forked from pool/openucx
Nicolas Morey
c9995295b3
- RDMA CORE (IB, ROCE, etc.)
- Added UCX_IB_GDA_RETAIN_INACTIVE_CTX option to control inactive CUDA context retention in GDAKI
- Added configure option to enable or disable GGA transport
- Fixed GDA DMA-BUF offset handling
- Fixed GDA WQE ordering by using CAS-based readiness marking
- Fixed GDAKI CUDA context handling during endpoint creation
- Fixed GDAKI NIC/GPU mapping when CUDA_VISIBLE_DEVICES hides physical GPUs
- ROCM
- Fixed hangs in HIP MPI and OMB tests
- Fixed endpoint flush for in-flight ROCm operations
- TCP
- Fixed interface selection by skipping IPv4 link-local addresses
- UCP
- Added UCX_PROTO_EMULATION_ENABLE option to force zero-copy RMA protocol selection
- Added UCX_MAX_HCA_PER_GPU policy to limit GPU memory registrations to nearest HCAs
- Added device lanes that can access host memory for GPU transfer fallback
- Enabled gdr_copy for memtype endpoint transport
- Fixed progress counter to return the actual operation status
- Fixed multi protocol minimum size handling for 1-byte operations
- Fixed endpoint finalization when no P2P or connection-manager lane is available
- UCS
- Reverted dynamically loaded external module/plugin support
- UCT
- Added device channel pool support
- Added CPU memory usage as AMO local buffer for device operations
- Fixed notify callback handling by adding a NULL check
Signed-off-by: Nicolas Morey <nmorey@suse.com>
Build Results
Current state of openucx in openSUSE:Factory is
The current state of openucx in the devel project build (science:HPC)