rust/library/std/src/sys_common/poison.rs

263 lines
8.1 KiB
Rust
Raw Normal View History

use crate::error::Error;
2019-02-10 19:23:21 +00:00
use crate::fmt;
use crate::sync::atomic::{AtomicBool, Ordering};
use crate::thread;
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
pub struct Flag {
failed: AtomicBool,
}
// Note that the Ordering uses to access the `failed` field of `Flag` below is
// always `Relaxed`, and that's because this isn't actually protecting any data,
// it's just a flag whether we've panicked or not.
//
// The actual location that this matters is when a mutex is **locked** which is
// where we have external synchronization ensuring that we see memory
// reads/writes to this flag.
//
// As a result, if it matters, we should see the correct value for `failed` in
// all cases.
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
impl Flag {
pub const fn new() -> Flag {
Flag { failed: AtomicBool::new(false) }
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
#[inline]
pub fn borrow(&self) -> LockResult<Guard> {
2015-02-17 23:10:25 +00:00
let ret = Guard { panicking: thread::panicking() };
if self.get() { Err(PoisonError::new(ret)) } else { Ok(ret) }
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
}
#[inline]
pub fn done(&self, guard: &Guard) {
2015-02-17 23:10:25 +00:00
if !guard.panicking && thread::panicking() {
self.failed.store(true, Ordering::Relaxed);
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
}
}
#[inline]
pub fn get(&self) -> bool {
self.failed.load(Ordering::Relaxed)
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
}
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
pub struct Guard {
2014-12-07 02:34:37 +00:00
panicking: bool,
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// A type of error which can be returned whenever a lock is acquired.
///
2017-02-24 19:56:04 +00:00
/// Both [`Mutex`]es and [`RwLock`]s are poisoned whenever a thread fails while the lock
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// is held. The precise semantics for when a lock is poisoned is documented on
/// each lock, but once a lock is poisoned then all future acquisitions will
/// return this error.
2017-02-24 19:56:04 +00:00
///
/// # Examples
///
/// ```
/// use std::sync::{Arc, Mutex};
/// use std::thread;
///
/// let mutex = Arc::new(Mutex::new(1));
///
/// // poison the mutex
/// let c_mutex = mutex.clone();
/// let _ = thread::spawn(move || {
/// let mut data = c_mutex.lock().unwrap();
/// *data = 2;
/// panic!();
/// }).join();
///
/// match mutex.lock() {
/// Ok(_) => unreachable!(),
/// Err(p_err) => {
/// let data = p_err.get_ref();
/// println!("recovered: {}", data);
/// }
/// };
/// ```
///
/// [`Mutex`]: crate::sync::Mutex
/// [`RwLock`]: crate::sync::RwLock
2015-01-24 05:48:20 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
pub struct PoisonError<T> {
guard: T,
}
/// An enumeration of possible errors associated with a [`TryLockResult`] which
2018-02-16 14:56:50 +00:00
/// can occur while trying to acquire a lock, from the [`try_lock`] method on a
/// [`Mutex`] or the [`try_read`] and [`try_write`] methods on an [`RwLock`].
///
/// [`Mutex`]: crate::sync::Mutex
/// [`RwLock`]: crate::sync::RwLock
/// [`try_lock`]: crate::sync::Mutex::try_lock
/// [`try_read`]: crate::sync::RwLock::try_read
/// [`try_write`]: crate::sync::RwLock::try_write
2015-01-24 05:48:20 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
pub enum TryLockError<T> {
/// The lock could not be acquired because another thread failed while holding
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// the lock.
2015-01-24 05:48:20 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
2016-02-20 00:08:36 +00:00
Poisoned(#[stable(feature = "rust1", since = "1.0.0")] PoisonError<T>),
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// The lock could not be acquired at this time because the operation would
/// otherwise block.
2015-01-24 05:48:20 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
WouldBlock,
}
/// A type alias for the result of a lock method which can be poisoned.
///
2017-02-24 19:56:04 +00:00
/// The [`Ok`] variant of this result indicates that the primitive was not
/// poisoned, and the `Guard` is contained within. The [`Err`] variant indicates
/// that the primitive was poisoned. Note that the [`Err`] variant *also* carries
/// the associated guard, and it can be acquired through the [`into_inner`]
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// method.
2017-02-24 19:56:04 +00:00
///
/// [`into_inner`]: PoisonError::into_inner
2015-01-24 05:48:20 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
pub type LockResult<Guard> = Result<Guard, PoisonError<Guard>>;
/// A type alias for the result of a nonblocking locking method.
///
2017-02-24 19:56:04 +00:00
/// For more information, see [`LockResult`]. A `TryLockResult` doesn't
/// necessarily hold the associated guard in the [`Err`] type as the lock may not
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// have been acquired for other reasons.
2015-01-24 05:48:20 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
pub type TryLockResult<Guard> = Result<Guard, TryLockError<Guard>>;
#[stable(feature = "rust1", since = "1.0.0")]
impl<T> fmt::Debug for PoisonError<T> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
"PoisonError { inner: .. }".fmt(f)
}
}
#[stable(feature = "rust1", since = "1.0.0")]
std: Rename Show/String to Debug/Display This commit is an implementation of [RFC 565][rfc] which is a stabilization of the `std::fmt` module and the implementations of various formatting traits. Specifically, the following changes were performed: [rfc]: https://github.com/rust-lang/rfcs/blob/master/text/0565-show-string-guidelines.md * The `Show` trait is now deprecated, it was renamed to `Debug` * The `String` trait is now deprecated, it was renamed to `Display` * Many `Debug` and `Display` implementations were audited in accordance with the RFC and audited implementations now have the `#[stable]` attribute * Integers and floats no longer print a suffix * Smart pointers no longer print details that they are a smart pointer * Paths with `Debug` are now quoted and escape characters * The `unwrap` methods on `Result` now require `Display` instead of `Debug` * The `Error` trait no longer has a `detail` method and now requires that `Display` must be implemented. With the loss of `String`, this has moved into libcore. * `impl<E: Error> FromError<E> for Box<Error>` now exists * `derive(Show)` has been renamed to `derive(Debug)`. This is not currently warned about due to warnings being emitted on stage1+ While backwards compatibility is attempted to be maintained with a blanket implementation of `Display` for the old `String` trait (and the same for `Show`/`Debug`) this is still a breaking change due to primitives no longer implementing `String` as well as modifications such as `unwrap` and the `Error` trait. Most code is fairly straightforward to update with a rename or tweaks of method calls. [breaking-change] Closes #21436
2015-01-20 23:45:07 +00:00
impl<T> fmt::Display for PoisonError<T> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
"poisoned lock: another task failed inside".fmt(f)
}
}
2015-11-16 16:54:28 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
impl<T> Error for PoisonError<T> {
#[allow(deprecated)]
fn description(&self) -> &str {
"poisoned lock: another task failed inside"
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
}
}
impl<T> PoisonError<T> {
/// Creates a `PoisonError`.
2017-02-24 19:56:04 +00:00
///
/// This is generally created by methods like [`Mutex::lock`] or [`RwLock::read`].
///
/// [`Mutex::lock`]: crate::sync::Mutex::lock
/// [`RwLock::read`]: crate::sync::RwLock::read
#[stable(feature = "sync_poison", since = "1.2.0")]
2015-01-23 19:58:49 +00:00
pub fn new(guard: T) -> PoisonError<T> {
2018-11-06 20:05:44 +00:00
PoisonError { guard }
2015-01-23 19:58:49 +00:00
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
/// Consumes this error indicating that a lock is poisoned, returning the
/// underlying guard to allow access regardless.
