From: Mingming Cao <cmm@us.ibm.com>
Subject: Re: [RFC PATCH] ext4: Use page_mkwrite vma_operations to get mmap
	write notification.
Date: Wed, 05 Mar 2008 15:37:18 -0800
Message-ID: <1204760238.3713.4.camel@localhost.localdomain>
References: <1204634542-16599-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
	 <1204677951.3605.6.camel@localhost.localdomain>
	 <20080305232927.GC18188@mit.edu>
Reply-To: cmm@us.ibm.com
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
	linux-ext4@vger.kernel.org
To: Theodore Tso <tytso@MIT.EDU>
In-Reply-To: <20080305232927.GC18188@mit.edu>
Sender: linux-ext4-owner@vger.kernel.org

On Wed, 2008-03-05 at 18:29 -0500, Theodore Tso wrote:
> On Tue, Mar 04, 2008 at 04:45:51PM -0800, Mingming Cao wrote:
> > > +	/* FIXME!! should we take inode->i_mutex ? Currently we can't because
> > > +	 * it has a circular locking dependency with DIO. But migrate expect
> > > +	 * i_mutex to ensure no i_data changes
> 
> Should I worry that we have something in the stable part of the patch
> queue with this FIXME!! comment?  :-)
> 

I think this comment could be moved to the migration.c. We can't take
i_mutex on mapped IO path. The i_data_mutex is the lock that should
protect the i_data concurrent changes, which is currently mapped IO
used. The race with migration could be addressed in migration instead of
here. I propose we drop this comment for now.

Mingming