xfs
[Top] [All Lists]

Re: [PATCH 3/6] xfs: xfs_vm_write_end truncates too much on failure

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH 3/6] xfs: xfs_vm_write_end truncates too much on failure
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 10 Apr 2014 03:35:55 -0700
Cc: xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <1397106053-7489-4-git-send-email-david@xxxxxxxxxxxxx>
References: <1397106053-7489-1-git-send-email-david@xxxxxxxxxxxxx> <1397106053-7489-4-git-send-email-david@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Apr 10, 2014 at 03:00:50PM +1000, Dave Chinner wrote:
> From: Dave Chinner <dchinner@xxxxxxxxxx>
> 
> Similar to the write_begin problem, xfs-vm_write_end will truncate
> back to the old EOF, potentially removing page cache from over the
> top of delalloc blocks with valid data in them. Fix this by
> truncating back to just the start of the failed write.
> 
> Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>

Looks good,

Reviewed-by: Christoph Hellwig <hch@xxxxxx>

Any reason you used max_t in the previous one, and opencode it here?

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