26 Jun 2025 |
m0lok | it says the C compiler cannot buidl the project | 04:04:31 |
m0lok | I can build the actual crate with nix build | 04:04:55 |
m0lok | but when I do it on the devShell I get that error | 04:05:05 |
m0lok | this is the devShell:
devShells.default = (
let
moldDevShell = craneLib.devShell.override {
# For example, use the mold linker
mkShell = pkgs.mkShell.override {
stdenv = pkgs.stdenvAdapters.useMoldLinker pkgs.clangStdenv;
};
};
in
moldDevShell {
RUSTC_WRAPPER = "${pkgs.sccache}/bin/sccache";
CARGO_BUILD_TARGET = "x86_64-unknown-linux-musl";
CARGO_BUILD_RUSTFLAGS = "-C target-feature=+crt-static -C linker=clang -C link-arg=-fuse-ld=${pkgs.mold}/bin/mold";
inherit (self.checks.${system}.pre-commit-check) shellHook;
nativeBuildInputs = with pkgs; [pkg-config clang];
buildInputs = [
(rustVersion.override {
extensions = ["rust-src" "rust-analyzer" "rustc" "cargo" "clippy"];
})
self.checks.${system}.pre-commit-check.enabledPackages
grpc_health_probe
];
packages = [
pkgs.upx
pkgs.coreutils
pkgs.rust-analyzer
pkgs.skopeo
pkgs.protobuf
pkgs.systemfd
pkgs.bacon
pkgs.openssl
pkgs.skopeo
pkgs.grpcurl
pkgs.sccache
];
}
)
| 04:05:29 |
| kernz9 joined the room. | 05:39:18 |
| blaxxmith joined the room. | 09:10:12 |
27 Jun 2025 |
| R̴̨͕͇͍̞̮̐̅͆̌̀̉̐͋̈́̃̀͒́̎̅̚̚̚͠͝Ĕ̵̡̛͖͖̟̙̫̱͈̘̞̭͍͍͑̌̄͑̓̋̓̀̈̏̈́͊̇͊͆̉͂̏̀̃̚͘͝͝ͅͅD̶̡̢͔̱̖̮͙͉̘̺͓͍̩̮͈͍͗̃̀̏͌͘͜ͅŚ̸̬̭̯̬͙͇͓̬̩̳̤͚͓̤̩̺͉͖̉͛̓̿̎͊̿̆́̐͂̇͌̄̇̓͘ͅͅT̴̞̫̘̝͇͔̟̪̪̦͂̔̎̀̎ͅŎ̷̡̬̹̪͈̭̣͈̭̭͉̦̖̝̘̪͖͔̥̦̘̻̳Ṋ̶̛̫͈̳̘͚̜̔̋͆̅̈́͊̑͊̉̌̈́̾͑̈́̚ͅË̸̡̨̨̛͇̜̖͔͖̻̟̗̠̙͓̘̗̥͉͇̜͑͆͊͑͑̀̓͒͜͝͝ changed their display name from Redstone to R̴̨͕͇͍̞̮̐̅͆̌̀̉̐͋̈́̃̀͒́̎̅̚̚̚͠͝Ĕ̵̡̛͖͖̟̙̫̱͈̘̞̭͍͍͑̌̄͑̓̋̓̀̈̏̈́͊̇͊͆̉͂̏̀̃̚͘͝͝ͅͅD̶̡̢͔̱̖̮͙͉̘̺͓͍̩̮͈͍͗̃̀̏͌͘͜ͅŚ̸̬̭̯̬͙͇͓̬̩̳̤͚͓̤̩̺͉͖̉͛̓̿̎͊̿̆́̐͂̇͌̄̇̓͘ͅͅT̴̞̫̘̝͇͔̟̪̪̦͂̔̎̀̎ͅŎ̷̡̬̹̪͈̭̣͈̭̭͉̦̖̝̘̪͖͔̥̦̘̻̳Ṋ̶̛̫͈̳̘͚̜̔̋͆̅̈́͊̑͊̉̌̈́̾͑̈́̚ͅË̸̡̨̨̛͇̜̖͔͖̻̟̗̠̙͓̘̗̥͉͇̜͑͆͊͑͑̀̓͒͜͝͝. | 00:54:17 |
| R̴̨͕͇͍̞̮̐̅͆̌̀̉̐͋̈́̃̀͒́̎̅̚̚̚͠͝Ĕ̵̡̛͖͖̟̙̫̱͈̘̞̭͍͍͑̌̄͑̓̋̓̀̈̏̈́͊̇͊͆̉͂̏̀̃̚͘͝͝ͅͅD̶̡̢͔̱̖̮͙͉̘̺͓͍̩̮͈͍͗̃̀̏͌͘͜ͅŚ̸̬̭̯̬͙͇͓̬̩̳̤͚͓̤̩̺͉͖̉͛̓̿̎͊̿̆́̐͂̇͌̄̇̓͘ͅͅT̴̞̫̘̝͇͔̟̪̪̦͂̔̎̀̎ͅŎ̷̡̬̹̪͈̭̣͈̭̭͉̦̖̝̘̪͖͔̥̦̘̻̳Ṋ̶̛̫͈̳̘͚̜̔̋͆̅̈́͊̑͊̉̌̈́̾͑̈́̚ͅË̸̡̨̨̛͇̜̖͔͖̻̟̗̠̙͓̘̗̥͉͇̜͑͆͊͑͑̀̓͒͜͝͝ changed their profile picture. | 00:55:39 |
| psibi left the room. | 09:38:19 |
| nbp changed their display name from nbp to nbp — PTO. | 17:26:33 |
emily | https://guix.gnu.org/en/blog/2025/a-new-rust-packaging-model/ it appears Guix has implemented almost exactly my Rust packaging proposal | 18:43:11 |
emily | (coming from the other end of fully separate full fat Rust library packages though, rather than from the other end of opaque FOD summaries that we're at) | 18:44:41 |
K900 | They're doing multiple versions of crates though? | 18:45:33 |
K900 | I think | 18:45:35 |
emily | that was also my proposal | 18:46:04 |
emily | well, deduplicating where possible within one SemVer major - maybe they're not doing that part | 18:46:27 |
K900 | Yeah I don't think they are | 18:46:44 |
K900 | I'm mostly worried about the duplicating dependency resolution part | 18:46:57 |
emily | cargo_metadata helps somewhat with that | 18:49:12 |
emily | though not entirely | 18:49:17 |
28 Jun 2025 |
| dramforever joined the room. | 07:32:05 |
dramforever | we should probably do #399014 for rust. problem: how do we check for target is not x86(-32)? | 07:32:59 |
dramforever | * we should probably do #399014 for rust. problem: how do we check for target is not x86-32? | 07:33:10 |
Alyssa Ross | We have a rustc wrapper | 07:40:33 |
Alyssa Ross | That I think is to some extent already target-aware | 07:40:58 |
dramforever | so the problem afaiaw | 07:42:56 |
dramforever | is that cargo really wants to use the same rustc for building build.rs and proc macros as the one for building for target | 07:43:17 |
dramforever | so basically it's trying to use the same rustc as both (b, b, b) and (b, b, h) | 07:43:57 |
Alyssa Ross | That's fine, isn't it? | 07:47:31 |
Alyssa Ross | The wrapper can do different things depending on the target | 07:47:42 |