pnfs: fix pnfs lock inversion of i_lock and cl_lock
The pnfs code was using throughout the lock order i_lock, cl_lock.
This conflicts with the nfs delegation code. Rework the pnfs code
to avoid taking both locks simultaneously.
Currently the code takes the double lock to add/remove the layout to a
nfs_client list, while atomically checking that the list of lsegs is
empty. To avoid this, we rely on existing serializations. When a
layout is initialized with lseg count equal zero, LAYOUTGET's
openstateid serialization is in effect, making it safe to assume it
stays zero unless we change it. And once a layout's lseg count drops
to zero, it is set as DESTROYED and so will stay at zero.
Signed-off-by: Fred Isaman <iisaman@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index 8958757..2f41dcce 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -188,10 +188,10 @@
rv = NFS4ERR_DELAY;
list_del_init(&lo->plh_bulk_recall);
spin_unlock(&ino->i_lock);
+ pnfs_free_lseg_list(&free_me_list);
put_layout_hdr(lo);
iput(ino);
}
- pnfs_free_lseg_list(&free_me_list);
return rv;
}