Hi Simon,
On 1/19/24 16:26, Simon Horman wrote:
On Thu, Jan 18, 2024 at 02:51:34AM +0000, Dmitry Safonov wrote:
From: Mohammad Nassiri mnassiri@ciena.com
The end_server() function only operates in the server thread and always takes an accept socket instead of a listen socket as its input argument. To align with this, invert the boolean values used when calling verify_counters() within the end_server() function.
Hi Dmitry,
I see that this is correct, but I am wondering if it fixes a user-visible bug. If so, could some text about that go here?
Yep. The test as the result doesn't correctly check tcp counters for non-symmetrical scenario, where peer-A uses a key <a:a2> to send data, but peer-B uses <b:b2> key to send its data. So, in simple terms, different keys for TX and RX on peers.
I'll send v2 with an improved patch message.
Fixes: 3c3ead555648 ("selftests/net: Add TCP-AO key-management test") Signed-off-by: Mohammad Nassiri mnassiri@ciena.com Link: https://lore.kernel.org/all/934627c5-eebb-4626-be23-cfb134c01d1a@arista.com/ [amended 'Fixes' tag and carried-over to lkml] Signed-off-by: Dmitry Safonov dima@arista.com
...
Thanks, Dmitry