it will be handled by j1939_cancel_all_active_sessions()
Signed-off-by: Oleksij Rempel <o.rempel@xxxxxxxxxxxxxx>
---
net/can/j1939/transport.c | 19 +++++++++++++++----
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c
index a2f3de289387..ecdcfb77c2e9 100644
--- a/net/can/j1939/transport.c
+++ b/net/can/j1939/transport.c
@@ -1078,7 +1078,8 @@ static enum hrtimer_restart j1939_tp_txtimer(struct hrtimer *hrtimer)
ret = j1939_xtp_txnext_receiver(session);
}
- if (ret == -ENOBUFS) {
+ switch (ret) {
+ case -ENOBUFS:
/* Retry limit is currently arbitrary chosen */
if (session->tx_retry < J1939_XTP_TX_RETRY_LIMIT) {
session->tx_retry++;
@@ -1091,7 +1092,19 @@ static enum hrtimer_restart j1939_tp_txtimer(struct hrtimer *hrtimer)
j1939_session_rxtimer_cancel(session);
j1939_session_deactivate_activate_next(session);
}
- } else if (ret) {
+ break;
+ case -ENETDOWN:
+ /* In this case we should get a netdev_event(), all active
+ * sessions will be cleared by
+ * j1939_cancel_all_active_sessions(). So handle this as an
+ * error, but let j1939_cancel_all_active_sessions() do the
+ * cleanup including propagation of the error to user space.
+ */
+ break;
+ case 0:
+ session->tx_retry = 0;
+ break;
+ default:
netdev_alert(priv->ndev, "%s: 0x%p: tx aborted with unknown reason: %i\n",
__func__, session, ret);
if (session->skcb.addr.type != J1939_SIMPLE) {
@@ -1103,8 +1116,6 @@ static enum hrtimer_restart j1939_tp_txtimer(struct hrtimer *hrtimer)
j1939_session_rxtimer_cancel(session);
j1939_session_deactivate_activate_next(session);
}
- } else {
- session->tx_retry = 0;
}
j1939_session_put(session);
--
2.20.1
