net/rds: Initialize ic->i_fastreg_wrs upon allocation

Otherwise, if an IB connection is torn down before "rds_ib_setup_qp"
is called, the value of "ic->i_fastreg_wrs" is still at zero
(as it wasn't initialized by "rds_ib_setup_qp").
Consequently "rds_ib_conn_path_shutdown" will spin forever,
waiting for it to go back to "RDS_IB_DEFAULT_FR_WR",
which of course will never happen as there are no
outstanding work requests.

Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c
index 1b6fd6c8b12b29bdcc24dc6875395ab061df798d..4de0214da63c4dd3aa66ef02d6c5e09337abb358 100644 (file)
--- a/net/rds/ib_cm.c
+++ b/net/rds/ib_cm.c
@@ -527,7 +527,6 @@ static int rds_ib_setup_qp(struct rds_connection *conn)
        attr.qp_type = IB_QPT_RC;
        attr.send_cq = ic->i_send_cq;
        attr.recv_cq = ic->i_recv_cq;
-       atomic_set(&ic->i_fastreg_wrs, RDS_IB_DEFAULT_FR_WR);
 
        /*
         * XXX this can fail if max_*_wr is too large?  Are we supposed
@@ -1139,6 +1138,7 @@ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp)
        spin_lock_init(&ic->i_ack_lock);
 #endif
        atomic_set(&ic->i_signaled_sends, 0);
+       atomic_set(&ic->i_fastreg_wrs, RDS_IB_DEFAULT_FR_WR);
 
        /*
         * rds_ib_conn_shutdown() waits for these to be emptied so they