Add support for IMPI Thread-Split mode using OpenMP #45
Conversation
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
Author
|
No need to merge yet, but sharing code here. |
Owner
|
@stwhite91 when MPI backend create comms thread according to openmp threads, does it be active when communication is not happening? I mean will it interfere with compute ops before and after communican, which also use openmp threads? |
Author
|
Sorry I didn't see your comment earlier. No the OpenMP threads I have added here are not active outside of the mpi_all_reduce, there is no problem using OpenMP threads elsewhere in the application. |
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Outdated suggestions cannot be applied.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.
Add BF16 and Float16 datatype support.
Add support for duplicating communicators for each OpenMP thread.
Use OpenMP inside mpi_all_reduce to split msg across threads/comms.
If HANDOFF is defined to 1 at top of coll_mpi.cpp, do not use thread-split.