f6e1e3d9ed
Add support for improved fault tolerance: non-blocking mode, new
init function with config, and ncclCommFinalize function.
Reintroduce collnet+chain algorithm, alongside collnet+direct.
Add LL protocol for intra-node P2P (on by default) and network
communication (off by default).
Use network instead of shared memory when performance is better.
Fix: wait for CUDA graph destroy before destroying comm with linked
graph resources.
Remove aggressive polling during enqueue.
Fix DMABUF fallback on MOFED 5.4 and earlier.
[ROCm/rccl commit: c4e2aa6c79]
7 lines
103 B
Makefile
7 lines
103 B
Makefile
##### version
|
|
NCCL_MAJOR := 2
|
|
NCCL_MINOR := 14
|
|
NCCL_PATCH := 3
|
|
NCCL_SUFFIX :=
|
|
PKG_REVISION := 1
|