网络源码分析
2021-03-12 本文已影响0人
grimlock44
ixbge网卡驱动
static int __init ixgbe_init_module(void)
{
int ret;
pr_info("%s\n", ixgbe_driver_string);
pr_info("%s\n", ixgbe_copyright);
ixgbe_wq = create_singlethread_workqueue(ixgbe_driver_name);
if (!ixgbe_wq) {
pr_err("%s: Failed to create workqueue\n", ixgbe_driver_name);
return -ENOMEM;
}
ixgbe_dbg_init();
ret = pci_register_driver(&ixgbe_driver);
if (ret) {
destroy_workqueue(ixgbe_wq);
ixgbe_dbg_exit();
return ret;
}
#ifdef CONFIG_IXGBE_DCA
dca_register_notify(&dca_notifier);
#endif
return 0;
}
ixgbe_wq 全局变量 workqueue
ixgbe_dbg_init 创建debugfs目录
pci_register_driver 注册驱动到pci总线
dca_register_notify Direct Cache Access注册 支持网卡DMA直接写进cpu cache，实现cpu处理之前的预读
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;
const struct ixgbe_info *ii = ixgbe_info_tbl[ent->driver_data];
int i, err, pci_using_dac, expected_gts;
unsigned int indices = MAX_TX_QUEUES;
u8 part_str[IXGBE_PBANUM_LENGTH];
bool disable_dev = false;
#ifdef IXGBE_FCOE
u16 device_caps;
#endif
u32 eec;
/* Catch broken hardware that put the wrong VF device ID in
* the PCIe SR-IOV capability.
*/
if (pdev->is_virtfn) {
WARN(1, KERN_ERR "%s (%hx:%hx) should not be a VF!\n",
pci_name(pdev), pdev->vendor, pdev->device);
return -EINVAL;
}
err = pci_enable_device_mem(pdev);
if (err)
return err;
if (!dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64))) {
pci_using_dac = 1;
} else {
err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
if (err) {
dev_err(&pdev->dev,
"No usable DMA configuration, aborting\n");
goto err_dma;
}
pci_using_dac = 0;
}
err = pci_request_mem_regions(pdev, ixgbe_driver_name);
if (err) {
dev_err(&pdev->dev,
"pci_request_selected_regions failed 0x%x\n", err);
goto err_pci_reg;
}
pci_enable_pcie_error_reporting(pdev);
pci_set_master(pdev);
pci_save_state(pdev);
if (ii->mac == ixgbe_mac_82598EB) {
#ifdef CONFIG_IXGBE_DCB
/* 8 TC w/ 4 queues per TC */
indices = 4 * MAX_TRAFFIC_CLASS;
#else
indices = IXGBE_MAX_RSS_INDICES;
#endif
}
netdev = alloc_etherdev_mq(sizeof(struct ixgbe_adapter), indices);
if (!netdev) {
err = -ENOMEM;
goto err_alloc_etherdev;
}
SET_NETDEV_DEV(netdev, &pdev->dev);
adapter = netdev_priv(netdev);
adapter->netdev = netdev;
adapter->pdev = pdev;
hw = &adapter->hw;
hw->back = adapter;
adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
hw->hw_addr = ioremap(pci_resource_start(pdev, 0),
pci_resource_len(pdev, 0));
adapter->io_addr = hw->hw_addr;
if (!hw->hw_addr) {
err = -EIO;
goto err_ioremap;
}
netdev->netdev_ops = &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;
if (ii->link_ops)
hw->link.ops = *ii->link_ops;
/* 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 */
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;
/* setup the private structure */
err = ixgbe_sw_init(adapter, ii);
if (err)
goto err_sw_init;
switch (adapter->hw.mac.type) {
case ixgbe_mac_X550:
case ixgbe_mac_X550EM_x:
netdev->udp_tunnel_nic_info = &ixgbe_udp_tunnels_x550;
break;
case ixgbe_mac_x550em_a:
netdev->udp_tunnel_nic_info = &ixgbe_udp_tunnels_x550em_a;
break;
default:
break;
}
/* Make sure the SWFW semaphore is in a valid state */
if (hw->mac.ops.init_swfw_sync)
hw->mac.ops.init_swfw_sync(hw);
/* Make it possible the adapter to be woken up via WOL */
switch (adapter->hw.mac.type) {
case ixgbe_mac_82599EB:
case ixgbe_mac_X540:
case ixgbe_mac_X550:
case ixgbe_mac_X550EM_x:
case ixgbe_mac_x550em_a:
IXGBE_WRITE_REG(&adapter->hw, IXGBE_WUS, ~0);
break;
default:
break;
}
/*
* If there is a fan on this device and it has failed log the
* failure.
