| *** JohnnyW557 is now known as JohnnyW55 | 04:13 | |
| opendevreview | Amit Uniyal proposed openstack/nova master: DNM: enable NFS backend job https://review.opendev.org/c/openstack/nova/+/944008 | 06:27 |
|---|---|---|
| opendevreview | Rajesh Tailor proposed openstack/nova master: Add support for showing instance-action finish_time https://review.opendev.org/c/openstack/nova/+/928933 | 06:51 |
| opendevreview | Michael Still proposed openstack/nova master: libvirt: Add extra spec for sound device. https://review.opendev.org/c/openstack/nova/+/926126 | 08:05 |
| opendevreview | Michael Still proposed openstack/nova master: Protect older compute managers from sound model requests. https://review.opendev.org/c/openstack/nova/+/940770 | 08:05 |
| opendevreview | Michael Still proposed openstack/nova master: libvirt: Add extra specs for USB redirection. https://review.opendev.org/c/openstack/nova/+/927354 | 08:05 |
| gibi | sean-k-mooney: I have a feeling that the pool matching logic in stats.py also intentionally handles the multi spec case. See https://github.com/openstack/nova/blob/0d484ce37d86e989c8abdf57aec5e334f68206ef/nova/pci/utils.py#L75 it basically implements an OR between the multiple specs | 08:22 |
| sean-k-mooney[m] | it did in the really early days | 08:22 |
| sean-k-mooney[m] | but i tough twe removed this like 10+ years ago | 08:23 |
| gibi | appreanlty we did not | 08:23 |
| gibi | PCI in Placement never supported it though I just broke the error reporting leading to http 500 | 08:23 |
| sean-k-mooney[m] | so the vfio-variant dirver live migration depend on 1:1 alias mappings | 08:23 |
| sean-k-mooney[m] | at least when not using pci in placement | 08:24 |
| gibi | sean-k-mooney: do you mean that the live migration will only work if the same exact device is on both sides and therefore having specs in OR relationship with different product id would not work? | 08:24 |
| sean-k-mooney[m] | that is my understanding | 08:25 |
| gibi | yeah I agree | 08:25 |
| sean-k-mooney[m] | i dont se how it could work any other way when the deviece is directly exposed to the guest kernel as a vf | 08:25 |
| sean-k-mooney[m] | i think we need to at least restict when oring can happen | 08:26 |
| gibi | the unplug/plug live migration for neutron requested PCI probably works as that is not real live migration of the device, but the variant driver based will lead to very sad guest kernels | 08:26 |
| sean-k-mooney[m] | yep | 08:27 |
| sean-k-mooney[m] | so i know why they orginally did this which is they were trying to make differnt configurations of intels QAT device look the same\ | 08:28 |
| sean-k-mooney[m] | and allow you to use ask for QAT | 08:28 |
| sean-k-mooney[m] | without careing which one you got | 08:28 |
| sean-k-mooney[m] | that only works if you never have to deal with live migration | 08:29 |
| sean-k-mooney[m] | so i think we need to disable this when live_migratable is set to true or you use pci_in_placment | 08:29 |
| sean-k-mooney[m] | either should result in a config error on startup | 08:30 |
| sean-k-mooney[m] | we can allow it for non llive migratbale device or when not using PCI in placement for backward compatibality | 08:31 |
| opendevreview | Rajesh Tailor proposed openstack/nova master: Add support for showing instance-action finish_time https://review.opendev.org/c/openstack/nova/+/928933 | 08:58 |
| sean-k-mooney | it is day 2 or alt world where my fundemental core belife that "their should only be 1 7:30 in a day and it should happen after noon" is being chllanged by the concept of 7 am... | 09:12 |
| sean-k-mooney | its currntly 09:13 which is still an hour before im useually awake | 09:13 |
| opendevreview | ribaudr proposed openstack/nova master: FUP improve comment accuracy and variable naming for tag removal https://review.opendev.org/c/openstack/nova/+/943124 | 11:43 |
| opendevreview | ribaudr proposed openstack/nova master: FUP Remove unnecessary PCI check https://review.opendev.org/c/openstack/nova/+/944105 | 11:43 |
| opendevreview | ribaudr proposed openstack/nova master: FUP improve and add integration tests for PCI SR-IOV servers https://review.opendev.org/c/openstack/nova/+/944106 | 11:43 |
| andrewbonney | sean-k-mooney: in case you have time today for another look, https://review.opendev.org/c/openstack/nova/+/919961 seems fine in CI | 11:52 |
| sean-k-mooney | andrewbonney: ill review it propelry shortly. if you need to respin it might be nice to add a fixes release note | 12:13 |
| sean-k-mooney | but thats not strictly required | 12:13 |
| andrewbonney | thanks | 12:13 |
| sean-k-mooney | so ya ill take a look shortly | 12:13 |
| sean-k-mooney | andrewbonney: +2 form me | 12:23 |
| sean-k-mooney | andrewbonney: this proably wont merge for RC1 but it can merge and be backported as normal after | 12:24 |
| andrewbonney | sure, just trying to get it ticked off before I head off for extended leave | 12:24 |
| sean-k-mooney | ack | 12:24 |
| opendevreview | Merged openstack/nova master: Fix serial console for ironic https://review.opendev.org/c/openstack/nova/+/942575 | 12:26 |
| *** elodilles is now known as elodilles_pto | 13:22 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!