[PATCH] j1939: fix unable to send messages with data length zero
From: Alexander Hölzl
Date: Wed Feb 05 2025 - 12:49:49 EST
The J1939 standard requires the transmission of messages of length 0.
For example the proprietary messages are specified with a data length
of 0 to 1785. The transmission of such messages was not possible.
Sending such a message resulted in no error being returned but no
corresponding can frame being generated.
This patch enables the transmission of zero length J1939 messages.
In order to facilitate this two changes were necessary.
First when the transmission of a new message is requested from userspace
the message is segmented in j1939_sk_send_loop. The segmentation did
account for zero length messages and terminated immediately without
queuing the corresponding skb.
Second in j1939_session_skb_get_by_offset, when selecting the next skb to
transmit for a session, it was not checked that there now might be a
zero length skb in the queue.
Signed-off-by: Alexander Hölzl <alexander.hoelzl@xxxxxxx>
---
net/can/j1939/socket.c | 4 ++--
net/can/j1939/transport.c | 5 +++--
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c
index 305dd72c844c..17226b2341d0 100644
--- a/net/can/j1939/socket.c
+++ b/net/can/j1939/socket.c
@@ -1132,7 +1132,7 @@ static int j1939_sk_send_loop(struct j1939_priv *priv, struct sock *sk,
todo_size = size;
- while (todo_size) {
+ do {
struct j1939_sk_buff_cb *skcb;
segment_size = min_t(size_t, J1939_MAX_TP_PACKET_SIZE,
@@ -1177,7 +1177,7 @@ static int j1939_sk_send_loop(struct j1939_priv *priv, struct sock *sk,
todo_size -= segment_size;
session->total_queued_size += segment_size;
- }
+ } while (todo_size);
switch (ret) {
case 0: /* OK */
diff --git a/net/can/j1939/transport.c b/net/can/j1939/transport.c
index 319f47df3330..99832e60c08d 100644
--- a/net/can/j1939/transport.c
+++ b/net/can/j1939/transport.c
@@ -382,8 +382,9 @@ sk_buff *j1939_session_skb_get_by_offset(struct j1939_session *session,
skb_queue_walk(&session->skb_queue, do_skb) {
do_skcb = j1939_skb_to_cb(do_skb);
- if (offset_start >= do_skcb->offset &&
- offset_start < (do_skcb->offset + do_skb->len)) {
+ if ((offset_start >= do_skcb->offset &&
+ offset_start < (do_skcb->offset + do_skb->len)) ||
+ (offset_start == 0 && do_skcb->offset == 0 && do_skb->len == 0)) {
skb = do_skb;
}
}
--
2.43.0