On 18 Nov 2019, at 09:29, Christian Reiss
<email(a)christian-reiss.de> wrote:
Ugh,
first off thanks for all your effort and information. Much obliged.
But this also means that I am looking at a loooong time before I can go to production
with this cluster. Now I am sad.
Would it be an option to run the engine on dedicated hardware and control the cluster
from there? Or is that thing in total not usable?
I’m curious why do you need virt-ssbd for hosted engine? What for?
-Chris.
On 18/11/2019 07:25, Juhani Rautiainen wrote:
> Hi!
> Had to get back to work to check which CPU we had. We have AMD Epyc
> 7281 and ovirt CPU Type is AMD EPYC IBPB SSBD. It seems that your CPU
> is the next generation (Zen2) and I'm pretty sure that problem is with
> Qemu version. As far as I can see from git there is not even Zen2
yes, AFAIK Zen2 is not yet supported in upstream. Once it is we’ll pick it up afterwards.
But it doesn’t mean you can’t use it for running stuff in EPYC or lower CPU family feature
set.
> support in latest Qemu (checking by target/i386/cpu.c)? I mean
they
> added Hygon Dhyana (never even heard about this chinese AMD EPYC
> clone) and in that discussion there was reference to Zen2
> architecture. So biggest problem for oVirt seems to come from
> upstream. I mean that Zen2 is quite good for virtualization and it's
> going to sell a lot. Maybe AMD should help with that push?
> -Juhani
> On Fri, Nov 15, 2019 at 9:03 PM Christian Reiss
> <email(a)christian-reiss.de <mailto:email@christian-reiss.de>> wrote:
>>
>> Sorry,
>>
>> I meant EPYC, not Ryzen.
>> How did you solve your EPYC issue?
>>
>> -Chris.
>>
>> On 15/11/2019 18:55, Juhani Rautiainen wrote:
>>> Hi!
>>>
>>> It might be that the Qemu in oVirt doesn't recognize the Ryzen. That
>>> was case with Epyc when I started using oVirt. It was reconized as a
>>> Opteron G2 which caused lot's of problems when upgrading to 4.3.
>>>
>>> -Juhani
>>>
>>> On Fri, Nov 15, 2019 at 6:45 PM Christian Reiss
>>> <email(a)christian-reiss.de> wrote:
>>>>
>>>> Hey folks,
>>>>
>>>> running an AMD Ryzen CPU here:
>>>>
>>>> processor : 0
>>>> vendor_id : AuthenticAMD
>>>> cpu family : 23
>>>> model : 49
>>>> model name : AMD EPYC 7282 16-Core Processor
>>>>
>>>> However, libvirt is detecting this as EPYC-IBPB without the ssbd flags?
>>>>
>>>> <cpu>
>>>> <arch>x86_64</arch>
>>>> <model>EPYC-IBPB</model>
>>>> <vendor>AMD</vendor>
>>>> <microcode version='137367580'/>
>>>> <counter name='tsc'
frequency='2799999000'/>
>>>> <topology sockets='1' cores='16'
threads='2'/>
>>>> <feature name='ht'/>
>>>> <feature name='osxsave'/>
>>>> <feature name='cmt'/>
>>>> <feature name='clwb'/>
>>>> <feature name='umip'/>
>>>> <feature name='xsaves'/>
>>>> <feature name='mbm_total'/>
>>>> <feature name='mbm_local'/>
>>>> <feature name='cmp_legacy'/>
>>>> <feature name='extapic'/>
>>>> <feature name='ibs'/>
>>>> <feature name='skinit'/>
>>>> <feature name='wdt'/>
>>>> <feature name='tce'/>
>>>> <feature name='topoext'/>
>>>> <feature name='perfctr_core'/>
>>>> <feature name='perfctr_nb'/>
>>>> <feature name='invtsc'/>
>>>> <feature name='wbnoinvd'/>
>>>> <pages unit='KiB' size='4'/>
>>>> <pages unit='KiB' size='2048'/>
>>>> <pages unit='KiB' size='1048576'/>
>>>> </cpu>
in 4.3 we still use flags instead of libvirt feature detection (which we’re fixing in 4.4,
but so far it’s playing to your advantage:)
So seeing “ssbd" below in cpuinfo should be enough to detect the oVirt type with
SSBD. Where exatly you see that not working?
