*** Dutch has joined #openstack-cyborg | 02:27 | |
*** crushil_ has joined #openstack-cyborg | 02:57 | |
*** crushil has joined #openstack-cyborg | 02:57 | |
*** crushil_ has quit IRC | 03:03 | |
*** crushil_ has joined #openstack-cyborg | 05:29 | |
*** Dutch has quit IRC | 05:53 | |
*** jkilpatr has quit IRC | 07:28 | |
*** crushil has quit IRC | 07:29 | |
*** jkilpatr has joined #openstack-cyborg | 07:39 | |
*** crushil has joined #openstack-cyborg | 07:47 | |
*** crushil has quit IRC | 08:27 | |
*** crushil has joined #openstack-cyborg | 08:37 | |
*** crushil__ has joined #openstack-cyborg | 08:39 | |
*** crushil_ has quit IRC | 08:41 | |
*** crushil has quit IRC | 08:43 | |
*** crushil__ has quit IRC | 08:45 | |
*** dims has quit IRC | 10:57 | |
*** dims has joined #openstack-cyborg | 10:58 | |
*** jkilpatr has quit IRC | 12:07 | |
*** jkilpatr has joined #openstack-cyborg | 12:09 | |
*** zhipeng has joined #openstack-cyborg | 14:17 | |
*** crushil__ has joined #openstack-cyborg | 14:23 | |
*** crushil_ has joined #openstack-cyborg | 14:24 | |
*** NokMikeR has joined #openstack-cyborg | 14:27 | |
*** crushil__ has quit IRC | 14:28 | |
*** Li_Liu has joined #openstack-cyborg | 14:46 | |
*** guhcampos has joined #openstack-cyborg | 14:57 | |
zhipeng | dear beloved | 15:02 |
---|---|---|
NokMikeR | We are gathered here today | 15:02 |
zhipeng | #startmeeting openstack-cyborg | 15:03 |
openstack | Meeting started Wed Jan 17 15:03:03 2018 UTC and is due to finish in 60 minutes. The chair is zhipeng. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:03 |
openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:03 |
*** openstack changes topic to " (Meeting topic: openstack-cyborg)" | 15:03 | |
openstack | The meeting name has been set to 'openstack_cyborg' | 15:03 |
zhipeng | NokMikeR to welcome you back to the team lol | 15:03 |
zhipeng | #topic Roll Call | 15:03 |
*** openstack changes topic to "Roll Call (Meeting topic: openstack-cyborg)" | 15:03 | |
NokMikeR | :) | 15:03 |
NokMikeR | #info Mike Rooke, Nokia. | 15:03 |
zhipeng | #Howard | 15:03 |
zhipeng | shit | 15:03 |
zhipeng | #info Howard | 15:03 |
jkilpatr | #info Justin | 15:04 |
NokMikeR | dont worry you can fix that in post production. | 15:04 |
jkilpatr | good morning everyone. | 15:04 |
zhipeng | morning, evening, half-day struggling everyone | 15:04 |
NokMikeR | morning / evening all. Im back, not that I was actualy away, just busy trying to get something installed. | 15:05 |
zhipeng | NokMikeR always great to have you | 15:06 |
*** Vipparthy has joined #openstack-cyborg | 15:06 | |
Vipparthy | Good Morning Folks | 15:06 |
zhuli | #info zhuli | 15:06 |
zhuli | morning/evening everyone | 15:08 |
zhipeng | #topic Queens Dev Progress | 15:09 |
*** openstack changes topic to "Queens Dev Progress (Meeting topic: openstack-cyborg)" | 15:09 | |
zhipeng | #link https://review.openstack.org/#/q/status:open+project:openstack/cyborg | 15:09 |
Li_Liu | #info Li_Liu | 15:09 |
zhipeng | glad we finally managed to land lots of patches last week | 15:09 |
zhipeng | let's comb through the ones we have | 15:09 |
zhipeng | #info FPGA data modeling | 15:10 |
zhipeng | #link https://review.openstack.org/526559 | 15:10 |
Li_Liu | I have just submitted a patch for zhuli's comments | 15:10 |
zhipeng | zhuli have you reviewed the updated patch ? | 15:10 |
zhuli | reviewing now | 15:11 |
