| *** bochecha has quit IRC | 00:55 | |
| *** alatiera_ has joined #buildstream | 03:42 | |
| *** tristan has joined #buildstream | 04:07 | |
| *** catonano has joined #buildstream | 04:38 | |
| *** ahmed89 has quit IRC | 04:54 | |
| *** tristan has quit IRC | 04:59 | |
| *** tristan has joined #buildstream | 05:27 | |
| *** toscalix has joined #buildstream | 07:41 | |
| *** catonano has quit IRC | 07:41 | |
| *** finn has joined #buildstream | 07:52 | |
| *** phildawson has joined #buildstream | 08:06 | |
| *** ahmed89 has joined #buildstream | 08:08 | |
| *** bochecha has joined #buildstream | 08:19 | |
| gitlab-br-bot | buildstream: merge request (juerg/cpucount->master: _platform: Fix get_cpu_count() with cap=None) #840 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/840 | 08:21 |
|---|---|---|
| gitlab-br-bot | buildstream: merge request (juerg/sandbox->master: sandbox/_sandboxremote.py: Use helpers to set up cwd and env) #841 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/841 | 08:37 |
| *** ChanServ sets mode: +o tristan | 08:45 | |
| tristan | jmac, I don't like what you did with element_enums.py in 3341948a6, I suppose this might have been to work around some circular import problem | 08:47 |
| tristan | jmac, I wonder if you added the new public module to the documentation index in the reference manual, even | 08:47 |
| tristan | jmac, rather than that; we might have a 'types.py' live at the top level, and then we could put Consistency in there too, and any basic types | 08:48 |
| tristan | jmac, also note that you named it with an underscore, whereas every other multiword module name has none | 08:49 |
| gitlab-br-bot | buildstream: merge request (juerg/cpucount->master: _platform: Fix get_cpu_count() with cap=None) #840 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/840 | 08:53 |
| *** tpollard has joined #buildstream | 09:02 | |
| gitlab-br-bot | buildstream: issue #654 ("Use Platform.artifact_cache() instead of reconstructing CASCache from context") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/654 | 09:03 |
| jmac | It was indeed to work around a circular import | 09:05 |
| tristan | jmac, right; I think it was a bit of a short cut, we should make it `types.py` or something generic, and have it as a location for defining low level, possibly public types; add Consistency there as well - and update the documentation so that it has a chapter in the reference manual | 09:12 |
| tristan | This is a problem that keeps popping up, so it will be useful to have a location for it, rather than growing new limbs on demand :) | 09:13 |
| tristan | jmac, another point, is that private symbols normally come after public ones | 09:14 |
| tiagogomes | 5 min vs 50+min \o/ | 09:15 |
| jmac | I will note all these and make an issue to fix them | 09:15 |
| tristan | jmac, fwiw, the documentation ToC to add a basic / foundation times place... is in doc/source/core_framework.rst | 09:16 |
| tristan | s/times/types | 09:16 |
| tristan | we might later even consider a _privatetypes.py if they become numerous, it will allow us to write _KeyStrength as KeyStrength, which would be aesthetically nicer | 09:19 |
| * tristan heads out to dinner | 09:19 | |
| *** catonano has joined #buildstream | 09:23 | |
| *** tiagogomes has quit IRC | 09:24 | |
| *** tiagogomes has joined #buildstream | 09:26 | |
| *** lachlan has joined #buildstream | 09:28 | |
| *** dtf has joined #buildstream | 09:46 | |
| *** ahmed89 has quit IRC | 09:59 | |
| *** ahmed89 has joined #buildstream | 09:59 | |
| jmac | tristan: Where are public modules listed in the documentation? As a public module it gets listed in the automatic documentation anyway, although the underscore does seem to confuse the autodocs somewhat | 10:04 |
| *** ahmed89 has quit IRC | 10:06 | |
| *** catonano has quit IRC | 10:14 | |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 10:25 |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 10:36 |
| gitlab-br-bot | buildstream: merge request (tiagogomes/ci-minor-tweaks->master: Minor CI tweaks) #842 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/842 | 10:39 |
| *** catonano has joined #buildstream | 10:53 | |
| *** mohan43u has quit IRC | 10:59 | |
