Namespaces
Variants
Views
Actions

std::memory_order

From cppreference.com
< cpp‎ | atomic
Defined in header <atomic>
enum memory_order {

    memory_order_relaxed,
    memory_order_consume,
    memory_order_acquire,
    memory_order_release,
    memory_order_acq_rel,
    memory_order_seq_cst

};
(since C++11)

std::memory_order specifies how regular, non-atomic memory accesses are to be ordered around an atomic operation. Absent any constraints on a multi-core system, when multiple threads simultaneously read and write to several variables, one thread can observe the values change in an order different from the order another thread wrote them. Indeed, the apparent order of changes can even differ among multiple reader threads.

The default behavior of the atomic operations in the library provides for sequentially consistent ordering (see discussion below). That default can hurt performance, but the library's atomic operations can be given an additional std::memory_order argument to specify the exact constraints, beyond atomicity, that the compiler and processor must enforce for that operation.

Contents

[edit] Constants

Defined in header <atomic>
Value Explanation
memory_order_relaxed Relaxed ordering: there are no synchronization or ordering constraints, only atomicity is required of this operation.
memory_order_consume A load operation with this memory order performs a consume operation on the affected memory location: prior writes to data-dependent memory locations made by the thread that did a release operation become visible to this thread's dependency chain.
memory_order_acquire A load operation with this memory order performs the acquire operation on the affected memory location: prior writes made to other memory locations by the thread that did the release become visible in this thread.
memory_order_release A store operation with this memory order performs the release operation: prior writes to other memory locations become visible to the threads that do a consume or an acquire on the same location.
memory_order_acq_rel A load operation with this memory order performs the acquire operation on the affected memory location and a store operation with this memory order performs the release operation.
memory_order_seq_cst Same as memory_order_acq_rel, plus a single total order exists in which all threads observe all modifications (see below) in the same order.

[edit] Formal description

Inter-thread synchronization and memory ordering determine how evaluations and side effects of expressions are ordered between different threads of execution. They are defined in the following terms:

[edit] Sequenced-before

Within the same thread, evaluation A may be sequenced-before evaluation B, as described in evaluation order.

[edit] Carries dependency

Within the same thread, evaluation A that is sequenced-before evaluation B may also carry a dependency into B (that is, B depends on A), if any of the following is true

1) The value of A is used as an operand of B, except
a) if B is a call to std::kill_dependency
b) if A is the left operand of the built-in &&, ||, ?:, or , operators.
2) A writes to a scalar object M, B reads from M, and A is sequenced-before B
3) A carries dependency into another evaluation X, and X carries dependency into B

[edit] Modification order

All modifications to any particular atomic variable occur in a total order that is specific to this one atomic variable.

The following three requirements are guaranteed for all atomic operations:

1) Write-write coherence: If evaluation A that modifies some atomic M (a write) happens-before evaluation B that modifies M, then A appears earlier than B in the modification order of M
2) Read-read coherence: if a value computation A of some atomic M (a read) happens-before a value computation B on M, and if the value of A comes from a write X on M, then the value of B is either the value stored by X, or the value stored by a side effect Y on M that appears later than X in the modification order of M.
3) Read-write coherence: if a value computation A of some atomic M (a read) happens-before an operation B on M (a write), then the value of A comes from a side-effect (a write) X that appears earlier than B in the modification order of M

[edit] Release sequence

After a release operation A is performed on an atomic object M, the longest continuous subsequence of the modification order of M that consists of

1) Writes performed by the same thread that performed A
2) Atomic read-modify-write operations made to M by any thread

is known as release sequence headed by A

[edit] Dependency-ordered before

Between threads, evaluation A is dependency-ordered before evaluation B if any of the following is true

1) A performs a release operation on some atomic M, and, in a different thread, B performs a consume operation on the same atomic M, and B reads a value written by any part of the release sequence headed by A.
2) A is dependency-ordered before X and X carries a dependency into B.