*/
if (adapter->flags & IXGBE_FLAG_FAN_FAIL_CAPABLE) {
u32 esdp = IXGBE_READ_REG(hw, IXGBE_ESDP);
if (esdp & IXGBE_ESDP_SDP1)
e_crit(probe, "Fan has stopped, replace the adapter\n");
}
if (allow_unsupported_sfp)
hw->allow_unsupported_sfp = allow_unsupported_sfp;
/* reset_hw fills in the perm_addr as well */
hw->phy.reset_if_overtemp = true;
err = hw->mac.ops.reset_hw(hw);
hw->phy.reset_if_overtemp = false;
ixgbe_set_eee_capable(adapter);
if (err == IXGBE_ERR_SFP_NOT_PRESENT) {
err = 0;
} else if (err == IXGBE_ERR_SFP_NOT_SUPPORTED) {
e_dev_err("failed to load because an unsupported SFP+ or QSFP module type was detected.\n");
e_dev_err("Reload the driver after installing a supported module.\n");
goto err_sw_init;
} else if (err) {
e_dev_err("HW Init failed: %d\n", err);
goto err_sw_init;
}
#ifdef CONFIG_PCI_IOV
/* SR-IOV not supported on the 82598 */
if (adapter->hw.mac.type == ixgbe_mac_82598EB)
goto skip_sriov;
/* Mailbox */
ixgbe_init_mbx_params_pf(hw);
hw->mbx.ops = ii->mbx_ops;
pci_sriov_set_totalvfs(pdev, IXGBE_MAX_VFS_DRV_LIMIT);
ixgbe_enable_sriov(adapter, max_vfs);
skip_sriov:
#endif
netdev->features = NETIF_F_SG |
NETIF_F_TSO |
NETIF_F_TSO6 |
NETIF_F_RXHASH |
NETIF_F_RXCSUM |
NETIF_F_HW_CSUM;
#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 | NETIF_F_GSO_UDP_L4;
#ifdef CONFIG_IXGBE_IPSEC
#define IXGBE_ESP_FEATURES (NETIF_F_HW_ESP | \
NETIF_F_HW_ESP_TX_CSUM | \
NETIF_F_GSO_ESP)
if (adapter->ipsec)
netdev->features |= IXGBE_ESP_FEATURES;
#endif
/* copy netdev features into list of user selectable features */
netdev->hw_features |= netdev->features |
NETIF_F_HW_VLAN_CTAG_FILTER |
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;
netdev->mpls_features |= NETIF_F_SG |
NETIF_F_TSO |
NETIF_F_TSO6 |
NETIF_F_HW_CSUM;
netdev->mpls_features |= IXGBE_GSO_PARTIAL_FEATURES;
/* set this bit last since it cannot be part of vlan_features */
netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER |
NETIF_F_HW_VLAN_CTAG_RX |
NETIF_F_HW_VLAN_CTAG_TX;
netdev->priv_flags |= IFF_UNICAST_FLT;
netdev->priv_flags |= IFF_SUPP_NOFCS;
/* MTU range: 68 - 9710 */
netdev->min_mtu = ETH_MIN_MTU;
netdev->max_mtu = IXGBE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN);
#ifdef CONFIG_IXGBE_DCB
if (adapter->flags & IXGBE_FLAG_DCB_CAPABLE)
netdev->dcbnl_ops = &ixgbe_dcbnl_ops;
#endif
#ifdef IXGBE_FCOE
if (adapter->flags & IXGBE_FLAG_FCOE_CAPABLE) {
unsigned int fcoe_l;
if (hw->mac.ops.get_device_caps) {
hw->mac.ops.get_device_caps(hw, &device_caps);
if (device_caps & IXGBE_DEVICE_CAPS_FCOE_OFFLOADS)
adapter->flags &= ~IXGBE_FLAG_FCOE_CAPABLE;
}
fcoe_l = min_t(int, IXGBE_FCRETA_SIZE, num_online_cpus());
adapter->ring_feature[RING_F_FCOE].limit = fcoe_l;
netdev->features |= NETIF_F_FSO |
NETIF_F_FCOE_CRC;
netdev->vlan_features |= NETIF_F_FSO |
NETIF_F_FCOE_CRC |
NETIF_F_FCOE_MTU;
}
#endif /* IXGBE_FCOE */
if (adapter->flags2 & IXGBE_FLAG2_RSC_CAPABLE)
netdev->hw_features |= NETIF_F_LRO;
if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
netdev->features |= NETIF_F_LRO;
if (ixgbe_check_fw_error(adapter)) {
err = -EIO;
goto err_sw_init;
}
/* make sure the EEPROM is good */
if (hw->eeprom.ops.validate_checksum(hw, NULL) < 0) {
e_dev_err("The EEPROM Checksum Is Not Valid\n");
err = -EIO;
goto err_sw_init;
}
eth_platform_get_mac_address(&adapter->pdev->dev,
adapter->hw.mac.perm_addr);
