diff options
Diffstat (limited to 'drivers/net/ethernet')
252 files changed, 21280 insertions, 7076 deletions
diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c index c89b9aeeceb6..5698f5354c0b 100644 --- a/drivers/net/ethernet/8390/ax88796.c +++ b/drivers/net/ethernet/8390/ax88796.c @@ -84,7 +84,6 @@ static u32 ax_msg_enable; struct ax_device { struct mii_bus *mii_bus; struct mdiobb_ctrl bb_ctrl; - struct phy_device *phy_dev; void __iomem *addr_memr; u8 reg_memr; int link; @@ -320,7 +319,7 @@ static void ax_block_output(struct net_device *dev, int count, static void ax_handle_link_change(struct net_device *dev) { struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; + struct phy_device *phy_dev = dev->phydev; int status_change = 0; if (phy_dev->link && ((ax->speed != phy_dev->speed) || @@ -369,8 +368,6 @@ static int ax_mii_probe(struct net_device *dev) phy_dev->supported &= PHY_BASIC_FEATURES; phy_dev->advertising = phy_dev->supported; - ax->phy_dev = phy_dev; - netdev_info(dev, "PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", phy_dev->drv->name, phydev_name(phy_dev), phy_dev->irq); @@ -410,7 +407,7 @@ static int ax_open(struct net_device *dev) ret = ax_mii_probe(dev); if (ret) goto failed_mii_probe; - phy_start(ax->phy_dev); + phy_start(dev->phydev); ret = ax_ei_open(dev); if (ret) @@ -421,7 +418,7 @@ static int ax_open(struct net_device *dev) return 0; failed_ax_ei_open: - phy_disconnect(ax->phy_dev); + phy_disconnect(dev->phydev); failed_mii_probe: ax_phy_switch(dev, 0); free_irq(dev->irq, dev); @@ -442,7 +439,7 @@ static int ax_close(struct net_device *dev) /* turn the phy off */ ax_phy_switch(dev, 0); - phy_disconnect(ax->phy_dev); + phy_disconnect(dev->phydev); free_irq(dev->irq, dev); return 0; @@ -450,8 +447,7 @@ static int ax_close(struct net_device *dev) static int ax_ioctl(struct net_device *dev, struct ifreq *req, int cmd) { - struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; + struct phy_device *phy_dev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -474,28 +470,6 @@ static void ax_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, pdev->name, sizeof(info->bus_info)); } -static int ax_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; - - if (!phy_dev) - return -ENODEV; - - return phy_ethtool_gset(phy_dev, cmd); -} - -static int ax_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; - - if (!phy_dev) - return -ENODEV; - - return phy_ethtool_sset(phy_dev, cmd); -} - static u32 ax_get_msglevel(struct net_device *dev) { struct ei_device *ei_local = netdev_priv(dev); @@ -512,12 +486,12 @@ static void ax_set_msglevel(struct net_device *dev, u32 v) static const struct ethtool_ops ax_ethtool_ops = { .get_drvinfo = ax_get_drvinfo, - .get_settings = ax_get_settings, - .set_settings = ax_set_settings, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, .get_msglevel = ax_get_msglevel, .set_msglevel = ax_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; #ifdef CONFIG_AX88796_93CX6 diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index 30defe6c81f2..cd7e2e5f496b 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -440,7 +440,6 @@ struct et131x_adapter { struct net_device *netdev; struct pci_dev *pdev; struct mii_bus *mii_bus; - struct phy_device *phydev; struct napi_struct napi; /* Flags that indicate current state of the adapter */ @@ -864,7 +863,7 @@ static void et1310_config_mac_regs2(struct et131x_adapter *adapter) { int32_t delay = 0; struct mac_regs __iomem *mac = &adapter->regs->mac; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; u32 cfg1; u32 cfg2; u32 ifctrl; @@ -1035,7 +1034,7 @@ static void et1310_setup_device_for_unicast(struct et131x_adapter *adapter) static void et1310_config_rxmac_regs(struct et131x_adapter *adapter) { struct rxmac_regs __iomem *rxmac = &adapter->regs->rxmac; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; u32 sa_lo; u32 sa_hi = 0; u32 pf_ctrl = 0; @@ -1230,7 +1229,7 @@ out: static int et131x_mii_read(struct et131x_adapter *adapter, u8 reg, u16 *value) { - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; if (!phydev) return -EIO; @@ -1311,7 +1310,7 @@ static void et1310_phy_read_mii_bit(struct et131x_adapter *adapter, static void et1310_config_flow_control(struct et131x_adapter *adapter) { - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; if (phydev->duplex == DUPLEX_HALF) { adapter->flow = FLOW_NONE; @@ -1456,7 +1455,7 @@ static int et131x_mdio_write(struct mii_bus *bus, int phy_addr, static void et1310_phy_power_switch(struct et131x_adapter *adapter, bool down) { u16 data; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; et131x_mii_read(adapter, MII_BMCR, &data); data &= ~BMCR_PDOWN; @@ -1469,7 +1468,7 @@ static void et1310_phy_power_switch(struct et131x_adapter *adapter, bool down) static void et131x_xcvr_init(struct et131x_adapter *adapter) { u16 lcr2; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; /* Set the LED behavior such that LED 1 indicates speed (off = * 10Mbits, blink = 100Mbits, on = 1000Mbits) and LED 2 indicates @@ -2111,7 +2110,7 @@ static int et131x_init_recv(struct et131x_adapter *adapter) /* et131x_set_rx_dma_timer - Set the heartbeat timer according to line rate */ static void et131x_set_rx_dma_timer(struct et131x_adapter *adapter) { - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; /* For version B silicon, we do not use the RxDMA timer for 10 and 100 * Mbits/s line rates. We do not enable and RxDMA interrupt coalescing. @@ -2426,7 +2425,7 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb) struct sk_buff *skb = tcb->skb; u32 nr_frags = skb_shinfo(skb)->nr_frags + 1; struct skb_frag_struct *frags = &skb_shinfo(skb)->frags[0]; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; dma_addr_t dma_addr; struct tx_ring *tx_ring = &adapter->tx_ring; @@ -2791,22 +2790,6 @@ static void et131x_handle_send_pkts(struct et131x_adapter *adapter) spin_unlock_irqrestore(&adapter->tcb_send_qlock, flags); } -static int et131x_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - struct et131x_adapter *adapter = netdev_priv(netdev); - - return phy_ethtool_gset(adapter->phydev, cmd); -} - -static int et131x_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - struct et131x_adapter *adapter = netdev_priv(netdev); - - return phy_ethtool_sset(adapter->phydev, cmd); -} - static int et131x_get_regs_len(struct net_device *netdev) { #define ET131X_REGS_LEN 256 @@ -2979,12 +2962,12 @@ static void et131x_get_drvinfo(struct net_device *netdev, } static struct ethtool_ops et131x_ethtool_ops = { - .get_settings = et131x_get_settings, - .set_settings = et131x_set_settings, .get_drvinfo = et131x_get_drvinfo, .get_regs_len = et131x_get_regs_len, .get_regs = et131x_get_regs, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* et131x_hwaddr_init - set up the MAC Address */ @@ -3098,7 +3081,7 @@ err_out: static void et131x_error_timer_handler(unsigned long data) { struct et131x_adapter *adapter = (struct et131x_adapter *)data; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; if (et1310_in_phy_coma(adapter)) { /* Bring the device immediately out of coma, to @@ -3168,7 +3151,7 @@ static int et131x_adapter_memory_alloc(struct et131x_adapter *adapter) static void et131x_adjust_link(struct net_device *netdev) { struct et131x_adapter *adapter = netdev_priv(netdev); - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = netdev->phydev; if (!phydev) return; @@ -3287,7 +3270,6 @@ static int et131x_mii_probe(struct net_device *netdev) phydev->advertising = phydev->supported; phydev->autoneg = AUTONEG_ENABLE; - adapter->phydev = phydev; phy_attached_info(phydev); @@ -3323,7 +3305,7 @@ static void et131x_pci_remove(struct pci_dev *pdev) unregister_netdev(netdev); netif_napi_del(&adapter->napi); - phy_disconnect(adapter->phydev); + phy_disconnect(netdev->phydev); mdiobus_unregister(adapter->mii_bus); mdiobus_free(adapter->mii_bus); @@ -3338,20 +3320,16 @@ static void et131x_pci_remove(struct pci_dev *pdev) static void et131x_up(struct net_device *netdev) { - struct et131x_adapter *adapter = netdev_priv(netdev); - et131x_enable_txrx(netdev); - phy_start(adapter->phydev); + phy_start(netdev->phydev); } static void et131x_down(struct net_device *netdev) { - struct et131x_adapter *adapter = netdev_priv(netdev); - /* Save the timestamp for the TX watchdog, prevent a timeout */ netif_trans_update(netdev); - phy_stop(adapter->phydev); + phy_stop(netdev->phydev); et131x_disable_txrx(netdev); } @@ -3684,12 +3662,10 @@ static int et131x_close(struct net_device *netdev) static int et131x_ioctl(struct net_device *netdev, struct ifreq *reqbuf, int cmd) { - struct et131x_adapter *adapter = netdev_priv(netdev); - - if (!adapter->phydev) + if (!netdev->phydev) return -EINVAL; - return phy_mii_ioctl(adapter->phydev, reqbuf, cmd); + return phy_mii_ioctl(netdev->phydev, reqbuf, cmd); } /* et131x_set_packet_filter - Configures the Rx Packet filtering */ @@ -4073,7 +4049,7 @@ out: return rc; err_phy_disconnect: - phy_disconnect(adapter->phydev); + phy_disconnect(netdev->phydev); err_mdio_unregister: mdiobus_unregister(adapter->mii_bus); err_mdio_free: diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index de2c4bf5fac4..6ffdff68bfc4 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -77,7 +77,6 @@ struct emac_board_info { int emacrx_completed_flag; - struct phy_device *phy_dev; struct device_node *phy_node; unsigned int link; unsigned int speed; @@ -115,7 +114,7 @@ static void emac_update_duplex(struct net_device *dev) static void emac_handle_link_change(struct net_device *dev) { struct emac_board_info *db = netdev_priv(dev); - struct phy_device *phydev = db->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; int status_change = 0; @@ -154,21 +153,22 @@ static void emac_handle_link_change(struct net_device *dev) static int emac_mdio_probe(struct net_device *dev) { struct emac_board_info *db = netdev_priv(dev); + struct phy_device *phydev; /* to-do: PHY interrupts are currently not supported */ /* attach the mac to the phy */ - db->phy_dev = of_phy_connect(db->ndev, db->phy_node, - &emac_handle_link_change, 0, - db->phy_interface); - if (!db->phy_dev) { + phydev = of_phy_connect(db->ndev, db->phy_node, + &emac_handle_link_change, 0, + db->phy_interface); + if (!phydev) { netdev_err(db->ndev, "could not find the PHY\n"); return -ENODEV; } /* mask with MAC supported features */ - db->phy_dev->supported &= PHY_BASIC_FEATURES; - db->phy_dev->advertising = db->phy_dev->supported; + phydev->supported &= PHY_BASIC_FEATURES; + phydev->advertising = phydev->supported; db->link = 0; db->speed = 0; @@ -179,10 +179,7 @@ static int emac_mdio_probe(struct net_device *dev) static void emac_mdio_remove(struct net_device *dev) { - struct emac_board_info *db = netdev_priv(dev); - - phy_disconnect(db->phy_dev); - db->phy_dev = NULL; + phy_disconnect(dev->phydev); } static void emac_reset(struct emac_board_info *db) @@ -208,8 +205,7 @@ static void emac_inblk_32bit(void __iomem *reg, void *data, int count) static int emac_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct emac_board_info *dm = netdev_priv(dev); - struct phy_device *phydev = dm->phy_dev; + struct phy_device *phydev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -229,33 +225,11 @@ static void emac_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info)); } -static int emac_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct emac_board_info *dm = netdev_priv(dev); - struct phy_device *phydev = dm->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int emac_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct emac_board_info *dm = netdev_priv(dev); - struct phy_device *phydev = dm->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static const struct ethtool_ops emac_ethtool_ops = { .get_drvinfo = emac_get_drvinfo, - .get_settings = emac_get_settings, - .set_settings = emac_set_settings, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static unsigned int emac_setup(struct net_device *ndev) @@ -744,7 +718,7 @@ static int emac_open(struct net_device *dev) return ret; } - phy_start(db->phy_dev); + phy_start(dev->phydev); netif_start_queue(dev); return 0; @@ -781,7 +755,7 @@ static int emac_stop(struct net_device *ndev) netif_stop_queue(ndev); netif_carrier_off(ndev); - phy_stop(db->phy_dev); + phy_stop(ndev->phydev); emac_mdio_remove(ndev); diff --git a/drivers/net/ethernet/altera/altera_tse.h b/drivers/net/ethernet/altera/altera_tse.h index 103c30ddddf7..e0052003d16f 100644 --- a/drivers/net/ethernet/altera/altera_tse.h +++ b/drivers/net/ethernet/altera/altera_tse.h @@ -473,7 +473,6 @@ struct altera_tse_private { int phy_addr; /* PHY's MDIO address, -1 for autodetection */ phy_interface_t phy_iface; struct mii_bus *mdio; - struct phy_device *phydev; int oldspeed; int oldduplex; int oldlink; diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c index be72e1e64525..7c367713c3e6 100644 --- a/drivers/net/ethernet/altera/altera_tse_ethtool.c +++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c @@ -233,40 +233,18 @@ static void tse_get_regs(struct net_device *dev, struct ethtool_regs *regs, buf[i] = csrrd32(priv->mac_dev, i * 4); } -static int tse_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct altera_tse_private *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; - - if (phydev == NULL) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int tse_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct altera_tse_private *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; - - if (phydev == NULL) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static const struct ethtool_ops tse_ethtool_ops = { .get_drvinfo = tse_get_drvinfo, .get_regs_len = tse_reglen, .get_regs = tse_get_regs, .get_link = ethtool_op_get_link, - .get_settings = tse_get_settings, - .set_settings = tse_set_settings, .get_strings = tse_gstrings, .get_sset_count = tse_sset_count, .get_ethtool_stats = tse_fill_stats, .get_msglevel = tse_get_msglevel, .set_msglevel = tse_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; void altera_tse_set_ethtool_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index f749e4d389eb..49025e99fb0e 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -625,7 +625,7 @@ out: static void altera_tse_adjust_link(struct net_device *dev) { struct altera_tse_private *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; int new_state = 0; /* only change config if there is a link */ @@ -845,7 +845,6 @@ static int init_phy(struct net_device *dev) netdev_dbg(dev, "attached to PHY %d UID 0x%08x Link = %d\n", phydev->mdio.addr, phydev->phy_id, phydev->link); - priv->phydev = phydev; return 0; } @@ -1172,8 +1171,8 @@ static int tse_open(struct net_device *dev) spin_unlock_irqrestore(&priv->rxdma_irq_lock, flags); - if (priv->phydev) - phy_start(priv->phydev); + if (dev->phydev) + phy_start(dev->phydev); napi_enable(&priv->napi); netif_start_queue(dev); @@ -1205,8 +1204,8 @@ static int tse_shutdown(struct net_device *dev) unsigned long int flags; /* Stop the PHY */ - if (priv->phydev) - phy_stop(priv->phydev); + if (dev->phydev) + phy_stop(dev->phydev); netif_stop_queue(dev); napi_disable(&priv->napi); @@ -1545,10 +1544,9 @@ err_free_netdev: static int altera_tse_remove(struct platform_device *pdev) { struct net_device *ndev = platform_get_drvdata(pdev); - struct altera_tse_private *priv = netdev_priv(ndev); - if (priv->phydev) - phy_disconnect(priv->phydev); + if (ndev->phydev) + phy_disconnect(ndev->phydev); platform_set_drvdata(pdev, NULL); altera_tse_mdio_destroy(ndev); diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index e0fb0f1122db..20760e10211a 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -509,8 +509,8 @@ static int au1000_mii_probe(struct net_device *dev) * on the current MAC's MII bus */ for (phy_addr = 0; phy_addr < PHY_MAX_ADDR; phy_addr++) - if (mdiobus_get_phy(aup->mii_bus, aup->phy_addr)) { - phydev = mdiobus_get_phy(aup->mii_bus, aup->phy_addr); + if (mdiobus_get_phy(aup->mii_bus, phy_addr)) { + phydev = mdiobus_get_phy(aup->mii_bus, phy_addr); if (!aup->phy_search_highest_addr) /* break out with first one found */ break; diff --git a/drivers/net/ethernet/arc/emac.h b/drivers/net/ethernet/arc/emac.h index ca562bc034c3..e4feb712d4f2 100644 --- a/drivers/net/ethernet/arc/emac.h +++ b/drivers/net/ethernet/arc/emac.h @@ -134,7 +134,6 @@ struct arc_emac_priv { /* Devices */ struct device *dev; - struct phy_device *phy_dev; struct mii_bus *bus; struct arc_emac_mdio_bus_data bus_data; diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c index a3a9392a4954..586bedac457d 100644 --- a/drivers/net/ethernet/arc/emac_main.c +++ b/drivers/net/ethernet/arc/emac_main.c @@ -47,7 +47,7 @@ static inline int arc_emac_tx_avail(struct arc_emac_priv *priv) static void arc_emac_adjust_link(struct net_device *ndev) { struct arc_emac_priv *priv = netdev_priv(ndev); - struct phy_device *phy_dev = priv->phy_dev; + struct phy_device *phy_dev = ndev->phydev; unsigned int reg, state_changed = 0; if (priv->link != phy_dev->link) { @@ -80,46 +80,6 @@ static void arc_emac_adjust_link(struct net_device *ndev) } /** - * arc_emac_get_settings - Get PHY settings. - * @ndev: Pointer to net_device structure. - * @cmd: Pointer to ethtool_cmd structure. - * - * This implements ethtool command for getting PHY settings. If PHY could - * not be found, the function returns -ENODEV. This function calls the - * relevant PHY ethtool API to get the PHY settings. - * Issue "ethtool ethX" under linux prompt to execute this function. - */ -static int arc_emac_get_settings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - struct arc_emac_priv *priv = netdev_priv(ndev); - - return phy_ethtool_gset(priv->phy_dev, cmd); -} - -/** - * arc_emac_set_settings - Set PHY settings as passed in the argument. - * @ndev: Pointer to net_device structure. - * @cmd: Pointer to ethtool_cmd structure. - * - * This implements ethtool command for setting various PHY settings. If PHY - * could not be found, the function returns -ENODEV. This function calls the - * relevant PHY ethtool API to set the PHY. - * Issue e.g. "ethtool -s ethX speed 1000" under linux prompt to execute this - * function. - */ -static int arc_emac_set_settings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - struct arc_emac_priv *priv = netdev_priv(ndev); - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - return phy_ethtool_sset(priv->phy_dev, cmd); -} - -/** * arc_emac_get_drvinfo - Get EMAC driver information. * @ndev: Pointer to net_device structure. * @info: Pointer to ethtool_drvinfo structure. @@ -137,10 +97,10 @@ static void arc_emac_get_drvinfo(struct net_device *ndev, } static const struct ethtool_ops arc_emac_ethtool_ops = { - .get_settings = arc_emac_get_settings, - .set_settings = arc_emac_set_settings, .get_drvinfo = arc_emac_get_drvinfo, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; #define FIRST_OR_LAST_MASK (FIRST_MASK | LAST_MASK) @@ -403,7 +363,7 @@ static void arc_emac_poll_controller(struct net_device *dev) static int arc_emac_open(struct net_device *ndev) { struct arc_emac_priv *priv = netdev_priv(ndev); - struct phy_device *phy_dev = priv->phy_dev; + struct phy_device *phy_dev = ndev->phydev; int i; phy_dev->autoneg = AUTONEG_ENABLE; @@ -474,7 +434,7 @@ static int arc_emac_open(struct net_device *ndev) /* Enable EMAC */ arc_reg_or(priv, R_CTRL, EN_MASK); - phy_start_aneg(priv->phy_dev); + phy_start_aneg(ndev->phydev); netif_start_queue(ndev); @@ -772,6 +732,7 @@ int arc_emac_probe(struct net_device *ndev, int interface) struct device *dev = ndev->dev.parent; struct resource res_regs; struct device_node *phy_node; + struct phy_device *phydev = NULL; struct arc_emac_priv *priv; const char *mac_addr; unsigned int id, clock_frequency, irq; @@ -887,16 +848,16 @@ int arc_emac_probe(struct net_device *ndev, int interface) goto out_clken; } - priv->phy_dev = of_phy_connect(ndev, phy_node, arc_emac_adjust_link, 0, - interface); - if (!priv->phy_dev) { + phydev = of_phy_connect(ndev, phy_node, arc_emac_adjust_link, 0, + interface); + if (!phydev) { dev_err(dev, "of_phy_connect() failed\n"); err = -ENODEV; goto out_mdio; } dev_info(dev, "connected to %s phy with id 0x%x\n", - priv->phy_dev->drv->name, priv->phy_dev->phy_id); + phydev->drv->name, phydev->phy_id); netif_napi_add(ndev, &priv->napi, arc_emac_poll, ARC_EMAC_NAPI_WEIGHT); @@ -910,8 +871,7 @@ int arc_emac_probe(struct net_device *ndev, int interface) out_netif_api: netif_napi_del(&priv->napi); - phy_disconnect(priv->phy_dev); - priv->phy_dev = NULL; + phy_disconnect(phydev); out_mdio: arc_mdio_remove(priv); out_clken: @@ -925,8 +885,7 @@ int arc_emac_remove(struct net_device *ndev) { struct arc_emac_priv *priv = netdev_priv(ndev); - phy_disconnect(priv->phy_dev); - priv->phy_dev = NULL; + phy_disconnect(ndev->phydev); arc_mdio_remove(priv); unregister_netdev(ndev); netif_napi_del(&priv->napi); diff --git a/drivers/net/ethernet/atheros/alx/alx.h b/drivers/net/ethernet/atheros/alx/alx.h index d02c4240b7df..8fc93c5f6abc 100644 --- a/drivers/net/ethernet/atheros/alx/alx.h +++ b/drivers/net/ethernet/atheros/alx/alx.h @@ -96,10 +96,6 @@ struct alx_priv { unsigned int rx_ringsz; unsigned int rxbuf_size; - struct page *rx_page; - unsigned int rx_page_offset; - unsigned int rx_frag_size; - struct napi_struct napi; struct alx_tx_queue txq; struct alx_rx_queue rxq; diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index c98acdc0d14f..e708e360a9e3 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -70,35 +70,6 @@ static void alx_free_txbuf(struct alx_priv *alx, int entry) } } -static struct sk_buff *alx_alloc_skb(struct alx_priv *alx, gfp_t gfp) -{ - struct sk_buff *skb; - struct page *page; - - if (alx->rx_frag_size > PAGE_SIZE) - return __netdev_alloc_skb(alx->dev, alx->rxbuf_size, gfp); - - page = alx->rx_page; - if (!page) { - alx->rx_page = page = alloc_page(gfp); - if (unlikely(!page)) - return NULL; - alx->rx_page_offset = 0; - } - - skb = build_skb(page_address(page) + alx->rx_page_offset, - alx->rx_frag_size); - if (likely(skb)) { - alx->rx_page_offset += alx->rx_frag_size; - if (alx->rx_page_offset >= PAGE_SIZE) - alx->rx_page = NULL; - else - get_page(page); - } - return skb; -} - - static int alx_refill_rx_ring(struct alx_priv *alx, gfp_t gfp) { struct alx_rx_queue *rxq = &alx->rxq; @@ -115,9 +86,22 @@ static int alx_refill_rx_ring(struct alx_priv *alx, gfp_t gfp) while (!cur_buf->skb && next != rxq->read_idx) { struct alx_rfd *rfd = &rxq->rfd[cur]; - skb = alx_alloc_skb(alx, gfp); + /* + * When DMA RX address is set to something like + * 0x....fc0, it will be very likely to cause DMA + * RFD overflow issue. + * + * To work around it, we apply rx skb with 64 bytes + * longer space, and offset the address whenever + * 0x....fc0 is detected. + */ + skb = __netdev_alloc_skb(alx->dev, alx->rxbuf_size + 64, gfp); if (!skb) break; + + if (((unsigned long)skb->data & 0xfff) == 0xfc0) + skb_reserve(skb, 64); + dma = dma_map_single(&alx->hw.pdev->dev, skb->data, alx->rxbuf_size, DMA_FROM_DEVICE); @@ -153,7 +137,6 @@ static int alx_refill_rx_ring(struct alx_priv *alx, gfp_t gfp) alx_write_mem16(&alx->hw, ALX_RFD_PIDX, cur); } - return count; } @@ -622,11 +605,6 @@ static void alx_free_rings(struct alx_priv *alx) kfree(alx->txq.bufs); kfree(alx->rxq.bufs); - if (alx->rx_page) { - put_page(alx->rx_page); - alx->rx_page = NULL; - } - dma_free_coherent(&alx->hw.pdev->dev, alx->descmem.size, alx->descmem.virt, @@ -681,7 +659,6 @@ static int alx_request_irq(struct alx_priv *alx) alx->dev->name, alx); if (!err) goto out; - /* fall back to legacy interrupt */ pci_disable_msi(alx->hw.pdev); } @@ -725,7 +702,6 @@ static int alx_init_sw(struct alx_priv *alx) struct pci_dev *pdev = alx->hw.pdev; struct alx_hw *hw = &alx->hw; int err; - unsigned int head_size; err = alx_identify_hw(alx); if (err) { @@ -741,12 +717,7 @@ static int alx_init_sw(struct alx_priv *alx) hw->smb_timer = 400; hw->mtu = alx->dev->mtu; - alx->rxbuf_size = ALX_MAX_FRAME_LEN(hw->mtu); - head_size = SKB_DATA_ALIGN(alx->rxbuf_size + NET_SKB_PAD) + - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); - alx->rx_frag_size = roundup_pow_of_two(head_size); - alx->tx_ringsz = 256; alx->rx_ringsz = 512; hw->imt = 200; @@ -848,7 +819,6 @@ static int alx_change_mtu(struct net_device *netdev, int mtu) { struct alx_priv *alx = netdev_priv(netdev); int max_frame = ALX_MAX_FRAME_LEN(mtu); - unsigned int head_size; if ((max_frame < ALX_MIN_FRAME_SIZE) || (max_frame > ALX_MAX_FRAME_SIZE)) @@ -860,9 +830,6 @@ static int alx_change_mtu(struct net_device *netdev, int mtu) netdev->mtu = mtu; alx->hw.mtu = mtu; alx->rxbuf_size = max(max_frame, ALX_DEF_RXBUF_SIZE); - head_size = SKB_DATA_ALIGN(alx->rxbuf_size + NET_SKB_PAD) + - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); - alx->rx_frag_size = roundup_pow_of_two(head_size); netdev_update_features(netdev); if (netif_running(netdev)) alx_reinit(alx); diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c index 08a23e6b60e9..dc2c35dce216 100644 --- a/drivers/net/ethernet/aurora/nb8800.c +++ b/drivers/net/ethernet/aurora/nb8800.c @@ -631,7 +631,7 @@ static void nb8800_mac_config(struct net_device *dev) static void nb8800_pause_config(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; u32 rxcr; if (priv->pause_aneg) { @@ -664,7 +664,7 @@ static void nb8800_pause_config(struct net_device *dev) static void nb8800_link_reconfigure(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; int change = 0; if (phydev->link) { @@ -690,7 +690,7 @@ static void nb8800_link_reconfigure(struct net_device *dev) } if (change) - phy_print_status(priv->phydev); + phy_print_status(phydev); } static void nb8800_update_mac_addr(struct net_device *dev) @@ -935,9 +935,10 @@ static int nb8800_dma_stop(struct net_device *dev) static void nb8800_pause_adv(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; u32 adv = 0; - if (!priv->phydev) + if (!phydev) return; if (priv->pause_rx) @@ -945,13 +946,14 @@ static void nb8800_pause_adv(struct net_device *dev) if (priv->pause_tx) adv ^= ADVERTISED_Asym_Pause; - priv->phydev->supported |= adv; - priv->phydev->advertising |= adv; + phydev->supported |= adv; + phydev->advertising |= adv; } static int nb8800_open(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev; int err; /* clear any pending interrupts */ @@ -969,10 +971,10 @@ static int nb8800_open(struct net_device *dev) nb8800_mac_rx(dev, true); nb8800_mac_tx(dev, true); - priv->phydev = of_phy_connect(dev, priv->phy_node, - nb8800_link_reconfigure, 0, - priv->phy_mode); - if (!priv->phydev) + phydev = of_phy_connect(dev, priv->phy_node, + nb8800_link_reconfigure, 0, + priv->phy_mode); + if (!phydev) goto err_free_irq; nb8800_pause_adv(dev); @@ -982,7 +984,7 @@ static int nb8800_open(struct net_device *dev) netif_start_queue(dev); nb8800_start_rx(dev); - phy_start(priv->phydev); + phy_start(phydev); return 0; @@ -997,8 +999,9 @@ err_free_dma: static int nb8800_stop(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; - phy_stop(priv->phydev); + phy_stop(phydev); netif_stop_queue(dev); napi_disable(&priv->napi); @@ -1007,8 +1010,7 @@ static int nb8800_stop(struct net_device *dev) nb8800_mac_rx(dev, false); nb8800_mac_tx(dev, false); - phy_disconnect(priv->phydev); - priv->phydev = NULL; + phy_disconnect(phydev); free_irq(dev->irq, dev); @@ -1019,9 +1021,7 @@ static int nb8800_stop(struct net_device *dev) static int nb8800_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct nb8800_priv *priv = netdev_priv(dev); - - return phy_mii_ioctl(priv->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static const struct net_device_ops nb8800_netdev_ops = { @@ -1035,34 +1035,14 @@ static const struct net_device_ops nb8800_netdev_ops = { .ndo_validate_addr = eth_validate_addr, }; -static int nb8800_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct nb8800_priv *priv = netdev_priv(dev); - - if (!priv->phydev) - return -ENODEV; - - return phy_ethtool_gset(priv->phydev, cmd); -} - -static int nb8800_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct nb8800_priv *priv = netdev_priv(dev); - - if (!priv->phydev) - return -ENODEV; - - return phy_ethtool_sset(priv->phydev, cmd); -} - static int nb8800_nway_reset(struct net_device *dev) { - struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; - if (!priv->phydev) + if (!phydev) return -ENODEV; - return genphy_restart_aneg(priv->phydev); + return genphy_restart_aneg(phydev); } static void nb8800_get_pauseparam(struct net_device *dev, @@ -1079,6 +1059,7 @@ static int nb8800_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pp) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; priv->pause_aneg = pp->autoneg; priv->pause_rx = pp->rx_pause; @@ -1088,8 +1069,8 @@ static int nb8800_set_pauseparam(struct net_device *dev, if (!priv->pause_aneg) nb8800_pause_config(dev); - else if (priv->phydev) - phy_start_aneg(priv->phydev); + else if (phydev) + phy_start_aneg(phydev); return 0; } @@ -1182,8 +1163,6 @@ static void nb8800_get_ethtool_stats(struct net_device *dev, } static const struct ethtool_ops nb8800_ethtool_ops = { - .get_settings = nb8800_get_settings, - .set_settings = nb8800_set_settings, .nway_reset = nb8800_nway_reset, .get_link = ethtool_op_get_link, .get_pauseparam = nb8800_get_pauseparam, @@ -1191,6 +1170,8 @@ static const struct ethtool_ops nb8800_ethtool_ops = { .get_sset_count = nb8800_get_sset_count, .get_strings = nb8800_get_strings, .get_ethtool_stats = nb8800_get_ethtool_stats, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int nb8800_hw_init(struct net_device *dev) diff --git a/drivers/net/ethernet/aurora/nb8800.h b/drivers/net/ethernet/aurora/nb8800.h index e5adbc2aac9f..6ec4a956e1e5 100644 --- a/drivers/net/ethernet/aurora/nb8800.h +++ b/drivers/net/ethernet/aurora/nb8800.h @@ -284,7 +284,6 @@ struct nb8800_priv { struct mii_bus *mii_bus; struct device_node *phy_node; - struct phy_device *phydev; /* PHY connection type from DT */ int phy_mode; diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index 18042c2460bd..d74a92e1c27d 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -139,26 +139,6 @@ config BNX2X_SRIOV Virtualization support in the 578xx and 57712 products. This allows for virtual function acceleration in virtual environments. -config BNX2X_VXLAN - bool "Virtual eXtensible Local Area Network support" - default n - depends on BNX2X && VXLAN && !