dlm: initialize file_lock struct in GETLK before copying conflicting lock
commit20d5a39929232a715f29e6cb7e3f0d0c790f41eb
authorJeff Layton <jlayton@redhat.com>
Wed, 21 Jan 2009 16:34:50 +0000 (21 11:34 -0500)
committerDavid Teigland <teigland@redhat.com>
Wed, 21 Jan 2009 21:28:45 +0000 (21 15:28 -0600)
tree6662d2a65f470526a8d6ec591b8429a2ca6024fb
parent24179f488092267c9a033d7e25ce7a58af50ff79
dlm: initialize file_lock struct in GETLK before copying conflicting lock

dlm_posix_get fills out the relevant fields in the file_lock before
returning when there is a lock conflict, but doesn't clean out any of
the other fields in the file_lock.

When nfsd does a NFSv4 lockt call, it sets the fl_lmops to
nfsd_posix_mng_ops before calling the lower fs. When the lock comes back
after testing a lock on GFS2, it still has that field set. This confuses
nfsd into thinking that the file_lock is a nfsd4 lock.

Fix this by making DLM reinitialize the file_lock before copying the
fields from the conflicting lock.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: David Teigland <teigland@redhat.com>
fs/dlm/plock.c