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

2.21.5-1

Browse Source 
Add support for IB SHARP 1PPN operation with user buffers.
Improve support for MNNVL, add NVLS support and multi-clique support.
 * Detect the NVLS clique through NVML
 * Exchange XML between peers in the same NVLS clique and fuse XMLs
   before creating the topology graph.
 * Rework bootstrap allgather algorithms to allow for large allgather
   operations intra-node (XML exchange).
Net/IB: add support for dynamic GID detection.
 * Automatically select RoCEv2/IPv4 interface by default. Allow to
   select IPv6 or even the network/mask.
Reduce NVLS memory usage.
 * Add stepSize as property of a connection to allow for different
   sizes on different peers; set it to 128K for NVLink SHARP.
Improve tuner loading
 * Look for more paths, be more consistent with the network device
   plugin.
 * Also search for tuner support inside the net plugin.
Improve tuner API
 * Add context to support multi-device per process.
Add magic number around comm object to detect comm corruption.
 * Add some basic check around communicators so that we can report a
   problem when a communicator gets corrupted or a wrong comm pointer
   is passed to NCCL.
Fix net/IB error path. Github PR #1164
Fix collnet rail mapping with split comm.
Fix packet reordering issue causing bootstrap mismatch
 * Use a different tag in ncclTransportP2pSetup for the connectInfo
   exchange and the following barrier.
Fix hang when crossNic is inconsistent between ranks.
Fix minCompCap/maxCompCap computation. Github issue #1184
This commit is contained in:
Sylvain Jeaugey
2024-03-26 06:08:55 -07:00
parent 6dd51f15bf
commit ab2b89c4c3
52 changed files with 2124 additions and 944 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/device/common.h
+8 -6
View File
@@ -18,19 +18,21 @@ typedef void(*ncclDevFuncPtr_t)();
extern __device__ ncclDevFuncPtr_t const ncclDevFuncTable[];
struct ncclShmemGroup {
ncclConnInfo *recvConns[NCCL_MAX_NVLS_ARITY];
ncclConnInfo *sendConns[NCCL_MAX_NVLS_ARITY];
void* srcs[NCCL_MAX_NVLS_ARITY+1];
void* dsts[NCCL_MAX_NVLS_ARITY+1];
ncclConnInfo *recvConns[NCCL_MAX_ARITY];
ncclConnInfo *sendConns[NCCL_MAX_ARITY];
void* userInput;
void* userOutput;
void* srcs[NCCL_MAX_ARITY+1];
void* dsts[NCCL_MAX_ARITY+1];
union {
unpackGroupShmem unpack;
} devicePlugin;
int32_t dstSizes[NCCL_MAX_NVLS_ARITY+1];
int32_t dstSizes[NCCL_MAX_ARITY+1];
};
struct ncclShmemData {
struct ncclShmemGroup groups[NCCL_MAX_GROUPS];
uint64_t redOpArgs[NCCL_MAX_NVLS_ARITY+1];
uint64_t redOpArgs[NCCL_MAX_ARITY+1];
int channelId;
int aborted;
alignas(16) struct ncclDevComm comm;