| 21 Nov 2025 |
 Grimmauld (any/all) | so it'll be a while until my system is on staging-next, i'll report once that is done | 11:17:37 |
 K900 | Can I have a weirdly philosophical take | 11:19:55 |
| K900 | Linux software generally sucks at telling the user what went wrong | 11:20:04 |
| K900 | SDL goes out of its way to have fallbacks for fallbacks for those cases | 11:20:12 |
 uep | i have webkitgtks and firefox and obs to build, will be a while.. will leave it overnight | 11:20:12 |
| K900 | I think we should respect thta | 11:20:16 |
| uep | I mean I could take them out of the system for this test, but whaever | 11:20:55 |
 Vladimír Čunát | After the night all linux binaries should be in cache.nixos.org, too. | 11:41:26 |
| Grimmauld (any/all) | uh i just got a gtk4 build fail on pkgsi686Linux | 11:43:42 |
| Grimmauld (any/all) | thats fun | 11:43:47 |
| uep | oh it was merged.. yeah, well, i wait till morning either way.. g'night | 11:44:17 |
| Grimmauld (any/all) | > ../gdk/gdkvulkancontext.c: In function 'gdk_vulkan_present_is_busy':
> ../gdk/gdkvulkancontext.c:691:29: warning: comparison between pointer and integer
> 691 | if (present->vk_swapchain == NULL)
> | ^~
> ../gdk/gdkvulkancontext.c: In function 'gdk_vulkan_context_surface_detach':
> ../gdk/gdkvulkancontext.c:1211:42: error: assignment to 'VkSwapchainKHR' {aka 'long long unsigned int'} from 'void *' makes integer from pointer without a cast [-Wint-conversion]
> 1211 | priv->presents[i].vk_swapchain = NULL;
> |
pain | 11:44:50 |
| Vladimír Čunát | People need pedantic linters in the upstream CIs. | 11:54:07 |
 sterni | what do we do now for low-ish rebuild PRs that rebuild the NixOS tests? staging-next? staging? | 14:26:05 |
| K900 | staging-nixos | 14:26:14 |
| sterni | ahh, a vague memory resurfaces | 14:28:24 |
|  isabel changed their profile picture. | 18:13:59 |
| uep | gnome-system-monitor from the new closure (not even bothering to switch) runs fine against gtk 4.20.3 | 20:11:48 |
| K900 | Good | 20:20:54 |
| uep | and fractal too | 20:33:40 |
| 22 Nov 2025 |
| Vladimír Čunát | I'd say we can merge in less than 24h. | 08:57:31 |
| Vladimír Čunát | This is the only thing I noticed that I'd want to resolve before merging ideally:
https://github.com/NixOS/nixpkgs/pull/456828#issuecomment-3566045924
(some of the participants will react fast, I hope) | 08:59:22 |
| K900 | I can fix MLT | 09:03:02 |
| K900 | Updating it should be fine really | 09:03:18 |
| Vladimír Čunát | If you think it's safe, it's probably easy to do. | 09:03:33 |
| K900 | It's a fairly safe update usually | 09:03:37 |
| K900 | And impact should not be big | 09:03:38 |
| K900 | @vcunat took some cmake wrangling but it should be good now | 09:42:58 |
| K900 | Rebuilds all of 14 things | 09:43:06 |
| K900 | Which I've verified locally | 09:43:10 |