///
/// # Examples
///
/// ```
/// use std::collections::HashSet;
/// use std::sync::{Arc, Mutex};
/// use std::thread;
///
/// let mutex = Arc::new(Mutex::new(HashSet::new()));
///
/// // poison the mutex
/// let c_mutex = mutex.clone();
/// let _ = thread::spawn(move || {
/// let mut data = c_mutex.lock().unwrap();
/// data.insert(10);
/// panic!();
/// }).join();
///
/// let p_err = mutex.lock().unwrap_err();
/// let data = p_err.into_inner();
/// println!("recovered {} items", data.len());
/// ```
#[stable(feature = "sync_poison", since = "1.2.0")]
pub fn into_inner(self) -> T {
self.guard
}
/// Reaches into this error indicating that a lock is poisoned, returning a
/// reference to the underlying guard to allow access regardless.
#[stable(feature = "sync_poison", since = "1.2.0")]
pub fn get_ref(&self) -> &T {
&self.guard
}
/// Reaches into this error indicating that a lock is poisoned, returning a
/// mutable reference to the underlying guard to allow access regardless.
#[stable(feature = "sync_poison", since = "1.2.0")]
pub fn get_mut(&mut self) -> &mut T {
&mut self.guard
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
}
2015-11-16 16:54:28 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
impl<T> From<PoisonError<T>> for TryLockError<T> {
fn from(err: PoisonError<T>) -> TryLockError<T> {
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
TryLockError::Poisoned(err)
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
}
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
impl<T> fmt::Debug for TryLockError<T> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
match *self {
TryLockError::Poisoned(..) => "Poisoned(..)".fmt(f),
TryLockError::WouldBlock => "WouldBlock".fmt(f),
}
}
}
#[stable(feature = "rust1", since = "1.0.0")]
impl<T> fmt::Display for TryLockError<T> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
match *self {
TryLockError::Poisoned(..) => "poisoned lock: another task failed inside",
TryLockError::WouldBlock => "try_lock failed because the operation would block",
}
.fmt(f)
}
}
2015-11-16 16:54:28 +00:00
#[stable(feature = "rust1", since = "1.0.0")]
impl<T> Error for TryLockError<T> {
#[allow(deprecated, deprecated_in_future)]
fn description(&self) -> &str {
match *self {
TryLockError::Poisoned(ref p) => p.description(),
TryLockError::WouldBlock => "try_lock failed because the operation would block",
}
}
#[allow(deprecated)]
fn cause(&self) -> Option<&dyn Error> {
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
match *self {
TryLockError::Poisoned(ref p) => Some(p),
_ => None,
std: Rewrite the `sync` module This commit is a reimplementation of `std::sync` to be based on the system-provided primitives wherever possible. The previous implementation was fundamentally built on top of channels, and as part of the runtime reform it has become clear that this is not the level of abstraction that the standard level should be providing. This rewrite aims to provide as thin of a shim as possible on top of the system primitives in order to make them safe. The overall interface of the `std::sync` module has in general not changed, but there are a few important distinctions, highlighted below: * The condition variable type, `Condvar`, has been separated out of a `Mutex`. A condition variable is now an entirely separate type. This separation benefits users who only use one mutex, and provides a clearer distinction of who's responsible for managing condition variables (the application). * All of `Condvar`, `Mutex`, and `RWLock` are now directly built on top of system primitives rather than using a custom implementation. The `Once`, `Barrier`, and `Semaphore` types are still built upon these abstractions of the system primitives. * The `Condvar`, `Mutex`, and `RWLock` types all have a new static type and constant initializer corresponding to them. These are provided primarily for C FFI interoperation, but are often useful to otherwise simply have a global lock. The types, however, will leak memory unless `destroy()` is called on them, which is clearly documented. * The `Condvar` implementation for an `RWLock` write lock has been removed. This may be added back in the future with a userspace implementation, but this commit is focused on exposing the system primitives first. * The fundamental architecture of this design is to provide two separate layers. The first layer is that exposed by `sys_common` which is a cross-platform bare-metal abstraction of the system