!UNVBThoJtlIiVwiDjU:nixos.org

Staging

318 Members
Staging merges | Find currently open staging-next PRs: https://github.com/NixOS/nixpkgs/pulls?q=is%3Apr+sort%3Aupdated-desc+head%3Astaging-next+head%3Astaging-next-21.05+is%3Aopen109 Servers

Load older messages


SenderMessageTime
21 Oct 2025
@vcunat:matrix.orgVladimír ČunátI noticed.08:03:01
@vcunat:matrix.orgVladimír ČunátI'm afraid that aarch64 had more issues than x86_6408:03:12
@vcunat:matrix.orgVladimír ČunátSo those remained unfixed so far.08:03:21
@k900:0upti.meK900https://hydra.nixos.org/build/310351910/nixlog/1/tail oh no08:09:39
@k900:0upti.meK900 @Wolfgang Walther look another diverging eval case 08:09:50
@wolfgangwalther:matrix.orgWolfgang WaltherDiverging Eval between hydra and GHA?08:11:51
@k900:0upti.meK900Between release-checks and GHA08:12:00
@wolfgangwalther:matrix.orgWolfgang Walther Right. GHA still only checks packages that are built on hydra. This package has hydraPlatforms = [ ]. 08:14:59
@k900:0upti.meK900Great08:15:09
@k900:0upti.meK900image.png
Download image.png
08:15:10
@k900:0upti.meK900Also what the fuck08:15:13
@k900:0upti.meK900 @SomeoneSerge (back on matrix) any ideas what THIS is 08:15:24
@wolfgangwalther:matrix.orgWolfgang WaltherWill fix this.08:16:02
@vcunat:matrix.orgVladimír ČunátIt reads OK in English, but I fail to see the point of such code.08:16:45
@k900:0upti.meK900 This makes no sense because qt6.full is just a buildEnv with all of those modules 08:17:04
@k900:0upti.meK900(and should be deleted the fuck out of nixpkgs, which I'm going to do, like, now)08:17:13
@vcunat:matrix.orgVladimír ČunátThe left sides should always exist, too (I expect).08:17:34
@k900:0upti.meK900Yes08:17:59
@k900:0upti.meK900 @Wolfgang Walther beat you to it :P 08:27:00
@k900:0upti.meK900Also I'm pretty sure the webkit thing is ALSO a red herring08:27:25
@k900:0upti.meK900Because the webkit being pulled in is qtwebkit via qt5.full which they should not have used ANYWAY08:27:40
@k900:0upti.meK900But when I gave it qtbase it failed with a haskell type error anyway08:28:09
@k900:0upti.meK900So I'm going to just call it broken and move on08:28:18
@k900:0upti.meK900 @Wolfgang Walther sniped again 10:24:36
@k900:0upti.meK900I'll wait for the CUDA people to weigh in10:24:44
@wolfgangwalther:matrix.orgWolfgang Walther:)10:25:03
@k900:0upti.meK900I bet they have some custom overlay or something that relies on this cursed behavior10:25:20
@grimmauld:grapevine.grimmauld.deGrimmauld (any/all)the bad webkitgtk was properly removed a few days ago, qt5 webkit needs to follow soon :P10:26:49
@k900:0upti.meK900qt5webkit has been marked insecure for ages now10:27:03
@k900:0upti.meK900It's probably yeetable tbh10:27:07

Show newer messages


Back to Room ListRoom Version: 6