net/nfp: move PF functions into its own file
authorHeinrich Kuhn <heinrich.kuhn@netronome.com>
Thu, 29 Jul 2021 13:47:10 +0000 (15:47 +0200)
committerFerruh Yigit <ferruh.yigit@intel.com>
Tue, 17 Aug 2021 16:26:57 +0000 (18:26 +0200)
Similar to the last commit, this changeset moves all the PF specific
functions to a new file called nfp_ethdev.c.

Signed-off-by: Heinrich Kuhn <heinrich.kuhn@netronome.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
drivers/net/nfp/meson.build
drivers/net/nfp/nfp_ethdev.c [new file with mode: 0644]
drivers/net/nfp/nfp_net.c

index 34f4054..ab64d0c 100644 (file)
@@ -22,4 +22,5 @@ sources = files(
         'nfp_rxtx.c',
         'nfp_cpp_bridge.c',
         'nfp_ethdev_vf.c',
+        'nfp_ethdev.c',
 )
diff --git a/drivers/net/nfp/nfp_ethdev.c b/drivers/net/nfp/nfp_ethdev.c
new file mode 100644 (file)
index 0000000..8b6e641
--- /dev/null
@@ -0,0 +1,1067 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright (c) 2014-2021 Netronome Systems, Inc.
+ * All rights reserved.
+ *
+ * Small portions derived from code Copyright(c) 2010-2015 Intel Corporation.
+ */
+
+/*
+ * vim:shiftwidth=8:noexpandtab
+ *
+ * @file dpdk/pmd/nfp_ethdev.c
+ *
+ * Netronome vNIC DPDK Poll-Mode Driver: Main entry point
+ */
+
+#include <rte_common.h>
+#include <ethdev_driver.h>
+#include <ethdev_pci.h>
+#include <rte_dev.h>
+#include <rte_ether.h>
+#include <rte_malloc.h>
+#include <rte_memzone.h>
+#include <rte_mempool.h>
+#include <rte_service_component.h>
+#include "eal_firmware.h"
+
+#include "nfpcore/nfp_cpp.h"
+#include "nfpcore/nfp_nffw.h"
+#include "nfpcore/nfp_hwinfo.h"
+#include "nfpcore/nfp_mip.h"
+#include "nfpcore/nfp_rtsym.h"
+#include "nfpcore/nfp_nsp.h"
+
+#include "nfp_net_pmd.h"
+#include "nfp_rxtx.h"
+#include "nfp_net_logs.h"
+#include "nfp_net_ctrl.h"
+#include "nfp_cpp_bridge.h"
+
+
+static int nfp_net_pf_read_mac(struct nfp_pf_dev *pf_dev, int port);
+static int nfp_net_start(struct rte_eth_dev *dev);
+static int nfp_net_stop(struct rte_eth_dev *dev);
+static int nfp_net_set_link_up(struct rte_eth_dev *dev);
+static int nfp_net_set_link_down(struct rte_eth_dev *dev);
+static int nfp_net_close(struct rte_eth_dev *dev);
+static int nfp_net_init(struct rte_eth_dev *eth_dev);
+static int nfp_fw_upload(struct rte_pci_device *dev,
+                        struct nfp_nsp *nsp, char *card);
+static int nfp_fw_setup(struct rte_pci_device *dev,
+                       struct nfp_cpp *cpp,
+                       struct nfp_eth_table *nfp_eth_table,
+                       struct nfp_hwinfo *hwinfo);
+static int nfp_init_phyports(struct nfp_pf_dev *pf_dev);
+static int nfp_pf_init(struct rte_pci_device *pci_dev);
+static int nfp_pf_secondary_init(struct rte_pci_device *pci_dev);
+static int nfp_pf_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
+                           struct rte_pci_device *dev);
+static int nfp_pci_uninit(struct rte_eth_dev *eth_dev);
+static int eth_nfp_pci_remove(struct rte_pci_device *pci_dev);
+
+static int
+nfp_net_pf_read_mac(struct nfp_pf_dev *pf_dev, int port)
+{
+       struct nfp_eth_table *nfp_eth_table;
+       struct nfp_net_hw *hw = NULL;
+
+       /* Grab a pointer to the correct physical port */
+       hw = pf_dev->ports[port];
+
+       nfp_eth_table = nfp_eth_read_ports(pf_dev->cpp);
+
+       nfp_eth_copy_mac((uint8_t *)&hw->mac_addr,
+                        (uint8_t *)&nfp_eth_table->ports[port].mac_addr);
+
+       free(nfp_eth_table);
+       return 0;
+}
+
+static int
+nfp_net_start(struct rte_eth_dev *dev)
+{
+       struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
+       struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
+       uint32_t new_ctrl, update = 0;
+       struct nfp_net_hw *hw;
+       struct nfp_pf_dev *pf_dev;
+       struct rte_eth_conf *dev_conf;
+       struct rte_eth_rxmode *rxmode;
+       uint32_t intr_vector;
+       int ret;
+
+       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+       pf_dev = NFP_NET_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+
+       PMD_INIT_LOG(DEBUG, "Start");
+
+       /* Disabling queues just in case... */
+       nfp_net_disable_queues(dev);
+
+       /* Enabling the required queues in the device */
+       nfp_net_enable_queues(dev);
+
+       /* check and configure queue intr-vector mapping */
+       if (dev->data->dev_conf.intr_conf.rxq != 0) {
+               if (pf_dev->multiport) {
+                       PMD_INIT_LOG(ERR, "PMD rx interrupt is not supported "
+                                         "with NFP multiport PF");
+                               return -EINVAL;
+               }
+               if (intr_handle->type == RTE_INTR_HANDLE_UIO) {
+                       /*
+                        * Better not to share LSC with RX interrupts.
+                        * Unregistering LSC interrupt handler
+                        */
+                       rte_intr_callback_unregister(&pci_dev->intr_handle,
+                               nfp_net_dev_interrupt_handler, (void *)dev);
+
+                       if (dev->data->nb_rx_queues > 1) {
+                               PMD_INIT_LOG(ERR, "PMD rx interrupt only "
+                                            "supports 1 queue with UIO");
+                               return -EIO;
+                       }
+               }
+               intr_vector = dev->data->nb_rx_queues;
+               if (rte_intr_efd_enable(intr_handle, intr_vector))
+                       return -1;
+
+               nfp_configure_rx_interrupt(dev, intr_handle);
+               update = NFP_NET_CFG_UPDATE_MSIX;
+       }
+
+       rte_intr_enable(intr_handle);
+
+       new_ctrl = nfp_check_offloads(dev);
+
+       /* Writing configuration parameters in the device */
+       nfp_net_params_setup(hw);
+
+       dev_conf = &dev->data->dev_conf;
+       rxmode = &dev_conf->rxmode;
+
+       if (rxmode->mq_mode & ETH_MQ_RX_RSS) {
+               nfp_net_rss_config_default(dev);
+               update |= NFP_NET_CFG_UPDATE_RSS;
+               new_ctrl |= NFP_NET_CFG_CTRL_RSS;
+       }
+
+       /* Enable device */
+       new_ctrl |= NFP_NET_CFG_CTRL_ENABLE;
+
+       update |= NFP_NET_CFG_UPDATE_GEN | NFP_NET_CFG_UPDATE_RING;
+
+       if (hw->cap & NFP_NET_CFG_CTRL_RINGCFG)
+               new_ctrl |= NFP_NET_CFG_CTRL_RINGCFG;
+
+       nn_cfg_writel(hw, NFP_NET_CFG_CTRL, new_ctrl);
+       if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+               return -EIO;
+
+       /*
+        * Allocating rte mbufs for configured rx queues.
+        * This requires queues being enabled before
+        */
+       if (nfp_net_rx_freelist_setup(dev) < 0) {
+               ret = -ENOMEM;
+               goto error;
+       }
+
+       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
+               /* Configure the physical port up */
+               nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 1);
+       else
+               nfp_eth_set_configured(dev->process_private,
+                                      hw->nfp_idx, 1);
+
+       hw->ctrl = new_ctrl;
+
+       return 0;
+
+error:
+       /*
+        * An error returned by this function should mean the app
+        * exiting and then the system releasing all the memory
+        * allocated even memory coming from hugepages.
+        *
+        * The device could be enabled at this point with some queues
+        * ready for getting packets. This is true if the call to
+        * nfp_net_rx_freelist_setup() succeeds for some queues but
+        * fails for subsequent queues.
+        *
+        * This should make the app exiting but better if we tell the
+        * device first.
+        */
+       nfp_net_disable_queues(dev);
+
+       return ret;
+}
+
+/* Stop device: disable rx and tx functions to allow for reconfiguring. */
+static int
+nfp_net_stop(struct rte_eth_dev *dev)
+{
+       int i;
+       struct nfp_net_hw *hw;
+       struct nfp_net_txq *this_tx_q;
+       struct nfp_net_rxq *this_rx_q;
+
+       PMD_INIT_LOG(DEBUG, "Stop");
+
+       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+       nfp_net_disable_queues(dev);
+
+       /* Clear queues */
+       for (i = 0; i < dev->data->nb_tx_queues; i++) {
+               this_tx_q = (struct nfp_net_txq *)dev->data->tx_queues[i];
+               nfp_net_reset_tx_queue(this_tx_q);
+       }
+
+       for (i = 0; i < dev->data->nb_rx_queues; i++) {
+               this_rx_q = (struct nfp_net_rxq *)dev->data->rx_queues[i];
+               nfp_net_reset_rx_queue(this_rx_q);
+       }
+
+       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
+               /* Configure the physical port down */
+               nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 0);
+       else
+               nfp_eth_set_configured(dev->process_private,
+                                      hw->nfp_idx, 0);
+
+       return 0;
+}
+
+/* Set the link up. */
+static int
+nfp_net_set_link_up(struct rte_eth_dev *dev)
+{
+       struct nfp_net_hw *hw;
+
+       PMD_DRV_LOG(DEBUG, "Set link up");
+
+       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
+               /* Configure the physical port down */
+               return nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 1);
+       else
+               return nfp_eth_set_configured(dev->process_private,
+                                             hw->nfp_idx, 1);
+}
+
+/* Set the link down. */
+static int
+nfp_net_set_link_down(struct rte_eth_dev *dev)
+{
+       struct nfp_net_hw *hw;
+
+       PMD_DRV_LOG(DEBUG, "Set link down");
+
+       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
+               /* Configure the physical port down */
+               return nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 0);
+       else
+               return nfp_eth_set_configured(dev->process_private,
+                                             hw->nfp_idx, 0);
+}
+
+/* Reset and stop device. The device can not be restarted. */
+static int
+nfp_net_close(struct rte_eth_dev *dev)
+{
+       struct nfp_net_hw *hw;
+       struct rte_pci_device *pci_dev;
+       struct nfp_pf_dev *pf_dev;
+       struct nfp_net_txq *this_tx_q;
+       struct nfp_net_rxq *this_rx_q;
+       int i;
+
+       if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+               return 0;
+
+       PMD_INIT_LOG(DEBUG, "Close");
+
+       pf_dev = NFP_NET_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+       pci_dev = RTE_ETH_DEV_TO_PCI(dev);
+
+       /*
+        * We assume that the DPDK application is stopping all the
+        * threads/queues before calling the device close function.
