)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"af6f8d952240c960790ea6203da824e4f97b0a25","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"e404744c_0acd6950","updated":"2024-02-15 11:16:47.000000000","message":"clean cherry pick","commit_id":"0b7a59ad2812c3501332155211b18f224cd55bca"},{"author":{"_account_id":17685,"name":"Elod Illes","email":"elod.illes@est.tech","username":"elod.illes"},"change_message_id":"06ee92a203602b185aaa53479677b967443fd96f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"534659a3_dd8252a6","updated":"2024-02-15 12:44:32.000000000","message":"my question: is that possible that this is used already for some operators (having intel hosts)? as this will remove the feature for them, so it would be kind of downgrading some functionality. (I do understand that this prevents AMD hosts to run Windows guests, but maybe both should be possible to work in the backport.)\nWhat do you think?\n\notherwise this is clean cherry pick.","commit_id":"0b7a59ad2812c3501332155211b18f224cd55bca"},{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"2d213d1216210cbdd8db14fad8bbcc5cd9dabc81","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"a9406133_f915f486","in_reply_to":"534659a3_dd8252a6","updated":"2024-02-15 12:53:25.000000000","message":"We discussed a similar idea on the original patch - namely, can we be smart about enabling evmcs only on Intel hosts, and while that\u0027s possible, it\u0027s essentially a small feature, so we decided to fix the bug first. It\u0027s not idea, because as you said, there\u0027s technically a small regression on Intel hosts, but we figured the outright breakage on AMD hosts takes priority.","commit_id":"0b7a59ad2812c3501332155211b18f224cd55bca"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"534983b7f9ad4ce73d712a139508c986eeb2b60a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"ec26168e_2c129583","in_reply_to":"534659a3_dd8252a6","updated":"2024-02-15 13:05:17.000000000","message":"just capturing what i said on irc.\n\nsupporting this on only a subset of x86 hosts would require a new trait and a new image/flavour property to request the new functionality explicitly.\n\nsince we cannot rely on this funcitonlay on all x86 hosts we need to have scheduler support for this adn a way to opt in to it explicitly rather then relying solely on (OS_TYPE\u003dwindows)\n\nwith this backport existing instance (on intel hosts) will continue to have the enlightenemtn enable until the next time the xml is regenated\n\nin practice this measn hard-reboot, cold migrate/resize, rescue, shelve or rebuild/evacuate.\n\nif the instance is live migrate after this change it will not modify the xml however live migrate is also broke due to a second enlignemtne which we also need to remove https://review.opendev.org/c/openstack/nova/+/904183","commit_id":"0b7a59ad2812c3501332155211b18f224cd55bca"},{"author":{"_account_id":17685,"name":"Elod Illes","email":"elod.illes@est.tech","username":"elod.illes"},"change_message_id":"5d24694a172ecf46fb207f780598689a42bfd377","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"1127e7c5_db337144","in_reply_to":"a9406133_f915f486","updated":"2024-02-15 13:07:41.000000000","message":"ACK, as I said on IRC [1], i also tend to agree on backporting this, though i let other stable cores to look at this patch and share their view as I\u0027m not super confident on this.\n\n[1] https://meetings.opendev.org/irclogs/%23openstack-nova/%23openstack-nova.2024-02-15.log.html#t2024-02-15T12:45:30","commit_id":"0b7a59ad2812c3501332155211b18f224cd55bca"}]}