网卡都是pci设备，因此这里每个网卡驱动其实就是一个pci驱动。并且intel这里是把好几个万兆网卡(82599/82598/x540)的驱动做在一起的。V4L2 一样几个类型摄像头合并在一起
先说一下 驱动总线平台；实际上就是platform_device(设备)与platform_driver（驱动）在platform(虚拟总线)上的注册、匹配，相互绑定。 但是最核心的还是cdev file_opterations 等
- ：struct bus_type-->它包含的最关键的函数：match() (要注意的是，这块由内核完成，我们不参与)
- ：struct platform_device-->注册：platform_device_register(unregister)
- ：struct platform_driver-->注册：platform_driver_register(unregister)
设备(或驱动)注册的时候，都会引发总线调用自己的match函数来寻找目前platform总线是否挂载有与该设备(或驱动)名字匹配的驱动（或设备），如果存在则将双方绑定；
platform_device:提供设备的硬件资源 比如 中断号 寄存器地址 DMA
platform_driver ： 提供 match id 绑定驱动和device资源的回调probe
详细 就不说了 很简单。。。。 之前在csdn上写过。。后续慢慢找回吧！！！！！！
假设：device先注册， 后续注册driver时，根据platform_match 匹配到了设备；
那么就会调用probe 实现资源的绑定！！！ 一般都是 创建设备 注册中断 初始化 file_opteriton 等回调函数
看下 ixgbe_probe 就知道：
static int ixgbe_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct net_device *netdev; struct ixgbe_adapter *adapter = NULL; struct ixgbe_hw *hw; /* static const struct ixgbe_info *ixgbe_info_tbl[] = { [board_82598] = &ixgbe_82598_info, [board_82599] = &ixgbe_82599_info, [board_X540] = &ixgbe_X540_info, [board_X550] = &ixgbe_X550_info, [board_X550EM_x] = &ixgbe_X550EM_x_info, [board_x550em_a] = &ixgbe_x550em_a_info, };有很多网卡，根据mactch的记过选择对用的设备信息 */ const struct ixgbe_info *ii = ixgbe_info_tbl[ent->driver_data];
根据硬件的参数来取得对应的devices info
通过对应的netdev的结构取得adapter，然后所有的核心操作都是保存在adapter中：
netdev = alloc_etherdev_mq(sizeof(struct ixgbe_adapter), indices); // 分配net_device和ixgbe_adapter，发送队列数为MAX_TX_QUEUE if (!netdev) { err = -ENOMEM; goto err_alloc_etherdev; } SET_NETDEV_DEV(netdev, &pdev->dev); adapter = netdev_priv(netdev);// 得到ixgbe_adapter的指针 adapter->netdev = netdev; adapter->pdev = pdev; hw = &adapter->hw;// 得到ixgbe_hw的指针 hw->back = adapter; adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); // 将BAR0中的总线地址映射成内存地址，赋给hw->hw_addr，允许网卡驱动通过hw->hw_addr访问网卡的BAR0对应的Memory空间 hw->hw_addr = ioremap(pci_resource_start(pdev, 0), pci_resource_len(pdev, 0)); adapter->io_addr = hw->hw_addr;
netdev->netdev_ops = &ixgbe_netdev_ops;// 注册ixgbe_netdev_ops ixgbe_set_ethtool_ops(netdev); netdev->watchdog_timeo = 5 * HZ; strlcpy(netdev->name, pci_name(pdev), sizeof(netdev->name)); /* Setup hw api */ hw->mac.ops = *ii->mac_ops; hw->mac.type = ii->mac; hw->mvals = ii->mvals; /* EEPROM */ hw->eeprom.ops = *ii->eeprom_ops; eec = IXGBE_READ_REG(hw, IXGBE_EEC(hw)); if (ixgbe_removed(hw->hw_addr)) { err = -EIO; goto err_ioremap; } /* If EEPROM is valid (bit 8 = 1), use default otherwise use bit bang */ if (!(eec & BIT(8))) hw->eeprom.ops.read = &ixgbe_read_eeprom_bit_bang_generic; /* PHY 硬件资源的初始化。。。。。。。。*/ hw->phy.ops = *ii->phy_ops; hw->phy.sfp_type = ixgbe_sfp_type_unknown; /* ixgbe_identify_phy_generic will set prtad and mmds properly */ hw->phy.mdio.prtad = MDIO_PRTAD_NONE; hw->phy.mdio.mmds = 0; hw->phy.mdio.mode_support = MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22; hw->phy.mdio.dev = netdev; hw->phy.mdio.mdio_read = ixgbe_mdio_read; hw->phy.mdio.mdio_write = ixgbe_mdio_write; ii->get_invariants(hw);
