Age | Commit message (Collapse) | Author |
|
1759: More docs for dir and mqueue r=rtzoeller a=asomers
Add doc comments for the `dir` and `mqueue` modules. Also, delete dead code in `mqueue`
1760: Add const constructors for TimeSpec and TimeVal r=rtzoeller a=asomers
These are basically the same as From<libc::timespec> and
From<libc::timeval>, but they're const and require less typing.
Co-authored-by: Alan Somers <asomers@gmail.com>
|
|
These are basically the same as From<libc::timespec> and
From<libc::timeval>, but they're const and require less typing.
|
|
|
|
Add the autocfg crate as a build dependency, and introduce
has_doc_alias as a conditional compilation symbol.
|
|
1748: Add format test to CI r=rtzoeller a=costinsin
To enforce uniformity for all PRs, the CI checks if the code
is formatted right using `cargo fmt` tool.
Results after implementing the format test in CicleCI, but before fixing the format errors: https://cirrus-ci.com/build/4684991404703744
Results after fixing the format errors: https://cirrus-ci.com/build/5423803479097344
Solves #770
Co-authored-by: Costin-Robert Sin <sin.costinrobert@gmail.com>
|
|
Signed-off-by: Costin-Robert Sin <sin.costinrobert@gmail.com>
|
|
Includes an enum to specify what to get resource usage for, and a new
struct that provides a more readable view into libc::rusage, including
using TimeVal for user and system CPU time.
|
|
1736: Fix socket address family checks r=rtzoeller a=qwandor
The `SockaddrLike::from_raw` implementations for `VsockAddr` and `SysControlAddr` were checking against the wrong address family constant. This PR makes them consistent with the values matched against in `SockaddrStorage::from_raw`.
Co-authored-by: Andrew Walbran <qwandor@google.com>
|
|
|
|
|
|
Remove obsolete references to target_env = wasi, target_os = nacl,
target_os = osx, and a typo'd target_os = fushsia that didn't compile
when fixed.
- target_env = wasi is dead: https://github.com/rust-lang/rust/pull/60117
- target_os = nacl is dead: https://github.com/rust-lang/rust/pull/45041
- target_os = osx is dead, but I can't find a link.
|
|
This was an oversight from #1684.
Fixes #1728
|
|
|
|
|
|
* enabled as much functionality and defines that match
updated libc definitions for haiku
|
|
The existing AIO implementation has some problems:
1) The in_progress field is checked at runtime, not compile time.
2) The mutable field is checked at runtime, not compile time.
3) A downstream lio_listio user must store extra state to track whether
the whole operation is partially, completely, or not at all
submitted.
4) Nix does heap allocation itself, rather than allowing the caller to
choose it. This can result in double (or triple, or quadruple)
boxing.
5) There's no easy way to use lio_listio to submit multiple operations with
a single syscall, but poll each individually.
6) The lio_listio usage is far from transparent and zero-cost.
7) No aio_readv or aio_writev support.
8) priority has type c_int; should be i32
9) aio_return should return a usize instead of an isize, since it only
uses negative values to indicate errors, which Rust represents via
the Result type.
This rewrite solves several problems:
1) Unsolved. I don't think it can be solved without something like
C++'s guaranteed type elision. It might require changing the
signature of Future::poll too.
2) Solved.
3) Solved, by the new in_progress method and by removing the complicated
lio_listio resubmit code.
4) Solved.
5) Solved.
6) Solved, by removing the lio_listo resubmit code. It can be
reimplemented downstream if necessary. Or even in Nix, but it
doesn't fit Nix's theme of zero-cost abstractions.
7) Solved.
8) Solved.
9) Solved.
The rewrite includes functions that don't work on FreeBSD, so add CI
testing for FreeBSD 14 too.
By default only enable tests that will pass on FreeBSD 12.3. But run a
CI job on FreeBSD 14 and set a flag that will enable such tests.
|
|
It just so happens that Redox, OpenBSD, Dragonfly, and uclibc don't use
some of the rules for two internal macros.
|
|
1716: Future-proof the kevent ABI r=rtzoeller a=asomers
FreeBSD 12 changes struct kevent. For now, libc always binds to the
11-compat ABI. But that will change some day. Adjust Nix's code to
build with either struct definition.
Co-authored-by: Alan Somers <asomers@gmail.com>
|
|
These constants are provided by Linux (linux/magic.h), not libc.
See rust-lang/libc#2639.
|
|
FreeBSD 12 changes struct kevent. For now, libc always binds to the
11-compat ABI. But that will change some day. Adjust Nix's code to
build with either struct definition.
|
|
Fixes #1710
|
|
SockaddrLike::size() is meant to return the amount of space that can be
used to store the sockaddr. But on Linux-based OSes, UnixAddr contains
an extra field to store the address's length. This field is not part of
the address, and should not contribute to the value of size().
This bug can't cause an out-of-bounds write, and every OS that we test
on can tolerate the greater-than-expected length, but it might confuse
applications that implement functions similar to getsockname in
userland.
|
|
features => feature
users => user
Neither of these features have yet been included in a release, so it's
ok to rename them.
|
|
1643: Replace the IoVec struct with IoSlice and IoSliceMut from the standard library r=asomers a=notgull
As per discussion in #1637, the `IoVec<&[u8]>` and `IoVec<&mut [u8]>` types have been replaced with `std::io::IoSlice` and `IoSliceMut`, respectively. Notable changes made in this pull request include:
- The complete replacement of `IoVec` with `IoSlice*` types in both public API, private API, and tests.
- Replacing `IoVec` with `IoSlice` in docs.
