Refactor CPU Topology Representation for Asymmetric/Heterogeneous Systems

[AutuSnow]

Refactor CPU topology representation to eliminate symmetric/uniform assumptions in CPU allocation logic, as discussed in #16 (comment).

Motivation
The previous CPUTopology methods (CPUsPerCore, CPUsPerSocket, CPUsPerUncore) computed global averages via simple division (e.g., NumCPUs / NumCores). This is incorrect when CPUsare offlined asymmetrically or on heterogeneous systems — all the math goes haywire as

Not fully resolved：

1. takePartialUncore still uses an average within the uncore cache scope. The cpusPerCore calculation (cpusInUncore.Size() / numCores, rounded up) is a per-uncore-cache average rather than the old global average. This is significantly more accurate, but still an approximation,However, this scenario is extremely rare in actual hardware - inside the uncore cache Core is usually symmetrical.

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: AutuSnow
Once this PR has been reviewed and has the lgtm label, please assign klueska for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[AutuSnow]

/assign @ffromani @pravk03

[ffromani]

This PR changes the cpumanager code we imported. At glance the changes seems sensible, but I need to careful review. So, the first question here is: is it time already to drift from the cpumanager reference code we imported?

[AutuSnow]

This PR changes the cpumanager code we imported. At glance the changes seems sensible, but I need to careful review. So, the first question here is: is it time already to drift from the cpumanager reference code we imported?


I believe it's the right time for this specific drift. The project README already states strict cpumanager compatibility is not a goal, and the codebase has already diverged with UncoreCache, CoreType, and NUMA distribution features. The removed methods (CPUsPerCore,CPUsPerSocket, CPUsPerUncore) had an explicit TODO (added in PR #16 review) requesting exactly this refactor. The change is narrowly scoped — it replaces global-average division with pre-computed per-entity maps while keeping the overall allocation algorithm structure intact. The upstream cpumanager is unlikely to address this since it's being superseded by DRA. The backport path remains unaffected since no files were restructured.


[k8s-ci-robot]

@AutuSnow: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
pull-dra-driver-cpu-e2e-device-mode-individual-amd64	f16a624	link	true	/test pull-dra-driver-cpu-e2e-device-mode-individual-amd64
pull-dra-driver-cpu-e2e-device-mode-grouped-amd64	f16a624	link	true	/test pull-dra-driver-cpu-e2e-device-mode-grouped-amd64

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.

[k8s-ci-robot]

PR needs rebase.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[AutuSnow]

/cc @ffromani I would like to hear your thoughts. Is it necessary to continue waiting for the current PR

[AutuSnow]

/close

[k8s-ci-robot]

@AutuSnow: Closed this PR.

Details

In response to this:

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.