Thanks,
michal
>>>>
>>>>
>>>> [root@node01 ~]# grep ssbd /var/cache/libvirt/qemu/capabilities/*.xml
>>>> <property name='ssbd' type='boolean'
value='false'/>
>>>> <property name='virt-ssbd' type='boolean'
value='false'/>
>>>> <property name='ssbd' type='boolean'
value='false'/>
>>>> <property name='virt-ssbd' type='boolean'
value='false'/>
>>>>
>>>> But the flag is there:
>>>>
>>>> [root@node01 ~]# grep ssbd /proc/cpuinfo | tail -n1
>>>> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
mca cmov
>>>> pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt
>>>> pdpe1gb rdtscp lm constant_tsc art rep_good nopl xtopology nonstop_tsc
>>>> extd_apicid aperfmperf eagerfpu pni pclmulqdq monitor ssse3 fma cx16
>>>> sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy
>>>> svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs
>>>> skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_l2 cpb
>>>> cat_l3 cdp_l3 hw_pstate sme retpoline_amd ssbd ibrs ibpb stibp vmmcall
>>>> fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb
>>>> sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total
>>>> cqm_mbm_local clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save
>>>> tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold
>>>> avic v_vmsave_vmload vgif umip overflow_recov succor smca
>>>>
>>>> I tried adding "options kvm_amd avic=1" as well as
"options kvm_amd
>>>> avic=0" to /etc/modprobe.d/kvm.conf (always with reboots), adding
>>>> mitigations=off to grub.. I can't think of any other solution.
>>>>
>>>> I just can't get the oVirt engine running with the ssbd flag. Seems
cpu
>>>> can do this, oVirt can do this, libvirt does not detect the cpu
>>>> correctly or at least ignores it. But the hosted engine demands it.
>>>>
>>>> I am at a loss. Any help is oh-so-greatly appreciated.
>>>>
>>>> -Chris.
>>>>
>>>> --
>>>> Christian Reiss - email(a)christian-reiss.de /"\ ASCII
Ribbon
>>>> support(a)alpha-labs.net \ / Campaign
>>>> X against HTML
>>>> WEB
alpha-labs.net / \ in eMails
>>>>
>>>> GPG Retrieval
https://gpg.christian-reiss.de
>>>> GPG ID ABCD43C5, 0x44E29126ABCD43C5
>>>> GPG fingerprint = 9549 F537 2596 86BA 733C A4ED 44E2 9126 ABCD 43C5
>>>>
>>>> "It's better to reign in hell than to serve in
heaven.",
>>>> John Milton, Paradise lost.
>>>> _______________________________________________
>>>> Users mailing list -- users(a)ovirt.org
>>>> To unsubscribe send an email to users-leave(a)ovirt.org
>>>> Privacy Statement:
https://www.ovirt.org/site/privacy-policy/
>>>> oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
>>>> List Archives:
https://lists.ovirt.org/archives/list/users@ovirt.org/message/PPXT55FJZES...
>>>
>>>
>>>
>>
>> --
>> Christian Reiss - email(a)christian-reiss.de /"\ ASCII Ribbon
>> support(a)alpha-labs.net \ / Campaign
>> X against HTML
>> WEB
alpha-labs.net / \ in eMails
>>
>> GPG Retrieval
https://gpg.christian-reiss.de
>> GPG ID ABCD43C5, 0x44E29126ABCD43C5
>> GPG fingerprint = 9549 F537 2596 86BA 733C A4ED 44E2 9126 ABCD 43C5
>>
>> "It's better to reign in hell than to serve in heaven.",
>> John Milton, Paradise lost.
--
Christian Reiss - email(a)christian-reiss.de <mailto:email@christian-reiss.de>
/"\ ASCII Ribbon
support(a)alpha-labs.net <mailto:support@alpha-labs.net>
\ / Campaign
X against HTML
WEB
alpha-labs.net <
http://alpha-labs.net/> / \
in eMails
GPG Retrieval
https://gpg.christian-reiss.de <
https://gpg.christian-reiss.de/>
GPG ID ABCD43C5, 0x44E29126ABCD43C5
GPG fingerprint = 9549 F537 2596 86BA 733C A4ED 44E2 9126 ABCD 43C5
"It's better to reign in hell than to serve in heaven.",
John Milton, Paradise lost.
_______________________________________________
Users mailing list -- users(a)ovirt.org <mailto:users@ovirt.org>
To unsubscribe send an email to users-leave(a)ovirt.org
<mailto:users-leave@ovirt.org>
Privacy Statement:
https://www.ovirt.org/site/privacy-policy/
<
https://www.ovirt.org/site/privacy-policy/>
oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
<
https://www.ovirt.org/community/about/community-guidelines/>
List Archives:
https://lists.ovirt.org/archives/list/users@ovirt.org/message/4LVQM5CDDEL...
<
https://lists.ovirt.org/archives/list/users@ovirt.org/message/4LVQM5CDDEL...