Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 27 Sep 2015 04:25:07 +0000 (UTC)
From:      Alan Cox <alc@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-10@freebsd.org
Subject:   svn commit: r288293 - stable/10/share/man/man9
Message-ID:  <201509270425.t8R4P7tY070733@repo.freebsd.org>

next in thread | raw e-mail | index | archive | help
Author: alc
Date: Sun Sep 27 04:25:07 2015
New Revision: 288293
URL: https://svnweb.freebsd.org/changeset/base/288293

Log:
  MFC r286513, r286784
    Revise the text about the atomicity of the defined operations across
    multiple processors.  In particular, clearly state that the operations
    are always atomic when they are applied to the default memory type
    that is used by the kernel (and applications).
  
    Stop describing an acquire operation as a read barrier and a release
    operation as a write barrier.  That description has never been correct,
    and it has caused confusion.
  
    Also, explicitly say that a thread doesn't see its own accesses being
    reordered.  The reordering of a thread's accesses is only (potentially)
    visible to another thread.

Modified:
  stable/10/share/man/man9/atomic.9
Directory Properties:
  stable/10/   (props changed)

Modified: stable/10/share/man/man9/atomic.9
==============================================================================
--- stable/10/share/man/man9/atomic.9	Sun Sep 27 04:17:08 2015	(r288292)
+++ stable/10/share/man/man9/atomic.9	Sun Sep 27 04:25:07 2015	(r288293)
@@ -23,7 +23,7 @@
 .\"
 .\" $FreeBSD$
 .\"
-.Dd June 20, 2015
+.Dd August 14, 2015
 .Dt ATOMIC 9
 .Os
 .Sh NAME
@@ -67,8 +67,8 @@
 .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.
+Each of the atomic operations is guaranteed to be atomic across multiple
+threads and 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
@@ -108,71 +108,94 @@ 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 an
+efficiently might not be available.
+.Ss Acquire and Release Operations
+By default, a thread's accesses to different memory locations might not be
+performed in
+.Em program order ,
+that is, the order in which the accesses appear in the source code.
+To optimize the program's execution, both the compiler and processor might
+reorder the thread's accesses.
+However, both ensure that their reordering of the accesses is not visible to
+the thread.
+Otherwise, the traditional memory model that is expected by single-threaded
+programs would be violated.
+Nonetheless, other threads in a multithreaded program, such as the
+.Fx
+kernel, might observe the reordering.
+Moreover, in some cases, such as the implementation of synchronization between
+threads, arbitrary reordering might result in the incorrect execution of the
+program.
+To constrain the reordering that both the compiler and processor might perform
+on a thread's accesses, the thread should use atomic operations with
 .Em acquire
-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
+and
+.Em release
+semantics.
+.Pp
+Most of the atomic operations on memory have three variants.
+The first variant performs the operation without imposing any ordering
+constraints on memory accesses to other locations.
+The second variant has acquire semantics, and the third variant has release
+semantics.
+In effect, operations with acquire and release semantics establish one-way
+barriers to reordering.
+.Pp
+When an atomic operation has acquire semantics, the effects of the operation
+must have completed before any subsequent load or store (by program order) is
+performed.
+Conversely, acquire semantics do not require that prior loads or stores have
+completed before the atomic operation is performed.
+To denote acquire semantics, 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
+For example, to subtract two integers ensuring that subsequent loads and
+stores happen after the subtraction is performed, use
 .Fn atomic_subtract_acq_int .
 .Pp
-The third variant of each operation includes a
-.Em release
-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
+When an atomic operation has release semantics, the effects of all prior
+loads or stores (by program order) must have completed before the operation
+is performed.
+Conversely, release semantics do not require that the effects of the
+atomic operation must have completed before any subsequent load or store is
+performed.
+To denote release semantics, 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
+For example, to add two long integers ensuring that all prior loads and
+stores happen before the addition, 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.
+The one-way barriers provided by acquire and release operations allow the
+implementations of common synchronization primitives to express their
+ordering requirements without also imposing unnecessary ordering.
+For example, for a critical section guarded by a mutex, an acquire operation
+when the mutex is locked and a release operation when the mutex is unlocked
+will prevent any loads or stores from moving outside of the critical
+section.
+However, they will not prevent the compiler or processor from moving loads
+or stores into the critical section, which does not violate the semantics of
+a mutex.
 .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
+In multiprocessor systems, the atomicity of the atomic operations on memory
+depends on support for cache coherence in the underlying architecture.
+In general, cache coherence on the default memory type,
+.Dv VM_MEMATTR_DEFAULT ,
+is guaranteed by all architectures that are supported by
+.Fx .
+For example, cache coherence is guaranteed on write-back memory by the
+.Tn amd64
+and
 .Tn i386
-architecture, the cache coherency model requires that the hardware perform
-this task, thus the atomic operations are atomic across multiple processors.
+architectures.
+However, on some architectures, cache coherence might not be enabled on all
+memory types.
+To determine if cache coherence is enabled for a non-default memory type,
+consult the architecture's documentation.
 On the
 .Tn ia64
 architecture, coherency is only guaranteed for pages that are configured to



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201509270425.t8R4P7tY070733>