+        */
+
+       nfp_net_disable_queues(dev);
+
+       /* Clear queues */
+       for (i = 0; i < dev->data->nb_tx_queues; i++) {
+               this_tx_q = (struct nfp_net_txq *)dev->data->tx_queues[i];
+               nfp_net_reset_tx_queue(this_tx_q);
+       }
+
+       for (i = 0; i < dev->data->nb_rx_queues; i++) {
+               this_rx_q = (struct nfp_net_rxq *)dev->data->rx_queues[i];
+               nfp_net_reset_rx_queue(this_rx_q);
+       }
+
+       /* Only free PF resources after all physical ports have been closed */
+       /* Mark this port as unused and free device priv resources*/
+       nn_cfg_writeb(hw, NFP_NET_CFG_LSC, 0xff);
+       pf_dev->ports[hw->idx] = NULL;
+       rte_eth_dev_release_port(dev);
+
+       for (i = 0; i < pf_dev->total_phyports; i++) {
+               /* Check to see if ports are still in use */
+               if (pf_dev->ports[i])
+                       return 0;
+       }
+
+       /* Now it is safe to free all PF resources */
+       PMD_INIT_LOG(INFO, "Freeing PF resources");
+       nfp_cpp_area_free(pf_dev->ctrl_area);
+       nfp_cpp_area_free(pf_dev->hwqueues_area);
+       free(pf_dev->hwinfo);
+       free(pf_dev->sym_tbl);
+       nfp_cpp_free(pf_dev->cpp);
+       rte_free(pf_dev);
+
+       rte_intr_disable(&pci_dev->intr_handle);
+
+       /* unregister callback func from eal lib */
+       rte_intr_callback_unregister(&pci_dev->intr_handle,
+                                    nfp_net_dev_interrupt_handler,
+                                    (void *)dev);
+
+       /*
+        * The ixgbe PMD driver disables the pcie master on the
+        * device. The i40e does not...
+        */
+
+       return 0;
+}
+
+/* Initialise and register driver with DPDK Application */
+static const struct eth_dev_ops nfp_net_eth_dev_ops = {
+       .dev_configure          = nfp_net_configure,
+       .dev_start              = nfp_net_start,
+       .dev_stop               = nfp_net_stop,
+       .dev_set_link_up        = nfp_net_set_link_up,
+       .dev_set_link_down      = nfp_net_set_link_down,
+       .dev_close              = nfp_net_close,
+       .promiscuous_enable     = nfp_net_promisc_enable,
+       .promiscuous_disable    = nfp_net_promisc_disable,
+       .link_update            = nfp_net_link_update,
+       .stats_get              = nfp_net_stats_get,
+       .stats_reset            = nfp_net_stats_reset,
+       .dev_infos_get          = nfp_net_infos_get,
+       .dev_supported_ptypes_get = nfp_net_supported_ptypes_get,
+       .mtu_set                = nfp_net_dev_mtu_set,
+       .mac_addr_set           = nfp_set_mac_addr,
+       .vlan_offload_set       = nfp_net_vlan_offload_set,
+       .reta_update            = nfp_net_reta_update,
+       .reta_query             = nfp_net_reta_query,
+       .rss_hash_update        = nfp_net_rss_hash_update,
+       .rss_hash_conf_get      = nfp_net_rss_hash_conf_get,
+       .rx_queue_setup         = nfp_net_rx_queue_setup,
+       .rx_queue_release       = nfp_net_rx_queue_release,
+       .tx_queue_setup         = nfp_net_tx_queue_setup,
+       .tx_queue_release       = nfp_net_tx_queue_release,
+       .rx_queue_intr_enable   = nfp_rx_queue_intr_enable,
+       .rx_queue_intr_disable  = nfp_rx_queue_intr_disable,
+};
+
+static int
+nfp_net_init(struct rte_eth_dev *eth_dev)
+{
+       struct rte_pci_device *pci_dev;
+       struct nfp_pf_dev *pf_dev;
+       struct nfp_net_hw *hw;
+       struct rte_ether_addr *tmp_ether_addr;
+
+       uint64_t tx_bar_off = 0, rx_bar_off = 0;
+       uint32_t start_q;
+       int stride = 4;
+       int port = 0;
+       int err;
+
+       PMD_INIT_FUNC_TRACE();
+
+       pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
+
+       /* Use backpointer here to the PF of this eth_dev */
+       pf_dev = NFP_NET_DEV_PRIVATE_TO_PF(eth_dev->data->dev_private);
+
+       /* NFP can not handle DMA addresses requiring more than 40 bits */
+       if (rte_mem_check_dma_mask(40)) {
+               RTE_LOG(ERR, PMD, "device %s can not be used:",
+                                  pci_dev->device.name);
+               RTE_LOG(ERR, PMD, "\trestricted dma mask to 40 bits!\n");
+               return -ENODEV;
+       };
+
+       port = ((struct nfp_net_hw *)eth_dev->data->dev_private)->idx;
+       if (port < 0 || port > 7) {
+               PMD_DRV_LOG(ERR, "Port value is wrong");
+               return -ENODEV;
+       }
+
+       /* Use PF array of physical ports to get pointer to
+        * this specific port
+        */
+       hw = pf_dev->ports[port];
+
+       PMD_INIT_LOG(DEBUG, "Working with physical port number: %d, "
+                           "NFP internal port number: %d",
+                           port, hw->nfp_idx);
+
+       eth_dev->dev_ops = &nfp_net_eth_dev_ops;
+       eth_dev->rx_queue_count = nfp_net_rx_queue_count;
+       eth_dev->rx_pkt_burst = &nfp_net_recv_pkts;
+       eth_dev->tx_pkt_burst = &nfp_net_xmit_pkts;
+
+       /* For secondary processes, the primary has done all the work */
+       if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+               return 0;
+
+       rte_eth_copy_pci_info(eth_dev, pci_dev);
+
+       hw->device_id = pci_dev->id.device_id;
+       hw->vendor_id = pci_dev->id.vendor_id;
+       hw->subsystem_device_id = pci_dev->id.subsystem_device_id;
+       hw->subsystem_vendor_id = pci_dev->id.subsystem_vendor_id;
+
+       PMD_INIT_LOG(DEBUG, "nfp_net: device (%u:%u) %u:%u:%u:%u",
+                    pci_dev->id.vendor_id, pci_dev->id.device_id,
+                    pci_dev->addr.domain, pci_dev->addr.bus,
+                    pci_dev->addr.devid, pci_dev->addr.function);
+
+       hw->ctrl_bar = (uint8_t *)pci_dev->mem_resource[0].addr;
+       if (hw->ctrl_bar == NULL) {
+               PMD_DRV_LOG(ERR,
+                       "hw->ctrl_bar is NULL. BAR0 not configured");
+               return -ENODEV;
+       }
+
+       if (port == 0) {
+               hw->ctrl_bar = pf_dev->ctrl_bar;
+       } else {
+               if (!pf_dev->ctrl_bar)
+                       return -ENODEV;
+               /* Use port offset in pf ctrl_bar for this
+                * ports control bar
+                */
+               hw->ctrl_bar = pf_dev->ctrl_bar +
+                              (port * NFP_PF_CSR_SLICE_SIZE);
+       }
+
+       PMD_INIT_LOG(DEBUG, "ctrl bar: %p", hw->ctrl_bar);
+
+       hw->max_rx_queues = nn_cfg_readl(hw, NFP_NET_CFG_MAX_RXRINGS);
+       hw->max_tx_queues = nn_cfg_readl(hw, NFP_NET_CFG_MAX_TXRINGS);
+
+       /* Work out where in the BAR the queues start. */
+       switch (pci_dev->id.device_id) {
+       case PCI_DEVICE_ID_NFP4000_PF_NIC:
+       case PCI_DEVICE_ID_NFP6000_PF_NIC:
+               start_q = nn_cfg_readl(hw, NFP_NET_CFG_START_TXQ);
+               tx_bar_off = (uint64_t)start_q * NFP_QCP_QUEUE_ADDR_SZ;
+               start_q = nn_cfg_readl(hw, NFP_NET_CFG_START_RXQ);
+               rx_bar_off = (uint64_t)start_q * NFP_QCP_QUEUE_ADDR_SZ;
+               break;
+       default:
+               PMD_DRV_LOG(ERR, "nfp_net: no device ID matching");
+               err = -ENODEV;
+               goto dev_err_ctrl_map;
+       }
+
+       PMD_INIT_LOG(DEBUG, "tx_bar_off: 0x%" PRIx64 "", tx_bar_off);
+       PMD_INIT_LOG(DEBUG, "rx_bar_off: 0x%" PRIx64 "", rx_bar_off);
+
+       hw->tx_bar = pf_dev->hw_queues + tx_bar_off;
+       hw->rx_bar = pf_dev->hw_queues + rx_bar_off;
+       eth_dev->data->dev_private = hw;
+
+       PMD_INIT_LOG(DEBUG, "ctrl_bar: %p, tx_bar: %p, rx_bar: %p",
+                    hw->ctrl_bar, hw->tx_bar, hw->rx_bar);
+
+       nfp_net_cfg_queue_setup(hw);
+
+       /* Get some of the read-only fields from the config BAR */
+       hw->ver = nn_cfg_readl(hw, NFP_NET_CFG_VERSION);
+       hw->cap = nn_cfg_readl(hw, NFP_NET_CFG_CAP);
+       hw->max_mtu = nn_cfg_readl(hw, NFP_NET_CFG_MAX_MTU);
+       hw->mtu = RTE_ETHER_MTU;
+
+       /* VLAN insertion is incompatible with LSOv2 */
+       if (hw->cap & NFP_NET_CFG_CTRL_LSO2)
+               hw->cap &= ~NFP_NET_CFG_CTRL_TXVLAN;
+
+       if (NFD_CFG_MAJOR_VERSION_of(hw->ver) < 2)
+               hw->rx_offset = NFP_NET_RX_OFFSET;
+       else
+               hw->rx_offset = nn_cfg_readl(hw, NFP_NET_CFG_RX_OFFSET_ADDR);
+
+       PMD_INIT_LOG(INFO, "VER: %u.%u, Maximum supported MTU: %d",
+                          NFD_CFG_MAJOR_VERSION_of(hw->ver),
+                          NFD_CFG_MINOR_VERSION_of(hw->ver), hw->max_mtu);
+
+       PMD_INIT_LOG(INFO, "CAP: %#x, %s%s%s%s%s%s%s%s%s%s%s%s%s%s", hw->cap,
+                    hw->cap & NFP_NET_CFG_CTRL_PROMISC ? "PROMISC " : "",
+                    hw->cap & NFP_NET_CFG_CTRL_L2BC    ? "L2BCFILT " : "",
+                    hw->cap & NFP_NET_CFG_CTRL_L2MC    ? "L2MCFILT " : "",
+                    hw->cap & NFP_NET_CFG_CTRL_RXCSUM  ? "RXCSUM "  : "",
+                    hw->cap & NFP_NET_CFG_CTRL_TXCSUM  ? "TXCSUM "  : "",
+                    hw->cap & NFP_NET_CFG_CTRL_RXVLAN  ? "RXVLAN "  : "",
+                    hw->cap & NFP_NET_CFG_CTRL_TXVLAN  ? "TXVLAN "  : "",
+                    hw->cap & NFP_NET_CFG_CTRL_SCATTER ? "SCATTER " : "",
+                    hw->cap & NFP_NET_CFG_CTRL_GATHER  ? "GATHER "  : "",
+                    hw->cap & NFP_NET_CFG_CTRL_LIVE_ADDR ? "LIVE_ADDR "  : "",
+                    hw->cap & NFP_NET_CFG_CTRL_LSO     ? "TSO "     : "",
+                    hw->cap & NFP_NET_CFG_CTRL_LSO2     ? "TSOv2 "     : "",
+                    hw->cap & NFP_NET_CFG_CTRL_RSS     ? "RSS "     : "",
+                    hw->cap & NFP_NET_CFG_CTRL_RSS2     ? "RSSv2 "     : "");
+
+       hw->ctrl = 0;
+
+       hw->stride_rx = stride;
+       hw->stride_tx = stride;
+
+       PMD_INIT_LOG(INFO, "max_rx_queues: %u, max_tx_queues: %u",
