The topic you requested is included in another documentation set. For convenience, it's displayed below. Choose Switch to see the topic in its original location.

InterlockedIncrement16NoFence function

Increments (increases by one) the value of the specified 16-bit variable as an atomic operation. The operation is performed atomically, but without using memory barriers.

To operate on 32-bit values, use the InterlockedIncrementNoFence function.

To operate on 64-bit values, use the InterlockedIncrementNoFence64 function.

Syntax

C++


short __cdecl InterlockedIncrement16NoFence(
  _Inout_ short volatile *Addend
);

Parameters

Addend [in, out]: A pointer to the variable to be incremented.

Return value

The function returns the resulting incremented value.

Remarks

The variable pointed to by the Addend parameter must be aligned on a 16-bit boundary; otherwise, this function will behave unpredictably on multiprocessor x86 systems and any non-x86 systems. See _aligned_malloc.

The interlocked functions provide a simple mechanism for synchronizing access to a variable that is shared by multiple threads. This function is atomic with respect to calls to other interlocked functions.

This function is implemented using a compiler intrinsic where possible. For more information, see the WinBase.h header file and _InterlockedIncrement16_nf.

This function generates no memory barriers (or fences) and does not guarantee that independent memory operations before or after it are completed in order.

Note This function is supported on Windows RT-based systems.

Requirements

Minimum supported client	Windows 8 [desktop apps only]
Minimum supported server	Windows Server 2012 [desktop apps only]
Header	Winnt.h (include Windows.h)