| *** swick has quit IRC | 00:43 | |
| *** nimish2711 has joined #buildstream | 01:27 | |
| *** nimish2711 has quit IRC | 02:55 | |
| *** nimish2711 has joined #buildstream | 02:58 | |
| *** nimish2711 has quit IRC | 03:08 | |
| *** nimish2711 has joined #buildstream | 03:08 | |
| *** nimish2711 has quit IRC | 03:19 | |
| *** swick has joined #buildstream | 03:51 | |
| *** mohan43u has joined #buildstream | 04:07 | |
| *** cs-shadow has quit IRC | 04:31 | |
| *** cs-shadow has joined #buildstream | 04:31 | |
| *** benschubert has quit IRC | 04:31 | |
| *** benschubert has joined #buildstream | 04:31 | |
| *** csoriano has quit IRC | 04:31 | |
| *** persia has quit IRC | 04:32 | |
| *** persia has joined #buildstream | 04:32 | |
| *** csoriano has joined #buildstream | 04:39 | |
| *** csoriano has quit IRC | 04:40 | |
| *** csoriano has joined #buildstream | 04:40 | |
| *** tristan has quit IRC | 05:06 | |
| *** tristan has joined #buildstream | 05:28 | |
| *** tristan has quit IRC | 07:22 | |
| *** toscalix has joined #buildstream | 08:16 | |
| *** toscalix has joined #buildstream | 08:18 | |
| *** tristan has joined #buildstream | 08:55 | |
| *** rdale has joined #buildstream | 09:21 | |
| *** raoul has joined #buildstream | 09:43 | |
| *** tpollard has joined #buildstream | 09:44 | |
| gitlab-br-bot | BenjaminSchubert opened MR !1254 (bschubert/optimize-dependencies->master: Rework Element.dependencies to be more efficient) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1254 | 09:49 |
|---|---|---|
| benschubert | tristan: ^ you might be interested | 09:49 |
| *** toscalix has quit IRC | 09:59 | |
| *** mohan43u has quit IRC | 10:00 | |
| *** toscalix has joined #buildstream | 10:03 | |
| *** mohan43u has joined #buildstream | 10:04 | |
| *** mohan43u has quit IRC | 10:08 | |
| *** jonathanmaw has joined #buildstream | 10:09 | |
| *** mohan43u has joined #buildstream | 10:14 | |
| *** mohan43u has quit IRC | 10:24 | |
| *** mohan43u has joined #buildstream | 10:27 | |
| gitlab-br-bot | BenjaminSchubert opened MR !1255 (bschubert/profiler-as-cm->master: Cleanup profiler and make is as a context manager) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1255 | 10:36 |
| *** lachlan has joined #buildstream | 10:40 | |
| *** toscalix has quit IRC | 10:48 | |
| *** toscalix has joined #buildstream | 10:53 | |
| *** csoriano has quit IRC | 10:59 | |
| *** mohan43u has quit IRC | 10:59 | |
| *** csoriano has joined #buildstream | 11:00 | |
| *** csoriano has joined #buildstream | 11:01 | |
| *** mohan43u has joined #buildstream | 11:03 | |
| *** nimish2711 has joined #buildstream | 11:03 | |
| gitlab-br-bot | tristanvb opened MR !1256 (tristan/backport-update-state-changes-1.2->bst-1.2: Tristan/backport update state changes 1.2) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1256 | 11:08 |
| *** gokcennurlu_ has joined #buildstream | 11:11 | |
| *** mohan43u has quit IRC | 11:12 | |
| *** mohan43u has joined #buildstream | 11:16 | |
| benschubert | juergbi: there is mention in the profiler and https://docs.buildstream.build/CONTRIBUTING.html#profiling-the-artifact-cache-receiver of being able to profile the cas server. However the relevant code is not in the casserver.py. Is this a mistake or should the profile and the docs not exist anymore? | 11:17 |
| *** ChanServ sets mode: +o tristan | 11:17 | |
| tristan | juergbi, raoul, I know you're waiting for feedback on !1214; I think I did have preference of queue separation in the past, and I still think it makes better sense than baking pushes into the fetch queue | 11:17 |
| gitlab-br-bot | MR !1214: Remote source cache https://gitlab.com/BuildStream/buildstream/merge_requests/1214 | 11:17 |
| tristan | I might be overlooking some things though ? | 11:17 |
| juergbi | benschubert: this is a leftover from the pre-CAS artifact server | 11:18 |
| benschubert | juergbi: so I can safely remove correct? | 11:18 |
| juergbi | yes. it might make sense adding profiling support in the future but let's just remove the leftovers from now | 11:19 |
| juergbi | (especially considering we might move to casd) | 11:19 |
| benschubert | thanks | 11:19 |
| tristan | juergbi, I think separating the tasks is more inline with our architecture in general, and one benefit I can see right away is that we should not be blocking builds which can commence after a fetch completes because the fetch is spending time pushing to a remote | 11:19 |
| raoul | tristan, The remaining points were what to name the new source push queue in the user interface, and whether to have the queue after the fetch stage or after artifact push. | 11:20 |
| juergbi | tristan: yes, separation is also how it's implemented right now. and generally does make sense | 11:20 |
| juergbi | but to avoid blocking build, we need to reorder the queues. right now it's between fetch and build | 11:20 |
| tristan | raoul, I see, I thought juergbi linked me in specifically to discuss whether it should be separated | 11:20 |
| juergbi | it should probably be between build and [artifact] push, or after [artifact] push | 11:20 |
| juergbi | tristan: that is/was one possible alternative | 11:21 |
| tristan | I see | 11:21 |
| juergbi | just wanted to mention the possibilities | 11:21 |
| Kinnison | Our WIP MR for the YAML rework is now posted. jennis and I would appreciate tactful eyes on it - there's still plenty of TODOs for us to resolve, but it'll be interesting to see what others thing over-all | 11:21 |
| tristan | That seems to be a bit flawed | 11:21 |
| tristan | juergbi, Maybe in some future, queues could be changed to not be a flat list, and orthogonal queues could occur ? | 11:22 |
| juergbi | yes, that's a limitation with the current scheduler | 11:22 |
| juergbi | however, I think we don't want to block the branch on lifting this restriction | 11:22 |
| tristan | it's not a huge change to implement | 11:22 |
| tristan | no no | 11:22 |
| tristan | I only mean, it's less of a concern if we consider that change on the roadmap | 11:22 |
| juergbi | sure | 11:23 |
| benschubert | tristan: I'm looking in reworking how we handle the queues | 11:23 |
| tristan | yeah :) | 11:23 |
| benschubert | so that problem would go away | 11:23 |
| juergbi | the remaining blocking questions are mainly about the (UI) name of the new queue, to avoid confusion with artifact push | 11:23 |
| tlater[m] | We're seeing an interesting performance trend for master: https://buildstream.gitlab.io/buildstream-benchmarking/benchmarks/public/average_time_master_Build_of_Baserock_stage1-binutils_for_x86_64_20190322-023246.html | 11:24 |
| tlater[m] | We aren't quite sure if this is a measurement error or a real effect | 11:24 |
| tlater[m] | Does anyone know what features might be responsible for that trend? | 11:24 |
| *** alatiera has joined #buildstream | 11:25 | |
| Kinnison | We recently got a few performance improvements merged I thought | 11:25 |
| Kinnison | benschubert: ^^ ? | 11:25 |
| benschubert | tlater[m]: trend being buildstream becomes more performant? (Am I reading the graph correctly?) | 11:25 |
| tlater[m] | benschubert: Yup | 11:25 |
| benschubert | yep, that's improvment from our performance work | 11:25 |
| tlater[m] | We're just not sure why it's that big of an improvement all of the sudden | 11:25 |
| benschubert | glad to see it's noticeable | 11:26 |
| tlater[m] | Cool I'll just take your word for it. Thanks benschubert :) | 11:26 |
| tristan | tlater[m], I have a hard time believing that the optimizations improve performance that much for such a small set of elements | 11:26 |
| tristan | are those minutes ? | 11:26 |
| tlater[m] | tristan: I'm not sure | 11:26 |
| tlater[m] | I haven't looked much into how the graphs work yet | 11:26 |
| tristan | I mean, we're really talking about something like < 20 elements | 11:26 |
| tlater[m] | lachlan? | 11:26 |
| juergbi | the staging improvements might also have had an impact here, difficult to say without more details | 11:26 |
| benschubert | ^ I second this | 11:26 |
| Kinnison | the first drop might be sourcecache populating | 11:26 |
| lachlan | Units are seconds | 11:26 |
| Kinnison | and the second it in use? | 11:26 |
| tristan | staging improvements are more likely | 11:26 |
| tlater[m] | i don't think we have more details atm, unfortunately | 11:27 |
| juergbi | and if the fetch job is not part of the performance test, source cache should also have provided a speedup | 11:27 |
| juergbi | (in staging sources) | 11:27 |
| Kinnison | tlater[m]: If you want to review a WIP MR, https://gitlab.com/BuildStream/buildstream/merge_requests/1257 would enjoy some love | 11:27 |
| tlater[m] | I'd love to write some more specific tests so we can tell what causes these improvements |: | 11:28 |
| tlater[m] | Kinnison: I was about to have a look, actually :) | 11:28 |
| benschubert | tlater[m]: that would be nice | 11:28 |
| jennis | pretty sure those benchmarks mount in a prepopulated source cache | 11:28 |
| Kinnison | While I was expecting some improvement, I think tlater[m] will be pleased with the numbers in the MR description | 11:28 |
| tlater[m] | Yes, they do, to avoid benchmarking network latency | 11:28 |
| benschubert | tristan: if you have time also, !1254 is the rework for Element.dependencies | 11:29 |
| gitlab-br-bot | MR !1254: Rework Element.dependencies to be more efficient https://gitlab.com/BuildStream/buildstream/merge_requests/1254 | 11:29 |
| juergbi | jennis, Kinnison: if this only reuses .cache/buildstream/sources and not CAS source cache, then source cache is expected to have a net negative effect (for now), though | 11:30 |
| Kinnison | aah | 11:34 |
| tristan | benschubert, Have been glancing at it | 11:35 |
| *** lachlan has quit IRC | 11:35 | |
| tristan | benschubert, you might also take a glance at https://gitlab.com/BuildStream/buildstream/merge_requests/1256 | 11:36 |
| benschubert | tristan: gah, seems like we would need to update our gitlab-ci.yml for this, seems the old runners are gone | 11:36 |
| tristan | benschubert, at a high level I dont have issues with the Element.dependencies() change - only I want to be sure that the behavior is the same | 11:37 |
| tristan | old runners gone ? | 11:37 |
| * tristan doesnt get it | 11:37 | |
| tristan | oh crap | 11:37 |
| benschubert | tristan: I'm pretty sure the behavior stayed the same since we pass all the tests | 11:37 |
| tristan | everything is broken on gitlab ? | 11:37 |
| tristan | yanked our machines ? | 11:38 |
| tristan | hehe | 11:38 |
| benschubert | for the old runners, it means the docker images | 11:38 |
| benschubert | we changed the ways we were handling docker images for running tests, and apparently removed the previous ones | 11:38 |
| tristan | They just went away ? I recall there was talk of reorganizing the runners | 11:38 |
| benschubert | cs-shadow: do you have more context about the runners? | 11:38 |
| tristan | but didnt make sure the CI was transitioned before deleting stuff ? | 11:38 |
| tristan | Seems easily reparable | 11:39 |
| benschubert | yep, updating the gitlab-ci.yml to point to the new images should do it | 11:39 |
| benschubert | I don't recall any other changes | 11:39 |
| cs-shadow | ah! this is 1.2. I remember updating it on the master branch, seems like 1.2 fell off the radar | 11:41 |
| tristan | benschubert, I think the cache key test *should* be enough to ensure the order of dependencies is stable, however I don't believe that test is | 11:41 |
| cs-shadow | i'll put in a MR in a moment | 11:41 |
| tristan | benschubert, i.e. we currently try to cover every feature (usually a yaml construct) in small test cases and ensure that every possible configuration of anything will result in the same key | 11:43 |
| tristan | And then we have a quite shallow project and `bst show` it | 11:43 |
| benschubert | tristan: not sure I understand your last comment. I meant for the Element.dependencies() changes, the test suite should cover all possible case so the result should bt stable | 11:43 |
| tristan | what it doesnt do, is construct a deep dependency graph | 11:43 |
| tristan | benschubert, That's what I'm not sure of - I'm talking about the cache key test because I think it's most likely to catch changes in dependency graph walking orders | 11:44 |
| tristan | (since dependency order is still relevant in cache keys, but might not be if we change that as we discussed, as an optimization) | 11:45 |
| tristan | Other than that, it is mostly relevant in staging order | 11:45 |
| tristan | So overlap tests should cover it a bit | 11:45 |
| benschubert | the cache keys tests were improved when working on loading rework, so I would think they are pretty much covering everythign currently :) | 11:46 |
| tristan | This actually reminds me | 11:46 |