| gitlab-br-bot | buildstream: merge request (tiagogomes/ci-minor-tweaks->master: Minor CI tweaks) #842 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/842 | 10:59 |
| juergbi | tiagogomes: I assume you've already noticed but the overnight tests keep failing due to insufficient disk space | 11:10 |
| juergbi | as long as this is not resolved, it probably doesn't make much sense to actually run them | 11:11 |
| tiagogomes | yup, this already being raised and valentind will take charge of this | 11:12 |
| juergbi | ok | 11:12 |
| tiagogomes | There's no harm in keeping running them meanwhile; and we could also spot if they failed before running out of disk space | 11:12 |
| juergbi | well, due to this the pipeline badge shows up as failure | 11:14 |
| juergbi | but that might be another issue, we probably want to consistently use the regular CI pipeline for that badge, not the overnight one | 11:14 |
| tiagogomes | Hmm | 11:15 |
| tiagogomes | I think you are right, I am going to disable them now | 11:16 |
| gitlab-br-bot | buildstream: issue #514 ("BuildStream CI is not configured to perform any system level testing") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/514 | 11:17 |
| tiagogomes | This bunch of messages looks very unpolished: | 11:27 |
| tiagogomes | [--:--:--][][] START cache_size.292.log | 11:27 |
| tiagogomes | [00:00:00][][] SUCCESS cache_size.292.log | 11:27 |
| tiagogomes | [--:--:--][][] START cache_size.295.log | 11:27 |
| tiagogomes | [00:00:00][][] SUCCESS cache_size.295.log | 11:27 |
| bochecha | tiagogomes: the accompanying huge amount of cache_size.*.log files created in ~/.cache/buildstream/ as well | 11:29 |
| bochecha | I have 1858 such files right now, 7.4MB | 11:30 |
| tiagogomes | This would indirectly fix this https://gitlab.com/BuildStream/buildstream/merge_requests/671, but it needs to be rebased now | 11:30 |
| tiagogomes | eek] | 11:30 |
| bochecha | they all contain ~2 lines like the ones you pasted | 11:30 |
| *** mohan43u has joined #buildstream | 11:41 | |
| *** lachlan has quit IRC | 12:07 | |
| gitlab-br-bot | buildstream: issue #687 ("Error when staging ostree sources when /tmp is mounted as a tmpfs on the host system - "Invalid cross-device link"") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/687 | 12:19 |
| * jmac wonders how often /tmp isn't a tmpfs | 12:19 | |
| Kinnison | Quite often | 12:20 |
| skullman | isn't by default on debian stretch | 12:20 |
| jmac | Huh, so it isn't | 12:21 |
| *** ahmed89 has joined #buildstream | 12:22 | |
| bochecha | I think it's only by default on Fedora and Ubuntu; maybe Arch as well? | 12:22 |
| *** lachlan has joined #buildstream | 13:04 | |
| gitlab-br-bot | buildstream: issue #688 ("Better handle artifact server unimplemented methods") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/688 | 13:21 |
| *** finn has quit IRC | 13:30 | |
| *** tiagogomes has quit IRC | 13:41 | |
| *** tiagogomes has joined #buildstream | 13:42 | |
| *** abderrahim has quit IRC | 13:42 | |
| *** abderrahim has joined #buildstream | 13:43 | |
| juergbi | WSalmon: you can specify the .pro file on the qmake command line. and with that out of source build seems to just work | 13:56 |
| juergbi | I tested with `qmake /usr/lib/qt5/examples/corelib/corelib.pro` in a temp directory | 13:57 |
| *** finn has joined #buildstream | 13:59 | |
| toscalix | added a couple more topics to the Gathering agenda | 14:11 |
| gitlab-br-bot | buildstream: merge request (lachlan/pickle-yaml-test-list-composite->master: Add yaml cache testing to yaml tests) #836 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/836 | 14:14 |
| gitlab-br-bot | buildstream: issue #668 ("Remote Execution client does not check for build action result exit code") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/668 | 14:21 |
| tpollard | is it possible to run the tests with --last-failed? | 14:23 |
| tpollard | that answer to that is yes | 14:25 |
| *** toscalix has quit IRC | 14:43 | |
| gitlab-br-bot | buildstream: merge request (juerg/cas-batch->master: CAS: Implement BatchUpdateBlobs support) #839 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/839 | 14:46 |
| *** ahmed89 has quit IRC | 14:49 | |
| *** ahmed89 has joined #buildstream | 14:49 | |
| gitlab-br-bot | buildstream: merge request (juerg/cas-batch->master: CAS: Implement BatchUpdateBlobs support) #839 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/839 | 14:58 |
