)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"31961c8c_c5fb6c25","updated":"2024-12-12 15:08:38.000000000","message":"my first notes","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"50137897_bbe2dd82","updated":"2025-01-07 13:23:47.000000000","message":"Looks mostly OK.I keep my -1 for two reasons:\n* lets be super clear about this feature talks about *live* migration\n* lets settle the upgrade comments.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"5798123ee3ab584376502554e947222175994f0e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"6d6dd958_db7baab3","updated":"2025-01-09 13:29:57.000000000","message":"I\u0027m +2 but I\u0027d want to discuss with Sean about the upgrade concern first","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"79171d0126997e1af993fb5e8af033241216ab5c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"bcd63efc_016c90a9","updated":"2025-01-09 12:46:40.000000000","message":"Overall looks good to me. But we need to settle the upgrade and defaulting questions. I suggest a high bandwidth video call later today.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7b210771ca2db7c619746a507e0770938b6b1749","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"8f348d2e_6027d15f","updated":"2025-01-08 17:57:56.000000000","message":"the object changes are not quite right but this is much closer.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"72f81d9001ab7153f531d9372ab644af908a11a5","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"b4927e52_a3b3fe60","updated":"2025-01-09 17:01:09.000000000","message":"The update should reflect what we discussed about: \n- default (alias and specs) usage.\n- upgrades.\n- data structure used to store pci info.","commit_id":"9c21ae4d38a5d94bf038b77505acaa1b27c3674f"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"68281c83233b680aaaa193efe66418cf4345f00d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"16431dae_1b2a254a","updated":"2025-01-09 17:44:09.000000000","message":"+2 but I think you need to fix the CI failure. No worries if you do that tomorrow.","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"cd8157d54463680783680e1c8e27289db9527d0b","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"0ad71012_22ecebc2","updated":"2025-01-10 08:09:23.000000000","message":"@sbauza@redhat.com was +2 in the previous PS and only CI fix was added in PS7 so I think we are good to go.","commit_id":"9b433f0025b2fd4f81e87f2e595a03808accba0b"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"705d818a_76d4107f","updated":"2025-01-09 18:52:32.000000000","message":"i still have some nits but i think this is close enough to move forward with","commit_id":"9b433f0025b2fd4f81e87f2e595a03808accba0b"}],"specs/2025.1/approved/migrate-vfio-devices-using-kernel-variant-drivers.rst":[{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":false,"context_lines":[{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."},{"line_number":54,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":55,"context_line":""},{"line_number":56,"context_line":""},{"line_number":57,"context_line":"When this tag is encountered by the `PCI resource tracker`__, the"}],"source_content_type":"text/x-rst","patch_set":1,"id":"2b4a86f2_f8477b97","line":54,"updated":"2024-12-12 15:08:38.000000000","message":"+1","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":60,"context_line":""},{"line_number":61,"context_line":".. note::"},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"  The PciDevice object remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, it will record an internal trait, HW_PCI_MIGRATABLE, in the"},{"line_number":66,"context_line":"resource provider. While this trait will not be utilized by the migration"}],"source_content_type":"text/x-rst","patch_set":1,"id":"5a32863c_02f3a16b","line":63,"updated":"2024-12-12 15:08:38.000000000","message":"not necessary and actually not exact, even if we don\u0027t need a new version for the object, the value of the dict will be different","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":60,"context_line":""},{"line_number":61,"context_line":".. note::"},{"line_number":62,"context_line":""},{"line_number":63,"context_line":"  The PciDevice object remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, it will record an internal trait, HW_PCI_MIGRATABLE, in the"},{"line_number":66,"context_line":"resource provider. While this trait will not be utilized by the migration"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ad4bc4b0_1c831cce","line":63,"in_reply_to":"5a32863c_02f3a16b","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":62,"context_line":""},{"line_number":63,"context_line":"  The PciDevice object remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, it will record an internal trait, HW_PCI_MIGRATABLE, in the"},{"line_number":66,"context_line":"resource provider. While this trait will not be utilized by the migration"},{"line_number":67,"context_line":"flow, it can serve as a reference for inventory purposes."},{"line_number":68,"context_line":""},{"line_number":69,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"7584bca3_9349d433","line":66,"range":{"start_line":65,"start_character":0,"end_line":66,"end_character":18},"updated":"2024-12-12 15:08:38.000000000","message":"rather : \"If PCI in Placement option is enabled, then it will record an internal trait for the related RP\"\n\nWe could also call Placement to look at that trait when pre-scheduling.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":62,"context_line":""},{"line_number":63,"context_line":"  The PciDevice object remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, it will record an internal trait, HW_PCI_MIGRATABLE, in the"},{"line_number":66,"context_line":"resource provider. While this trait will not be utilized by the migration"},{"line_number":67,"context_line":"flow, it can serve as a reference for inventory purposes."},{"line_number":68,"context_line":""},{"line_number":69,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"796b7843_ffaa6f48","line":66,"range":{"start_line":65,"start_character":0,"end_line":66,"end_character":18},"in_reply_to":"7584bca3_9349d433","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"0999d85d_3271d4b2","line":83,"updated":"2024-12-12 15:08:38.000000000","message":"correct, because if by default we would not look at whether we should use a migratable PCI device, then when live-migrating to a target, the compute service could then provide back a target PCI device that is not enabled to be migrating (while the original PCI device was able to do it)","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"6b5508a2_52a95dee","line":83,"in_reply_to":"0999d85d_3271d4b2","updated":"2025-01-07 19:41:24.000000000","message":"I\u0027m not entirely convinced we should default to migratable\u003d\u0027no\u0027 if unset.\n\nI would just not express a preference.\n\nlong term the hope woudl be that new hardware woudl embarce this approch \nand migratableity woudl become the default in the ecosystem.\n\nos i sould make this a trindary default to None and only avoid migratbale device if it has live_migratable\u003dno explictly.\n\ni partly want that to be the bevior so that nic will not avoid migatable devices in the filter_pools implmation since they wont express a request for live migratablity\nin there pci request.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"bb2b5d7582002d9f479e7e93ca2be87d9fad5e97","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"4e935898_8b5368c3","line":83,"in_reply_to":"0ed2a3e3_d695944d","updated":"2025-01-09 18:13:13.000000000","message":"I think I have covered that with line 62.\n```\n- If not specified, the default behavior will be equivalent to\n  live_migratable\u003d\u0027no\u0027. However, this value will not be persisted in the\n  PciDevice object.\n```\n\nIs it something else ?","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"4997d1c74095ae62beacf7bb33379c59cc92c7db","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"0875dc76_e8289d7d","line":83,"in_reply_to":"4e935898_8b5368c3","updated":"2025-01-09 20:23:46.000000000","message":"nope that covers it i just had this comment pending for a few yours i wrote it before we chatted.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"7404a5476be0078fa4128b7bb0ed9461f579cd26","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"e9800fb0_c0076dff","line":83,"in_reply_to":"6b5508a2_52a95dee","updated":"2025-01-09 11:37:15.000000000","message":"I\u0027m not sure I understand everything.\nHowever, if we are not setting the default to \u0027no,\u0027 I dislike the fact that running a live migration can result in success or failure depending on the source device initially chosen by the scheduler.\n\nIf the chosen device is not migratable, the live migration will always fail.\nIf the chosen device is migratable, then the live migration can succeed, provided there are other migratable devices available on the host.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"bff7f819954920c637789e44e744042dff995802","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"0ed2a3e3_d695944d","line":83,"in_reply_to":"756fe570_4f15608e","updated":"2025-01-09 17:59:44.000000000","message":"for proparity.\n\nyes if you dont have live_migatble set in the dev spec i dont want to inject\nlive_migratble\u003dno into all the json blobs for existing pci_devices.\n\ni think that pretty heavy wight and woudl prefer to only do that if you add a tag explictly.\n\nwe can set a default in memory when we load the db object to avoid that data migration.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"5798123ee3ab584376502554e947222175994f0e","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"756fe570_4f15608e","line":83,"in_reply_to":"c362f9da_46115a21","updated":"2025-01-09 13:29:57.000000000","message":"Gibi provided exactly my concern when I discussed that with uggla and that\u0027s why we should default to \u0027no\u0027 as the behaviour wouldn\u0027t change when upgrading for pci devices that are not set.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"79171d0126997e1af993fb5e8af033241216ab5c","unresolved":true,"context_lines":[{"line_number":80,"context_line":"  potential migrations to a similar device on another host."},{"line_number":81,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":82,"context_line":"  non-migratable device, making migration impossible."},{"line_number":83,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":84,"context_line":""},{"line_number":85,"context_line":""},{"line_number":86,"context_line":"Live migration modifications:"}],"source_content_type":"text/x-rst","patch_set":1,"id":"c362f9da_46115a21","line":83,"in_reply_to":"e9800fb0_c0076dff","updated":"2025-01-09 12:46:40.000000000","message":"@smooney@redhat.com Do I understand correctly that you suggest not to persist lived_migratable\u003dno in the PCIDevice object if the device_spec does not have the tag set? But still make the implementation work in a way that if PCIDevice.live_migratable is None then the code assumes that such device is not live migratable today and does not try to start the live migration in libvirt. This would be OK to me. \n\nWhat would not be OK to me is to still try to live migrate a VM with a device where PCIDevice.live_migratable is None. As that is just a gamble and might kills the VM in case of a post-copy live migration mode.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":false,"context_lines":[{"line_number":87,"context_line":""},{"line_number":88,"context_line":"Verify in _check_can_migrate_pci() whether the source instance contains"},{"line_number":89,"context_line":"migratable devices. If no migratable devices are found, raise an exception"},{"line_number":90,"context_line":"indicating that the migration is not possible."},{"line_number":91,"context_line":""},{"line_number":92,"context_line":"Modify stats.py in the filter_pools() function to handle PCI requests for"},{"line_number":93,"context_line":"migratable devices. Ensure it retrieves hosts with the appropriate number"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ad23f1eb_da2d56d9","line":90,"updated":"2024-12-12 15:08:38.000000000","message":"+1","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":false,"context_lines":[{"line_number":91,"context_line":""},{"line_number":92,"context_line":"Modify stats.py in the filter_pools() function to handle PCI requests for"},{"line_number":93,"context_line":"migratable devices. Ensure it retrieves hosts with the appropriate number"},{"line_number":94,"context_line":"of migratable devices by adding a new filter."},{"line_number":95,"context_line":""},{"line_number":96,"context_line":"Should we add a supplemental check on destination host to ensure the devices"},{"line_number":97,"context_line":"are migratable ? I don\u0027t think so because the above filter should send use"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ff089c31_2107d3e6","line":94,"updated":"2024-12-12 15:08:38.000000000","message":"+1","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":95,"context_line":""},{"line_number":96,"context_line":"Should we add a supplemental check on destination host to ensure the devices"},{"line_number":97,"context_line":"are migratable ? I don\u0027t think so because the above filter should send use"},{"line_number":98,"context_line":"the correct devices."},{"line_number":99,"context_line":""},{"line_number":100,"context_line":"Inspired by _claim_pci_for_instance_vifs() create a new method to"},{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."}],"source_content_type":"text/x-rst","patch_set":1,"id":"1b7bc0af_5fa3aac5","line":98,"updated":"2024-12-12 15:08:38.000000000","message":"indeed, the filter_pools() will look at whether we need to provide migratable PCI devices by looking at the pci_request\u0027s alias value of the tag you add.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":95,"context_line":""},{"line_number":96,"context_line":"Should we add a supplemental check on destination host to ensure the devices"},{"line_number":97,"context_line":"are migratable ? I don\u0027t think so because the above filter should send use"},{"line_number":98,"context_line":"the correct devices."},{"line_number":99,"context_line":""},{"line_number":100,"context_line":"Inspired by _claim_pci_for_instance_vifs() create a new method to"},{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."