Starting from Linux 5.16 kernel, Tx timeout mechanism was added
in the virtio_net driver which prints the "Tx timeout" warning
message when a packet stays in Tx queue for too long. Below is an
example of the reported message:
"[494105.316739] virtio_net virtio1 tmfifo_net0: TX timeout on
queue: 0, sq: output.0, vq: 0×1, name: output.0, usecs since
last trans: 3079892256".
This issue could happen when external host driver which drains the
FIFO is restared, stopped or upgraded. To avoid such confusing
"Tx timeout" messages, this commit adds logic to drop the outstanding
Tx packet if it's not able to transmit in two seconds due to Tx FIFO
full, which can be considered as congestion or out-of-resource drop.
This commit also handles the special case that the packet is half-
transmitted into the Tx FIFO. In such case, the packet is discarded
with remaining length stored in vring->rem_padding. So paddings with
zeros can be sent out when Tx space is available to maintain the
integrity of the packet format. The padded packet will be dropped on
the receiving side.
Signed-off-by: Liming Sun <email address hidden>
Reviewed-by: Ilpo Järvinen <email address hidden>
Link: https://<email address hidden>
Signed-off-by: Hans de Goede <email address hidden>
(cherry picked from commit 8cbc756b802605dee3dd40019bd75960772bacf5)
Signed-off-by: Liming Sun <email address hidden>
Acked-by: Bartlomiej Zolnierkiewicz <email address hidden>
Acked-by: Andrei Gherzan <email address hidden>
Signed-off-by: Bartlomiej Zolnierkiewicz <email address hidden>
