ofi/btl: Reuse MTL's domain and fabric in BTL #13415
Open
+79
−13
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
shijin-aws
reviewed
Sep 24, 2025
Share the domain among BTL and MTL to reduce the number of domains needed. This prevents hitting the resource limit on systems with high core counts. Signed-off-by: Jessie Yang <jiaxiyan@amazon.com>
MTL is initialized and finalized before BTL. When MTL and BTL share the same fabric and domain, BTL must be responsible for closing them to ensure the correct order of resource finalization. This commit adds flags to indicate whether the fabric and domain are shared with BTL, allowing MTL to skip closing these objects during its finalization. Signed-off-by: Jessie Yang <jiaxiyan@amazon.com>
shijin-aws
approved these changes
Sep 24, 2025
|
@bwbarrett @hppritcha Can you review this? |
Move cm before ob1 during pml selection to ensure MTL initialization occurs before BTL initialization. This addresses non-deterministic component ordering caused by filesystem discovery that could result in ob1 listed before cm. This allows BTL to reuse MTL's fabric and domain. Signed-off-by: Jessie Yang <jiaxiyan@amazon.com>
bwbarrett
requested changes
Sep 29, 2025
| #include "opal/mca/btl/btl.h" | ||
| #include "opal/mca/common/ofi/common_ofi.h" | ||
| #include "opal/mca/hwloc/base/base.h" | ||
| #include "ompi/mca/mtl/base/base.h" |
There was a problem hiding this comment.
This is an abstraction break. The BTL can't include MTL headers. So whatever you really need here needs to be logic that ends up in the common/ofi component.
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Outdated suggestions cannot be applied.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.
Share the domain among BTL and MTL to reduce the number of domains needed. This prevents hitting the resource limit on systems with high core counts.