| *** Sundar has joined #openstack-cyborg | 02:53 | |
| *** Li_Liu has joined #openstack-cyborg | 02:54 | |
| *** Coco_gao has joined #openstack-cyborg | 02:58 | |
| Coco_gao | Hi all | 02:58 |
|---|---|---|
| Li_Liu | Hi Coco | 02:59 |
| *** ikuo_o has joined #openstack-cyborg | 02:59 | |
| yikun | hello | 02:59 |
| ikuo_o | hello, yikun. | 03:00 |
| Sundar | Hi all | 03:00 |
| yikun | ikuo_o: hi | 03:00 |
| Sundar | #startmeeting openstack-cyborg | 03:00 |
| openstack | Meeting started Wed Apr 24 03:00:48 2019 UTC and is due to finish in 60 minutes. The chair is Sundar. Information about MeetBot at http://wiki.debian.org/MeetBot. | 03:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 03:00 |
| *** openstack changes topic to " (Meeting topic: openstack-cyborg)" | 03:00 | |
| openstack | The meeting name has been set to 'openstack_cyborg' | 03:00 |
| Coco_gao | #info Coco_gao | 03:00 |
| Li_Liu | #info Li_Liu | 03:01 |
| Sundar | #info Sundar | 03:01 |
| yikun | #info yikun | 03:01 |
| Sundar | Good to have many of us right at the beginning | 03:01 |
| Sundar | Agenda: PTG session time allocation, Summit project update slides | 03:02 |
| Sundar | Anything else? | 03:02 |
| Li_Liu | Have we decided the team dinner date yet? | 03:03 |
| Sundar | It is Thu evening. But we haven't decided a place yet. I'll leave that to all of you | 03:03 |
| Li_Liu | ok | 03:04 |
| yikun | 2th May, right? | 03:04 |
| Sundar | yikun: Yes | 03:04 |
| yikun | Sundar: ok | 03:04 |
| Coco_gao | OK, maybe we can keep the old place. | 03:04 |
| Sundar | #topic PTG session planning | 03:04 |
| *** openstack changes topic to "PTG session planning (Meeting topic: openstack-cyborg)" | 03:04 | |
| Sundar | We have 3.5 hours on Thu afternoon (1:30-5) and 2.5 hours on Fri afternoon (1:30-4) because I need to leave by 4 to catch my flight | 03:05 |
| Sundar | Not much time! We need to plan the sessions tightly | 03:05 |
| Sundar | You can all continue on Fri eve and on Sat too, when I am gone | 03:05 |
| *** wangzhh has joined #openstack-cyborg | 03:06 | |
| Sundar | BTW, Fri morn are the cross-projects with Nova and ironic | 03:06 |
| wangzhh | #info wangzhh | 03:06 |
| *** Yumeng has joined #openstack-cyborg | 03:06 | |
| Sundar | Shall we start allocating times to the topics in Looking at: https://etherpad.openstack.org/p/cyborg-ptg-train ? | 03:07 |
| *** xinranwang has joined #openstack-cyborg | 03:07 | |
| Sundar | s/Looking at// | 03:07 |
| xinranwang | Hi all | 03:08 |
| yikun | xinranwang: hi | 03:08 |
| Sundar | Hi xinranwang, wangzhh and Yumeng | 03:08 |
| xinranwang | #info xinranwang | 03:08 |
| xinranwang | hi Sundar yikun | 03:08 |
| Yumeng | Hi Sundar, and all | 03:08 |
| Yumeng | # info Yumeng | 03:08 |
| wangzhh | Hi Sundar, hi all. | 03:09 |
| Sundar | Retrospective: keep it to 10 min? | 03:09 |
| Sundar | Too short? | 03:09 |
| yikun | Sundar: I think we should classify these topic in etherpad, such as, driver, api, common/design...and then do the allocation tiems | 03:10 |
| yikun | * times | 03:10 |
| Coco_gao | 10min if no other people join, because we all familar with the project status. | 03:11 |
| Coco_gao | Sundar, agree with yikun | 03:11 |
| Sundar | yikun: ok. The main topics in my mind are: Train Goals, Networked FPGAs, mapping names to UUIDs, new drivers/devices (incl. Ascend) | 03:11 |
| Coco_gao | Better to classify | 03:11 |
| wangzhh | Yikun, do you need more time for topic #line 58? | 03:12 |
