06:04:35 #startmeeting cyborg 06:04:35 Meeting started Fri Oct 29 06:04:35 2021 UTC and is due to finish in 60 minutes. The chair is brinzhang_. Information about MeetBot at http://wiki.debian.org/MeetBot. 06:04:35 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 06:04:35 The meeting name has been set to 'cyborg' 06:04:58 #topic roll call 06:05:02 #topic roll call 06:05:08 #info brinzhang 06:05:20 #info songwenping 06:06:09 #topic specs proposed in Yoga 06:07:33 wenpingsong__: with the enable/disable device status spec, I left some comments, and you can check again 06:07:42 https://review.opendev.org/c/openstack/cyborg-specs/+/815460 06:08:17 do you have some questions about this topic? 06:08:38 vmaccel: hi are you around? 06:09:23 i havenot look into your comment. 06:09:55 dont worry, you can review it after meeting 06:10:05 hi eric-xie 06:10:12 i have the question about the show api return value. 06:10:56 wenpingsong__: you mean the attributes_list column? 06:11:03 Hi all 06:12:11 hi, welcome join in our meeting 06:12:31 yes, and num_accelerators etc. 06:13:28 ah, num_accelerators also a new column will add with this feature 06:14:13 and she is not describe in *propose changes* 06:15:16 wenpingsong: "attributes_list" you can paste your local device show, then we trace it to 06:16:02 i think we shouldnot change the return value except is_maintaining related this feature. 06:17:17 eric-xie: you can attach the device show return value as my env is not available. 06:17:40 for this spec: v 06:17:42 https://review.opendev.org/c/openstack/cyborg-specs/+/815460 06:19:17 as discussed on PTG, we should record the num of the device used by instance, if you dont add it, how to check the device is available or in-use? 06:20:15 we can add it in other feature 06:22:03 songweiping: attach the return value here? 06:22:21 to the spec pls 06:23:59 have added it. 06:25:42 thanks 06:27:35 songwenping: no, we can talked under the spec, num_accelerators is useful for this feature, IMO, it should added in this spec, and it also describe in spec 06:30:54 ok, but there are also some extra values such as rp_uuid and bitstream_id. 06:31:56 another spec is xilin driver spec need to reiew: https://review.opendev.org/c/openstack/cyborg-specs/+/803103 06:32:44 songwenping: yeah, we should have a clear decision with disable/enable feature changes 06:33:14 pls add your doubt on the spec 06:33:27 ok i donot know where she get the return value. 06:34:53 eric-xie: xilin driver spec need to fix the pep8 failed 06:35:03 I changed it and submitted it based on ptg's comments. 06:35:21 OK, i will fix it as soon. 06:35:36 ack, added it in my list 06:35:54 the content is LGTM 06:38:10 ok 06:38:47 another spec is pmem driver spec: https://review.opendev.org/c/openstack/cyborg-specs/+/799173 06:40:34 Shuai Qian proposed openstack/cyborg master: support recreating mdevs after node reboots https://review.opendev.org/c/openstack/cyborg/+/815935 06:40:34 eric-xie: but i still have some nits for your spec, i will comment later. 06:41:08 songwenping: got it. 06:41:27 the pmem driver still have question about the data to report 06:41:57 as nova doesnot report the pmem as a rp 06:43:46 i have talked with xinran on wechat but has no response 06:44:24 ack 06:44:30 #topic pending patches 06:44:41 cyborg project: https://review.opendev.org/q/project:openstack%252Fcyborg+status:open 06:44:46 python-cyborgclient: https://review.opendev.org/q/project:openstack%252Fpython-cyborgclient+status:open 06:44:51 cyborg-tempest-plugin: https://review.opendev.org/q/project:openstack%252Fcyborg-tempest-plugin+status:open 06:45:02 songwenping, eric-xie: pls review 06:46:08 ack 06:47:23 ack 06:56:02 we can let this patch go:https://review.opendev.org/c/openstack/python-cyborgclient/+/809093 06:56:46 brinzhang: if there are nothing propose, we can end the meeting. 06:57:00 yeah, let's ending 06:57:49 #end meeting 06:57:54 #endmeeting 06:58:29 #endmeeting 06:58:41 #endmeeting cyborg 07:04:27 brinzhang: Error: Can't start another meeting, one is in progress. Use #endmeeting first. 07:04:39 #endmeeting