Minchan Kim 7e5a5104c6 zram: Fix deadlock bug in partial read/write
Now zram allocates new page with GFP_KERNEL in zram I/O path
if IO is partial. Unfortunately, It may cause deadlock with
reclaim path like below.

write_page from fs
fs_lock
allocation(GFP_KERNEL)
reclaim
pageout
				write_page from fs
				fs_lock <-- deadlock

This patch fixes it by using GFP_NOIO.  In read path, we
reorganize code flow so that kmap_atomic is called after the
GFP_NOIO allocation.

Cc: stable@vger.kernel.org
Acked-by: Jerome Marchand <jmarchand@redhat.com>
Acked-by: Nitin Gupta <ngupta@vflare.org>
[ penberg@kernel.org: don't use GFP_ATOMIC ]
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Signed-off-by: Minchan Kim <minchan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2013-02-03 17:52:49 -08:00
..
2013-01-23 20:30:52 -08:00
2012-12-11 14:08:47 -08:00
2012-12-18 15:19:06 +10:30
2013-01-23 20:30:52 -08:00
2013-01-22 16:36:23 -08:00
2012-12-12 11:45:16 -08:00
2013-01-08 18:53:56 -08:00
2013-01-25 21:25:02 -08:00
2013-01-08 18:53:56 -08:00
2012-12-19 08:19:07 -08:00
2012-12-11 13:13:55 -08:00
2013-01-16 15:57:54 +01:00
2012-12-12 12:05:15 -08:00
2013-01-15 10:45:26 -07:00
2013-01-18 12:02:52 -08:00