| gitlab-br-bot | cs-shadow opened MR !1258 (chandan/fix-ci-1.2->bst-1.2: .gitlab-ci.yml: Update CI to use images from GitLab Registry) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1258 | 11:47 |
| tristan | phil, you are working on making sure external plugins have good support from BuildStream for the purpose of testing correct ? | 11:47 |
| tristan | I have mixed some people up last week, hope I am not guessing the wrong person :) | 11:47 |
| cs-shadow | can someone please sanity check ^^ before I hand it over to marge? :) | 11:48 |
| *** alatiera_ has joined #buildstream | 11:48 | |
| *** alatiera has quit IRC | 11:48 | |
| tristan | phil, I think we have a good framework with the update.py script and cache key test harness, and we probably want to expose that in a sane way for external plugins to test for cache key breakages | 11:48 |
| *** alatiera_ is now known as alatiera | 11:49 | |
| gitlab-br-bot | tristanvb approved MR !1258 (chandan/fix-ci-1.2->bst-1.2: .gitlab-ci.yml: Update CI to use images from GitLab Registry) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1258 | 11:49 |
| tristan | cs-shadow, looks good to me | 11:49 |
| cs-shadow | tristan: thanks! | 11:49 |
| *** lachlan has joined #buildstream | 11:53 | |
| *** alatiera has quit IRC | 11:58 | |
| *** lachlan has quit IRC | 12:04 | |
| phil | tristan, You are correct. I'm not really familiar with our cache key tests, but from a brief look I think I agree that it would nice to expose them. I'll open an issue to that effect. | 12:05 |
| tlater[m] | Kinnison: The timing results definitely look promising - but a single commit that touches 105 files? :| | 12:05 |
| * tlater[m] rolls up his sleeves | 12:05 | |
| tristan | phil, Thanks ! | 12:06 |
| jennis | heh, tlater[m], we were following the commit policy | 12:06 |
| tristan | phil, it is pretty simple once you look at it :) | 12:06 |
| jennis | "all commits must pass the tests", or something along those lines | 12:06 |
| tristan | Hah | 12:06 |
| tlater[m] | This sounds like a good situation in which to break that policy x) | 12:07 |
| Kinnison | Not my call :D | 12:07 |
| * Kinnison -> hospital for to have mould stuffed into a hole in his arm :D | 12:07 | |
| *** lachlan has joined #buildstream | 12:07 | |
| juergbi | tristan: thanks for the comment with regards to the scheduler. any opinion on the names for the two push queues? https://gitlab.com/BuildStream/buildstream/merge_requests/1214/#note_151570265 | 12:10 |
| tristan | I have worse :-S | 12:11 |
| tristan | juergbi, Commented | 12:18 |
| juergbi | ta | 12:18 |
| tristan | juergbi, We probably dont want to block on that comment, but it is quite annoying :-S | 12:18 |
| tristan | you'll see heh | 12:18 |
| *** lachlan has quit IRC | 12:18 | |
| juergbi | tristan: raoul did increase the max queue name length from 5 to 8 to allow src-push | 12:19 |
| juergbi | (in frontend widget) | 12:20 |
| raoul | Yeah was just about to say, I didn't seem to mess anything else up from what I saw | 12:20 |
| juergbi | one question is whether this will break anything. the other question is what to call the artifact push queue | 12:20 |
| juergbi | in the branch it's currently art-push but I don't like that too much | 12:20 |
| juergbi | we could just leave it as push but that could be confusing | 12:21 |
| raoul | tristan, for reference made this asciinema of it working https://asciinema.org/a/OIObuXAhwhYi39nXQWkXAJkUW | 12:21 |
| * juergbi would keep 'fetch' instead of 'src-pull' | 12:23 | |
| juergbi | if we wanted to change that, we should consistently change it also for commands and documentation. but I don't think we want that, at least not in this MR | 12:23 |
| juergbi | and in the pipeline summary I wouldn't use any abbreviations | 12:24 |
| juergbi | SKIPPED Art-pull also seems odd. why do we abbreviate that? | 12:24 |
| gitlab-br-bot | BenjaminSchubert approved MR !1258 (chandan/fix-ci-1.2->bst-1.2: .gitlab-ci.yml: Update CI to use images from GitLab Registry) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1258 | 12:25 |
| raoul | The pipeline summary just uses the same name variables, I can extend it to be Artifact Pull etc. | 12:25 |
| raoul | same for the SKIPPED | 12:26 |
