_InterlockedExchangeAdd Intrinsic Functions

Article
12/18/2017

The latest version of this topic can be found at _InterlockedExchangeAdd Intrinsic Functions.

Microsoft Specific**

Provide compiler intrinsic support for the Win32 Windows SDK _InterlockedExchangeAdd Intrinsic Functions function.

Syntax

long _InterlockedExchangeAdd(  
   long volatile * Addend,  
   long Value  
);  
long _InterlockedExchangeAdd_acq(  
   long volatile * Addend,  
   long Value  
);  
long _InterlockedExchangeAdd_rel(  
   long volatile * Addend,  
   long Value  
);  
long _InterlockedExchangeAdd_nf(  
   long volatile * Addend,  
   long Value  
);  
long _InterlockedExchangeAdd_HLEAcquire(  
   long volatile * Addend,  
   long Value  
);  
long _InterlockedExchangeAdd_HLERelease(  
   long volatile * Addend,  
   long Value  
);  
char _InterlockedExchangeAdd8(  
   char volatile * Addend,  
   char Value  
);  
char _InterlockedExchangeAdd8_acq(  
   char volatile * Addend,  
   char Value  
);  
char _InterlockedExchangeAdd8_rel(  
   char volatile * Addend,  
   char Value  
);  
char _InterlockedExchangeAdd8_nf(  
   char volatile * Addend,  
   char Value  
);  
short _InterlockedExchangeAdd16(  
   short volatile * Addend,  
   short Value  
);  
short _InterlockedExchangeAdd16_acq(  
   short volatile * Addend,  
   short Value  
);  
short _InterlockedExchangeAdd16_rel(  
   short volatile * Addend,  
   short Value  
);  
short _InterlockedExchangeAdd16_nf(  
   short volatile * Addend,  
   short Value  
);  
__int64 _InterlockedExchangeAdd64(  
   __int64 volatile * Addend,  
   __int64 Value  
);  
__int64 _InterlockedExchangeAdd64_acq(  
   __int64 volatile * Addend,  
   __int64 Value  
);  
__int64 _InterlockedExchangeAdd64_rel(  
   __int64 volatile * Addend,  
   __int64 Value  
);  
__int64 _InterlockedExchangeAdd64_nf(  
   __int64 volatile * Addend,  
   __int64 Value  
);  
__int64 _InterlockedExchangeAdd64_HLEAcquire(  
   __int64 volatile * Addend,  
   __int64 Value  
);  
__int64 _InterlockedExchangeAdd64_HLERelease(  
   __int64 volatile * Addend,  
   __int64 Value  
);

Parameters

[in, out] Addend
The value to be added to; replaced by the result of the addition.

[in] Value
The value to add.

Return Value

The return value is the initial value of the variable pointed to by the Addend parameter.

Requirements

Intrinsic	Architecture	Header
`_InterlockedExchangeAdd`, `_InterlockedExchangeAdd8`, `_InterlockedExchangeAdd16`, `_InterlockedExchangeAdd64`	x86, ARM, x64	<intrin.h>
`_InterlockedExchangeAdd_acq`, `_InterlockedExchangeAdd_rel`, `_InterlockedExchangeAdd_nf`, `_InterlockedExchangeAdd8_acq`, `_InterlockedExchangeAdd8_rel`, `_InterlockedExchangeAdd8_nf`,`_InterlockedExchangeAdd16_acq`, `_InterlockedExchangeAdd16_rel`, `_InterlockedExchangeAdd16_nf`, `_InterlockedExchangeAdd64_acq`, `_InterlockedExchangeAdd64_rel`, `_InterlockedExchangeAdd64_nf`	ARM	<intrin.h>
`_InterlockedExchangeAdd_HLEAcquire`, `_InterlockedExchangeAdd_HLERelease`, `_InterlockedExchangeAdd64_HLEAcquire`, `_InterlockedExchangeAdd64_HLErelease`	x86, x64	<immintrin.h>

Remarks

There are several variations on _InterlockedExchangeAdd that vary based on the data types they involve and whether processor-specific acquire or release semantics is used.

While the _InterlockedExchangeAdd function operates on 32-bit integer values, _InterlockedExchangeAdd8 operates on 8-bit integer values, _InterlockedExchangeAdd16 operates on 16-bit integer values, and _InterlockedExchangeAdd64 operates on 64-bit integer values.

On ARM platforms, use the intrinsics with _acq and _rel suffixes if you need acquire and release semantics, such as at the beginning and end of a critical section. The intrinsics with an _nf ("no fence") suffix do not act as a memory barrier.

On Intel platforms that support Hardware Lock Elision (HLE) instructions, the intrinsics with _HLEAcquire and _HLERelease suffixes include a hint to the processor that can accelerate performance by eliminating a lock write step in hardware. If these intrinsics are called on platforms that do not support HLE, the hint is ignored.

These routines are only available as intrinsics. Thus, they are intrinsic whether or not /Oi or #pragma intrinsic is used. It is not possible to use #pragma function on these intrinsics.

Example

For a sample of how to use _InterlockedExchangeAdd, see _InterlockedDecrement.

Partager via