)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"bb7dc394f923c805c803ccc2ce7a5c8edfb7e5bd","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"c8e47cb2_c9ff4ffb","updated":"2023-09-20 12:30:17.000000000","message":"you shoudl not have moved this it should have been copied","commit_id":"a26616590c18c1c2c9b2063ccd509f90bb4efe90"},{"author":{"_account_id":35489,"name":"Danylo Vodopianov","email":"dvo-plv@napatech.com","username":"dvo-plv"},"change_message_id":"f2f88851145a3cfde1e48f4c9405fdec61f20d22","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"1f2869bf_ea4d5b6c","in_reply_to":"c8e47cb2_c9ff4ffb","updated":"2023-09-20 12:35:58.000000000","message":"Fixed","commit_id":"a26616590c18c1c2c9b2063ccd509f90bb4efe90"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"a136e6867098907a3198b42cc3373276dc3ba87d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"47ee0254_b69da4c3","updated":"2023-09-20 12:44:19.000000000","message":"im happy with this reproposal to continue in the 2024.1 cycle","commit_id":"ad467a23913c135fb84f179766613aa404462930"},{"author":{"_account_id":6962,"name":"Kashyap Chamarthy","email":"kchamart@redhat.com","username":"kashyapc"},"change_message_id":"15273b0ac6d40b32727bd0cdb81d4e1adc68b743","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"3b29accc_15323ece","updated":"2023-10-26 10:05:08.000000000","message":"FWIW, the core idea looks good to me: to gain better performance from using the more efficient virtq structure.\n\nTo quote the documentation from libvirt about its benefit:\n\n\n\"The attribute packed controls if QEMU should try to use packed virtqueues. Compared to regular split queues, packed queues consist of only a single descriptor ring replacing available and used ring, index and descriptor buffer. This can result in better cache utilization and performance. [...]\"","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"60cd35efb69c33443b0a6c4686feb76bff759996","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"ad0c6c71_e6701fe8","updated":"2023-11-07 10:32:20.000000000","message":"While I understand the usecase and the solution, I still have some comments for this spec :\n- IMO, you should tell in the dependencies that we need to use a specific os-vif release\n- I think you should tell operators that if they upgrade their computes by rolling, only the upgraded computes would be used.","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"950cf3413e632c03fcaaaa27709450a20b869049","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"11e96632_7ce5285f","updated":"2023-11-07 14:39:46.000000000","message":"Now that I better understand the required changes, I debate the need for adding a specific trait that would only have a very short life.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"e841744a96c93fd0714ee89155231649061a826e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"68f669c8_06b5a58e","updated":"2023-11-07 14:19:38.000000000","message":"to be clear, the nova spec seems quite simple to me, but we still need to explain the dependencies and how we could test that. Can we run something in the gate ?","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"61c6abc632464782f4d255f9f6327eb04aff7116","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"f6266170_2cfb28a2","in_reply_to":"11e96632_7ce5285f","updated":"2023-11-07 15:24:14.000000000","message":"the trait would not be short lived.\nit woudl be tere as long as we supprot more then just ironic and libvirt as hypervior dirvers\n\nalthough that might be short lived since we are removing vmware and hyperv this cycle.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"83efac1718d0af9e9629dc057ebedea0b72deb96","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":8,"id":"f8975a95_6c8b3db6","updated":"2023-11-23 12:16:46.000000000","message":"i would prefer to move forward with the spec as is and focus on the impleation review and amend if we have any concerns based on that.\n\ni have not looked at the implmeation for a while but i tought this was pretty close to being merged last release so im not expecting there to be large changes required at this point","commit_id":"10c06719960f31c2224003a040f3e69bcc849d87"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"0b323fec37d3719d0f749e461722c55223d68b14","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":8,"id":"230abbe0_027946d6","updated":"2023-11-23 12:15:04.000000000","message":"im still ok with this spec over all so re-adding +2","commit_id":"10c06719960f31c2224003a040f3e69bcc849d87"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"b49fb3375f9ab4fb901f2b36a84cc13d515cd375","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":8,"id":"32ce1997_ddde154c","updated":"2023-11-30 09:35:40.000000000","message":"looks good","commit_id":"10c06719960f31c2224003a040f3e69bcc849d87"}],"specs/2024.1/approved/virtio_packedring_configuration_support.rst":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"30cb0b179bcb8e17d25d0f4dac18ae46d59418bd","unresolved":true,"context_lines":[{"line_number":182,"context_line":""},{"line_number":183,"context_line":"   * - Release Name"},{"line_number":184,"context_line":"     - Description"},{"line_number":185,"context_line":"   * - 2023.1 Bobcat"},{"line_number":186,"context_line":"     - Introduced"},{"line_number":187,"context_line":"   * - 2023.2 Bobcat"},{"line_number":188,"context_line":"     - Accepted"}],"source_content_type":"text/x-rst","patch_set":4,"id":"3aa593e4_aa2339d6","line":185,"updated":"2023-09-20 12:44:58.000000000","message":"actully this should be antelope","commit_id":"ad467a23913c135fb84f179766613aa404462930"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"60cd35efb69c33443b0a6c4686feb76bff759996","unresolved":true,"context_lines":[{"line_number":121,"context_line":""},{"line_number":122,"context_line":"Upgrade impact"},{"line_number":123,"context_line":"--------------"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"* This spec will update scheduling process. New trait"},{"line_number":126,"context_line":"  ``COMPUTE_NET_VIRTIO_PACKED`` will be set to the resource provider trait list"},{"line_number":127,"context_line":"  automatically if this feaure is supported on the host."}],"source_content_type":"text/x-rst","patch_set":5,"id":"45179989_50f10441","line":124,"updated":"2023-11-07 10:32:20.000000000","message":"there should be none, since we automatically schedule to computes that support the trait.\n\nThat said, that means for rolling upgrades that only new upgraded computes would be used, so it could be a performance question.","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"60cd35efb69c33443b0a6c4686feb76bff759996","unresolved":true,"context_lines":[{"line_number":138,"context_line":"  justas_napa on IRC and Gerrit"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The feature can be implemented by the Napatech devs dvo-plv@napatech.com and"},{"line_number":141,"context_line":"obu-plv@napatech.com."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Feature Liaison"},{"line_number":144,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":5,"id":"ed2463b6_c1d788cd","line":141,"updated":"2023-11-07 10:32:20.000000000","message":"nit: you can just add those people into additional assignees","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"60cd35efb69c33443b0a6c4686feb76bff759996","unresolved":true,"context_lines":[{"line_number":148,"context_line":"Work Items"},{"line_number":149,"context_line":"----------"},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"N/A at this stage."},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Dependencies"},{"line_number":154,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"4969b54f_6ebe1bf0","line":151,"updated":"2023-11-07 10:32:20.000000000","message":"really ? why aren\u0027t you just adding bullet points from what you said in the proposed change ?","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"60cd35efb69c33443b0a6c4686feb76bff759996","unresolved":true,"context_lines":[{"line_number":153,"context_line":"Dependencies"},{"line_number":154,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"None"},{"line_number":157,"context_line":""},{"line_number":158,"context_line":"Testing"},{"line_number":159,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"4e913403_e17478db","line":156,"updated":"2023-11-07 10:32:20.000000000","message":"technically, this depends on some Neutron and os-vif support that was merged before, right?","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"950cf3413e632c03fcaaaa27709450a20b869049","unresolved":false,"context_lines":[{"line_number":153,"context_line":"Dependencies"},{"line_number":154,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"None"},{"line_number":157,"context_line":""},{"line_number":158,"context_line":"Testing"},{"line_number":159,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"aee9e45f_b1566b38","line":156,"in_reply_to":"4e913403_e17478db","updated":"2023-11-07 14:39:46.000000000","message":"\u003e technically, this depends on some Neutron and os-vif support that was merged before, right?\n\nI\u0027m stupid, I mixed two different things. To be clear, this sole spec is only about letting nova-compute to decorate with a new trait if it supports that new QEMU capability.","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"60cd35efb69c33443b0a6c4686feb76bff759996","unresolved":true,"context_lines":[{"line_number":187,"context_line":"   * - 2023.2 Bobcat"},{"line_number":188,"context_line":"     - Accepted"},{"line_number":189,"context_line":"   * - 2024.1 Caracal"},{"line_number":190,"context_line":"     - Reproposed"}],"source_content_type":"text/x-rst","patch_set":5,"id":"86da538e_addc90ac","line":190,"updated":"2023-11-07 10:32:20.000000000","message":"FWIW, this isn\u0027t exactly a plain reproposal, I can see a lot of differences :\n\nhttps://paste.opendev.org/show/bey4AkZ4w7uhps1NE7Cj/","commit_id":"60491067ffd8c65eab113b9f5e66ce360dfcea55"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"950cf3413e632c03fcaaaa27709450a20b869049","unresolved":true,"context_lines":[{"line_number":62,"context_line":"* Provide new compute ``COMPUTE_NET_VIRTIO_PACKED`` capablity trait."