memcpy(netdev->dev_addr, hw->mac.perm_addr, netdev->addr_len);
if (!is_valid_ether_addr(netdev->dev_addr)) {
e_dev_err("invalid MAC address\n");
err = -EIO;
goto err_sw_init;
}
/* Set hw->mac.addr to permanent MAC address */
ether_addr_copy(hw->mac.addr, hw->mac.perm_addr);
ixgbe_mac_set_default_filter(adapter);
timer_setup(&adapter->service_timer, ixgbe_service_timer, 0);
if (ixgbe_removed(hw->hw_addr)) {
err = -EIO;
goto err_sw_init;
}
INIT_WORK(&adapter->service_task, ixgbe_service_task);
set_bit(__IXGBE_SERVICE_INITED, &adapter->state);
clear_bit(__IXGBE_SERVICE_SCHED, &adapter->state);
err = ixgbe_init_interrupt_scheme(adapter);
if (err)
goto err_sw_init;
for (i = 0; i < adapter->num_rx_queues; i++)
u64_stats_init(&adapter->rx_ring[i]->syncp);
for (i = 0; i < adapter->num_tx_queues; i++)
u64_stats_init(&adapter->tx_ring[i]->syncp);
for (i = 0; i < adapter->num_xdp_queues; i++)
u64_stats_init(&adapter->xdp_ring[i]->syncp);
/* WOL not supported for all devices */
adapter->wol = 0;
hw->eeprom.ops.read(hw, 0x2c, &adapter->eeprom_cap);
hw->wol_enabled = ixgbe_wol_supported(adapter, pdev->device,
pdev->subsystem_device);
if (hw->wol_enabled)
adapter->wol = IXGBE_WUFC_MAG;
device_set_wakeup_enable(&adapter->pdev->dev, adapter->wol);
/* save off EEPROM version number */
ixgbe_set_fw_version(adapter);
/* pick up the PCI bus settings for reporting later */
if (ixgbe_pcie_from_parent(hw))
ixgbe_get_parent_bus_info(adapter);
else
hw->mac.ops.get_bus_info(hw);
/* calculate the expected PCIe bandwidth required for optimal
* performance. Note that some older parts will never have enough
* bandwidth due to being older generation PCIe parts. We clamp these
* parts to ensure no warning is displayed if it can't be fixed.
*/
switch (hw->mac.type) {
case ixgbe_mac_82598EB:
expected_gts = min(ixgbe_enumerate_functions(adapter) * 10, 16);
break;
default:
expected_gts = ixgbe_enumerate_functions(adapter) * 10;
break;
}
/* don't check link if we failed to enumerate functions */
if (expected_gts > 0)
ixgbe_check_minimum_link(adapter, expected_gts);
err = ixgbe_read_pba_string_generic(hw, part_str, sizeof(part_str));
if (err)
strlcpy(part_str, "Unknown", sizeof(part_str));
if (ixgbe_is_sfp(hw) && hw->phy.sfp_type != ixgbe_sfp_type_not_present)
e_dev_info("MAC: %d, PHY: %d, SFP+: %d, PBA No: %s\n",
hw->mac.type, hw->phy.type, hw->phy.sfp_type,
part_str);
else
e_dev_info("MAC: %d, PHY: %d, PBA No: %s\n",
hw->mac.type, hw->phy.type, part_str);
e_dev_info("%pM\n", netdev->dev_addr);
/* reset the hardware with the new settings */
err = hw->mac.ops.start_hw(hw);
if (err == IXGBE_ERR_EEPROM_VERSION) {
/* We are running on a pre-production device, log a warning */
e_dev_warn("This device is a pre-production adapter/LOM. "
"Please be aware there may be issues associated "
"with your hardware. If you are experiencing "
"problems please contact your Intel or hardware "
"representative who provided you with this "
"hardware.\n");
}
strcpy(netdev->name, "eth%d");
pci_set_drvdata(pdev, adapter);
err = register_netdev(netdev);
if (err)
goto err_register;
/* power down the optics for 82599 SFP+ fiber */
if (hw->mac.ops.disable_tx_laser)
hw->mac.ops.disable_tx_laser(hw);
/* carrier off reporting is important to ethtool even BEFORE open */
netif_carrier_off(netdev);
#ifdef CONFIG_IXGBE_DCA