| Sundar | Some of them are cross-cutting across these categories, esp. first 2 | 03:12 |
| yikun | wangzhh: checking... | 03:12 |
| yikun | wangzhh: bitstream management? So what's the "more time" mean? you mean the more time to discuss in PTG? | 03:13 |
| wangzhh | Yep, I saw you leave comment 'expand the bitstream to more generic level'. So... | 03:14 |
| Sundar | yikun, all: In my mind, bitstream mgmt means Cyborg provides APIs to parse metadata for bitstreams, validate and upload in Glance. What else do you have in mind? | 03:14 |
| Sundar | I think firmware update etc. are usually done as part of host config, with pupper/ansible/... | 03:15 |
| Sundar | *puppet | 03:15 |
| yikun | wangzhh: ok, make sense, it is just some idea for "Do not provide specific API for specific type driver." | 03:16 |
| wangzhh | Got it. | 03:16 |
| ikuo_o | Sundar: I'm thinking bitstream md5 checksum to check integrity. | 03:16 |
| Coco_gao | I thought there were a generic Driver in code, but we don't use that or inherit from that. | 03:16 |
| Sundar | ikuo_o: That is up to the vendor stack, because they may use different checksums, authentication/decryption schemes, or even do it in hardware in some cases | 03:17 |
| Coco_gao | About the bug, do we need specific time for that? | 03:18 |
| wangzhh | Coco, Yep, but, actually, it is out of date. Should be improved as our new design. | 03:18 |
| yikun | Coco_gao: yes, https://github.com/openstack/cyborg/blob/master/cyborg/accelerator/drivers/generic_driver.py | 03:18 |
| Coco_gao | Thanks, yikun. The generic driver is too old. | 03:18 |
| Sundar | Have you all looked at the driver API in Line 396 https://review.opendev.org/#/c/608624/4/specs/stein/approved/cyborg-api-wflows-for-instance-ops.rst ? The can_handle API can be dropped but can you review the rest? | 03:18 |
| yikun | So these APIs is belong to the generic_driver.py? | 03:20 |
| Sundar | Anyways, I'll take a first stab like this: 30 min each for Train Goals, Networked FPGAs, mapping names to UUIDs, new drivers/devices, NTT's discussion. That is 2.5 hours. Now we can add more time to some of these topics and fit more on Friday | 03:20 |
| Coco_gao | Sundar, OK | 03:20 |
| yikun | Sundar: sounds good | 03:21 |
| Sundar | May be new devices needs more time | 03:21 |
| ikuo_o | Thanks, Sundar! | 03:22 |
| Sundar | ikuo_o: Welcome :) | 03:22 |
| Coco_gao | I need some time in Friday to dicuss the dirty data caused by conductor diff. | 03:23 |
| Sundar | Coco_gao: Good. Conductor diff: 15-30 min? | 03:23 |
| Coco_gao | That's good | 03:23 |
| Sundar | PCI slot change on reboot -- why is it any different than Nova's PCI subsystem? | 03:23 |
| Coco_gao | I will make sure with shaohe, and see if we need to discuss that. | 03:25 |
| Sundar | wangzhh, Coco_gao: any performance countres planned for GPUs? I am thinking just report driver-reported perf counters from agent to a collectd plugin. From there, it can be consumed by telemetry/ceilometer, prometheus, .... | 03:25 |
| Coco_gao | Right now , we can skip that | 03:25 |
| Sundar | Coco_gao: ok | 03:26 |
| yikun | Who is the owner of L42 idea? I plus some info on it | 03:27 |
| Sundar | yikun: do you mean SmartNIC? | 03:28 |
| yikun | "Performance/health monitoring for GPUs or any other devices?" | 03:28 |
| wangzhh | Whuuu, difficult to monitor it now, because when we passthrough the GPU in the vm. Host will lost control of this device. I'll consider about that. | 03:30 |
