https://arolstar52-zhtest.hf.space/index.php?action=history&feed=atom&title=Read-modify-write
Read-modify-write - 版本历史
2026-06-28T01:44:18Z
在这个wiki上该页的修订历史
MediaWiki 1.43.8
https://arolstar52-zhtest.hf.space/index.php?title=Read-modify-write&diff=2976568&oldid=prev
imported>Jimmy-bot：机器人: 尚未清空的已重定向分类（Cat:協同控制→Cat:并发控制）
2023-12-05T02:21:53Z
<p>机器人: 尚未清空的已重定向分类（<a href="/index.php?title=Cat:%E5%8D%94%E5%90%8C%E6%8E%A7%E5%88%B6&action=edit&redlink=1" class="new" title="Cat:協同控制（页面不存在）">Cat:協同控制</a>→<a href="/index.php?title=Cat:%E5%B9%B6%E5%8F%91%E6%8E%A7%E5%88%B6&action=edit&redlink=1" class="new" title="Cat:并发控制（页面不存在）">Cat:并发控制</a>）</p>
<p><b>新页面</b></p><div>{{noteTA<br />
|G1=IT<br />
}}<br />
<br />
'''读-修改-写'''（read-modify-write）是计算机科学中的一个原子操作（atomic operation，类似的还有[[test-and-set]], [[fetch-and-add]], [[compare-and-swap]]等），操作过程是读一个内存位置（或IO端口），修改其值，再写回原位置。<ref>[http://www.ddj.com/embedded/210604448?pgno=3 "Writing Lock-Free Code: A Corrected Queue"]. Dr. Dobb's Journal. {{Wayback|url=http://www.ddj.com/embedded/210604448?pgno=3 |date=20081120035340 }}<br />
by [[Herb Sutter]]: "Compare-and-swap (CAS) is ... widely available ... However, some systems instead provide the equivalently powerful load-linked/store-conditional (LL/SC) instead."</ref><ref>{{cite journal |last=Herlihy |first=Maurice |title=Wait-free synchronization |journal=ACM Transactions on Programming Languages and Systems |volume=13 |issue=1 |date=January 1991 |pages=124–149 |url=http://www.cs.brown.edu/~mph/Herlihy91/p124-herlihy.pdf |accessdate=2007-05-20 |doi=10.1145/114005.102808 |archive-date=2011-06-05 |archive-url=https://web.archive.org/web/20110605003208/http://www.cs.brown.edu/~mph/Herlihy91/p124-herlihy.pdf |dead-url=no }}</ref><br />
<br />
必须要先读操作的一个原因是，系统架构往往只允许字（word）级的读写，必须先读出那些不做修改的位元，保持不变再写回。写成[[C语言]]语句类似于：<br />
*pRegister |= SOME_BIT;<br />
<br />
Read-modify-write指令用于IO端口时，可能会产生出乎意料的结果，如无法给一个比特置位。这往往是因为写操作并不影响到读操作的源寄存器。<ref>{{Cite web |url=http://techref.massmind.org/techref/readmodwrite.htm |title=Massmind: "The read–modify–write problem" |accessdate=2017-11-22 |archive-date=2017-12-01 |archive-url=https://web.archive.org/web/20171201033821/http://techref.massmind.org/techref/readmodwrite.htm |dead-url=no| website = techref.massmind.org }}</ref><br />
<br />
[[RAID]]也使用这一术语描述原子操作的read-modify-write序列。<ref>{{cite web| url = http://www.ecs.umass.edu/ece/koren/architecture/Raid/basicRAID.html| title = Basic RAID Organizations | accessdate = 2013-10-04| website = umass.edu| archive-date = 2013-10-24| archive-url = https://web.archive.org/web/20131024014416/http://www.ecs.umass.edu/ece/koren/architecture/Raid/basicRAID.html| dead-url = no}}</ref> Such RAID levels include [[RAID&nbsp;4]], [[RAID&nbsp;5]] and [[RAID&nbsp;6]].<br />
<br />
== 参见 ==<br />
* {{tsl|en|Linearizability|可线性化}}<br />
* {{tsl|en|Read-erase-modify-write}}<br />
<br />
== 参考文献 ==<br />
{{Reflist}}<br />
<br />
<br />
[[Category:并发控制]]</div>
imported>Jimmy-bot