GDA conduit for AMD Pensando IONIC
Changed
- Dependency libraries are loaded dynamically
- The following APIs now have an implementation for the GDA conduit
rocshmem_p- fetching atomics
rochsmem_<TYPE>_fetch_<op> - collective APIs
- The following APIs now have an implementation for the IPC conduit
rocshmem_<TYPE>_atomic_{and,or,xor,swap}rocshmem_<TYPE>_atomic_fetch_{and,or,xor,swap}
Known issues
- Only 64bit rocSHMEM atomic APIs are implemented for the GDA conduit
ROCm v7.1.1 (rocSHMEM 3.1.0)
ROCm release v7.1.1
rocSHMEM 3.1.0
Added
- Allowed IPC, RO, and GDA backends to be selected at runtime.
- GDA (GPUDirect Async) conduit for different NIC vendors:
- Broadcom BNXT_RE (Thor 2)
- Mellanox MLX5 (IB and RoCE ConnectX-7)
- New APIs:
rocshmem_get_device_ctx
Changed
- The following APIs have been deprecated:
rocshmem_wg_initrocshmem_wg_finalizerocshmem_wg_init_thread
rocshmem_ptrcan now return non-null pointer to a shared memory region when the IPC transport is available to reach that region. Previously, it would return a null pointer.ROCSHMEM_RO_DISABLE_IPCis renamed toROCSHMEM_DISABLE_MIXED_IPC.- This environment variable wasn’t documented in earlier releases. It’s now documented.
Removed
- rocSHMEM no longer requires rocPRIM and rocThrust as dependencies.
- Removed MPI compile-time dependency.
Known issues
- Only a subset of rocSHMEM APIs are implemented for the GDA conduit.
rocm-7.1.0
ROCm release v7.1.0
rocm-7.0.2
ROCm release v7.0.2
rocm-6.4.4
ROCm release v6.4.4
rocm-7.0.1
ROCm release v7.0.1
rocm-7.0.0
ROCm release v7.0.0
rocSHMEM 3.0.0 for ROCm 7.0.0
Added
- Added the Reverse Offload conduit
- Added new APIs:
rocshmem_ctx_barrierrocshmem_ctx_barrier_waverocshmem_ctx_barrier_wgrocshmem_barrier_allrocshmem_barrier_all_waverocshmem_barrier_all_wgrocshmem_ctx_syncrocshmem_ctx_sync_waverocshmem_ctx_sync_wgrocshmem_sync_allrocshmem_sync_all_waverocshmem_sync_all_wgrocshmem_init_attrrocshmem_get_uniqueidrocshmem_set_attr_uniqueid_args
- Added dlmalloc based allocator
- Added XNACK support
- Added support for initialization with MPI communicators other than
MPI_COMM_WORLD
Changed
- Changed collective APIs to use
_wgsuffix rather than_wg_infix
Resolved Issues
- Resolved segfault in
rocshmem_wg_ctx_create, now provides nullptr if ctx cannot be created
rocm-6.4.3
ROCm release v6.4.3