| 26 Feb 2026 |
 rosssmyth | The issue author says it is because a different pthread library is being picked up | 20:56:32 |
| rosssmyth | But that is the error that occurs | 20:57:02 |
 dramforever | so there are two screenshots | 20:58:00 |
| dramforever | the first one is using the wrong ld which we can just ignore | 20:58:13 |
| dramforever | and the second one is also using the wrong ld | 20:59:39 |
| dramforever | it's the unwrapped ld | 20:59:46 |
| dramforever | oh no | 20:59:47 |
 emily | do I get points for guessing right? | 21:02:18 |
| dramforever | no because i can't reproduce it | 21:04:14 |
| dramforever | * not yet because i can't reproduce it | 21:04:19 |
 Randy Eckenrode | Does it do the right thing if I add pkgsCross.<target>.package to buildInputs? | 21:14:08 |
| Randy Eckenrode | * Does it do the right thing if I add pkgsCross.<target>.<package> to buildInputs? | 21:14:21 |
 Artturin | Yeah since buildInputs is __spliced.hostTarget | 21:15:37 |
| dramforever | rosssmyth: do you still have some exact log of pkgsCross.mingwW64.ripgrep failing? | 21:15:42 |
| dramforever | instead of just > Failure | 21:16:22 |
| Artturin | * Yeah since buildInputs is __spliced.hostTarget
Any further issues caused by dependencies for the wrong system in the dep attributes aren't splicing issues
| 21:16:42 |
| dramforever | my "favorite" part is that this is also clearly not broken for what we've seen above where this started | 21:20:28 |
| dramforever | wait nevermind that's just crane | 21:21:07 |
| rosssmyth | I don't think so :(
Last time I looked into this was a while ago. | 21:21:54 |
| rosssmyth | I'll post on that issue to see if other people can supply it | 21:22:05 |
| rosssmyth | If it is still an issue | 21:22:11 |
| dramforever | i can't reproduce this on the base of https://github.com/NixOS/nixpkgs/pull/457066 either | 21:23:06 |
| rosssmyth | Yeah I can't either | 21:23:51 |
| rosssmyth | Maybe it has somehow been fixed | 21:25:10 |
| rosssmyth | I touch a bunch of the windows package last year so maybe using not jank pthreads fixed it | 21:25:40 |
| dramforever | i'm gonna try 24.05 | 21:25:48 |
| dramforever | or something like that | 21:26:11 |
| dramforever | in case it was somehow broken on an ancient version | 21:26:20 |
| rosssmyth | Cause I remember working on 457066 and testing it, and I rememeber running it once and it was broken, then I ran it again to see see more of the output later and it wasn't. | 21:27:12 |
| rosssmyth | That was the last time I saw it, but that was also the last time I did any real Windows work | 21:28:24 |