}],"source_content_type":"text/x-rst","patch_set":1,"id":"a86a39b9_45bf35d1","line":98,"in_reply_to":"1b7bc0af_5fa3aac5","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":98,"context_line":"the correct devices."},{"line_number":99,"context_line":""},{"line_number":100,"context_line":"Inspired by _claim_pci_for_instance_vifs() create a new method to"},{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"Update the LiveMigrateData or LiveMigrateDataLibvirt object to include the"},{"line_number":104,"context_line":"PCI addresses of the destination hosts."}],"source_content_type":"text/x-rst","patch_set":1,"id":"e3395cd2_74ff6239","line":101,"updated":"2024-12-12 15:08:38.000000000","message":"that\u0027s just an implementation nit, I think you don\u0027t need to tell it or you would rather explain before that you need to change some stuff related to the SRIOV live-migrations since we don\u0027t use VIFs.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":98,"context_line":"the correct devices."},{"line_number":99,"context_line":""},{"line_number":100,"context_line":"Inspired by _claim_pci_for_instance_vifs() create a new method to"},{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"Update the LiveMigrateData or LiveMigrateDataLibvirt object to include the"},{"line_number":104,"context_line":"PCI addresses of the destination hosts."}],"source_content_type":"text/x-rst","patch_set":1,"id":"914f6ff6_999f5f71","line":101,"in_reply_to":"e3395cd2_74ff6239","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":100,"context_line":"Inspired by _claim_pci_for_instance_vifs() create a new method to"},{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"Update the LiveMigrateData or LiveMigrateDataLibvirt object to include the"},{"line_number":104,"context_line":"PCI addresses of the destination hosts."},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"Update _live_migration_operation() to update the destination XML file based on"}],"source_content_type":"text/x-rst","patch_set":1,"id":"71b400f9_830b7992","line":103,"range":{"start_line":103,"start_character":27,"end_line":103,"end_character":53},"updated":"2024-12-12 15:08:38.000000000","message":"remove it, since we don\u0027t have specific checks in the libvirt drivers","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":100,"context_line":"Inspired by _claim_pci_for_instance_vifs() create a new method to"},{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"Update the LiveMigrateData or LiveMigrateDataLibvirt object to include the"},{"line_number":104,"context_line":"PCI addresses of the destination hosts."},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"Update _live_migration_operation() to update the destination XML file based on"}],"source_content_type":"text/x-rst","patch_set":1,"id":"d40a9dfe_2629d9be","line":103,"range":{"start_line":103,"start_character":27,"end_line":103,"end_character":53},"in_reply_to":"71b400f9_830b7992","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"Update the LiveMigrateData or LiveMigrateDataLibvirt object to include the"},{"line_number":104,"context_line":"PCI addresses of the destination hosts."},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"Update _live_migration_operation() to update the destination XML file based on"},{"line_number":107,"context_line":"the PCI addresses collected in the LiveMigrateData or LiveMigrateDataLibvirt"}],"source_content_type":"text/x-rst","patch_set":1,"id":"c50a4ae7_4b4483fc","line":104,"updated":"2024-12-12 15:08:38.000000000","message":"you need to explain the change","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":101,"context_line":"claim_pci_devices and retrieve the PCI addresses of the destination host."},{"line_number":102,"context_line":""},{"line_number":103,"context_line":"Update the LiveMigrateData or LiveMigrateDataLibvirt object to include the"},{"line_number":104,"context_line":"PCI addresses of the destination hosts."},{"line_number":105,"context_line":""},{"line_number":106,"context_line":"Update _live_migration_operation() to update the destination XML file based on"},{"line_number":107,"context_line":"the PCI addresses collected in the LiveMigrateData or LiveMigrateDataLibvirt"}],"source_content_type":"text/x-rst","patch_set":1,"id":"520eaf46_22640f19","line":104,"in_reply_to":"c50a4ae7_4b4483fc","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":false,"context_lines":[{"line_number":105,"context_line":""},{"line_number":106,"context_line":"Update _live_migration_operation() to update the destination XML file based on"},{"line_number":107,"context_line":"the PCI addresses collected in the LiveMigrateData or LiveMigrateDataLibvirt"},{"line_number":108,"context_line":"object."},{"line_number":109,"context_line":""},{"line_number":110,"context_line":""},{"line_number":111,"context_line":"Alternatives"}],"source_content_type":"text/x-rst","patch_set":1,"id":"cd6b1f3b_db21de9b","line":108,"updated":"2024-12-12 15:08:38.000000000","message":"+1","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":117,"context_line":"REST API impact"},{"line_number":118,"context_line":"---------------"},{"line_number":119,"context_line":""},{"line_number":120,"context_line":"NA"},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"Data model impact"},{"line_number":123,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"2e164ca0_dec158e9","line":120,"updated":"2024-12-12 15:08:38.000000000","message":"shouldn\u0027t be but we will need to modify the schema that\u0027s checked on the API https://github.com/openstack/nova/blob/b27447d55dbe6660eae7283ff7c32259d31967c7/nova/pci/request.py#L72-L117\n\nNo microversion needed AFAIK.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":117,"context_line":"REST API impact"},{"line_number":118,"context_line":"---------------"},{"line_number":119,"context_line":""},{"line_number":120,"context_line":"NA"},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"Data model impact"},{"line_number":123,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1a690860_f635f7a5","line":120,"in_reply_to":"2e164ca0_dec158e9","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":122,"context_line":"Data model impact"},{"line_number":123,"context_line":"-----------------"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"LiveMigrateDate or LiveMigrateDataLibvirt object will be extended to supply"},{"line_number":126,"context_line":"the PCI addresses of the destination host."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Security impact"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1aad2949_ea61d660","line":125,"range":{"start_line":125,"start_character":15,"end_line":125,"end_character":42},"updated":"2024-12-12 15:08:38.000000000","message":"remove it","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":122,"context_line":"Data model impact"},{"line_number":123,"context_line":"-----------------"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"LiveMigrateDate or LiveMigrateDataLibvirt object will be extended to supply"},{"line_number":126,"context_line":"the PCI addresses of the destination host."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Security impact"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3fe16ed2_f11ebc96","line":125,"range":{"start_line":125,"start_character":15,"end_line":125,"end_character":42},"in_reply_to":"1aad2949_ea61d660","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":124,"context_line":""},{"line_number":125,"context_line":"LiveMigrateDate or LiveMigrateDataLibvirt object will be extended to supply"},{"line_number":126,"context_line":"the PCI addresses of the destination host."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Security impact"},{"line_number":129,"context_line":"---------------"},{"line_number":130,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"3bd635e5_bd687cd7","line":127,"updated":"2024-12-12 15:08:38.000000000","message":"here that\u0027s where you need to explain the object change.\nI\u0027m proposing to add a ListOfObjects(\u0027PCIDevice\u0027) field so we would have everything we need (including the \u0027managed\u0027 value in the extra_info dict that we need to use for VFIO variant driver PCI devices)","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":124,"context_line":""},{"line_number":125,"context_line":"LiveMigrateDate or LiveMigrateDataLibvirt object will be extended to supply"},{"line_number":126,"context_line":"the PCI addresses of the destination host."},{"line_number":127,"context_line":""},{"line_number":128,"context_line":"Security impact"},{"line_number":129,"context_line":"---------------"},{"line_number":130,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"f2c03332_2f9acb24","line":127,"in_reply_to":"3bd635e5_bd687cd7","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":143,"context_line":""},{"line_number":144,"context_line":"- Device specifications and aliases."},{"line_number":145,"context_line":"- Flavors."},{"line_number":146,"context_line":""},{"line_number":147,"context_line":"Performance Impact"},{"line_number":148,"context_line":"------------------"},{"line_number":149,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"b401e366_058d8583","line":146,"updated":"2024-12-12 15:08:38.000000000","message":"there is a huge impact : users will need to use different flavors per GPU type.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":143,"context_line":""},{"line_number":144,"context_line":"- Device specifications and aliases."},{"line_number":145,"context_line":"- Flavors."},{"line_number":146,"context_line":""},{"line_number":147,"context_line":"Performance Impact"},{"line_number":148,"context_line":"------------------"},{"line_number":149,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"7b7bb1b1_9be6827e","line":146,"in_reply_to":"b401e366_058d8583","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":147,"context_line":"Performance Impact"},{"line_number":148,"context_line":"------------------"},{"line_number":149,"context_line":""},{"line_number":150,"context_line":"This `bug`__ should be taken into account as it may impact performance."},{"line_number":151,"context_line":""},{"line_number":152,"context_line":"`Mitigation measures`__ are currently being developed to minimize this impact."},{"line_number":153,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"c41024ff_6d7599ee","line":150,"updated":"2024-12-12 15:08:38.000000000","message":"only if PCI in Placement is enabled (which is default to false as of now), we just need to see at the end of Epoxy whether we need to document that.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"cfd5bccd40609cefb0b78003497f2d60bcec9f3e","unresolved":false,"context_lines":[{"line_number":147,"context_line":"Performance Impact"},{"line_number":148,"context_line":"------------------"},{"line_number":149,"context_line":""},{"line_number":150,"context_line":"This `bug`__ should be taken into account as it may impact performance."},{"line_number":151,"context_line":""},{"line_number":152,"context_line":"`Mitigation measures`__ are currently being developed to minimize this impact."},{"line_number":153,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"2aca0c38_0e334872","line":150,"in_reply_to":"c41024ff_6d7599ee","updated":"2024-12-12 20:26:49.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":173,"context_line":"- The configuration of device specifications and aliases."},{"line_number":174,"context_line":"- Update their flavors to specify the correct aliases."},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"In Nova\u0027s data model, this is only achievable via a resize operation."},{"line_number":177,"context_line":""},{"line_number":178,"context_line":"Performing a reshape in place is not feasible because it would require"},{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"}],"source_content_type":"text/x-rst","patch_set":1,"id":"1f1cd62d_c921a5c0","line":176,"updated":"2024-12-12 15:08:38.000000000","message":"Which is a stopping operation, which is a HUGE upgrade impact.\nSince vGPUs are something very different in previous releases, this is something we can\u0027t just migrate from previously to now (and that\u0027s horribly sad)","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":false,"context_lines":[{"line_number":173,"context_line":"- The configuration of device specifications and aliases."},{"line_number":174,"context_line":"- Update their flavors to specify the correct aliases."},{"line_number":175,"context_line":""},{"line_number":176,"context_line":"In Nova\u0027s data model, this is only achievable via a resize operation."},{"line_number":177,"context_line":""},{"line_number":178,"context_line":"Performing a reshape in place is not feasible because it would require"},{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"}],"source_content_type":"text/x-rst","patch_set":1,"id":"7ac7003f_0a4b44d4","line":176,"in_reply_to":"1f1cd62d_c921a5c0","updated":"2025-01-08 09:16:35.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":177,"context_line":""},{"line_number":178,"context_line":"Performing a reshape in place is not feasible because it would require"},{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"},{"line_number":180,"context_line":"entries, and placement inventories."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Given the complexity involved, this approach should not be considered."},{"line_number":183,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"b7e2a0e1_fed41658","line":180,"updated":"2024-12-12 15:08:38.000000000","message":"we are not able to know the resource classes, because they couldn\u0027t be only VGPU so that\u0027s also a difficult question to fix it.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":177,"context_line":""},{"line_number":178,"context_line":"Performing a reshape in place is not feasible because it would require"},{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"},{"line_number":180,"context_line":"entries, and placement inventories."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Given the complexity involved, this approach should not be considered."},{"line_number":183,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"f50e80b6_35cd5994","line":180,"in_reply_to":"b7e2a0e1_fed41658","updated":"2025-01-07 19:41:24.000000000","message":"we can proabvly shorten this to \n\nIn Nova\u0027s data model, the only way to move between mdev pashtough to vrio-variant driver is through a resize operation, which requires stopping the instance.