Re: DAX data corruption for mmaped and written files

[Jan Kara <jack@suse.cz>]

On Tue 29-03-16 09:39:16, Dave Chinner wrote:
> On Thu, Mar 24, 2016 at 02:12:23PM +0100, Jan Kara wrote:
> > Hello,
> > 
> > yesterday I have been stress-testing mmap code with my new fault locking
> > patches and I have found a data corruption issue when file is written both
> > via mmap and standard write(2). The problem is following:
> > 
> > CPU1					CPU2
> > dax_io()				dax_fault()
> >   get_block() - allocates block
> >   ...					  get_block() - finds allocated block
> > 					    - zeroes it inside fs
> > 					  fault completese
> > 
> >   if (buffer_unwritten(bh) || buffer_new(bh)) -> new buffer
> >     dax_new_buf() -> zeroes buffer which may
> > 		     overwrite user data
> 
> Which filesystem? XFS should be, in both cases, zeroing the
> block entirely inside get_block() when it is first allocated. i.e we
> should see:
> 
> CPU1					CPU2
> dax_io()				dax_fault()
>   get_block() - allocates block
>               - zeroes it inside fs
>   ...					  get_block() - finds allocated block
> 					  fault completes
> 
>   buffer returned is not new or unwritten.
> 
> 
> > In some cases the race can also go the other way around and we lose data
> > written by write.

Yeah, correct. This is ext4 specific issue because ext4 doesn't return
prezeroed blocks from get_block() callback used for dax_io().

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR