2.22.1 fails to build due to fs_at crate
0323pin opened this issue · 9 comments
0323pin commented
Describe the bug
2.22.1 fails to build on NetBSD.
Versions:
- OS: NetBSD-current (development branch)
- Rust-1.68.2
Additional context
Build fails due to fs_at as follows:
Compiling fs_at v0.1.4
error[E0425]: cannot find value `O_PATH` in crate `libc`
--> /usr/pkgsrc/wip/navi/work/vendor/fs_at-0.1.4/src/unix.rs:130:55
|
130 | libc::O_RDONLY | libc::O_NOFOLLOW | libc::O_PATH | libc::O_CLOEXEC | libc::O_NOCTTY;
| ^^^^^^ not found in `libc`
For more information about this error, try `rustc --explain E0425`.
error: could not compile `fs_at` due to previous error
warning: build failed, waiting for other jobs to finish...
*** Error code 101
Stop.
make[1]: stopped in /usr/pkgsrc/wip/navi
*** Error code 1
Stop.
welcome commented
Thanks for opening your first issue here! In case you're facing a bug, please update navi to the latest version first. Maybe the bug is already solved! :)
denisidoro commented
Could you please try to use the same versions as defined by the repo?
Namely, rust 1.67.0 instead of 1.68.2 with locked dependencies from Cargo.lock?
0323pin commented
I'm the NetBSD package maintainer and all our package builds are done in offline mode. The crate dependency versions used are the ones specified in Cargo.lock.
I don't see how downgrading Rust would help solving an issue related to a unknown variable.
I understand that this is basically not a navi issue, rather a problem in fs_at but, our package is currently stuck at 2.20.1 which, didn't depend on fs_at.
EDIT: Building fs_at standalone:
fs_at-0.1.4 > cargo build --release --locked
Updating crates.io index
Downloaded static_assertions v1.1.0
Downloaded nix v0.26.2
Downloaded cfg-if v0.1.10
Downloaded bitflags v1.3.2
Downloaded libc v0.2.140
Downloaded cvt v0.1.1
Downloaded cfg-if v1.0.0
Downloaded 7 crates (1010.8 KB) in 1.19s
Compiling libc v0.2.140
Compiling bitflags v1.3.2
Compiling static_assertions v1.1.0
Compiling cfg-if v0.1.10
Compiling cfg-if v1.0.0
Compiling cvt v0.1.1
Compiling nix v0.26.2
Compiling fs_at v0.1.4 (/home/pin/Downloads/fs_at-0.1.4)
error[E0425]: cannot find value `O_PATH` in crate `libc`
--> src/unix.rs:130:55
|
130 | libc::O_RDONLY | libc::O_NOFOLLOW | libc::O_PATH | libc::O_CLOEXEC | libc::O_NOCTTY;
| ^^^^^^ not found in `libc`
For more information about this error, try `rustc --explain E0425`.
error: could not compile `fs_at` due to previous error
Exception: cargo exited with 101
0323pin commented
@denisidoro This is worst than I thought, although I was afraid this would be the case :(
~ > grep -r "O_PATH" libc-0.2.141
...
libc-0.2.141/src/unix/bsd/freebsdlike/freebsd/mod.rs:pub const O_PATH: ::c_int = 0x00400000;
libc-0.2.141/src/unix/linux_like/android/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/emscripten/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b32/mod.rs: pub const O_PATH: ::c_int = 0x1000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b32/mod.rs: pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/aarch64/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/loongarch64/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/mips64/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/powerpc64/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/riscv64/mod.rs:pub const O_PATH: ::c_int = 2097152;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/s390x.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/sparc64/mod.rs:pub const O_PATH: ::c_int = 0x1000000;
libc-0.2.141/src/unix/linux_like/linux/gnu/b64/x86_64/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/musl/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/uclibc/arm/mod.rs:pub const O_PATH: ::c_int = 0o10000000;
libc-0.2.141/src/unix/linux_like/linux/uclibc/mips/mod.rs:pub const O_PATH: ::c_int = 0o010000000;
No NetBSD support in libc :(
denisidoro commented
When I find some time, I'll check cargo tree to understand why fs_at is being included.
If it's not being used by the core logic, in theory I could put the logic behind a cargo feature so that we can build a more stripped down version for NetBSD
0323pin commented
Seria óptimo, se for possível. Obrigado!