},{"line_number":63,"context_line":"  This trait can be required/forbidden by user. Nova-compute agent"},{"line_number":64,"context_line":"  will automatically set this trait to the resource provider summary"},{"line_number":65,"context_line":"  if libvirt version is higher than 6.3"},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"* This spec will update scheduling process. ALL_REQUEST_FILTERS will be"},{"line_number":68,"context_line":"  extended with new filter packed_virtqueue_filter. It will update RequestSpec"}],"source_content_type":"text/x-rst","patch_set":6,"id":"32f82052_813de18c","line":65,"updated":"2023-11-07 14:39:46.000000000","message":"Could we try to avoid to add yet another trait for something like just \"this is my libvirt version\" ?\n\nGiven https://docs.openstack.org/nova/latest/reference/libvirt-distro-support-matrix.html#min-libvirt-qemu-version-and-next-min-libvirt-qemu-version-table Bobcat computes are already above that libvirt version so the libvirt detection would be only for Antelope computes (we don\u0027t support more than N-2 rolling upgrades given Caracal is a SLURP release). Again, is this really worth it, and couldn\u0027t we just add a prefilter that would check the compute versions ?\n\nSince removing a prefilter is cheap, we could just remove that once we\u0027re on D.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"acfbc6d2eb9ea1fc1ed4bc10a16497389cee1a2f","unresolved":true,"context_lines":[{"line_number":62,"context_line":"* Provide new compute ``COMPUTE_NET_VIRTIO_PACKED`` capablity trait."},{"line_number":63,"context_line":"  This trait can be required/forbidden by user. Nova-compute agent"},{"line_number":64,"context_line":"  will automatically set this trait to the resource provider summary"},{"line_number":65,"context_line":"  if libvirt version is higher than 6.3"},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"* This spec will update scheduling process. ALL_REQUEST_FILTERS will be"},{"line_number":68,"context_line":"  extended with new filter packed_virtqueue_filter. It will update RequestSpec"}],"source_content_type":"text/x-rst","patch_set":6,"id":"5d1c9dea_bf9b7350","line":65,"in_reply_to":"32f82052_813de18c","updated":"2023-11-07 15:13:50.000000000","message":"the prefilter is interesting idea\nwe woudl have to make the prefilter mandatory if we were to rely on it.\nthis feels kind of like an abuse of prefilters however.\n\nit would be more correct to do the min compute service check in the api\n\na prefilter would be correct if its enfrocing the presence of the trait but i dont think prefilter shoudl do min comptue service checks.\n\n\nremoving a libvirt version check is also cheap for what its worth but to your point about is this need. no our min version on master is now above the required version so we can remove all libvirt version check related to this form the nova drive rchange.\n\n\nto summerisie:\n \nwe can simply document that this feature requires libvirt 6.3+ and is only supported in a fully upgraded cloud.\n\n \nwe should not requrie a compute service bump for this feature as its a virt driver speciffic capablity not a compute manager capablity.\n\nwe shoudl ahve a prefilter to transform the extra spec/image property into a required trait and that shoudl be always enabled.\n\nwe do not need a min libvirt check in the code since our min libvirt is higher then\nthe version where this was intoduced on master. SLURP shoudl not affect that because the prefilter will enfoce that vms can only go to the host that support this.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"e12190b33f79173bf825fccf7d3f14e5c28436aa","unresolved":true,"context_lines":[{"line_number":62,"context_line":"* Provide new compute ``COMPUTE_NET_VIRTIO_PACKED`` capablity trait."},{"line_number":63,"context_line":"  This trait can be required/forbidden by user. Nova-compute agent"},{"line_number":64,"context_line":"  will automatically set this trait to the resource provider summary"},{"line_number":65,"context_line":"  if libvirt version is higher than 6.3"},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"* This spec will update scheduling process. ALL_REQUEST_FILTERS will be"},{"line_number":68,"context_line":"  extended with new filter packed_virtqueue_filter. It will update RequestSpec"}],"source_content_type":"text/x-rst","patch_set":6,"id":"7c37fe25_0c523958","line":65,"in_reply_to":"5d1c9dea_bf9b7350","updated":"2023-11-07 17:43:11.000000000","message":"\u003e the prefilter is interesting idea\n\u003e we woudl have to make the prefilter mandatory if we were to rely on it.\n\u003e this feels kind of like an abuse of prefilters however.\n\u003e \n\u003e it would be more correct to do the min compute service check in the api\n\u003e \n\nThat\u0027s the usual pattern we do with service version checks, I agree.\n\n\u003e a prefilter would be correct if its enfrocing the presence of the trait but i dont think prefilter shoudl do min comptue service checks.