zhipeng | other than zhuli's comment, any other concerns about this spec ? | 15:11 |
*** Dutch has joined #openstack-cyborg | 15:14 | |
zhipeng | if there are no other concerns, we should land this spec today | 15:15 |
zhipeng | #action landing fpga data modeling spec today | 15:15 |
zhipeng | #info zhuli's api/db patch | 15:15 |
zhipeng | #link https://review.openstack.org/527396 | 15:15 |
zhipeng | jkilpatr crushil could either of you guys give a workflow +1 for this patch ? | 15:16 |
zhipeng | crushil_ | 15:16 |
jkilpatr | I'm still confused, the tox command should run linters but not actual test code. | 15:17 |
jkilpatr | unless the tox.ini is setup to run actual test code in the linter categories which I don't think it is | 15:17 |
zhuli | liliu, you mean the attributes are constant which will be initialized with deployable object together and do not need further methods to modify? | 15:18 |
zhipeng | Li_Liu per zhuli's question ? | 15:21 |
zhipeng | zhuli could you take a look at jkilpatr's problem ? | 15:22 |
Li_Liu | Zhuli, yes, that's what I meant | 15:22 |
Li_Liu | the attributes table should be transparent to users | 15:24 |
zhuli | jkilpatr, tox can support both linters check and unit test according to your command, for instance 'tox -epep8' for linters and 'tox -epy27' for unittest | 15:25 |
zhuli | the openstack-tox-pep8 and openstack-tox-py27 gerrit jobs run against these two situation | 15:27 |
jkilpatr | ok there we go I'll wf it | 15:28 |
jkilpatr | unless crushil_ has objections? | 15:29 |
zhuli | jkilpatr, you can see the detail in these links | 15:31 |
zhuli | http://logs.openstack.org/96/527396/11/check/openstack-tox-pep8/185864a/job-output.txt.gz#_2018-01-13_18_12_47_975849 | 15:31 |
zhuli | http://logs.openstack.org/96/527396/11/check/openstack-tox-py27/7cf69da/job-output.txt.gz#_2018-01-13_18_08_40_533805 | 15:31 |
*** zhipeng_ has joined #openstack-cyborg | 15:31 | |
*** zhipeng has quit IRC | 15:32 | |
*** Awagh has joined #openstack-cyborg | 15:32 | |
*** Awagh has quit IRC | 15:32 | |
zhipeng_ | #info crushil_'s generic driver | 15:34 |
zhipeng_ | #link https://review.openstack.org/525057 | 15:34 |
zhipeng_ | crushil_ any ETA on a full implementation ready for review and landing ? | 15:34 |
zhuli | liliu, ok, I'm ok with this, will help merge the patch if there is no other objections | 15:34 |
Li_Liu | Thanks a lot Zhuli | 15:35 |
crushil_ | zhipeng_, It's almost ready. Need to write accompanying unit tests | 15:35 |
crushil_ | You can start reviewing now | 15:35 |
zhipeng_ | sounds gr8t ! | 15:35 |
zhipeng_ | #action land zhuli's pacth today | 15:36 |
zhipeng_ | #action start reviewing crushil_'s generic driver patch | 15:36 |
zhipeng_ | i think what we miss now is the conductor's report functionality to sync up resource info with Placement | 15:37 |
zhipeng_ | jkilpatr could you help on this ? | 15:37 |
zhipeng_ | on the basis of zhuli's patch | 15:37 |
Vipparthy | Also Zhipeng, Any feedback on Cyborg-Nova Integration Code | 15:37 |
jkilpatr | zhipeng_, this is on crushil_'s patch? | 15:38 |
jkilpatr | or zhuli's patch? | 15:38 |
zhipeng_ | zhuli's | 15:38 |
jkilpatr | I'll look at it | 15:39 |
zhipeng_ | something like adding a report.py from nova-compute to cyborg-conductor | 15:39 |
zhipeng_ | with the corresponding modifications | 15:39 |
zhipeng_ | #action jkilpatr to start working on the cyborg-conductor report function | 15:40 |
