8000 Tags · crownk1997/nccl · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Tags: crownk1997/nccl

Tags

v2.5.7-1

Toggle v2.5.7-1's commit message 
2.5.7-1

v2.5.6-2

Toggle v2.5.6-2's commit message 
2.5.6-2

Fix PPC64 Debian packaging

v2.5.6-1

Toggle v2.5.6-1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Verified
Learn about vigilant mode 
2.5.6-1 (NVIDIA#255)

Add LL128 Protocol.

Rewrite the topology detection and tree/ring creation (NVIDIA#179). Improve
tree performance by sending/receiving from different GPUs. Add
model-based tuning to switch between the different algorithms and
protocols.

Rework P2P/SHM detection in containers (NVIDIA#155, NVIDIA#248).

Detect duplicated devices and return an error (NVIDIA#231).

Add tuning for GCP

v2.4.8-1

Toggle v2.4.8-1's commit message 
2.4.8-1

Fix NVIDIA#209: improve socket transport performance
  Split transfers over multiple sockets
  Launch multiple threads to drive sockets
  Detect AWS NICs and set nsockets/nthreads accordingly

v2.4.7-1

Toggle v2.4.7-1's commit message 
NCCL 2.4.7-1

    Performance tweaks for PowerPC builds only;
      Set default NCCL_MIN_NRINGS to 4
      Disable PCI-E NUMA distance detection

v2.4.6-1

Toggle v2.4.6-1's commit message 
NCCL 2.4.6-1

    Added detection of IBM/Power NVLink bridge device.
    Add NUMA support to PCI distance calculations.
    Added NCCL_IGNORE_CPU_AFFINITY env var.
    Fix memory leaks; GithubIssue#180
    Compiler warning fix; GithubIssue#178
    Replace non-standard variable length arrays. GithubIssue#171
    Fix Tree+Shared Memory crash. GithubPR#185
    Fix LL cleanup hang during long running DL jobs.
    Fix NCCL_RINGS environment variable handling.
    Added extra checks to catch repeat calls to ncclCommDestroy() GithubIssue#191
    Improve bootstrap socket connection reliability at scale.
    Fix hostname hashing issue. GithubIssue#187
    Code cleanup to rename all non device files from *.cu to *.cc

v2.4.2-1

Toggle v2.4.2-1's commit message 
2.4.2-1

Add tree algorithms for allreduce to improve performance at scale.
Add ncclCommAbort() and ncclCommGetAsyncError() to properly handle
network errors and be permit recover.
Detect initial CPU affinity and no longer escape it.

v2.3.7-1

Toggle v2.3.7-1's commit message 
2.3.7-1

Improved LL tuning for multi-node jobs.
Improved bootstrap for large job scaling.
Fixed a hang during bootstrap due to socket reuse.
Added operation name to the COLL INFO logging.

v2.3.5-5

Toggle v2.3.5-5's commit message 
2.3.5-5

Add support for inter-node communication using sockets and InfiniBand/RoCE.
Improve latency.
Add support for aggregation.
Improve LL/regular tuning.
Remove tests as those are now at github.com/nvidia/nccl-tests .

v1.3.4-1

Toggle v1.3.4-1's commit message 
Added Pascal nvcc flags, bumped version
0