| 13 Nov 2025 |
 Grimmauld (any/all) | currently just experimenting with llvm musl, no cross yet | 10:47:48 |
 K900 | Huh, what are you overriding there? | 10:54:48 |
| K900 | Or just removing the GTK dependency? | 10:55:15 |
| Grimmauld (any/all) | ye just setting withGtk3 = false | 10:57:18 |
| Grimmauld (any/all) | qtbase and gst-bad are the two things pulling gtk3, and both can be killed by override | 10:57:53 |
| Grimmauld (any/all) | i have already killed go, haskell and fortran. Now there is no gtk3 or qt5 either. I'll want qt6 (because quickshell), and i need sdl3/sdl2 (because i like ffplay, and qemu display is sdl) | 10:59:15 |
| Grimmauld (any/all) | but i am slowly killing all the ugly garbage | 10:59:37 |
| K900 | What are you trying to build? | 11:00:02 |
| Grimmauld (any/all) | my musl vm, today building with llvm instead of gcc (and doing some more cleanup) | 11:00:26 |
| Grimmauld (any/all) | I have found some crimes in nixpkgs already | 11:00:53 |
| K900 | No I mean like, what's in the closure | 11:01:10 |
| Grimmauld (any/all) | E.g. https://github.com/NixOS/nixpkgs/blob/9da7f1cf7f8a6e2a7cb3001b048546c92a8258b4/pkgs/by-name/x2/x265/package.nix#L130-L133 can just be removed. | 11:01:21 |
| Grimmauld (any/all) | its a normal-ish nixos system. There is systemd, sway, quickshell, servo, polkit, pam, networkd, nftables, all the good stuff | 11:02:05 |
| Grimmauld (any/all) | and quickshell pulls qt6, but i don't want gtk3 on here | 11:02:43 |
| Grimmauld (any/all) | i am hoping i can get this system small enough to actually track staging instead of staging-next half way into the cycle | 11:03:47 |
| K900 | Speaking of | 11:05:10 |
| 14 Nov 2025 |
| Grimmauld (any/all) | Btw, is this intended:
https://github.com/NixOS/nixpkgs/blob/0d9bcf4d56844168da8ea9d32b01eed507bb0371/pkgs/development/libraries/qt-6/default.nix#L65-L68
https://github.com/NixOS/nixpkgs/blob/0d9bcf4d56844168da8ea9d32b01eed507bb0371/pkgs/development/libraries/qt-6/modules/qtbase/default.nix#L78
withGtk3 defaults to force, but is then forced to true. Feels wonky. Shouldn't the default itself be !withMingw if that is really what we want? | 08:22:36 |
| K900 | Probably should yeah | 08:27:05 |
| Grimmauld (any/all) | There is also something wrong with the cross check:
https://github.com/NixOS/nixpkgs/blob/0d9bcf4d56844168da8ea9d32b01eed507bb0371/pkgs/development/libraries/qt-6/modules/qtbase/default.nix#L90
it is a canExecute check, but this fails on cross targets that can actually be executed, such as building x86_64-unknown-linux-musl as cross (as in, useLLVM = true in the host platform config while on a glibc build system). This makes cmake confused and complain about a missing DQT_HOST_PATH.
Can be fixed by adding (lib.cmakeBool "CMAKE_CROSSCOMPILING" isCrossBuild) cmake flag to force cmake to treat executable cross as native build. | 08:27:11 |
| Grimmauld (any/all) | Btw, is this intended:
https://github.com/NixOS/nixpkgs/blob/0d9bcf4d56844168da8ea9d32b01eed507bb0371/pkgs/development/libraries/qt-6/default.nix#L65-L68
https://github.com/NixOS/nixpkgs/blob/0d9bcf4d56844168da8ea9d32b01eed507bb0371/pkgs/development/libraries/qt-6/modules/qtbase/default.nix#L78
withGtk3 defaults to false but is then forced to true. Feels wonky. Shouldn't the default itself be !withMingw if that is really what we want? | 08:27:22 |
| K900 | Qt cross is still extremely cursed | 08:27:43 |
| Grimmauld (any/all) | yes | 08:27:48 |
| Grimmauld (any/all) | but i am doing it and i think this is a fix we should do | 08:28:00 |
| Grimmauld (any/all) | (if it works, we'll see) | 08:28:08 |
| Grimmauld (any/all) | feels weird to have executable cross fail while non-executable cross just works | 08:29:08 |
| Grimmauld (any/all) | though the issue is the same in other qt components | 08:30:11 |
| Grimmauld (any/all) | its a bit painful | 08:30:14 |
| Grimmauld (any/all) | * though the issue is the same in other qt components too | 08:30:27 |
| K900 | Any effort towards unfucking it will be greatly appreciated | 08:31:07 |
| Grimmauld (any/all) | the proper fix is probably some cmake fuckery... | 08:33:20 |
