| 15 Sep 2025 |
 matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | * nixpkgs.localSystem = {
system = "aarch64-unknown-linux-gnu";
};
nixpkgs.crossSystem = {
config = "aarch64-unknown-linux-musl";
useLLVM = true;
linker = "lld";
};
``
| 19:58:00 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | * nixpkgs.localSystem = {
system = "aarch64-unknown-linux-gnu";
};
nixpkgs.crossSystem = {
config = "aarch64-unknown-linux-musl";
useLLVM = true;
linker = "lld";
};
| 19:58:05 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | And in that setup, I get the relocation issue | 19:58:21 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | Tristan Ross: in pkgsLLVM.pkgsMusl.buildPlatform the linker is linker = "bfd"; | 20:07:15 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | pkgsLLVM.buildPlatform's linker is also bfd | 20:07:36 |
 Tristan Ross | Yes, that's right | 20:07:43 |
| Tristan Ross | That's the default on GNU | 20:07:50 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | nix-repl> pkgsLLVM.pkgsMusl.hostPlatform.linker
"bfd"
| 20:07:58 |
| Tristan Ross | nix-repl> legacyPackages.aarch64-linux.hostPlatform.linker
"bfd"
| 20:08:02 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | and is that also correct? | 20:08:02 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) |
nix-repl> pkgsLLVM.hostPlatform.linker
"lld"
nix-repl> pkgsLLVM.pkgsMusl.hostPlatform.linker
"bfd"
| 20:08:17 |
| Tristan Ross | Right eval, wrong order | 20:08:18 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | * nix-repl> pkgsLLVM.hostPlatform.linker
"lld"
nix-repl> pkgsLLVM.pkgsMusl.hostPlatform.linker
"bfd"
| 20:08:23 |
| Tristan Ross | pkgsMusl.pkgsLLVM | 20:08:30 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | Ah.. I see | 20:08:35 |
| Tristan Ross | It's the // problem | 20:08:43 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | well that's why it builds then I guess | 20:08:52 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | I got the order wrong | 20:08:56 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | nix build nixpkgs#pkgsMusl.pkgsLLVM.netbsd.getent should fail | 20:09:20 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | Tristan Ross: actually pkgsMusl.pkgsLLVM sets the buildPlatform to musl | 20:17:20 |
| Tristan Ross | Huh | 20:17:33 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | nix build github:nixos/nixpkgs/staging-next#pkgsMusl.pkgsLLVM.netbsd.getent | 20:17:54 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | yes, this fails in the same way | 20:17:57 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | So that's your reproducer | 20:18:44 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | Even when I set the buildPlatform to musl, it fails in the same way | 20:25:06 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | it only worked when I got the order wrong then, that's all | 20:25:19 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | so that derivation just doesn't build for musl atm | 20:25:24 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | * so that derivation just doesn't build for musl + llvm atm | 20:25:41 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | I cannot get it to stop stripping | 21:02:33 |
| matthewcroughan @ 39c3 (DECT 94667 or 97340 or 67192) | dontStrip has no impact on the netbsd derivations | 21:02:38 |
