| *** fiddletwix has quit IRC | 02:35 | |
| *** fiddletwix has joined #kata-dev | 02:40 | |
| *** EricRen has quit IRC | 03:04 | |
| *** igordc has quit IRC | 06:30 | |
| *** changcheng has joined #kata-dev | 07:39 | |
| *** tmhoang has joined #kata-dev | 07:41 | |
| *** sgarzare has joined #kata-dev | 07:50 | |
| *** jodh has joined #kata-dev | 08:24 | |
| *** gwhaley has joined #kata-dev | 08:53 | |
| *** sameo has joined #kata-dev | 08:56 | |
| *** davidgiluk has joined #kata-dev | 09:02 | |
| *** davidgiluk has quit IRC | 09:03 | |
| *** tmhoang_ has joined #kata-dev | 09:03 | |
| *** tmhoang has quit IRC | 09:04 | |
| *** davidgiluk has joined #kata-dev | 09:04 | |
| *** changcheng has quit IRC | 09:24 | |
| *** dklyle has quit IRC | 09:25 | |
| *** dklyle has joined #kata-dev | 09:27 | |
| *** sameo has quit IRC | 10:49 | |
| kata-irc-bot | <graham.whaley> @james.o.hunt /me sees golang 1.12 now released :slightly_smiling_face: | 11:25 |
|---|---|---|
| *** manny_ has joined #kata-dev | 12:30 | |
| *** manny has quit IRC | 12:30 | |
| *** fiddletwix has quit IRC | 12:30 | |
| *** jdewald has quit IRC | 12:30 | |
| *** sgarzare has quit IRC | 12:33 | |
| *** tmhoang_ is now known as tmhoang | 12:42 | |
| *** sgarzare has joined #kata-dev | 12:49 | |
| *** devimc has joined #kata-dev | 13:18 | |
| kata-irc-bot | <jose.carlos.venegas.m> @eric.ernst https://github.com/kata-containers/runtime/pull/1282 | 13:48 |
| kata-irc-bot | <jose.carlos.venegas.m> https://github.com/kata-containers/runtime/pull/1283 | 13:48 |
| kata-irc-bot | <jose.carlos.venegas.m> I think those PRs are good to merge the errors the two errors seems that are not releated with the PR | 13:50 |
| kata-irc-bot | <jose.carlos.venegas.m> and would be great have them on the stable releases today | 13:50 |
| *** sameo has joined #kata-dev | 13:53 | |
| kata-irc-bot | Action: graham.whaley sees kernel 4.19.25 is out already! | 13:56 |
| *** gwhaley has quit IRC | 14:00 | |
| *** fuentess has joined #kata-dev | 14:12 | |
| *** fuentess has quit IRC | 14:21 | |
| *** fuentess has joined #kata-dev | 14:22 | |
| *** fuentess has quit IRC | 14:37 | |
| *** fuentess has joined #kata-dev | 14:38 | |
| kata-irc-bot | <jose.carlos.venegas.m> @james.o.hunt ping | 15:00 |
| kata-irc-bot | <james.o.hunt> Hi | 15:16 |
| kata-irc-bot | <jose.carlos.venegas.m> Yesterday we were talking about if https://github.com/pholleran/toc-me | 15:17 |
| kata-irc-bot | <jose.carlos.venegas.m> could help us in our docs | 15:17 |
| kata-irc-bot | <jose.carlos.venegas.m> to generate ToC | 15:17 |
| kata-irc-bot | <jose.carlos.venegas.m> or you prefer to modify them manually ? | 15:18 |
| kata-irc-bot | <james.o.hunt> Looks interesting, but I'm not sure if the time taken to setup the service is going to outweigh the benefit honestly. `.rst` files get tocs "for free" but we've already had 2 rounds of Holy Wars on doc formats :slightly_smiling_face: I think Graham has played around with probot so might have some input on that. | 15:21 |
| kata-irc-bot | <jose.carlos.venegas.m> so probably we need a tool md -> rst :P | 15:23 |
| kata-irc-bot | <jose.carlos.venegas.m> yeah I remember the format wars | 15:23 |
| kata-irc-bot | <sebastien.boeuf> davidgiluk around? | 15:32 |
| davidgiluk | sebastien.boeuf: Yep | 15:33 |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: just saw your answer :slightly_smiling_face: | 15:34 |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: I can change the virtio-mmio implementation in Firecracker to include those extra registers, but would that help getting those accepted? | 15:36 |
| kata-irc-bot | <sebastien.boeuf> I mean, if it makes sense for the spec, things should just be accepted anyway. | 15:36 |
| davidgiluk | sebastien.boeuf: I think so; there's 'makes sense' and 'makes sense and actually works' | 15:37 |
| kata-irc-bot | <sebastien.boeuf> ok got it | 15:37 |
| kata-irc-bot | <sebastien.boeuf> with the change you're proposing to the virtio-mmio spec, can you share multiple memory regions? | 15:38 |
| davidgiluk | Yes, the idea is they have an 'id'; you write the ID register and then read the length register to see if that ID exists | 15:39 |
| kata-irc-bot | <sebastien.boeuf> ok so that's good because it's generic enough to be reused by any virtio device | 15:39 |
| davidgiluk | so we have an ID for the DAX area, another one for the metadata stuff we're trying etc | 15:39 |
| kata-irc-bot | <sebastien.boeuf> sounds good | 15:39 |
| *** sgarzare_ has joined #kata-dev | 15:42 | |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: do you have a PDF version of your spec changes? | 15:42 |
| kata-irc-bot | <sebastien.boeuf> I tried to render latex, but it didn't go well... | 15:42 |