\n\u003e \n\nCool with that.\n\u003e \n\u003e removing a libvirt version check is also cheap for what its worth but to your point about is this need. no our min version on master is now above the required version so we can remove all libvirt version check related to this form the nova drive rchange.\n\u003e \n\u003e \n\nAgreed.\n\n\u003e to summerisie:\n\u003e  \n\u003e we can simply document that this feature requires libvirt 6.3+ and is only supported in a fully upgraded cloud.\n\u003e \n\nSounds good to me.\n\n\u003e  \n\u003e we should not requrie a compute service bump for this feature as its a virt driver speciffic capablity not a compute manager capablity.\n\nAbsolutely yes, please KISS.\n\n\u003e \n\u003e we shoudl ahve a prefilter to transform the extra spec/image property into a required trait and that shoudl be always enabled.\n\nThat\u0027s where I disagree : I only see the need of a trait if that capability was virt-driver agnostic. This is not the case (very specific to QEMU) and except saying \u0027look, I\u0027m libvirt\u0027, I don\u0027t really see the need.\nIn other words, it would only ensure you land on a libvirt node, which is something operators already set up with aggregates and flavor extra specs in case of an heteregonous cloud in terms of hypervisors.\n\n\u003e \n\u003e we do not need a min libvirt check in the code since our min libvirt is higher then\n\u003e the version where this was intoduced on master. SLURP shoudl not affect that because the prefilter will enfoce that vms can only go to the host that support this.\n\nI\u0027d rather prefer we put some service version check in the API only for the time of this cycle, rather than selling a trait which would only be necessary for mixed hypervisors cloud (and even not necessary if the operator did the necessary actions for ensuring the shard correctly)","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"36857e56b6fba00ea4e006b8f32bcb9bb1acdfc1","unresolved":true,"context_lines":[{"line_number":62,"context_line":"* Provide new compute ``COMPUTE_NET_VIRTIO_PACKED`` capablity trait."},{"line_number":63,"context_line":"  This trait can be required/forbidden by user. Nova-compute agent"},{"line_number":64,"context_line":"  will automatically set this trait to the resource provider summary"},{"line_number":65,"context_line":"  if libvirt version is higher than 6.3"},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"* This spec will update scheduling process. ALL_REQUEST_FILTERS will be"},{"line_number":68,"context_line":"  extended with new filter packed_virtqueue_filter. It will update RequestSpec"}],"source_content_type":"text/x-rst","patch_set":6,"id":"ddedaa18_952d60d7","line":65,"in_reply_to":"7c37fe25_0c523958","updated":"2023-11-07 18:44:29.000000000","message":"the prefilter and triat would be considtnet for what we do for cyborg\n\nhttps://github.com/openstack/nova/blob/master/nova/scheduler/request_filter.py#L258-L271\n\nand ephmeral disk encyption\nhttps://github.com/openstack/nova/blob/master/nova/scheduler/request_filter.py#L396-L428\n\nthe image prprites could be a minor extention to the generic image metadtat prefilter\n\nhttps://github.com/openstack/nova/blob/master/nova/scheduler/request_filter.py#L198-L239\n\n\n\nfor vtpm, sev and pci numa affinity polices instead of prefilters\n\nhttps://github.com/openstack/nova/blob/b64ecb0cc776bd3eced674b0f879bb23c8a4b486/nova/scheduler/utils.py#L274-L397\n\nwe incldue the traits in ResourceRequest.from_request_spec\n\nhttps://github.com/openstack/nova/blob/b64ecb0cc776bd3eced674b0f879bb23c8a4b486/nova/scheduler/utils.py#L158-L195\n\nthat is arguablel where the trait request should be added not a prefilter if we were to have a trait for this capablity.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"e841744a96c93fd0714ee89155231649061a826e","unresolved":true,"context_lines":[{"line_number":122,"context_line":"Upgrade impact"},{"line_number":123,"context_line":"--------------"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"None"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Implementation"},{"line_number":128,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":6,"id":"d154b125_1a4efbab","line":125,"updated":"2023-11-07 14:19:38.000000000","message":"as I said in PS5, that means that if operators upgrade their computes one by one, only new upgraded computes will support that feature for a while. In terms of capacity planning, this shouldn\u0027t be a problem but this is worth noting here.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"acfbc6d2eb9ea1fc1ed4bc10a16497389cee1a2f","unresolved":true,"context_lines":[{"line_number":122,"context_line":"Upgrade impact"},{"line_number":123,"context_line":"--------------"},{"line_number":124,"context_line":""},{"line_number":125,"context_line":"None"},{"line_number":126,"context_line":""},{"line_number":127,"context_line":"Implementation"},{"line_number":128,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":6,"id":"d5f25afc_10f5b02a","line":125,"in_reply_to":"d154b125_1a4efbab","updated":"2023-11-07 15:13:50.000000000","message":"fair.