| raoul | probably just need to add a full_name to queues and use that where appropriate | 12:26 |
| tristan | (A) agree with juergbi that I would keep 'fetch' instead of 'src-pull'... (B) I don't mind 'art-pull' so much, but it would be nice to have long/short names now that it is becoming more important.. (C) I personally would not want to block this technical branch *too* much on UI improvements | 12:28 |
| tristan | it's probably a better strategy to do concentrated efforts on improving the UI separately from concentrated efforts in the core | 12:29 |
| juergbi | makes sense | 12:29 |
| gitlab-br-bot | cs-shadow approved MR !1254 (bschubert/optimize-dependencies->master: Rework Element.dependencies to be more efficient) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1254 | 12:29 |
| juergbi | wondering whether we should just go with a new src-push queue for this MR and leave all existing queue names the same for now | 12:30 |
| tristan | I would be happy with that | 12:30 |
| juergbi | that way we reduce the risk of renaming the same queue multiple times | 12:30 |
| tristan | Sad to increase the minimum length to 8 | 12:31 |
| tristan | that screen realestate is precious | 12:31 |
| raoul | cool, I can change some of the names back and then I think that's ready to be merged? | 12:32 |
| raoul | (haven't push removing the push queue from source fetch and moving the source push queue after the build queue) | 12:33 |
| *** alatiera has joined #buildstream | 12:41 | |
| *** CTtpollard has joined #buildstream | 12:43 | |
| *** tpollard has quit IRC | 12:43 | |
| *** tristan has quit IRC | 12:44 | |
| *** CTtpollard has quit IRC | 12:49 | |
| gitlab-br-bot | marge-bot123 merged MR !1258 (chandan/fix-ci-1.2->bst-1.2: .gitlab-ci.yml: Update CI to use images from GitLab Registry) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1258 | 12:50 |
| gitlab-br-bot | marge-bot123 merged MR !1254 (bschubert/optimize-dependencies->master: Rework Element.dependencies to be more efficient) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1254 | 13:00 |
| *** tristan has joined #buildstream | 13:16 | |
| *** alatiera has quit IRC | 13:37 | |
| *** alatiera has joined #buildstream | 13:41 | |
| *** nimish2711 has quit IRC | 13:48 | |
| *** nimish2711 has joined #buildstream | 13:51 | |
| *** tpollard has joined #buildstream | 14:12 | |
| gitlab-br-bot | BenjaminSchubert closed issue #864 (Local cache doesn't cleanup properly) on buildstream https://gitlab.com/BuildStream/buildstream/issues/864 | 14:12 |
| *** tpollard has quit IRC | 14:12 | |
| *** nimish2711 has quit IRC | 14:24 | |
| jjardon | cs-shadow: when we talk about the deprecation of the docker images, we have always said they will be eventually removed, but not a specific date (I would imagine at least a year minimum) seems now they are removed and we are breaking the CI in some projects | 14:29 |
| *** tpollard has joined #buildstream | 14:29 | |
| jjardon | Why the rush on removing them? | 14:29 |
| *** nimish2711 has joined #buildstream | 14:30 | |
| *** mohan43u has quit IRC | 14:34 | |
| *** mohan43u has joined #buildstream | 14:37 | |
| *** lachlan has joined #buildstream | 14:46 | |
| raoul | Tristan, juergbi, I've pushed the remote source cache with just src-push name and everything else the same, (had to faff around with the tests that checked for names) and changed the PushQueue to ArtifactPushQueue, so I think it's ready to be merged | 14:49 |
| raoul | After I rebase onto master at least | 14:51 |
| cs-shadow | jjardon: I asked around about the dates in https://mail.gnome.org/archives/buildstream-list/2019-February/msg00047.html but didn't get any response so I assumed they were okay. | 14:53 |
| cs-shadow | The reason for removing them is that it's not really recommended to use images that aren't getting updates. | 14:53 |
| cs-shadow | buildstream-fedora should still be around until May as per the message, but I need to look into what happened to the older tags | 14:53 |
| jjardon | cs-shadow: agreed, but that's a problem to downstream to solve. We even broke our own project here (1.2 branch); eave the images in dockerhub doesn't cost anything, no reason really to remove them without time to downstream to move to the new ones | 14:56 |
| *** lachlan has quit IRC | 15:01 | |
