On 4/12/2018 8:41 AM, Matej Vido wrote:
> + if (pci_dev->id.device_id == PCI_DEVICE_ID_NETCOPE_NFB200G2QL) {
> + unsigned int i;
> + unsigned int rx_queues = max_rx_queues / max_ports;
> + unsigned int tx_queues = max_tx_queues / max_ports;
> +
> + /*
> + * Number of queues reported by szedata_ifaces_available()
> + * is the number of all queues from all DMA controllers which
> + * may reside at different numa locations.
> + * All queues from the same DMA controller have the same numa
> + * node.
> + * Numa node from the first queue of each DMA controller is
> + * retrieved.
> + * If the numa node differs from the numa node of the queues
> + * from the previous DMA controller the queues are assigned
> + * to the next port.
> + */
> +
> + for (i = 0; i < max_ports; i++) {
> + int numa_rx = szedata_get_area_numa_node(szedata_temp,
> + SZE2_DIR_RX, rx_queues * i);
> + int numa_tx = szedata_get_area_numa_node(szedata_temp,
Hi Matej,
Where szedata_get_area_numa_node() is defined?
Is it possible that you are missing a patch?
Thanks,
ferruh