+                    hw->max_rx_queues, hw->max_tx_queues);
+
+       /* Initializing spinlock for reconfigs */
+       rte_spinlock_init(&hw->reconfig_lock);
+
+       /* Allocating memory for mac addr */
+       eth_dev->data->mac_addrs = rte_zmalloc("mac_addr",
+                                              RTE_ETHER_ADDR_LEN, 0);
+       if (eth_dev->data->mac_addrs == NULL) {
+               PMD_INIT_LOG(ERR, "Failed to space for MAC address");
+               err = -ENOMEM;
+               goto dev_err_queues_map;
+       }
+
+       nfp_net_pf_read_mac(pf_dev, port);
+       nfp_net_write_mac(hw, (uint8_t *)&hw->mac_addr);
+
+       tmp_ether_addr = (struct rte_ether_addr *)&hw->mac_addr;
+       if (!rte_is_valid_assigned_ether_addr(tmp_ether_addr)) {
+               PMD_INIT_LOG(INFO, "Using random mac address for port %d",
+                                  port);
+               /* Using random mac addresses for VFs */
+               rte_eth_random_addr(&hw->mac_addr[0]);
+               nfp_net_write_mac(hw, (uint8_t *)&hw->mac_addr);
+       }
+
+       /* Copying mac address to DPDK eth_dev struct */
+       rte_ether_addr_copy((struct rte_ether_addr *)hw->mac_addr,
+                       &eth_dev->data->mac_addrs[0]);
+
+       if (!(hw->cap & NFP_NET_CFG_CTRL_LIVE_ADDR))
+               eth_dev->data->dev_flags |= RTE_ETH_DEV_NOLIVE_MAC_ADDR;
+
+       eth_dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS;
+
+       PMD_INIT_LOG(INFO, "port %d VendorID=0x%x DeviceID=0x%x "
+                    "mac=%02x:%02x:%02x:%02x:%02x:%02x",
+                    eth_dev->data->port_id, pci_dev->id.vendor_id,
+                    pci_dev->id.device_id,
+                    hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
+                    hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
+
+       if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
+               /* Registering LSC interrupt handler */
+               rte_intr_callback_register(&pci_dev->intr_handle,
+                                          nfp_net_dev_interrupt_handler,
+                                          (void *)eth_dev);
+               /* Telling the firmware about the LSC interrupt entry */
+               nn_cfg_writeb(hw, NFP_NET_CFG_LSC, NFP_NET_IRQ_LSC_IDX);
+               /* Recording current stats counters values */
+               nfp_net_stats_reset(eth_dev);
+       }
+
+       return 0;
+
+dev_err_queues_map:
+               nfp_cpp_area_free(hw->hwqueues_area);
+dev_err_ctrl_map:
+               nfp_cpp_area_free(hw->ctrl_area);
+
+       return err;
+}
+
+#define DEFAULT_FW_PATH       "/lib/firmware/netronome"
+
+static int
+nfp_fw_upload(struct rte_pci_device *dev, struct nfp_nsp *nsp, char *card)
+{
+       struct nfp_cpp *cpp = nsp->cpp;
+       void *fw_buf;
+       char fw_name[125];
+       char serial[40];
+       size_t fsize;
+
+       /* Looking for firmware file in order of priority */
+
+       /* First try to find a firmware image specific for this device */
+       snprintf(serial, sizeof(serial),
+                       "serial-%02x-%02x-%02x-%02x-%02x-%02x-%02x-%02x",
+               cpp->serial[0], cpp->serial[1], cpp->serial[2], cpp->serial[3],
+               cpp->serial[4], cpp->serial[5], cpp->interface >> 8,
+               cpp->interface & 0xff);
+
+       snprintf(fw_name, sizeof(fw_name), "%s/%s.nffw", DEFAULT_FW_PATH,
+                       serial);
+
+       PMD_DRV_LOG(DEBUG, "Trying with fw file: %s", fw_name);
+       if (rte_firmware_read(fw_name, &fw_buf, &fsize) == 0)
+               goto load_fw;
+       /* Then try the PCI name */
+       snprintf(fw_name, sizeof(fw_name), "%s/pci-%s.nffw", DEFAULT_FW_PATH,
+                       dev->device.name);
+
+       PMD_DRV_LOG(DEBUG, "Trying with fw file: %s", fw_name);
+       if (rte_firmware_read(fw_name, &fw_buf, &fsize) == 0)
+               goto load_fw;
+
+       /* Finally try the card type and media */
+       snprintf(fw_name, sizeof(fw_name), "%s/%s", DEFAULT_FW_PATH, card);
+       PMD_DRV_LOG(DEBUG, "Trying with fw file: %s", fw_name);
+       if (rte_firmware_read(fw_name, &fw_buf, &fsize) < 0) {
+               PMD_DRV_LOG(INFO, "Firmware file %s not found.", fw_name);
+               return -ENOENT;
+       }
+
+load_fw:
+       PMD_DRV_LOG(INFO, "Firmware file found at %s with size: %zu",
+               fw_name, fsize);
+       PMD_DRV_LOG(INFO, "Uploading the firmware ...");
+       nfp_nsp_load_fw(nsp, fw_buf, fsize);
+       PMD_DRV_LOG(INFO, "Done");
+
+       free(fw_buf);
+
+       return 0;
+}
+
+static int
+nfp_fw_setup(struct rte_pci_device *dev, struct nfp_cpp *cpp,
+            struct nfp_eth_table *nfp_eth_table, struct nfp_hwinfo *hwinfo)
+{
+       struct nfp_nsp *nsp;
+       const char *nfp_fw_model;
+       char card_desc[100];
+       int err = 0;
+
+       nfp_fw_model = nfp_hwinfo_lookup(hwinfo, "assembly.partno");
+
+       if (nfp_fw_model) {
+               PMD_DRV_LOG(INFO, "firmware model found: %s", nfp_fw_model);
+       } else {
+               PMD_DRV_LOG(ERR, "firmware model NOT found");
+               return -EIO;
+       }
+
+       if (nfp_eth_table->count == 0 || nfp_eth_table->count > 8) {
+               PMD_DRV_LOG(ERR, "NFP ethernet table reports wrong ports: %u",
+                      nfp_eth_table->count);
+               return -EIO;
+       }
+
+       PMD_DRV_LOG(INFO, "NFP ethernet port table reports %u ports",
+                          nfp_eth_table->count);
+
+       PMD_DRV_LOG(INFO, "Port speed: %u", nfp_eth_table->ports[0].speed);
+
+       snprintf(card_desc, sizeof(card_desc), "nic_%s_%dx%d.nffw",
+                       nfp_fw_model, nfp_eth_table->count,
+                       nfp_eth_table->ports[0].speed / 1000);
+
+       nsp = nfp_nsp_open(cpp);
+       if (!nsp) {
+               PMD_DRV_LOG(ERR, "NFP error when obtaining NSP handle");
+               return -EIO;
+       }
+
+       nfp_nsp_device_soft_reset(nsp);
+       err = nfp_fw_upload(dev, nsp, card_desc);
+
+       nfp_nsp_close(nsp);
+       return err;
+}
+
+static int nfp_init_phyports(struct nfp_pf_dev *pf_dev)
+{
+       struct nfp_net_hw *hw;
+       struct rte_eth_dev *eth_dev;
+       struct nfp_eth_table *nfp_eth_table = NULL;
+       int ret = 0;
+       int i;
+
+       nfp_eth_table = nfp_eth_read_ports(pf_dev->cpp);
+       if (!nfp_eth_table) {
+               PMD_INIT_LOG(ERR, "Error reading NFP ethernet table");
+               ret = -EIO;
+               goto error;
+       }
+
+       /* Loop through all physical ports on PF */
+       for (i = 0; i < pf_dev->total_phyports; i++) {
+               const unsigned int numa_node = rte_socket_id();
+               char port_name[RTE_ETH_NAME_MAX_LEN];
+
+               snprintf(port_name, sizeof(port_name), "%s_port%d",
+                        pf_dev->pci_dev->device.name, i);
+
+               /* Allocate a eth_dev for this phyport */
+               eth_dev = rte_eth_dev_allocate(port_name);
+               if (!eth_dev) {
+                       ret = -ENODEV;
+                       goto port_cleanup;
+               }
+
+               /* Allocate memory for this phyport */
+               eth_dev->data->dev_private =
+                       rte_zmalloc_socket(port_name, sizeof(struct nfp_net_hw),
+                                          RTE_CACHE_LINE_SIZE, numa_node);
+               if (!eth_dev->data->dev_private) {
+                       ret = -ENOMEM;
+                       rte_eth_dev_release_port(eth_dev);
+                       goto port_cleanup;
+               }
+
+               hw = NFP_NET_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
+
+               /* Add this device to the PF's array of physical ports */
+               pf_dev->ports[i] = hw;
+
+               hw->pf_dev = pf_dev;
+               hw->cpp = pf_dev->cpp;
+               hw->eth_dev = eth_dev;
+               hw->idx = i;
+               hw->nfp_idx = nfp_eth_table->ports[i].index;
+               hw->is_phyport = true;
+
+               eth_dev->device = &pf_dev->pci_dev->device;
+
+               /* ctrl/tx/rx BAR mappings and remaining init happens in
+                * nfp_net_init
+                */
+               ret = nfp_net_init(eth_dev);
+
+               if (ret) {
+                       ret = -ENODEV;
+                       goto port_cleanup;
+               }
+
+               rte_eth_dev_probing_finish(eth_dev);
+
+       } /* End loop, all ports on this PF */
+       ret = 0;
+       goto eth_table_cleanup;
+
+port_cleanup:
+       for (i = 0; i < pf_dev->total_phyports; i++) {
+               if (pf_dev->ports[i] && pf_dev->ports[i]->eth_dev) {
+                       struct rte_eth_dev *tmp_dev;
+                       tmp_dev = pf_dev->ports[i]->eth_dev;
+                       rte_eth_dev_release_port(tmp_dev);
+                       pf_dev->ports[i] = NULL;
+               }
+       }
+eth_table_cleanup:
+       free(nfp_eth_table);
+error:
+       return ret;
+}
+
+static int nfp_pf_init(struct rte_pci_device *pci_dev)
+{
+       struct nfp_pf_dev *pf_dev = NULL;
+       struct nfp_cpp *cpp;
+       struct nfp_hwinfo *hwinfo;
+       struct nfp_rtsym_table *sym_tbl;
+       struct nfp_eth_table *nfp_eth_table = NULL;
+       char name[RTE_ETH_NAME_MAX_LEN];
+       int total_ports;
+       int ret = -ENODEV;
+       int err;
+
+       if (!pci_dev)
+               return ret;
+
+       /*
+        * When device bound to UIO, the device could be used, by mistake,
+        * by two DPDK apps, and the UIO driver does not avoid it. This
+        * could lead to a serious problem when configuring the NFP CPP
+        * interface. Here we avoid this telling to the CPP init code to
+        * use a lock file if UIO is being used.