| *** lachlan has quit IRC | 14:59 | |
| gitlab-br-bot | buildstream: merge request (jmac/vdir_import_unreadable_files->master: WIP: Import unreadable files) #837 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/837 | 15:00 |
| WSalmon | juergbi, i was talking to daniel about this, as we dont know the .pro file already although we do know were it is we were considering ether cd conf-root, then qmake -o"comand-dir" or qmake "conf-root/*.pro" but I have a good example to test with, which did you use? | 15:01 |
| WSalmon | *I dont have a good example to test with | 15:02 |
| juergbi | WSalmon: *.pro wouldn't always be equivalent to the default project finding logic in qmake. however, you can simply pass the path to the source directory to qmake | 15:05 |
| juergbi | qmake %{conf-root} | 15:05 |
| juergbi | the example project I'm using is part of upstream Qt5 (qtbase) | 15:06 |
| WSalmon | why do the doc's not make this obvious rant... many thanks | 15:07 |
| *** lachlan has joined #buildstream | 15:07 | |
| juergbi | no idea, looking at the source was simpler | 15:08 |
| WSalmon | Receiving objects: 42% (219836/520800), 215.47 MiB | 561.00 KiB/s this makes me a little sad | 15:18 |
| WSalmon | qtbase is big | 15:19 |
| *** toscalix has joined #buildstream | 15:23 | |
| gitlab-br-bot | buildstream: merge request (jmac/vdir_import_unreadable_files->master: WIP: Import unreadable files) #837 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/837 | 15:29 |
| gitlab-br-bot | buildstream: issue #677 ("CAS Client: Use BatchUpdateBlobs") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/677 | 15:32 |
| gitlab-br-bot | buildstream: issue #676 ("CAS Server: Implement BatchUpdateBlobs") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/676 | 15:32 |
| gitlab-br-bot | buildstream: merge request (juerg/cas-batch->master: CAS: Implement BatchUpdateBlobs support) #839 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/839 | 15:32 |
| gitlab-br-bot | buildstream: issue #689 ("Build should fail if `type:` dependecy build is not a supported type") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/689 | 15:36 |
| gitlab-br-bot | buildstream: merge request (juerg/cas-batch-1.2->bst-1.2: _artifactcache/cascache.py: Use BatchReadBlobs) #838 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/838 | 15:42 |
| gitlab-br-bot | buildstream: merge request (jmac/vdir_import_unreadable_files->master: WIP: Import unreadable files) #837 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/837 | 15:45 |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 16:04 |
| *** lachlan has quit IRC | 16:08 | |
| jennis | This cache size logging is super spammy | 16:12 |
| tiagogomes | yeah… and inefficient | 16:17 |
| jmac | #684 & #429 both raise the problem of files without read permission and I'm not sure how to solve it at the moment | 16:24 |
| jmac | If you alter the permissions on the source file then you have a potential race condition with two elements trying to import from the same directory | 16:25 |
| *** raoul has joined #buildstream | 16:27 | |
| *** toscalix has quit IRC | 16:27 | |
| *** toscalix has joined #buildstream | 16:28 | |
| *** lachlan has joined #buildstream | 16:34 | |
| *** lachlan has quit IRC | 16:44 | |
| tpollard | am I right in thinking for instance 'tests/testutils/runcli.py' is only linted https://gitlab.com/BuildStream/buildstream/-/jobs/103217282? I asked because one of its included tests fails when run locally for me | 16:57 |
| *** lachlan has joined #buildstream | 16:57 | |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 16:59 |
| *** toscalix has quit IRC | 18:18 | |
| *** catonano has quit IRC | 18:28 | |
| *** lachlan has quit IRC | 19:14 | |
| *** lachlan has joined #buildstream | 19:21 | |
| *** lachlan has quit IRC | 19:25 | |
| *** catonano has joined #buildstream | 19:58 | |
| *** dtf has quit IRC | 20:03 | |
| *** dtf has joined #buildstream | 20:03 | |
| *** xjuan has joined #buildstream | 20:31 | |
| *** catonano has quit IRC | 21:11 | |
| *** alatiera_ has quit IRC | 21:19 | |
| *** tristan has quit IRC | 21:29 | |
| *** cs-shadow has quit IRC | 21:29 | |
| *** bochecha has quit IRC | 21:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!