Re: [PATCH 1/6] xfs: mmap write/read leaves bad state on pages

From: Chris Mason <clm@fb.com>
To: Christoph Hellwig <hch@infradead.org>,
	Dave Chinner <david@fromorbit.com>
Cc: linux-fsdevel@vger.kernel.org, xfs@oss.sgi.com
Subject: Re: [PATCH 1/6] xfs: mmap write/read leaves bad state on pages
Date: Thu, 21 Aug 2014 11:21:34 -0400	[thread overview]
Message-ID: <53F60E7E.2060905@fb.com> (raw)
In-Reply-To: <20140821130812.GC1567@infradead.org>

On 08/21/2014 09:08 AM, Christoph Hellwig wrote:
> On Thu, Aug 21, 2014 at 03:09:09PM +1000, Dave Chinner wrote:
>> From: Dave Chinner <dchinner@redhat.com>
>>
>> generic/263 is failing fsx at this point with a page spanning
>> EOF that cannot be invalidated. The operations are:
>>
>> 1190 mapwrite   0x52c00 thru    0x5e569 (0xb96a bytes)
>> 1191 mapread    0x5c000 thru    0x5d636 (0x1637 bytes)
>> 1192 write      0x5b600 thru    0x771ff (0x1bc00 bytes)
>>
>> where 1190 extents EOF from 0x54000 to 0x5e569. When the direct IO
>> write attempts to invalidate the cached page over this range, it
>> fails with -EBUSY and so we fire this assert:
>>
>> XFS: Assertion failed: ret < 0 || ret == count, file: fs/xfs/xfs_file.c, line: 676
>>
>> because the kernel is trying to fall back to buffered IO on the
>> direct IO path (which XFS does not do).
>>
>> The real question is this: Why can't that page be invalidated after
>> it has been written to disk an cleaned?
>>
>> Well, there's data on the first two buffers in the page (1k block
>> size, 4k page), but the third buffer on the page (i.e. beyond EOF)
>> is failing drop_buffers because it's bh->b_state == 0x3, which is
>> BH_Uptodate | BH_Dirty.  IOWs, there's dirty buffers beyond EOF. Say
>> what?
>>
>> OK, set_buffer_dirty() is called on all buffers from
>> __set_page_buffers_dirty(), regardless of whether the buffer is
>> beyond EOF or not, which means that when we get to ->writepage,
>> we have buffers marked dirty beyond EOF that we need to clean.
>> So, we need to implement our own .set_page_dirty method that
>> doesn't dirty buffers beyond EOF.
> 
> Shouldn't this be fixed in __set_page_buffers_dirty itself?  This
> doesn't seem an XFS-specific issue.
> 

block_write_full_page() is invalidating buffers past eof.  Probably
because we used to be able to dirty buffers without holding the page
lock, and it's much easier to trust i_size at writepage time.

I think we have the page locked for all the dirties now, so this isn't
as important as in the past?

-chris

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs