[PATCH net 2/2] net: mctp: test: Use dev_direct_xmit for TX to our test device

[Jeremy Kerr <jk@codeconstruct.com.au>]

In our test cases, we typically feed a packet sequence into the routing
code, then inspect the device's TXed skbs to assert specific behaviours.

Using dev_queue_xmit() for our TX path introduces a fair bit of
complexity between the test packet sequence and the test device's
ndo_start_xmit callback; which may mean that the skbs have not hit the
device at the point we're inspecting the TXed skb list.

Use dev_direct_xmit instead, as we want a direct a path as possible
here, and the test dev does not need any queueing, scheduling or flow
control.

Fixes: 6ab578739a4c ("net: mctp: test: move TX packetqueue from dst to dev")
Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202604281320.525eee17-lkp@intel.com
Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
---
 net/mctp/test/utils.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mctp/test/utils.c b/net/mctp/test/utils.c
index c3987d5ade7a..6eef8d485c25 100644
--- a/net/mctp/test/utils.c
+++ b/net/mctp/test/utils.c
@@ -116,7 +116,7 @@ void mctp_test_destroy_dev(struct mctp_test_dev *dev)
 static int mctp_test_dst_output(struct mctp_dst *dst, struct sk_buff *skb)
 {
 	skb->dev = dst->dev->dev;
-	dev_queue_xmit(skb);
+	dev_direct_xmit(skb, 0);
 
 	return 0;
 }

-- 
2.39.5