devices: fix issue of per_interface_next_index

Per interface, next hop graph node can be customized
with vnet_hw_interface_rx_redirect_to_node function,
but it doesn't work well for af-packet type interface.

In current implementation, with function
af_packet_set_interface_next_node invoked next hop graph
node index can be set to apif->per_interface_next_index,
but it's not set to next0 properly for packet processing
in af_packet_device_input_fn.

Type: fix

Signed-off-by: Michael Yu <michael.a.yu@nokia-sbell.com>
Change-Id: I8e132ddd1c3c01b6f476de78546d4a9389b3ff87
Signed-off-by: Michael Yu <michael.a.yu@nokia-sbell.com>
(cherry picked from commit 90b34ed67a516c4391ad353ba431f8419b582d50)
This commit is contained in:
Michael Yu
2019-08-30 00:50:16 +08:00
committed by Andrew Yourtchenko
parent a2c0637129
commit 61b1d46db0

View File

@ -195,9 +195,6 @@ af_packet_device_input_fn (vlib_main_t * vm, vlib_node_runtime_t * node,
u32 n_buffer_bytes = vlib_buffer_get_default_data_size (vm);
u32 min_bufs = apif->rx_req->tp_frame_size / n_buffer_bytes;
if (apif->per_interface_next_index != ~0)
next_index = apif->per_interface_next_index;
n_free_bufs = vec_len (apm->rx_buffers[thread_index]);
if (PREDICT_FALSE (n_free_bufs < VLIB_FRAME_SIZE))
{
@ -302,6 +299,10 @@ af_packet_device_input_fn (vlib_main_t * vm, vlib_node_runtime_t * node,
else
{
next0 = VNET_DEVICE_INPUT_NEXT_ETHERNET_INPUT;
if (PREDICT_FALSE (apif->per_interface_next_index != ~0))
next0 = apif->per_interface_next_index;
/* redirect if feature path enabled */
vnet_feature_start_device_input_x1 (apif->sw_if_index, &next0,
first_b0);