| *** pro[m] has joined #buildstream | 00:37 | |
| *** verdre[m]1 has joined #buildstream | 00:40 | |
| *** tristan has quit IRC | 04:18 | |
| *** tristan has joined #buildstream | 04:42 | |
| *** tristan has quit IRC | 07:35 | |
| *** pointswaves has joined #buildstream | 07:46 | |
| *** tristan has joined #buildstream | 07:54 | |
| *** bochecha has joined #buildstream | 08:02 | |
| *** raoul has joined #buildstream | 08:10 | |
| *** pointswaves has quit IRC | 08:22 | |
| jennis | juergbi, it looks like sorting by the depth hasn't introduced any slowdowns, I've posted some benchmarks here: https://gitlab.com/BuildStream/buildstream/merge_requests/1344#note_177638654 please let me know if you think the discussion is resolved | 08:38 |
|---|---|---|
| jennis | (this discussion: https://gitlab.com/BuildStream/buildstream/merge_requests/1344#note_177267201) | 08:38 |
| *** rdale has joined #buildstream | 08:55 | |
| *** jonathanmaw has joined #buildstream | 09:21 | |
| benschubert | WSalmon: are you around? | 09:39 |
| WSalmon | benschubert, yep | 09:42 |
| gitlab-br-bot | danielsilverstone-ct approved MR !1370 (bschubert/optimize-node_get->master: Optimize _yaml.node_get) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1370 | 09:44 |
| *** lachlan has joined #buildstream | 09:44 | |
| gitlab-br-bot | BenjaminSchubert opened MR !1371 (bschubert/cythonize-valid-char-names->master: _loader/loader: cythonize valid_chars_name) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1371 | 10:00 |
| *** lachlan has quit IRC | 10:24 | |
| *** lachlan has joined #buildstream | 10:30 | |
| jennis | has marge kicked it again? | 10:37 |
| phildawson_ | It does look that way. | 10:43 |
| jennis | 😤 | 10:43 |
| *** lachlan has quit IRC | 10:45 | |
| gitlab-br-bot | marge-bot123 merged MR !1370 (bschubert/optimize-node_get->master: Optimize _yaml.node_get) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1370 | 10:46 |
| Kinnison | triumph? | 10:47 |
| tpollard | xD | 10:47 |
| phildawson_ | \o/ | 10:47 |
| jennis | Kinnison, a poorly named emoji | 10:47 |
| *** lachlan has joined #buildstream | 10:47 | |
| jennis | I asked for marge to be restarted again, looks like that worked | 10:48 |
| Kinnison | so triumph :D | 10:48 |
| *** lachlan has quit IRC | 10:52 | |
| *** raoul_ has joined #buildstream | 11:01 | |
| *** raoul has quit IRC | 11:02 | |
| *** lachlan has joined #buildstream | 11:02 | |
| *** bochecha_ has joined #buildstream | 11:02 | |
| *** bochecha has quit IRC | 11:04 | |
| *** bochecha_ is now known as bochecha | 11:04 | |
| *** raoul_ has quit IRC | 11:14 | |
| *** lachlan has quit IRC | 11:25 | |
| *** lachlan has joined #buildstream | 11:30 | |
| *** bochecha has quit IRC | 11:39 | |
| *** raoul_ has joined #buildstream | 11:40 | |
| *** lachlan has quit IRC | 11:40 | |
| *** raoul_ has quit IRC | 11:53 | |
| *** raoul_ has joined #buildstream | 11:53 | |
| *** lachlan has joined #buildstream | 11:54 | |
| *** raoul_ has quit IRC | 11:55 | |
| *** raoul_ has joined #buildstream | 11:55 | |
| *** lachlan has quit IRC | 12:08 | |
| *** pointswaves has joined #buildstream | 12:14 | |
| *** bochecha has joined #buildstream | 12:36 | |
| *** tristan has quit IRC | 12:45 | |
| benschubert | we are not using ostree at all in the artifactcache correct? Any message mentioning that would be wrong, true? | 12:46 |
| tpollard | sounds wrong to me | 12:47 |
| *** phildawson has quit IRC | 12:48 | |
| benschubert | ok, I'll clear mentions of ostree there then, thanks! | 12:48 |
| tpollard | is this from tests? | 12:50 |
| benschubert | yes | 12:50 |
| *** bilelmoussaoui has joined #buildstream | 13:00 | |
| *** tristan has joined #buildstream | 13:04 | |
| *** lachlan has joined #buildstream | 13:15 | |
| *** jude has joined #buildstream | 13:16 | |
| *** jude has quit IRC | 13:17 | |
| lachlan | I've put logging into one of the benchmark tests and it seems to have significantly changed the result for the worse - is this a known issue? | 13:22 |
| lachlan | Further information can be found at https://gitlab.com/BuildStream/benchmarks/issues/52 | 13:22 |
| *** raoul_ is now known as raoul | 13:26 | |
| tpollard | it's adding extra work, so I wouldn't find it surprising that it's added an overhead | 13:44 |
| juergbi | lachlan: --log-file outputs log messages additionally to the log file. for non-interactive mode I'd rather redirect stdout/stderr and skip --log-file | 13:44 |
