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

2.20.3-1

Browse Source 
Add support for alternating rings, allow for cross-nic rings without
cross-rail communication.
Add support for user buffer registration for network send/recv.
Optimize aggregated operations to better utilize all channels.
Add flattening for BCM PCI gen5 switches.
Add support for inter-node NVLink communication
Add support for port fusion in NET/IB.
Add support for ReduceScatter and AllGather using Collnet.
Update net API to v8.
Fix hang during A2A connection.
This commit is contained in:
Sylvain Jeaugey
2024-02-05 05:06:02 -08:00
parent b6d7438d31
commit b6475625fb
74 changed files with 4632 additions and 2165 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/include/enqueue.h
+4 -2
View File
@@ -12,8 +12,10 @@
#include "collectives.h"
#include "utils.h"
#define NCCL_MIN_CHANNEL_SIZE (NCCL_LL_THREAD_THRESHOLD*64)
#define NCCL_AGG_CHANNEL_SIZE (1LL << 21) /* 2 MiB, ideal per-channel size to fully utilize bandwidth */
#define NCCL_LL_ALIGNMENT_PER_THREAD sizeof(uint64_t)
#define NCCL_LL128_ALIGNMENT_PER_WARP 480
#define NCCL_SIMPLE_ALIGNMENT (WARP_SIZE * 8LL * 16LL)
#define NCCL_BYTES_ALIGNMENT 16
ncclResult_t ncclInitKernelsForDevice(int cudaArch, size_t* maxStackSize);
ncclResult_t ncclEnqueueCheck(struct ncclInfo* info);