[v5.0.x] ofi: follow user specified include/exclude list to select providers #12251
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
This PR addresses open-mpi#12233 Since 5.0.x, we introduced an optional FI_HMEM capability in ofi provider selection logic(both mtl and btl) in order to support accelerator memory. As described in the issue, this introduced a bug that can cause the wrong ofi provider to be selected, even if the user explicitly includes/excludes the provider name. This change refactors the selection logic to correctly handle the include/exclude list, and therefore fixes the bug. Signed-off-by: Wenduo Wang <[email protected]> (cherry picked from commit 29efcef)
|
why don't you backport the FI_REMOTE_COMM addition? |
@shijin-aws Added thanks for reminder |
hppritcha
approved these changes
Jan 19, 2024
shijin-aws
approved these changes
Jan 19, 2024
wenduwan
force-pushed
the
5_backport_pr12234
branch
from
9771c19 to
75e726c
Compare
|
@hppritcha @shijin-aws I removed FI_REMOTE_COMM commit due to #12255 |
|
@wenduwan giving it a look now! Thanks |
|
@wenduwan Verified this change! OPX provider is found without issue, I tested using OSU MB's, and with the following options, which are typically preferred for OPX runs |
|
@tmh97 Thank you! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This PR addresses #12233
Since 5.0.x, we introduced an optional FI_HMEM capability in ofi provider selection logic(both mtl and btl) in order to support accelerator memory. As described in the issue, this introduced a bug that can cause the wrong ofi provider to be selected, even if the user explicitly includes/excludes the provider name.
This change refactors the selection logic to correctly handle the include/exclude list, and therefore fixes the bug.
Signed-off-by: Wenduo Wang [email protected]
(cherry picked from commit 29efcef)