| lachlan | juergbi: Strangely it doesn't seem to affect other tests (to the same degree) logged in exactly the same way (e.g. the bstgen test we have which has a far higher logging loading than the default benchmark). Redirecting as you suggest might introduce a number of issues into benchmarking, but it is perhaps something that needs revisiting. | 13:50 |
| juergbi | the logging overhead might impact scheduling differently for different element graph shapes | 13:51 |
| lachlan | Possibly, it certainly doesn't seem to scale with the fundamental level of output. | 13:53 |
| lachlan | I will revert in the case of the default benchmarking test (to prevent further corruption) and review the approach to see if some other redirection approach will work. | 13:56 |
| lachlan | juergbi: Thanks | 13:56 |
| *** rdale has quit IRC | 14:00 | |
| *** rdale has joined #buildstream | 14:02 | |
| *** lachlan has quit IRC | 14:23 | |
| *** lachlan has joined #buildstream | 14:27 | |
| *** lachlan has quit IRC | 14:32 | |
| *** lachlan has joined #buildstream | 14:52 | |
| *** lachlan has quit IRC | 14:55 | |
| *** lachlan has joined #buildstream | 15:07 | |
| *** lachlan has quit IRC | 15:14 | |
| *** lachlan has joined #buildstream | 15:28 | |
| *** lachlan has quit IRC | 15:34 | |
| *** lachlan has joined #buildstream | 15:37 | |
| *** lachlan has quit IRC | 15:45 | |
| *** lachlan has joined #buildstream | 15:52 | |
| *** pointswaves has quit IRC | 16:26 | |
| *** bochecha has quit IRC | 16:34 | |
| gitlab-br-bot | jennis opened MR !1372 (jennis/load_sometimes_overworks->master: Don't overwork in load when we don't need too) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1372 | 16:37 |
| *** phil has joined #buildstream | 16:41 | |
| *** phildawson_ has quit IRC | 16:43 | |
| *** pointswaves has joined #buildstream | 16:47 | |
| jennis | !1344 has all discussions resolved now and there are some pretty substantial gains for it. One thing that's not perfect is that we're setting Element._depth (depth within the current session's dependency graph), obviously not an inherent property of Element, but we're unsure of how we could alternatively do this, see: https://gitlab.com/BuildStrea | 16:51 |
| jennis | m/buildstream/merge_requests/1344#note_177267201 for more information. If anyone feels strongly against this or has a better suggestion please let me know! | 16:51 |
| gitlab-br-bot | MR !1344: Push based pipeline https://gitlab.com/BuildStream/buildstream/merge_requests/1344 | 16:51 |
| jennis | (tristan I thought this might particularly interest you) | 16:52 |
| *** pointswaves_ has joined #buildstream | 17:06 | |
| *** pointswaves has quit IRC | 17:08 | |
| *** bilelmoussaoui has quit IRC | 17:08 | |
| *** bilelmoussaoui has joined #buildstream | 17:09 | |
| *** jonathanmaw has quit IRC | 17:11 | |
| *** phil has quit IRC | 17:17 | |
| *** pointswaves_ has quit IRC | 17:21 | |
| *** lachlan has quit IRC | 17:22 | |
| *** rdale has quit IRC | 17:35 | |
| *** raoul has quit IRC | 18:11 | |
| benschubert | Do we have a good way for linking to the BuildStream doc from a plugin's docs? | 18:14 |
| *** slaf_ has joined #buildstream | 18:26 | |
| *** slaf_ has joined #buildstream | 18:26 | |
| *** slaf_ has joined #buildstream | 18:26 | |
| *** slaf_ has joined #buildstream | 18:26 | |
| *** slaf_ has joined #buildstream | 18:27 | |
| *** slaf_ has joined #buildstream | 18:27 | |
| *** slaf_ has joined #buildstream | 18:27 | |
| *** slaf_ has joined #buildstream | 18:28 | |
| *** slaf has quit IRC | 18:28 | |
| *** slaf has joined #buildstream | 18:28 | |
| *** toscalix has joined #buildstream | 18:41 | |
| *** pointswaves_ has joined #buildstream | 18:50 | |
| *** slaf has quit IRC | 19:00 | |
| *** slaf has joined #buildstream | 19:01 | |
| *** slaf has joined #buildstream | 19:01 | |
| *** slaf has joined #buildstream | 19:01 | |
| *** slaf has joined #buildstream | 19:02 | |
| *** slaf has quit IRC | 19:05 | |
| *** bilelmoussaoui has quit IRC | 19:13 | |
| *** bilelmoussaoui has joined #buildstream | 19:36 | |
| *** bilelmoussaoui has quit IRC | 19:59 | |
| *** bilelmoussaoui has joined #buildstream | 20:01 | |
| *** bilelmoussaoui has quit IRC | 21:46 | |
| *** bilelmoussaoui has joined #buildstream | 21:46 | |
| *** Kinnison has quit IRC | 22:06 | |
| *** slaf has joined #buildstream | 22:26 | |
| *** slaf has joined #buildstream | 22:26 | |
| *** slaf has joined #buildstream | 22:27 | |
| *** slaf has joined #buildstream | 22:27 | |
| *** slaf has joined #buildstream | 22:27 | |
| *** slaf has joined #buildstream | 22:27 | |
| *** slaf has joined #buildstream | 22:28 | |
| *** slaf has joined #buildstream | 22:28 | |
| *** pointswaves_ has quit IRC | 22:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!