\nPerforming a reshape in place is not feasible.\n\nwith that said i woudl be inclined to delete this paragraph entirely\n\nMoving from our current VGPU feature to this new one is not in the scope fo enabling live migration capability for VF with vfio variant drivers.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":177,"context_line":""},{"line_number":178,"context_line":"Performing a reshape in place is not feasible because it would require"},{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"},{"line_number":180,"context_line":"entries, and placement inventories."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Given the complexity involved, this approach should not be considered."},{"line_number":183,"context_line":""}],"source_content_type":"text/x-rst","patch_set":1,"id":"625a88cf_ca5aac9c","line":180,"in_reply_to":"f50e80b6_35cd5994","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"4d9df698ec15e61cbfbc935ce96835875a735fcd","unresolved":true,"context_lines":[{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"},{"line_number":180,"context_line":"entries, and placement inventories."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Given the complexity involved, this approach should not be considered."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ded069ed_0163b9bc","line":182,"range":{"start_line":182,"start_character":0,"end_line":182,"end_character":2},"updated":"2024-12-12 15:08:38.000000000","message":"that part should mostly be in the alternatives section","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"},{"line_number":180,"context_line":"entries, and placement inventories."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Given the complexity involved, this approach should not be considered."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":1,"id":"f951a93b_fd577327","line":182,"range":{"start_line":182,"start_character":0,"end_line":182,"end_character":2},"in_reply_to":"ded069ed_0163b9bc","updated":"2025-01-07 19:41:24.000000000","message":"we can proably drop this line.","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":179,"context_line":"changes to allocations, flavor content, configuration values, database"},{"line_number":180,"context_line":"entries, and placement inventories."},{"line_number":181,"context_line":""},{"line_number":182,"context_line":"Given the complexity involved, this approach should not be considered."},{"line_number":183,"context_line":""},{"line_number":184,"context_line":""},{"line_number":185,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":1,"id":"635c1ab1_918ef7ac","line":182,"range":{"start_line":182,"start_character":0,"end_line":182,"end_character":2},"in_reply_to":"f951a93b_fd577327","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"82744730b28336487fed793b50f53c79bbc63f78"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":19,"context_line":"Support for devices using the variant driver interface is detailed in this"},{"line_number":20,"context_line":"`specification`__."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"However, the migration process is not covered here. This is addressed in"},{"line_number":23,"context_line":"the following section, which describes the Nova updates required to migrate"},{"line_number":24,"context_line":"SR-IOV devices to the new kernel variant driver interface."},{"line_number":25,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"44d66059_913da484","line":22,"range":{"start_line":22,"start_character":46,"end_line":22,"end_character":50},"updated":"2024-12-13 13:32:51.000000000","message":"there","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":19,"context_line":"Support for devices using the variant driver interface is detailed in this"},{"line_number":20,"context_line":"`specification`__."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"However, the migration process is not covered here. This is addressed in"},{"line_number":23,"context_line":"the following section, which describes the Nova updates required to migrate"},{"line_number":24,"context_line":"SR-IOV devices to the new kernel variant driver interface."},{"line_number":25,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3dd94b3b_4ab21807","line":22,"range":{"start_line":22,"start_character":46,"end_line":22,"end_character":50},"in_reply_to":"44d66059_913da484","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":28,"context_line":"Use Cases"},{"line_number":29,"context_line":"---------"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate SR-IOV devices that require variant"},{"line_number":32,"context_line":"  drivers."},{"line_number":33,"context_line":""},{"line_number":34,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":35,"context_line":"  non-migratable."}],"source_content_type":"text/x-rst","patch_set":2,"id":"4687d77b_8362516f","line":32,"range":{"start_line":31,"start_character":0,"end_line":32,"end_character":10},"updated":"2024-12-13 13:32:51.000000000","message":"nit ... I want to migrate VMs with SR-IOV devices if such operation is supported by the variant driver.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":28,"context_line":"Use Cases"},{"line_number":29,"context_line":"---------"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate SR-IOV devices that require variant"},{"line_number":32,"context_line":"  drivers."},{"line_number":33,"context_line":""},{"line_number":34,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":35,"context_line":"  non-migratable."}],"source_content_type":"text/x-rst","patch_set":2,"id":"cd04a61a_62e78cf0","line":32,"range":{"start_line":31,"start_character":0,"end_line":32,"end_character":10},"in_reply_to":"4687d77b_8362516f","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":42,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"Configuring PCI device specification:"},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"Users must specify whether the device is eligible for migration to a similar"},{"line_number":47,"context_line":"device on another compute node."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":50,"context_line":"specification."}],"source_content_type":"text/x-rst","patch_set":2,"id":"21e6a750_58d339d6","line":47,"range":{"start_line":45,"start_character":0,"end_line":47,"end_character":31},"updated":"2024-12-13 13:32:51.000000000","message":"nit: Not the user. It is the admin/deployer","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":42,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":43,"context_line":""},{"line_number":44,"context_line":"Configuring PCI device specification:"},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"Users must specify whether the device is eligible for migration to a similar"},{"line_number":47,"context_line":"device on another compute node."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":50,"context_line":"specification."}],"source_content_type":"text/x-rst","patch_set":2,"id":"158ea480_49a23415","line":47,"range":{"start_line":45,"start_character":0,"end_line":47,"end_character":31},"in_reply_to":"21e6a750_58d339d6","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":47,"context_line":"device on another compute node."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":50,"context_line":"specification."},{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."}],"source_content_type":"text/x-rst","patch_set":2,"id":"4b29bee0_88131c87","line":50,"updated":"2024-12-13 13:32:51.000000000","message":"add a ``migratable`` tag to the device specification in ``[pci]dev_spec`` config.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":47,"context_line":"device on another compute node."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":50,"context_line":"specification."},{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."}],"source_content_type":"text/x-rst","patch_set":2,"id":"bf70ce0e_a97ea7f6","line":50,"in_reply_to":"4b29bee0_88131c87","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":49,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":50,"context_line":"specification."},{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."},{"line_number":54,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":55,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"4859b01c_b6e77516","line":52,"updated":"2024-12-13 13:32:51.000000000","message":"... can be live migrated","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":49,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":50,"context_line":"specification."},{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."},{"line_number":54,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":55,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3a990083_c060491c","line":52,"in_reply_to":"4859b01c_b6e77516","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":50,"context_line":"specification."},{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."},{"line_number":54,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":55,"context_line":""},{"line_number":56,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"a1c8a45c_dc5cec7b","line":53,"updated":"2024-12-13 13:32:51.000000000","message":"... cannot be live migrated","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":50,"context_line":"specification."},{"line_number":51,"context_line":""},{"line_number":52,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be migrated."},{"line_number":53,"context_line":"- ``migratable\u003d\u0027no\u0027`` means that the device cannot be migrated."},{"line_number":54,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"},{"line_number":55,"context_line":""},{"line_number":56,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"06141828_b1f118ed","line":53,"in_reply_to":"a1c8a45c_dc5cec7b","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":62,"context_line":""},{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."}],"source_content_type":"text/x-rst","patch_set":2,"id":"5ee5305a_49d23a21","line":65,"range":{"start_line":65,"start_character":64,"end_line":65,"end_character":72},"updated":"2024-12-13 13:32:51.000000000","message":"a new standard","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":62,"context_line":""},{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."}],"source_content_type":"text/x-rst","patch_set":2,"id":"e0d0cd11_7c2b6302","line":65,"range":{"start_line":65,"start_character":64,"end_line":65,"end_character":72},"in_reply_to":"5ee5305a_49d23a21","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"67fa8bb3_6f4f589e","line":66,"range":{"start_line":66,"start_character":7,"end_line":66,"end_character":24},"updated":"2024-12-13 13:32:51.000000000","message":"After reading the spec mixing migration with live migration, I\u0027m wondering if it would be better naming this trait as HW_PCI_LIVE_MIGRATABLE","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"13fb50de_b7bfb4a0","line":66,"range":{"start_line":66,"start_character":33,"end_line":66,"end_character":50},"updated":"2024-12-13 13:32:51.000000000","message":"resource provider representing the physical device.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"921e373c_01dcba78","line":66,"range":{"start_line":66,"start_character":33,"end_line":66,"end_character":50},"in_reply_to":"13fb50de_b7bfb4a0","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"f26f6e0a_393286da","line":66,"range":{"start_line":66,"start_character":7,"end_line":66,"end_character":24},"in_reply_to":"67fa8bb3_6f4f589e","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""},{"line_number":70,"context_line":".. note::"},{"line_number":71,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"7b0ffdf3_6aa2b395","line":68,"range":{"start_line":66,"start_character":52,"end_line":68,"end_character":41},"updated":"2024-12-13 13:32:51.000000000","message":"This can be dropped or rephrased to. The PCI in Placement code path can be extended later to automatically request this trait if the PCI alias requests migratable\u003dyes device(s)","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":false,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""},{"line_number":70,"context_line":".. note::"},{"line_number":71,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9a7fdea7_10971258","line":68,"range":{"start_line":66,"start_character":52,"end_line":68,"end_character":41},"in_reply_to":"529dc5ee_b5bd4faf","updated":"2025-01-07 13:23:47.000000000","message":"Looks OK to me.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":true,"context_lines":[{"line_number":63,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Additionally, if pci in placement is enabled, it will record an internal"},{"line_number":66,"context_line":"trait, HW_PCI_MIGRATABLE, in the resource provider. While this trait will"},{"line_number":67,"context_line":"not be utilized by the migration flow, it can serve as a reference for"},{"line_number":68,"context_line":"inventory or for pre-scheduling purposes."},{"line_number":69,"context_line":""},{"line_number":70,"context_line":".. note::"},{"line_number":71,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"529dc5ee_b5bd4faf","line":68,"range":{"start_line":66,"start_character":52,"end_line":68,"end_character":41},"in_reply_to":"7b0ffdf3_6aa2b395","updated":"2025-01-03 17:11:30.000000000","message":"I have not entirely replaced I keep the idea of inventory. Let me know what you think about it.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":79,"context_line":"requires a migratable device."},{"line_number":80,"context_line":""},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"The proposed solution is to add a new ``migratable`` key to the PCI alias"},{"line_number":83,"context_line":"definition."},{"line_number":84,"context_line":""},{"line_number":85,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the user wants a migratable device allowing"},{"line_number":86,"context_line":"  potential migrations to a similar device on another host."}],"source_content_type":"text/x-rst","patch_set":2,"id":"4947894b_d290a3ae","line":83,"range":{"start_line":82,"start_character":0,"end_line":83,"end_character":11},"updated":"2024-12-13 13:32:51.000000000","message":"... PCI alias definition in the [pci]alias config.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":79,"context_line":"requires a migratable device."