+        */
+       if (pci_dev->kdrv == RTE_PCI_KDRV_VFIO)
+               cpp = nfp_cpp_from_device_name(pci_dev, 0);
+       else
+               cpp = nfp_cpp_from_device_name(pci_dev, 1);
+
+       if (!cpp) {
+               PMD_INIT_LOG(ERR, "A CPP handle can not be obtained");
+               ret = -EIO;
+               goto error;
+       }
+
+       hwinfo = nfp_hwinfo_read(cpp);
+       if (!hwinfo) {
+               PMD_INIT_LOG(ERR, "Error reading hwinfo table");
+               ret = -EIO;
+               goto error;
+       }
+
+       nfp_eth_table = nfp_eth_read_ports(cpp);
+       if (!nfp_eth_table) {
+               PMD_INIT_LOG(ERR, "Error reading NFP ethernet table");
+               ret = -EIO;
+               goto hwinfo_cleanup;
+       }
+
+       if (nfp_fw_setup(pci_dev, cpp, nfp_eth_table, hwinfo)) {
+               PMD_INIT_LOG(ERR, "Error when uploading firmware");
+               ret = -EIO;
+               goto eth_table_cleanup;
+       }
+
+       /* Now the symbol table should be there */
+       sym_tbl = nfp_rtsym_table_read(cpp);
+       if (!sym_tbl) {
+               PMD_INIT_LOG(ERR, "Something is wrong with the firmware"
+                               " symbol table");
+               ret = -EIO;
+               goto eth_table_cleanup;
+       }
+
+       total_ports = nfp_rtsym_read_le(sym_tbl, "nfd_cfg_pf0_num_ports", &err);
+       if (total_ports != (int)nfp_eth_table->count) {
+               PMD_DRV_LOG(ERR, "Inconsistent number of ports");
+               ret = -EIO;
+               goto sym_tbl_cleanup;
+       }
+
+       PMD_INIT_LOG(INFO, "Total physical ports: %d", total_ports);
+
+       if (total_ports <= 0 || total_ports > 8) {
+               PMD_INIT_LOG(ERR, "nfd_cfg_pf0_num_ports symbol with wrong value");
+               ret = -ENODEV;
+               goto sym_tbl_cleanup;
+       }
+       /* Allocate memory for the PF "device" */
+       snprintf(name, sizeof(name), "nfp_pf%d", 0);
+       pf_dev = rte_zmalloc(name, sizeof(*pf_dev), 0);
+       if (!pf_dev) {
+               ret = -ENOMEM;
+               goto sym_tbl_cleanup;
+       }
+
+       /* Populate the newly created PF device */
+       pf_dev->cpp = cpp;
+       pf_dev->hwinfo = hwinfo;
+       pf_dev->sym_tbl = sym_tbl;
+       pf_dev->total_phyports = total_ports;
+
+       if (total_ports > 1)
+               pf_dev->multiport = true;
+
+       pf_dev->pci_dev = pci_dev;
+
+       /* Map the symbol table */
+       pf_dev->ctrl_bar = nfp_rtsym_map(pf_dev->sym_tbl, "_pf0_net_bar0",
+                                    pf_dev->total_phyports * 32768,
+                                    &pf_dev->ctrl_area);
+       if (!pf_dev->ctrl_bar) {
+               PMD_INIT_LOG(ERR, "nfp_rtsym_map fails for _pf0_net_ctrl_bar");
+               ret = -EIO;
+               goto pf_cleanup;
+       }
+
+       PMD_INIT_LOG(DEBUG, "ctrl bar: %p", pf_dev->ctrl_bar);
+
+       /* configure access to tx/rx vNIC BARs */
+       pf_dev->hw_queues = nfp_cpp_map_area(pf_dev->cpp, 0, 0,
+                                             NFP_PCIE_QUEUE(0),
+                                             NFP_QCP_QUEUE_AREA_SZ,
+                                             &pf_dev->hwqueues_area);
+       if (!pf_dev->hw_queues) {
+               PMD_INIT_LOG(ERR, "nfp_rtsym_map fails for net.qc");
+               ret = -EIO;
+               goto ctrl_area_cleanup;
+       }
+
+       PMD_INIT_LOG(DEBUG, "tx/rx bar address: 0x%p", pf_dev->hw_queues);
+
+       /* Initialize and prep physical ports now
+        * This will loop through all physical ports
+        */
+       ret = nfp_init_phyports(pf_dev);
+       if (ret) {
+               PMD_INIT_LOG(ERR, "Could not create physical ports");
+               goto hwqueues_cleanup;
+       }
+
+       /* register the CPP bridge service here for primary use */
+       nfp_register_cpp_service(pf_dev->cpp);
+
+       return 0;
+
+hwqueues_cleanup:
+       nfp_cpp_area_free(pf_dev->hwqueues_area);
+ctrl_area_cleanup:
+       nfp_cpp_area_free(pf_dev->ctrl_area);
+pf_cleanup:
+       rte_free(pf_dev);
+sym_tbl_cleanup:
+       free(sym_tbl);
+eth_table_cleanup:
+       free(nfp_eth_table);
+hwinfo_cleanup:
+       free(hwinfo);
+error:
+       return ret;
+}
+
+/*
+ * When attaching to the NFP4000/6000 PF on a secondary process there
+ * is no need to initialise the PF again. Only minimal work is required
+ * here
+ */
+static int nfp_pf_secondary_init(struct rte_pci_device *pci_dev)
+{
+       struct nfp_cpp *cpp;
+       struct nfp_rtsym_table *sym_tbl;
+       int total_ports;
+       int i;
+       int err;
+
+       if (!pci_dev)
+               return -ENODEV;
+
+       /*
+        * When device bound to UIO, the device could be used, by mistake,
+        * by two DPDK apps, and the UIO driver does not avoid it. This
+        * could lead to a serious problem when configuring the NFP CPP
+        * interface. Here we avoid this telling to the CPP init code to
+        * use a lock file if UIO is being used.
+        */
+       if (pci_dev->kdrv == RTE_PCI_KDRV_VFIO)
+               cpp = nfp_cpp_from_device_name(pci_dev, 0);
+       else
+               cpp = nfp_cpp_from_device_name(pci_dev, 1);
+
+       if (!cpp) {
+               PMD_INIT_LOG(ERR, "A CPP handle can not be obtained");
+               return -EIO;
+       }
+
+       /*
+        * We don't have access to the PF created in the primary process
+        * here so we have to read the number of ports from firmware
+        */
+       sym_tbl = nfp_rtsym_table_read(cpp);
+       if (!sym_tbl) {
+               PMD_INIT_LOG(ERR, "Something is wrong with the firmware"
+                               " symbol table");
+               return -EIO;
+       }
+
+       total_ports = nfp_rtsym_read_le(sym_tbl, "nfd_cfg_pf0_num_ports", &err);
+
+       for (i = 0; i < total_ports; i++) {
+               struct rte_eth_dev *eth_dev;
+               char port_name[RTE_ETH_NAME_MAX_LEN];
+
+               snprintf(port_name, sizeof(port_name), "%s_port%d",
+                        pci_dev->device.name, i);
+
+               PMD_DRV_LOG(DEBUG, "Secondary attaching to port %s",
+                   port_name);
+               eth_dev = rte_eth_dev_attach_secondary(port_name);
+               if (!eth_dev) {
+                       RTE_LOG(ERR, EAL,
+                       "secondary process attach failed, "
+                       "ethdev doesn't exist");
+                       return -ENODEV;
+               }
+               eth_dev->process_private = cpp;
+               eth_dev->dev_ops = &nfp_net_eth_dev_ops;
+               eth_dev->rx_queue_count = nfp_net_rx_queue_count;
+               eth_dev->rx_pkt_burst = &nfp_net_recv_pkts;
+               eth_dev->tx_pkt_burst = &nfp_net_xmit_pkts;
+               rte_eth_dev_probing_finish(eth_dev);
+       }
+
+       /* Register the CPP bridge service for the secondary too */
+       nfp_register_cpp_service(cpp);
+
+       return 0;
+}
+
+static int nfp_pf_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
+                           struct rte_pci_device *dev)
+{
+       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
+               return nfp_pf_init(dev);
+       else
+               return nfp_pf_secondary_init(dev);
+}
+
+static const struct rte_pci_id pci_id_nfp_pf_net_map[] = {
+       {
+               RTE_PCI_DEVICE(PCI_VENDOR_ID_NETRONOME,
+                              PCI_DEVICE_ID_NFP4000_PF_NIC)
+       },
+       {
+               RTE_PCI_DEVICE(PCI_VENDOR_ID_NETRONOME,
+                              PCI_DEVICE_ID_NFP6000_PF_NIC)
+       },
+       {
+               .vendor_id = 0,
+       },
+};
+
+static int nfp_pci_uninit(struct rte_eth_dev *eth_dev)
+{
+       struct rte_pci_device *pci_dev;
+       uint16_t port_id;
+
+       pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
+
+       /* Free up all physical ports under PF */
+       RTE_ETH_FOREACH_DEV_OF(port_id, &pci_dev->device)
+               rte_eth_dev_close(port_id);
+       /*
+        * Ports can be closed and freed but hotplugging is not
+        * currently supported
+        */
+       return -ENOTSUP;
+}
+
+static int eth_nfp_pci_remove(struct rte_pci_device *pci_dev)
+{
+       return rte_eth_dev_pci_generic_remove(pci_dev, nfp_pci_uninit);
+}
+
+static struct rte_pci_driver rte_nfp_net_pf_pmd = {
+       .id_table = pci_id_nfp_pf_net_map,
+       .drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC,
+       .probe = nfp_pf_pci_probe,
+       .remove = eth_nfp_pci_remove,
+};
+
+RTE_PMD_REGISTER_PCI(net_nfp_pf, rte_nfp_net_pf_pmd);
+RTE_PMD_REGISTER_PCI_TABLE(net_nfp_pf, pci_id_nfp_pf_net_map);
+RTE_PMD_REGISTER_KMOD_DEP(net_nfp_pf, "* igb_uio | uio_pci_generic | vfio");
+/*
+ * Local variables:
+ * c-file-style: "Linux"
+ * indent-tabs-mode: t
+ * End:
+ */
index a46c4dd..04288a3 100644 (file)
@@ -30,8 +30,6 @@
 #include <rte_spinlock.h>
 #include <rte_service_component.h>
 
-#include "eal_firmware.h"
-
 #include "nfpcore/nfp_cpp.h"
 #include "nfpcore/nfp_nffw.h"
 #include "nfpcore/nfp_hwinfo.h"
 #include "nfpcore/nfp_rtsym.h"
 #include "nfpcore/nfp_nsp.h"
 
-#include "nfp_net_pmd.h"
+#include "nfp_common.h"
 #include "nfp_rxtx.h"
-#include "nfp_net_logs.h"
-#include "nfp_net_ctrl.h"
+#include "nfp_logs.h"
+#include "nfp_ctrl.h"
 #include "nfp_cpp_bridge.h"
 
 #include <sys/types.h>
 #include <sys/ioctl.h>
 #include <errno.h>
 
-/* Prototypes */
-static int nfp_net_close(struct rte_eth_dev *dev);
-static int nfp_net_init(struct rte_eth_dev *eth_dev);
-static int nfp_pf_init(struct rte_pci_device *pci_dev);
-static int nfp_pf_secondary_init(struct rte_pci_device *pci_dev);
-static int nfp_net_pf_read_mac(struct nfp_pf_dev *pf_dev, int port);
-static int nfp_pci_uninit(struct rte_eth_dev *eth_dev);
-static int nfp_init_phyports(struct nfp_pf_dev *pf_dev);
-static int nfp_net_stop(struct rte_eth_dev *dev);
-static int nfp_fw_setup(struct rte_pci_device *dev,
-                       struct nfp_cpp *cpp,
-                       struct nfp_eth_table *nfp_eth_table,
-                       struct nfp_hwinfo *hwinfo);
-
 static int
