Date: Thu, 21 May 2009 12:10:59 -0400 From: John Baldwin <jhb@freebsd.org> To: Scott Long <scottl@samsco.org> Cc: src-committers@freebsd.org, svn-src-all@freebsd.org, attilio@freebsd.org, svn-src-head@freebsd.org, rwatson@freebsd.org, kostikbel@gmail.com, "M. Warner Losh" <imp@bsdimp.com> Subject: Re: svn commit: r192535 - head/sys/kern Message-ID: <200905211211.00168.jhb@freebsd.org> In-Reply-To: <4A157919.7040103@samsco.org> References: <3bbf2fe10905210629p46c7a204v6863aaba77354462@mail.gmail.com> <20090521.094100.70797067.imp@bsdimp.com> <4A157919.7040103@samsco.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 21 May 2009 11:54:01 am Scott Long wrote: > M. Warner Losh wrote: > > In message: <alpine.BSF.2.00.0905211610140.18790@fledge.watson.org> > > Robert Watson <rwatson@FreeBSD.org> writes: > > : On Thu, 21 May 2009, John Baldwin wrote: > > : > > : >>>> Move the M_WAITOK flag in notify() into an M_NOWAIT one in order to > > : > match > > : >>>> the behaviour alredy present with the further malloc() call in > > : >>>> devctl_notify(). > > : >>>> This fixes a bug in the CAM layer where the camisr handler finished to > > : >>>> call camperiphfree() (and subsequently destroy_dev() resulting in a new > > : >>>> dev notify) while the xpt lock is held. > > : >>> This is wrong. You cannot call destroy_dev() while holding any mutex. > > : >>> Taking this into account, it makes no sense to use M_NOWAIT in notify(). > > : >> > > : >> As long as devctl_notify() also calls M_NOWAIT and if not available skips > > : >> "silently" it just does the same thing, I think this approach is more > > : >> consistent. > > : >> > > : >> It remains, though, the fact to fix CAM when calling destroy_dev(). Maybe > > : >> we should add a witness_warn() there? > > : > > > : > I agree with kib, this should be reverted and CAM fixed instead. I also > > : > agree that M_NOWAIT use should be limited where possible. > > : > > : devctl_notify() probably needs to grow a sleepable flag, or perhaps we need > > : two variations, one that can sleep. > > > > devctl_notify() has expanded well beyond its original needs. Having > > an extra case for sleeping is the wrong way to solve this problem. > > Really. We're adding hacks on hacks on hacks here and we need to step > > back and think. > > > > I specifically didn't put in CDEV notifications into devd when I > > originally did it because one can get the same notification via > > kevents on /dev. Maybe the right answer is to remove this stuff > > entirely and update devd to do that instead? It isn't a lot of code, > > and should provide equivalent functionality without needing to change > > the rules of the game when it comes to destroy_dev(). Especially this > > close to the code slush... > > > > Comments? > > > > Warner > > Very much in agreement here. I would also love to have destroy_dev() > and make_dev() be locking-neutral. Having sleepable locks in leaf APIs > is unpleasant for consumers of those APIs. destroy_dev() does not use a sleepable lock, the problem is it drains so it can provide sane semantics to a caller who wants to ensure that all outside references to a cdev are gone when it returns. You can't provide that without doing some sort of synchronization with the other threads trying to call d_open(), etc. And you most certainly can't do it if you call destroy_dev() while holding your driver's mutex as you then have the problem that some other thread could be blocked on that mutex already in your d_open() routine when you call destroy_dev(). These sane semantics are needed so drivers can do things like safely free softcs and destroy locks, etc. -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200905211211.00168.jhb>