staging/lustre/ptlrpc: lost bulk leads to a hang

The reverse order of request_out_callback() and reply_in_callback()
puts the RPC into UNREGISTERING state, which is waiting for RPC &
bulk md unlink, whereas only RPC md unlink has been called so far.
If bulk is lost, even expired_set does not check for UNREGISTERING
state.

The same for write if server returns an error.

This phase is ambiguous, split to UNREG_RPC and UNREG_BULK.

Signed-off-by: Vitaly Fertman <[email protected]>
Seagate-bug-id:  MRP-2953, MRP-3206
Reviewed-by: Andriy Skulysh <[email protected]>
Reviewed-by: Alexey Leonidovich Lyashkov <[email protected]>
Tested-by: Elena V. Gryaznova <[email protected]>
Reviewed-on: http://review.whamcloud.com/19953
Reviewed-by: Chris Horn <[email protected]>
Reviewed-by: Ann Koehler <[email protected]>
Reviewed-by: Andreas Dilger <[email protected]>
Signed-off-by: Oleg Drokin <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

0 files changed, 0 insertions, 0 deletions