 __nfp_net_reconfig(struct nfp_net_hw *hw, uint32_t update)
 {
@@ -268,24 +252,6 @@ nfp_eth_copy_mac(uint8_t *dst, const uint8_t *src)
                dst[i] = src[i];
 }
 
-static int
-nfp_net_pf_read_mac(struct nfp_pf_dev *pf_dev, int port)
-{
-       struct nfp_eth_table *nfp_eth_table;
-       struct nfp_net_hw *hw = NULL;
-
-       /* Grab a pointer to the correct physical port */
-       hw = pf_dev->ports[port];
-
-       nfp_eth_table = nfp_eth_read_ports(pf_dev->cpp);
-
-       nfp_eth_copy_mac((uint8_t *)&hw->mac_addr,
-                        (uint8_t *)&nfp_eth_table->ports[port].mac_addr);
-
-       free(nfp_eth_table);
-       return 0;
-}
-
 void
 nfp_net_write_mac(struct nfp_net_hw *hw, uint8_t *mac)
 {
@@ -438,282 +404,6 @@ nfp_check_offloads(struct rte_eth_dev *dev)
        return ctrl;
 }
 
-static int
-nfp_net_start(struct rte_eth_dev *dev)
-{
-       struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
-       struct rte_intr_handle *intr_handle = &pci_dev->intr_handle;
-       uint32_t new_ctrl, update = 0;
-       struct nfp_net_hw *hw;
-       struct nfp_pf_dev *pf_dev;
-       struct rte_eth_conf *dev_conf;
-       struct rte_eth_rxmode *rxmode;
-       uint32_t intr_vector;
-       int ret;
-
-       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-       pf_dev = NFP_NET_DEV_PRIVATE_TO_PF(dev->data->dev_private);
-
-       PMD_INIT_LOG(DEBUG, "Start");
-
-       /* Disabling queues just in case... */
-       nfp_net_disable_queues(dev);
-
-       /* Enabling the required queues in the device */
-       nfp_net_enable_queues(dev);
-
-       /* check and configure queue intr-vector mapping */
-       if (dev->data->dev_conf.intr_conf.rxq != 0) {
-               if (pf_dev->multiport) {
-                       PMD_INIT_LOG(ERR, "PMD rx interrupt is not supported "
-                                         "with NFP multiport PF");
-                               return -EINVAL;
-               }
-               if (intr_handle->type == RTE_INTR_HANDLE_UIO) {
-                       /*
-                        * Better not to share LSC with RX interrupts.
-                        * Unregistering LSC interrupt handler
-                        */
-                       rte_intr_callback_unregister(&pci_dev->intr_handle,
-                               nfp_net_dev_interrupt_handler, (void *)dev);
-
-                       if (dev->data->nb_rx_queues > 1) {
-                               PMD_INIT_LOG(ERR, "PMD rx interrupt only "
-                                            "supports 1 queue with UIO");
-                               return -EIO;
-                       }
-               }
-               intr_vector = dev->data->nb_rx_queues;
-               if (rte_intr_efd_enable(intr_handle, intr_vector))
-                       return -1;
-
-               nfp_configure_rx_interrupt(dev, intr_handle);
-               update = NFP_NET_CFG_UPDATE_MSIX;
-       }
-
-       rte_intr_enable(intr_handle);
-
-       new_ctrl = nfp_check_offloads(dev);
-
-       /* Writing configuration parameters in the device */
-       nfp_net_params_setup(hw);
-
-       dev_conf = &dev->data->dev_conf;
-       rxmode = &dev_conf->rxmode;
-
-       if (rxmode->mq_mode & ETH_MQ_RX_RSS) {
-               nfp_net_rss_config_default(dev);
-               update |= NFP_NET_CFG_UPDATE_RSS;
-               new_ctrl |= NFP_NET_CFG_CTRL_RSS;
-       }
-
-       /* Enable device */
-       new_ctrl |= NFP_NET_CFG_CTRL_ENABLE;
-
-       update |= NFP_NET_CFG_UPDATE_GEN | NFP_NET_CFG_UPDATE_RING;
-
-       if (hw->cap & NFP_NET_CFG_CTRL_RINGCFG)
-               new_ctrl |= NFP_NET_CFG_CTRL_RINGCFG;
-
-       nn_cfg_writel(hw, NFP_NET_CFG_CTRL, new_ctrl);
-       if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
-               return -EIO;
-
-       /*
-        * Allocating rte mbufs for configured rx queues.
-        * This requires queues being enabled before
-        */
-       if (nfp_net_rx_freelist_setup(dev) < 0) {
-               ret = -ENOMEM;
-               goto error;
-       }
-
-       if (hw->is_phyport) {
-               if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-                       /* Configure the physical port up */
-                       nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 1);
-               else
-                       nfp_eth_set_configured(dev->process_private,
-                                              hw->nfp_idx, 1);
-       }
-
-       hw->ctrl = new_ctrl;
-
-       return 0;
-
-error:
-       /*
-        * An error returned by this function should mean the app
-        * exiting and then the system releasing all the memory
-        * allocated even memory coming from hugepages.
-        *
-        * The device could be enabled at this point with some queues
-        * ready for getting packets. This is true if the call to
-        * nfp_net_rx_freelist_setup() succeeds for some queues but
-        * fails for subsequent queues.
-        *
-        * This should make the app exiting but better if we tell the
-        * device first.
-        */
-       nfp_net_disable_queues(dev);
-
-       return ret;
-}
-
-/* Stop device: disable rx and tx functions to allow for reconfiguring. */
-static int
-nfp_net_stop(struct rte_eth_dev *dev)
-{
-       int i;
-       struct nfp_net_hw *hw;
-
-       PMD_INIT_LOG(DEBUG, "Stop");
-
-       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-
-       nfp_net_disable_queues(dev);
-
-       /* Clear queues */
-       for (i = 0; i < dev->data->nb_tx_queues; i++) {
-               nfp_net_reset_tx_queue(
-                       (struct nfp_net_txq *)dev->data->tx_queues[i]);
-       }
-
-       for (i = 0; i < dev->data->nb_rx_queues; i++) {
-               nfp_net_reset_rx_queue(
-                       (struct nfp_net_rxq *)dev->data->rx_queues[i]);
-       }
-
-       if (hw->is_phyport) {
-               if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-                       /* Configure the physical port down */
-                       nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 0);
-               else
-                       nfp_eth_set_configured(dev->process_private,
-                                              hw->nfp_idx, 0);
-       }
-
-       return 0;
-}
-
-/* Set the link up. */
-static int
-nfp_net_set_link_up(struct rte_eth_dev *dev)
-{
-       struct nfp_net_hw *hw;
-
-       PMD_DRV_LOG(DEBUG, "Set link up");
-
-       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-
-       if (!hw->is_phyport)
-               return -ENOTSUP;
-
-       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-               /* Configure the physical port down */
-               return nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 1);
-       else
-               return nfp_eth_set_configured(dev->process_private,
-                                             hw->nfp_idx, 1);
-}
-
-/* Set the link down. */
-static int
-nfp_net_set_link_down(struct rte_eth_dev *dev)
-{
-       struct nfp_net_hw *hw;
-
-       PMD_DRV_LOG(DEBUG, "Set link down");
-
-       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-
-       if (!hw->is_phyport)
-               return -ENOTSUP;
-
-       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-               /* Configure the physical port down */
-               return nfp_eth_set_configured(hw->cpp, hw->nfp_idx, 0);
-       else
-               return nfp_eth_set_configured(dev->process_private,
-                                             hw->nfp_idx, 0);
-}
-
-/* Reset and stop device. The device can not be restarted. */
-static int
-nfp_net_close(struct rte_eth_dev *dev)
-{
-       struct nfp_net_hw *hw;
-       struct rte_pci_device *pci_dev;
-       int i;
-
-       if (rte_eal_process_type() != RTE_PROC_PRIMARY)
-               return 0;
-
-       PMD_INIT_LOG(DEBUG, "Close");
-
-       hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
-       pci_dev = RTE_ETH_DEV_TO_PCI(dev);
-
-       /*
-        * We assume that the DPDK application is stopping all the
-        * threads/queues before calling the device close function.
-        */
-
-       nfp_net_disable_queues(dev);
-
-       /* Clear queues */
-       for (i = 0; i < dev->data->nb_tx_queues; i++) {
-               nfp_net_reset_tx_queue(
-                       (struct nfp_net_txq *)dev->data->tx_queues[i]);
-       }
-
-       for (i = 0; i < dev->data->nb_rx_queues; i++) {
-               nfp_net_reset_rx_queue(
-                       (struct nfp_net_rxq *)dev->data->rx_queues[i]);
-       }
-
-       /* Only free PF resources after all physical ports have been closed */
-       if (pci_dev->id.device_id == PCI_DEVICE_ID_NFP4000_PF_NIC ||
-           pci_dev->id.device_id == PCI_DEVICE_ID_NFP6000_PF_NIC) {
-               struct nfp_pf_dev *pf_dev;
-               pf_dev = NFP_NET_DEV_PRIVATE_TO_PF(dev->data->dev_private);
-
-               /* Mark this port as unused and free device priv resources*/
-               nn_cfg_writeb(hw, NFP_NET_CFG_LSC, 0xff);
-               pf_dev->ports[hw->idx] = NULL;
-               rte_eth_dev_release_port(dev);
-
-               for (i = 0; i < pf_dev->total_phyports; i++) {
-                       /* Check to see if ports are still in use */
-                       if (pf_dev->ports[i])
-                               return 0;
-               }
-
-               /* Now it is safe to free all PF resources */
-               PMD_INIT_LOG(INFO, "Freeing PF resources");
-               nfp_cpp_area_free(pf_dev->ctrl_area);
-               nfp_cpp_area_free(pf_dev->hwqueues_area);
-               free(pf_dev->hwinfo);
-               free(pf_dev->sym_tbl);
-               nfp_cpp_free(pf_dev->cpp);
-               rte_free(pf_dev);
-       }
-
-       rte_intr_disable(&pci_dev->intr_handle);
-
-       /* unregister callback func from eal lib */
-       rte_intr_callback_unregister(&pci_dev->intr_handle,
-                                    nfp_net_dev_interrupt_handler,
-                                    (void *)dev);
-
-       /*
-        * The ixgbe PMD driver disables the pcie master on the
-        * device. The i40e does not...
