Xiubo Li 8e4473bb50 ceph: do not execute direct write in parallel if O_APPEND is specified
In O_APPEND & O_DIRECT mode, the data from different writers will
be possibly overlapping each other since they take the shared lock.

For example, both Writer1 and Writer2 are in O_APPEND and O_DIRECT
mode:

          Writer1                         Writer2

     shared_lock()                   shared_lock()
     getattr(CAP_SIZE)               getattr(CAP_SIZE)
     iocb->ki_pos = EOF              iocb->ki_pos = EOF
     write(data1)
                                     write(data2)
     shared_unlock()                 shared_unlock()

The data2 will overlap the data1 from the same file offset, the
old EOF.

Switch to exclusive lock instead when O_APPEND is specified.

Signed-off-by: Xiubo Li <xiubli@redhat.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
2020-02-11 17:04:04 +01:00
..
2019-09-27 15:10:34 -07:00
2020-02-03 17:03:42 +00:00
2019-09-19 09:42:37 -07:00
2019-11-11 09:21:59 -05:00
2019-09-19 09:42:37 -07:00
2020-02-05 05:28:20 +00:00
2020-02-07 17:50:21 -08:00
\n
2020-01-30 15:37:41 -08:00
2019-09-19 09:42:37 -07:00
2020-01-14 13:28:28 -08:00
2020-02-07 14:39:38 +09:00
2019-11-30 10:44:49 -08:00
2019-08-07 21:51:47 -04:00
2020-01-29 19:38:34 -08:00
2020-02-07 14:48:35 -05:00
2020-02-07 14:48:35 -05:00
2020-01-29 18:53:37 -08:00
2020-02-06 06:33:17 +00:00
2020-02-08 13:44:41 -08:00
2020-02-09 15:51:46 -08:00
2020-02-09 15:51:46 -08:00
2020-02-01 10:36:49 -08:00
2020-02-03 21:23:33 -05:00
2020-02-04 11:45:21 +00:00
2019-12-08 14:37:36 +01:00