Files

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

31 wiersze
1.2 KiB
Plaintext

2016-06-15 16:54:57 -07:00
Source: nccl
Section: libs
Maintainer: cudatools <cudatools@nvidia.com>
Priority: optional
Build-depends: debhelper(>=9)
Standards-Version: 3.9.5
Package: libnccl${nccl:Major}
Section: libs
2018-09-24 16:06:59 -07:00
Architecture: ${pkg:Arch}
2016-06-15 16:54:57 -07:00
Depends: ${misc:Depends}, ${shlibs:Depends}
2020-09-04 14:35:05 -07:00
Description: NVIDIA Collective Communication Library (NCCL) Runtime
2016-06-15 16:54:57 -07:00
NCCL (pronounced "Nickel") is a stand-alone library of standard collective
2018-09-24 16:06:59 -07:00
communication routines for GPUs, implementing all-reduce, all-gather, reduce,
broadcast, and reduce-scatter.
It has been optimized to achieve high bandwidth on any platform using PCIe,
NVLink, NVswitch, as well as networking using InfiniBand Verbs or TCP/IP
sockets.
2016-06-15 16:54:57 -07:00
Package: libnccl-dev
Section: libdevel
2018-09-24 16:06:59 -07:00
Architecture: ${pkg:Arch}
2016-06-15 16:54:57 -07:00
Depends: ${misc:Depends}, ${shlibs:Depends}, libnccl${nccl:Major} (= ${binary:Version})
2020-09-04 14:35:05 -07:00
Description: NVIDIA Collective Communication Library (NCCL) Development Files
2016-06-15 16:54:57 -07:00
NCCL (pronounced "Nickel") is a stand-alone library of standard collective
2018-09-24 16:06:59 -07:00
communication routines for GPUs, implementing all-reduce, all-gather, reduce,
broadcast, and reduce-scatter.
It has been optimized to achieve high bandwidth on any platform using PCIe,
NVLink, NVswitch, as well as networking using InfiniBand Verbs or TCP/IP
sockets.