| 21 Nov 2025 |
|  amadaluzia changed their display name from amadaluzia (in πΉπ· til 25) to amadaluzia (πΉπ· til 25th). | 14:45:28 |
|  isabel changed their profile picture. | 18:14:27 |
| 22 Nov 2025 |
|  @kaya:catnip.ee left the room. | 15:47:32 |
| 24 Nov 2025 |
| amadaluzia changed their display name from amadaluzia (πΉπ· til 25th) to amadaluzia. | 12:58:09 |
| 27 Nov 2025 |
|  γΈγ changed their profile picture. | 02:16:06 |
 Puna | throwing this in here, in case the Go ppl of Nixpkgs have an opinion on this:
{go,go_latest}: Default to gccgo on platforms that normal Go doesn't support | https://github.com/NixOS/nixpkgs/pull/461644 | 15:04:58 |
 jrick | is gccgo relevant at all anymore? iirc they don't implement any of the new language features since generics were introduced | 16:07:21 |
 dish [Fox/It/She] | they dont even support generics, however the one good thing is that it can be used in a bootstrap chain to bootstrap a go compiler from source. That's not currently done in nixpkgs though, and I imagine most modern go code won't work at all on it. | 16:10:14 |
| jrick | looks like gccgo is stuck at go 1.18 | 16:11:15 |
 Sandro π§ | then I am against using that per default on those platforms as we don't gain much with it | 16:15:20 |
 K900 | +1 against because if we do that it'll only be a matter of time before Go package maintainers get yelled at because their shit doesn't build on ppc32 or whatever the fuck | 16:18:42 |
| K900 | And it will be gccgo's fault | 16:18:58 |
| Puna | bootstrapping Go via gccgo is cool, but it doesn't help me because Go is intentionally-borked on ppc64 baseline, and will likely never get fixed unless someone forks & maintains it.
and i definitely don't have the spoons for that π« . | 16:42:42 |
| dish [Fox/It/She] | 1.18 without generics, even | 16:46:13 |
| dish [Fox/It/She] | so definitely opposing this PR being merged | 16:46:20 |
| Puna | that would happen like once in a decade or so, but fair ig.
option to buildGoModule that marks gccgo-support (default false) and affects the derivation's platforms / badPlatforms / broken / whatever? | 16:55:01 |
| Puna | like with the python versions or smth | 16:59:11 |
| K900 | This would absolutely happen tbh | 16:59:15 |
| K900 | And faster than you think | 16:59:26 |
 Paul Meyer (katexochen) | In reply to @sandro:supersandro.de
then I am against using that per default on those platforms as we don't gain much with it I agree | 18:44:07 |
| 28 Nov 2025 |
| Puna | i guess we have more nixos users on wii & wii u than i was anticipating π₯΄.
fine. closed for now, since i don't think that a now-properly-marked-unsupported go prevents anything important from being built anymore (was previously blocking libcap)
| 00:23:06 |
| Puna | * i guess we have more nixos users on wii & wii u than i was anticipating π₯΄.
fine. closed for now, since i don't think that a now-properly-marked-unsupported-on-ppc64 go prevents anything important from being built anymore (was previously blocking libcap)
| 00:30:38 |
| jrick | um, neither of those are ppc64, they're 32-bit | 00:30:45 |
| Puna | yes, but i'm mainly doing this for ppc64 rn | 00:31:11 |
| Puna | issue there: see above | 00:31:29 |
| Puna | non-ppc64le POWER then, I guess | 00:32:00 |
| Puna | * yes, but i'm mainly doing this for ppc64 rn, ppc32 was brought up by k900 | 00:37:26 |
| Puna | * issue with ppc64: see above | 00:37:35 |
| Puna | * yes, but i'm mainly doing this for ppc64 rn. ppc32 was brought up by k900 | 00:37:54 |
| 29 Nov 2025 |
| amadaluzia changed their profile picture. | 11:40:54 |
