Patronage & Donations
If you use the Slackware UK services on a regular basis, and would like to contribute to running costs, make a continuing patronage (and receive special benefits), or make a one off PayPal payment, please use one of the following buttons:
Patreon Logo
For regular monthly donations
PayPal Logo
For one time donations
All donations will be gratefully received - thank you!
Toggle theme:Toggle Theme Button
Content of directory's README

NCCL (pronounced "Nickel") is a stand-alone library of standard
communication routines for GPUs, implementing all-reduce, all-gather,
reduce, broadcast, reduce-scatter, as well as any send/receive based
communication pattern. It has been optimized to achieve high bandwidth
on platforms using PCIe, NVLink, NVswitch, as well as networking using
InfiniBand Verbs or TCP/IP sockets. NCCL supports an arbitrary number
of GPUs installed in a single node or across multiple nodes, and can
be used in either single- or multi-process (e.g., MPI) applications.

Although this SlackBuild REQUIRES="cudatoolkit_13", cudatoolkit_12 has
also been successfully tested and may be substituted if neccessary.
Other CUDA toolkit versions may also be suitable but have not been
tested and are not supported.

Building nccl requires one of the cudatoolkits to be installed. Since
none of the cudatoolkits support anything other than x86_64 systems,
only x86_64 system architecture is supported by this SlackBuild.

Icon  Name                    Last modified      Size  Description
[<--] Parent Directory - [!!!] README 2026-06-06 01:56 1.0K [SPT] nccl.SlackBuild 2026-09-26 02:09 3.6K [   ] nccl.info 2026-10-03 02:54 334 [TXT] slack-desc 2026-06-06 01:56 1.1K
Server (facade0) bandwidth utilisation
0  Bandwidth graph measuring   0.02 Gbit/s  10 Gbit/s