add KSM and THP tuning opt-ins

[michalskrivanek]

add opt-in MachineConfigs for KSM zero-only and THP tuning

Two new opt-in features gated on HCO annotations:

• ksm-zero-only (platform.kubevirt.io/enable-ksm-zero-only):
  Enables KSM in redhat_only_zero_pages mode with adaptive scan rate.
  Formula: pages_to_scan=1000, sleep=min(50, 25600/node_gb).
• thp-tuning (platform.kubevirt.io/enable-thp-tuning):
  Sets THP to madvise and tunes khugepaged for KVM workloads.
  Formula: pages_to_scan=130000, sleep=1000+64000/node_gb.

Both deploy a Python script via systemd oneshot that reads node
memory size and writes computed parameters to sysfs at boot.

[openshift-ci]

Hi @michalskrivanek. Thanks for your PR.

I'm waiting for a openshift-virtualization member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work.

Tip

We noticed you've done this a few times! Consider joining the org to skip this step and gain /lgtm and other bot rights. We recommend asking approvers on your previous PRs to sponsor you.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

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.

[michalskrivanek]

@fabiand @tiraboschi wdyt?

tried to balance the CPU and the scanning rate and the (currently inevitable) conflict between KSM and THP.
This way for ~90GB worth of VM memory it takes 20m for KSM sweep (40m to merge zero pages), and ~4m for THP to collapse, with <5% CPU usage for ksmd (with spike to <15% when merging) and <1% khugepaged (with spikes as well, depending on available memory)
With some proportionality on smaller or larger systems.
They could be deployed individually and also combined.
The KSM really only makes sense for Windows (or DISA STIG Linux, with the same zeroing memory feature).

corresponding metrics PR openshift-virtualization/kubevirt-metrics-exporter#18

[tiraboschi]

/ok-to-test

[tiraboschi]

I'm wondering if we should do the opposite and threat HCO CR as read-only honoring the new annotation only when ksmConfiguration is null.

[michalskrivanek]

zero-only-pages conflict with kubevirt's KSM management. I think it's the right direction, overriding possible full KSM config when zero-only-pages are requested. Or did you mean something else?

[tiraboschi]

same

[github-actions]

❌ Generated Files Verification Failed

One or more generated files in this PR are out of sync:

• CRDs: Run make update-crds if CRD verification failed
• RBAC: Run make generate-rbac if RBAC verification failed

Please regenerate the files locally and commit the changes.

[github-actions]

❌ Generated Files Verification Failed

One or more generated files in this PR are out of sync:

• CRDs: Run make update-crds if CRD verification failed
• RBAC: Run make generate-rbac if RBAC verification failed

Please regenerate the files locally and commit the changes.

[tiraboschi]

/approve

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: tiraboschi

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

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [tiraboschi]

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

[tiraboschi]

/cherry-pick release-4.22

[openshift-cherrypick-robot]

@tiraboschi: once the present PR merges, I will cherry-pick it on top of release-4.22 in a new PR and assign it to you.

Details

In response to this:

/cherry-pick release-4.22

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.

[openshift-cherrypick-robot]

@tiraboschi: #288 failed to apply on top of branch "release-4.22":

Applying: fix RBAC scanner to handle multi-line template comments
Applying: feat: add generic template_params and topology condition to asset engine
Using index info to reconstruct a base tree...
M	pkg/assets/registry.go
M	pkg/assets/registry_test.go
M	pkg/context/render_context.go
M	pkg/controller/platform_controller.go
M	pkg/engine/renderer.go
M	pkg/engine/renderer_test.go
Falling back to patching base and 3-way merge...
Auto-merging pkg/assets/registry.go
Auto-merging pkg/assets/registry_test.go
Auto-merging pkg/context/render_context.go
Auto-merging pkg/controller/platform_controller.go
Auto-merging pkg/engine/renderer.go
Auto-merging pkg/engine/renderer_test.go
CONFLICT (content): Merge conflict in pkg/engine/renderer_test.go
error: Failed to merge in the changes.
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config set advice.mergeConflict false"
Patch failed at 0002 feat: add generic template_params and topology condition to asset engine

Details

In response to this:

/cherry-pick release-4.22

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.