设置网卡属性 ethtool 可以查看
#define IXGBE_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE | NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_IPXIP6 | NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_UDP_TUNNEL_CSUM) netdev->gso_partial_features = IXGBE_GSO_PARTIAL_FEATURES; netdev->features |= NETIF_F_GSO_PARTIAL | IXGBE_GSO_PARTIAL_FEATURES; if (hw->mac.type >= ixgbe_mac_82599EB) netdev->features |= NETIF_F_SCTP_CRC;ixgbe_msix_clean_rings /* copy netdev features into list of user selectable features */ netdev->hw_features |= netdev->features | NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_RXALL | NETIF_F_HW_L2FW_DOFFLOAD; if (hw->mac.type >= ixgbe_mac_82599EB) netdev->hw_features |= NETIF_F_NTUPLE | NETIF_F_HW_TC; if (pci_using_dac) netdev->features |= NETIF_F_HIGHDMA; netdev->vlan_features |= netdev->features | NETIF_F_TSO_MANGLEID; netdev->hw_enc_features |= netdev->vlan_features;
还有注册中断
在enable 设备时 就会ixgbe_request_irq
err = request_irq(entry->vector, &ixgbe_msix_clean_rings, 0, q_vector->name, q_vector);
static irqreturn_t ixgbe_msix_clean_rings(int irq, void *data)
{
struct ixgbe_q_vector *q_vector = data;
/* EIAM disabled interrupts (on this vector) for us */
if (q_vector->rx.ring || q_vector->tx.ring)
napi_schedule_irqoff(&q_vector->napi);
return IRQ_HANDLED;
}
也就是 注册了napi 回调函数。。。 最后执行其对应的poll
/* initialize NAPI */ netif_napi_add(adapter->netdev, &q_vector->napi, ixgbe_poll, 64);
void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
int (*poll)(struct napi_struct *, int), int weight)
{
INIT_LIST_HEAD(&napi->poll_list);
hrtimer_init(&napi->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED);
napi->timer.function = napi_watchdog;
napi->gro_count = 0;
napi->gro_list = NULL;
napi->skb = NULL;
napi->poll = poll;
if (weight > NAPI_POLL_WEIGHT)
pr_err_once("netif_napi_add() called with weight %d on device %s ",
weight, dev->name);
napi->weight = weight;
list_add(&napi->dev_list, &dev->napi_list);
napi->dev = dev;
#ifdef CONFIG_NETPOLL
spin_lock_init(&napi->poll_lock);
napi->poll_owner = -1;
#endif
set_bit(NAPI_STATE_SCHED, &napi->state);
napi_hash_add(napi);
}
注册网络设备dev
strcpy(netdev->name, "eth%d"); err = register_netdev(netdev);----->if (dev->netdev_ops->ndo_init) {ret = dev->netdev_ops->ndo_init(dev)
netdev->netdev_ops = &ixgbe_netdev_ops;
poll：NAPI驱动最终是会调用网卡驱动挂载的poll回调，在ixgbe中，对应的回调就是ixgbe_poll
/** * ixgbe_poll - NAPI Rx polling callback * @napi: structure for representing this polling device * @budget: how many packets driver is allowed to clean * * This function is used for legacy and MSI, NAPI mode **/ int ixgbe_poll(struct napi_struct *napi, int budget) { struct ixgbe_q_vector *q_vector = container_of(napi, struct ixgbe_q_vector, napi); struct ixgbe_adapter *adapter = q_vector->adapter; struct ixgbe_ring *ring; int per_ring_budget, work_done = 0; bool clean_complete = true; #ifdef CONFIG_IXGBE_DCA