Caching qemu capabilities and KubeVirt

Thu Sep 8 14:22:17 UTC 2022

On Thu, Sep 8, 2022 at 4:04 PM Daniel P. Berrangé <berrange at redhat.com>
wrote:

> On Thu, Sep 08, 2022 at 02:23:31PM +0100, Daniel P. Berrangé wrote:
> > On Thu, Sep 08, 2022 at 03:10:09PM +0200, Roman Mohr wrote:
> > > On Thu, Sep 8, 2022 at 2:56 PM Daniel P. Berrangé <berrange at redhat.com
> >
> > > wrote:
> > >
> > > > On Thu, Sep 08, 2022 at 02:24:00PM +0200, Roman Mohr wrote:
> > > > > Hi,
> > > > >
> > > > >
> > > > > I have a question regarding capability caching  in the context of
> > > > KubeVirt.
> > > > > Since we start in KubeVirt one libvirt instance per VM, libvirt
> has to
> > > > > re-discover on every VM start the qemu capabilities which leads to
> a
> > > > 1-2s+
> > > > > delay in startup.
> > > > >
> > > > > We already discover the features in a dedicated KubeVirt pod on
> each
> > > > node.
> > > > > Therefore I tried to copy the capabilities over to see if that
> would
> > > > work.
> > > > >
> > > > > It looks like in general it could work, but libvirt seems to
> detect a
> > > > > mismatch in the exposed KVM CPU ID in every pod. Therefore it
> invalidates
> > > > > the cache. The recreated capability cache looks esctly like the
> original
> > > > > one though ...
> > > > >
> > > > > The check responsible for the invalidation is this:
> > > > >
> > > > > ```
> > > > > Outdated capabilities for '%s': host cpuid changed
> > > > > ```
> > > > >
> > > > > So the KVM_GET_SUPPORTED_CPUID call seems to return
> > > > > slightly different values in different containers.
> > > > >
> > > > > After trying out the attached golang scripts in different
> containers, I
> > > > > could indeed see differences.
> > > > >
> > > > > I can however not really judge what the differences in these KVM
> function
> > > > > registers mean and I am curious if someone else knows. The files
> are
> > > > > attached too (as json for easy diffing).
> > > >
> > > > Can you confirm whether the two attached data files were captured
> > > > by containers running on the same physical host, or could each
> > > > container have run on a different host.
> > > >
> > >
> > > They are coming from the same host, that is the most surprising bit
> for me.
> > > I am also very sure that this is the case, because I only had one k8s
> node
> > > from where I took these.
> > > The containers however differ (obviously) on namespaces and on the
> > > privilege level (less obvious). The handler dump is from a fully
> privileged
> > > container.
> >
> > The privilege level sounds like something that might be impactful,
> > so I'll investigate that.  I'd be pretty surprised for namespaces
> > to have any impact thnough.
>
> The privilege level is a red herring. Peter reminded me that we have
> to filter out some parts of CPUID because the APIC IDs vary depending
> on what host CPU the task executes on.
>
>
> https://gitlab.com/libvirt/libvirt/-/blob/master/src/util/virhostcpu.c#L1346
>
> In the 2 jSON files you provide, the differences i see should already
> be matched by
>
>         /* filter out local apic id */
>         if (entry->function == 0x01 && entry->index == 0x00)
>             entry->ebx &= 0x00ffffff;
>         if (entry->function == 0x0b)
>             entry->edx &= 0xffffff00;
>
> so those differences ought not to be causing the cache to be
> invalidated.
>

Hm, maybe I misinterpreted the logs then. The snipped I looked at was this:

```
{"component":"virt-launcher","level":"info","msg":"/dev/kvm has changed
(1661786802 vs
0)","pos":"virQEMUCapsKVMUsable:4850","subcomponent":"libvirt","thread":"25","timestamp":"2022-08-29T15:26:42.936000Z"}
{"component":"virt-launcher","level":"info","msg":"a=0x7f8138153ba0,
b=0x7f818001c480","pos":"virCPUDataIsIdentical:1178","subcomponent":"libvirt","thread":"25","timestamp":"2022-08-29T15:26:42.939000Z"}
{"component":"virt-launcher","level":"info","msg":"Outdated capabilities
for '/usr/bin/qemu-system-x86_64': host cpuid
changed","pos":"virQEMUCapsIsValid:4993","subcomponent":"libvirt","thread":"25","timestamp":"2022-08-29T15:26:42.939000Z"}
{"component":"virt-launcher","level":"info","msg":"Outdated cached
capabilities
'/var/cache/libvirt/qemu/capabilities/926803a9278e445ec919c2b6cbd8c1c449c75b26dcb1686b774314180376c725.xml'
for
'/usr/bin/qemu-system-x86_64'","pos":"virFileCacheLoad:163","subcomponent":"libvirt","thread":"25","timestamp":"2022-08-29T15:26:42.939000Z"}
```

I had the impression from the code that the `/dev/kvm` change (because the
containers are not created at the same time) does not invalidate it either.

I added the whole debug log, maybe I missed something obvious.

Does it make a difference if the cache is created via `virsh
domcapabilities` and `virsh capabilities` or via defining the first domain?

Best regards,
Roman

>
> With regards,
> Daniel
> --
> |: https://berrange.com      -o-
> https://www.flickr.com/photos/dberrange :|
> |: https://libvirt.org         -o-
> https://fstop138.berrange.com :|
> |: https://entangle-photo.org    -o-
> https://www.instagram.com/dberrange :|
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/libvirt-users/attachments/20220908/af7f961f/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: debug.log.gz
Type: application/gzip
Size: 169826 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/libvirt-users/attachments/20220908/af7f961f/attachment-0001.gz>