- Replacing `&[IoVec<&mut [u8]>]` with `&mut [IoSliceMut]`, note that the slice requires a mutable reference now. This is how it's done in the standard library, and there might be a soundness issue in doing it the other way.
Resolves #1637
Co-authored-by: not_a_seagull <notaseagull048@gmail.com>
|
|
|
|
IP_DONTFRAG: iOS, macOS
IPV6_DONTFRAG: android, iOS, linux and macOS
Test: `cargo test --test test dontfrag_opts`
Some CI tests running ENOPROTOOPT are disabled (qemu-based).
|
|
The statfs magic constants of file systems types are available on
target_os android and the cfg guard is updated accordingly.
Sync the list of constant with the constants declared in libc.
Fixes #1689
|
|
1672: Make `uname` always safe r=asomers a=koute
Currently `uname` doesn't check for errors and just blindly assumes that it always succeeds. According to the manpage this function can fail, even though no actual errors are defined:
```
RETURN VALUE
Upon successful completion, a non-negative value shall be returned. Otherwise, -1 shall be returned and errno set to indicate the error.
ERRORS
No errors are defined.
The following sections are informative.
```
Looking at [the glibc's sources](https://github.com/bminor/glibc/blob/b92a49359f33a461db080a33940d73f47c756126/posix/uname.c#L29) we can see that it indeed could fail if the internal `gethostname` call fails for some reason.
This code also assumes that every field of `utsname` is going to be initialized by the call to `uname`, which apparently is also not true. Even though the interface doesn't expose this field so it's not a problem in practice (although it might be UB since we do call `assume_init` on the whole struct) [the `utsname` does have a `domainname` field](https://docs.rs/libc/0.2.119/libc/struct.utsname.html) which glibc doesn't initialize.
The code also assumes that every field is a valid UTF-8 string, which is also technically not guaranteed.
The code also assumes that every field will be null terminated, which might not be true if any of the strings are too long (since glibc uses `strncpy` which will *not* null-terminate the string if it ends up running out of space).
This PR should fix all of these problems.
This is a breaking change.
Co-authored-by: Jan Bujak <jan@parity.io>
|
|
1686: [skip ci] better docs for SockaddrLike::from_raw r=asomers a=asomers
Fixes #1680
Co-authored-by: Alan Somers <asomers@gmail.com>
|
|
This fixes several issues with the current `uname` bindings:
- Do not ignore `uname` errors; at least on glibc `uname` can fail,
so now it returns a `Result` instead of assuming that the call
will always succeed.
- Do not assume `uname` will initialize every member of `utsname`;
not every implementation initializes every field, so internally
the struct is now zero-initialized.
- Do not blindly assume strings returned by `uname` will always be valid UTF-8;
`UtsName`'s accessors will now return `&OsStr`s instead of `&str`s.
|
|
Fixes #1680
|
|
Because they're redundant with types in the standard library.
Fixes #1681
|
|
The SockAddr enum is quite large, and the user must allocate space for
the whole thing even though he usually knows what type he needs.
Furthermore, thanks to the sa_family field, the sockaddr types are
basically an enum even in C.
So replace the ungainly enum with a SockaddrLike trait implemented by
all sockaddr types and a SockaddrStorage union that has safe accessors.
Also, deprecate InetAddr, which only existed to support SockAddr.
Supplants #1504
Fixes #1544
|
|
|
|
|
|
This should've been done as part of #1675
|
|
Fixes #1666
|
|
The function assumed something about the values of the sockaddr_dl's
fields. But because the inner type is public, we musn't do that. The
only solution is to change the function's signature to return an Option.
|
|
1671: docs: fix link in doc for cmsg_space in sys/socket r=rtzoeller a=liubin
Link to cmsg_space macro for recvmmsg and recvmsg should
be the doc's root of crate as macro is an exported macro.
Signed-off-by: bin liu <liubin0329@gmail.com>
Co-authored-by: bin liu <liubin0329@gmail.com>
|
|
waitid() has a number of additional features that waitpid() is missing:
- WNOWAIT is only accepted for waitid() on Linux (and possibly other
platforms)
- Support for waiting on PID file descriptors on Linux
For now support is added for all platforms with waitid() that have proper
siginfo_t support in libc. NetBSD support is currently a work in progress
[1].
Tests for the signal/exit code are currently skipped on MIPS platforms due
to bugs in qemu-user's translation of siginfo_t (fixed in [2] and [3]; the
second fix is not in a released qemu version yet).
[1] https://github.com/rust-lang/libc/pull/2476
[2] https://lists.nongnu.org/archive/html/qemu-devel/2021-01/msg04810.html
[3] https://lists.nongnu.org/archive/html/qemu-devel/2021-10/msg05433.html
|
|
Link to cmsg_space macro for recvmmsg and recvmsg should
be the doc's root of crate as macro is an exported macro.
Signed-off-by: bin liu <liubin0329@gmail.com>
|
|
1670: Add `MsgFlag::MSG_NOSIGNAL` r=rtzoeller a=i509VCB
(I guess I will see if I did the cfg block correctly)
Co-authored-by: i509VCB <git@i509.me>
|
|
|
|
Upstream-Status: Pending
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
|
|
|
|
|
CHANGELOG: add process_vm_* entry
process_vm_*: fix documentation for android
expose process_vm_readv, process_vm_writev for android
Signed-off-by: rupansh-arch <rupanshsekar@hotmail.com>
|
|
1653: Document inotify, mman, personality, reboot, timerfd r=asomers a=rtzoeller
Document the `inotify`, `mman`, `personality`, `reboot`, and `timerfd` modules.
Co-authored-by: Ryan Zoeller <rtzoeller@rtzoeller.com>
|
|
|