Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 1 Mar 2010 17:22:27 -0500
From:      John Baldwin <jhb@freebsd.org>
To:        Kostik Belousov <kostikbel@gmail.com>
Cc:        svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org
Subject:   Re: svn commit: r204475 - head/sys/fs/msdosfs
Message-ID:  <201003011722.27155.jhb@freebsd.org>
In-Reply-To: <20100301221213.GY2489@deviant.kiev.zoral.com.ua>
References:  <201002281719.o1SHJM6u042302@svn.freebsd.org> <201003011030.51154.jhb@freebsd.org> <20100301221213.GY2489@deviant.kiev.zoral.com.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 01 March 2010 5:12:13 pm Kostik Belousov wrote:
> On Mon, Mar 01, 2010 at 10:30:51AM -0500, John Baldwin wrote:
> > On Sunday 28 February 2010 12:19:22 pm Konstantin Belousov wrote:
> > > Author: kib
> > > Date: Sun Feb 28 17:19:22 2010
> > > New Revision: 204475
> > > URL: http://svn.freebsd.org/changeset/base/204475
> > > 
> > > Log:
> > >   Mark msdosfs as mpsafe.
> > 
> > Are you planning on implementing support for LOOKUP_SHARED and
> > EXTENDED_SHARED as well?
> 
> Do you think it makes sense to do this for msdosfs ?
> Please note that current msdosfs code exclusively uses exclusive locks
> on the vnodes.
> 
> One obvious issue with *_SHARED is the fat cache in the denode.

Oh, hmm.  I guess we would need to first allow it to use shared locks 
generically (VN_LOCK_ASHARE() for VOP_READ() and the like).  I do think we 
should try to make our filesystems concurrent when it is not too terribly 
difficult.

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201003011722.27155.jhb>