Date: Thu, 19 Aug 2004 16:50:52 -0600 From: Scott Long <scottl@samsco.org> To: Roman Kurakin <rik@cronyx.ru> Cc: freebsd-current@freebsd.org Subject: Re: Tracking down LORs Message-ID: <41252ECC.5000203@samsco.org> In-Reply-To: <41252924.4020305@cronyx.ru> References: <41252924.4020305@cronyx.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
Roman Kurakin wrote: > Hi, > > Currently I am trying to track down a couple of LORS > in my code. But it seems that I do not undestand smth or all > things id realy so bad. > > So I want to ask some questions to find out if my thoughts > correct or wrong. > > 1. If I am right LOR means that we have at least two mutexs. > Lets call them a and b. If we set a, then b in first case > and b then a in second we could get dead loop, and thus LOR. Correct. > > 2. If I have some driver that have mutex a, and we have some > sytem code that could call this driver with Giant (b), we would > get LOR if driver lock a and some other part of system will > try to lock Giant? The general rule is that NO other locks should be held when Giant is grabbed. Scott
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41252ECC.5000203>