synchronization primitives. No attempt is made at making this layer safe, and it is quite unsafe to use! It is currently not exported as part of the API of the standard library, but the stabilization of the `sys` module will ensure that these will be exposed in time. The purpose of this layer is to provide the core cross-platform abstractions if necessary to implementors. The second layer is the layer provided by `std::sync` which is intended to be the thinnest possible layer on top of `sys_common` which is entirely safe to use. There are a few concerns which need to be addressed when making these system primitives safe: * Once used, the OS primitives can never be **moved**. This means that they essentially need to have a stable address. The static primitives use `&'static self` to enforce this, and the non-static primitives all use a `Box` to provide this guarantee. * Poisoning is leveraged to ensure that invalid data is not accessible from other tasks after one has panicked. In addition to these overall blanket safety limitations, each primitive has a few restrictions of its own: * Mutexes and rwlocks can only be unlocked from the same thread that they were locked by. This is achieved through RAII lock guards which cannot be sent across threads. * Mutexes and rwlocks can only be unlocked if they were previously locked. This is achieved by not exposing an unlocking method. * A condition variable can only be waited on with a locked mutex. This is achieved by requiring a `MutexGuard` in the `wait()` method. * A condition variable cannot be used concurrently with more than one mutex. This is guaranteed by dynamically binding a condition variable to precisely one mutex for its entire lifecycle. This restriction may be able to be relaxed in the future (a mutex is unbound when no threads are waiting on the condvar), but for now it is sufficient to guarantee safety. * Condvars now support timeouts for their blocking operations. The implementation for these operations is provided by the system. Due to the modification of the `Condvar` API, removal of the `std::sync::mutex` API, and reimplementation, this is a breaking change. Most code should be fairly easy to port using the examples in the documentation of these primitives. [breaking-change] Closes #17094 Closes #18003
2014-11-24 19:16:40 +00:00
}
}
}
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
pub fn map_result<T, U, F>(result: LockResult<T>, f: F) -> LockResult<U>
where
F: FnOnce(T) -> U,
{
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
match result {
Ok(t) => Ok(f(t)),
Err(PoisonError { guard }) => Err(PoisonError::new(f(guard))),
std: Return Result from RWLock/Mutex methods All of the current std::sync primitives have poisoning enable which means that when a task fails inside of a write-access lock then all future attempts to acquire the lock will fail. This strategy ensures that stale data whose invariants are possibly not upheld are never viewed by other tasks to help propagate unexpected panics (bugs in a program) among tasks. Currently there is no way to test whether a mutex or rwlock is poisoned. One method would be to duplicate all the methods with a sister foo_catch function, for example. This pattern is, however, against our [error guidelines][errors]. As a result, this commit exposes the fact that a task has failed internally through the return value of a `Result`. [errors]: https://github.com/rust-lang/rfcs/blob/master/text/0236-error-conventions.md#do-not-provide-both-result-and-fail-variants All methods now return a `LockResult<T>` or a `TryLockResult<T>` which communicates whether the lock was poisoned or not. In a `LockResult`, both the `Ok` and `Err` variants contains the `MutexGuard<T>` that is being returned in order to allow access to the data if poisoning is not desired. This also means that the lock is *always* held upon returning from `.lock()`. A new type, `PoisonError`, was added with one method `into_guard` which can consume the assertion that a lock is poisoned to gain access to the underlying data. This is a breaking change because the signatures of these methods have changed, often incompatible ways. One major difference is that the `wait` methods on a condition variable now consume the guard and return it in as a `LockResult` to indicate whether the lock was poisoned while waiting. Most code can be updated by calling `.unwrap()` on the return value of `.lock()`. [breaking-change]
2014-12-09 04:20:03 +00:00
}
}