Share via


Cache Support Using Streaming SIMD Extensions 

Microsoft Specific

The following intrinsics provide ways to make efficient use of the cache. For an explanation of the syntax used in code samples in this topic, see Floating-Point Intrinsics Using Streaming SIMD Extensions.

Intrinsic Description

_mm_prefetch

Loads one cache line of data from address p to a location closer to the processor.

_mm_stream_pi

Stores the data in a to the address p without polluting the caches.

_mm_stream_ps

Stores the data in a to the address p without polluting the caches.

_mm_sfence

Guarantees that every preceding store is globally visible before any subsequent store.

Requirements

Header: xmmintrin.h

END Microsoft Specific

See Also

Reference

Streaming SIMD Extensions (SSE)