xfs
[Top] [All Lists]

Re: [PATCH 4/6] xfs: DAX does not use IO completion callbacks

To: Brian Foster <bfoster@xxxxxxxxxx>
Subject: Re: [PATCH 4/6] xfs: DAX does not use IO completion callbacks
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 30 Oct 2015 10:39:11 +1100
Cc: xfs@xxxxxxxxxxx, ross.zwisler@xxxxxxxxxxxxxxx, jack@xxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <20151029142957.GF11663@xxxxxxxxxxxxxxx>
References: <1445225238-30413-1-git-send-email-david@xxxxxxxxxxxxx> <1445225238-30413-5-git-send-email-david@xxxxxxxxxxxxx> <20151029142957.GF11663@xxxxxxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Oct 29, 2015 at 10:29:57AM -0400, Brian Foster wrote:
> On Mon, Oct 19, 2015 at 02:27:16PM +1100, Dave Chinner wrote:
> > From: Dave Chinner <dchinner@xxxxxxxxxx>
> > 
> > For DAX, we are now doing block zeroing and
> > we are updating the file size during allocation. This means we no
> > longer need an IO completion callback to do these things, so remove
> > the completion callbacks from the __dax_fault and __dax_mkwrite
> > calls.
> > 
> 
> Where do we "update the file size during allocation?"

Stale comment. For page faults, we'll never update the file size
(segv if fault is beyond EOF), and DIO still does IO completion
based file size updates.

Cheers,

Dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

<Prev in Thread] Current Thread [Next in Thread>