ck_pr_fence_atomic(3)

Usage

#include <ck_pr.h>

void

ck_pr_fence_atomic(void);

void

ck_pr_fence_strict_atomic(void);

Library

Concurrency Kit (libck, -lck)

Overview

The ck_pr_fence_atomic() function enforces the ordering of any atomic read-modify-write operations relative to the invocation of the function. This function always serve as an implicit compiler barrier. On architectures implementing CK_MD_TSO, this operation only serves as a compiler barrier and no fences are emitted. On architectures implementing CK_MD_PSO and CK_MD_RMO, a store fence is emitted. To force the unconditional emission of a fence, use ck_pr_fence_strict_atomic().

Examples

#include <ck_pr.h>
static int a = 0;
static int b = 0;
static int c = 0;
void
function(void)
{
	ck_pr_fas_int(&a, 1);
	/*
	 * Guarantee that the update to a is completed
	 * with respect to the updates of b and c.
	 */
	ck_pr_fence_atomic();
	ck_pr_fas_int(&b, 2);
	ck_pr_fas_int(&c, 2);
	return;
}

Return values

This function has no return value.

Releases

5 releases since Debian 8 (Jessie); the text changed 2 times.

ck_pr_stall(3) busy-wait primitive ck_pr_fence_atomic_store(3) enforce ordering of atomic read-modify-write operations to store operations ck_pr_fence_atomic_load(3) enforce ordering of atomic read-modify-write operations to load operations ck_pr_fence_store(3) enforce partial ordering of store operations ck_pr_fence_load(3) enforce partial ordering of load operations ck_pr_fence_load_atomic(3) enforce ordering of load operations to atomic read-modify-write operations ck_pr_fence_load_store(3) enforce ordering of load operations to store operations ck_pr_fence_load_depends(3) data dependency barrier ck_pr_fence_memory(3) enforce partial ordering of all memory operations ck_pr_barrier(3) compiler optimization barrier ck_pr_fas(3) atomic swap operations ck_pr_load(3) atomic volatile load operations ck_pr_store(3) atomic volatile store operations ck_pr_faa(3) atomic fetch-and-add operations ck_pr_inc(3) atomic increment operations ck_pr_dec(3) atomic decrement operations ck_pr_neg(3) atomic negation operations ck_pr_not(3) atomic complement operations ck_pr_add(3) atomic addition operations ck_pr_sub(3) atomic subtraction operations ck_pr_and(3) atomic bitwise-and operations ck_pr_or(3) atomic bitwise-or operations ck_pr_xor(3) atomic bitwise-xor operations ck_pr_cas(3) atomic compare-and-swap operations ck_pr_btc(3) atomic bit test-and-complement operations ck_pr_bts(3) atomic bit test-and-set operations ck_pr_btr(3) atomic bit test-and-reset operations

Referred to by

ck_pr_fence_load(3) enforce partial ordering of load operations ck_pr_fence_store(3) enforce partial ordering of store operations ck_pr_fence_memory(3) enforce partial ordering of all memory operations ck_pr_fence_acquire(3) enforce acquire semantics ck_pr_fence_release(3) enforce release semantics ck_pr_fence_load_store(3) enforce ordering of load operations to store operations ck_pr_fence_store_load(3) enforce ordering of store operations to load operations ck_pr_fence_atomic_load(3) enforce ordering of atomic read-modify-write operations to load operations ck_pr_fence_load_atomic(3) enforce ordering of load operations to atomic read-modify-write operations ck_pr_fence_atomic_store(3) enforce ordering of atomic read-modify-write operations to store operations ck_pr_fence_load_depends(3) data dependency barrier ck_pr_fence_store_atomic(3) enforce ordering of store operations to load operations