<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Sep 8, 2022 at 4:04 PM Daniel P. Berrangé <<a href="mailto:berrange@redhat.com">berrange@redhat.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Thu, Sep 08, 2022 at 02:23:31PM +0100, Daniel P. Berrangé wrote:<br> > On Thu, Sep 08, 2022 at 03:10:09PM +0200, Roman Mohr wrote:<br> > > On Thu, Sep 8, 2022 at 2:56 PM Daniel P. Berrangé <<a href="mailto:berrange@redhat.com" target="_blank">berrange@redhat.com</a>><br> > > wrote:<br> > > <br> > > > On Thu, Sep 08, 2022 at 02:24:00PM +0200, Roman Mohr wrote:<br> > > > > Hi,<br> > > > ><br> > > > ><br> > > > > I have a question regarding capability caching in the context of<br> > > > KubeVirt.<br> > > > > Since we start in KubeVirt one libvirt instance per VM, libvirt has to<br> > > > > re-discover on every VM start the qemu capabilities which leads to a<br> > > > 1-2s+<br> > > > > delay in startup.<br> > > > ><br> > > > > We already discover the features in a dedicated KubeVirt pod on each<br> > > > node.<br> > > > > Therefore I tried to copy the capabilities over to see if that would<br> > > > work.<br> > > > ><br> > > > > It looks like in general it could work, but libvirt seems to detect a<br> > > > > mismatch in the exposed KVM CPU ID in every pod. Therefore it invalidates<br> > > > > the cache. The recreated capability cache looks esctly like the original<br> > > > > one though ...<br> > > > ><br> > > > > The check responsible for the invalidation is this:<br> > > > ><br> > > > > ```<br> > > > > Outdated capabilities for '%s': host cpuid changed<br> > > > > ```<br> > > > ><br> > > > > So the KVM_GET_SUPPORTED_CPUID call seems to return<br> > > > > slightly different values in different containers.<br> > > > ><br> > > > > After trying out the attached golang scripts in different containers, I<br> > > > > could indeed see differences.<br> > > > ><br> > > > > I can however not really judge what the differences in these KVM function<br> > > > > registers mean and I am curious if someone else knows. The files are<br> > > > > attached too (as json for easy diffing).<br> > > ><br> > > > Can you confirm whether the two attached data files were captured<br> > > > by containers running on the same physical host, or could each<br> > > > container have run on a different host.<br> > > ><br> > > <br> > > They are coming from the same host, that is the most surprising bit for me.<br> > > I am also very sure that this is the case, because I only had one k8s node<br> > > from where I took these.<br> > > The containers however differ (obviously) on namespaces and on the<br> > > privilege level (less obvious). The handler dump is from a fully privileged<br> > > container.<br> > <br> > The privilege level sounds like something that might be impactful,<br> > so I'll investigate that. I'd be pretty surprised for namespaces<br> > to have any impact thnough.<br> <br> The privilege level is a red herring. Peter reminded me that we have<br> to filter out some parts of CPUID because the APIC IDs vary depending<br> on what host CPU the task executes on.<br> <br> <a href="https://gitlab.com/libvirt/libvirt/-/blob/master/src/util/virhostcpu.c#L1346" rel="noreferrer" target="_blank">https://gitlab.com/libvirt/libvirt/-/blob/master/src/util/virhostcpu.c#L1346</a><br> <br> In the 2 jSON files you provide, the differences i see should already<br> be matched by<br> <br> /* filter out local apic id */<br> if (entry->function == 0x01 && entry->index == 0x00)<br> entry->ebx &= 0x00ffffff;<br> if (entry->function == 0x0b)<br> entry->edx &= 0xffffff00;<br> <br> so those differences ought not to be causing the cache to be<br> invalidated.<br></blockquote><div><br></div><div>Hm, maybe I misinterpreted the logs then. The snipped I looked at was this:<br><br><br>```<br>{"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"}<br>{"component":"virt-launcher","level":"info","msg":"a=0x7f8138153ba0, b=0x7f818001c480","pos":"virCPUDataIsIdentical:1178","subcomponent":"libvirt","thread":"25","timestamp":"2022-08-29T15:26:42.939000Z"}<br>{"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"}<br>{"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"}<br>```<br><br>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.<br><br>I added the whole debug log, maybe I missed something obvious.</div><div><br></div><div>Does it make a difference if the cache is created via `virsh domcapabilities` and `virsh capabilities` or via defining the first domain?</div><div><br></div><div>Best regards,</div><div>Roman</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> <br> With regards,<br> Daniel<br> -- <br> |: <a href="https://berrange.com" rel="noreferrer" target="_blank">https://berrange.com</a> -o- <a href="https://www.flickr.com/photos/dberrange" rel="noreferrer" target="_blank">https://www.flickr.com/photos/dberrange</a> :|<br> |: <a href="https://libvirt.org" rel="noreferrer" target="_blank">https://libvirt.org</a> -o- <a href="https://fstop138.berrange.com" rel="noreferrer" target="_blank">https://fstop138.berrange.com</a> :|<br> |: <a href="https://entangle-photo.org" rel="noreferrer" target="_blank">https://entangle-photo.org</a> -o- <a href="https://www.instagram.com/dberrange" rel="noreferrer" target="_blank">https://www.instagram.com/dberrange</a> :|<br> <br> </blockquote></div></div>