Rollup merge of rust-lang#148765 - joboet:split-up-thread, r=ChrisDenton

std: split up the `thread` module

Almost all functionality in `std::thread` is currently implemented in `thread/mod.rs`, resulting in a *huge* file with more than 2000 lines and multiple, interoperating `unsafe` sections. This PR splits the file up into multiple different private modules, each implementing mostly independent parts of the functionality. The only remaining `unsafe` interplay is that of the `lifecycle` and `scope` modules, the `spawn_scoped` implementation relies on the live thread count being updated correctly by the `lifecycle` module.

This PR contains no functional changes and only moves code around for the most part, with a few notable exceptions:
* `with_current_name` is moved to the already existing `current` module and now uses the `name` method instead of calculating the name from private fields. The old code was just a reimplementation of that method anyway.
* The private `JoinInner` type used to implement both join handles now has some more methods (`is_finished`, `thread` and the `AsInner`/`IntoInner` implementations) to avoid having to expose private fields and their invariants.
* The private `spawn_unchecked_` (note the underscore) method of `Builder` is now a freestanding function in `lifecycle`.

The rest of the changes are just visibility annotations.

I realise this PR ended up quite large – let me know if there is anyway I can aid the review process.

Edit: I've simplified the diff by adding an intermediate commit that creates all the new files by duplicating `mod.rs`. The actual changes in the second commit thus appear to delete the non-relevant parts from the respective file.

