| 12 May 2022 |
 Julio Sueiras | since it seem that nix itself uses a binary ghc 8.10.7 for bootstrapping later ghc | 04:27:28 |
 Rick (Mindavi) | In reply to @matthewcroughan:defenestrate.it
Even if you have a content addressed store, should we always expect the output of a cross-compiler to differ from native? I give it a good chance it'll differ, but only one way to find out :) | 05:18:32 |
| Rick (Mindavi) | It does help that the same compiler (gcc) is used, I'd say | 05:19:25 |
| 13 May 2022 |
|  Janne Heß joined the room. | 12:08:21 |
| Janne Heß | Fyi, Hydra will not properly work on non-x86 and non-aarch64 anymore: https://github.com/NixOS/nixpkgs/pull/172378/files#diff-b8f33964a93f34e19fdee3d032e64396c384b7b652f1e167a708b21ab79a0ba9R24500 | 12:09:00 |
| Janne Heß | Letting you know to save you from wasting your time with troubleshooting | 12:09:20 |
| Rick (Mindavi) | Luckily the tests from UUID4-Tiny will alert you to that | 12:19:09 |
| Rick (Mindavi) | But yeah, I really hope it'll be replaced with a dependency that's a bit more cross-platform | 12:19:37 |
| Janne Heß | Now that I think of it… why are we not calling gcc in the postPatch phase? | 12:19:47 |
| Janne Heß | (the call from your comment) | 12:20:00 |
| Rick (Mindavi) | It'd need some cleanup of the output, but definitely an option! | 12:23:27 |
| 17 May 2022 |
|  v0|d joined the room. | 18:04:30 |
 qbit | matthewcroughan - nix.how: out of the gate, bootstrap.sh fails | 19:27:37 |
| qbit | configure.ac:1: warning: AC_INIT: not a literal: "m4_esyscmd(bash -c "echo -n $(cat ./.version)$VERSION_SUFFIX")"
configure.ac:108: error: possibly undefined macro: AC_MSG_ERROR
If this token and others are legitimate, please use m4_pattern_allow.
See the Autoconf documentation.
autoreconf-2.71: error: /usr/local/bin/autoconf-2.71 failed with exit status: 1
| 19:27:42 |
| qbit | ill try the release'd version (with the configure script) - but it fails too iirc | 19:28:08 |
 matthewcroughan - nix.how | do things as specified in hacking.md | 19:28:18 |
| qbit | is that not the first step? | 19:28:49 |
| qbit | it is :P | 19:28:52 |
| matthewcroughan - nix.how | You need to get the correct dependencies though. | 19:29:07 |
| qbit | To build Nix itself in this shell:
[nix-shell]$ ./bootstrap.sh
[nix-shell]$ ./configure $configureFlags --prefix=$(pwd)/outputs/out
[nix-shell]$ make -j $NIX_BUILD_CORES
| 19:29:07 |
| matthewcroughan - nix.how | Which is usually waht nix develop will do. | 19:29:11 |
| qbit | i have them | 19:29:13 |
| matthewcroughan - nix.how | * Which is usually what nix develop will do. | 19:29:15 |
| qbit | https://github.com/jasperla/openbsd-wip/blob/master/devel/nix/Makefile#L35-L49 | 19:29:30 |
| matthewcroughan - nix.how | I'm not sure if you do, the cmd errors in a strange way | 19:29:32 |
| Janne Heß left the room. | 19:29:39 |
| matthewcroughan - nix.how | The problem with the BSDs seems to be that you can't recreate the stdenv | 19:29:49 |
| matthewcroughan - nix.how | Nix has a stdenv, and you may not be creating the exact one it expects. | 19:30:02 |
| matthewcroughan - nix.how | even if it's slight version mismatches | 19:30:13 |
| matthewcroughan - nix.how | Unless you have reason to believe the error you sent is BSD specific | 19:30:32 |
