[Runtime] Introduce MSCCLPP with NCCL equivalent interface - #16804
Merged
tqchen merged 2 commits intoMar 29, 2024
Merged
Conversation
* Add a fast and simple AllReduce kernel (sum only) using using mscclpp smChannel scratch for small reductions up to 2**24 bytes.
csullivan
force-pushed
the
feature/2024-03-20/mscclpp-allreduce
branch
4 times, most recently
from
March 28, 2024 18:36
471fb9a to
b78c1c3
Compare
csullivan
force-pushed
the
feature/2024-03-20/mscclpp-allreduce
branch
from
March 28, 2024 21:05
b78c1c3 to
e2222d6
Compare
tqchen
approved these changes
Mar 29, 2024
Member
|
Thanks @csullivan . looking forward to the disco integration |
Ubospica
pushed a commit
to Ubospica/tvm-develop
that referenced
this pull request
Apr 18, 2024
…pache#16804)" This reverts commit 64db9f7.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Currently only supports ncclAllreduce equivalent, with op=Sum up to 2**24 bytes as noted below.
Plan to integrate with disco in a follow up PR.