},{"line_number":80,"context_line":""},{"line_number":81,"context_line":""},{"line_number":82,"context_line":"The proposed solution is to add a new ``migratable`` key to the PCI alias"},{"line_number":83,"context_line":"definition."},{"line_number":84,"context_line":""},{"line_number":85,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the user wants a migratable device allowing"},{"line_number":86,"context_line":"  potential migrations to a similar device on another host."}],"source_content_type":"text/x-rst","patch_set":2,"id":"36ad8dff_1b2f6e85","line":83,"range":{"start_line":82,"start_character":0,"end_line":83,"end_character":11},"in_reply_to":"4947894b_d290a3ae","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":83,"context_line":"definition."},{"line_number":84,"context_line":""},{"line_number":85,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the user wants a migratable device allowing"},{"line_number":86,"context_line":"  potential migrations to a similar device on another host."},{"line_number":87,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":88,"context_line":"  non-migratable device, making migration impossible."},{"line_number":89,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"}],"source_content_type":"text/x-rst","patch_set":2,"id":"8c05887b_187a40c1","line":86,"updated":"2024-12-13 13:32:51.000000000","message":"... means that the user wants a device(s) allowing live migration to a similar device(s) on another host.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":83,"context_line":"definition."},{"line_number":84,"context_line":""},{"line_number":85,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the user wants a migratable device allowing"},{"line_number":86,"context_line":"  potential migrations to a similar device on another host."},{"line_number":87,"context_line":"- ``migratable\u003d\u0027no\u0027`` This explicitly indicates that the user requires a"},{"line_number":88,"context_line":"  non-migratable device, making migration impossible."},{"line_number":89,"context_line":"- If not specified then the default is ``migratable\u003d\u0027no\u0027``"}],"source_content_type":"text/x-rst","patch_set":2,"id":"4771d70f_ba743baa","line":86,"in_reply_to":"8c05887b_187a40c1","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":92,"context_line":"Live migration modifications:"},{"line_number":93,"context_line":""},{"line_number":94,"context_line":"Verify in _check_can_migrate_pci() whether the source instance contains"},{"line_number":95,"context_line":"migratable devices. If no migratable devices are found, raise an exception"},{"line_number":96,"context_line":"indicating that the migration is not possible."},{"line_number":97,"context_line":""},{"line_number":98,"context_line":"Modify stats.py in the filter_pools() function to handle PCI requests for"}],"source_content_type":"text/x-rst","patch_set":2,"id":"5720da9a_d6502d59","line":95,"range":{"start_line":95,"start_character":20,"end_line":95,"end_character":55},"updated":"2024-12-13 13:32:51.000000000","message":"note that the VM on the source host might have PCI devices attached that are not related to any PCI alias, but it is there because of neutron direct or direct-physical ports. In this case nova should do what it does today, detach these ports at the start of the migration and re-attach them on the dest after the migration. Also such PCI devices having no migratable\u003dyes key in their extra_info should not prevent the live migration to be accepted.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":92,"context_line":"Live migration modifications:"},{"line_number":93,"context_line":""},{"line_number":94,"context_line":"Verify in _check_can_migrate_pci() whether the source instance contains"},{"line_number":95,"context_line":"migratable devices. If no migratable devices are found, raise an exception"},{"line_number":96,"context_line":"indicating that the migration is not possible."},{"line_number":97,"context_line":""},{"line_number":98,"context_line":"Modify stats.py in the filter_pools() function to handle PCI requests for"}],"source_content_type":"text/x-rst","patch_set":2,"id":"73bd4920_3970cabe","line":95,"range":{"start_line":95,"start_character":20,"end_line":95,"end_character":55},"in_reply_to":"5720da9a_d6502d59","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":112,"context_line":""},{"line_number":113,"context_line":""},{"line_number":114,"context_line":".. __: https://github.com/openstack/nova/blob/f98f414f971b6c897bf48781a579730419b5a93d/nova/compute/pci_placement_translator.py#L597-L600"},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"Alternatives"},{"line_number":117,"context_line":"------------"},{"line_number":118,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"25e2a295_4d59dba6","line":115,"updated":"2024-12-13 13:32:51.000000000","message":"Please note that\n* If PCI in Placement is enabled then live migration will work as today for neutron requested PCI devices (i.e. legacy behavior works)\n* If PCI in Placement is enabled then SR-IOV live migration proposed in this spec will still work (i.e. new functionality works)\n* Optionally PCI in Placement will be extended to automatically request HW_PCI_MIGRATABLE trait if the alias has migratable\u003d\"yes\".\n   * A further enhancement would be to extend the translation of the [pci]alias spec to placement RequestGroups to support forbidden traits. So when migratable\u003dno is present in the alias the HW_PCI_MIGRATABLE trait is requested as frobidden.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":112,"context_line":""},{"line_number":113,"context_line":""},{"line_number":114,"context_line":".. __: https://github.com/openstack/nova/blob/f98f414f971b6c897bf48781a579730419b5a93d/nova/compute/pci_placement_translator.py#L597-L600"},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"Alternatives"},{"line_number":117,"context_line":"------------"},{"line_number":118,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"9eeeaad2_a82e13a7","line":115,"in_reply_to":"25e2a295_4d59dba6","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":183,"context_line":""},{"line_number":184,"context_line":"Upgrade impact"},{"line_number":185,"context_line":"--------------"},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Users should review the following:"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"- The configuration of device specifications and aliases."}],"source_content_type":"text/x-rst","patch_set":2,"id":"ad871849_31da49d0","line":186,"updated":"2024-12-13 13:32:51.000000000","message":"This is a copy-paste from the parent spec. This makes me wonder what is the base version of the upgrade described here.\na) It is Dalmatian. Then we need the full consideration of how to move mdev based VGPU users to SR-IOV based VGPUs.\nb) It is the implementation of the previous spec which means the affected users are moved to SR-IOV already.\n\nIn any case we need to consider that there might be a user using SR-IOV GPUs based on the parent spec. And then want to upgrade to the implementation of this spec. How can that user change an existing VM from non migratable to migratable. I think the answer is resize again. As changing the alias config for a running VM is not feasible.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"bdbdd6f93aa54e975a7d093edfc9b09cab9808f2","unresolved":true,"context_lines":[{"line_number":183,"context_line":""},{"line_number":184,"context_line":"Upgrade impact"},{"line_number":185,"context_line":"--------------"},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Users should review the following:"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"- The configuration of device specifications and aliases."}],"source_content_type":"text/x-rst","patch_set":2,"id":"7b2ecf27_e5163c96","line":186,"in_reply_to":"38c29ccb_e99c87b4","updated":"2025-01-08 11:25:26.000000000","message":"\u003e for the live migration feature however there isned realy any upgrade impact\n\u003e \n\u003e the admin jsut need to upgrade the alias and device_spect to makr live migratable devices as live migratiabel and restart the comptue agent\n\u003e\n\u003e there is no need to create new flavors or resize in that case.\n\nHm, I might be mixing up something but I thought that the pci alias also need the migratable\u003dtrue flag so that when the live migration is requested the scheduler will try to find a target host with a migratable device. See L79-89 above.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"5798123ee3ab584376502554e947222175994f0e","unresolved":false,"context_lines":[{"line_number":183,"context_line":""},{"line_number":184,"context_line":"Upgrade impact"},{"line_number":185,"context_line":"--------------"},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Users should review the following:"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"- The configuration of device specifications and aliases."}],"source_content_type":"text/x-rst","patch_set":2,"id":"2a32fa57_de274457","line":186,"in_reply_to":"4dab161e_4092930a","updated":"2025-01-09 13:29:57.000000000","message":"Acknowledged","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7b210771ca2db7c619746a507e0770938b6b1749","unresolved":true,"context_lines":[{"line_number":183,"context_line":""},{"line_number":184,"context_line":"Upgrade impact"},{"line_number":185,"context_line":"--------------"},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Users should review the following:"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"- The configuration of device specifications and aliases."}],"source_content_type":"text/x-rst","patch_set":2,"id":"f8de7cad_17bf8ae0","line":186,"in_reply_to":"7b2ecf27_e5163c96","updated":"2025-01-08 17:57:56.000000000","message":"so we do not use the alis direclty wehn selcting a device\n\nthe request for a pci device is carried by the approately named  InstancePCIRequest\nobject\n\nhttps://github.com/openstack/nova/blob/master/nova/objects/instance_pci_requests.py#L21-L47\n\nthis is contcxuted in the api with refence to the pci apias and then included in the request spec.\n\nonly the api and compute nodes have the alis defiend in tehre config so its not avaiable to the schdluer.\n\nas such we will need to extend the https://github.com/openstack/nova/blob/master/nova/objects/instance_pci_requests.py#L21-L47 with a new live_migtabel filed\n\nas we did when we added the numa policy \n\nhttps://github.com/openstack/nova/commit/0cd858b4ab6a500cb6c20447bd7bc1c7136e1bc5\n\n\nwe should actully update the spec to incldue that change in thge \"Data model impact\" section.\n\n\nfor neutron based port request teh alis name file dis empty which is how we know if its a flavor or port based request\n\nhttps://github.com/openstack/nova/blob/master/nova/objects/instance_pci_requests.py#L48-L54","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":183,"context_line":""},{"line_number":184,"context_line":"Upgrade impact"},{"line_number":185,"context_line":"--------------"},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Users should review the following:"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"- The configuration of device specifications and aliases."}],"source_content_type":"text/x-rst","patch_set":2,"id":"38c29ccb_e99c87b4","line":186,"in_reply_to":"ad871849_31da49d0","updated":"2025-01-07 19:41:24.000000000","message":"to be honest there is very little upgade impact.\n\nmovign from mdev based vgpus to variatn drivers is changing the underly ign technology and howe its modled in nova\n\nso we cant in place upgrade them without a resize.\n\nso i dont think we shoudl be providing any upgrade path outside of that.\nadmisn cna defien new flavors with the new alianse and then user can resize.\n\n\nfor the live migration feature however there isned realy any upgrade impact\n\nthe admin jsut need to upgrade the alias and device_spect to makr live migratable devices as live migratiabel and restart the comptue agent\n\nthere is no need to create new flavors or resize in that case.\n\nwe will need a min compute server version check to know if we shoudl reject the request in the api or defer ti to the condocutor but beyond that the upgrade impact of this is very small.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"79171d0126997e1af993fb5e8af033241216ab5c","unresolved":true,"context_lines":[{"line_number":183,"context_line":""},{"line_number":184,"context_line":"Upgrade impact"},{"line_number":185,"context_line":"--------------"},{"line_number":186,"context_line":""},{"line_number":187,"context_line":"Users should review the following:"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"- The configuration of device specifications and aliases."}],"source_content_type":"text/x-rst","patch_set":2,"id":"4dab161e_4092930a","line":186,"in_reply_to":"f8de7cad_17bf8ae0","updated":"2025-01-09 12:46:40.000000000","message":"You are right but I think missed my point. As you write the InstancePCIRequest is generated from the alias config (in case of flavor based request). The scheduler acts based on the InstancePCIRequest to find a dest host for the live migration. So the InstancePCIRequest needs to request a migratable device during live migration. Therefore the alias the InstancePCIRequest is generated from needs to request a migratable device.\n\nA VM that started today (or started with the feature described with the previous spec implementing managed\u003dno) has no data about migratability in the InstancePCIRequest stored in the RequestSpec. So even if the alias is updated in the config, the VM\u0027s IntancePCIRequests persisted in the request_spec will not be re-created based on the new alias tags. Therefore when this VM is live migrated the VM will be rejected as not live migratable.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":219,"context_line":"Work Items"},{"line_number":220,"context_line":"----------"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."}],"source_content_type":"text/x-rst","patch_set":2,"id":"ff37e36a_bc95f49b","line":222,"updated":"2024-12-13 13:32:51.000000000","message":"* [pci]device_spec\n* s/managed/migratable/\n* also the [pci]alias parsing needs to be extended to support the new migratable parameter.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":false,"context_lines":[{"line_number":219,"context_line":"Work Items"},{"line_number":220,"context_line":"----------"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."}],"source_content_type":"text/x-rst","patch_set":2,"id":"4377f47a_a4cb5ca6","line":222,"in_reply_to":"f5d245cd_fcff7c35","updated":"2025-01-08 09:16:35.