| 23 Jan 2026 |
 Randy Eckenrode | For some reason, swift build is only using the linker in my toolset configuration. | 03:11:23 |
| Randy Eckenrode | That’s not very helpful. | 03:11:35 |
 Ivy | like link them into zig cache? | 03:12:46 |
| Randy Eckenrode | No idea for Zig. https://github.com/NixOS/nixpkgs/blob/b494a615768aac2f4b0e2839fd23ec02236368fa/pkgs/tools/security/pinentry/mac.nix is an example of a package that does it for .nib files with xcbuild to avoid depending on ibtool. I assume something could be done similarly for Zig. | 03:19:22 |
| Randy Eckenrode | After messing with it for a bit, I don’t think Swift SDKs are going to do what we need. Swift doesn’t pick up the toolchain stuff from it. We’d have to merge the Swift stdlib and Darwin SDK to make it work on Darwin. It also seems to get funny ideas about triples and deployment targets. | 03:44:31 |
|  @buttergrillcorn:matrix.org left the room. | 17:42:13 |
 Sarah Clark | Heads-up: with the latest staging merge into master, nix-manual only builds on Darwin with the sandbox off.
> error: filesystem error: in posix_stat: failed to determine attributes for the specified path: Operation not permitted ["/nix/var/nix/profiles/per-user/root/channels/nixpkgs"]
>
> [9/148] Generating source/command-ref/experimental-features-shortlist.md with a custom command (wrapped by meson to set env, to capture output)
> [10/148] Generating source/store/types with a custom command (wrapped by meson to set env)
> [11/148] Generating source/command-ref/conf-file.body.md with a custom command (wrapped by meson to set env, to capture output)
> [12/148] Generating source/command-ref/new-cli with a custom command (wrapped by meson to set env)
> ninja: build stopped: subcommand failed.
For full logs, run 'nix log /nix/store/9n8f9fa8ifdjpqxfj0wqc6hdwms1fchw-nix-manual-2.31.3.drv'.
| 18:26:21 |
|  vczf changed their profile picture. | 18:47:01 |
| 24 Jan 2026 |
|  kra3 joined the room. | 14:06:38 |
| 25 Jan 2026 |
|  kfiz left the room. | 07:00:06 |
| kfiz joined the room. | 07:00:25 |
|  Contact @rappet:rappet.xyz instead (I will not respond here) changed their display name from Contact @rappet:rappet.xyz instead to Contact @rappet:rappet.xyz instead (I will not respond here). | 21:01:06 |
| 26 Jan 2026 |
 debtquity | swift-5.10.1 no longer compiling for me on unstable. related issue - https://github.com/NixOS/nixpkgs/issues/483584
pinning nixpkgs to 70801e06d9730c4f1704fbd3bbf5b8e11c03a2a7 seems to do the trick [1]
[1] https://github.com/NixOS/nixpkgs/issues/483584#issuecomment-3796732786
| 09:08:25 |
 alexfmpe | yeah it kind of looks like that's what's happening | 09:52:54 |
| alexfmpe | the include chain goes through gcc, sdk and then back again | 09:55:32 |
| alexfmpe | In file included from /nix/store/8pk2m0fn7z5dlb86z5qlv1pxa8w68idx-apple-sdk-14.4/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/secure/_stdio.h:31,
from /nix/store/nvx5zr97w93krpl08x796sqai7z43smi-gcc-15.2.0/lib/gcc/arm64-apple-darwin/15.2.0/include-fixed/stdio.h:423,
from lib/stdio.h:55,
from lib/error.h:47,
from src/hello.c:21:
lib/string.h:754:20: error: expected declaration specifiers or '...' before '__builtin_object_size'
754 | _GL_EXTERN_C void *memcpy (void *__dest, const void *__src, size_t __n)
|
| 09:55:48 |
| alexfmpe | these top-level lib/ come bundled with gnu hello for whatever reason and then it does a tour
gnu -> gcc -> sdk -> gnu | 10:01:47 |
| alexfmpe | what fresh hell is this
/* Declarations for ISO C N3322. */
#if defined __GNUC__ && __GNUC__ >= 15 && !defined __clang__
_GL_EXTERN_C void *memcpy (void *__dest, const void *__src, size_t __n)
_GL_ATTRIBUTE_NONNULL_IF_NONZERO (1, 3)
_GL_ATTRIBUTE_NONNULL_IF_NONZERO (2, 3);
_GL_EXTERN_C void *memccpy (void *__dest, const void *__src, int __c, size_t __n)
_GL_ATTRIBUTE_NONNULL_IF_NONZERO (1, 4)
_GL_ATTRIBUTE_NONNULL_IF_NONZERO (2, 4);
| 10:02:45 |
| alexfmpe | tempted to file this under "gnu hello's problem" | 10:03:14 |
| 27 Jan 2026 |
|  Theo Paris joined the room. | 00:33:53 |
 eveeifyeve | Discussion from the last nix-meeting.
We are going to wait for hydra code-base to rebuilt in rust, before taking on how the scheduling with the jobserver protocol will work.
There is discussions of issues
But this will need to be relooked at later on. | 03:28:23 |
| eveeifyeve | Also Randy Eckenrode I have an issue that you might want to look into: https://github.com/NixOS/nixpkgs/issues/483584 | 03:28:49 |
| eveeifyeve | * Also Randy Eckenrode I have an issue that you might want to look into: https://github.com/NixOS/nixpkgs/issues/483584 (Swift build failure) | 03:29:05 |
| Randy Eckenrode | The compiler crash is a known problem. It seems some people are looking into it. My time is limited, so I am focused on the Swift 6 rewrite. I unfortunately lost some time this week due to dbus and sqlite issues. | 08:19:44 |
| Randy Eckenrode | What is this from? | 08:22:03 |
| eveeifyeve | Last meeting that happen not today but last wednestay when we discussed about the parallelism. | 08:22:59 |
| eveeifyeve | * Last meeting that happen, not today but last wednestay when we discussed about the parallelism. | 08:23:09 |
| Randy Eckenrode | * The compiler crash is a known problem. It seems some people are looking into it. My time is limited, so I am focused on the Swift 6 rewrite.
I unfortunately lost some time this week due to dbus and sqlite issues and have had some life issues come up that have further limited my time.
| 08:23:23 |
| eveeifyeve | * Last meeting that happen, not today but last wednestay when we discussed about the parallelism issues (asking when they will be finished or touched). | 08:23:42 |
| eveeifyeve | Well tonight I will be building sqlite while I am asleep. | 08:24:48 |
