Work in progress This page is under construction

Experimental Feature The functionality described on this page is part of the Transactional Memory Technical Specification ISO/IEC TS 19841:2015(TM TS)

Transactional memory is a concurrency synchronization mechanism that combines groups of statements in transactions, that are

• atomic (either all statements occur, or nothing occurs)
• isolated (statements in a transaction may not observe half-written writes made by another transaction, even if they execute in parallel)

Typical implementations use hardware transactional memory where supported and to the limits that it is available (e.g. until the changeset is saturated) and fall back to software transactional memory, usually implemented with optimistic concurrency: if another transaction updated some of the variables used by a transaction, it is silently retried. For that reason, retriable transactions ("atomic blocks") can only call transaction-safe functions.

Note that accessing a variable in a transaction and out of a transaction without other external synchronization is a data race.

If feature testing is supported, the features described here are indicated by the macro constant__cpp_transactional_memory with a value equal or greater201505.

[edit]Synchronized blocks

synchronizedcompound-statement

Executes thecompound statement as if under a global lock: all outermost synchronized blocks in the program execute in a single total order. The end of each synchronized block synchronizes with the beginning of the next synchronized block in that order. Synchronized blocks that are nested within other synchronized blocks have no special semantics.

Synchronized blocks are not transactions (unlike the atomic blocks below) and may call transaction-unsafe functions.

#include <iostream>#include <thread>#include <vector> int f(){staticint i=0;    synchronized{// begin synchronized blockstd::cout<< i<<" -> ";++i;// each call to f() obtains a unique value of istd::cout<< i<<'\n';return i;// end synchronized block}} int main(){std::vector<std::thread> v(10);for(auto& t: v)        t=std::thread([]{for(int n=0; n<10;++n) f();});for(auto& t: v)        t.join();}

0 -> 11 -> 22 -> 3...99 -> 100

Leaving a synchronized block by any means (reaching the end, executing goto, break, continue, or return, or throwing an exception) exits the block and synchronizes-with the next block in the single total order if the exited block was an outer block. The behavior is undefined ifstd::longjmp is used to exit a synchronized block.

Entering a synchronized block by goto or switch is not allowed.

Although synchronized blocks execute as-if under a global lock, the implementations are expected to examine the code within each block and use optimistic concurrency (backed up by hardware transactional memory where available) for transaction-safe code and minimal locking for non-transaction safe code. When a synchronized block makes a call to a non-inlined function, the compiler may have to drop out of speculative execution and hold a lock around the entire call unless the function is declaredtransaction_safe (see below) or the attribute[[optimize_for_synchronized]] (see below) is used.

[edit]Atomic blocks

atomic_noexceptcompound-statement

atomic_cancelcompound-statement

atomic_commitcompound-statement

The exceptions used for transaction cancellation inatomic_cancel blocks arestd::bad_alloc,std::bad_array_new_length,std::bad_cast,std::bad_typeid,std::bad_exception,std::exception and all standard library exceptions derived from it, and the special exception typestd::tx_exception<T>.

Thecompound-statement in an atomic block is not allowed to execute any expression or statement or call any function that isn'ttransaction_safe (this is a compile time error).

// each call to f() retrieves a unique value of i, even when done in parallelint f(){staticint i=0;    atomic_noexcept{// begin transaction//  printf("before %d\n", i); // error: cannot call a non transaction-safe function++i;return i;// commit transaction}}

Leaving an atomic block by any means other than exception (reaching the end, goto, break, continue, return) commits the transaction. The behavior is undefined ifstd::longjmp is used to exit an atomic block.

[edit]Transaction-safe functions

A function can be explicitly declared to be transaction-safe by using the keywordtransaction_safe in its declaration.

