xprtrdma: Fix completion wait during device removal

commit 13cb886c59 upstream.

I've found that on occasion, "rmmod <dev>" will hang while if an NFS
is under load.

Ensure that ri_remove_done is initialized only just before the
transport is woken up to force a close. This avoids the completion
possibly getting initialized again while the CM event handler is
waiting for a wake-up.

Fixes: bebd031866 ("xprtrdma: Support unplugging an HCA from under an NFS mount")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Chuck Lever 2020-01-03 11:52:17 -05:00 committed by Greg Kroah-Hartman
parent b1152098d4
commit f1e47dcc65

View File

@ -248,6 +248,7 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event)
ia->ri_device->name,
rpcrdma_addrstr(xprt), rpcrdma_portstr(xprt));
#endif
init_completion(&ia->ri_remove_done);
set_bit(RPCRDMA_IAF_REMOVING, &ia->ri_flags);
ep->rep_connected = -ENODEV;
xprt_force_disconnect(&xprt->rx_xprt);
@ -306,7 +307,6 @@ rpcrdma_create_id(struct rpcrdma_xprt *xprt, struct rpcrdma_ia *ia)
trace_xprtrdma_conn_start(xprt);
init_completion(&ia->ri_done);
init_completion(&ia->ri_remove_done);
id = rdma_create_id(xprt->rx_xprt.xprt_net, rpcrdma_conn_upcall,
xprt, RDMA_PS_TCP, IB_QPT_RC);