[edit] Inter-thread happens-before

Between threads, evaluation A inter-thread happens before evaluation B if any of the following is true

1) A synchronizes-with B
2) A is dependency-ordered before B
3) A synchronizes-with some evaluation X, and X is sequenced-before B
3) A is sequenced-before some evaluation X, and X inter-thread happens-before B
4) A inter-thread happens-before some evaluation X, and X inter-thread happens-before B

[edit] Happens-before

Regardless of threads, evaluation A happens-before evaluation B if any of the following is true:

1) A is sequenced-before B
2) A inter-thread happens before B

If one evaluation modifies a memory location, and the other reads or modifies the same memory location, and if at least one of the evaluations is not an atomic operation, the behavior of the program is undefined (the program has a data race) unless there exists a happens-before relationship between these two evaluations.

[edit] Visible side-effects

The side-effect A on a scalar M (a write) is visible with respect to value computation B on M (a read) if both of the following are true:

1) A happens-before B
2) There is no other side effect X to M where A happens-before X and X happens-before B

If side-effect A is visible with respect to the value computation B, then the longest contiguous subset of the side-effects to M, in modification order, where B does not happen-before it is known as the visible sequence of side-effects. (the value of M, determined by B, will be the value stored by one of these side effects)

Note: inter-thread synchronization boils down to defining which side effects become visible under what conditions

[edit] Consume operation

Atomic load with memory_order_consume or stronger is a consume operation. Note that std::atomic_thread_fence is not a consume operation.

[edit] Acquire operation

Atomic load with memory_order_acquire or stronger is an acquire operation. The lock() operation on a Mutex is also an acquire operation. Note that std::atomic_thread_fence is not an acquire operation.

[edit] Release operation

Atomic store with memory_order_release or stronger is a release operation. The unlock() operation on a Mutex is also a release operation. Note that std::atomic_thread_fence is not a release operation.

[edit] Explanation

[edit] Relaxed ordering

Atomic operations tagged std::memory_order_relaxed are not synchronization operations, they do not order memory. They only guarantee atomicity and modification order consistency.

For example, with x and y initially zero,

// Thread 1:
r1 = y.load(memory_order_relaxed); // A
x.store(r1, memory_order_relaxed); // B
// Thread 2:
r2 = x.load(memory_order_relaxed); // C 
y.store(42, memory_order_relaxed); // D

is allowed to produce r1 == r2 == 42 because, although A is sequenced-before B and C is sequenced before D, nothing prevents D from appearing before A in the modification order of y, and B from appearing before C in the modification order of x.

Typical use for relaxed memory ordering is updating counters, such as the reference counters of std::shared_ptr, since this only requires atomicity, but not ordering or synchronization.

#include <vector>
#include <iostream>
#include <thread>
#include <atomic>
 
std::atomic<int> cnt = {0};
 
void f()
{
    for (int n = 0; n < 1000; ++n) {
        cnt.fetch_add(1, std::memory_order_relaxed);
    }
}
 
int main()
{
    std::vector<std::thread> v;
    for (int n = 0; n < 10; ++n) {
        v.emplace_back(f);
    }
    for (auto& t : v) {
        t.join();
    }
    std::cout << "Final counter value is " << cnt << '\n';
}

Output:

Final counter value is 10000

[edit] Release-Acquire ordering

If an atomic store in thread A is tagged std::memory_order_release and an atomic load in thread B from the same variable is tagged std::memory_order_acquire, all memory writes (non-atomic and relaxed atomic) that happened-before the atomic store from the point of view of thread A, become visible side-effects in thread B, that is, once the atomic load is completed, thread B is guaranteed to see everything thread A wrote to memory.

The synchronization is established only between the threads releasing and acquiring the same atomic variable. Other threads can see different order of memory accesses than either or both of the synchronized threads.

On strongly-ordered systems (x86, SPARC TSO, IBM mainframe), release-acquire ordering is automatic for the majority of operations. No additional CPU instructions are issued for this synchronization mode, only certain compiler optimizations are affected (e.g. the compiler is prohibited from moving non-atomic stores past the atomic store-release or perform non-atomic loads earlier than the atomic load-acquire). On weakly-ordered systems (ARM, Itanium, PowerPC), special CPU load or memory fence instructions have to be used.