zhipeng_ | Vipparthy the above mentioned code is about the cyborg-nova integration | 15:41 |
zhipeng_ | Dutch dude are you still around ? | 15:41 |
zhipeng_ | $topic Xilinx SDAccel introduction | 15:43 |
zhipeng_ | #topic Xilinx SDAccel introduction | 15:44 |
*** zhipeng_ has quit IRC | 15:44 | |
*** zhipeng has joined #openstack-cyborg | 15:45 | |
zhipeng | #topic Xilinx SDAccel introduction | 15:45 |
*** openstack changes topic to "Xilinx SDAccel introduction (Meeting topic: openstack-cyborg)" | 15:45 | |
zhipeng | So I'm glad to have Dutch from Xilinx to join our dev team | 15:45 |
zhipeng | Dutch the floor is yours sir | 15:45 |
zhuli | dutch welcome | 15:46 |
Dutch | Thanks! Hey everyone, I work with the Xilinx SDAccel team on our cloud program. I can help provide any resources or support from Xilinx. Also can help with SDAccel framework in general | 15:47 |
Dutch | We are still coming up to speed on the Cyborg project but the project looks great and we are happy to get involved | 15:48 |
zhipeng | Dutch could you provide some info on SDAccel ? | 15:49 |
Dutch | Sure, SDAccel is basically the framework for PCIe attached Xilinx FPGAs. Currently we use two PFs. PF0 is user space for DMA and PF1 is management for ioctls | 15:50 |
Dutch | Are people familiar with the driver at all? | 15:52 |
zhipeng | I think you could give a quick overview :P | 15:52 |
Dutch | Haha, I can't say I have done it over IRC before, this is usually consumed in slides with lots of images | 15:52 |
Dutch | Also, caution, I am not on the drivers team, so there are times when I don't know all the details | 15:53 |
zhipeng | haha i know | 15:53 |
zhipeng | maybe you could drop a link here ? | 15:53 |
zhipeng | for a good overview slide ? | 15:54 |
Dutch | Sure #link https://www.xilinx.com/html_docs/xilinx2017_4/sdaccel_doc/index.html | 15:54 |
zhuli | thanks, that's very helpful | 15:54 |
Dutch | These are most of the user guides in one place instead of dealing with the PDFs, contents are the same | 15:55 |
zhipeng | gr8t thx! | 15:55 |
zhipeng | are there any support yet in the OpenStack for SDAccel ? | 15:55 |
zhipeng | for example nova driver ? | 15:55 |
Dutch | There is also a helpful platform guide UG1164, this is still a PDF | 15:56 |
Dutch | #link https://www.xilinx.com/support/documentation/sw_manuals/xilinx2017_4/ug1164-sdaccel-platform-development.pdf | 15:56 |
NokMikeR | are there any xilinx parts with an open bitstream? (and tools to program them)? | 15:57 |
Dutch | Still no nova support, our openstack cluster is using pass-through, which as you can image is a serious limitation | 15:57 |
Dutch | The drivers could be open, but bitstream generation is IP so unlikely we will ever see that open | 15:59 |
Dutch | NokMikeR, what in particular do you want open? | 16:00 |
NokMikeR | any way to program the device via e.g an arm cpu so the synthesis and routing is done on the board - that way the entry level for the programmer is lower. | 16:00 |
Dutch | For drivers and runtime, absolutely. For synthesis and P&R, I am less sure. | 16:03 |
Dutch | I am also not sure what "done on the board" means | 16:04 |
NokMikeR | pcie board with fpga + additional cpu to generate the bitstream. you send that vhdl/verilog, "other code" and it all magically works :) | 16:04 |
NokMikeR | self contained basically. no reliance on vivado etc. | 16:05 |
