bug: rdma exlusive handling #603
Merged
Conversation
This file contains 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
Pull Request Test Coverage Report for Build 11577952041Details
💛 - Coveralls |
|
Note that |
ykulazhenkov
approved these changes
Oct 23, 2024
e0ne
approved these changes
Oct 23, 2024
almaslennikov
approved these changes
Oct 24, 2024
pkg/devices/rdma.go
Outdated
| default: | ||
| return false | ||
| } | ||
| // Checking for netlink param for exclusive RDMA use case |
There was a problem hiding this comment.
can you add more information here why we need to check netlink param in this case. (requested by sebastian)
There was a problem hiding this comment.
Done.
@SchSeba PTAL
In case a RDMA device in exclusive mode is in use by a Pod, the DP was not reporting it as a resource after DP restart. Following changes are introduced in RdmaSpec: - isRdma: in case of no rdma resources, check if netlink "enable_rdma" is available. - GetRdmaDeviceSpec: the device specs are retrieved dynamically and not on discovery stage as before. Dynamic RDMA specs computation vs on discovery, comes to solve following scenario for exlusive mode: - Discover RDMA device - Allocate to Pod (resources are hidden on host) - Restart DP pod - Deallocate - Reallocate Fixes k8snetworkplumbingwg#565 Signed-off-by: Fred Rolland <[email protected]>
adrianchiris
approved these changes
Oct 30, 2024
|
@SchSeba can you PTAL? |
SchSeba
approved these changes
Nov 3, 2024
There was a problem hiding this comment.
I tested this one and also add a functional test to cover this one in the operator https://github.com/k8snetworkplumbingwg/sriov-network-operator/pull/799/files#diff-909069834ea269a01a51f28b8830efebec16799766767ebcd01b58f966ddc5c5R226 (real mlx device is needed for the test to run)
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.
In case a RDMA device in exclusive mode is in use by a Pod, the DP was not reporting it as a resource after DP restart.
Following changes are introduced in RdmaSpec:
Dynamic RDMA specs computation vs on discovery, comes to solve following scenario for exlusive mode:
Fixes #565