| *** cs-shadow has quit IRC | 02:06 | |
| *** tristan has quit IRC | 05:57 | |
| *** tristan has joined #buildstream | 06:27 | |
| *** tristan has quit IRC | 07:12 | |
| *** tristan has joined #buildstream | 07:17 | |
| *** bochecha has joined #buildstream | 07:48 | |
| juergbi | benschubert: I have buildbox MRs to fix the excessive memory usage on upload. Small changes, so hopefully I can get those in today and then tag/update buildbox for BuildStream | 08:04 |
|---|---|---|
| juergbi | or at least one instance of excessive memory usage. there might be similar issues in other parts | 08:05 |
| benschubert | juergbi: I'm happy to test ;) | 08:18 |
| benschubert | turns out building the whole llvm suite with -flto=full is a good way of stress testing memory usage on that ;) | 08:18 |
| Kinnison | benschubert: !1605 could do with some attention from you if you have time -- I think, reading the commits, the requisite stuff is merged into casd | 08:29 |
| gitlab-br-bot | MR !1605: WIP: Report when Buildbox-casd is not alive at the end of a run https://gitlab.com/BuildStream/buildstream/merge_requests/1605 | 08:29 |
| benschubert | Kinnison: we are still waiting on a last MR from juerb before updating all our testing infra and the docs | 08:29 |
| benschubert | (cf just before :) ) | 08:30 |
| Kinnison | aah right | 08:30 |
| Kinnison | ok | 08:30 |
| benschubert | but it's still in my head :) not forgotten | 08:30 |
| Kinnison | I should get tea, it's 2+ hours into the working day and I've still not had any | 08:30 |
| *** tiagogomes has joined #buildstream | 08:37 | |
| *** tristan has quit IRC | 09:01 | |
| *** tpollard has joined #buildstream | 09:02 | |
| *** jonathanmaw has joined #buildstream | 09:07 | |
| *** tristan has joined #buildstream | 09:12 | |
| *** traveltissues has joined #buildstream | 09:14 | |
| Kinnison | traveltissues: The performance change noted in the benchmarking results on !1563 is very confusing to me. I don't see how your changes could have negatively affected performance on a benchmark which doesn't use workspaces | 09:41 |
| gitlab-br-bot | MR !1563: workspaces via sourcecache 1 https://gitlab.com/BuildStream/buildstream/merge_requests/1563 | 09:41 |
| *** phil has joined #buildstream | 09:46 | |
| *** phildawson_ has quit IRC | 09:47 | |
| coldtom | benschubert: if you have time, could you take a look at !1607 please? | 09:50 |
| gitlab-br-bot | MR !1607: WIP: Run standardised source tests from external plugins https://gitlab.com/BuildStream/buildstream/merge_requests/1607 | 09:50 |
| *** phil has quit IRC | 09:59 | |
| *** phil has joined #buildstream | 09:59 | |
| benschubert | coldtom: sure I'll look start of this afternoon. Is this based on my ML thread? | 10:06 |
| coldtom | benschubert: thanks, it is indeed based on your ML thread | 10:06 |
| benschubert | thanks a lot for this work :) | 10:07 |
| gitlab-br-bot | traveltissues opened issue #1140 (Follow-up from "workspaces via sourcecache 1" - workspace.init_workspace should raise a SourceError) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1140 | 10:40 |
| *** phoenix has joined #buildstream | 10:41 | |
| traveltissues | Kinnison, i have some cleanup to do and i'll rebenchmark, i had better results on py36 | 10:58 |
| traveltissues | but i'll be looking into that | 10:58 |
| coldtom | just wanted to say, `bst artifact log` is a wonderful addition and thank you for adding it | 11:04 |
| tpollard | it is good! | 11:10 |
| benschubert | ^^ completely agree :D | 11:26 |
| Kinnison | traveltissues: Mmm, I just find it odd that it's not almost exactly the same | 11:39 |
| traveltissues | i have a theory Kinnison | 11:40 |
| *** cs-shadow has joined #buildstream | 13:07 | |
| *** tristan has quit IRC | 13:10 | |
| coldtom | would it be possible to get review on the several plugin tests I've added to bst-plugins-experimental please? | 13:30 |
| coldtom | on a related note, I think we need to rethink the base of the test project - the alpine image keeps needing additional dependencies adding, and for the qmake plugin I think I'd need to add gcc + glibc | 13:32 |
| coldtom | would there be much opposition to maintaining a bst project to act as the base image for the plugin tests? | 13:32 |
| tpollard | We've gone down that road once | 13:36 |
| tpollard | https://gitlab.com/BuildStream/buildstream/merge_requests/1300 | 13:38 |
| benschubert | meh, freedesktop and a specialized docker image is quite different. One would be lightweight and easy to update, the other one would require caching/rebuilding a good part of a stack we don't really care about. I prefer the path of having a tailored image as a base | 13:47 |
| coldtom | a tailored image would *probably* be necessary anyway | 13:50 |
| coldtom | i was thinking of using the alpine image as a base and chopping and changing things | 13:50 |
| coldtom | but we'd still need to build more stuff on top of freedesktop-sdk anyway, especially if more weird and wonderful plugins appeared | 13:50 |
| coldtom | also we don't *want* frequent updates in a test image | 13:53 |
| benschubert | coldtom: about your prs on bst-plugins-experimental any reason they are not working with buildbox? | 14:51 |
| coldtom | ah oops, i just copied the cmake ones and missed that mark, I'm not sure if they work with buildbox or not | 14:51 |
| benschubert | would you mind checking? :) | 14:52 |
| coldtom | i'll check whether the cmake one is still broken while i'm at it :) | 14:53 |
| benschubert | awesome | 15:02 |
| coldtom | i think this was only a problem in buildstream CI where the plugins were tested in the `test-buildbox` tests, judging by the comment remarking things only failed in CI | 15:05 |
| *** phil has quit IRC | 15:09 | |
| *** phildawson has joined #buildstream | 15:09 | |
| *** phoenix has quit IRC | 15:44 | |
| gitlab-br-bot | juergbi opened (was WIP) MR !1601 (juerg/casd->master: Do not directly communicate with CAS server) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1601 | 15:49 |
| juergbi | can I please get an approval for https://gitlab.com/BuildStream/website/merge_requests/127 ? | 16:14 |
| juergbi | also, will marge !1601 soon unless there are any last minute objections | 16:15 |
| gitlab-br-bot | BenjaminSchubert approved MR !1601 (juerg/casd->master: Do not directly communicate with CAS server) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1601 | 16:33 |
| gitlab-br-bot | BenjaminSchubert opened (was WIP) MR !1605 (bschubert/ensure-buildbox-alive->master: Report when Buildbox-casd is not alive at the end of a run) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1605 | 16:33 |
| benschubert | coldtom: want me to merge and rebase all your PRs to bst-plugins-experimental? | 16:40 |
| gitlab-br-bot | marge-bot123 merged MR !1601 (juerg/casd->master: Do not directly communicate with CAS server) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1601 | 16:44 |
| *** lachlan has joined #buildstream | 16:53 | |
| *** lachlan has quit IRC | 17:04 | |
| *** traveltissues has quit IRC | 17:27 | |
| *** phoenix has joined #buildstream | 18:27 | |
| *** hergertme has quit IRC | 19:07 | |
| *** hergertme has joined #buildstream | 19:11 | |
| *** hergertme has joined #buildstream | 19:12 | |
| *** bochecha has quit IRC | 22:41 | |
| *** phoenix has quit IRC | 23:22 | |
| *** cs-shadow has quit IRC | 23:45 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!