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":219,"context_line":"Work Items"},{"line_number":220,"context_line":"----------"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."}],"source_content_type":"text/x-rst","patch_set":2,"id":"20261d15_4aff5ebb","line":222,"in_reply_to":"ff37e36a_bc95f49b","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7e3ed641562dc287be22da5a26417e6373d40b7f","unresolved":true,"context_lines":[{"line_number":219,"context_line":"Work Items"},{"line_number":220,"context_line":"----------"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."}],"source_content_type":"text/x-rst","patch_set":2,"id":"f5d245cd_fcff7c35","line":222,"in_reply_to":"ff37e36a_bc95f49b","updated":"2025-01-07 13:42:40.000000000","message":"we need both managed and migratabel.\n\nthey are modeling two different things.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":220,"context_line":"----------"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."},{"line_number":226,"context_line":"- Unit tests + functional tests."}],"source_content_type":"text/x-rst","patch_set":2,"id":"dfe98fa9_ecb83e81","line":223,"updated":"2024-12-13 13:32:51.000000000","message":"drop this as it is not described in this spec","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":220,"context_line":"----------"},{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."},{"line_number":226,"context_line":"- Unit tests + functional tests."}],"source_content_type":"text/x-rst","patch_set":2,"id":"039a120f_e2154f17","line":223,"in_reply_to":"dfe98fa9_ecb83e81","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."},{"line_number":226,"context_line":"- Unit tests + functional tests."},{"line_number":227,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"3c7497a7_4b5064a5","line":224,"updated":"2024-12-13 13:32:51.000000000","message":"that feels like a copy paste from the parent spec. The XML does not change directly due to the migratable flag","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":221,"context_line":""},{"line_number":222,"context_line":"- Parse managed parameter from whitelist."},{"line_number":223,"context_line":"- Sanitize device specification."},{"line_number":224,"context_line":"- Change XML generation to deal with migratable parameter."},{"line_number":225,"context_line":"- Documentation updates."},{"line_number":226,"context_line":"- Unit tests + functional tests."},{"line_number":227,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"11a2755c_e62866bb","line":224,"in_reply_to":"3c7497a7_4b5064a5","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"384cb9dfc30d368a62eddf10105937c9f7ef05b5","unresolved":true,"context_lines":[{"line_number":228,"context_line":"Dependencies"},{"line_number":229,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":230,"context_line":""},{"line_number":231,"context_line":"- Performance impact bug."},{"line_number":232,"context_line":""},{"line_number":233,"context_line":""},{"line_number":234,"context_line":"Testing"}],"source_content_type":"text/x-rst","patch_set":2,"id":"39ad28ff_4133ce20","line":231,"updated":"2024-12-13 13:32:51.000000000","message":"I guess the parent spec is a direct hard dependency.","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"ace9c901d5b792531bb7a56e07220d68435e5ee3","unresolved":false,"context_lines":[{"line_number":228,"context_line":"Dependencies"},{"line_number":229,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":230,"context_line":""},{"line_number":231,"context_line":"- Performance impact bug."},{"line_number":232,"context_line":""},{"line_number":233,"context_line":""},{"line_number":234,"context_line":"Testing"}],"source_content_type":"text/x-rst","patch_set":2,"id":"70a2b1ae_d5a33219","line":231,"in_reply_to":"39ad28ff_4133ce20","updated":"2025-01-03 17:11:30.000000000","message":"Done","commit_id":"d21e0686c0b114bd78460c07f27e3c60d73562ef"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/migrate-vfio-devices-using-kernel-variant-drivers"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This spec outlines the necessary steps to migrate SR-IOV devices"},{"line_number":14,"context_line":"using the new kernel variant driver interface."},{"line_number":15,"context_line":""},{"line_number":16,"context_line":"Problem description"}],"source_content_type":"text/x-rst","patch_set":3,"id":"a78f469d_01b2e8e4","line":13,"range":{"start_line":13,"start_character":42,"end_line":13,"end_character":49},"updated":"2025-01-07 13:23:47.000000000","message":"live migrate\n\nThis also makes me realize that cold migration supposed to work out of the box. Maybe it make sense to emphasize that this spec only talks about live migration support.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/migrate-vfio-devices-using-kernel-variant-drivers"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This spec outlines the necessary steps to migrate SR-IOV devices"},{"line_number":14,"context_line":"using the new kernel variant driver interface."},{"line_number":15,"context_line":""},{"line_number":16,"context_line":"Problem description"}],"source_content_type":"text/x-rst","patch_set":3,"id":"8768bf51_1815f924","line":13,"range":{"start_line":13,"start_character":42,"end_line":13,"end_character":49},"in_reply_to":"0d1ef14c_aa601ec5","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/migrate-vfio-devices-using-kernel-variant-drivers"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This spec outlines the necessary steps to migrate SR-IOV devices"},{"line_number":14,"context_line":"using the new kernel variant driver interface."},{"line_number":15,"context_line":""},{"line_number":16,"context_line":"Problem description"}],"source_content_type":"text/x-rst","patch_set":3,"id":"0d1ef14c_aa601ec5","line":13,"range":{"start_line":13,"start_character":42,"end_line":13,"end_character":49},"in_reply_to":"a78f469d_01b2e8e4","updated":"2025-01-08 09:16:35.000000000","message":"+1","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/migrate-vfio-devices-using-kernel-variant-drivers"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This spec outlines the necessary steps to migrate SR-IOV devices"},{"line_number":14,"context_line":"using the new kernel variant driver interface."},{"line_number":15,"context_line":""},{"line_number":16,"context_line":"Problem description"},{"line_number":17,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"55a17f0f_a02eae97","line":14,"range":{"start_line":14,"start_character":13,"end_line":14,"end_character":21},"updated":"2025-01-08 09:16:35.000000000","message":"vfio-pci variant driver kernel interface","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":11,"context_line":"https://blueprints.launchpad.net/nova/+spec/migrate-vfio-devices-using-kernel-variant-drivers"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"This spec outlines the necessary steps to migrate SR-IOV devices"},{"line_number":14,"context_line":"using the new kernel variant driver interface."},{"line_number":15,"context_line":""},{"line_number":16,"context_line":"Problem description"},{"line_number":17,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"2a0a15c6_4436dd4f","line":14,"range":{"start_line":14,"start_character":13,"end_line":14,"end_character":21},"in_reply_to":"55a17f0f_a02eae97","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":20,"context_line":"`specification`__."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"However, the migration process is not covered there. This is addressed in"},{"line_number":23,"context_line":"the following section, which describes the Nova updates required to migrate"},{"line_number":24,"context_line":"SR-IOV devices to the new kernel variant driver interface."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":".. __: https://specs.openstack.org/openstack/nova-specs/specs/2025.1/approved/enable-vfio-devices-with-kernel-variant-drivers.html"},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Use Cases"}],"source_content_type":"text/x-rst","patch_set":3,"id":"1e464523_c8b414c6","line":25,"range":{"start_line":23,"start_character":29,"end_line":25,"end_character":0},"updated":"2025-01-07 13:23:47.000000000","message":"I missed this last time. I think the main goal of this spec is not to describe a migration path to the new kernel variant driver interface. It is more like to describe the way nova could live migration VMs with devices that are (already) using SR-IOV devices with kernel variant drivers that support live migration.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7e3ed641562dc287be22da5a26417e6373d40b7f","unresolved":true,"context_lines":[{"line_number":20,"context_line":"`specification`__."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"However, the migration process is not covered there. This is addressed in"},{"line_number":23,"context_line":"the following section, which describes the Nova updates required to migrate"},{"line_number":24,"context_line":"SR-IOV devices to the new kernel variant driver interface."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":".. __: https://specs.openstack.org/openstack/nova-specs/specs/2025.1/approved/enable-vfio-devices-with-kernel-variant-drivers.html"},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Use Cases"}],"source_content_type":"text/x-rst","patch_set":3,"id":"505ce0dc_153f721f","line":25,"range":{"start_line":23,"start_character":29,"end_line":25,"end_character":0},"in_reply_to":"1e464523_c8b414c6","updated":"2025-01-07 13:42:40.000000000","message":"correct.\n\nthis this spec is inteded to decribe the changes requried to enable live migration\n\nit is not in socpe of this to descrbe how to move exisitng vms to use this feature.\n\nthe only way to do that woudl be resize to a new flavor but that out of scope.\n\nwith that said i need ot compelter reviewing the other spec first before looping back to this one.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":20,"context_line":"`specification`__."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"However, the migration process is not covered there. This is addressed in"},{"line_number":23,"context_line":"the following section, which describes the Nova updates required to migrate"},{"line_number":24,"context_line":"SR-IOV devices to the new kernel variant driver interface."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":".. __: https://specs.openstack.org/openstack/nova-specs/specs/2025.1/approved/enable-vfio-devices-with-kernel-variant-drivers.html"},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Use Cases"}],"source_content_type":"text/x-rst","patch_set":3,"id":"d21813fa_02488bd9","line":25,"range":{"start_line":23,"start_character":29,"end_line":25,"end_character":0},"in_reply_to":"505ce0dc_153f721f","updated":"2025-01-08 09:16:35.000000000","message":"Yeah, René, please update that phrase : \"nova updates required for SRIOV devices using vfio-pci variant driver to be live-migrated to other hosts supporting the same devices\"","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":20,"context_line":"`specification`__."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"However, the migration process is not covered there. This is addressed in"},{"line_number":23,"context_line":"the following section, which describes the Nova updates required to migrate"},{"line_number":24,"context_line":"SR-IOV devices to the new kernel variant driver interface."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":".. __: https://specs.openstack.org/openstack/nova-specs/specs/2025.1/approved/enable-vfio-devices-with-kernel-variant-drivers.html"},{"line_number":27,"context_line":""},{"line_number":28,"context_line":"Use Cases"}],"source_content_type":"text/x-rst","patch_set":3,"id":"b0c5285b_9af27a85","line":25,"range":{"start_line":23,"start_character":29,"end_line":25,"end_character":0},"in_reply_to":"d21813fa_02488bd9","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":28,"context_line":"Use Cases"},{"line_number":29,"context_line":"---------"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."}],"source_content_type":"text/x-rst","patch_set":3,"id":"db581113_d3f86e44","line":31,"range":{"start_line":31,"start_character":28,"end_line":31,"end_character":35},"updated":"2025-01-07 13:23:47.000000000","message":"nit: live migrate","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":28,"context_line":"Use Cases"},{"line_number":29,"context_line":"---------"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."}],"source_content_type":"text/x-rst","patch_set":3,"id":"f51b17d0_921929da","line":31,"range":{"start_line":31,"start_character":28,"end_line":31,"end_character":35},"in_reply_to":"db581113_d3f86e44","updated":"2025-01-08 09:16:35.000000000","message":"+1","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":28,"context_line":"Use Cases"},{"line_number":29,"context_line":"---------"},{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."}],"source_content_type":"text/x-rst","patch_set":3,"id":"841a329b_6601f9ff","line":31,"range":{"start_line":31,"start_character":28,"end_line":31,"end_character":35},"in_reply_to":"f51b17d0_921929da","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"}],"source_content_type":"text/x-rst","patch_set":3,"id":"d6149508_39349377","line":33,"range":{"start_line":33,"start_character":55,"end_line":33,"end_character":65},"updated":"2025-01-07 13:23:47.000000000","message":"live migratable","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"}],"source_content_type":"text/x-rst","patch_set":3,"id":"10d46820_725e3efd","line":33,"range":{"start_line":33,"start_character":28,"end_line":33,"end_character":34},"updated":"2025-01-07 19:41:24.000000000","message":"nit: i would say declare rather then define\n\nif it was automatically discoverable i think we would just automate this and remove the need to declare this in the config.\n\nso its less about the operator saying you may migrate this device, rather you are saying i as an operator know its live migrateable.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"}],"source_content_type":"text/x-rst","patch_set":3,"id":"d0c26613_f0730eeb","line":33,"range":{"start_line":33,"start_character":28,"end_line":33,"end_character":34},"in_reply_to":"10d46820_725e3efd","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":30,"context_line":""},{"line_number":31,"context_line":"- As an operator, I want to migrate VMs with SR-IOV devices if such operation"},{"line_number":32,"context_line":"  is supported by the variant driver."