NokMikeR | anyhow we are offtopic here, so Howard please carry on :) | 16:06 |
zhipeng | no this is actually an interesting topic :P | 16:06 |
Dutch | It is nice to think about but yeah, I would say that is a ways off | 16:06 |
zhipeng | NokMikeR I know there are open source bitstream FPGA cards, but not sure if it is production level | 16:07 |
Dutch | It might be possible to run Vivado on the add-in-card but as far as Vivado being open that seems very unlikely | 16:07 |
Dutch | I would push the currently open project to do something like this | 16:08 |
zhipeng | Dutch so coming back to SDAccel, the software defined here refering to ? | 16:08 |
Dutch | SDAccell was initially built to support OpenCL memory model, and there is also support for OpenCL kernel code | 16:09 |
Dutch | but it has advanced some to support iteration of HLS C and RTL IP | 16:09 |
zhipeng | got it | 16:10 |
Dutch | It was designed to be flexible so that it would support future models, MPI for example | 16:10 |
zhipeng | folks thx very much for the gr8t discussion today, i'm afraid I have to close the meeting now | 16:11 |
zhipeng | please go help reviewing the patch or drafting one, we have two weeks left before Queens code freeze | 16:11 |
zhipeng | so let's squeeze in as many stuff as we could :) | 16:12 |
NokMikeR | was the devstack bug report about the permissions for cyborg solved? | 16:12 |
zhipeng | yes NokMikeR | 16:13 |
NokMikeR | Great :) | 16:13 |
zhipeng | okey meeting adjourned | 16:14 |
zhipeng | #endmeeting | 16:14 |
*** openstack changes topic to "#openstack-cyborg" | 16:14 | |
openstack | Meeting ended Wed Jan 17 16:14:13 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:14 |
openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2018/openstack_cyborg.2018-01-17-15.03.html | 16:14 |
openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg/2018/openstack_cyborg.2018-01-17-15.03.txt | 16:14 |
openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2018/openstack_cyborg.2018-01-17-15.03.log.html | 16:14 |
Vipparthy | Thank you folks | 16:14 |
NokMikeR | Thanks all, bye | 16:14 |
*** NokMikeR has quit IRC | 16:14 | |
Dutch | Thanks! | 16:15 |
*** Dutch has left #openstack-cyborg | 16:15 | |
*** zhipeng has quit IRC | 16:18 | |
*** Vipparthy has quit IRC | 16:19 | |
*** guhcampos has quit IRC | 16:26 | |
*** guhcampos has joined #openstack-cyborg | 16:40 | |
guhcampos | Hi folks, a quick question: I may need to implement our driver for cyborg. Is there a way to write it as a plugin? Like a separate python package that can be installed alongside cyborg? Or do I have to add it to cyborg's tree alongside the generic driver? | 17:51 |
crushil_ | guhcampos, Add it to Cyborg's tree alongside the generic driver | 17:54 |
guhcampos | crushil_, thanks. I needed this confirmation to pressure management to sign the contributor agreement | 17:55 |
*** guhcampos has quit IRC | 18:30 | |
*** openstackstatus has quit IRC | 18:40 | |
*** openstackstatus has joined #openstack-cyborg | 18:41 | |
*** ChanServ sets mode: +v openstackstatus | 18:41 | |
-openstackstatus- NOTICE: Zuul will be offline for a few minutes; existing changes will be re-enqueued; approvals during the downtime will need to be re-added. | 20:38 | |
*** jkilpatr has quit IRC | 22:54 | |
*** crushil has joined #openstack-cyborg | 23:10 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!