if (dca_add_requester(&pdev->dev) == 0) {
adapter->flags |= IXGBE_FLAG_DCA_ENABLED;
ixgbe_setup_dca(adapter);
}
#endif
if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
e_info(probe, "IOV is enabled with %d VFs\n", adapter->num_vfs);
for (i = 0; i < adapter->num_vfs; i++)
ixgbe_vf_configuration(pdev, (i | 0x10000000));
}
/* firmware requires driver version to be 0xFFFFFFFF
* since os does not support feature
*/
if (hw->mac.ops.set_fw_drv_ver)
hw->mac.ops.set_fw_drv_ver(hw, 0xFF, 0xFF, 0xFF, 0xFF,
sizeof(UTS_RELEASE) - 1,
UTS_RELEASE);
/* add san mac addr to netdev */
ixgbe_add_sanmac_netdev(netdev);
e_dev_info("%s\n", ixgbe_default_device_descr);
#ifdef CONFIG_IXGBE_HWMON
if (ixgbe_sysfs_init(adapter))
e_err(probe, "failed to allocate sysfs resources\n");
#endif /* CONFIG_IXGBE_HWMON */
ixgbe_dbg_adapter_init(adapter);
/* setup link for SFP devices with MNG FW, else wait for IXGBE_UP */
if (ixgbe_mng_enabled(hw) && ixgbe_is_sfp(hw) && hw->mac.ops.setup_link)
hw->mac.ops.setup_link(hw,
IXGBE_LINK_SPEED_10GB_FULL | IXGBE_LINK_SPEED_1GB_FULL,
true);
err = ixgbe_mii_bus_init(hw);
if (err)
goto err_netdev;
return 0;
err_netdev:
unregister_netdev(netdev);
err_register:
ixgbe_release_hw_control(adapter);
ixgbe_clear_interrupt_scheme(adapter);
err_sw_init:
ixgbe_disable_sriov(adapter);
adapter->flags2 &= ~IXGBE_FLAG2_SEARCH_FOR_SFP;
iounmap(adapter->io_addr);
kfree(adapter->jump_tables[0]);
kfree(adapter->mac_table);
kfree(adapter->rss_key);
bitmap_free(adapter->af_xdp_zc_qps);
err_ioremap:
disable_dev = !test_and_set_bit(__IXGBE_DISABLED, &adapter->state);
free_netdev(netdev);
err_alloc_etherdev:
pci_release_mem_regions(pdev);
err_pci_reg:
err_dma:
if (!adapter || disable_dev)
pci_disable_device(pdev);
return err;
}
ixgbe_probe->ixgbe_init_interrupt_scheme->ixgbe_alloc_q_vectors->netif_napi_add(ixgbe_poll) 注册poll函数到q_vector->napi
ixgbe_open->ixgbe_request_irq->ixgbe_request_msix_irqs->request_irq(ixgbe_intr) 注册硬中断处理函数 取得q_vector
ixgbe_intr->napi_schedule_irqoff->list_add_tail(&napi->poll_list, &sd->poll_list); 将q_vectorr->napi挂到sd->poll_list 触发软中断
open_softirq(NET_RX_SOFTIRQ, net_rx_action); 系统初始化时注册软中断
net_rx_action->napi_poll 遍历sd->poll_list 调用poll函数 也就是ixgbe_poll
ixgbe_poll->ixgbe_clean_rx_irq->ixgbe_process_skb_fields 填写包2层头
ixgbe_rx_skb->napi_gro_receive->dev_gro_receive->napi_skb_finish->gro_normal_one->gro_normal_list gro处理
netif_receive_skb_list_internal->netif_receive_skb_list->__netif_receive_skb_list_core->__netif_receive_skb_core 协议栈处理
static int __netif_receive_skb_core(struct sk_buff **pskb, bool pfmemalloc,
struct packet_type **ppt_prev)
{
list_for_each_entry_rcu(ptype, &ptype_all, list) { //tcpdump回调处理
if (pt_prev)
ret = deliver_skb(skb, pt_prev, orig_dev);
pt_prev = ptype;
}
list_for_each_entry_rcu(ptype, &skb->dev->ptype_all, list) {
if (pt_prev)
ret = deliver_skb(skb, pt_prev, orig_dev);
pt_prev = ptype;
}
rx_handler = rcu_dereference(skb->dev->rx_handler); //ovs回调处理
type = skb->protocol;
/* deliver only exact match when indicated */
if (likely(!deliver_exact)) {
deliver_ptype_list_skb(skb, &pt_prev, orig_dev, type, //网络3层协议处理
&ptype_base[ntohs(type) &
PTYPE_HASH_MASK]);
}
deliver_ptype_list_skb(skb, &pt_prev, orig_dev, type,
&orig_dev->ptype_specific);
if (unlikely(skb->dev != orig_dev)) {
deliver_ptype_list_skb(skb, &pt_prev, orig_dev, type,
&skb->dev->ptype_specific);
}
}