0323pin commented
navi v2.22.1 (/home/pin/Downloads/navi)
├── anyhow v1.0.70
├── clap v4.2.1
│ ├── clap_builder v4.2.1
│ │ ├── anstream v0.2.6
│ │ │ ├── anstyle v0.3.5
│ │ │ ├── anstyle-parse v0.1.1
│ │ │ │ └── utf8parse v0.2.1
│ │ │ ├── concolor-override v1.0.0
│ │ │ ├── concolor-query v0.3.3
│ │ │ ├── is-terminal v0.4.6
│ │ │ │ ├── io-lifetimes v1.0.9
│ │ │ │ │ └── libc v0.2.140
│ │ │ │ └── rustix v0.37.7
│ │ │ │ ├── bitflags v1.3.2
│ │ │ │ ├── errno v0.3.0
│ │ │ │ │ └── libc v0.2.140
│ │ │ │ ├── io-lifetimes v1.0.9 (*)
│ │ │ │ └── libc v0.2.140
│ │ │ └── utf8parse v0.2.1
│ │ ├── anstyle v0.3.5
│ │ ├── bitflags v1.3.2
│ │ ├── clap_lex v0.4.1
│ │ ├── once_cell v1.17.1
│ │ └── strsim v0.10.0
│ ├── clap_derive v4.2.0 (proc-macro)
│ │ ├── heck v0.4.0
│ │ ├── proc-macro2 v1.0.56
│ │ │ └── unicode-ident v1.0.8
│ │ ├── quote v1.0.26
│ │ │ └── proc-macro2 v1.0.56 (*)
│ │ └── syn v2.0.13
│ │ ├── proc-macro2 v1.0.56 (*)
│ │ ├── quote v1.0.26 (*)
│ │ └── unicode-ident v1.0.8
│ └── once_cell v1.17.1
├── crossterm v0.26.1
│ ├── bitflags v1.3.2
│ ├── libc v0.2.140
│ ├── mio v0.8.3
│ │ ├── libc v0.2.140
│ │ └── log v0.4.14
│ │ └── cfg-if v1.0.0
│ ├── parking_lot v0.12.0
│ │ ├── lock_api v0.4.6
│ │ │ └── scopeguard v1.1.0
│ │ └── parking_lot_core v0.9.1
│ │ ├── cfg-if v1.0.0
│ │ ├── libc v0.2.140
│ │ └── smallvec v1.10.0
│ ├── signal-hook v0.3.13
│ │ ├── libc v0.2.140
│ │ └── signal-hook-registry v1.4.0
│ │ └── libc v0.2.140
│ └── signal-hook-mio v0.2.3
│ ├── libc v0.2.140
│ ├── mio v0.8.3 (*)
│ └── signal-hook v0.3.13 (*)
├── directories-next v2.0.0
│ ├── cfg-if v1.0.0
│ └── dirs-sys-next v0.1.2
│ └── libc v0.2.140
├── dns_common v0.2.1
│ ├── anyhow v1.0.70
│ ├── directories-next v2.0.0 (*)
│ ├── serde v1.0.159
│ │ └── serde_derive v1.0.159 (proc-macro)
│ │ ├── proc-macro2 v1.0.56 (*)
│ │ ├── quote v1.0.26 (*)
│ │ └── syn v2.0.13 (*)
│ ├── serde_json v1.0.95
│ │ ├── itoa v1.0.6
│ │ ├── ryu v1.0.9
│ │ └── serde v1.0.159 (*)
│ ├── serde_yaml v0.9.21
│ │ ├── indexmap v1.9.3
│ │ │ └── hashbrown v0.12.3
│ │ │ [build-dependencies]
│ │ │ └── autocfg v1.0.1
│ │ ├── itoa v1.0.6
│ │ ├── ryu v1.0.9
│ │ ├── serde v1.0.159 (*)
│ │ └── unsafe-libyaml v0.2.8
│ ├── tracing v0.1.37
│ │ ├── cfg-if v1.0.0
│ │ ├── pin-project-lite v0.2.9
│ │ ├── tracing-attributes v0.1.23 (proc-macro)
│ │ │ ├── proc-macro2 v1.0.56 (*)
│ │ │ ├── quote v1.0.26 (*)
│ │ │ └── syn v1.0.109
│ │ │ ├── proc-macro2 v1.0.56 (*)
│ │ │ ├── quote v1.0.26 (*)
│ │ │ └── unicode-ident v1.0.8
│ │ └── tracing-core v0.1.30
