.\" Copyright (c) 2000-2001 John H. Baldwin <jhb@FreeBSD.org>
.\" All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE DEVELOPERS ``AS IS'' AND ANY EXPRESS OR
.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
.\" IN NO EVENT SHALL THE DEVELOPERS BE LIABLE FOR ANY DIRECT, INDIRECT,
.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
.\"
.\" $FreeBSD$
.\"
.Dd August 20, 2013
.Dt ATOMIC 9
.Os
.Sh NAME
.Nm atomic_add ,
.Nm atomic_clear ,
.Nm atomic_cmpset ,
.Nm atomic_fetchadd ,
.Nm atomic_load ,
.Nm atomic_readandclear ,
.Nm atomic_set ,
.Nm atomic_subtract ,
.Nm atomic_store
.Nd atomic operations
.Sh SYNOPSIS
.In sys/types.h
.In machine/atomic.h
.Ft void
.Fn atomic_add_[acq_|rel_]<type> "volatile <type> *p" "<type> v"
.Ft void
.Fn atomic_clear_[acq_|rel_]<type> "volatile <type> *p" "<type> v"
.Ft int
.Fo atomic_cmpset_[acq_|rel_]<type>
.Fa "volatile <type> *dst"
.Fa "<type> old"
.Fa "<type> new"
.Fc
.Ft <type>
.Fn atomic_fetchadd_<type> "volatile <type> *p" "<type> v"
.Ft <type>
.Fn atomic_load_acq_<type> "volatile <type> *p"
.Ft <type>
.Fn atomic_readandclear_<type> "volatile <type> *p"
.Ft void
.Fn atomic_set_[acq_|rel_]<type> "volatile <type> *p" "<type> v"
.Ft void
.Fn atomic_subtract_[acq_|rel_]<type> "volatile <type> *p" "<type> v"
.Ft void
.Fn atomic_store_rel_<type> "volatile <type> *p" "<type> v"
.Ft <type>
.Fn atomic_swap_<type> "volatile <type> *p" "<type> v"
.Ft int
.Fn atomic_testandset_<type> "volatile <type> *p" "u_int v"
.Sh DESCRIPTION
Each of the atomic operations is guaranteed to be atomic in the presence of
interrupts.
They can be used to implement reference counts or as building blocks for more
advanced synchronization primitives such as mutexes.
.Ss Types
Each atomic operation operates on a specific
.Fa type .
The type to use is indicated in the function name.
The available types that can be used are:
.Pp
.Bl -tag -offset indent -width short -compact
.It Li int
unsigned integer
.It Li long
unsigned long integer
.It Li ptr
unsigned integer the size of a pointer
.It Li 32
unsigned 32-bit integer
.It Li 64
unsigned 64-bit integer
.El
.Pp
For example, the function to atomically add two integers is called
.Fn atomic_add_int .
.Pp
Certain architectures also provide operations for types smaller than
.Dq Li int .
.Pp
.Bl -tag -offset indent -width short -compact
.It Li char
unsigned character
.It Li short
unsigned short integer
.It Li 8
unsigned 8-bit integer
.It Li 16
unsigned 16-bit integer
.El
.Pp
These must not be used in MI code because the instructions to implement them
efficiently may not be available.
.Ss Memory Barriers
Memory barriers are used to guarantee the order of data accesses in
two ways.
First, they specify hints to the compiler to not re-order or optimize the
operations.
Second, on architectures that do not guarantee ordered data accesses,
special instructions or special variants of instructions are used to indicate
to the processor that data accesses need to occur in a certain order.
As a result, most of the atomic operations have three variants in order to
include optional memory barriers.
The first form just performs the operation without any explicit barriers.
The second form uses a read memory barrier, and the third variant uses a write
memory barrier.
.Pp
The second variant of each operation includes a read memory barrier.
This barrier ensures that the effects of this operation are completed before the
effects of any later data accesses.
As a result, the operation is said to have acquire semantics as it acquires a
pseudo-lock requiring further operations to wait until it has completed.
To denote this, the suffix
.Dq Li _acq
is inserted into the function name immediately prior to the
.Dq Li _ Ns Aq Fa type
suffix.
For example, to subtract two integers ensuring that any later writes will
happen after the subtraction is performed, use
.Fn atomic_subtract_acq_int .
.Pp
The third variant of each operation includes a write memory barrier.
This ensures that all effects of all previous data accesses are completed
before this operation takes place.
As a result, the operation is said to have release semantics as it releases
any pending data accesses to be completed before its operation is performed.
To denote this, the suffix
.Dq Li _rel
is inserted into the function name immediately prior to the
.Dq Li _ Ns Aq Fa type
suffix.
For example, to add two long integers ensuring that all previous
writes will happen first, use
.Fn atomic_add_rel_long .
.Pp
A practical example of using memory barriers is to ensure that data accesses
that are protected by a lock are all performed while the lock is held.
To achieve this, one would use a read barrier when acquiring the lock to
guarantee that the lock is held before any protected operations are performed.
Finally, one would use a write barrier when releasing the lock to ensure that
all of the protected operations are completed before the lock is released.
.Ss Multiple Processors
The current set of atomic operations do not necessarily guarantee atomicity
across multiple processors.
To guarantee atomicity across processors, not only does the individual
operation need to be atomic on the processor performing the operation, but
the result of the operation needs to be pushed out to stable storage and the
caches of all other processors on the system need to invalidate any cache
lines that include the affected memory region.
On the
.Tn i386
architecture, the cache coherency model requires that the hardware perform
this task, thus the atomic operations are atomic across multiple processors.
