f93fe9bfd9
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 .
6 lines
231 B
Plaintext
6 lines
231 B
Plaintext
nccl (${nccl:Major}.${nccl:Minor}.${nccl:Patch}${nccl:Suffix}-${pkg:Revision}+cuda${cuda:Major}.${cuda:Minor}) trusty; urgency=medium
|
|
|
|
* Automatic Debian package from build
|
|
|
|
-- cudatools <cudatools@nvidia.com> ${pkg:Timestamp}
|