Bug 1747185

Verify:
qemu-kvm-2.12.0-86.module+el8.1.0+4146+4ed2d185
kernel-4.18.0-141.el8.x86_64
libvirt-client-4.5.0-33.scrmod+el8.1.0+4090+e8e6ad83.x86_64

Host: intel-purley-lr-03.khw1.lab.eng.bos.redhat.com 
# lscpu
Model name:          Intel(R) Xeon(R) Platinum 8260L CPU @ 2.40GHz
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req pku ospke avx512_vnni md_clear flush_l1d arch_capabilities


The arch-capabilities bits (rdctl-no, ibrs-all, skip-l1dfl-vmentry, mds-no) are reported by "virsh domcapabilities", except "ssb-no", "rsba".

# virsh domcapabilities
   <mode name='host-model' supported='yes'>
      <model fallback='forbid'>Cascadelake-Server</model>
      <vendor>Intel</vendor>
      <feature policy='require' name='ss'/>
      <feature policy='require' name='vmx'/>
      <feature policy='require' name='hypervisor'/>
      <feature policy='require' name='tsc_adjust'/>
      <feature policy='require' name='umip'/>
      <feature policy='require' name='pku'/>
      <feature policy='require' name='md-clear'/>
      <feature policy='require' name='stibp'/>
      <feature policy='require' name='arch-capabilities'/>
      <feature policy='require' name='xsaves'/>
      <feature policy='require' name='invtsc'/>
      <feature policy='require' name='rdctl-no'/>
      <feature policy='require' name='ibrs-all'/>
      <feature policy='require' name='skip-l1dfl-vmentry'/>
      <feature policy='require' name='mds-no'/>
    </mode>

When use host-model, QEMU cli is,
#...  -cpu Cascadelake-Server,ss=on,vmx=on,hypervisor=on,tsc-adjust=on,umip=on,pku=on,md-clear=on,stibp=on,arch-capabilities=on,xsaves=on,rdctl-no=on,ibrs-all=on,skip-l1dfl-vmentry=on,mds-no=on,hv_time,hv_vapic,hv_spinlocks=0x1000


Checked cpu vulnerability info in guest, all are same to host.
# grep . /sys/devices/system/cpu/vulnerabilities/*
/sys/devices/system/cpu/vulnerabilities/l1tf:Not affected
/sys/devices/system/cpu/vulnerabilities/mds:Not affected
/sys/devices/system/cpu/vulnerabilities/meltdown:Not affected
/sys/devices/system/cpu/vulnerabilities/spec_store_bypass:Mitigation: Speculative Store Bypass disabled via prctl and seccomp
/sys/devices/system/cpu/vulnerabilities/spectre_v1:Mitigation: usercopy/swapgs barriers and __user pointer sanitization
/sys/devices/system/cpu/vulnerabilities/spectre_v2:Mitigation: Enhanced IBRS, IBPB: conditional, RSB filling

Hi Eduardo,

Would you please check above verification? Is it expected that "ssb-no" and "rsba" are not reported in virsh domcapabilities? And I'm wondering if there is a way to verify from QEMU other than libvirt. 

Thanks,
Yumei Huang

(In reply to Yumei Huang from comment #11)
> Hi Eduardo,
> 
> Would you please check above verification? Is it expected that "ssb-no" and
> "rsba" are not reported in virsh domcapabilities?

It depends on the host.  Unfortunately the only way to get the MSR bits from the host is using the `rdmsr 0x0000010a` command and decoding the bits by hand.  RSBA is bit 2, SSB_NO is bit 4.

We can see that this host doesn't have SSB_NO set, because it would make vulnerabilities/spec_store_bypass say "Not affected".

RSBA seems to be ignored by Linux, so we shouldn't worry about it.  But you can check it using `rdmsr 0x0000010a` if you want to.  My guess is that the host doesn't have RSBA set.



> And I'm wondering if there
> is a way to verify from QEMU other than libvirt. 

It is possible to use the `qom-get path=... property=unavailable-features` QMP command to get the property from the CPU objects (look for the CPU QOM path on `query-cpus`).  But as this is an integration bug, testing if libvirt works is the most important part of the validation of this bug.

(In reply to Eduardo Habkost from comment #12)
> (In reply to Yumei Huang from comment #11)
> > Hi Eduardo,
> > 
> > Would you please check above verification? Is it expected that "ssb-no" and
> > "rsba" are not reported in virsh domcapabilities?
> 
> It depends on the host.  Unfortunately the only way to get the MSR bits from
> the host is using the `rdmsr 0x0000010a` command and decoding the bits by
> hand.  RSBA is bit 2, SSB_NO is bit 4.
> 
> We can see that this host doesn't have SSB_NO set, because it would make
> vulnerabilities/spec_store_bypass say "Not affected".
> 
> RSBA seems to be ignored by Linux, so we shouldn't worry about it.  But you
> can check it using `rdmsr 0x0000010a` if you want to.  My guess is that the
> host doesn't have RSBA set.
> 

Checked host, both RSBA and SSB_NO bit is 0.

# rdmsr 0x0000010a
2b

> 
> > And I'm wondering if there
> > is a way to verify from QEMU other than libvirt. 
> 
> It is possible to use the `qom-get path=... property=unavailable-features`
> QMP command to get the property from the CPU objects (look for the CPU QOM
> path on `query-cpus`).  But as this is an integration bug, testing if
> libvirt works is the most important part of the validation of this bug.

Seems it can't by using qom-get as it returns empty.

{"execute": "qom-get","arguments": { "path": "/machine/unattached/device[2]","property": "unavailable-features"}}
{"return": []}


However, I'm moving to verified per the test with libvirt in comment 10. Thanks.

(In reply to Yumei Huang from comment #13)
> Seems it can't by using qom-get as it returns empty.
> 
> {"execute": "qom-get","arguments": { "path":
> "/machine/unattached/device[2]","property": "unavailable-features"}}
> {"return": []}

This is expected.  unavailable-features is the QMP equivalent of the "feature not supported by host" warnings on the console.  If you see no warnings, unavailable-features will be empty.

e.g.: using "-cpu host,+rsba,+ssb-no" should make the property return ["rsba", "ssb-no"].

(In reply to Eduardo Habkost from comment #14)
> (In reply to Yumei Huang from comment #13)
> > Seems it can't by using qom-get as it returns empty.
> > 
> > {"execute": "qom-get","arguments": { "path":
> > "/machine/unattached/device[2]","property": "unavailable-features"}}
> > {"return": []}
> 
> This is expected.  unavailable-features is the QMP equivalent of the
> "feature not supported by host" warnings on the console.  If you see no
> warnings, unavailable-features will be empty.
> 
> e.g.: using "-cpu host,+rsba,+ssb-no" should make the property return
> ["rsba", "ssb-no"].

Yes, you are right, thanks.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2019:3345