)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"4c30b49b4cb7391901caa929997c453d6bb38238","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"39427fd4_dbb92267","updated":"2026-01-20 16:40:21.000000000","message":"-1 just to have the discussion on the root cause here","commit_id":"ee070aa2c3d07d5878b1bab7e1aee34cf657dfa8"},{"author":{"_account_id":21798,"name":"Bernard Cafarelli","email":"bcafarel@redhat.com","username":"bcafarel"},"change_message_id":"fae8f51a3f70b9f82d26a94b5b77e72c10fce978","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"fbc60c0f_33e66ab8","updated":"2026-01-20 15:51:06.000000000","message":"Small change to help with that issue (and also avoids re-applying existing state)","commit_id":"ee070aa2c3d07d5878b1bab7e1aee34cf657dfa8"}],"neutron/plugins/ml2/drivers/mech_sriov/agent/pci_lib.py":[{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"4c30b49b4cb7391901caa929997c453d6bb38238","unresolved":true,"context_lines":[{"line_number":86,"context_line":"        # See LP#2138636, setting the VF state (even the same one), could"},{"line_number":87,"context_line":"        # disrupt the traffic."},{"line_number":88,"context_line":"        if self.get_vf_state(vf_index) \u003d\u003d LinkState(link_state).name:"},{"line_number":89,"context_line":"            return"},{"line_number":90,"context_line":""},{"line_number":91,"context_line":"        vf_config \u003d {\u0027vf\u0027: vf_index, \u0027link_state\u0027: link_state}"},{"line_number":92,"context_line":"        ip.link.set_vf_feature(vf_config)"}],"source_content_type":"text/x-python","patch_set":2,"id":"a56bf7e1_a7f32e44","line":89,"updated":"2026-01-20 16:40:21.000000000","message":"So while I think the change is Ok, the bug is/was actually due to a change in the i40e driver, subsequently fixed with [0]. It was something one of our customers ran into a few months back (i had a similar patch ready) but we fixed the kernel instead.\n\nThe better fix would be/should be to check state in the DPDK code (not neutron) but that was found to be too difficult to do in the short term.\n\nFor that reason I don\u0027t think we need this change, thoughts?\n\n[0] https://lore.kernel.org/netdev/20251021154439.180838-1-robert.malz@canonical.com/T/#u","commit_id":"ee070aa2c3d07d5878b1bab7e1aee34cf657dfa8"},{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"52fe0c3f1fe3f0524483f932389736cede0f5263","unresolved":true,"context_lines":[{"line_number":86,"context_line":"        # See LP#2138636, setting the VF state (even the same one), could"},{"line_number":87,"context_line":"        # disrupt the traffic."},{"line_number":88,"context_line":"        if self.get_vf_state(vf_index) \u003d\u003d LinkState(link_state).name:"},{"line_number":89,"context_line":"            return"},{"line_number":90,"context_line":""},{"line_number":91,"context_line":"        vf_config \u003d {\u0027vf\u0027: vf_index, \u0027link_state\u0027: link_state}"},{"line_number":92,"context_line":"        ip.link.set_vf_feature(vf_config)"}],"source_content_type":"text/x-python","patch_set":2,"id":"6254e5f5_479af5d9","line":89,"in_reply_to":"a56bf7e1_a7f32e44","updated":"2026-01-21 07:00:58.000000000","message":"Sorry, I don\u0027t have access to the link you provided.\n\nThis is related to [1]. The fix was merged some months ago but still not released. The problem is that there could be some users with older drivers hitting this problem for a while.\n\nWe can add a note in the related code with a recommendation to remove it once we consider that all users will have a minimum version having this patch (that could be in some releases).\n\n[1]https://github.com/torvalds/linux/commit/a7ae783da0b919550e260aebfca1c6ef030b99a4","commit_id":"ee070aa2c3d07d5878b1bab7e1aee34cf657dfa8"}],"neutron/tests/unit/plugins/ml2/drivers/mech_sriov/agent/test_pci_lib.py":[{"author":{"_account_id":21798,"name":"Bernard Cafarelli","email":"bcafarel@redhat.com","username":"bcafarel"},"change_message_id":"fae8f51a3f70b9f82d26a94b5b77e72c10fce978","unresolved":true,"context_lines":[{"line_number":102,"context_line":"        self.mock_ip_device.link.set_vf_feature.assert_called_once_with(vf)"},{"line_number":103,"context_line":""},{"line_number":104,"context_line":"    def test_set_vf_state_same_value(self):"},{"line_number":105,"context_line":"        # The currennt value of the link_state is now the same as the one that"},{"line_number":106,"context_line":"        # needs to be set. In this case, there is no need to write it again."},{"line_number":107,"context_line":"        # state\u003dTrue, auto\u003dFalse --\u003e link_state\u003denable"},{"line_number":108,"context_line":"        link_state \u003d 1"}],"source_content_type":"text/x-python","patch_set":2,"id":"c6ca370c_7fbf20b9","line":105,"range":{"start_line":105,"start_character":14,"end_line":105,"end_character":22},"updated":"2026-01-20 15:51:06.000000000","message":"I will nit on the typo, but definitely not worth a respin","commit_id":"ee070aa2c3d07d5878b1bab7e1aee34cf657dfa8"}]}