Commit Graph

2 Commits

Author SHA1 Message Date
Nusrat Islam 5545392913 ext-src: Fix compiler warnings for MSCCLPP integration (#1368)
[ROCm/rccl commit: 6160603d4c]
2024-10-10 08:20:02 -05:00
Nusrat Islam f61053dcba Add a custom allreduce algorithm in MSCCLPP for cpx mode (#1362)
* cmake: remove mscclpp patch after build is complete

To enable mscclpp in cpx mode, a patch cpx.patch needs to be applied.
This patch can be removed after building is done. This helps with the
build process the following time.

* Use read-based mscclpp allreduce from rccl

MSCCLPP by default uses remote write in the allreduce kernel for
large (> 1MB) messages. This PR adds an allreduce kernel that uses
remote read. It needs the users to use an environment variable
MSCCLPP_READ_ALLRED=1.

[ROCm/rccl commit: 4d68751ce1]
2024-10-08 14:42:12 -05:00