[apple/libc.git] / sys / barrier.3

.Dd May 26, 2004
.Dt BARRIER 3
.Os Darwin
.Sh NAME
.Nm OSMemoryBarrier
.Nd memory barrier to order loads and stores
.Sh LIBRARY
.Lb libc
.Sh SYNOPSIS
.In libkern/OSAtomic.h
.Ft void
.Fn OSMemoryBarrier "void"
.Sh DESCRIPTION
.Fn OSMemoryBarrier
strictly orders memory accesses in a weakly ordered memory model such as with PowerPC,
by creating a barrier.  All loads and stores executed in sequential program order before
the barrier will complete with respect to the memory coherence mechanism, before any
load or store executed after the barrier.  Used with an atomic operation, the barrier
can be used to create custom synchronization protocols as an alternative to the
spinlock or queue/dequeue operations.  Note that this barrier does not order uncached loads
and stores.  On a uniprocessor, the barrier operation is typically optimized into a nop.
.Sh SEE ALSO
.Xr atomic 3 ,
.Xr atomicqueue 3 ,
.Xr spinlock 3
Commit	Line	Data
59e0d9fe A	1	.Dd May 26, 2004
	2	.Dt BARRIER 3
	3	.Os Darwin
	4	.Sh NAME
	5	.Nm OSMemoryBarrier
	6	.Nd memory barrier to order loads and stores
	7	.Sh LIBRARY
	8	.Lb libc
	9	.Sh SYNOPSIS
	10	.In libkern/OSAtomic.h
	11	.Ft void
	12	.Fn OSMemoryBarrier "void"
	13	.Sh DESCRIPTION
	14	.Fn OSMemoryBarrier
	15	strictly orders memory accesses in a weakly ordered memory model such as with PowerPC,
	16	by creating a barrier. All loads and stores executed in sequential program order before
	17	the barrier will complete with respect to the memory coherence mechanism, before any
	18	load or store executed after the barrier. Used with an atomic operation, the barrier
	19	can be used to create custom synchronization protocols as an alternative to the
	20	spinlock or queue/dequeue operations. Note that this barrier does not order uncached loads
	21	and stores. On a uniprocessor, the barrier operation is typically optimized into a nop.
	22	.Sh SEE ALSO
	23	.Xr atomic 3 ,
	24	.Xr atomicqueue 3 ,
	25	.Xr spinlock 3