From 218caa410aa38c29984be31a5229b9fa717560ee Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Wed, 17 Apr 2024 14:19:13 +0200 Subject: Merging upstream version 1.68.2+dfsg1. Signed-off-by: Daniel Baumann --- library/std/src/sys/unix/fs.rs | 154 +++++++++++++++++++++++------------------ 1 file changed, 85 insertions(+), 69 deletions(-) (limited to 'library/std/src/sys/unix/fs.rs') diff --git a/library/std/src/sys/unix/fs.rs b/library/std/src/sys/unix/fs.rs index 37a49f2d7..8e1f35d6c 100644 --- a/library/std/src/sys/unix/fs.rs +++ b/library/std/src/sys/unix/fs.rs @@ -149,12 +149,13 @@ cfg_has_statx! {{ ) -> Option> { use crate::sync::atomic::{AtomicU8, Ordering}; - // Linux kernel prior to 4.11 or glibc prior to glibc 2.28 don't support `statx` - // We store the availability in global to avoid unnecessary syscalls. - // 0: Unknown - // 1: Not available - // 2: Available - static STATX_STATE: AtomicU8 = AtomicU8::new(0); + // Linux kernel prior to 4.11 or glibc prior to glibc 2.28 don't support `statx`. + // We check for it on first failure and remember availability to avoid having to + // do it again. + #[repr(u8)] + enum STATX_STATE{ Unknown = 0, Present, Unavailable } + static STATX_SAVED_STATE: AtomicU8 = AtomicU8::new(STATX_STATE::Unknown as u8); + syscall! { fn statx( fd: c_int, @@ -165,31 +166,44 @@ cfg_has_statx! {{ ) -> c_int } - match STATX_STATE.load(Ordering::Relaxed) { - 0 => { - // It is a trick to call `statx` with null pointers to check if the syscall - // is available. According to the manual, it is expected to fail with EFAULT. - // We do this mainly for performance, since it is nearly hundreds times - // faster than a normal successful call. - let err = cvt(statx(0, ptr::null(), 0, libc::STATX_ALL, ptr::null_mut())) - .err() - .and_then(|e| e.raw_os_error()); - // We don't check `err == Some(libc::ENOSYS)` because the syscall may be limited - // and returns `EPERM`. Listing all possible errors seems not a good idea. - // See: https://github.com/rust-lang/rust/issues/65662 - if err != Some(libc::EFAULT) { - STATX_STATE.store(1, Ordering::Relaxed); - return None; - } - STATX_STATE.store(2, Ordering::Relaxed); - } - 1 => return None, - _ => {} + if STATX_SAVED_STATE.load(Ordering::Relaxed) == STATX_STATE::Unavailable as u8 { + return None; } let mut buf: libc::statx = mem::zeroed(); if let Err(err) = cvt(statx(fd, path, flags, mask, &mut buf)) { - return Some(Err(err)); + if STATX_SAVED_STATE.load(Ordering::Relaxed) == STATX_STATE::Present as u8 { + return Some(Err(err)); + } + + // Availability not checked yet. + // + // First try the cheap way. + if err.raw_os_error() == Some(libc::ENOSYS) { + STATX_SAVED_STATE.store(STATX_STATE::Unavailable as u8, Ordering::Relaxed); + return None; + } + + // Error other than `ENOSYS` is not a good enough indicator -- it is + // known that `EPERM` can be returned as a result of using seccomp to + // block the syscall. + // Availability is checked by performing a call which expects `EFAULT` + // if the syscall is usable. + // See: https://github.com/rust-lang/rust/issues/65662 + // FIXME this can probably just do the call if `EPERM` was received, but + // previous iteration of the code checked it for all errors and for now + // this is retained. + // FIXME what about transient conditions like `ENOMEM`? + let err2 = cvt(statx(0, ptr::null(), 0, libc::STATX_ALL, ptr::null_mut())) + .err() + .and_then(|e| e.raw_os_error()); + if err2 == Some(libc::EFAULT) { + STATX_SAVED_STATE.store(STATX_STATE::Present as u8, Ordering::Relaxed); + return Some(Err(err)); + } else { + STATX_SAVED_STATE.store(STATX_STATE::Unavailable as u8, Ordering::Relaxed); + return None; + } } // We cannot fill `stat64` exhaustively because of private padding fields. @@ -243,17 +257,15 @@ struct InnerReadDir { pub struct ReadDir { inner: Arc, - #[cfg(not(any( - target_os = "android", - target_os = "linux", - target_os = "solaris", - target_os = "illumos", - target_os = "fuchsia", - target_os = "redox", - )))] end_of_stream: bool, } +impl ReadDir { + fn new(inner: InnerReadDir) -> Self { + Self { inner: Arc::new(inner), end_of_stream: false } + } +} + struct Dir(*mut libc::DIR); unsafe impl Send for Dir {} @@ -332,11 +344,23 @@ pub struct FileTimes { modified: Option, } -#[derive(Copy, Clone, PartialEq, Eq, Hash, Debug)] +#[derive(Copy, Clone, Eq, Debug)] pub struct FileType { mode: mode_t, } +impl PartialEq for FileType { + fn eq(&self, other: &Self) -> bool { + self.masked() == other.masked() + } +} + +impl core::hash::Hash for FileType { + fn hash(&self, state: &mut H) { + self.masked().hash(state); + } +} + #[derive(Debug)] pub struct DirBuilder { mode: mode_t, @@ -548,7 +572,11 @@ impl FileType { } pub fn is(&self, mode: mode_t) -> bool { - self.mode & libc::S_IFMT == mode + self.masked() == mode + } + + fn masked(&self) -> mode_t { + self.mode & libc::S_IFMT } } @@ -578,18 +606,26 @@ impl Iterator for ReadDir { target_os = "illumos" ))] fn next(&mut self) -> Option> { + if self.end_of_stream { + return None; + } + unsafe { loop { // As of POSIX.1-2017, readdir() is not required to be thread safe; only // readdir_r() is. However, readdir_r() cannot correctly handle platforms - // with unlimited or variable NAME_MAX. Many modern platforms guarantee + // with unlimited or variable NAME_MAX. Many modern platforms guarantee // thread safety for readdir() as long an individual DIR* is not accessed // concurrently, which is sufficient for Rust. super::os::set_errno(0); let entry_ptr = readdir64(self.inner.dirp.0); if entry_ptr.is_null() { - // null can mean either the end is reached or an error occurred. - // So we had to clear errno beforehand to check for an error now. + // We either encountered an error, or reached the end. Either way, + // the next call to next() should return None. + self.end_of_stream = true; + + // To distinguish between errors and end-of-directory, we had to clear + // errno beforehand to check for an error now. return match super::os::errno() { 0 => None, e => Some(Err(Error::from_raw_os_error(e))), @@ -1347,18 +1383,7 @@ pub fn readdir(path: &Path) -> io::Result { } else { let root = path.to_path_buf(); let inner = InnerReadDir { dirp: Dir(ptr), root }; - Ok(ReadDir { - inner: Arc::new(inner), - #[cfg(not(any( - target_os = "android", - target_os = "linux", - target_os = "solaris", - target_os = "illumos", - target_os = "fuchsia", - target_os = "redox", - )))] - end_of_stream: false, - }) + Ok(ReadDir::new(inner)) } } @@ -1739,12 +1764,16 @@ mod remove_dir_impl { use crate::os::unix::io::{AsRawFd, FromRawFd, IntoRawFd}; use crate::os::unix::prelude::{OwnedFd, RawFd}; use crate::path::{Path, PathBuf}; - use crate::sync::Arc; use crate::sys::common::small_c_string::run_path_with_cstr; use crate::sys::{cvt, cvt_r}; - #[cfg(not(all(target_os = "macos", not(target_arch = "aarch64")),))] + #[cfg(not(any( + all(target_os = "linux", target_env = "gnu"), + all(target_os = "macos", not(target_arch = "aarch64")) + )))] use libc::{fdopendir, openat, unlinkat}; + #[cfg(all(target_os = "linux", target_env = "gnu"))] + use libc::{fdopendir, openat64 as openat, unlinkat}; #[cfg(all(target_os = "macos", not(target_arch = "aarch64")))] use macos_weak::{fdopendir, openat, unlinkat}; @@ -1811,21 +1840,8 @@ mod remove_dir_impl { // a valid root is not needed because we do not call any functions involving the full path // of the DirEntrys. let dummy_root = PathBuf::new(); - Ok(( - ReadDir { - inner: Arc::new(InnerReadDir { dirp, root: dummy_root }), - #[cfg(not(any( - target_os = "android", - target_os = "linux", - target_os = "solaris", - target_os = "illumos", - target_os = "fuchsia", - target_os = "redox", - )))] - end_of_stream: false, - }, - new_parent_fd, - )) + let inner = InnerReadDir { dirp, root: dummy_root }; + Ok((ReadDir::new(inner), new_parent_fd)) } #[cfg(any( -- cgit v1.2.3