Bug 1382434

Summary: CPU to NUMA node QMP mapping interface (libvirt)
Product: Red Hat Enterprise Linux 9 Reporter: Hai Huang <hhuang>
Component: libvirtAssignee: Virtualization Maintenance <virt-maint>
Status: CLOSED DUPLICATE QA Contact: Jing Qi <jinqi>
Severity: medium Docs Contact:
Priority: medium    
Version: unspecifiedCC: chhu, drjones, dyuan, imammedo, jinqi, jsuchane, jtomko, knoel, lhuang, mprivozn, mtessun, pkrempa, virt-bugs, virt-maint, xuzhang, yalzhang
Target Milestone: rcKeywords: FutureFeature, Reopened, Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 1382425 Environment:
Last Closed: 2021-09-30 13:58:12 UTC Type: Feature Request
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 1382425, 1562805, 1678085    
Bug Blocks:    

Description Hai Huang 2016-10-06 17:22:20 UTC 
+++ This bug was initially created as a clone of Bug #1382425 +++

Description of problem:

Current -numa CLI interface is quite limited in terms that allow map CPUs to NUMA nodes as it requires to provide cpu_index values which are non obvious and depend on machine/arch. As result libvirt has to assume/re-implement cpu_index allocation logic to provide valid values for -numa cpus=... QEMU CLI option.

Now QEMU has in place generic CPU hotplug interface and ability to query possible CPUs layout (with QMP command query-hotpluggable-cpus),
however it requires to run QEMU once per each machine type and topology configuration (-M & -smp combination) which would be too taxing for mgmt layer to do.
Currently proposed idea to solve the issue is to do NUMA mapping at runtime:
 1. start QEMU in stopped mode with needed -M & -smp configuration
    but leave out "-numa cpus" options
 2. query possible cpus layout (query-hotpluggable-cpus)
 3. use new QMP command to map CPUs to NUMA node in terms of generic CPU
    hotplug interface (socket/core/thread)
 3.1 potentially translate mapping as new CLI options so the same configuration
     could be started without runtime configuration step
 4. continue VM exection
Comment 2 Igor Mammedov 2017-05-18 08:33:03 UTC 
CLI part of interface is merged in qemu-2.10, it added new CLI option
  -numa cpu=[socket-id/core-id/thread-id]

see commit 419fcdec3 "numa: add '-numa cpu,...' option for property based node mapping"
for more details.
Comment 9 Igor Mammedov 2020-10-20 10:53:54 UTC 
We probably should keep it open (CCing Michal)

PS:
One way to raise interest in implementing it, could be deprecating legacy
-numa node,cpu='' on QEMU side, to 'help' users adopt 'new' interface.
Comment 10 Michal Privoznik 2020-10-20 14:32:11 UTC 
Yeah, for that we firstly need to address bug 1678085 and then we can look into this.
Comment 11 Jaroslav Suchanek 2020-10-30 15:26:22 UTC 
Moving to advanced virt where this will be addressed first. Also resetting the stale date as this is still desired to be supported.
Comment 13 RHEL Program Management 2021-05-01 07:30:14 UTC 
After evaluating this issue, there are no plans to address it further or fix it in an upcoming release.  Therefore, it is being closed.  If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened.
Comment 15 John Ferlan 2021-09-09 16:03:50 UTC 
Bulk update: Move RHEL-AV bugs to RHEL9. If necessary to resolve in RHEL8, then clone to the current RHEL8 release.
Comment 16 Peter Krempa 2021-09-30 13:58:12 UTC 

*** This bug has been marked as a duplicate of bug 1678085 ***