-        */
-
-       return 0;
-}
-
 int
 nfp_net_promisc_enable(struct rte_eth_dev *dev)
 {
@@ -1622,746 +1312,6 @@ nfp_net_rss_config_default(struct rte_eth_dev *dev)
        return ret;
 }
 
-
-/* Initialise and register driver with DPDK Application */
-static const struct eth_dev_ops nfp_net_eth_dev_ops = {
-       .dev_configure          = nfp_net_configure,
-       .dev_start              = nfp_net_start,
-       .dev_stop               = nfp_net_stop,
-       .dev_set_link_up        = nfp_net_set_link_up,
-       .dev_set_link_down      = nfp_net_set_link_down,
-       .dev_close              = nfp_net_close,
-       .promiscuous_enable     = nfp_net_promisc_enable,
-       .promiscuous_disable    = nfp_net_promisc_disable,
-       .link_update            = nfp_net_link_update,
-       .stats_get              = nfp_net_stats_get,
-       .stats_reset            = nfp_net_stats_reset,
-       .dev_infos_get          = nfp_net_infos_get,
-       .dev_supported_ptypes_get = nfp_net_supported_ptypes_get,
-       .mtu_set                = nfp_net_dev_mtu_set,
-       .mac_addr_set           = nfp_set_mac_addr,
-       .vlan_offload_set       = nfp_net_vlan_offload_set,
-       .reta_update            = nfp_net_reta_update,
-       .reta_query             = nfp_net_reta_query,
-       .rss_hash_update        = nfp_net_rss_hash_update,
-       .rss_hash_conf_get      = nfp_net_rss_hash_conf_get,
-       .rx_queue_setup         = nfp_net_rx_queue_setup,
-       .rx_queue_release       = nfp_net_rx_queue_release,
-       .tx_queue_setup         = nfp_net_tx_queue_setup,
-       .tx_queue_release       = nfp_net_tx_queue_release,
-       .rx_queue_intr_enable   = nfp_rx_queue_intr_enable,
-       .rx_queue_intr_disable  = nfp_rx_queue_intr_disable,
-};
-
-
-static int
-nfp_net_init(struct rte_eth_dev *eth_dev)
-{
-       struct rte_pci_device *pci_dev;
-       struct nfp_pf_dev *pf_dev;
-       struct nfp_net_hw *hw;
-
-       uint64_t tx_bar_off = 0, rx_bar_off = 0;
-       uint32_t start_q;
-       int stride = 4;
-       int port = 0;
-       int err;
-
-       PMD_INIT_FUNC_TRACE();
-
-       pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
-
-       /* Use backpointer here to the PF of this eth_dev */
-       pf_dev = NFP_NET_DEV_PRIVATE_TO_PF(eth_dev->data->dev_private);
-
-       /* NFP can not handle DMA addresses requiring more than 40 bits */
-       if (rte_mem_check_dma_mask(40)) {
-               RTE_LOG(ERR, PMD, "device %s can not be used:",
-                                  pci_dev->device.name);
-               RTE_LOG(ERR, PMD, "\trestricted dma mask to 40 bits!\n");
-               return -ENODEV;
-       };
-
-       if ((pci_dev->id.device_id == PCI_DEVICE_ID_NFP4000_PF_NIC) ||
-           (pci_dev->id.device_id == PCI_DEVICE_ID_NFP6000_PF_NIC)) {
-               port = ((struct nfp_net_hw *)eth_dev->data->dev_private)->idx;
-               if (port < 0 || port > 7) {
-                       PMD_DRV_LOG(ERR, "Port value is wrong");
-                       return -ENODEV;
-               }
-
-               /* Use PF array of physical ports to get pointer to
-                * this specific port
-                */
-               hw = pf_dev->ports[port];
-
-               PMD_INIT_LOG(DEBUG, "Working with physical port number: %d, "
-                                   "NFP internal port number: %d",
-                                   port, hw->nfp_idx);
-
-       } else {
-               hw = NFP_NET_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
-       }
-
-       eth_dev->dev_ops = &nfp_net_eth_dev_ops;
-       eth_dev->rx_queue_count = nfp_net_rx_queue_count;
-       eth_dev->rx_pkt_burst = &nfp_net_recv_pkts;
-       eth_dev->tx_pkt_burst = &nfp_net_xmit_pkts;
-
-       /* For secondary processes, the primary has done all the work */
-       if (rte_eal_process_type() != RTE_PROC_PRIMARY)
-               return 0;
-
-       rte_eth_copy_pci_info(eth_dev, pci_dev);
-
-       hw->device_id = pci_dev->id.device_id;
-       hw->vendor_id = pci_dev->id.vendor_id;
-       hw->subsystem_device_id = pci_dev->id.subsystem_device_id;
-       hw->subsystem_vendor_id = pci_dev->id.subsystem_vendor_id;
-
-       PMD_INIT_LOG(DEBUG, "nfp_net: device (%u:%u) %u:%u:%u:%u",
-                    pci_dev->id.vendor_id, pci_dev->id.device_id,
-                    pci_dev->addr.domain, pci_dev->addr.bus,
-                    pci_dev->addr.devid, pci_dev->addr.function);
-
-       hw->ctrl_bar = (uint8_t *)pci_dev->mem_resource[0].addr;
-       if (hw->ctrl_bar == NULL) {
-               PMD_DRV_LOG(ERR,
-                       "hw->ctrl_bar is NULL. BAR0 not configured");
-               return -ENODEV;
-       }
-
-       if (hw->is_phyport) {
-               if (port == 0) {
-                       hw->ctrl_bar = pf_dev->ctrl_bar;
-               } else {
-                       if (!pf_dev->ctrl_bar)
-                               return -ENODEV;
-                       /* Use port offset in pf ctrl_bar for this
-                        * ports control bar
-                        */
-                       hw->ctrl_bar = pf_dev->ctrl_bar +
-                                      (port * NFP_PF_CSR_SLICE_SIZE);
-               }
-       }
-
-       PMD_INIT_LOG(DEBUG, "ctrl bar: %p", hw->ctrl_bar);
-
-       hw->max_rx_queues = nn_cfg_readl(hw, NFP_NET_CFG_MAX_RXRINGS);
-       hw->max_tx_queues = nn_cfg_readl(hw, NFP_NET_CFG_MAX_TXRINGS);
-
-       /* Work out where in the BAR the queues start. */
-       switch (pci_dev->id.device_id) {
-       case PCI_DEVICE_ID_NFP4000_PF_NIC:
-       case PCI_DEVICE_ID_NFP6000_PF_NIC:
-       case PCI_DEVICE_ID_NFP6000_VF_NIC:
-               start_q = nn_cfg_readl(hw, NFP_NET_CFG_START_TXQ);
-               tx_bar_off = (uint64_t)start_q * NFP_QCP_QUEUE_ADDR_SZ;
-               start_q = nn_cfg_readl(hw, NFP_NET_CFG_START_RXQ);
-               rx_bar_off = (uint64_t)start_q * NFP_QCP_QUEUE_ADDR_SZ;
-               break;
-       default:
-               PMD_DRV_LOG(ERR, "nfp_net: no device ID matching");
-               err = -ENODEV;
-               goto dev_err_ctrl_map;
-       }
-
-       PMD_INIT_LOG(DEBUG, "tx_bar_off: 0x%" PRIx64 "", tx_bar_off);
-       PMD_INIT_LOG(DEBUG, "rx_bar_off: 0x%" PRIx64 "", rx_bar_off);
-
-       if (hw->is_phyport) {
-               hw->tx_bar = pf_dev->hw_queues + tx_bar_off;
-               hw->rx_bar = pf_dev->hw_queues + rx_bar_off;
-               eth_dev->data->dev_private = hw;
-       } else {
-               hw->tx_bar = (uint8_t *)pci_dev->mem_resource[2].addr +
-                            tx_bar_off;
-               hw->rx_bar = (uint8_t *)pci_dev->mem_resource[2].addr +
-                            rx_bar_off;
-       }
-
-       PMD_INIT_LOG(DEBUG, "ctrl_bar: %p, tx_bar: %p, rx_bar: %p",
-                    hw->ctrl_bar, hw->tx_bar, hw->rx_bar);
-
-       nfp_net_cfg_queue_setup(hw);
-
-       /* Get some of the read-only fields from the config BAR */
-       hw->ver = nn_cfg_readl(hw, NFP_NET_CFG_VERSION);
-       hw->cap = nn_cfg_readl(hw, NFP_NET_CFG_CAP);
-       hw->max_mtu = nn_cfg_readl(hw, NFP_NET_CFG_MAX_MTU);
-       hw->mtu = RTE_ETHER_MTU;
-
-       /* VLAN insertion is incompatible with LSOv2 */
-       if (hw->cap & NFP_NET_CFG_CTRL_LSO2)
-               hw->cap &= ~NFP_NET_CFG_CTRL_TXVLAN;
-
-       if (NFD_CFG_MAJOR_VERSION_of(hw->ver) < 2)
-               hw->rx_offset = NFP_NET_RX_OFFSET;
-       else
-               hw->rx_offset = nn_cfg_readl(hw, NFP_NET_CFG_RX_OFFSET_ADDR);
-
-       PMD_INIT_LOG(INFO, "VER: %u.%u, Maximum supported MTU: %d",
-                          NFD_CFG_MAJOR_VERSION_of(hw->ver),
-                          NFD_CFG_MINOR_VERSION_of(hw->ver), hw->max_mtu);
-
-       PMD_INIT_LOG(INFO, "CAP: %#x, %s%s%s%s%s%s%s%s%s%s%s%s%s%s", hw->cap,
-                    hw->cap & NFP_NET_CFG_CTRL_PROMISC ? "PROMISC " : "",
-                    hw->cap & NFP_NET_CFG_CTRL_L2BC    ? "L2BCFILT " : "",
-                    hw->cap & NFP_NET_CFG_CTRL_L2MC    ? "L2MCFILT " : "",
-                    hw->cap & NFP_NET_CFG_CTRL_RXCSUM  ? "RXCSUM "  : "",
-                    hw->cap & NFP_NET_CFG_CTRL_TXCSUM  ? "TXCSUM "  : "",
-                    hw->cap & NFP_NET_CFG_CTRL_RXVLAN  ? "RXVLAN "  : "",
-                    hw->cap & NFP_NET_CFG_CTRL_TXVLAN  ? "TXVLAN "  : "",
-                    hw->cap & NFP_NET_CFG_CTRL_SCATTER ? "SCATTER " : "",
-                    hw->cap & NFP_NET_CFG_CTRL_GATHER  ? "GATHER "  : "",
-                    hw->cap & NFP_NET_CFG_CTRL_LIVE_ADDR ? "LIVE_ADDR "  : "",
-                    hw->cap & NFP_NET_CFG_CTRL_LSO     ? "TSO "     : "",
-                    hw->cap & NFP_NET_CFG_CTRL_LSO2     ? "TSOv2 "     : "",
-                    hw->cap & NFP_NET_CFG_CTRL_RSS     ? "RSS "     : "",
-                    hw->cap & NFP_NET_CFG_CTRL_RSS2     ? "RSSv2 "     : "");
-
-       hw->ctrl = 0;
-
-       hw->stride_rx = stride;
-       hw->stride_tx = stride;
-
-       PMD_INIT_LOG(INFO, "max_rx_queues: %u, max_tx_queues: %u",
-                    hw->max_rx_queues, hw->max_tx_queues);
-
-       /* Initializing spinlock for reconfigs */
-       rte_spinlock_init(&hw->reconfig_lock);
-
-       /* Allocating memory for mac addr */
-       eth_dev->data->mac_addrs = rte_zmalloc("mac_addr",
-                                              RTE_ETHER_ADDR_LEN, 0);
-       if (eth_dev->data->mac_addrs == NULL) {
