| 9 Mar 2026 |
 K900 | Or have broken Pipewire socket activation SOMETIMES for a release | 15:33:46 |
| K900 | * Or have broken Pipewire socket activation SOMETIMES for a cycle | 15:33:49 |
| 10 Mar 2026 |
| K900 | A Go rebuild bubbled into next from master | 06:35:48 |
| K900 | smdh | 06:35:50 |
| K900 | Anyway I guess that means I can safely merge Pipewire | 06:36:01 |
 Sandro 🐧 | probably because staging bumped to go 1.26 by default | 12:58:05 |
| K900 | Yeah | 12:58:25 |
|  @mtheil:scs.ems.host left the room. | 14:24:30 |
| Sandro 🐧 | Can someone restart https://hydra.nixos.org/job/nixos/release-25.11/nixpkgs.linuxKernel.kernels.linux_6_18.x86_64-linux ? | 22:44:36 |
| 12 Mar 2026 |
| K900 | Anyone looked into openblas on aa64-darwin? | 11:53:26 |
| K900 | I have none darwin currently | 11:53:39 |
 vcunat | I'm not aware of anyone looking into that, which is why I raised it as
https://github.com/NixOS/nixpkgs/pull/497493#issuecomment-4038052079 | 11:57:08 |
 emily | https://github.com/OpenMathLib/OpenBLAS/pull/5423 https://github.com/NixOS/nixpkgs/pull/441795 | 11:58:28 |
| emily | there is a "remove this when … is resolved" comment, linking to a resolved issue :) | 11:58:52 |
| emily | not sure if it will fix the build or not by itself though | 11:59:06 |
| emily | I think we also set a dodgy arch for it perhaps | 11:59:16 |
| K900 | I don't understand that build system and it scares me | 11:59:30 |
| emily | I don't have time to look at it right now but it should hopefully be a simple fix unless upstream did something wrong again | 11:59:59 |
| emily | the duplicate symbols are a bit weird | 12:00:41 |
| emily | my guess is that https://github.com/OpenMathLib/OpenBLAS/pull/5423 messed something up while fixing the issue we patch | 12:01:12 |
| emily | https://github.com/OpenMathLib/OpenBLAS/pull/5423/commits/55a10c748df2bdc5e4bf678457e446e23e9996fc | 12:04:04 |
| emily | wondering if we should be using the CMake build | 12:04:09 |
| emily | https://github.com/OpenMathLib/OpenBLAS/issues/5625 | 12:05:34 |
| emily | looks maybe related. | 12:06:01 |
| emily | I would try HEAD and try CMake I suppose. can't spend more time on it now though | 12:07:34 |
 Lun | https://github.com/NixOS/nixpkgs/pull/499299 | 16:26:27 |
