| 26 Sep 2021 |
 Vladimír Čunát | unstable-small has it already. | 17:38:30 |
 lukegb (he/him) | Reviewing the diffs, we have:
bumps for:
- imagemagick
- eigen
- mandoc
- python3.sqlalchemy
- gd (to 2.3.2)
- texinfo
- linuxHeaders
- pcre2
- gdb
- xmrig
- libva
- valgrind
as well as:
- imagemagick JPEG XL build option
- mandoc fixes
- llvm_5 fixes
- xcbuild build fix
- {cc,binutils}-wrapper fixes
- autoPatchelfHook fixups
- mesa OpenCL support
- qt5 deps
| 17:42:27 |
| Vladimír Čunát | The staging builds on Hydra were unblocked today, so it should be possible. | 17:55:57 |
| lukegb (he/him) | https://github.com/NixOS/nixpkgs/pull/139554 | 17:56:14 |
|  das_j joined the room. | 18:20:39 |
| das_j changed their display name from Janne Heß to das_j. | 18:21:35 |
 sterni | hm the eval from yesterday still has 32000 queued build? these are still the aarch64-darwin build, probably? | 18:49:47 |
| lukegb (he/him) | yeah | 18:49:58 |
| lukegb (he/him) | hey das_j, was wondering if you were about | 18:50:05 |
| lukegb (he/him) | wanted to try to minimize the changes that would land alongside the coreutils bump, although... it shouldn't be too disruptive? | 18:51:17 |
| das_j | It looks like it's less disruptive than I was initially expecting. Maybe Darwin will give me trouble, I'm currently trying to build stdenv | 18:52:14 |
| das_j | My *-linux builds are about halfway through now without any problems so if problems should come up, they will probably not be in the -small set | 18:53:15 |
| 27 Sep 2021 |
| Vladimír Čunát | In reply to @sternenseemann:systemli.org
hm the eval from yesterday still has 32000 queued build? these are still the aarch64-darwin build, probably? I usually click the eval and put the name of the platform into the eval box. It's all darwin but 18:12 for x86:aarch. | 05:50:03 |
 j-k | When does staging merge into staging-next? Is there an open PR somewhere? | 10:18:15 |
| Vladimír Čunát | We don't do PRs for that. | 10:18:47 |
| Vladimír Čunát | The merge happens manually and then the resulting staging-next gets a PR against master. | 10:19:11 |
| Vladimír Čunát | * The merge happens "silently" and then the resulting staging-next gets a PR against master. | 10:19:27 |
| Vladimír Čunát | * The merge happens "silently" and then the resulting staging-next gets a PR against master. (the last link above is the current one) | 10:20:17 |
 trofi | Is it a fast-forward most of the time? | 16:34:03 |
| Vladimír Čunát | I think nowadays usually people don't use --no-ff when merging staging into staging-next, and it tends to be fast-forward. | 16:39:06 |
| Vladimír Čunát | As for staging-next to master, there I'd dislike a fast-forward merge. | 16:39:57 |
| 28 Sep 2021 |
 happysalada | It's a littlle odd at the moment, staging-next is a priority but the unstable build has a lower priority and so the last staging-next changes haven't made it onto unstable yet | 23:09:48 |
| lukegb (he/him) | I think that's mostly down to the fact that unstable is broken | 23:27:07 |
| lukegb (he/him) | so the channel won't advance | 23:27:17 |
| 29 Sep 2021 |
| Vladimír Čunát | trunk-combined jobset has a bit less "build shares" configuration than staging-next (one third, roughly). I agree it would be better to reverse their order. | 05:47:02 |
| Vladimír Čunát | trunk has more shares than the other two (which helps trunk-combined except for NixOS stuff, i.e. mainlytests). | 05:48:21 |
| Vladimír Čunát | By the way, there's been lots of work scheduled in the past few days, so I wasn't running into adding more so far, e.g. https://github.com/NixOS/nixpkgs/pull/139514#issuecomment-927260195
I suppose we'll also want a staging-21.05 iteration soon, too, as we have more than week-old pushes in there with CVEs. | 05:52:58 |
| happysalada | no problem at all, I'm still getting feedback on it. I've tested building chromium and it works fine, so hopefully it should be fairly ready to merge. | 10:01:02 |
| happysalada | In reply to @lukegb:zxcvbnm.ninja
so the channel won't advance the channel seems broken just because of cancel buids (It seems). It looks like the channel is a little starved, just like vladimir was saying | 10:02:28 |
| lukegb (he/him) | It'd be quite nice if we could update the topic of the room to mention the latest PR... | 12:01:37 |
