Fix highest capacity memspace with OS provider integration. #390
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
igchor
force-pushed
the
memspace_highest_capacity_alloc
branch
5 times, most recently
from
416c8f0 to
e5926d8
Compare
bratpiorka
approved these changes
Mar 26, 2024
igchor
force-pushed
the
memspace_highest_capacity_alloc
branch
2 times, most recently
from
be8cf0a to
b2d3822
Compare
kswiecicki
approved these changes
Mar 26, 2024
The test was using get_mempolicy to verify whether allocation was done on a specific node but params to get_mem_policy were wrong. Instead of returning the nodeId, get_mempolicy was returning a policy associated with the address which was always: all nodes set.
when creating provider from highest capacity memspace
igchor
force-pushed
the
memspace_highest_capacity_alloc
branch
from
b2d3822 to
2a2680e
Compare
Id stored in memory target is physcial, not logical id so use hwloc_get_numanode_obj_by_os_index
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Outdated suggestions cannot be applied.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.
The integration was not correct - umfMemoryProviderCreateFromMemspace was passing nodemask with all bits set to OS provider (with MODE_BIND) which resulted in memory being allocated from the arbitrary node. This wasn't caught by the test due to wrong way of checking allocation location.
Fix this, by changing the implementation of numa_memory_target to only pass a single node to OS provider for now. This should be changed once we have implemented memspace policies and/or extend OS provider to possibly support setting priority for nodes from which we want to allocate.
Checklist