Author: Zalathar

library/std/src/thread/builder.rs

@@ -0,0 +1,267 @@
+use super::join_handle::JoinHandle;
+use super::lifecycle::spawn_unchecked;
+use crate::io;
+
+/// Thread factory, which can be used in order to configure the properties of
+/// a new thread.
+///
+/// Methods can be chained on it in order to configure it.
+///
+/// The two configurations available are:
+///
+/// - [`name`]: specifies an [associated name for the thread][naming-threads]
+/// - [`stack_size`]: specifies the [desired stack size for the thread][stack-size]
+///
+/// The [`spawn`] method will take ownership of the builder and create an
+/// [`io::Result`] to the thread handle with the given configuration.
+///
+/// The [`thread::spawn`] free function uses a `Builder` with default
+/// configuration and [`unwrap`]s its return value.
+///
+/// You may want to use [`spawn`] instead of [`thread::spawn`], when you want
+/// to recover from a failure to launch a thread, indeed the free function will
+/// panic where the `Builder` method will return a [`io::Result`].
+///
+/// # Examples
+///
+/// ```
+/// use std::thread;
+///
+/// let builder = thread::Builder::new();
+///
+/// let handler = builder.spawn(|| {
+///     // thread code
+/// }).unwrap();
+///
+/// handler.join().unwrap();
+/// ```
+///
+/// [`stack_size`]: Builder::stack_size
+/// [`name`]: Builder::name
+/// [`spawn`]: Builder::spawn
+/// [`thread::spawn`]: super::spawn
+/// [`io::Result`]: crate::io::Result
+/// [`unwrap`]: crate::result::Result::unwrap
+/// [naming-threads]: ./index.html#naming-threads
+/// [stack-size]: ./index.html#stack-size
+#[must_use = "must eventually spawn the thread"]
+#[stable(feature = "rust1", since = "1.0.0")]
+#[derive(Debug)]
+pub struct Builder {
+    /// A name for the thread-to-be, for identification in panic messages
+    pub(super) name: Option<String>,
+    /// The size of the stack for the spawned thread in bytes
+    pub(super) stack_size: Option<usize>,
+    /// Skip running and inheriting the thread spawn hooks
+    pub(super) no_hooks: bool,
+}
+
+impl Builder {
+    /// Generates the base configuration for spawning a thread, from which
+    /// configuration methods can be chained.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use std::thread;
+    ///
+    /// let builder = thread::Builder::new()
+    ///                               .name("foo".into())
+    ///                               .stack_size(32 * 1024);
+    ///
+    /// let handler = builder.spawn(|| {
+    ///     // thread code
+    /// }).unwrap();
+    ///
+    /// handler.join().unwrap();
+    /// ```
+    #[stable(feature = "rust1", since = "1.0.0")]
+    pub fn new() -> Builder {
+        Builder { name: None, stack_size: None, no_hooks: false }
+    }
+
+    /// Names the thread-to-be. Currently the name is used for identification
+    /// only in panic messages.
+    ///
+    /// The name must not contain null bytes (`\0`).
+    ///
+    /// For more information about named threads, see
+    /// [this module-level documentation][naming-threads].
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use std::thread;
+    ///
+    /// let builder = thread::Builder::new()
+    ///     .name("foo".into());
+    ///
+    /// let handler = builder.spawn(|| {
+    ///     assert_eq!(thread::current().name(), Some("foo"))
+    /// }).unwrap();
+    ///
+    /// handler.join().unwrap();
+    /// ```
+    ///
+    /// [naming-threads]: ./index.html#naming-threads
+    #[stable(feature = "rust1", since = "1.0.0")]
+    pub fn name(mut self, name: String) -> Builder {
+        self.name = Some(name);
+        self
+    }
+
+    /// Sets the size of the stack (in bytes) for the new thread.
+    ///
+    /// The actual stack size may be greater than this value if
+    /// the platform specifies a minimal stack size.
+    ///
+    /// For more information about the stack size for threads, see
+    /// [this module-level documentation][stack-size].
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use std::thread;
+    ///
+    /// let builder = thread::Builder::new().stack_size(32 * 1024);
+    /// ```
+    ///
+    /// [stack-size]: ./index.html#stack-size
+    #[stable(feature = "rust1", since = "1.0.0")]
+    pub fn stack_size(mut self, size: usize) -> Builder {
+        self.stack_size = Some(size);
+        self
+    }
+
+    /// Disables running and inheriting [spawn hooks].
+    ///
+    /// Use this if the parent thread is in no way relevant for the child thread.
+    /// For example, when lazily spawning threads for a thread pool.
+    ///
+    /// [spawn hooks]: super::add_spawn_hook
+    #[unstable(feature = "thread_spawn_hook", issue = "132951")]
+    pub fn no_hooks(mut self) -> Builder {
+        self.no_hooks = true;
+        self
+    }
+
+    /// Spawns a new thread by taking ownership of the `Builder`, and returns an
+    /// [`io::Result`] to its [`JoinHandle`].
+    ///
+    /// The spawned thread may outlive the caller (unless the caller thread
+    /// is the main thread; the whole process is terminated when the main
+    /// thread finishes). The join handle can be used to block on
+    /// termination of the spawned thread, including recovering its panics.
+    ///
+    /// For a more complete documentation see [`thread::spawn`].
+    ///
+    /// # Errors
+    ///
+    /// Unlike the [`spawn`] free function, this method yields an
+    /// [`io::Result`] to capture any failure to create the thread at
+    /// the OS level.
+    ///
+    /// [`io::Result`]: crate::io::Result
+    ///
+    /// # Panics
+    ///
+    /// Panics if a thread name was set and it contained null bytes.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use std::thread;
+    ///
+    /// let builder = thread::Builder::new();
+    ///
+    /// let handler = builder.spawn(|| {
+    ///     // thread code
+    /// }).unwrap();
+    ///
+    /// handler.join().unwrap();
+    /// ```
+    ///
+    /// [`thread::spawn`]: super::spawn
+    /// [`spawn`]: super::spawn
+    #[stable(feature = "rust1", since = "1.0.0")]
+    #[cfg_attr(miri, track_caller)] // even without panics, this helps for Miri backtraces
+    pub fn spawn<F, T>(self, f: F) -> io::Result<JoinHandle<T>>
+    where
+        F: FnOnce() -> T,
+        F: Send + 'static,
+        T: Send + 'static,
+    {
+        unsafe { self.spawn_unchecked(f) }
+    }
+
+    /// Spawns a new thread without any lifetime restrictions by taking ownership
+    /// of the `Builder`, and returns an [`io::Result`] to its [`JoinHandle`].
+    ///
+    /// The spawned thread may outlive the caller (unless the caller thread
+    /// is the main thread; the whole process is terminated when the main
+    /// thread finishes). The join handle can be used to block on
+    /// termination of the spawned thread, including recovering its panics.
+    ///
+    /// This method is identical to [`thread::Builder::spawn`][`Builder::spawn`],
+    /// except for the relaxed lifetime bounds, which render it unsafe.
+    /// For a more complete documentation see [`thread::spawn`].
+    ///
+    /// # Errors
+    ///
+    /// Unlike the [`spawn`] free function, this method yields an
+    /// [`io::Result`] to capture any failure to create the thread at
+    /// the OS level.
+    ///
+    /// # Panics
+    ///
+    /// Panics if a thread name was set and it contained null bytes.
+    ///
+    /// # Safety
+    ///
+    /// The caller has to ensure that the spawned thread does not outlive any
+    /// references in the supplied thread closure and its return type.
+    /// This can be guaranteed in two ways:
+    ///
+    /// - ensure that [`join`][`JoinHandle::join`] is called before any referenced
+    /// data is dropped
+    /// - use only types with `'static` lifetime bounds, i.e., those with no or only
+    /// `'static` references (both [`thread::Builder::spawn`][`Builder::spawn`]
+    /// and [`thread::spawn`] enforce this property statically)
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use std::thread;
+    ///
+    /// let builder = thread::Builder::new();
+    ///
+    /// let x = 1;
+    /// let thread_x = &x;
+    ///
+    /// let handler = unsafe {
+    ///     builder.spawn_unchecked(move || {
+    ///         println!("x = {}", *thread_x);
+    ///     }).unwrap()
+    /// };
+    ///
+    /// // caller has to ensure `join()` is called, otherwise
+    /// // it is possible to access freed memory if `x` gets
+    /// // dropped before the thread closure is executed!
+    /// handler.join().unwrap();
+    /// ```
+    ///
+    /// [`io::Result`]: crate::io::Result
+    /// [`thread::spawn`]: super::spawn
+    /// [`spawn`]: super::spawn
+    #[stable(feature = "thread_spawn_unchecked", since = "1.82.0")]
+    #[cfg_attr(miri, track_caller)] // even without panics, this helps for Miri backtraces
+    pub unsafe fn spawn_unchecked<F, T>(self, f: F) -> io::Result<JoinHandle<T>>
+    where
+        F: FnOnce() -> T,
+        F: Send,
+        T: Send,
+    {
+        let Builder { name, stack_size, no_hooks } = self;
+        Ok(JoinHandle(unsafe { spawn_unchecked(name, stack_size, no_hooks, None, f) }?))
+    }
+}