│ │ └── once_cell v1.17.1
│ └── tracing-subscriber v0.3.16
│ ├── matchers v0.1.0
│ │ └── regex-automata v0.1.10
│ │ └── regex-syntax v0.6.29
│ ├── nu-ansi-term v0.46.0
│ │ └── overload v0.1.1
│ ├── once_cell v1.17.1
│ ├── regex v1.7.3
│ │ ├── aho-corasick v0.7.18
│ │ │ └── memchr v2.4.0
│ │ ├── memchr v2.4.0
│ │ └── regex-syntax v0.6.29
│ ├── sharded-slab v0.1.4
│ │ └── lazy_static v1.4.0
│ ├── smallvec v1.10.0
│ ├── thread_local v1.1.7
│ │ ├── cfg-if v1.0.0
│ │ └── once_cell v1.17.1
│ ├── tracing v0.1.37 (*)
│ ├── tracing-core v0.1.30 (*)
│ └── tracing-log v0.1.3
│ ├── lazy_static v1.4.0
│ ├── log v0.4.14 (*)
│ └── tracing-core v0.1.30 (*)
├── dns_common_derive v0.2.1 (proc-macro)
│ ├── proc-macro2 v1.0.56 (*)
│ ├── quote v1.0.26 (*)
│ └── synstructure v0.12.6
│ ├── proc-macro2 v1.0.56 (*)
│ ├── quote v1.0.26 (*)
│ ├── syn v1.0.109 (*)
│ └── unicode-xid v0.2.2
├── edit v0.1.4
│ ├── tempfile v3.2.0
│ │ ├── cfg-if v1.0.0
│ │ ├── libc v0.2.140
│ │ ├── rand v0.8.3
│ │ │ ├── libc v0.2.140
│ │ │ ├── rand_chacha v0.3.1
│ │ │ │ ├── ppv-lite86 v0.2.10
│ │ │ │ └── rand_core v0.6.2
│ │ │ │ └── getrandom v0.2.3
│ │ │ │ ├── cfg-if v1.0.0
│ │ │ │ └── libc v0.2.140
│ │ │ └── rand_core v0.6.2 (*)
│ │ └── remove_dir_all v0.5.3
│ └── which v4.1.0
│ ├── either v1.6.1
│ └── libc v0.2.140
├── lazy_static v1.4.0
├── regex v1.7.3 (*)
├── remove_dir_all v0.8.2
│ ├── cfg-if v1.0.0
│ ├── cvt v0.1.2
│ │ └── cfg-if v1.0.0
│ ├── fs_at v0.1.4
│ │ ├── cfg-if v1.0.0
│ │ ├── cvt v0.1.2 (*)
│ │ ├── libc v0.2.140
│ │ └── nix v0.26.2
│ │ ├── bitflags v1.3.2
│ │ ├── cfg-if v1.0.0
│ │ ├── libc v0.2.140
│ │ └── static_assertions v1.1.0
│ ├── lazy_static v1.4.0
│ ├── libc v0.2.140
│ └── normpath v1.1.1
├── serde v1.0.159 (*)
├── serde_yaml v0.9.21 (*)
├── shellwords v1.1.0
│ ├── lazy_static v1.4.0
│ └── regex v1.7.3 (*)
├── strip-ansi-escapes v0.1.1
│ └── vte v0.10.1
│ ├── arrayvec v0.5.2
│ ├── utf8parse v0.2.1
│ └── vte_generate_state_changes v0.1.1 (proc-macro)
│ ├── proc-macro2 v1.0.56 (*)
│ └── quote v1.0.26 (*)
├── thiserror v1.0.40
│ └── thiserror-impl v1.0.40 (proc-macro)
│ ├── proc-macro2 v1.0.56 (*)
│ ├── quote v1.0.26 (*)
│ └── syn v2.0.13 (*)
├── unicode-width v0.1.10
└── walkdir v2.3.3
└── same-file v1.0.6
Looks like remove_dir_all v0.8.2 is pulling it in.
0323pin commented
@denisidoro Workaround patch submitted, rbtcollins/fs_at#74
...
Compiling navi v2.22.1 (/usr/pkgsrc/wip/navi/work/navi-2.22.1)
Finished release [optimized] target(s) in 3m 48s
0323pin commented