Mutual exclusion locks (such as std::mutex or atomic spinlock) are an example of release-acquire synchronization: when the lock is released by thread A and acquired by thread B, everything that took place in the critical section (before the release) in the context of thread A has to be visible to thread B (after the acquire) which is executing the same critical section.

#include <thread>
#include <atomic>
#include <cassert>
#include <string>
 
std::atomic<std::string*> ptr;
int data;
 
void producer()
{
    std::string* p  = new std::string("Hello");
    data = 42;
    ptr.store(p, std::memory_order_release);
}
 
void consumer()
{
    std::string* p2;
    while (!(p2 = ptr.load(std::memory_order_acquire)))
        ;
    assert(*p2 == "Hello"); // never fires
    assert(data == 42); // never fires
}
 
int main()
{
    std::thread t1(producer);
    std::thread t2(consumer);
    t1.join(); t2.join();
}


The following example demonstrates transitive release-acquire ordering across three threads

#include <thread>
#include <atomic>
#include <cassert>
#include <vector>
 
std::vector<int> data;
std::atomic<int> flag = {0};
 
void thread_1()
{
    data.push_back(42);
    flag.store(1, std::memory_order_release);
}
 
void thread_2()
{
    int expected=1;
    while (!flag.compare_exchange_strong(expected, 2, std::memory_order_acq_rel)) {
        expected = 1;
    }
}
 
void thread_3()
{
    while (flag.load(std::memory_order_acquire) < 2)
        ;
    assert(data.at(0) == 42); // will never fire
}
 
int main()
{
    std::thread a(thread_1);
    std::thread b(thread_2);
    std::thread c(thread_3);
    a.join(); b.join(); c.join();
}



[edit] Release-Consume ordering

If an atomic store in thread A is tagged std::memory_order_release and an atomic load in thread B from the same variable is tagged std::memory_order_consume, all memory writes (non-atomic and relaxed atomic) that are dependency-ordered-before the atomic store from the point of view of thread A, become visible side-effects within those operations in thread B into which the load operation carries dependency, that is, once the atomic load is completed, those operators and functions in thread B that use the value obtained from the load are guaranteed to see what thread A wrote to memory.

The synchronization is established only between the threads releasing and acquiring the same atomic variable. Other threads can see different order of memory accesses than either or both of the synchronized threads.

On all mainstream CPUs other than DEC Alpha, dependency ordering is automatic, no additional CPU instructions are issued for this synchronization mode, only certain compiler optimizations are affected (e.g. the compiler is prohibited from performing speculative loads on the objects that are involved in the dependency chain).

Typical use cases for this ordering involve read access to rarely written concurrent data structures (routing tables, configuration, security policies, firewall rules, etc) and publisher-subscriber situations with pointer-mediated publication, that is, when the producer publishes a pointer through which the consumer can access information: there is no need to make everything else the producer wrote to memory visible to the consumer (which may be an expensive operation on weakly-ordered architectures). An example of such scenario is rcu_dereference.

See also std::kill_dependency and [[carries_dependency]] for fine-grained dependency chain control.

This example demonstrates dependency-ordered synchronization for pointer-mediated publication: the integer data is not related to the pointer to string by a data-dependency relationship, thus its value is undefined in the consumer.

#include <thread>
#include <atomic>
#include <cassert>
#include <string>
 
std::atomic<std::string*> ptr;
int data;
 
void producer()
{
    std::string* p  = new std::string("Hello");
    data = 42;
    ptr.store(p, std::memory_order_release);
}
 
void consumer()
{
    std::string* p2;
    while (!(p2 = ptr.load(std::memory_order_consume)))
        ;
    assert(*p2 == "Hello"); // never fires: *p2 carries dependency from ptr
    assert(data == 42); // may or may not fire: data does not carry dependency from ptr
}
 
int main()
{
    std::thread t1(producer);
    std::thread t2(consumer);
    t1.join(); t2.join();
}



[edit] Sequentially-consistent ordering

Atomic operations tagged std::memory_order_seq_cst not only order memory the same way as release/acquire ordering (everything that happened-before a store in one thread becomes a visible side effect in the thread that did a load), but also establish a single total modification order of all atomic operations that are so tagged.

