| 18 Sep 2025 |
 emily | I'll check back in tomorrow and if all is going well we can get the hook-dropping patch stuff reviewed and hit the button | 01:51:23 |
| emily | or rather the many buttons:
- https://github.com/NixOS/nixpkgs/pull/394444
- https://github.com/NixOS/nixpkgs/pull/442167
- https://github.com/NixOS/nixpkgs/pull/441569
- https://github.com/NixOS/nixpkgs/pull/441576
- https://github.com/NixOS/nixpkgs/pull/443806
- https://github.com/NixOS/nixpkgs/pull/443820
- https://github.com/NixOS/nixpkgs/pull/443821
- https://github.com/NixOS/nixpkgs/pull/443852
- https://github.com/NixOS/nixpkgs/pull/443853
| 01:52:33 |
 Lun | Yay | 01:57:32 |
| emily | I am looking forward to throwing away my staging mega-merge and starting a new one | 02:02:01 |
| emily | it was just starting to become untenable from having to rebase over merge conflicts and then undo stuff to avoid mass rebuilds | 02:02:20 |
| emily | reviews of the outstanding PRs v. much appreciated. someone should check that the glaxnimate plugins that use PIL and stuff actually work… | 02:02:57 |
| emily | nim65s: btw I tweaked some of the fixes from your branch but most of them look good to send out as-is to me. have comments on a few of them, but, tomorrow 😅 | 02:03:31 |
| emily | incredible just how much worse CMake 4 was than LLVM 19→21 | 02:06:02 |
| emily | at least the individual fixes are simple | 02:06:09 |
| emily | but there are so so many of them | 02:06:13 |
| Lun | got barely anywhere on rocm due to not having https://github.com/ROCm/llvm-project/commit/a18cc4c7cb51f94182b6018c7c73acde1b8ebddb | 03:57:23 |
| Lun | * got barely anywhere on rocm due to not having https://github.com/ROCm/llvm-project/commit/a18cc4c7cb51f94182b6018c7c73acde1b8ebddb
dubious quality patch turned into PR https://github.com/NixOS/nixpkgs/pull/443947 | 05:00:55 |
 K900 | All right, going to run it in a bit | 05:34:11 |
| K900 | OK I pressed all the buttons except the big ones | 05:44:18 |
| K900 | Also pushed a new cmake4-integration branch that just rebases the last two on top of current staging | 05:51:49 |
| K900 | (which includes all of current master) | 05:51:54 |
 nim65s | Wow, thanks, my understanding of this syntax was totally wrong. It make a lot of sense now :D | 06:24:24 |
| K900 | That's the typical cmake experience | 06:27:05 |
| K900 | Actually no it's not the typical cmake experience | 06:27:10 |
| K900 | The typical cmake experience is "my understanding of this syntax was totally wrong and WHY DID THEY DO IT LIKE THIS" | 06:27:22 |
| nim65s | Got side tracked on a quest to properly package all https://github.com/cnr-isti-vclab/ stuff upstream (meshlab is on on list of CMake v4 failures), but this is taking ages :/ | 06:28:29 |
| nim65s | and clearly not a priority, but I don't know anything about priority here anyways | 06:28:59 |
| K900 | We don't really have a priority tbh | 06:30:16 |
| K900 | I'm just building the ISO and hoping this gets most things | 06:30:25 |
