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: Al Viro: Re: Kernel SCM saga..</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 Al Viro" href="/groupie.php?aid=24619" /><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/2005/4"> [Apr]</a> 聽 <a class="nb" href="/lkml/2005/4/7"> [7]</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/2005/4/7/176" onclick="this.href='/lkml/headers'+'/2005/4/7/176';">[headers]</a>聽 <a href="/lkml/bounce/2005/4/7/176">[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/2005/4/6/121">First message in thread</a></li><li><a href="/lkml/2005/4/7/20">David Woodhouse</a><ul><li><a href="/lkml/2005/4/7/54">Andrew Morton</a><ul><li><a href="/lkml/2005/4/7/59">Paul Mackerras</a><ul><li><a href="/lkml/2005/4/7/66">Andrew Morton</a><ul><li><a href="/lkml/2005/4/7/88">Paul Mackerras</a></li></ul></li><li><a href="/lkml/2005/4/7/84">Geert Uytterhoeven</a></li></ul></li><li><a href="/lkml/2005/4/7/61">David Woodhouse</a><ul><li><a href="/lkml/2005/4/7/68">Andrew Morton</a></li><li><a href="/lkml/2005/4/7/71">Russell King</a><ul><li><a href="/lkml/2005/4/7/75">David Woodhouse</a></li></ul></li></ul></li><li><a href="/lkml/2005/4/7/64">David Vrabel</a></li></ul></li><li><a href="/lkml/2005/4/7/60">Sergei Organov</a><ul><li><a href="/lkml/2005/4/7/81">Matthias Andree</a><ul><li><a href="/lkml/2005/4/7/85">Andrew Walrond</a></li><li><a href="/lkml/2005/4/9/73">David Roundy</a><ul><li><a href="/lkml/2005/4/10/16">Giuseppe Bilotta</a></li></ul></li></ul></li></ul></li><li><a href="/lkml/2005/4/7/150">Linus Torvalds</a><ul><li><a href="/lkml/2005/4/7/175">Daniel Phillips</a></li><li class="origin"><a href="/lkml/2005/4/7/185">Al Viro</a><ul><li><a href="/lkml/2005/4/7/185">Linus Torvalds</a><ul><li><a href="/lkml/2005/4/7/192">=?iso-8859-1?Q?J=F6rn?= Engel</a></li><li><a href="/lkml/2005/4/7/326">Jeff Garzik</a></li></ul></li><li><a href="/lkml/2005/4/7/188">Bartlomiej Zolnierkiewicz</a></li><li><a href="/lkml/2005/4/7/190">Daniel Phillips</a><ul><li><a href="/lkml/2005/4/7/197">Dmitry Yusupov</a></li><li><a href="/lkml/2005/4/8/198">Jon Masters</a></li></ul></li></ul></li><li><a href="/lkml/2005/4/8/289">Roman Zippel</a><ul><li><a href="/lkml/2005/4/8/299">Tupshin Harper</a><ul><li><a href="/lkml/2005/4/8/314">Roman Zippel</a></li></ul></li><li><a href="/lkml/2005/4/9/79">"Eric D. Mudama"</a><ul><li><a href="/lkml/2005/4/9/85">Roman Zippel</a></li></ul></li></ul></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">Date</td><td class="rp" itemprop="datePublished">Thu, 7 Apr 2005 18:10:07 +0100</td></tr><tr><td class="lp">From</td><td class="rp" itemprop="author">Al Viro <></td></tr><tr><td class="lp">Subject</td><td class="rp" itemprop="name">Re: Kernel SCM saga..</td></tr></table></td><td></td></tr></table><pre itemprop="articleBody">On Thu, Apr 07, 2005 at 08:32:04AM -0700, Linus Torvalds wrote:<br />> Also, there's actually a second reason why I've decided that cherry-<br />> picking is wrong, and it's non-technical. <br />> <br />> The thing is, cherry-picking very much implies that the people "up" the <br />> foodchain end up editing the work of the people "below" them. The whole <br />> reason you want cherry-picking is that you want to fix up somebody elses <br />> mistakes, ie something you disagree with.<br /><br />No. There's another reason - when you are cherry-picking and reordering<br />*your* *own* *patches*. That's what I had been unable to explain to<br />Larry and that's what made BK unusable for me.<br /><br />As for the immutable history... Ever had to read or grade students'<br />homework?<br /> * the dumbest kind: "here's an answer <expression>, whaddya<br />mean 'where's the solution'?".<br /> * next one: "here's how I've solved the problem: <pages of text<br />documenting the attempts, with many 'oops, there had been a mistake,<br />here's how we fix it'>". <br /> * what you really want to see: series of steps leading to answer,<br />with clean logical structure that allows to understand what's being<br />done and verify correctness.<br /><br />The first corresponds to "here's a half-meg of patch, it fixes everything".<br />The second is chronological history (aka "this came from our CVS, all bugs<br />are fixed by now, including those introduced in the middle of it; see<br />CVS history for details"). The third is a decent patch series.<br /><br />And to get from "here's how I came up to solution" to "here's a clean way<br />to reach the solution" you _have_ to reorder. There's also "here are<br />misc notes from today, here are misc notes from yesterday, etc." and to<br />get that into sane shape you will need to split, reorder and probably<br />collapse several into combined delta (possibly getting an empty delta<br />as the result, if later ones negate the prior).<br /><br />The point being, both history and well, publishable result can be expressed<br />as series of small steps, but they are not the same thing. So far all I've<br />seen in the area (and that includes BK) is heavily biased towards history part<br />and attempts to use this stuff for manipulating patch series turn into fighting<br />the tool.<br /><br />I'd *love* to see something that can handle both - preferably with<br />history of reordering, etc. being kept. IOW, not just a tree of changesets<br />but a lattice - with multiple paths leading to the same node. So far<br />I've seen nothing of that kind ;-/<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: 2005-04-07 19:14 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>