| tlater[m] | jjardon: tbf, this was announced a fair while ago. There was time to move to the new images. | 15:02 |
| *** lachlan has joined #buildstream | 15:02 | |
| * tlater[m] imagines removing them later would just have broken downstream later | 15:03 | |
| benschubert | and the test images were private to the project, so we are the only ones who should break | 15:04 |
| jjardon | And still we broke our own project :) I still argue that they should not be removed in a long long time | 15:04 |
| jjardon | benschubert: not sure I understand, those images are public in dockerhub | 15:05 |
| benschubert | https://gitlab.com/BuildStream/buildstream-docker-images#testsuite-images jjardon | 15:06 |
| tlater[m] | jjardon: We've removed them from the public space specifically because they were supposed to be internal | 15:06 |
| cs-shadow | bst-1.2 was on oversight, for others, I am not sure if leaving it indefinitely is a better alternative. Personally, I'd be in favor of removing the image vs. shipping an image with vulnerabilities because it hasn't been rebuilt in ages | 15:08 |
| jjardon | Right, I guess buildstream-fedora images are not being removed then? | 15:08 |
| cs-shadow | not until May anyway :) | 15:09 |
| cs-shadow | feel free to reply to the ML thread or https://gitlab.com/BuildStream/buildstream-docker-images/issues/13 if you think we should keep them around for longer | 15:09 |
| tlater[m] | Side note - because of this restructure we realized that the benchmarks were always testing buildstream master | 15:10 |
| tlater[m] | Because we were running on buildstream-fedora with a pre-installed buildstream | 15:10 |
| jjardon | cs-shadow: I didn't say keeping them forever, only saying removing them will break CI of projects we are not aware of, and that can make people not very happy | 15:10 |
| *** lachlan has quit IRC | 15:11 | |
| * cs-shadow isn't sure how to avoid CI breakages of projects we don't know about | 15:11 | |
| jjardon | More if we take in account keeping them cost us literally nothing | 15:12 |
| cs-shadow | I disagree that it costs nothing. https://hub.docker.com/u/buildstream used to have two pages of images which made it harder for users browsing the page which one is relevant for their use-case | 15:13 |
| jjardon | cs-shadow: sure, but that's a documentation problem really | 15:15 |
| jjardon | But still, give one month from the deprecation notice to actually removing the images from dockerhub is way too short in my opinion. Sorry I didn't said anything before, somehow I missed that email. BTW, this is in no means a criticism of the move itself! I think the organization of the images is much clear now | 15:20 |
| jjardon | I will reply to the email about buildstream-fedora one | 15:20 |
| jjardon | But problem is not that much project can not update, but that they will not be able to build old releases anymore | 15:21 |
| *** lachlan has joined #buildstream | 15:27 | |
| *** mohan43u has quit IRC | 15:33 | |
| *** tpollard has quit IRC | 15:49 | |
| *** lachlan has quit IRC | 15:52 | |
| *** lachlan has joined #buildstream | 15:54 | |
| *** lachlan has quit IRC | 16:03 | |
| *** lachlan has joined #buildstream | 16:12 | |
| gitlab-br-bot | raoul.hidalgocharman opened MR !1259 (raoul/965-AaaP-service->master: Artifact as a Proto: protos and service) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1259 | 16:23 |
| *** toscalix has quit IRC | 17:00 | |
| *** nimish2711 has quit IRC | 17:11 | |
| *** nimish2711 has joined #buildstream | 17:52 | |
| *** raoul has quit IRC | 17:59 | |
| *** jonathanmaw has quit IRC | 17:59 | |
| *** nimish2711 has quit IRC | 18:00 | |
| benschubert | tristan: are you around? I'm trying to understand why "a.bst" is at this place for https://gitlab.com/BuildStream/buildstream/blob/master/tests/frontend/order.py#L76 (I would expect it to be first) | 18:06 |
| *** alatiera_ has joined #buildstream | 18:12 | |
| *** alatiera has quit IRC | 18:12 | |
| *** alatiera_ is now known as alatiera | 18:13 | |
| benschubert | nevermind, got it | 18:24 |
| *** mablanch has quit IRC | 18:31 | |
| *** lachlan has quit IRC | 19:35 | |
| *** rdale has quit IRC | 19:57 | |
| *** alatiera has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!