CINXE.COM

<?xml version="1.0" encoding="UTF-8" standalone="yes"?> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>LKML: Nikita Danilov: Re: [PATCH 1/19] MUTEX: Introduce simple mutex implementation</title><link href="/css/message.css" rel="stylesheet" type="text/css" /><link href="/css/wrap.css" rel="alternate stylesheet" type="text/css" title="wrap" /><link href="/css/nowrap.css" rel="stylesheet" type="text/css" title="nowrap" /><link href="/favicon.ico" rel="shortcut icon" /><script src="/js/simple-calendar.js" type="text/javascript"></script><script src="/js/styleswitcher.js" type="text/javascript"></script><link rel="alternate" type="application/rss+xml" title="lkml.org : last 100 messages" href="/rss.php" /><link rel="alternate" type="application/rss+xml" title="lkml.org : last messages by Nikita Danilov" href="/groupie.php?aid=3176" /><script> var _paq = window._paq = window._paq || []; /* tracker methods like "setCustomDimension" should be called before "trackPageView" */ _paq.push(["setDoNotTrack", true]); _paq.push(["disableCookies"]); _paq.push(['trackPageView']); _paq.push(['enableLinkTracking']); (function() { var u="//m.lkml.org/"; _paq.push(['setTrackerUrl', u+'matomo.php']); _paq.push(['setSiteId', '1']); var d=document, g=d.createElement('script'), s=d.getElementsByTagName('script')[0]; g.async=true; g.src=u+'matomo.js'; s.parentNode.insertBefore(g,s); })(); </script></head><body onload="es.jasper.simpleCalendar.init();" itemscope="itemscope" itemtype="http://schema.org/BlogPosting"><table border="0" cellpadding="0" cellspacing="0"><tr><td width="180" align="center"><a href="/"><img style="border:0;width:135px;height:32px" src="/images/toprowlk.gif" alt="lkml.org" /></a></td><td width="32">聽</td><td class="nb"><div><a class="nb" href="/lkml"> [lkml]</a> 聽 <a class="nb" href="/lkml/2005"> [2005]</a> 聽 <a class="nb" href="/lkml/2005/12"> [Dec]</a> 聽 <a class="nb" href="/lkml/2005/12/15"> [15]</a> 聽 <a class="nb" href="/lkml/last100"> [last100]</a> 聽 <a href="/rss.php"><img src="/images/rss-or.gif" border="0" alt="RSS Feed" /></a></div><div>Views: <a href="#" class="nowrap" onclick="setActiveStyleSheet('wrap');return false;">[wrap]</a><a href="#" class="wrap" onclick="setActiveStyleSheet('nowrap');return false;">[no wrap]</a> 聽 <a class="nb" href="/lkml/mheaders/2005/12/15/269" onclick="this.href='/lkml/headers'+'/2005/12/15/269';">[headers]</a>聽 <a href="/lkml/bounce/2005/12/15/269">[forward]</a>聽 </div></td><td width="32">聽</td></tr><tr><td valign="top"><div class="es-jasper-simpleCalendar" baseurl="/lkml/"></div><div class="threadlist">Messages in this thread</div><ul class="threadlist"><li class="root"><a href="/lkml/2005/12/15/149">First message in thread</a></li><li><a href="/lkml/2005/12/15/149"> linux@horizon ...</a><ul><li><a href="/lkml/2005/12/15/201">Linus Torvalds</a><ul><li><a href="/lkml/2005/12/15/216">Erik Mouw</a><ul><li><a href="/lkml/2005/12/15/224">(Dick Streefland)</a></li><li><a href="/lkml/2005/12/16/61">Erik Mouw</a><ul><li><a href="/lkml/2005/12/17/22">Sander</a></li></ul></li></ul></li><li class="origin"><a href="">Nikita Danilov</a></li><li><a href="/lkml/2005/12/15/274"> linux@horizon ...</a><ul><li><a href="/lkml/2005/12/15/286">Linus Torvalds</a><ul><li><a href="/lkml/2005/12/15/422"> linux@horizon ...</a></li></ul></li><li><a href="/lkml/2005/12/15/315">Steven Rostedt</a></li></ul></li><li><a href="/lkml/2005/12/15/304">Steven Rostedt</a></li></ul></li></ul></li></ul></td><td width="32" rowspan="2" class="c" valign="top"><img src="/images/icornerl.gif" width="32" height="32" alt="/" /></td><td class="c" rowspan="2" valign="top" style="padding-top: 1em"><table><tr><td><table><tr><td class="lp">From</td><td class="rp" itemprop="author">Nikita Danilov <></td></tr><tr><td class="lp">Date</td><td class="rp" itemprop="datePublished">Thu, 15 Dec 2005 22:02:59 +0300</td></tr><tr><td class="lp">Subject</td><td class="rp" itemprop="name">Re: [PATCH 1/19] MUTEX: Introduce simple mutex implementation</td></tr></table></td><td></td></tr></table><pre itemprop="articleBody">Linus Torvalds writes:<br /> > <br /> > <br /> > On Thu, 15 Dec 2005, linux@horizon.com wrote:<br /> > > <br /> > > A counting semaphore is NOT a perfectly fine mutex, and it SHOULD be changed.