| davidgiluk | haha yes, I'll send it you; it took a bit of a fight for PDF as well | 15:43 |
| kata-irc-bot | <sebastien.boeuf> thank you very much! | 15:43 |
| * davidgiluk thinks he has 1600 tex packages installed :-( | 15:44 | |
| *** sgarzare has quit IRC | 15:44 | |
| *** gwhaley has joined #kata-dev | 15:45 | |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: ok then that's definitely why I need the PDF version from you haha! | 15:45 |
| * davidgiluk throws the pdf in sebastien's direction | 15:46 | |
| kata-irc-bot | <sebastien.boeuf> received! thanks! | 15:50 |
| kata-irc-bot | Action: graham.whaley made a docker image to do his last pdf/latex/Rmarkdown build :-) | 15:53 |
| *** sameo has quit IRC | 15:54 | |
| *** jodh has quit IRC | 15:55 | |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: one nit, but for consistency, I think `SHMSel` would be more appropriate than `SHMId`. It would match `QueueSel` which has the exact same behavior | 15:59 |
| davidgiluk | yeh that's reasonable | 16:00 |
| kata-irc-bot | <krsna1729> cc @graham.whaley | 16:20 |
| *** tmhoang has quit IRC | 16:22 | |
| kata-irc-bot | <graham.whaley> Heh, ok, who started the format wars again... ;) We settled on markdown as it is just easier - easier for devs to contribute new docs. We all appreciate rst is better formatted and can do more fancy things, but it is a little more complex to write. We can raise the question of converting all md to rst for all repos and docs if you want - it would have to be an arch committee discussion imo. | 16:26 |
| kata-irc-bot | <graham.whaley> as for probot - if that bot is a *built in base probot* app, then we can use it without having to find somewhere to host it I believe - but, again, we are wary in kata of using 3rd party items that we have little control over - so rather than use the probot hosting (we would want to check into who really hosts that for instance), we could wait until github actions are out of beta and host the probot in an action - we are more | 16:27 |
| kata-irc-bot | comfortable with that I think. | 16:27 |
| davidgiluk | I'm hoping we can get this virtio change in and then I can ignore TeX for another 25 years | 16:37 |
| kata-irc-bot | <krsna1729> will leave this here https://probot.github.io/docs/serverless-deployment/ | 16:37 |
| kata-irc-bot | <krsna1729> github actions definitely seems more elegant :slightly_smiling_face: | 16:39 |
| *** sgarzare_ has quit IRC | 17:09 | |
| *** tmhoang has joined #kata-dev | 17:16 | |
| kata-irc-bot | <graham.whaley> @salvador.fuentes @sebastien.boeuf - I tweaked the kata metrics bounds to try and account for the new 4.19 kernel, but it looks like at least the memory footprint is not there yet. I'll check in the morning to see how it is looking, and re-tweak if I need. but, in the meantime, we might get some metrics CI fails. | 17:17 |
| kata-irc-bot | <salvador.fuentes> ok, thanks for the notice @graham.whaley | 17:21 |
| kata-irc-bot | <sebastien.boeuf> thx @graham.whaley! | 17:36 |
| kata-irc-bot | <jose.carlos.venegas.m> @james.o.hunt @graham.whaley @salvador.fuentes https://github.com/kata-containers/tests/pull/1215 | 17:47 |
| kata-irc-bot | <jose.carlos.venegas.m> take a look, @james.o.hunt metion it has some common things with the kernel PR, lets merge (if all passes) first, then refactor in the other PR | 17:49 |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: I realized the register `SHMBaseLow` and `SHMBaseHigh` from your proposal are supposed to be written by the driver? Can you explain why? My understanding was that the VMM would create a memory region for DAX and then share the `addr` and `size` with the driver. So I was expecting those registers to be tagged as `R` instead of `W` | 19:05 |
| *** igordc has joined #kata-dev | 19:08 | |
| kata-irc-bot | <sebastien.boeuf> davidgiluk: also, IIUC the virtqueues are the way from the driver to specify where to map things, not the virtio config | 19:14 |
| *** fiddletwix has joined #kata-dev | 19:19 | |
| *** fuentes2 has joined #kata-dev | 19:20 | |
| *** fuentess has quit IRC | 19:21 | |
| *** eguan has quit IRC | 20:12 | |
| *** eguan has joined #kata-dev | 20:12 | |
| *** davidgiluk has quit IRC | 20:16 | |
| *** tmhoang has quit IRC | 20:24 | |
| *** tmhoang has joined #kata-dev | 20:24 | |
| *** gwhaley has quit IRC | 20:30 | |
| *** devimc_ has joined #kata-dev | 21:21 | |
| *** gmmaharaj has joined #kata-dev | 21:24 | |
| *** devimc has quit IRC | 21:24 | |
| *** gmmaha has quit IRC | 21:24 | |
| *** gmmaharaj is now known as gmmaha | 21:24 | |
| *** devimc_ has quit IRC | 22:27 | |
| *** fuentes2 has quit IRC | 22:34 | |
| *** tmhoang has quit IRC | 23:40 | |
| *** sameo has joined #kata-dev | 23:43 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!