library/std/src/thread/current.rs

@@ -1,6 +1,9 @@
-use super::{Thread, ThreadId, imp};
+use super::id::ThreadId;
+use super::main_thread;
+use super::thread::Thread;
 use crate::mem::ManuallyDrop;
 use crate::ptr;
+use crate::sys::thread as imp;
 use crate::sys::thread_local::local_pointer;
 
 const NONE: *mut () = ptr::null_mut();
@@ -184,7 +187,7 @@ pub(crate) fn current_os_id() -> u64 {
 
 /// Gets a reference to the handle of the thread that invokes it, if the handle
 /// has been initialized.
-pub(super) fn try_with_current<F, R>(f: F) -> R
+fn try_with_current<F, R>(f: F) -> R
 where
     F: FnOnce(Option<&Thread>) -> R,
 {
@@ -202,6 +205,36 @@ where
     }
 }
 
+/// Run a function with the current thread's name.
+///
+/// Modulo thread local accesses, this function is safe to call from signal
+/// handlers and in similar circumstances where allocations are not possible.
+pub(crate) fn with_current_name<F, R>(f: F) -> R
+where
+    F: FnOnce(Option<&str>) -> R,
+{
+    try_with_current(|thread| {
+        let name = if let Some(thread) = thread {
+            // If there is a current thread handle, try to use the name stored
+            // there.
+            thread.name()
+        } else if let Some(main) = main_thread::get()
+            && let Some(id) = id::get()
+            && id == main
+        {
+            // The main thread doesn't always have a thread handle, we must
+            // identify it through its ID instead. The checks are ordered so
+            // that the current ID is only loaded if it is actually needed,
+            // since loading it from TLS might need multiple expensive accesses.
+            Some("main")
+        } else {
+            None
+        };
+
+        f(name)
+    })
+}
+
 /// Gets a handle to the thread that invokes it. If the handle stored in thread-
 /// local storage was already destroyed, this creates a new unnamed temporary
 /// handle to allow thread parking in nearly all situations.