[dpdk-users] rte_eth_dev_start failing after rte_eth_dev_stop: ixgbe

[Ori Zakin]

Hi,

If still relevant, I posted two possible workarounds for this on the dpdk-dev list, see link below.

http://dpdk.org/ml/archives/dev/2016-August/045096.html

Ori Zakin

Hello,

I am seeing strange behavior I haven't seen on some other other drivers.
When using the ixgbe driver with an I540 10Gb interface we see this
problem.

We have the traditional callback function:

kni_config_network_interface


Which gets called as interface is brought up/down.  Within that function it
will execute a stop/start of the rte_eth_dev_start / rte_eth_dev_stop.

The issue we see is once the ixgbe device is EVER stopped, when calling
start we get the following errors:

PMD: ixgbe_set_rx_function(): Vector rx enabled, please make sure RX burst
size no less than 4 (port=0).
PMD: ixgbe_alloc_rx_queue_mbufs(): RX mbuf alloc failed queue_id=0
PMD: ixgbe_dev_rx_queue_start(): Could not alloc mbuf for queue:0
PMD: ixgbe_dev_start(): Unable to start rxtx queues
PMD: ixgbe_dev_start(): failure in ixgbe_dev_start(): -1
APP: Failed to start port 0

Is this a known issue/bug/workaround (I can comment out all the start/stop
but that defeats the purpose).

Shawn