},{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"}],"source_content_type":"text/x-rst","patch_set":3,"id":"888d5d3a_ba0c2b52","line":33,"range":{"start_line":33,"start_character":55,"end_line":33,"end_character":65},"in_reply_to":"d6149508_39349377","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"},{"line_number":37,"context_line":"  non-migratable devices."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"60a81f46_ef606174","line":36,"range":{"start_line":36,"start_character":52,"end_line":36,"end_character":62},"updated":"2025-01-07 13:23:47.000000000","message":"live migratable","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"},{"line_number":37,"context_line":"  non-migratable devices."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"73f80900_4443b99d","line":36,"range":{"start_line":36,"start_character":52,"end_line":36,"end_character":62},"in_reply_to":"01be22cf_d6abc405","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":33,"context_line":"- As an operator, I want to define whether a device is migratable or"},{"line_number":34,"context_line":"  non-migratable."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"- As an operator, I want to define flavors that use migratable or"},{"line_number":37,"context_line":"  non-migratable devices."},{"line_number":38,"context_line":""},{"line_number":39,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"01be22cf_d6abc405","line":36,"range":{"start_line":36,"start_character":52,"end_line":36,"end_character":62},"in_reply_to":"60a81f46_ef606174","updated":"2025-01-07 19:41:24.000000000","message":"the use case is valid but you shoudl not need to do anythign special in the falvor\n\nagain it should be hidden behind hte alias and depdnign on which alise you refence it will be migratable or not as a result.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":40,"context_line":"Proposed change"},{"line_number":41,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":42,"context_line":""},{"line_number":43,"context_line":"Configuring PCI device specification:"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."}],"source_content_type":"text/x-rst","patch_set":3,"id":"4ef1515f_3ef5fc36","line":43,"updated":"2025-01-08 09:16:35.000000000","message":"nit: please use a section style https://docutils.sourceforge.io/docs/ref/rst/restructuredtext.html#sections\n\n```\nConfiguring PCI device specification\n------------------------------------\n```","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":40,"context_line":"Proposed change"},{"line_number":41,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":42,"context_line":""},{"line_number":43,"context_line":"Configuring PCI device specification:"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."}],"source_content_type":"text/x-rst","patch_set":3,"id":"908f1254_cd5d0827","line":43,"in_reply_to":"4ef1515f_3ef5fc36","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":42,"context_line":""},{"line_number":43,"context_line":"Configuring PCI device specification:"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."},{"line_number":47,"context_line":""},{"line_number":48,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"}],"source_content_type":"text/x-rst","patch_set":3,"id":"d4776488_aeef76fb","line":45,"range":{"start_line":45,"start_character":62,"end_line":45,"end_character":71},"updated":"2025-01-07 13:23:47.000000000","message":"live migration","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":42,"context_line":""},{"line_number":43,"context_line":"Configuring PCI device specification:"},{"line_number":44,"context_line":""},{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."},{"line_number":47,"context_line":""},{"line_number":48,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"}],"source_content_type":"text/x-rst","patch_set":3,"id":"c757a006_d41b55f0","line":45,"range":{"start_line":45,"start_character":62,"end_line":45,"end_character":71},"in_reply_to":"d4776488_aeef76fb","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a3f4f50b989ad3c7ac2f66bc7c6b19457dc2aa88","unresolved":true,"context_lines":[{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."},{"line_number":47,"context_line":""},{"line_number":48,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":49,"context_line":"specification in [pci]dev_spec config."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be live migrated."}],"source_content_type":"text/x-rst","patch_set":3,"id":"3d420c36_782049b9","line":48,"range":{"start_line":48,"start_character":36,"end_line":48,"end_character":46},"updated":"2025-01-07 13:23:47.000000000","message":"I start to feel that this should be also called live_migratable","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."},{"line_number":47,"context_line":""},{"line_number":48,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":49,"context_line":"specification in [pci]dev_spec config."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be live migrated."}],"source_content_type":"text/x-rst","patch_set":3,"id":"fc118562_a73711b7","line":48,"range":{"start_line":48,"start_character":36,"end_line":48,"end_character":46},"in_reply_to":"3d420c36_782049b9","updated":"2025-01-08 09:16:35.000000000","message":"+1","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."},{"line_number":47,"context_line":""},{"line_number":48,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":49,"context_line":"specification in [pci]dev_spec config."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be live migrated."}],"source_content_type":"text/x-rst","patch_set":3,"id":"74b9d982_0fd2480c","line":48,"range":{"start_line":48,"start_character":36,"end_line":48,"end_character":46},"in_reply_to":"3d420c36_782049b9","updated":"2025-01-07 19:41:24.000000000","message":"+1\n\nits more explcit so i think that a good thing","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":45,"context_line":"Administrator must specify whether the device is eligible for migration to"},{"line_number":46,"context_line":"a similar device on another compute node."},{"line_number":47,"context_line":""},{"line_number":48,"context_line":"The proposed solution is to add a ``migratable`` tag to the device"},{"line_number":49,"context_line":"specification in [pci]dev_spec config."},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"- ``migratable\u003d\u0027yes\u0027`` means that the device can be live migrated."}],"source_content_type":"text/x-rst","patch_set":3,"id":"474ecd77_054591dd","line":48,"range":{"start_line":48,"start_character":36,"end_line":48,"end_character":46},"in_reply_to":"fc118562_a73711b7","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":74,"context_line":"  separate commits."},{"line_number":75,"context_line":""},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Configuring PCI aliases:"},{"line_number":78,"context_line":""},{"line_number":79,"context_line":"Users must specify whether the PCI request, and consequently the flavor,"},{"line_number":80,"context_line":"requires a migratable device."}],"source_content_type":"text/x-rst","patch_set":3,"id":"8dd95655_bd964e44","line":77,"updated":"2025-01-08 09:16:35.000000000","message":"ditto nit","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":74,"context_line":"  separate commits."},{"line_number":75,"context_line":""},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Configuring PCI aliases:"},{"line_number":78,"context_line":""},{"line_number":79,"context_line":"Users must specify whether the PCI request, and consequently the flavor,"},{"line_number":80,"context_line":"requires a migratable device."}],"source_content_type":"text/x-rst","patch_set":3,"id":"748fbefb_62a35f86","line":77,"in_reply_to":"8dd95655_bd964e44","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":109,"context_line":"migratable devices. Ensure it retrieves hosts with the appropriate number"},{"line_number":110,"context_line":"of migratable devices by adding a new filter."},{"line_number":111,"context_line":""},{"line_number":112,"context_line":"Since VIFs are not used in this context, we need to claim PCI devices and"},{"line_number":113,"context_line":"retrieve the PCI addresses of the destination host."},{"line_number":114,"context_line":""},{"line_number":115,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"}],"source_content_type":"text/x-rst","patch_set":3,"id":"272bbe63_8af94dcc","line":112,"range":{"start_line":112,"start_character":6,"end_line":112,"end_character":10},"updated":"2025-01-07 19:41:24.000000000","message":"you mean the VIF field in the migration data object yes.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":109,"context_line":"migratable devices. Ensure it retrieves hosts with the appropriate number"},{"line_number":110,"context_line":"of migratable devices by adding a new filter."},{"line_number":111,"context_line":""},{"line_number":112,"context_line":"Since VIFs are not used in this context, we need to claim PCI devices and"},{"line_number":113,"context_line":"retrieve the PCI addresses of the destination host."},{"line_number":114,"context_line":""},{"line_number":115,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"}],"source_content_type":"text/x-rst","patch_set":3,"id":"58da9061_864de717","line":112,"range":{"start_line":112,"start_character":6,"end_line":112,"end_character":10},"in_reply_to":"272bbe63_8af94dcc","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":121,"context_line":"collected in the LiveMigrateData object into the destination XML file."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":".. note::"},{"line_number":124,"context_line":"  - If PCI in Placement is enabled then live migration will work as today"},{"line_number":125,"context_line":"    for neutron requested PCI devices (i.e. legacy behavior works)"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"  - If PCI in Placement is enabled then SR-IOV live migration proposed in"},{"line_number":128,"context_line":"    this spec will still work (i.e. new functionality works)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"059905cf_313a5705","line":125,"range":{"start_line":124,"start_character":0,"end_line":125,"end_character":66},"updated":"2025-01-07 19:41:24.000000000","message":"so on that.\n\nwe really shoudl be enabling neutron live migration without hotplug of port with vnic_type\u003ddirect the VF is migratable.\n\nwe know that melanox connect-x 7 nic will also be supproted by vfio-variant dirvers so we shoudl try to make sure they \"just work\" as aprt of this btu im ok with that being a streach goal. i.e. put it at the end of the service btu i think we shoudl try and do that this cycle if we have time.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"bdbdd6f93aa54e975a7d093edfc9b09cab9808f2","unresolved":true,"context_lines":[{"line_number":121,"context_line":"collected in the LiveMigrateData object into the destination XML file."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":".. note::"},{"line_number":124,"context_line":"  - If PCI in Placement is enabled then live migration will work as today"},{"line_number":125,"context_line":"    for neutron requested PCI devices (i.e. legacy behavior works)"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"  - If PCI in Placement is enabled then SR-IOV live migration proposed in"},{"line_number":128,"context_line":"    this spec will still work (i.e. new functionality works)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"6f118f71_1a5eb05d","line":125,"range":{"start_line":124,"start_character":0,"end_line":125,"end_character":66},"in_reply_to":"059905cf_313a5705","updated":"2025-01-08 11:25:26.000000000","message":"We touched on this yesterday on a call with Rene and Sylvain. We ended up saying that it feels like an extra complexity on the implementation side and would like to phase it out to later. I.e. by the current proposal a PCI device in of the VM is either based on a NIC and live migrated via hotplug or not a NIC but migratable\u003dyes and live migrated via the variant driver support. But there are no PCI devices that both a neutron requested NIC and also migratable\u003dyes. So the existing hotplug codepath does not need to change. Just a new codepath needed to be added to migratable\u003dyes.\n\nI defer to Rene if he feels tackling the mixed case within this spec and within this cycle or if he rather do the separate case first and have the mixed case later.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7b210771ca2db7c619746a507e0770938b6b1749","unresolved":true,"context_lines":[{"line_number":121,"context_line":"collected in the LiveMigrateData object into the destination XML file."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":".. note::"},{"line_number":124,"context_line":"  - If PCI in Placement is enabled then live migration will work as today"},{"line_number":125,"context_line":"    for neutron requested PCI devices (i.e. legacy behavior works)"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"  - If PCI in Placement is enabled then SR-IOV live migration proposed in"},{"line_number":128,"context_line":"    this spec will still work (i.e. new functionality works)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"b2606c2b_cd0b4ee7","line":125,"range":{"start_line":124,"start_character":0,"end_line":125,"end_character":66},"in_reply_to":"1dd52cba_54b6b3b3","updated":"2025-01-08 17:57:56.000000000","message":"right but the chagne to enabel it are pretty minimal\nits litryally ensureign the ath souce pci device is migratble and if so requireing the same for the dest and then jsut not hotpluging.\n\nill alos not that the original plan was to enable live migration of vfio variant dirver using Melonox connectx-7 nics first before trying to enable it for any other device because the upstream kernel enablement is more mature and less complicated for the connectx-7 then for other devices.\n\n\nits perfiectly valid to set both  migratable\u003dyes and physical_network\u003d\"something\" in teh device_spec.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":true,"context_lines":[{"line_number":121,"context_line":"collected in the LiveMigrateData object into the destination XML file."