koordlet: support hugepage reporting

[peiqiaoWang]

Ⅰ. Describe what this PR does

支持了大页上报的功能，包括2M和1G大页

Ⅱ. Does this pull request fix one issue?

无

Ⅲ. Describe how to verify it

可通过k8s kubectl get noderesourcetopologies 查看

Ⅳ. Special notes for reviews

支持了大页内存的上报，其中原本的内存memory上送的数目也进行了修改，将上报内存 设置为 总内存大小 减去 大页内存的数量。

func (s *nodeTopoInformer) calTopologyZoneList(nodeCPUInfo *metriccache.NodeCPUInfo) 需重点看下。

V. Checklist

• I have written necessary docs and comments
• I have added necessary unit tests and integration tests
• All checks passed in make test

[codecov]

Codecov Report

Attention: 34 lines in your changes are missing coverage. Please review.

Comparison is base (b6a7d9e) 66.08% compared to head (3e8552a) 66.32%.
Report is 18 commits behind head on main.

Files	Patch %	Lines
pkg/koordlet/util/meminfo.go	74.54%	20 Missing and 8 partials ⚠️
pkg/koordlet/util/system/system_file.go	0.00%	4 Missing ⚠️
...statesinformer/impl/states_noderesourcetopology.go	91.30%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1744      +/-   ##
==========================================
+ Coverage   66.08%   66.32%   +0.23%     
==========================================
  Files         388      395       +7     
  Lines       42373    43536    +1163     
==========================================
+ Hits        28003    28874     +871     
- Misses      12296    12512     +216     
- Partials     2074     2150      +76     

Flag	Coverage Δ
unittests	66.32% <75.71%> (+0.23%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[saintube]

/lgtm
PTAL /cc @zwzhang0107

[hormes]

@peiqiaoWang please add a feature-gate for it, thanks.

[zwzhang0107]

I think we need to clarify whether koordlet needs to do something on huge-page considering NUMA (i.e. set cpuset.mems or not set yet)

[peiqiaoWang]

@zwzhang0107 kubelet will do this thing that set value on hugetlb.1GB.limit_in_bytes.

[zwzhang0107]

@zwzhang0107 kubelet will do this thing that set value on hugetlb.1GB.limit_in_bytes.

@peiqiaoWang yep, i mean the cpuset.mems. We should add some comments to clarify whether limit the hugepages on single numa node.

[peiqiaoWang]

@zwzhang0107 kubelet will do this thing that set value on hugetlb.1GB.limit_in_bytes.

@peiqiaoWang yep, i mean the cpuset.mems. We should add some comments to clarify whether limit the hugepages on single numa node.

这个功能只支持大页上报，设置numa信息由调度器分配，然后作用到pod是由csi根据调度器分配的numa信息把预分配大页绑上去，此处是否设置绑定namespace的numa使用关系倒是不大

[peiqiaoWang]

@zwzhang0107 kubelet will do this thing that set value on hugetlb.1GB.limit_in_bytes.

@peiqiaoWang yep, i mean the cpuset.mems. We should add some comments to clarify whether limit the hugepages on single numa node.

这个功能只支持大页上报，设置numa信息由调度器分配，然后作用到pod是由csi根据调度器分配的numa信息把预分配大页绑上去，此处是否设置绑定namespace的numa使用关系倒是不大

@zwzhang0107

// This feature supports reporting of hugepages.
// The koord-scheduler will allocate hugepage information based on the user's hugepage request and add it to the Pod's annotations.
// Format: scheduling.koordinator.sh/resource-status: '{"numaNodeResources":[{"node":1,"resources":{"hugepages-1Gi":"50Gi"}}]}'.
// Backend applications can enable the hugepages based on the allocation results.
// For example, the CSI mounts the pre-allocated hugepages into the pod.

Add some comments in pkg/features/koordlet_features.go

[zwzhang0107]

/lgtm

[zwzhang0107]

/approve

[saintube]

/lgtm

[koordinator-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: hormes, zwzhang0107

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

The pull request process is described here

Needs approval from an approver in each of these files:

• pkg/features/OWNERS [hormes]
• pkg/koordlet/OWNERS [hormes,zwzhang0107]

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