\n\nwe can note it more for documenation then anything else.\nthe other thing to condier is that just because we enabel the ablity to use virtio-ring packed format, it does not mean the vm actully will enabel it.\nform a schudling point of view we have to assume tehy are however.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"e841744a96c93fd0714ee89155231649061a826e","unresolved":true,"context_lines":[{"line_number":151,"context_line":"Dependencies"},{"line_number":152,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"None"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"Testing"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":6,"id":"f358acc5_82966337","line":154,"updated":"2023-11-07 14:19:38.000000000","message":"Ahem, really? Shouldn\u0027t we depend on https://review.opendev.org/c/openstack/os-vif/+/859574 and https://review.opendev.org/c/openstack/neutron/+/869510 ?","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"950cf3413e632c03fcaaaa27709450a20b869049","unresolved":false,"context_lines":[{"line_number":151,"context_line":"Dependencies"},{"line_number":152,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":153,"context_line":""},{"line_number":154,"context_line":"None"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"Testing"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":6,"id":"2d2e5803_b76aabcc","line":154,"in_reply_to":"f358acc5_82966337","updated":"2023-11-07 14:39:46.000000000","message":"\u003e Ahem, really? Shouldn\u0027t we depend on https://review.opendev.org/c/openstack/os-vif/+/859574 and https://review.opendev.org/c/openstack/neutron/+/869510 ?\n\nas I explain just below, I mixed two things. Nevermind","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"e841744a96c93fd0714ee89155231649061a826e","unresolved":true,"context_lines":[{"line_number":156,"context_line":"Testing"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"None"},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"Documentation Impact"},{"line_number":162,"context_line":"\u003d\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":6,"id":"068f873b_aa1d231b","line":159,"updated":"2023-11-07 14:19:38.000000000","message":"really? Shouldn\u0027t we somehow do functional tests ? Also, shall we maybe have either some upstream jobs, a 3rd-party CI for this or some other way to know it works fine ?","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":7166,"name":"Sylvain Bauza","email":"sbauza@redhat.com","username":"sbauza"},"change_message_id":"950cf3413e632c03fcaaaa27709450a20b869049","unresolved":true,"context_lines":[{"line_number":156,"context_line":"Testing"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"None"},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"Documentation Impact"},{"line_number":162,"context_line":"\u003d\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":6,"id":"9f5f019f_77d548bd","line":159,"in_reply_to":"068f873b_aa1d231b","updated":"2023-11-07 14:39:46.000000000","message":"this looks very doable as part of a Tempest test.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"4ab900f98308237dc794f163d24fa68e3b24d0b2","unresolved":true,"context_lines":[{"line_number":156,"context_line":"Testing"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"None"},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"Documentation Impact"},{"line_number":162,"context_line":"\u003d\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":6,"id":"5d0dfc90_24f92464","line":159,"in_reply_to":"289e6fb0_56487a9a","updated":"2023-11-07 15:15:21.000000000","message":"by the way we generally require unit and funcitonal tests for all feature like this in the libvirt driver so i expect thos to be added too.\n\nwe can explictly document that if we want too as well bug unit and funcitonal tests are our geneal baseline for all new features and tempest is encurraged.\n\ni dont think we need any 3rd party ci here .","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"acfbc6d2eb9ea1fc1ed4bc10a16497389cee1a2f","unresolved":true,"context_lines":[{"line_number":156,"context_line":"Testing"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":158,"context_line":""},{"line_number":159,"context_line":"None"},{"line_number":160,"context_line":""},{"line_number":161,"context_line":"Documentation Impact"},{"line_number":162,"context_line":"\u003d\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":6,"id":"289e6fb0_56487a9a","line":159,"in_reply_to":"9f5f019f_77d548bd","updated":"2023-11-07 15:13:50.000000000","message":"we shoudl be able to test this in tempest yes\nlikely we will need a new comptue feature flag and we can enabel this in any of our jobs since the versision provdied by ubuntu 22.04 shoudl be sufficent.\n\nthe reason we will need to compute feature flag is because master tempest is used on branches before we added this feature.","commit_id":"d165f96b009389973891c57b08b8f3ce9d62bc41"}]}