| Sundar | yikun: I responded too. I would like to find an owner. | 03:30 |
| ikuo_o | I'm not the owner but the function seems useful. | 03:31 |
| wangzhh | Does FPGA or other devices have the same scenario? | 03:31 |
| Sundar | wangzhh: yea, same problem for FPGAs, but we should have the PF in the host from where we can get shell counters. | 03:31 |
| Sundar | Besides, we also have dependence on the vendor stack | 03:31 |
| Sundar | yikun: Does Ascend support SR-IOV? | 03:32 |
| wangzhh | Sundar, Yep. Depend on vendor stack. | 03:32 |
| Sundar | If Ascend supports SR-IOV, and we can get performance counters from its PF, we have a potential use case. Otherwise, we'll focus on other practical things and then come to this. | 03:33 |
| yikun | Sundar: ascend 310 don't support sr-iov yet | 03:35 |
| Sundar | Then shall we deprioritize this topic for PTG? | 03:35 |
| xinranwang | ascend support entire pci passthrough right | 03:36 |
| yikun | yes | 03:36 |
| yikun | passthrough is supported. | 03:36 |
| wangzhh | Sundar, Depend on the priorities and our time. | 03:36 |
| yikun | Sundar: I think we still can add this topic in PTG | 03:36 |
| Sundar | OK, I only said de-prioritized. So, if there's time after topics, sure | 03:37 |
| Sundar | Driver support for Movidius - who owns this? | 03:37 |
| yikun | Shaohe? | 03:38 |
| xinranwang | Sundar: i think is Shaohe | 03:38 |
| Coco_gao | Coco | 03:38 |
| Sundar | RISC-V? | 03:38 |
| Coco_gao | We use Movidius in the Edge. | 03:39 |
| yikun | Movidius is a USB device, right? | 03:39 |
| Sundar | Coco_gao: Cool. Maybe both you and Shaohe can drive it? | 03:39 |
| Coco_gao | Yes | 03:39 |
| Coco_gao | It's USB right now, in order to use it, we need to install openvino | 03:40 |
| Sundar | I should a topic for post-PTG plans. That's important. Merge pilot with master, add FPGA CI, etc. are important for Nova intgration | 03:40 |
| Coco_gao | Sundar, agree | 03:41 |
| Coco_gao | the most important thing is nova integration | 03:41 |
| yikun | Sundar: thanks for your hard work, and the pilot is the most import one. | 03:41 |
| yikun | * important | 03:41 |
| Sundar | Xinranwang is looking at tempest plugins for Cyborg CI, right? I can help her. Anybody else wants to help? It will be a set of CI test cases. So, could be lots of work. | 03:42 |
| xinranwang | yes, that's also a part of post ptg plan | 03:43 |
| wangzhh | I can take part of these if it is not emergency. | 03:44 |
| Sundar | wangzhh: It is not an emergency, but until we get the CI going, Nova will probably not merge. | 03:44 |
| Sundar | Please see https://etherpad.openstack.org/p/ptg-train-xproj-nova-cyborg | 03:45 |
| xinranwang | does nova guys require to have CI done before they merge? | 03:45 |
| Yumeng | I can also join once there is time. | 03:46 |
| Yumeng | btw, my visa is still on administrative processing. only 4days to go. I'm highly probably not be able to attend the summit and ptg. the new project update looks great for me! | 03:46 |
| xinranwang | thanks wangzhh Yumeng | 03:46 |
| Sundar | xinranwang: Please see that link. | 03:46 |
| Sundar | Yumeng: That's tough. We'll keep you updated | 03:47 |
| Sundar | Thanks for the slides review | 03:47 |
| Yumeng | but I will join remotely, will we have remote access just like last ptg did? | 03:48 |
| xinranwang | Sundar: ok, will look into it. Thanks | 03:48 |
| Sundar | Yumeng: I'll check and get back | 03:48 |
| Yumeng | Sundar:ok. thanks | 03:48 |
