RDMA/iw_cxgb4: move QP -> ERROR on fatal disconnect errors
In c4iw_ep_disconnect(), if we fail to initiate a close operation, then
move the qp to ERROR to disassociate the ep from the qp. Failure to do
this will leak the ep resources.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 44e0bc4..aea69ca 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -3509,6 +3509,19 @@
stop_ep_timer(ep);
close_complete_upcall(ep, -EIO);
}
+ if (ep->com.qp) {
+ struct c4iw_qp_attributes attrs;
+
+ attrs.next_state = C4IW_QP_STATE_ERROR;
+ ret = c4iw_modify_qp(ep->com.qp->rhp,
+ ep->com.qp,
+ C4IW_QP_ATTR_NEXT_STATE,
+ &attrs, 1);
+ if (ret)
+ pr_err(MOD
+ "%s - qp <- error failed!\n",
+ __func__);
+ }
fatal = 1;
}
}