-               PMD_INIT_LOG(ERR, "Failed to space for MAC address");
-               err = -ENOMEM;
-               goto dev_err_queues_map;
-       }
-
-       if (hw->is_phyport) {
-               nfp_net_pf_read_mac(pf_dev, port);
-               nfp_net_write_mac(hw, (uint8_t *)&hw->mac_addr);
-       }
-
-       if (!rte_is_valid_assigned_ether_addr(
-                   (struct rte_ether_addr *)&hw->mac_addr)) {
-               PMD_INIT_LOG(INFO, "Using random mac address for port %d",
-                                  port);
-               /* Using random mac addresses for VFs */
-               rte_eth_random_addr(&hw->mac_addr[0]);
-               nfp_net_write_mac(hw, (uint8_t *)&hw->mac_addr);
-       }
-
-       /* Copying mac address to DPDK eth_dev struct */
-       rte_ether_addr_copy((struct rte_ether_addr *)hw->mac_addr,
-                       &eth_dev->data->mac_addrs[0]);
-
-       if (!(hw->cap & NFP_NET_CFG_CTRL_LIVE_ADDR))
-               eth_dev->data->dev_flags |= RTE_ETH_DEV_NOLIVE_MAC_ADDR;
-
-       eth_dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS;
-
-       PMD_INIT_LOG(INFO, "port %d VendorID=0x%x DeviceID=0x%x "
-                    "mac=%02x:%02x:%02x:%02x:%02x:%02x",
-                    eth_dev->data->port_id, pci_dev->id.vendor_id,
-                    pci_dev->id.device_id,
-                    hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
-                    hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
-
-       if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
-               /* Registering LSC interrupt handler */
-               rte_intr_callback_register(&pci_dev->intr_handle,
-                                          nfp_net_dev_interrupt_handler,
-                                          (void *)eth_dev);
-               /* Telling the firmware about the LSC interrupt entry */
-               nn_cfg_writeb(hw, NFP_NET_CFG_LSC, NFP_NET_IRQ_LSC_IDX);
-               /* Recording current stats counters values */
-               nfp_net_stats_reset(eth_dev);
-       }
-
-       return 0;
-
-dev_err_queues_map:
-               nfp_cpp_area_free(hw->hwqueues_area);
-dev_err_ctrl_map:
-               nfp_cpp_area_free(hw->ctrl_area);
-
-       return err;
-}
-
-#define DEFAULT_FW_PATH       "/lib/firmware/netronome"
-
-static int
-nfp_fw_upload(struct rte_pci_device *dev, struct nfp_nsp *nsp, char *card)
-{
-       struct nfp_cpp *cpp = nsp->cpp;
-       void *fw_buf;
-       char fw_name[125];
-       char serial[40];
-       size_t fsize;
-
-       /* Looking for firmware file in order of priority */
-
-       /* First try to find a firmware image specific for this device */
-       snprintf(serial, sizeof(serial),
-                       "serial-%02x-%02x-%02x-%02x-%02x-%02x-%02x-%02x",
-               cpp->serial[0], cpp->serial[1], cpp->serial[2], cpp->serial[3],
-               cpp->serial[4], cpp->serial[5], cpp->interface >> 8,
-               cpp->interface & 0xff);
-
-       snprintf(fw_name, sizeof(fw_name), "%s/%s.nffw", DEFAULT_FW_PATH,
-                       serial);
-       PMD_DRV_LOG(DEBUG, "Trying with fw file: %s", fw_name);
-       if (rte_firmware_read(fw_name, &fw_buf, &fsize) == 0)
-               goto load_fw;
-
-       /* Then try the PCI name */
-       snprintf(fw_name, sizeof(fw_name), "%s/pci-%s.nffw", DEFAULT_FW_PATH,
-                       dev->device.name);
-       PMD_DRV_LOG(DEBUG, "Trying with fw file: %s", fw_name);
-       if (rte_firmware_read(fw_name, &fw_buf, &fsize) == 0)
-               goto load_fw;
-
-       /* Finally try the card type and media */
-       snprintf(fw_name, sizeof(fw_name), "%s/%s", DEFAULT_FW_PATH, card);
-       PMD_DRV_LOG(DEBUG, "Trying with fw file: %s", fw_name);
-       if (rte_firmware_read(fw_name, &fw_buf, &fsize) < 0) {
-               PMD_DRV_LOG(INFO, "Firmware file %s not found.", fw_name);
-               return -ENOENT;
-       }
-
-load_fw:
-       PMD_DRV_LOG(INFO, "Firmware file found at %s with size: %zu",
-               fw_name, fsize);
-
-       PMD_DRV_LOG(INFO, "Uploading the firmware ...");
-       nfp_nsp_load_fw(nsp, fw_buf, fsize);
-       PMD_DRV_LOG(INFO, "Done");
-
-       free(fw_buf);
-       return 0;
-}
-
-static int
-nfp_fw_setup(struct rte_pci_device *dev, struct nfp_cpp *cpp,
-            struct nfp_eth_table *nfp_eth_table, struct nfp_hwinfo *hwinfo)
-{
-       struct nfp_nsp *nsp;
-       const char *nfp_fw_model;
-       char card_desc[100];
-       int err = 0;
-
-       nfp_fw_model = nfp_hwinfo_lookup(hwinfo, "assembly.partno");
-
-       if (nfp_fw_model) {
-               PMD_DRV_LOG(INFO, "firmware model found: %s", nfp_fw_model);
-       } else {
-               PMD_DRV_LOG(ERR, "firmware model NOT found");
-               return -EIO;
-       }
-
-       if (nfp_eth_table->count == 0 || nfp_eth_table->count > 8) {
-               PMD_DRV_LOG(ERR, "NFP ethernet table reports wrong ports: %u",
-                      nfp_eth_table->count);
-               return -EIO;
-       }
-
-       PMD_DRV_LOG(INFO, "NFP ethernet port table reports %u ports",
-                          nfp_eth_table->count);
-
-       PMD_DRV_LOG(INFO, "Port speed: %u", nfp_eth_table->ports[0].speed);
-
-       snprintf(card_desc, sizeof(card_desc), "nic_%s_%dx%d.nffw",
-                       nfp_fw_model, nfp_eth_table->count,
-                       nfp_eth_table->ports[0].speed / 1000);
-
-       nsp = nfp_nsp_open(cpp);
-       if (!nsp) {
-               PMD_DRV_LOG(ERR, "NFP error when obtaining NSP handle");
-               return -EIO;
-       }
-
-       nfp_nsp_device_soft_reset(nsp);
-       err = nfp_fw_upload(dev, nsp, card_desc);
-
-       nfp_nsp_close(nsp);
-       return err;
-}
-
-static int nfp_init_phyports(struct nfp_pf_dev *pf_dev)
-{
-       struct nfp_net_hw *hw;
-       struct rte_eth_dev *eth_dev;
-       struct nfp_eth_table *nfp_eth_table = NULL;
-       int ret = 0;
-       int i;
-
-       nfp_eth_table = nfp_eth_read_ports(pf_dev->cpp);
-       if (!nfp_eth_table) {
-               PMD_INIT_LOG(ERR, "Error reading NFP ethernet table");
-               ret = -EIO;
-               goto error;
-       }
-
-       /* Loop through all physical ports on PF */
-       for (i = 0; i < pf_dev->total_phyports; i++) {
-               const unsigned int numa_node = rte_socket_id();
-               char port_name[RTE_ETH_NAME_MAX_LEN];
-
-               snprintf(port_name, sizeof(port_name), "%s_port%d",
-                        pf_dev->pci_dev->device.name, i);
-
-               /* Allocate a eth_dev for this phyport */
-               eth_dev = rte_eth_dev_allocate(port_name);
-               if (!eth_dev) {
-                       ret = -ENODEV;
-                       goto port_cleanup;
-               }
-
-               /* Allocate memory for this phyport */
-               eth_dev->data->dev_private =
-                       rte_zmalloc_socket(port_name, sizeof(struct nfp_net_hw),
-                                          RTE_CACHE_LINE_SIZE, numa_node);
-               if (!eth_dev->data->dev_private) {
-                       ret = -ENOMEM;
-                       rte_eth_dev_release_port(eth_dev);
-                       goto port_cleanup;
-               }
-
-               hw = NFP_NET_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
-
-               /* Add this device to the PF's array of physical ports */
-               pf_dev->ports[i] = hw;
-
-               hw->pf_dev = pf_dev;
-               hw->cpp = pf_dev->cpp;
-               hw->eth_dev = eth_dev;
-               hw->idx = i;
-               hw->nfp_idx = nfp_eth_table->ports[i].index;
-               hw->is_phyport = true;
-
-               eth_dev->device = &pf_dev->pci_dev->device;
-
-               /* ctrl/tx/rx BAR mappings and remaining init happens in
-                * nfp_net_init
-                */
-               ret = nfp_net_init(eth_dev);
-
-               if (ret) {
-                       ret = -ENODEV;
-                       goto port_cleanup;
-               }
-
-               rte_eth_dev_probing_finish(eth_dev);
-
-       } /* End loop, all ports on this PF */
-       ret = 0;
-       goto eth_table_cleanup;
-
-port_cleanup:
-       for (i = 0; i < pf_dev->total_phyports; i++) {
-               if (pf_dev->ports[i] && pf_dev->ports[i]->eth_dev) {
-                       struct rte_eth_dev *tmp_dev;
-                       tmp_dev = pf_dev->ports[i]->eth_dev;
-                       rte_eth_dev_release_port(tmp_dev);
-                       pf_dev->ports[i] = NULL;
-               }
-       }
-eth_table_cleanup:
-       free(nfp_eth_table);
-error:
-       return ret;
-}
-
-static int nfp_pf_init(struct rte_pci_device *pci_dev)
-{
-       struct nfp_pf_dev *pf_dev = NULL;
-       struct nfp_cpp *cpp;
-       struct nfp_hwinfo *hwinfo;
-       struct nfp_rtsym_table *sym_tbl;
-       struct nfp_eth_table *nfp_eth_table = NULL;
-       char name[RTE_ETH_NAME_MAX_LEN];
-       int total_ports;
-       int ret = -ENODEV;
-       int err;
-
-       if (!pci_dev)
-               return ret;
-
-       /*
-        * When device bound to UIO, the device could be used, by mistake,
-        * by two DPDK apps, and the UIO driver does not avoid it. This
-        * could lead to a serious problem when configuring the NFP CPP
-        * interface. Here we avoid this telling to the CPP init code to
-        * use a lock file if UIO is being used.