(BNX2X=y && VXLAN=m) - ---help--- - This enables hardward offload support for VXLAN protocol over the - NetXtremeII series adapters. - Say Y here if you want to enable hardware offload support for - Virtual eXtensible Local Area Network (VXLAN) in the driver. - -config BNX2X_GENEVE - bool "Generic Network Virtualization Encapsulation (GENEVE) support" - depends on BNX2X && GENEVE && !(BNX2X=y && GENEVE=m) - ---help--- - This allows one to create GENEVE virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. GENEVE is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to enable hardware offload support for - Generic Network Virtualization Encapsulation (GENEVE) in the driver. - config BGMAC tristate "BCMA bus GBit core support" depends on BCMA && BCMA_HOST_SOC @@ -186,7 +166,6 @@ config SYSTEMPORT config BNXT tristate "Broadcom NetXtreme-C/E support" depends on PCI - depends on VXLAN || VXLAN=n select FW_LOADER select LIBCRC32C ---help--- diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 543bf38105c9..834afbb51aff 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -96,28 +96,6 @@ static inline void tdma_port_write_desc_addr(struct bcm_sysport_priv *priv, } /* Ethtool operations */ -static int bcm_sysport_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct bcm_sysport_priv *priv = netdev_priv(dev); - - if (!netif_running(dev)) - return -EINVAL; - - return phy_ethtool_sset(priv->phydev, cmd); -} - -static int bcm_sysport_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct bcm_sysport_priv *priv = netdev_priv(dev); - - if (!netif_running(dev)) - return -EINVAL; - - return phy_ethtool_gset(priv->phydev, cmd); -} - static int bcm_sysport_set_rx_csum(struct net_device *dev, netdev_features_t wanted) { @@ -1127,7 +1105,7 @@ static void bcm_sysport_tx_timeout(struct net_device *dev) static void bcm_sysport_adj_link(struct net_device *dev) { struct bcm_sysport_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; unsigned int changed = 0; u32 cmd_bits = 0, reg; @@ -1182,7 +1160,7 @@ static void bcm_sysport_adj_link(struct net_device *dev) umac_writel(priv, reg, UMAC_CMD); } - phy_print_status(priv->phydev); + phy_print_status(phydev); } static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv, @@ -1525,7 +1503,7 @@ static void bcm_sysport_netif_start(struct net_device *dev) /* Enable RX interrupt and TX ring full interrupt */ intrl2_0_mask_clear(priv, INTRL2_0_RDMA_MBDONE | INTRL2_0_TX_RING_FULL); - phy_start(priv->phydev); + phy_start(dev->phydev); /* Enable TX interrupts for the 32 TXQs */ intrl2_1_mask_clear(priv, 0xffffffff); @@ -1546,6 +1524,7 @@ static void rbuf_init(struct bcm_sysport_priv *priv) static int bcm_sysport_open(struct net_device *dev) { struct bcm_sysport_priv *priv = netdev_priv(dev); + struct phy_device *phydev; unsigned int i; int ret; @@ -1570,9 +1549,9 @@ static int bcm_sysport_open(struct net_device *dev) /* Read CRC forward */ priv->crc_fwd = !!(umac_readl(priv, UMAC_CMD) & CMD_CRC_FWD); - priv->phydev = of_phy_connect(dev, priv->phy_dn, bcm_sysport_adj_link, - 0, priv->phy_interface); - if (!priv->phydev) { + phydev = of_phy_connect(dev, priv->phy_dn, bcm_sysport_adj_link, + 0, priv->phy_interface); + if (!phydev) { netdev_err(dev, "could not attach to PHY\n"); return -ENODEV; } @@ -1650,7 +1629,7 @@ out_free_tx_ring: out_free_irq0: free_irq(priv->irq0, dev); out_phy_disconnect: - phy_disconnect(priv->phydev); + phy_disconnect(phydev); return ret; } @@ -1661,7 +1640,7 @@ static void bcm_sysport_netif_stop(struct net_device *dev) /* stop all software from updating hardware */ netif_tx_stop_all_queues(dev); napi_disable(&priv->napi); - phy_stop(priv->phydev); + phy_stop(dev->phydev); /* mask all interrupts */ intrl2_0_mask_set(priv, 0xffffffff); @@ -1708,14 +1687,12 @@ static int bcm_sysport_stop(struct net_device *dev) free_irq(priv->irq1, dev); /* Disconnect from PHY */ - phy_disconnect(priv->phydev); + phy_disconnect(dev->phydev); return 0; } static struct ethtool_ops bcm_sysport_ethtool_ops = { - .get_settings = bcm_sysport_get_settings, - .set_settings = bcm_sysport_set_settings, .get_drvinfo = bcm_sysport_get_drvinfo, .get_msglevel = bcm_sysport_get_msglvl, .set_msglevel = bcm_sysport_set_msglvl, @@ -1727,6 +1704,8 @@ static struct ethtool_ops bcm_sysport_ethtool_ops = { .set_wol = bcm_sysport_set_wol, .get_coalesce = bcm_sysport_get_coalesce, .set_coalesce = bcm_sysport_set_coalesce, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops bcm_sysport_netdev_ops = { @@ -1929,7 +1908,7 @@ static int bcm_sysport_suspend(struct device *d) bcm_sysport_netif_stop(dev); - phy_suspend(priv->phydev); + phy_suspend(dev->phydev); netif_device_detach(dev); @@ -2055,7 +2034,7 @@ static int bcm_sysport_resume(struct device *d) goto out_free_rx_ring; } - phy_resume(priv->phydev); + phy_resume(dev->phydev); bcm_sysport_netif_start(dev); diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h index f28bf545d7f4..1c82e3da69a7 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.h +++ b/drivers/net/ethernet/broadcom/bcmsysport.h @@ -670,7 +670,6 @@ struct bcm_sysport_priv { /* PHY device */ struct device_node *phy_dn; - struct phy_device *phydev; phy_interface_t phy_interface; int old_pause; int old_link; diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index ee5f431ab32a..b045dc072c40 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -246,6 +246,8 @@ err_dma_head: err_drop: dev_kfree_skb(skb); + net_dev->stats.tx_dropped++; + net_dev->stats.tx_errors++; return NETDEV_TX_OK; } @@ -267,15 +269,16 @@ static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring) while (ring->start != ring->end) { int slot_idx = ring->start % BGMAC_TX_RING_SLOTS; struct bgmac_slot_info *slot = &ring->slots[slot_idx]; - u32 ctl1; + u32 ctl0, ctl1; int len; if (slot_idx == empty_slot) break; + ctl0 = le32_to_cpu(ring->cpu_base[slot_idx].ctl0); ctl1 = le32_to_cpu(ring->cpu_base[slot_idx].ctl1); len = ctl1 & BGMAC_DESC_CTL1_LEN; - if (ctl1 & BGMAC_DESC_CTL0_SOF) + if (ctl0 & BGMAC_DESC_CTL0_SOF) /* Unmap no longer used buffer */ dma_unmap_single(dma_dev, slot->dma_addr, len, DMA_TO_DEVICE); @@ -284,6 +287,8 @@ static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring) DMA_TO_DEVICE); if (slot->skb) { + bgmac->net_dev->stats.tx_bytes += slot->skb->len; + bgmac->net_dev->stats.tx_packets++; bytes_compl += slot->skb->len; pkts_compl++; @@ -464,6 +469,7 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, bgmac_err(bgmac, "Found poisoned packet at slot %d, DMA issue!\n", ring->start); put_page(virt_to_head_page(buf)); + bgmac->net_dev->stats.rx_errors++; break; } @@ -471,6 +477,8 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, bgmac_err(bgmac, "Found oversized packet at slot %d, DMA issue!