[PATCH net 3/6] can: j1939: fix wrong RX timeout for CTS hold messages

[Marc Kleine-Budde <mkl@pengutronix.de>]

From: Alexander Hölzl <alexander.hoelzl@gmx.net>

In J1939 segmented transport, a CTS message with data byte 2 set to zero is
interpreted as a hold message. This instructs the transmitter of the
segmented message to hold the connection open but to delay sending.

According to the J1939-21 standard, section 5.10.2.4 the timeout T4 after
which an held open session is invalidated is 1050 ms, not 550 as
implemented currently. The 550 ms are problematic if a device uses hold
messages and assumes it can wait for more than 550 ms before it has to
resend the hold message.

Fix the RX timeout by changing the T4 timeout from 550 ms to 1050.

Signed-off-by: Alexander Hölzl <alexander.hoelzl@gmx.net>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://patch.msgid.link/20260421153152.87772-3-alexander.hoelzl@gmx.net
[mkl: rewrap long lines in patch description, use imperative mood in last section]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 net/can/j1939/transport.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c
index df93d57907da..7ad56b5f17b9 100644
--- a/net/can/j1939/transport.c
+++ b/net/can/j1939/transport.c
@@ -1479,7 +1479,7 @@ j1939_xtp_rx_cts_one(struct j1939_session *session, struct sk_buff *skb)
 		}
 	} else {
 		/* CTS(0) */
-		j1939_tp_set_rxtimeout(session, 550);
+		j1939_tp_set_rxtimeout(session, 1050);
 	}
 	return;
 
-- 
2.53.0