if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) ixgbe_update_dca(q_vector); #endif ixgbe_for_each_ring(ring, q_vector->tx) { TX if (!ixgbe_clean_tx_irq(q_vector, ring, budget)) clean_complete = false; } /* Exit if we are called by netpoll or busy polling is active */ if ((budget <= 0) || !ixgbe_qv_lock_napi(q_vector)) return budget; /* attempt to distribute budget to each queue fairly, but don't allow * the budget to go below 1 because we'll exit polling */ if (q_vector->rx.count > 1) per_ring_budget = max(budget/q_vector->rx.count, 1); else per_ring_budget = budget; ixgbe_for_each_ring(ring, q_vector->rx) {RX int cleaned = ixgbe_clean_rx_irq(q_vector, ring, per_ring_budget); work_done += cleaned; if (cleaned >= per_ring_budget) clean_complete = false; } ixgbe_qv_unlock_napi(q_vector); /* If all work not completed, return budget and keep polling */ if (!clean_complete) return budget; /* all work done, exit the polling mode */ napi_complete_done(napi, work_done); if (adapter->rx_itr_setting & 1) ixgbe_set_itr(q_vector); if (!test_bit(__IXGBE_DOWN, &adapter->state)) ixgbe_irq_enable_queues(adapter, BIT_ULL(q_vector->v_idx)); return min(work_done, budget - 1); }
ndo_start_xmit---->ixgbe_xmit_frame，这个回调就是驱动提供给协议栈的发送回调接口。我们来看这个函数.
它的实现很简单，就是选取对应的队列，然后调用ixgbe_xmit_frame_ring来发送数据。
static netdev_tx_t __ixgbe_xmit_frame(struct sk_buff *skb, struct net_device *netdev, struct ixgbe_ring *ring) { struct ixgbe_adapter *adapter = netdev_priv(netdev); struct ixgbe_ring *tx_ring; /* * The minimum packet size for olinfo paylen is 17 so pad the skb * in order to meet this minimum size requirement. */ if (skb_put_padto(skb, 17)) return NETDEV_TX_OK; tx_ring = ring ? ring : adapter->tx_ring[skb->queue_mapping]; return ixgbe_xmit_frame_ring(skb, adapter, tx_ring); }
用ixgbe_tx_map来发送数据。而ixgbe_tx_map所做的最主要是两步，第一步请求DMA，第二步写寄存器，通知网卡发送数据.
netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct ixgbe_adapter *adapter, struct ixgbe_ring *tx_ring) { struct ixgbe_tx_buffer *first; int tso; u32 tx_flags = 0; unsigned short f; u16 count = TXD_USE_COUNT(skb_headlen(skb)); __be16 protocol = skb->protocol; u8 hdr_len = 0; /* * need: 1 descriptor per page * PAGE_SIZE/IXGBE_MAX_DATA_PER_TXD, * + 1 desc for skb_headlen/IXGBE_MAX_DATA_PER_TXD, * + 2 desc gap to keep tail from touching head, * + 1 desc for context descriptor, * otherwise try next time */ for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); if (ixgbe_maybe_stop_tx(tx_ring, count + 3)) { tx_ring->tx_stats.tx_busy++; return NETDEV_TX_BUSY; } /* record the location of the first descriptor for this packet */ first = &tx_ring->tx_buffer_info[tx_ring->next_to_use]; first->skb = skb; first->bytecount = skb->len; first->gso_segs = 1; ------------------------------------- ixgbe_tx_map(tx_ring, first, hdr_len); return NETDEV_TX_OK; out_drop: dev_kfree_skb_any(first->skb); first->skb = NULL; return NETDEV_TX_OK; }
上面的操作是异步的，此时内核还不能释放SKB，而是网卡硬件发送完数据之后，会再次产生中断通知内核，然后内核才能释放内存 ; 此时可以看到 napi的poll 回调函数 会清除tx 队列