.Ss Semantics
This section describes the semantics of each operation using a C like notation.
.Bl -hang
.It Fn atomic_add p v
.Bd -literal -compact
*p += v;
.Ed
.It Fn atomic_clear p v
.Bd -literal -compact
*p &= ~v;
.Ed
.It Fn atomic_cmpset dst old new
.Bd -literal -compact
if (*dst == old) {
*dst = new;
return (1);
} else
return (0);
.Ed
.El
.Pp
The
.Fn atomic_cmpset
functions are not implemented for the types
.Dq Li char ,
.Dq Li short ,
.Dq Li 8 ,
and
.Dq Li 16 .
.Bl -hang
.It Fn atomic_fetchadd p v
.Bd -literal -compact
tmp = *p;
*p += v;
return (tmp);
.Ed
.El
.Pp
The
.Fn atomic_fetchadd
functions are only implemented for the types
.Dq Li int ,
.Dq Li long
and
.Dq Li 32
and do not have any variants with memory barriers at this time.
.Bl -hang
.It Fn atomic_load p
.Bd -literal -compact
return (*p);
.Ed
.El
.Pp
The
.Fn atomic_load
functions are only provided with acquire memory barriers.
.Bl -hang
.It Fn atomic_readandclear p
.Bd -literal -compact
tmp = *p;
*p = 0;
return (tmp);
.Ed
.El
.Pp
The
.Fn atomic_readandclear
functions are not implemented for the types
.Dq Li char ,
.Dq Li short ,
.Dq Li ptr ,
.Dq Li 8 ,
and
.Dq Li 16
and do not have any variants with memory barriers at this time.
.Bl -hang
.It Fn atomic_set p v
.Bd -literal -compact
*p |= v;
.Ed
.It Fn atomic_subtract p v
.Bd -literal -compact
*p -= v;
.Ed
.It Fn atomic_store p v
.Bd -literal -compact
*p = v;
.Ed
.El
.Pp
The
.Fn atomic_store
functions are only provided with release memory barriers.
.Bl -hang
.It Fn atomic_swap p v
.Bd -literal -compact
tmp = *p;
*p = v;
return (tmp);
.Ed
.El
.Pp
The
.Fn atomic_swap
functions are not implemented for the types
.Dq Li char ,
.Dq Li short ,
.Dq Li ptr ,
.Dq Li 8 ,
and
.Dq Li 16
and do not have any variants with memory barriers at this time.
.Bl -hang
.It Fn atomic_testandset p v
.Bd -literal -compact
bit = 1 << (v % (sizeof(*p) * NBBY));
tmp = (*p & bit) != 0;
*p |= bit;
return (tmp);
.Ed
.El
.Pp
The
.Fn atomic_testandset
functions are only implemented for the types
.Dq Li int ,
.Dq Li long
and
.Dq Li 32
and do not have any variants with memory barriers at this time.
.Pp
The type
.Dq Li 64
is currently not implemented for any of the atomic operations on the
.Tn arm ,
.Tn i386 ,
and
.Tn powerpc
architectures.
.Sh RETURN VALUES
The
.Fn atomic_cmpset
function returns the result of the compare operation.
The
.Fn atomic_fetchadd ,
.Fn atomic_load ,
.Fn atomic_readandclear ,
and
.Fn atomic_swap
functions return the value at the specified address.
The
.Fn atomic_testandset
function returns the result of the test operation.
.Sh EXAMPLES
This example uses the
.Fn atomic_cmpset_acq_ptr
and
.Fn atomic_set_ptr
functions to obtain a sleep mutex and handle recursion.
Since the
.Va mtx_lock
member of a
.Vt "struct mtx"
is a pointer, the
.Dq Li ptr
type is used.
.Bd -literal
/* Try to obtain mtx_lock once. */
#define _obtain_lock(mp, tid) \\
atomic_cmpset_acq_ptr(&(mp)->mtx_lock, MTX_UNOWNED, (tid))
/* Get a sleep lock, deal with recursion inline. */
#define _get_sleep_lock(mp, tid, opts, file, line) do { \\
uintptr_t _tid = (uintptr_t)(tid); \\
\\
if (!_obtain_lock(mp, tid)) { \\
if (((mp)->mtx_lock & MTX_FLAGMASK) != _tid) \\
_mtx_lock_sleep((mp), _tid, (opts), (file), (line));\\
else { \\
atomic_set_ptr(&(mp)->mtx_lock, MTX_RECURSE); \\
(mp)->mtx_recurse++; \\
} \\
} \\
} while (0)
.Ed
.Sh HISTORY
The
.Fn atomic_add ,
.Fn atomic_clear ,
.Fn atomic_set ,
and
.Fn atomic_subtract
operations were first introduced in
.Fx 3.0 .
This first set only supported the types
.Dq Li char ,
.Dq Li short ,
.Dq Li int ,
and
.Dq Li long .
The
.Fn atomic_cmpset ,
.Fn atomic_load ,
.Fn atomic_readandclear ,
and
.Fn atomic_store
operations were added in
.Fx 5.0 .
The types
.Dq Li 8 ,
.Dq Li 16 ,
.Dq Li 32 ,
.Dq Li 64 ,
and
.Dq Li ptr
and all of the acquire and release variants
were added in
.Fx 5.0
as well.
The
.Fn atomic_fetchadd
operations were added in
.Fx 6.0 .
The
.Fn atomic_swap
and
.Fn atomic_testandset
operations were added in
.Fx 10.0 .
