LKML: Greg KH: Re: PCI: fix an oops in some pci devices on hotplug remove when their resources are being freed.

<?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: Greg KH: Re: PCI: fix an oops in some pci devices on hotplug remove when their resources are being freed.</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 Greg KH" href="/groupie.php?aid=45" /><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/2005/4/1"> [1]</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/2005/4/1/338" onclick="this.href='/lkml/headers'+'/2005/4/1/338';">[headers]</a>聽 <a href="/lkml/bounce/2005/4/1/338">[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/2005/4/1/275">First message in thread</a></li><li><a href="/lkml/2005/4/1/307">Greg KH</a><ul><li><a href="/lkml/2005/4/1/304">Greg KH</a><ul><li><a href="/lkml/2005/4/1/305">Greg KH</a><ul><li><a href="/lkml/2005/4/1/302">Greg KH</a><ul><li><a href="/lkml/2005/4/1/309">Greg KH</a></li></ul></li><li><a href="/lkml/2005/4/2/24">Rolf Eike Beer</a><ul><li><a href="/lkml/2005/4/2/64">Andrew Morton</a></li><li><a href="/lkml/2005/4/7/285">Greg KH</a></li></ul></li></ul></li></ul></li><li><a href="/lkml/2005/4/1/311">Matthew Wilcox</a><ul><li class="origin"><a href="/lkml/2005/4/1/340">Greg KH</a><ul><li><a href="/lkml/2005/4/1/340">Matthew Wilcox</a><ul><li><a href="/lkml/2005/4/1/342">Greg KH</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">Fri, 1 Apr 2005 19:31:41 -0800</td></tr><tr><td class="lp">From</td><td class="rp" itemprop="author">Greg KH <></td></tr><tr><td class="lp">Subject</td><td class="rp" itemprop="name">Re: PCI: fix an oops in some pci devices on hotplug remove when their resources are being freed.</td></tr></table></td><td></td></tr></table><pre itemprop="articleBody">On Sat, Apr 02, 2005 at 02:10:23AM +0100, Matthew Wilcox wrote:<br />> On Fri, Apr 01, 2005 at 03:47:50PM -0800, Greg KH wrote:<br />> > PCI: fix an oops in some pci devices on hotplug remove when their resources are being freed.<br />> > <br />> > As reported by Prarit Bhargava <prarit@sgi.com><br />> > Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de><br />> > <br />> > diff -Nru a/drivers/pci/remove.c b/drivers/pci/remove.c<br />> > for (i = 0; i < PCI_NUM_RESOURCES; i++) {<br />> > struct resource *res = dev->resource + i;<br />> > - if (res->parent)<br />> > + if (res && res->parent)<br />> > release_resource(res);<br />> <br />> I can't believe this fixes anything. How can res possibly be NULL here?<br />> It's a pointer into a pci_dev.<br /><br />I agree. However, SGI seems to have some majorly <insert expletive here><br />hardware and drivers that cause this line to release a already released<br />resource. See the other part of this patch for the part where this<br />resource is supposedly freed up.<br /><br />I took the patch as it doesn't hurt anyone, and it gets them off of my<br />back. But if you so much as think this patch isn't needed, I'll gladly<br />revert it, as I'm really not trusting any PCI hotplug patches coming out<br />from them anymore...<br /><br />thanks,<br /><br />greg k-h<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: 2005-04-06 13:31 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>

CINXE.COM

LKML: Greg KH: Re: PCI: fix an oops in some pci devices on hotplug remove when their resources are being freed.