-        */
-       if (pci_dev->kdrv == RTE_PCI_KDRV_VFIO)
-               cpp = nfp_cpp_from_device_name(pci_dev, 0);
-       else
-               cpp = nfp_cpp_from_device_name(pci_dev, 1);
-
-       if (!cpp) {
-               PMD_INIT_LOG(ERR, "A CPP handle can not be obtained");
-               ret = -EIO;
-               goto error;
-       }
-
-       hwinfo = nfp_hwinfo_read(cpp);
-       if (!hwinfo) {
-               PMD_INIT_LOG(ERR, "Error reading hwinfo table");
-               ret = -EIO;
-               goto error;
-       }
-
-       nfp_eth_table = nfp_eth_read_ports(cpp);
-       if (!nfp_eth_table) {
-               PMD_INIT_LOG(ERR, "Error reading NFP ethernet table");
-               ret = -EIO;
-               goto hwinfo_cleanup;
-       }
-
-       if (nfp_fw_setup(pci_dev, cpp, nfp_eth_table, hwinfo)) {
-               PMD_INIT_LOG(ERR, "Error when uploading firmware");
-               ret = -EIO;
-               goto eth_table_cleanup;
-       }
-
-       /* Now the symbol table should be there */
-       sym_tbl = nfp_rtsym_table_read(cpp);
-       if (!sym_tbl) {
-               PMD_INIT_LOG(ERR, "Something is wrong with the firmware"
-                               " symbol table");
-               ret = -EIO;
-               goto eth_table_cleanup;
-       }
-
-       total_ports = nfp_rtsym_read_le(sym_tbl, "nfd_cfg_pf0_num_ports", &err);
-       if (total_ports != (int)nfp_eth_table->count) {
-               PMD_DRV_LOG(ERR, "Inconsistent number of ports");
-               ret = -EIO;
-               goto sym_tbl_cleanup;
-       }
-
-       PMD_INIT_LOG(INFO, "Total physical ports: %d", total_ports);
-
-       if (total_ports <= 0 || total_ports > 8) {
-               PMD_INIT_LOG(ERR, "nfd_cfg_pf0_num_ports symbol with wrong value");
-               ret = -ENODEV;
-               goto sym_tbl_cleanup;
-       }
-       /* Allocate memory for the PF "device" */
-       snprintf(name, sizeof(name), "nfp_pf%d", 0);
-       pf_dev = rte_zmalloc(name, sizeof(*pf_dev), 0);
-       if (!pf_dev) {
-               ret = -ENOMEM;
-               goto sym_tbl_cleanup;
-       }
-
-       /* Populate the newly created PF device */
-       pf_dev->cpp = cpp;
-       pf_dev->hwinfo = hwinfo;
-       pf_dev->sym_tbl = sym_tbl;
-       pf_dev->total_phyports = total_ports;
-
-       if (total_ports > 1)
-               pf_dev->multiport = true;
-
-       pf_dev->pci_dev = pci_dev;
-
-       /* Map the symbol table */
-       pf_dev->ctrl_bar = nfp_rtsym_map(pf_dev->sym_tbl, "_pf0_net_bar0",
-                                    pf_dev->total_phyports * 32768,
-                                    &pf_dev->ctrl_area);
-       if (!pf_dev->ctrl_bar) {
-               PMD_INIT_LOG(ERR, "nfp_rtsym_map fails for _pf0_net_ctrl_bar");
-               ret = -EIO;
-               goto pf_cleanup;
-       }
-
-       PMD_INIT_LOG(DEBUG, "ctrl bar: %p", pf_dev->ctrl_bar);
-
-       /* configure access to tx/rx vNIC BARs */
-       pf_dev->hw_queues = nfp_cpp_map_area(pf_dev->cpp, 0, 0,
-                                             NFP_PCIE_QUEUE(0),
-                                             NFP_QCP_QUEUE_AREA_SZ,
-                                             &pf_dev->hwqueues_area);
-       if (!pf_dev->hw_queues) {
-               PMD_INIT_LOG(ERR, "nfp_rtsym_map fails for net.qc");
-               ret = -EIO;
-               goto ctrl_area_cleanup;
-       }
-
-       PMD_INIT_LOG(DEBUG, "tx/rx bar address: 0x%p", pf_dev->hw_queues);
-
-       /* Initialize and prep physical ports now
-        * This will loop through all physical ports
-        */
-       ret = nfp_init_phyports(pf_dev);
-       if (ret) {
-               PMD_INIT_LOG(ERR, "Could not create physical ports");
-               goto hwqueues_cleanup;
-       }
-
-       /* register the CPP bridge service here for primary use */
-       nfp_register_cpp_service(pf_dev->cpp);
-
-       return 0;
-
-hwqueues_cleanup:
-       nfp_cpp_area_free(pf_dev->hwqueues_area);
-ctrl_area_cleanup:
-       nfp_cpp_area_free(pf_dev->ctrl_area);
-pf_cleanup:
-       rte_free(pf_dev);
-sym_tbl_cleanup:
-       free(sym_tbl);
-eth_table_cleanup:
-       free(nfp_eth_table);
-hwinfo_cleanup:
-       free(hwinfo);
-error:
-       return ret;
-}
-
-/*
- * When attaching to the NFP4000/6000 PF on a secondary process there
- * is no need to initialize the PF again. Only minimal work is required
- * here
- */
-static int nfp_pf_secondary_init(struct rte_pci_device *pci_dev)
-{
-       struct nfp_cpp *cpp;
-       struct nfp_rtsym_table *sym_tbl;
-       int total_ports;
-       int i;
-       int err;
-
-       if (!pci_dev)
-               return -ENODEV;
-
-       /*
-        * When device bound to UIO, the device could be used, by mistake,
-        * by two DPDK apps, and the UIO driver does not avoid it. This
-        * could lead to a serious problem when configuring the NFP CPP
-        * interface. Here we avoid this telling to the CPP init code to
-        * use a lock file if UIO is being used.
-        */
-       if (pci_dev->kdrv == RTE_PCI_KDRV_VFIO)
-               cpp = nfp_cpp_from_device_name(pci_dev, 0);
-       else
-               cpp = nfp_cpp_from_device_name(pci_dev, 1);
-
-       if (!cpp) {
-               PMD_INIT_LOG(ERR, "A CPP handle can not be obtained");
-               return -EIO;
-       }
-
-       /*
-        * We don't have access to the PF created in the primary process
-        * here so we have to read the number of ports from firmware
-        */
-       sym_tbl = nfp_rtsym_table_read(cpp);
-       if (!sym_tbl) {
-               PMD_INIT_LOG(ERR, "Something is wrong with the firmware"
-                               " symbol table");
-               return -EIO;
-       }
-
-       total_ports = nfp_rtsym_read_le(sym_tbl, "nfd_cfg_pf0_num_ports", &err);
-
-       for (i = 0; i < total_ports; i++) {
-               struct rte_eth_dev *eth_dev;
-               char port_name[RTE_ETH_NAME_MAX_LEN];
-
-               snprintf(port_name, sizeof(port_name), "%s_port%d",
-                        pci_dev->device.name, i);
-
-               PMD_DRV_LOG(DEBUG, "Secondary attaching to port %s",
-                   port_name);
-               eth_dev = rte_eth_dev_attach_secondary(port_name);
-               if (!eth_dev) {
-                       RTE_LOG(ERR, EAL,
-                       "secondary process attach failed, "
-                       "ethdev doesn't exist");
-                       return -ENODEV;
-               }
-               eth_dev->process_private = cpp;
-               eth_dev->dev_ops = &nfp_net_eth_dev_ops;
-               eth_dev->rx_queue_count = nfp_net_rx_queue_count;
-               eth_dev->rx_pkt_burst = &nfp_net_recv_pkts;
-               eth_dev->tx_pkt_burst = &nfp_net_xmit_pkts;
-               rte_eth_dev_probing_finish(eth_dev);
-       }
-
-       /* Register the CPP bridge service for the secondary too */
-       nfp_register_cpp_service(cpp);
-
-       return 0;
-}
-
-static int nfp_pf_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
-                           struct rte_pci_device *dev)
-{
-       if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-               return nfp_pf_init(dev);
-       else
-               return nfp_pf_secondary_init(dev);
-}
-
-static const struct rte_pci_id pci_id_nfp_pf_net_map[] = {
-       {
-               RTE_PCI_DEVICE(PCI_VENDOR_ID_NETRONOME,
-                              PCI_DEVICE_ID_NFP4000_PF_NIC)
-       },
-       {
-               RTE_PCI_DEVICE(PCI_VENDOR_ID_NETRONOME,
-                              PCI_DEVICE_ID_NFP6000_PF_NIC)
-       },
-       {
-               .vendor_id = 0,
-       },
-};
-
-static int nfp_pci_uninit(struct rte_eth_dev *eth_dev)
-{
-       struct rte_pci_device *pci_dev;
-       uint16_t port_id;
-
-       pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
-
-       if (pci_dev->id.device_id == PCI_DEVICE_ID_NFP4000_PF_NIC ||
-           pci_dev->id.device_id == PCI_DEVICE_ID_NFP6000_PF_NIC) {
-               /* Free up all physical ports under PF */
-               RTE_ETH_FOREACH_DEV_OF(port_id, &pci_dev->device)
-                       rte_eth_dev_close(port_id);
-               /*
-                * Ports can be closed and freed but hotplugging is not
-                * currently supported
-                */
-               return -ENOTSUP;
-       }
-
-       /* VF cleanup, just free private port data */
-       return nfp_net_close(eth_dev);
-}
-
-static int eth_nfp_pci_remove(struct rte_pci_device *pci_dev)
-{
-       return rte_eth_dev_pci_generic_remove(pci_dev, nfp_pci_uninit);
-}
-
-static struct rte_pci_driver rte_nfp_net_pf_pmd = {
-       .id_table = pci_id_nfp_pf_net_map,
-       .drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC,
-       .probe = nfp_pf_pci_probe,
-       .remove = eth_nfp_pci_remove,
-};
-
-RTE_PMD_REGISTER_PCI(net_nfp_pf, rte_nfp_net_pf_pmd);
-RTE_PMD_REGISTER_PCI_TABLE(net_nfp_pf, pci_id_nfp_pf_net_map);
-RTE_PMD_REGISTER_KMOD_DEP(net_nfp_pf, "* igb_uio | uio_pci_generic | vfio");
 RTE_LOG_REGISTER_SUFFIX(nfp_logtype_init, init, NOTICE);
 RTE_LOG_REGISTER_SUFFIX(nfp_logtype_driver, driver, NOTICE);
 /*