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":".. note::"},{"line_number":124,"context_line":"  - If PCI in Placement is enabled then live migration will work as today"},{"line_number":125,"context_line":"    for neutron requested PCI devices (i.e. legacy behavior works)"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"  - If PCI in Placement is enabled then SR-IOV live migration proposed in"},{"line_number":128,"context_line":"    this spec will still work (i.e. new functionality works)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"1dd52cba_54b6b3b3","line":125,"range":{"start_line":124,"start_character":0,"end_line":125,"end_character":66},"in_reply_to":"6f118f71_1a5eb05d","updated":"2025-01-08 17:04:41.000000000","message":"I would like to keep it separated, as this cycle is shorter.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"7404a5476be0078fa4128b7bb0ed9461f579cd26","unresolved":true,"context_lines":[{"line_number":121,"context_line":"collected in the LiveMigrateData object into the destination XML file."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":".. note::"},{"line_number":124,"context_line":"  - If PCI in Placement is enabled then live migration will work as today"},{"line_number":125,"context_line":"    for neutron requested PCI devices (i.e. legacy behavior works)"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"  - If PCI in Placement is enabled then SR-IOV live migration proposed in"},{"line_number":128,"context_line":"    this spec will still work (i.e. new functionality works)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"e7f9d1af_1503cc3c","line":125,"range":{"start_line":124,"start_character":0,"end_line":125,"end_character":66},"in_reply_to":"b2606c2b_cd0b4ee7","updated":"2025-01-09 11:37:15.000000000","message":"OK is there a way to put that in the spec but skip it if we are lacking time ?","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"79171d0126997e1af993fb5e8af033241216ab5c","unresolved":true,"context_lines":[{"line_number":121,"context_line":"collected in the LiveMigrateData object into the destination XML file."},{"line_number":122,"context_line":""},{"line_number":123,"context_line":".. note::"},{"line_number":124,"context_line":"  - If PCI in Placement is enabled then live migration will work as today"},{"line_number":125,"context_line":"    for neutron requested PCI devices (i.e. legacy behavior works)"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"  - If PCI in Placement is enabled then SR-IOV live migration proposed in"},{"line_number":128,"context_line":"    this spec will still work (i.e. new functionality works)"}],"source_content_type":"text/x-rst","patch_set":3,"id":"3e709e7f_4fa8bc49","line":125,"range":{"start_line":124,"start_character":0,"end_line":125,"end_character":66},"in_reply_to":"e7f9d1af_1503cc3c","updated":"2025-01-09 12:46:40.000000000","message":"I can rephrase my statement. I rather get the simple case merged in E than trying \nto do the mixed case and missing E freeze with the whole thing. And yes, while logically physical_network and live-migratable\u003dyes should not be mutually exclusive at the end we can make them temporarily exclusive to have a phased approach. If time allows I\u0027m happy to review the mixed case implementation as well in E, but I don\u0027t want to jeopardize the base case availability in E.\n\n@rene.ribaud@gmail.com: Just write it explicitly in the spec that you will treat this as a different, independent implementation step that will not block merging the base case implementation.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":179,"context_line":"The user is fully responsible for configuring the following:"},{"line_number":180,"context_line":""},{"line_number":181,"context_line":"- Device specifications and aliases."},{"line_number":182,"context_line":"- Flavors: Major impact. If users need to support multiple types of"},{"line_number":183,"context_line":"  VFs, they must use different flavors for each VF type."},{"line_number":184,"context_line":""},{"line_number":185,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"f70f321d_92db73a5","line":182,"range":{"start_line":182,"start_character":59,"end_line":182,"end_character":64},"updated":"2025-01-08 09:16:35.000000000","message":"nit: kinds","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":179,"context_line":"The user is fully responsible for configuring the following:"},{"line_number":180,"context_line":""},{"line_number":181,"context_line":"- Device specifications and aliases."},{"line_number":182,"context_line":"- Flavors: Major impact. If users need to support multiple types of"},{"line_number":183,"context_line":"  VFs, they must use different flavors for each VF type."},{"line_number":184,"context_line":""},{"line_number":185,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"65097423_6d6b6791","line":182,"range":{"start_line":182,"start_character":59,"end_line":182,"end_character":64},"in_reply_to":"f70f321d_92db73a5","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"353a7d272cd14c5f4335990403d10bac667237a5","unresolved":true,"context_lines":[{"line_number":186,"context_line":"Performance Impact"},{"line_number":187,"context_line":"------------------"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"If PCI in placement is enable, this `bug`__ should be taken into account"},{"line_number":190,"context_line":"as it may impact performance."},{"line_number":191,"context_line":""},{"line_number":192,"context_line":"`Mitigation measures`__ are currently being developed to minimize this impact."}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa036ad5_d24b292a","line":189,"range":{"start_line":189,"start_character":23,"end_line":189,"end_character":29},"updated":"2025-01-08 09:16:35.000000000","message":"enabled","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":186,"context_line":"Performance Impact"},{"line_number":187,"context_line":"------------------"},{"line_number":188,"context_line":""},{"line_number":189,"context_line":"If PCI in placement is enable, this `bug`__ should be taken into account"},{"line_number":190,"context_line":"as it may impact performance."},{"line_number":191,"context_line":""},{"line_number":192,"context_line":"`Mitigation measures`__ are currently being developed to minimize this impact."}],"source_content_type":"text/x-rst","patch_set":3,"id":"de7f29c4_d7777158","line":189,"range":{"start_line":189,"start_character":23,"end_line":189,"end_character":29},"in_reply_to":"fa036ad5_d24b292a","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":208,"context_line":"Upgrade impact"},{"line_number":209,"context_line":"--------------"},{"line_number":210,"context_line":""},{"line_number":211,"context_line":"Users should review the following:"},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"- The configuration of device specifications and aliases."},{"line_number":214,"context_line":"- Update their flavors to specify the correct aliases."}],"source_content_type":"text/x-rst","patch_set":3,"id":"497926e4_770f1545","line":211,"range":{"start_line":211,"start_character":0,"end_line":211,"end_character":5},"updated":"2025-01-07 19:41:24.000000000","message":"operators","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":208,"context_line":"Upgrade impact"},{"line_number":209,"context_line":"--------------"},{"line_number":210,"context_line":""},{"line_number":211,"context_line":"Users should review the following:"},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"- The configuration of device specifications and aliases."},{"line_number":214,"context_line":"- Update their flavors to specify the correct aliases."}],"source_content_type":"text/x-rst","patch_set":3,"id":"ab498156_ca450b57","line":211,"range":{"start_line":211,"start_character":0,"end_line":211,"end_character":5},"in_reply_to":"497926e4_770f1545","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"84df61f025b4a10b286a0edee919924d317a42e5","unresolved":true,"context_lines":[{"line_number":211,"context_line":"Users should review the following:"},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"- The configuration of device specifications and aliases."},{"line_number":214,"context_line":"- Update their flavors to specify the correct aliases."},{"line_number":215,"context_line":""},{"line_number":216,"context_line":"In Nova\u0027s data model, this is only achievable through a resize operation,"},{"line_number":217,"context_line":"which requires stopping the instance."}],"source_content_type":"text/x-rst","patch_set":3,"id":"a45085a8_c77ca197","line":214,"range":{"start_line":214,"start_character":0,"end_line":214,"end_character":2},"updated":"2025-01-07 19:41:24.000000000","message":"you shoudl never update flavor extra spec once the flavor is in use.\nso we shoudl not suggest they do that here.\n\nthey should define new flavor for the new alsias.\nor update the existing alisas/device_specs in place, if an only if all devices refence by the alias are actully live_migarable already.\n\nthis really only matters if the live migration part is not in the same release as the generic supprot for vfio-varint dirvers however as 99% of operators wont have any existing device that supprot this.\n\na small subset may have connectx-7 nic but for the most part they wont have exsting workload using VF that could supprot live migration before this feature is released.","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"4fe6410518b4138186919f5e07b84676f9cdcd21","unresolved":false,"context_lines":[{"line_number":211,"context_line":"Users should review the following:"},{"line_number":212,"context_line":""},{"line_number":213,"context_line":"- The configuration of device specifications and aliases."},{"line_number":214,"context_line":"- Update their flavors to specify the correct aliases."},{"line_number":215,"context_line":""},{"line_number":216,"context_line":"In Nova\u0027s data model, this is only achievable through a resize operation,"},{"line_number":217,"context_line":"which requires stopping the instance."}],"source_content_type":"text/x-rst","patch_set":3,"id":"ccccafe4_c7d69547","line":214,"range":{"start_line":214,"start_character":0,"end_line":214,"end_character":2},"in_reply_to":"a45085a8_c77ca197","updated":"2025-01-08 17:04:41.000000000","message":"Done","commit_id":"6b76defb2c44c71728f336f29f11ffc269cdf5fd"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7b210771ca2db7c619746a507e0770938b6b1749","unresolved":true,"context_lines":[{"line_number":54,"context_line":""},{"line_number":55,"context_line":"- ``live_migratable\u003d\u0027yes\u0027`` means that the device can be live migrated."},{"line_number":56,"context_line":"- ``live_migratable\u003d\u0027no\u0027`` means that the device cannot be live migrated."},{"line_number":57,"context_line":"- If not specified then the default is ``live_migratable\u003d\u0027no\u0027``"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":""},{"line_number":60,"context_line":"When this tag is encountered by the `PCI resource tracker`__, the"}],"source_content_type":"text/x-rst","patch_set":4,"id":"bf561952_7e996d25","line":57,"updated":"2025-01-08 17:57:56.000000000","message":"again inm not really happy with this default.\n\nif we start with this we will have to remvoe it in the fure so i would prefer not do default to no for now.\n\ni.e. i dont think we shoudl filter out migrable devices if you dont request a migratble device.\n\nthat i think weill intoduce upgrade impacts when we go to enable this for neutron ports.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"5798123ee3ab584376502554e947222175994f0e","unresolved":true,"context_lines":[{"line_number":54,"context_line":""},{"line_number":55,"context_line":"- ``live_migratable\u003d\u0027yes\u0027`` means that the device can be live migrated."},{"line_number":56,"context_line":"- ``live_migratable\u003d\u0027no\u0027`` means that the device cannot be live migrated."},{"line_number":57,"context_line":"- If not specified then the default is ``live_migratable\u003d\u0027no\u0027``"},{"line_number":58,"context_line":""},{"line_number":59,"context_line":""},{"line_number":60,"context_line":"When this tag is encountered by the `PCI resource tracker`__, the"}],"source_content_type":"text/x-rst","patch_set":4,"id":"f020c34c_2dfa9da1","line":57,"in_reply_to":"bf561952_7e996d25","updated":"2025-01-09 13:29:57.000000000","message":"see my comment above","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"7b210771ca2db7c619746a507e0770938b6b1749","unresolved":true,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"},{"line_number":122,"context_line":"addresses, live migratable, managed) for the destination host by"},{"line_number":123,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Update the _live_migration_operation() function, with a specific focus"},{"line_number":126,"context_line":"on the get_updated_guest_xml() function, to incorporate the PCI addresses"}],"source_content_type":"text/x-rst","patch_set":4,"id":"770e3b30_78190e54","line":123,"updated":"2025-01-08 17:57:56.000000000","message":"this is proably a bit to hevey wight.\n\nwe could pass the full pci device object\n\nhttps://github.com/openstack/nova/blob/master/nova/objects/pci_device.py#L52\n\nbut what we actually need is a map of source pci_adress to dest pci adress\n \na list is not really sufficient because we can have many flavour based pci requests so we need to have a 1:1 mapping\n\ni dont think relying on the order of devices in the list is a good thing as that is technially not required to be preserved wehn you serialie into json.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"7404a5476be0078fa4128b7bb0ed9461f579cd26","unresolved":true,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"},{"line_number":122,"context_line":"addresses, live migratable, managed) for the destination host by"},{"line_number":123,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Update the _live_migration_operation() function, with a specific focus"},{"line_number":126,"context_line":"on the get_updated_guest_xml() function, to incorporate the PCI addresses"}],"source_content_type":"text/x-rst","patch_set":4,"id":"7c54dd44_6136b2ec","line":123,"in_reply_to":"770e3b30_78190e54","updated":"2025-01-09 11:37:15.000000000","message":"I understand it is advice about the implementation, not something that must be in the spec.\n\nI usually like to use objects instead of dict because I feel that more straightforward to understand the content. Anyway I will see because yes maybe the object here is a bit overshoot.