Formally,

Each memory_order_seq_cst operation B that loads from atomic variable M, observes one of the following:

  • the result of the last operation A that modified M, which appears before B in the single total order
  • OR, if there was such an A, B may observe the result of some modification on M that is not memory_order_seq_cst and does not happen-before A
  • OR, if there wasn't such an A, B may observe the result of some unrelated modification of M that is not memory_order_seq_cst

If there was a memory_order_seq_cst std::atomic_thread_fence operation X sequenced-before B, then B observes one of the following:

  • the last memory_order_seq_cst modification of M that appears before X in the single total order
  • some unrelated modification of M that appears later in M's modification order

For a pair of atomic operations on M called A and B, where A writes and B reads M's value, if there are two memory_order_seq_cst std::atomic_thread_fences X and Y, and if A is sequenced-before X, Y is sequenced-before B, and X appears before Y in the Single Total Order, then B observes either:

  • the effect of A
  • some unrelated modification of M that appears after A in M's modification order

For a pair of atomic modifications of M called A and B, B occurs after A in M's modification order if

  • there is a memory_order_seq_cst std::atomic_thread_fence X such that A is sequenced-before X and X appears before B in the Single Total Order
  • or, there is a memory_order_seq_cst std::atomic_thread_fence Y such that Y is sequenced-before B and A appears before Y in the Single Total Order
  • or, there are memory_order_seq_cst std::atomic_thread_fences X and Y such that A is sequenced-before X, Y is sequenced-before B, and X appears before Y in the Single Total Order.

Note that the means that:

1) as soon as atomic operations that are not tagged memory_order_seq_cst enter the picture, the sequential consistency is lost
2) the sequentially-consistent fences are only establishing total ordering for the fences themselves, not for the atomic operations in the general case (sequenced-before is not a cross-thread relationship, unlike happens-before)

Sequential ordering may be necessary for multiple producer-multiple consumer situations where all consumers must observe the actions of all producers occurring in the same order.

Total sequential ordering requires a full memory fence CPU instruction on all multi-core systems. This may become a performance bottleneck since it forces the affected memory accesses to propagate to every core.

This example demonstrates a situation where sequential ordering is necessary. Any other ordering may trigger the assert because it would be possible for the threads c and d to observe changes to the atomics x and y in opposite order.

#include <thread>
#include <atomic>
#include <cassert>
 
std::atomic<bool> x = {false};
std::atomic<bool> y = {false};
std::atomic<int> z = {0};
 
void write_x()
{
    x.store(true, std::memory_order_seq_cst);
}
 
void write_y()
{
    y.store(true, std::memory_order_seq_cst);
}
 
void read_x_then_y()
{
    while (!x.load(std::memory_order_seq_cst))
        ;
    if (y.load(std::memory_order_seq_cst)) {
        ++z;
    }
}
 
void read_y_then_x()
{
    while (!y.load(std::memory_order_seq_cst))
        ;
    if (x.load(std::memory_order_seq_cst)) {
        ++z;
    }
}
 
int main()
{
    std::thread a(write_x);
    std::thread b(write_y);
    std::thread c(read_x_then_y);
    std::thread d(read_y_then_x);
    a.join(); b.join(); c.join(); d.join();
    assert(z.load() != 0);  // will never happen
}


[edit] Relationship with volatile

Within a thread of execution, accesses (reads and writes) to all volatile objects are guaranteed to not be reordered relative to each other, but this order is not guaranteed to be observed by another thread, since volatile access does not establish inter-thread synchronization.

In addition, volatile accesses are not atomic (concurrent read and write is a data race) and do not order memory (non-volatile memory accesses may be freely reordered around the volatile access).

One notable exception is Visual Studio, where, with default settings, every volatile write has release semantics and every volatile read has acquire semantics (MSDN), and thus volatiles may be used for inter-thread synchronization. Standard volatile semantics are not applicable to multithreaded programming, although they are sufficient for e.g. communication with a signal handler (see also std::atomic_signal_fence)