In alambda declaration, it appears either immediately after the capture list, or immediately after the (keywordmutable (if one is used).

externvolatileint* p=0;struct S{virtual ~S();};int f() transaction_safe{int x=0;// ok: not volatile    p=&x;// ok: the pointer is not volatileint i=*p;// error: read through volatile glvalue    S s;// error: invocation of unsafe destructor}

int f(int x){// implicitly transaction-safeif(x<=0)return0;return x+ f(x-1);}

If a function that is not transaction-safe is called through a reference or pointer to a transaction-safe function, the behavior is undefined.

Pointers to transaction-safe functions and pointers to transaction-safe member functions are implicitly convertible to pointers to functions and pointers to member functions respectively. It is unspecified if the resulting pointer compares equal to the original.

[edit]Transaction-safe virtual functions

If the final overrider of atransaction_safe_dynamic function is not declaredtransaction_safe, calling it in an atomic block is undefined behavior.

[edit]Standard library

Besides introducing the new exception templatestd::tx_exception, the transactional memory technical specification makes the following changes to the standard library:

• makes the following functions explicitlytransaction_safe:

• std::forward,std::move,std::move_if_noexcept,std::align,std::abort, global defaultoperator new, global defaultoperator delete,std::allocator::construct if the invoked constructor is transaction-safe,std::allocator::destroy if the invoked destructor is transaction-safe,std::get_temporary_buffer,std::return_temporary_buffer,std::addressof,std::pointer_traits::pointer_to, each non-virtual member function of all exception types that support transaction cancellation (seeatomic_cancel above)

  This section is incomplete Reason: there's more

• makes the following functions explicitlytransaction_safe_dynamic

• each virtual member function of all exception types that support transaction cancellation (seeatomic_cancel above)

• requires that all operations that are transaction-safe on anAllocator X are transaction-safe onX::rebind<>::other

[edit]Attributes

The attribute[[optimize_for_synchronized]] may be applied to a declarator in a function declaration and must appear on the first declaration of the function.

If a function is declared[[optimize_for_synchronized]] in one translation unit and the same function is declared without[[optimize_for_synchronized]] in another translation unit, the program is ill-formed; no diagnostic required.

It indicates that a the function definition should be optimized for invocation from asynchronized statement. In particular, it avoids serializing synchronized blocks that make a call to a function that is transaction-safe for the majority of calls, but not for all calls (e.g. hash table insertion that may have to rehash, allocator that may have to request a new block, a simple function that may rarely log).

std::atomic<bool> rehash{false}; // maintenance thread runs this loopvoid maintenance_thread(void*){while(!shutdown){        synchronized{if(rehash){                hash.rehash();                rehash=false;}}}} // worker threads execute hundreds of thousands of calls to this function// every second. Calls to insert_key() from synchronized blocks in other// translation units will cause those blocks to serialize, unless insert_key()// is marked [[optimize_for_synchronized]][[optimize_for_synchronized]]void insert_key(char* key,char* value){bool concern= hash.insert(key, value);if(concern)        rehash=true;}

GCC assembly without the attribute: the entire function is serialized

insert_key(char*, char*):subq$8, %rspmovq%rsi, %rdxmovq%rdi, %rsimovl$hash, %edicallHash::insert(char*, char*)testb%al, %alje.L20movb$1, rehash(%rip)mfence.L20:addq$8, %rspret

GCC assembly with the attribute:

transaction clone for insert_key(char*, char*):subq$8, %rspmovq%rsi, %rdxmovq%rdi, %rsimovl$hash, %edicalltransaction clone for Hash::insert(char*, char*)testb%al, %alje.L27xorl%edi, %edicall_ITM_changeTransactionMode # Note: this is the serialization pointmovb$1, rehash(%rip)mfence.L27:addq$8, %rspret

[edit]Notes

[edit]Keywords

atomic_cancel,atomic_commit,atomic_noexcept,synchronized,transaction_safe,transaction_safe_dynamic

[edit]Compiler support

This technical specification is supported by GCC as of version 6.1 (requires-fgnu-tm to enable). An older variant of this specification wassupported in GCC as of 4.7.

• In other languages

• 日本語
• Русский
• 中文