| xinranwang | nice to have remote access, I can join remotely also :) | 03:49 |
| Sundar | @all, shall I take a stab at time assignments based on the above, and you can add your comments if you want changes? | 03:50 |
| Sundar | Silence means yes, I think :) | 03:52 |
| ikuo_o | It's ok. | 03:52 |
| Sundar | OK, cool. I'll update and ping you all | 03:53 |
| Sundar | #topic Tomorrow's Zoom meeting | 03:53 |
| *** openstack changes topic to "Tomorrow's Zoom meeting (Meeting topic: openstack-cyborg)" | 03:53 | |
| Sundar | I will present the PTG demo. This is highly important to get consensus across projects. Please attend and provide your feedback. | 03:54 |
| ikuo_o | I will attend the meeting. | 03:54 |
| Sundar | Should we take up anything else tomorrow? | 03:54 |
| Sundar | ikuo_o: Thanks | 03:54 |
| ikuo_o | I'm preparing material for PTG, but nothing for tomorrow. | 03:55 |
| Sundar | Cool. | 03:55 |
| Sundar | #topic AoB | 03:55 |
| *** openstack changes topic to "AoB (Meeting topic: openstack-cyborg)" | 03:55 | |
| Sundar | Anything else? | 03:55 |
| ikuo_o | It's ok, thanks Sundar. | 03:56 |
| Sundar | Great. See you all tomorrow. Have a good day! | 03:57 |
| Sundar | #endmeeting | 03:57 |
| *** openstack changes topic to "Pending patches (Meeting topic: openstack-cyborg)" | 03:57 | |
| openstack | Meeting ended Wed Apr 24 03:57:10 2019 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 03:57 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2019/openstack_cyborg.2019-04-24-03.00.html | 03:57 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg/2019/openstack_cyborg.2019-04-24-03.00.txt | 03:57 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2019/openstack_cyborg.2019-04-24-03.00.log.html | 03:57 |
| *** Sundar has quit IRC | 04:00 | |
| *** Yumeng has quit IRC | 04:03 | |
| *** Li_Liu has quit IRC | 04:08 | |
| *** ikuo_o has quit IRC | 05:36 | |
| *** Coco_gao has quit IRC | 06:08 | |
| *** wangzhh has quit IRC | 06:15 | |
| *** xinranwang has quit IRC | 07:22 | |
| *** helenafm has joined #openstack-cyborg | 08:04 | |
| openstackgerrit | Merged openstack/cyborg master: Use opendev repository https://review.opendev.org/654899 | 08:17 |
| *** tetsuro has joined #openstack-cyborg | 09:06 | |
| *** tetsuro has quit IRC | 09:24 | |
| *** links has joined #openstack-cyborg | 09:31 | |
| *** tetsuro has joined #openstack-cyborg | 09:31 | |
| *** tetsuro has quit IRC | 09:34 | |
| *** tetsuro has joined #openstack-cyborg | 09:41 | |
| *** tetsuro has quit IRC | 09:56 | |
| *** helenafm has quit IRC | 11:45 | |
| *** irclogbot_2 has quit IRC | 12:55 | |
| *** irclogbot_1 has joined #openstack-cyborg | 12:55 | |
| *** altlogbot_3 has quit IRC | 12:57 | |
| *** altlogbot_1 has joined #openstack-cyborg | 12:57 | |
| *** helenafm has joined #openstack-cyborg | 13:02 | |
| *** helenafm has quit IRC | 13:46 | |
| *** helenafm has joined #openstack-cyborg | 14:04 | |
| *** links has quit IRC | 14:51 | |
| *** helenafm has quit IRC | 16:00 | |
| *** efried is now known as efried_rollin | 16:04 | |
| *** altlogbot_1 has quit IRC | 16:09 | |
| *** altlogbot_1 has joined #openstack-cyborg | 16:11 | |
| *** altlogbot_1 has quit IRC | 16:43 | |
| *** altlogbot_3 has joined #openstack-cyborg | 16:44 | |
| *** altlogbot_3 has quit IRC | 16:53 | |
| *** altlogbot_0 has joined #openstack-cyborg | 16:54 | |
| *** openstackgerrit has quit IRC | 18:57 | |
| *** efried_rollin is now known as efried | 20:14 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!