| 7 Oct 2025 |
K900 | (yes there's no 6.10 release so I'm pulling patches into 6.9.3) | 15:39:43 |
K900 | https://github.com/NixOS/nixpkgs/pull/449572 | 17:13:01 |
K900 | OK 6.10 is ~done | 17:13:06 |
| updown joined the room. | 19:48:05 |
| 9 Oct 2025 |
| tavinator joined the room. | 22:17:03 |
| 10 Oct 2025 |
K900 | Anyone want a very cursed snipe? | 07:05:19 |
K900 | https://github.com/NixOS/nixpkgs/pull/450577 | 07:05:20 |
K900 | And also here's a fun post-25.11 exercise | 07:33:18 |
K900 | https://github.com/NixOS/nixpkgs/pull/450591 | 07:33:19 |
Grimmauld (any/all) | In reply to @k900:0upti.me https://github.com/NixOS/nixpkgs/pull/450577 What's the point of doing this? | 08:23:25 |
K900 | What? | 08:24:39 |
K900 | jemalloc? | 08:24:43 |
Grimmauld (any/all) | Ye | 08:24:53 |
K900 | Supposedly 20-30% fast | 08:25:23 |
K900 | Supposedly for free | 08:25:26 |
Grimmauld (any/all) | Huh interesting | 08:25:41 |
K900 | QML is a JS engine and not a very good one | 08:25:54 |
K900 | The allocator stress test comes with the territory | 08:26:13 |
| 11 Oct 2025 |
| somasis changed their display name from Kylie McClain to somasis. | 03:47:38 |
| 13 Oct 2025 |
| vivekanandan_ks joined the room. | 18:58:09 |
| 15 Oct 2025 |
Sandro 🐧 | arch does that, too https://gitlab.archlinux.org/archlinux/packaging/packages/qt6-base/-/blob/main/PKGBUILD?ref_type=heads#L68 | 13:47:36 |
K900 | Well our pyside just segfaults immediately when built with jemalloc | 13:48:52 |
K900 | So I don't know what Arch is doing that's making it work | 13:48:58 |
ghpzin | It should be default off, so even Arch does not seem to use it.
Other than "declaring" that it depends on it. | 13:57:04 |
ghpzin | * It should be default off, so even Arch does not seem to use it.
Other than "declaring" that it depends on it in PKGBUILD. | 13:59:48 |
Sandro 🐧 | they set this https://gitlab.archlinux.org/archlinux/packaging/packages/pyside6/-/blob/main/PKGBUILD?ref_type=heads#L62-63 which we don't set | 14:02:16 |
Sandro 🐧 | yep, seems like it https://github.com/qt/qtbase/blob/104ad65a173e27a54650ef0c165f82e2fdcb2301/src/corelib/configure.cmake#L786 | 14:04:44 |
Sandro 🐧 | I tried building that with jemalloc on pyside6 and got this:
-- Found WrapAtomic: TRUE
CMake Warning at /nix/store/zb7axlmjysv7dd5f67ljjs38nsikd8va-qtbase-6.10.0/lib/cmake/Qt6/Qt6Config.cmake:233 (find_package):
Found package configuration file:
/nix/store/zb7axlmjysv7dd5f67ljjs38nsikd8va-qtbase-6.10.0/lib/cmake/Qt6Core/Qt6CoreConfig.cmake
but it set Qt6Core_FOUND to FALSE so package "Qt6Core" is considered to be
NOT FOUND. Reason given by package:
Qt6Core could not be found because dependency JeMalloc could not be found.
Configuring with --debug-find-pkg=JeMalloc might reveal details why the
package was not found.
Configuring with -DQT_DEBUG_FIND_PACKAGE=ON will print the values of some
of the path variables that find_package uses to try and find the package.
Call Stack (most recent call first):
CMakeLists.txt:18 (find_package)
CMake Error at CMakeLists.txt:18 (find_package):
Found package configuration file:
/nix/store/zb7axlmjysv7dd5f67ljjs38nsikd8va-qtbase-6.10.0/lib/cmake/Qt6/Qt6Config.cmake
but it set Qt6_FOUND to FALSE so package "Qt6" is considered to be NOT
FOUND. Reason given by package:
Failed to find required Qt component "Core".
Expected Config file at
"/nix/store/zb7axlmjysv7dd5f67ljjs38nsikd8va-qtbase-6.10.0/lib/cmake/Qt6Core/Qt6CoreConfig.cmake"
exists
Configuring with --debug-find-pkg=Qt6Core might reveal details why the
package was not found.
Configuring with -DQT_DEBUG_FIND_PACKAGE=ON will print the values of some
of the path variables that find_package uses to try and find the package.
-- Configuring incomplete, errors occurred!
| 15:07:43 |
emily | (can we just mimalloc system-wide or something) | 15:09:04 |
K900 | Use my PR | 15:10:45 |