Bug 1275088
Summary: | Intel Core i7 4790k (Haswell, Devil's Canyon) CPU not properly detected | ||
---|---|---|---|
Product: | [oVirt] ovirt-engine | Reporter: | fdinoto |
Component: | BLL.Virt | Assignee: | Michal Skrivanek <michal.skrivanek> |
Status: | CLOSED CURRENTRELEASE | QA Contact: | Nisim Simsolo <nsimsolo> |
Severity: | unspecified | Docs Contact: | |
Priority: | unspecified | ||
Version: | 3.4.0 | CC: | bugs, fdinoto, jdenemar, jsuchane, mavital, tjelinek |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2016-04-13 06:46:43 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | Virt | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
fdinoto
2015-10-25 19:22:55 UTC
I made another datacenter/cluster, this time of cpu-type Sandybridge. I was able to add the host to the cluster, but now virtual machines will not start. "VM c7-1 is down with error. Exit message: internal error: Cannot find suitable CPU model for given data." I assume this is because the virtual machines are seeing the CPUs as Haswell and are expecting Sandybridge because of the cluster definition. I guess this is not a bug the 'K'(Devil's Canyon) designated Haswell CPUs have some features disabled. http://ark.intel.com/compare/75122,80806,80807 I have not correlated the features listed in the Intel docs to the flags in /proc/cpuinfo, but I was able to work around the issue using these settings in /usr/share/libvirt/cpu_map.xml: <model name='Haswell'> <model name='SandyBridge'/> <feature name='fma'/> <feature name='pcid'/> <feature name='movbe'/> <feature name='fsgsbase'/> <feature name='bmi1'/> <!-- <feature name='hle'/> --> <feature name='avx2'/> <feature name='smep'/> <feature name='bmi2'/> <feature name='erms'/> <feature name='invpcid'/> <!-- <feature name='rtm'/> --> </model> do we(you) need a new CPU definition? (In reply to Michal Skrivanek from comment #3) > do we(you) need a new CPU definition? Probably not, as the definition from comment 2 corresponds to Haswell-noTSX. Jirka can you please confirm? But this might be present in rhel-7.2 (centos 7.2) though. Thanks. To fdinoto, please provide also exact versions of used components, specifically libvirt in this case. Thanks. Yeah, the CPU is apparently Haswell-noTSX, which was added in 7.2. SandyBridge should work from libvirt's point of view. Just to be sure, could you provide "virsh capabilities" output (in addition to the libvirt version used)? any news? otherwise closing as working in 3.6 The Haswell-noTSX has been added to libvirt in centos 7.2 -> closing as currentrelease |