Huge pages KEP: support multiple sizes huge pages on a container level

[bart0sh]

This change adds support of multiple sizes huge pages on a
container level to support the following use cases:

• VMs running on a Kubernetes infrastructure (QEMU, libvirt, etc)
• Applications using more than one huge page size

[k8s-ci-robot]

Welcome @bart0sh!

It looks like this is your first PR to kubernetes/enhancements 🎉. Please refer to our pull request process documentation to help your PR have a smooth ride to approval.

You will be prompted by a bot to use commands during the review process. Do not be afraid to follow the prompts! It is okay to experiment. Here is the bot commands documentation.

You can also check if kubernetes/enhancements has its own contribution guidelines.

You may want to refer to our testing guide if you run into trouble with your tests not passing.

If you are having difficulty getting your pull request seen, please follow the recommended escalation practices. Also, for tips and tricks in the contribution process you may want to read the Kubernetes contributor cheat sheet. We want to make sure your contribution gets all the attention it needs!

Thank you, and welcome to Kubernetes. 😃

[derekwaynecarr]

/assign

[derekwaynecarr]

Just a question on the validation

[odinuge]

Thanks for opening! Other than some small comments, this looks good to me! :)

PR to add node level support is here kubernetes/kubernetes#82820, and I have already been looking at implementing this, so i'll be happy to help with the implementation.

[odinuge]

I think we should rewrite (or remove) this chapter:

enhancements/keps/sig-node/20190129-hugepages.md

Line 145 in 404fb3d

A system may support multiple huge page sizes. It is assumed that most nodes

I think there is some kind of ambiguity in It is assumed that most nodes will be configured to primarily use the default huge page size as returned via grep Hugepagesize /proc/meminfo. regarding support for multiple page sizes at node level, and that is needed in order to supporting on pod level.

[bart0sh]

@derekwaynecarr @odinuge Thank you for the review!

I've updated the PR according to your suggestions.

[odinuge]

With the latest changes, this looks good to me.

It is really nice with all the clear examples. I cannot see any explicit info (or example) about the fact that mounting hugetlbfs is optional. A pod can request 1GiB of 2MiB pages, without using a volume mount to mount a hugetlbfs. May be nice to have it explicitly stated.

/lgtm

[bg-chun]

Updates looks good.

I have just one thing to suggest here.
How about putting a yaml example for the case that Pod consumes without emptyDir backing.
It would look like below.

apiVersion: v1
kind: Pod
metadata:
  name: example
spec:
  containers:
...
    resources:
      requests:
        hugepages-1Gi: 1Gi
      limits:
        hugepages-1Gi: 1Gi

As i know(and KEP says) there are three ways to consume hugepages.
Those are :

1. mmap() a file backed by hugetlbfs / sample
2. mmap with MAP_ANONYMOUS / sample
3. shmget() with SHM_HUGETLB / sample

[bart0sh]

@odinuge @bg-chun Thank you for the review and suggestions!

I've added the following text to explicitly mention that emptyDir(hugetlbfs mounts) are optional:

Also, it is important to note that emptyDir usage is not required if pod
consumes huge pages via shmat/shmget system calls or mmap with MAP_HUGETLB.
This is an example of the pod that consumes 1Gi huge pages of size 2Mi and
2Gi huge pages of size 1Gi without using emptyDir:

apiVersion: v1
kind: Pod
metadata:
  name: example
spec:
  containers:
...
    resources:
      requests:
        hugepages-2Mi: 1Gi
        hugepages-1Gi: 2Gi
      limits:
        hugepages-2Mi: 1Gi
        hugepages-1Gi: 2Gi

Does this look more clear now?

[derekwaynecarr]

Thank you for the updates and clarifying the validation error scenarios.

/approve
/lgtm

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: bart0sh, derekwaynecarr, odinuge

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:

• keps/sig-node/OWNERS [derekwaynecarr]

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