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"79171d0126997e1af993fb5e8af033241216ab5c","unresolved":true,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"},{"line_number":122,"context_line":"addresses, live migratable, managed) for the destination host by"},{"line_number":123,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Update the _live_migration_operation() function, with a specific focus"},{"line_number":126,"context_line":"on the get_updated_guest_xml() function, to incorporate the PCI addresses"}],"source_content_type":"text/x-rst","patch_set":4,"id":"d164e6b9_26e75603","line":123,"in_reply_to":"7c54dd44_6136b2ec","updated":"2025-01-09 12:46:40.000000000","message":"I\u0027m fine to discuss the details of the model change in the implementation patch. I think we all agree that extra information is needed in the LiveMigrateData object.\n\nPersonally I would need to look at the code that uses the data to express the my preferences.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"96af04838fbc8033e5647fceb2dcbbcde551a5f5","unresolved":true,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"},{"line_number":122,"context_line":"addresses, live migratable, managed) for the destination host by"},{"line_number":123,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Update the _live_migration_operation() function, with a specific focus"},{"line_number":126,"context_line":"on the get_updated_guest_xml() function, to incorporate the PCI addresses"}],"source_content_type":"text/x-rst","patch_set":4,"id":"36d510c4_22efd679","line":123,"in_reply_to":"c3d876d4_c4568072","updated":"2025-01-09 18:01:23.000000000","message":"im ok to defer this to the impleation patch if an only if we come back and update the spec with the details of the object change before we merge the imepleation patch or when we merge it.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"5798123ee3ab584376502554e947222175994f0e","unresolved":true,"context_lines":[{"line_number":120,"context_line":""},{"line_number":121,"context_line":"Update the ``LiveMigrateData`` object to include PCI device information (e.g.,"},{"line_number":122,"context_line":"addresses, live migratable, managed) for the destination host by"},{"line_number":123,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"Update the _live_migration_operation() function, with a specific focus"},{"line_number":126,"context_line":"on the get_updated_guest_xml() function, to incorporate the PCI addresses"}],"source_content_type":"text/x-rst","patch_set":4,"id":"c3d876d4_c4568072","line":123,"in_reply_to":"d164e6b9_26e75603","updated":"2025-01-09 13:29:57.000000000","message":"Agreed, that\u0027s also what we did for the vgpu mdev live-migration spec, we needed to modify the object when implementing the series.","commit_id":"48c3bb127aa3a8729ef28479b22fd058002485a7"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"68281c83233b680aaaa193efe66418cf4345f00d","unresolved":true,"context_lines":[{"line_number":61,"context_line":""},{"line_number":62,"context_line":"- If not specified, the default behavior will be equivalent to"},{"line_number":63,"context_line":"  live_migratable\u003d\u0027no\u0027. However, this value will not be persisted in the"},{"line_number":64,"context_line":"  PciDevice object."},{"line_number":65,"context_line":""},{"line_number":66,"context_line":".. note::"},{"line_number":67,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"3a12799f_f22df49f","line":64,"updated":"2025-01-09 17:44:09.000000000","message":"++","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":true,"context_lines":[{"line_number":61,"context_line":""},{"line_number":62,"context_line":"- If not specified, the default behavior will be equivalent to"},{"line_number":63,"context_line":"  live_migratable\u003d\u0027no\u0027. However, this value will not be persisted in the"},{"line_number":64,"context_line":"  PciDevice object."},{"line_number":65,"context_line":""},{"line_number":66,"context_line":".. note::"},{"line_number":67,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"7992b294_8fcd23d5","line":64,"in_reply_to":"3a12799f_f22df49f","updated":"2025-01-09 18:52:32.000000000","message":"+1","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"68281c83233b680aaaa193efe66418cf4345f00d","unresolved":true,"context_lines":[{"line_number":67,"context_line":""},{"line_number":68,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":69,"context_line":""},{"line_number":70,"context_line":"Additionally, if pci in placement is enabled and ``live_migrateable\u003d\u0027yes\u0027``,"},{"line_number":71,"context_line":"it will record a new standard trait, HW_PCI_LIVE_MIGRATABLE, in the resource"},{"line_number":72,"context_line":"provider representing the physical device. While this trait will not be"},{"line_number":73,"context_line":"utilized by the migration flow, it can serve as a reference for inventory"}],"source_content_type":"text/x-rst","patch_set":6,"id":"40cde852_4174301d","line":70,"range":{"start_line":70,"start_character":51,"end_line":70,"end_character":67},"updated":"2025-01-09 17:44:09.000000000","message":"nit (can be fixed in a FUP) : live-migratable","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":16207,"name":"ribaudr","display_name":"uggla","email":"rene.ribaud@gmail.com","username":"uggla","status":"Red Hat"},"change_message_id":"b676d342ee0c557b2b5a0d0ec6089bceed6a1c94","unresolved":false,"context_lines":[{"line_number":67,"context_line":""},{"line_number":68,"context_line":"  The PciDevice object version remains unchanged."},{"line_number":69,"context_line":""},{"line_number":70,"context_line":"Additionally, if pci in placement is enabled and ``live_migrateable\u003d\u0027yes\u0027``,"},{"line_number":71,"context_line":"it will record a new standard trait, HW_PCI_LIVE_MIGRATABLE, in the resource"},{"line_number":72,"context_line":"provider representing the physical device. While this trait will not be"},{"line_number":73,"context_line":"utilized by the migration flow, it can serve as a reference for inventory"}],"source_content_type":"text/x-rst","patch_set":6,"id":"a87eecb9_0dbb178e","line":70,"range":{"start_line":70,"start_character":51,"end_line":70,"end_character":67},"in_reply_to":"40cde852_4174301d","updated":"2025-01-15 18:10:41.000000000","message":"Done","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"68281c83233b680aaaa193efe66418cf4345f00d","unresolved":true,"context_lines":[{"line_number":129,"context_line":".. note::"},{"line_number":130,"context_line":"  The data structure required to store this information will be defined as"},{"line_number":131,"context_line":"  the implementation progresses. This specification will then be amended"},{"line_number":132,"context_line":"  accordingly."},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"for the destination host by"},{"line_number":135,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."}],"source_content_type":"text/x-rst","patch_set":6,"id":"7b0761d0_8f008db2","line":132,"updated":"2025-01-09 17:44:09.000000000","message":"++","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":false,"context_lines":[{"line_number":129,"context_line":".. note::"},{"line_number":130,"context_line":"  The data structure required to store this information will be defined as"},{"line_number":131,"context_line":"  the implementation progresses. This specification will then be amended"},{"line_number":132,"context_line":"  accordingly."},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"for the destination host by"},{"line_number":135,"context_line":"introducing a new ``pci_devices`` field as a ListOfObjectsField(\"PciDevice\")."}],"source_content_type":"text/x-rst","patch_set":6,"id":"4e28aaa9_7d4f6696","line":132,"in_reply_to":"7b0761d0_8f008db2","updated":"2025-01-09 18:52:32.000000000","message":"ack that works for me.","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"68281c83233b680aaaa193efe66418cf4345f00d","unresolved":true,"context_lines":[{"line_number":160,"context_line":"However, this change will:"},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"- Be implemented in a separate patch to allow the base case to land first."},{"line_number":163,"context_line":"- Ensure that such NICs are properly live migrated using the new code path."},{"line_number":164,"context_line":""},{"line_number":165,"context_line":".. __: https://github.com/openstack/nova/blob/f98f414f971b6c897bf48781a579730419b5a93d/nova/compute/pci_placement_translator.py#L597-L600"},{"line_number":166,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"29abbf60_e7054908","line":163,"updated":"2025-01-09 17:44:09.000000000","message":"++","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":false,"context_lines":[{"line_number":160,"context_line":"However, this change will:"},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"- Be implemented in a separate patch to allow the base case to land first."},{"line_number":163,"context_line":"- Ensure that such NICs are properly live migrated using the new code path."},{"line_number":164,"context_line":""},{"line_number":165,"context_line":".. __: https://github.com/openstack/nova/blob/f98f414f971b6c897bf48781a579730419b5a93d/nova/compute/pci_placement_translator.py#L597-L600"},{"line_number":166,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"c34a754d_f242c596","line":163,"in_reply_to":"29abbf60_e7054908","updated":"2025-01-09 18:52:32.000000000","message":"Acknowledged","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"68281c83233b680aaaa193efe66418cf4345f00d","unresolved":true,"context_lines":[{"line_number":241,"context_line":""},{"line_number":242,"context_line":"This can be achieved by resizing the VM and changing its flavor to the new one."},{"line_number":243,"context_line":""},{"line_number":244,"context_line":"For NICs, an alternative approach could be to detach and reattach the device."},{"line_number":245,"context_line":""},{"line_number":246,"context_line":""},{"line_number":247,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":6,"id":"fc718bea_6ec7a3ea","line":244,"updated":"2025-01-09 17:44:09.000000000","message":"++","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":true,"context_lines":[{"line_number":241,"context_line":""},{"line_number":242,"context_line":"This can be achieved by resizing the VM and changing its flavor to the new one."},{"line_number":243,"context_line":""},{"line_number":244,"context_line":"For NICs, an alternative approach could be to detach and reattach the device."},{"line_number":245,"context_line":""},{"line_number":246,"context_line":""},{"line_number":247,"context_line":"Implementation"}],"source_content_type":"text/x-rst","patch_set":6,"id":"97f8cdc9_94fa3885","line":244,"in_reply_to":"fc718bea_6ec7a3ea","updated":"2025-01-09 18:52:32.000000000","message":"+1","commit_id":"ad4ab77cc43c367c5ef1f72940c64b10780f81ef"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":true,"context_lines":[{"line_number":77,"context_line":".. note::"},{"line_number":78,"context_line":""},{"line_number":79,"context_line":"  Since this is not mandatory for the migration, it will be included in"},{"line_number":80,"context_line":"  separate commits."},{"line_number":81,"context_line":""},{"line_number":82,"context_line":""},{"line_number":83,"context_line":"Configuring PCI aliases:"}],"source_content_type":"text/x-rst","patch_set":7,"id":"4089498e_78e440d3","line":80,"updated":"2025-01-09 18:52:32.000000000","message":"while it true the pci passthough filter will account for this\n\nwe should modify from_request_spec\n\nhttps://github.com/openstack/nova/blob/634be5191e0fde60aac774fb7917868de9a2b29c/nova/scheduler/utils.py#L80\n\nby adding a function like _translate_pci_numa_affinity_policy\nto add the appropriate traits to the request groups when pci in placement is used but yes that can be a follow-up performance enhancement.","commit_id":"9b433f0025b2fd4f81e87f2e595a03808accba0b"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"79586bfef17bf8d135a43ed340862f6a0e0be01c","unresolved":true,"context_lines":[{"line_number":146,"context_line":"    this spec will still work (i.e. new functionality works)"},{"line_number":147,"context_line":""},{"line_number":148,"context_line":"  - Optionally PCI in Placement will be extended to automatically request"},{"line_number":149,"context_line":"    HW_PCI_LIVE_MIGRATABLE trait if the alias has live_migratable\u003d\"yes\"."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"    - A further enhancement would be to extend the translation of the"},{"line_number":152,"context_line":"      [pci]alias spec to placement RequestGroups to support forbidden traits."}],"source_content_type":"text/x-rst","patch_set":7,"id":"d2c47dc3_662e9f60","line":149,"updated":"2025-01-09 18:52:32.000000000","message":"we can probably discuss this in the implementation but one open i have is where will we reprot the trait.\nwe have two options, for pci in placment it shoudl be repoted on the pci device resouce provider and the trait request shoudl be part of the request group created from the alisa or neutron port when we get to supproting that.\n\n\nif we are not reporting pci device isn placemnet \nwe have 2 options 1 just dont report the trait at all or two report it on the root resource provider.\n\n\nI\u0027m inclined to say that we should only report HW_PCI_LIVE_MIGRATABLE in the PCI RP which means it would not be reported at all if not using pci in placement.","commit_id":"9b433f0025b2fd4f81e87f2e595a03808accba0b"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"a6a07cc87082af6825f738fb74f603c7356a4db4","unresolved":true,"context_lines":[{"line_number":146,"context_line":"    this spec will still work (i.e. new functionality works)"},{"line_number":147,"context_line":""},{"line_number":148,"context_line":"  - Optionally PCI in Placement will be extended to automatically request"},{"line_number":149,"context_line":"    HW_PCI_LIVE_MIGRATABLE trait if the alias has live_migratable\u003d\"yes\"."},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"    - A further enhancement would be to extend the translation of the"},{"line_number":152,"context_line":"      [pci]alias spec to placement RequestGroups to support forbidden traits."}],"source_content_type":"text/x-rst","patch_set":7,"id":"5872df0c_bbf76883","line":149,"in_reply_to":"d2c47dc3_662e9f60","updated":"2025-01-10 08:08:30.000000000","message":"Yeah I agree with Sean here.","commit_id":"9b433f0025b2fd4f81e87f2e595a03808accba0b"}]}