Rollup merge of #120306 - safinaskar:clone3-clean-up, r=petrochenkov

Clean up after clone3 removal from pidfd code (docs and tests)

https://github.com/rust-lang/rust/pull/113939 removed clone3 from pidfd code. This patchset does necessary clean up: fixes docs and tests
This commit is contained in:
Matthias Krüger 2024-01-25 17:39:28 +01:00 committed by GitHub
commit 8750bec42a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 32 additions and 63 deletions

View File

@ -149,8 +149,7 @@ pub trait CommandExt: Sealed {
/// The pidfd can be retrieved from the child with [`pidfd`] or [`take_pidfd`].
///
/// A pidfd will only be created if it is possible to do so
/// in a guaranteed race-free manner (e.g. if the `clone3` system call
/// is supported). Otherwise, [`pidfd`] will return an error.
/// in a guaranteed race-free manner. Otherwise, [`pidfd`] will return an error.
///
/// If a pidfd has been successfully created and not been taken from the `Child`
/// then calls to `kill()`, `wait()` and `try_wait()` will use the pidfd

View File

@ -147,8 +147,7 @@ impl Command {
#[cfg(not(target_os = "linux"))]
let pidfd = -1;
// Safety: We obtained the pidfd from calling `clone3` with
// `CLONE_PIDFD` so it's valid an otherwise unowned.
// Safety: We obtained the pidfd (on Linux) using SOCK_SEQPACKET, so it's valid.
let mut p = unsafe { Process::new(pid, pidfd) };
let mut bytes = [0; 8];

View File

@ -62,13 +62,14 @@ fn test_command_fork_no_unwind() {
}
#[test]
#[cfg(target_os = "linux")]
#[cfg(target_os = "linux")] // pidfds are a linux-specific concept
fn test_command_pidfd() {
use crate::assert_matches::assert_matches;
use crate::os::fd::{AsRawFd, RawFd};
use crate::os::linux::process::{ChildExt, CommandExt};
use crate::process::Command;
// pidfds require the pidfd_open syscall
let our_pid = crate::process::id();
let pidfd = unsafe { libc::syscall(libc::SYS_pidfd_open, our_pid, 0) };
let pidfd_open_available = if pidfd >= 0 {
@ -81,7 +82,9 @@ fn test_command_pidfd() {
// always exercise creation attempts
let mut child = Command::new("false").create_pidfd(true).spawn().unwrap();
// but only check if we know that the kernel supports pidfds
// but only check if we know that the kernel supports pidfds.
// We don't assert the precise value, since the standard library
// might have opened other file descriptors before our code runs.
if pidfd_open_available {
assert!(child.pidfd().is_ok());
}
@ -97,4 +100,17 @@ fn test_command_pidfd() {
child.kill().expect("failed to kill child");
let status = child.wait().expect("error waiting on pidfd");
assert_eq!(status.signal(), Some(libc::SIGKILL));
let _ = Command::new("echo")
.create_pidfd(false)
.spawn()
.unwrap()
.pidfd()
.expect_err("pidfd should not have been created when create_pid(false) is set");
let _ = Command::new("echo")
.spawn()
.unwrap()
.pidfd()
.expect_err("pidfd should not have been created");
}

View File

@ -106,7 +106,18 @@ mod imp {
// supported on the current kernel.
//
// Also fall back in case it is disabled by something like
// seccomp or inside of virtual machines.
// seccomp or inside of docker.
//
// If the `getrandom` syscall is not implemented in the current kernel version it should return an
// `ENOSYS` error. Docker also blocks the whole syscall inside unprivileged containers, and
// returns `EPERM` (instead of `ENOSYS`) when a program tries to invoke the syscall. Because of
// that we need to check for *both* `ENOSYS` and `EPERM`.
//
// Note that Docker's behavior is breaking other projects (notably glibc), so they're planning
// to update their filtering to return `ENOSYS` in a future release:
//
// https://github.com/moby/moby/issues/42680
//
GETRANDOM_UNAVAILABLE.store(true, Ordering::Relaxed);
return false;
} else if err == libc::EAGAIN {

View File

@ -1,56 +0,0 @@
// run-pass
// only-linux - pidfds are a linux-specific concept
#![feature(linux_pidfd)]
#![feature(rustc_private)]
extern crate libc;
use std::io::Error;
use std::os::linux::process::{ChildExt, CommandExt};
use std::process::Command;
fn has_clone3() -> bool {
let res = unsafe { libc::syscall(libc::SYS_clone3, 0, 0) };
let err = (res == -1)
.then(|| Error::last_os_error())
.expect("probe syscall should not succeed");
// If the `clone3` syscall is not implemented in the current kernel version it should return an
// `ENOSYS` error. Docker also blocks the whole syscall inside unprivileged containers, and
// returns `EPERM` (instead of `ENOSYS`) when a program tries to invoke the syscall. Because of
// that we need to check for *both* `ENOSYS` and `EPERM`.
//
// Note that Docker's behavior is breaking other projects (notably glibc), so they're planning
// to update their filtering to return `ENOSYS` in a future release:
//
// https://github.com/moby/moby/issues/42680
//
err.raw_os_error() != Some(libc::ENOSYS) && err.raw_os_error() != Some(libc::EPERM)
}
fn main() {
// pidfds require the clone3 syscall
if !has_clone3() {
return;
}
// We don't assert the precise value, since the standard library
// might have opened other file descriptors before our code runs.
let _ = Command::new("echo")
.create_pidfd(true)
.spawn()
.unwrap()
.pidfd().expect("failed to obtain pidfd");
let _ = Command::new("echo")
.create_pidfd(false)
.spawn()
.unwrap()
.pidfd().expect_err("pidfd should not have been created when create_pid(false) is set");
let _ = Command::new("echo")
.spawn()
.unwrap()
.pidfd().expect_err("pidfd should not have been created");
}