Enable NUMA support in arm64 builds
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
libvirt (Debian) |
New
|
Unknown
|
|||
libvirt (Ubuntu) |
Fix Released
|
High
|
dann frazier | ||
Xenial |
Fix Released
|
High
|
Stefan Bader |
Bug Description
[Impact]
The host capabilities for arm64 NUMA systems are incomplete - one effect of which is that nova is unable to make use of hugetlbfs for backing VMs with huge pages.
[Test Case]
Examine the output of 'virsh capabilities' on an arm64 NUMA system. The topology section for all but the first node are currently inaccurate and lacks page size entries. Sample output from a 2-node NUMA system is attached, both pre and post fix.
[Regression Risk]
Risk is minimized by the fact that this change is just enabling the same code for arm64 that is already enabled for the other NUMA-enabled Ubuntu architectures.
[racb] What if, for example, numa support is completely broken on arm64 and this is why it was disabled? By landing this, we could be regressing arm64 support.
Changed in libvirt (Ubuntu Xenial): | |
status: | New → Confirmed |
importance: | Undecided → High |
assignee: | nobody → dann frazier (dannf) |
Changed in libvirt (Debian): | |
status: | Unknown → New |
Changed in libvirt (Ubuntu): | |
status: | Confirmed → In Progress |
Changed in libvirt (Ubuntu Xenial): | |
status: | Confirmed → Fix Committed |
tags: |
added: verification-done removed: verification-needed |
tags: |
added: verification-done removed: verification-needed |
Hi Dann, please check how much that would need the on top that is in bug 1621121.
If you just want to get where others already are I think you are fine.
If you expect all numa* stuff the other bug is missing for you (as well as for the others).