<br /> > <br /> > Don't be silly.<br /> > <br /> > First off, the data structure is called a "semaphore", and always has <br /> > been. It's _never_ been called a "mutex" in the first place, and the <br /> > operations have been called "down()" and "up()", because I thought calling <br /> > them P() and V() was just too damn traditional and confusing (I don't <br /> > speak dutch, and even if I did, I think shortening names to that degree is <br /> > just evil).<br /> > <br /> > And dammit, a counting semaphore (and usually you don't even say the <br /> > "counting" part, since counting is really always there) is just about <br /> > _the_ classical mutual exclusion mechanism. If somebody doesn't know that, <br /> > he has absolutely _no_ place talking about mutexes etc.<br /><br />Dijkstra (that cannot talk about this due to much more serious reasons)<br />didn't know this, because semaphores were initially used as a<br />wait/signal mechanism to provide concurrency control between "process<br />context" and "interrupts" however they were called at the time, and<br />calling this "just mutual exclusion" is stretching a bit far.<br /><br />Mutex implies usage pattern much narrower than generic semaphore. <br /><br /> > <br /> > And a semaphore _is_ a mutex. <br /><br />Nope, a mutex is a semaphore and not other way around. For one thing, a<br />notion of ownership is well-defined for the mutex, but it is not for a<br />semaphore. This is what they call "sub-type".<br /><br /> > Anybody who disputes that is just being a <br /> > total troll. <br /><br />Oh wait... what is that thing on the right of my screen? This<br />is... gnome task-list!<br /><br />[...]<br /><br /> > <br /> > And I can't understand how somebody has the balls to even say that a <br /> > semaphore isn't a mutex. That's like saying that an object of type "long" <br /> > isn't an integer, because only "int" objects are integers. That's just <br /> > INSANE.<br /><br />And the person that claims that "long" is an "int" is non-portable. :-)<br /><br />[...]<br /><br /> > <br /> > Linus<br /><br />Nikita.<br />-<br />To unsubscribe from this list: send the line "unsubscribe linux-kernel" in<br />the body of a message to majordomo@vger.kernel.org<br />More majordomo info at <a href="http://vger.kernel.org/majordomo-info.html">http://vger.kernel.org/majordomo-info.html</a><br />Please read the FAQ at <a href="http://www.tux.org/lkml/">http://www.tux.org/lkml/</a><br /><br /></pre></td><td width="32" rowspan="2" class="c" valign="top"><img src="/images/icornerr.gif" width="32" height="32" alt="\" /></td></tr><tr><td align="right" valign="bottom"> 聽 </td></tr><tr><td align="right" valign="bottom">聽</td><td class="c" valign="bottom" style="padding-bottom: 0px"><img src="/images/bcornerl.gif" width="32" height="32" alt="\" /></td><td class="c">聽</td><td class="c" valign="bottom" style="padding-bottom: 0px"><img src="/images/bcornerr.gif" width="32" height="32" alt="/" /></td></tr><tr><td align="right" valign="top" colspan="2"> 聽 </td><td class="lm">Last update: 2005-12-15 20:05 聽聽 [from the cache]<br />漏2003-2020 <a href="http://blog.jasper.es/"><span itemprop="editor">Jasper Spaans</span></a>|hosted at <a href="https://www.digitalocean.com/?refcode=9a8e99d24cf9">Digital Ocean</a> and my Meterkast|<a href="http://blog.jasper.es/categories.html#lkml-ref">Read the blog</a></td><td>聽</td></tr></table><script language="javascript" src="/js/styleswitcher.js" type="text/javascript"></script></body></html>