\n", ring->start); put_page(virt_to_head_page(buf)); + bgmac->net_dev->stats.rx_length_errors++; + bgmac->net_dev->stats.rx_errors++; break; } @@ -481,6 +489,7 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, if (unlikely(!skb)) { bgmac_err(bgmac, "build_skb failed\n"); put_page(virt_to_head_page(buf)); + bgmac->net_dev->stats.rx_errors++; break; } skb_put(skb, BGMAC_RX_FRAME_OFFSET + @@ -490,6 +499,8 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, skb_checksum_none_assert(skb); skb->protocol = eth_type_trans(skb, bgmac->net_dev); + bgmac->net_dev->stats.rx_bytes += len; + bgmac->net_dev->stats.rx_packets++; napi_gro_receive(&bgmac->napi, skb); handled++; } while (0); @@ -1310,9 +1321,10 @@ static int bgmac_open(struct net_device *net_dev) } napi_enable(&bgmac->napi); - phy_start(bgmac->phy_dev); + phy_start(net_dev->phydev); + + netif_start_queue(net_dev); - netif_carrier_on(net_dev); return 0; } @@ -1322,7 +1334,7 @@ static int bgmac_stop(struct net_device *net_dev) netif_carrier_off(net_dev); - phy_stop(bgmac->phy_dev); + phy_stop(net_dev->phydev); napi_disable(&bgmac->napi); bgmac_chip_intrs_off(bgmac); @@ -1360,12 +1372,10 @@ static int bgmac_set_mac_address(struct net_device *net_dev, void *addr) static int bgmac_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd) { - struct bgmac *bgmac = netdev_priv(net_dev); - if (!netif_running(net_dev)) return -EINVAL; - return phy_mii_ioctl(bgmac->phy_dev, ifr, cmd); + return phy_mii_ioctl(net_dev->phydev, ifr, cmd); } static const struct net_device_ops bgmac_netdev_ops = { @@ -1382,20 +1392,125 @@ static const struct net_device_ops bgmac_netdev_ops = { * ethtool_ops **************************************************/ -static int bgmac_get_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) +struct bgmac_stat { + u8 size; + u32 offset; + const char *name; +}; + +static struct bgmac_stat bgmac_get_strings_stats[] = { + { 8, BGMAC_TX_GOOD_OCTETS, "tx_good_octets" }, + { 4, BGMAC_TX_GOOD_PKTS, "tx_good" }, + { 8, BGMAC_TX_OCTETS, "tx_octets" }, + { 4, BGMAC_TX_PKTS, "tx_pkts" }, + { 4, BGMAC_TX_BROADCAST_PKTS, "tx_broadcast" }, + { 4, BGMAC_TX_MULTICAST_PKTS, "tx_multicast" }, + { 4, BGMAC_TX_LEN_64, "tx_64" }, + { 4, BGMAC_TX_LEN_65_TO_127, "tx_65_127" }, + { 4, BGMAC_TX_LEN_128_TO_255, "tx_128_255" }, + { 4, BGMAC_TX_LEN_256_TO_511, "tx_256_511" }, + { 4, BGMAC_TX_LEN_512_TO_1023, "tx_512_1023" }, + { 4, BGMAC_TX_LEN_1024_TO_1522, "tx_1024_1522" }, + { 4, BGMAC_TX_LEN_1523_TO_2047, "tx_1523_2047" }, + { 4, BGMAC_TX_LEN_2048_TO_4095, "tx_2048_4095" }, + { 4, BGMAC_TX_LEN_4096_TO_8191, "tx_4096_8191" }, + { 4, BGMAC_TX_LEN_8192_TO_MAX, "tx_8192_max" }, + { 4, BGMAC_TX_JABBER_PKTS, "tx_jabber" }, + { 4, BGMAC_TX_OVERSIZE_PKTS, "tx_oversize" }, + { 4, BGMAC_TX_FRAGMENT_PKTS, "tx_fragment" }, + { 4, BGMAC_TX_UNDERRUNS, "tx_underruns" }, + { 4, BGMAC_TX_TOTAL_COLS, "tx_total_cols" }, + { 4, BGMAC_TX_SINGLE_COLS, "tx_single_cols" }, + { 4, BGMAC_TX_MULTIPLE_COLS, "tx_multiple_cols" }, + { 4, BGMAC_TX_EXCESSIVE_COLS, "tx_excessive_cols" }, + { 4, BGMAC_TX_LATE_COLS, "tx_late_cols" }, + { 4, BGMAC_TX_DEFERED, "tx_defered" }, + { 4, BGMAC_TX_CARRIER_LOST, "tx_carrier_lost" }, + { 4, BGMAC_TX_PAUSE_PKTS, "tx_pause" }, + { 4, BGMAC_TX_UNI_PKTS, "tx_unicast" }, + { 4, BGMAC_TX_Q0_PKTS, "tx_q0" }, + { 8, BGMAC_TX_Q0_OCTETS, "tx_q0_octets" }, + { 4, BGMAC_TX_Q1_PKTS, "tx_q1" }, + { 8, BGMAC_TX_Q1_OCTETS, "tx_q1_octets" }, + { 4, BGMAC_TX_Q2_PKTS, "tx_q2" }, + { 8, BGMAC_TX_Q2_OCTETS, "tx_q2_octets" }, + { 4, BGMAC_TX_Q3_PKTS, "tx_q3" }, + { 8, BGMAC_TX_Q3_OCTETS, "tx_q3_octets" }, + { 8, BGMAC_RX_GOOD_OCTETS, "rx_good_octets" }, + { 4, BGMAC_RX_GOOD_PKTS, "rx_good" }, + { 8, BGMAC_RX_OCTETS, "rx_octets" }, + { 4, BGMAC_RX_PKTS, "rx_pkts" }, + { 4, BGMAC_RX_BROADCAST_PKTS, "rx_broadcast" }, + { 4, BGMAC_RX_MULTICAST_PKTS, "rx_multicast" }, + { 4, BGMAC_RX_LEN_64, "rx_64" }, + { 4, BGMAC_RX_LEN_65_TO_127, "rx_65_127" }, + { 4, BGMAC_RX_LEN_128_TO_255, "rx_128_255" }, + { 4, BGMAC_RX_LEN_256_TO_511, "rx_256_511" }, + { 4, BGMAC_RX_LEN_512_TO_1023, "rx_512_1023" }, + { 4, BGMAC_RX_LEN_1024_TO_1522, "rx_1024_1522" }, + { 4, BGMAC_RX_LEN_1523_TO_2047, "rx_1523_2047" }, + { 4, BGMAC_RX_LEN_2048_TO_4095, "rx_2048_4095" }, + { 4, BGMAC_RX_LEN_4096_TO_8191, "rx_4096_8191" }, + { 4, BGMAC_RX_LEN_8192_TO_MAX, "rx_8192_max" }, + { 4, BGMAC_RX_JABBER_PKTS, "rx_jabber" }, + { 4, BGMAC_RX_OVERSIZE_PKTS, "rx_oversize" }, + { 4, BGMAC_RX_FRAGMENT_PKTS, "rx_fragment" }, + { 4, BGMAC_RX_MISSED_PKTS, "rx_missed" }, + { 4, BGMAC_RX_CRC_ALIGN_ERRS, "rx_crc_align" }, + { 4, BGMAC_RX_UNDERSIZE, "rx_undersize" }, + { 4, BGMAC_RX_CRC_ERRS, "rx_crc" }, + { 4, BGMAC_RX_ALIGN_ERRS, "rx_align" }, + { 4, BGMAC_RX_SYMBOL_ERRS, "rx_symbol" }, + { 4, BGMAC_RX_PAUSE_PKTS, "rx_pause" }, + { 4, BGMAC_RX_NONPAUSE_PKTS, "rx_nonpause" }, + { 4, BGMAC_RX_SACHANGES, "rx_sa_changes" }, + { 4, BGMAC_RX_UNI_PKTS, "rx_unicast" }, +}; + +#define BGMAC_STATS_LEN ARRAY_SIZE(bgmac_get_strings_stats) + +static int bgmac_get_sset_count(struct net_device *dev, int string_set) { - struct bgmac *bgmac = netdev_priv(net_dev); + switch (string_set) { + case ETH_SS_STATS: + return BGMAC_STATS_LEN; + } - return phy_ethtool_gset(bgmac->phy_dev, cmd); + return -EOPNOTSUPP; } -static int bgmac_set_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) +static void bgmac_get_strings(struct net_device *dev, u32 stringset, + u8 *data) { - struct bgmac *bgmac = netdev_priv(net_dev); + int i; + + if (stringset != ETH_SS_STATS) + return; - return phy_ethtool_sset(bgmac->phy_dev, cmd); + for (i = 0; i < BGMAC_STATS_LEN; i++) + strlcpy(data + i * ETH_GSTRING_LEN, + bgmac_get_strings_stats[i].name, ETH_GSTRING_LEN); +} + +static void bgmac_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *ss, uint64_t *data) +{ + struct bgmac *bgmac = netdev_priv(dev); + const struct bgmac_stat *s; + unsigned int i; + u64 val; + + if (!netif_running(dev)) + return; + + for (i = 0; i < BGMAC_STATS_LEN; i++) { + s = &bgmac_get_strings_stats[i]; + val = 0; + if (s->size == 8) + val = (u64)bgmac_read(bgmac, s->offset + 4) << 32; + val |= bgmac_read(bgmac, s->offset); + data[i] = val; + } } static void bgmac_get_drvinfo(struct net_device *net_dev, @@ -1406,9 +1521,12 @@ static void bgmac_get_drvinfo(struct net_device *net_dev, } static const struct ethtool_ops bgmac_ethtool_ops = { - .get_settings = bgmac_get_settings, - .set_settings = bgmac_set_settings, + .get_strings = bgmac_get_strings, + .get_sset_count = bgmac_get_sset_count, + .get_ethtool_stats = bgmac_get_ethtool_stats, .get_drvinfo = bgmac_get_drvinfo, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /************************************************** @@ -1429,7 +1547,7 @@ static int bgmac_mii_write(struct mii_bus *bus, int mii_id, int regnum, static void bgmac_adjust_link(struct net_device *net_dev) { struct bgmac *bgmac = netdev_priv(net_dev); - struct phy_device *phy_dev = bgmac->phy_dev; + struct phy_device *phy_dev = net_dev->phydev; bool update = false; if (phy_dev->link) { @@ -1473,8 +1591,6 @@ static int bgmac_fixed_phy_register(struct bgmac *bgmac) return err; } - bgmac->phy_dev = phy_dev; - return err; } @@ -1519,7 +1635,6 @@ static int bgmac_mii_register(struct bgmac *bgmac) err = PTR_ERR(phy_dev); goto err_unregister_bus; } - bgmac->phy_dev = phy_dev; return err; @@ -1588,6 +1703,7 @@ static int bgmac_probe(struct bcma_device *core) bgmac->net_dev = net_dev; bgmac->core = core; bcma_set_drvdata(core, bgmac); + SET_NETDEV_DEV(net_dev, &core->dev); /* Defaults */ memcpy(bgmac->net_dev->dev_addr, mac, ETH_ALEN); diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 9a03c142b742..99beb181f577 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -123,7 +123,7 @@ #define BGMAC_TX_LEN_1024_TO_1522 0x334 #define BGMAC_TX_LEN_1523_TO_2047 0x338 #define BGMAC_TX_LEN_2048_TO_4095 0x33c -#define BGMAC_TX_LEN_4095_TO_8191 0x340 +#define BGMAC_TX_LEN_4096_TO_8191 0x340 #define BGMAC_TX_LEN_8192_TO_MAX 0x344 #define BGMAC_TX_JABBER_PKTS 0x348 /* Error */ #define BGMAC_TX_OVERSIZE_PKTS 0x34c /* Error */ @@ -166,7 +166,7 @@ #define BGMAC_RX_LEN_1024_TO_1522 0x3e4 #define BGMAC_RX_LEN_1523_TO_2047 0x3e8 #define BGMAC_RX_LEN_2048_TO_4095 0x3ec -#define BGMAC_RX_LEN_4095_TO_8191 0x3f0 +#define BGMAC_RX_LEN_4096_TO_8191 0x3f0 #define BGMAC_RX_LEN_8192_TO_MAX 0x3f4 #define BGMAC_RX_JABBER_PKTS 0x3f8 /* Error */ #define BGMAC_RX_OVERSIZE_PKTS 0x3fc /* Error */ @@ -441,7 +441,6 @@ struct bgmac { struct net_device *net_dev; struct napi_struct napi; struct mii_bus *mii_bus; - struct phy_device *phy_dev; /* DMA */ struct bgmac_dma_ring tx_ring[BGMAC_MAX_TX_RINGS]; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index c5fe915870ad..97e892511666 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -59,9 +59,6 @@ #include <linux/semaphore.h> #include <linux/stringify.h> #include <linux/vmalloc.h> -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) -#include <net/geneve.h> -#endif #include "bnx2x.h" #include "bnx2x_init.h" #include "bnx2x_init_ops.h" @@ -10076,7 +10073,6 @@ static void bnx2x_parity_recover(struct bnx2x *bp) } } -#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_BNX2X_GENEVE) static int bnx2x_udp_port_update(struct bnx2x *bp) { struct bnx2x_func_switch_update_params *switch_update_params; @@ -10177,47 +10173,42 @@ static void __bnx2x_del_udp_port(struct bnx2x *bp, u16 port, DP(BNX2X_MSG_SP, "Deleted UDP tunnel [%d] port %d\n", type, port); } -#endif - -#ifdef CONFIG_BNX2X_VXLAN -static void bnx2x_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); - - __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); -} - -static void bnx2x_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); - - __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); -} -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) -static void bnx2x_add_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void bnx2x_udp_tunnel_add(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); + break; + case UDP_TUNNEL_TYPE_GENEVE: + __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + break; + default: + break; + } } -static void bnx2x_del_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void bnx2x_udp_tunnel_del(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); + break; + case UDP_TUNNEL_TYPE_GENEVE: + __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + break; + default: + break; + } } -#endif static int bnx2x_close(struct net_device *dev); @@ -10325,7 +10316,6 @@ sp_rtnl_not_reset: &bp->sp_rtnl_state)) bnx2x_update_mng_version(bp); -#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_BNX2X_GENEVE) if (test_and_clear_bit(BNX2X_SP_RTNL_CHANGE_UDP_PORT, &bp->sp_rtnl_state)) { if (bnx2x_udp_port_update(bp)) { @@ -10335,20 +10325,14 @@ sp_rtnl_not_reset: BNX2X_UDP_PORT_MAX); } else { /* Since we don't store additional port information, - * if no port is configured for any feature ask for + * if no ports are configured for any feature ask for * information about currently configured ports. */ -#ifdef CONFIG_BNX2X_VXLAN - if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count) - vxlan_get_rx_port(bp->dev); -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) - if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) - geneve_get_rx_port(bp->dev); -#endif + if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count && + !bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) + udp_tunnel_get_rx_info(bp->dev); } } -#endif /* work which needs rtnl lock not-taken (as it takes the lock itself and * can be called from other contexts as well) @@ -12551,14 +12535,8 @@ static int bnx2x_open(struct net_device *dev) if (rc) return rc; -#ifdef CONFIG_BNX2X_VXLAN - if (IS_PF(bp)) - vxlan_get_rx_port(dev); -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) if (IS_PF(bp)) - geneve_get_rx_port(dev); -#endif + udp_tunnel_get_rx_info(dev); return 0; } @@ -12895,52 +12873,71 @@ static int __bnx2x_vlan_configure_vid(struct bnx2x *bp, u16 vid, bool add) return rc; } -int bnx2x_vlan_reconfigure_vid(struct bnx2x *bp) +static int bnx2x_vlan_configure_vid_list(struct bnx2x *bp) { struct bnx2x_vlan_entry *vlan; int rc = 0; - if (!bp->vlan_cnt) { - DP(NETIF_MSG_IFUP, "No need to re-configure vlan filters\n"); - return 0; - } - + /* Configure all non-configured entries */ list_for_each_entry(vlan, &bp->vlan_reg, link) { - /* Prepare for cleanup in case of errors */ - if (rc) { - vlan->hw = false; - continue; - } - - if (!vlan->hw) + if (vlan->hw) continue; - DP(NETIF_MSG_IFUP, "Re-configuring vlan 0x%04x\n", vlan->vid); + if (bp->vlan_cnt >= bp->vlan_credit) + return -ENOBUFS; rc = __bnx2x_vlan_configure_vid(bp, vlan->vid, true); if (rc) { - BNX2X_ERR("Unable to configure VLAN %d\n", vlan->vid); - vlan->hw = false; - rc = -EINVAL; - continue; + BNX2X_ERR("Unable to config VLAN %d\n", vlan->vid); + return rc; } + + DP(NETIF_MSG_IFUP, "HW configured for VLAN %d\n", vlan->vid); + vlan->hw = true; + bp->vlan_cnt++; } - return rc; + return 0; +} + +static void bnx2x_vlan_configure(struct bnx2x *bp, bool set_rx_mode) +{ + bool need_accept_any_vlan; + + need_accept_any_vlan = !!bnx2x_vlan_configure_vid_list(bp); + + if (bp->accept_any_vlan != need_accept_any_vlan) { + bp->accept_any_vlan = need_accept_any_vlan; + DP(NETIF_MSG_IFUP, "Accept all VLAN %s\n", + bp->accept_any_vlan ? "raised" : "cleared"); + if (set_rx_mode) { + if (IS_PF(bp)) + bnx2x_set_rx_mode_inner(bp); + else + bnx2x_vfpf_storm_rx_mode(bp); + } + } +} + +int bnx2x_vlan_reconfigure_vid(struct bnx2x *bp) +{ + struct bnx2x_vlan_entry *vlan; + + /* The hw forgot all entries after reload */ + list_for_each_entry(vlan, &bp->vlan_reg, link) + vlan->hw = false; + bp->vlan_cnt = 0; + + /* Don't set rx mode here. Our caller will do it. */ + bnx2x_vlan_configure(bp, false); + + return 0; } static int bnx2x_vlan_rx_add_vid(struct net_device *dev, __be16 proto, u16 vid) { struct bnx2x *bp = netdev_priv(dev); struct bnx2x_vlan_entry *vlan; - bool hw = false; - int rc = 0; - - if (!netif_running(bp->dev)) { - DP(NETIF_MSG_IFUP, - "Ignoring VLAN configuration the interface is down\n"); - return -EFAULT; - } DP(NETIF_MSG_IFUP, "Adding VLAN %d\n", vid); @@ -12948,93 +12945,47 @@ static int bnx2x_vlan_rx_add_vid(struct net_device *dev, __be16 proto, u16 vid) if (!vlan) return -ENOMEM; - bp->vlan_cnt++; - if (bp->vlan_cnt > bp->vlan_credit && !bp->accept_any_vlan) { - DP(NETIF_MSG_IFUP, "Accept all VLAN raised\n"); - bp->accept_any_vlan = true; - if (IS_PF(bp)) - bnx2x_set_rx_mode_inner(bp); - else - bnx2x_vfpf_storm_rx_mode(bp); - } else if (bp->vlan_cnt <= bp->vlan_credit) { - rc = __bnx2x_vlan_configure_vid(bp, vid, true); - hw = true; - } - vlan->vid = vid; - vlan->hw = hw; - - if (!rc) { - list_add(&vlan->link, &bp->vlan_reg); - } else { - bp->vlan_cnt--; - kfree(vlan); - } + vlan->hw = false; + list_add_tail(&vlan->link, &bp->vlan_reg); - DP(NETIF_MSG_IFUP, "Adding VLAN result %d\n", rc); + if (netif_running(dev)) + bnx2x_vlan_configure(bp, true); - return rc; + return 0; } static int bnx2x_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid) { struct bnx2x *bp = netdev_priv(dev); struct bnx2x_vlan_entry *vlan; + bool found = false; int rc = 0; - if (!netif_running(bp->dev)) { - DP(NETIF_MSG_IFUP, - "Ignoring VLAN configuration the interface is down\n"); - return -EFAULT; - } - DP(NETIF_MSG_IFUP, "Removing VLAN %d\n", vid); - if (!bp->vlan_cnt) { - BNX2X_ERR("Unable to kill VLAN %d\n", vid); - return -EINVAL; - } - list_for_each_entry(vlan, &bp->vlan_reg, link) - if (vlan->vid == vid) + if (vlan->vid == vid) { + found = true; break; + } - if (vlan->vid != vid) { + if (!found) { BNX2X_ERR("Unable to kill VLAN %d - not found\n", vid); return -EINVAL; } - if (vlan->hw) + if (netif_running(dev) && vlan->hw) { rc = __bnx2x_vlan_configure_vid(bp, vid, false); + DP(NETIF_MSG_IFUP, "HW deconfigured for VLAN %d\n", vid); + bp->vlan_cnt--; + } list_del(&vlan->link); kfree(vlan); - bp->vlan_cnt--; - - if (bp->vlan_cnt <= bp->vlan_credit && bp->accept_any_vlan) { - /* Configure all non-configured entries */ - list_for_each_entry(vlan, &bp->vlan_reg, link) { - if (vlan->hw) - continue; - - rc = __bnx2x_vlan_configure_vid(bp, vlan->vid, true); - if (rc) { - BNX2X_ERR("Unable to config VLAN %d\n", - vlan->vid); - continue; - } - DP(NETIF_MSG_IFUP, "HW configured for VLAN %d\n", - vlan->vid); - vlan->hw = true; - } - DP(NETIF_MSG_IFUP, "Accept all VLAN Removed\n"); - bp->accept_any_vlan = false; - if (IS_PF(bp)) - bnx2x_set_rx_mode_inner(bp); - else - bnx2x_vfpf_storm_rx_mode(bp); - } + if (netif_running(dev)) + bnx2x_vlan_configure(bp, true); DP(NETIF_MSG_IFUP, "Removing VLAN result %d\n", rc); @@ -13072,14 +13023,8 @@ static const struct net_device_ops bnx2x_netdev_ops = { .ndo_get_phys_port_id = bnx2x_get_phys_port_id, .ndo_set_vf_link_state = bnx2x_set_vf_link_state, .ndo_features_check = bnx2x_features_check, -#ifdef CONFIG_BNX2X_VXLAN - .ndo_add_vxlan_port = bnx2x_add_vxlan_port, - .ndo_del_vxlan_port = bnx2x_del_vxlan_port, -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) - .ndo_add_geneve_port = bnx2x_add_geneve_port, - .ndo_del_geneve_port = bnx2x_del_geneve_port, -#endif + .ndo_udp_tunnel_add = bnx2x_udp_tunnel_add, + .ndo_udp_tunnel_del = bnx2x_udp_tunnel_del, }; static int bnx2x_set_coherency_mask(struct bnx2x *bp) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 72a2efff8e49..70b148a10ec8 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -37,9 +37,7 @@ #include <net/udp.h> #include <net/checksum.h> #include <net/ip6_checksum.h> -#if defined(CONFIG_VXLAN) || defined(CONFIG_VXLAN_MODULE) -#include <net/vxlan.h> -#endif +#include <net/udp_tunnel.h> #ifdef CONFIG_NET_RX_BUSY_POLL #include <net/busy_poll.h> #endif @@ -75,12 +73,22 @@ enum board_idx { BCM57301, BCM57302, BCM57304, + BCM57311, + BCM57312, BCM57402, BCM57404, BCM57406, + BCM57404_NPAR, + BCM57412, + BCM57414, + BCM57416, + BCM57417, + BCM57414_NPAR, BCM57314, BCM57304_VF, BCM57404_VF, + BCM57414_VF, + BCM57314_VF, }; /* indexed by enum above */ @@ -90,25 +98,45 @@ static const struct { { "Broadcom BCM57301 NetXtreme-C Single-port 10Gb Ethernet" }, { "Broadcom BCM57302 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, + { "Broadcom BCM57311 NetXtreme-C Single-port 10Gb Ethernet" }, + { "Broadcom BCM57312 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57402 NetXtreme-E Dual-port 10Gb Ethernet" }, { "Broadcom BCM57404 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57406 NetXtreme-E Dual-port 10GBase-T Ethernet" }, + { "Broadcom BCM57404 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57412 NetXtreme-E Dual-port 10Gb Ethernet" }, + { "Broadcom BCM57414 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, + { "Broadcom BCM57416 NetXtreme-E Dual-port 10GBase-T Ethernet" }, + { "Broadcom BCM57417 NetXtreme-E Dual-port 10GBase-T Ethernet" }, + { "Broadcom BCM57414 NetXtreme-E Ethernet Partition" }, { "Broadcom BCM57314 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Ethernet Virtual Function" }, { "Broadcom BCM57404 NetXtreme-E Ethernet Virtual Function" }, + { "Broadcom BCM57414 NetXtreme-E Ethernet Virtual Function" }, + { "Broadcom BCM57314 NetXtreme-E Ethernet Virtual Function" }, }; static const struct pci_device_id bnxt_pci_tbl[] = { { PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 }, { PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 }, { PCI_VDEVICE(BROADCOM, 0x16ca), .driver_data = BCM57304 }, + { PCI_VDEVICE(BROADCOM, 0x16ce), .driver_data = BCM57311 }, + { PCI_VDEVICE(BROADCOM, 0x16cf), .driver_data = BCM57312 }, { PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 }, { PCI_VDEVICE(BROADCOM, 0x16d1), .driver_data = BCM57404 }, { PCI_VDEVICE(BROADCOM, 0x16d2), .driver_data = BCM57406 }, + { PCI_VDEVICE(BROADCOM, 0x16d4), .driver_data = BCM57404_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16d6), .driver_data = BCM57412 }, + { PCI_VDEVICE(BROADCOM, 0x16d7), .driver_data = BCM57414 }, + { PCI_VDEVICE(BROADCOM, 0x16d8), .driver_data = BCM57416 }, + { PCI_VDEVICE(BROADCOM, 0x16d9), .driver_data = BCM57417 }, + { PCI_VDEVICE(BROADCOM, 0x16de), .driver_data = BCM57414_NPAR }, { PCI_VDEVICE(BROADCOM, 0x16df), .driver_data = BCM57314 }, #ifdef CONFIG_BNXT_SRIOV { PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = BCM57304_VF }, { PCI_VDEVICE(BROADCOM, 0x16d3), .driver_data = BCM57404_VF }, + { PCI_VDEVICE(BROADCOM, 0x16dc), .driver_data = BCM57414_VF }, + { PCI_VDEVICE(BROADCOM, 0x16e1), .driver_data = BCM57314_VF }, #endif { 0 } }; @@ -125,12 +153,14 @@ static const u16 bnxt_async_events_arr[] = { HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE, HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD, HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED, + HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE, HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE, }; static bool bnxt_vf_pciid(enum board_idx idx) { - return (idx == BCM57304_VF || idx == BCM57404_VF); + return (idx == BCM57304_VF || idx == BCM57404_VF || + idx == BCM57314_VF || idx == BCM57414_VF); } #define DB_CP_REARM_FLAGS (DB_KEY_CP | DB_IDX_VALID) @@ -286,7 +316,9 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev) cpu_to_le32(DB_KEY_TX_PUSH | DB_LONG_TX_PUSH | prod); txr->tx_prod = prod; + tx_buf->is_push = 1; netdev_tx_sent_queue(txq, skb->len); + wmb(); /* Sync is_push and byte queue before pushing data */ push_len = (length + sizeof(*tx_push) + 7) / 8; if (push_len > 16) { @@ -298,7 +330,6 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev) push_len); } - tx_buf->is_push = 1; goto tx_done; } @@ -919,6 +950,7 @@ static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, } tpa_info->flags2 = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_flags2); tpa_info->metadata = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_metadata); + tpa_info->hdr_info = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_hdr_info); rxr->rx_prod = NEXT_RX(prod); cons = NEXT_RX(cons); @@ -937,32 +969,102 @@ static void bnxt_abort_tpa(struct bnxt *bp, struct bnxt_napi *bnapi, bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs); } +static struct sk_buff *bnxt_gro_func_5731x(struct bnxt_tpa_info *tpa_info, + int payload_off, int tcp_ts, + struct sk_buff *skb) +{ +#ifdef CONFIG_INET + struct tcphdr *th; + int len, nw_off; + u16 outer_ip_off, inner_ip_off, inner_mac_off; + u32 hdr_info = tpa_info->hdr_info; + bool loopback = false; + + inner_ip_off = BNXT_TPA_INNER_L3_OFF(hdr_info); + inner_mac_off = BNXT_TPA_INNER_L2_OFF(hdr_info); + outer_ip_off = BNXT_TPA_OUTER_L3_OFF(hdr_info); + + /* If the packet is an internal loopback packet, the offsets will + * have an extra 4 bytes. + */ + if (inner_mac_off == 4) { + loopback = true; + } else if (inner_mac_off > 4) { + __be16 proto = *((__be16 *)(skb->data + inner_ip_off - + ETH_HLEN - 2)); + + /* We only support inner iPv4/ipv6. If we don't see the + * correct protocol ID, it must be a loopback packet where + * the offsets are off by 4. + */ + if (proto != htons(ETH_P_IP) && proto && htons(ETH_P_IPV6)) + loopback = true; + } + if (loopback) { + /* internal loopback packet, subtract all offsets by 4 */ + inner_ip_off -= 4; + inner_mac_off -= 4; + outer_ip_off -= 4; + } + + nw_off = inner_ip_off - ETH_HLEN; + skb_set_network_header(skb, nw_off); + if (tpa_info->flags2 & RX_TPA_START_CMP_FLAGS2_IP_TYPE) { + struct ipv6hdr *iph = ipv6_hdr(skb); + + skb_set_transport_header(skb, nw_off + sizeof(struct ipv6hdr)); + len = skb->len - skb_transport_offset(skb); + th = tcp_hdr(skb); + th->check = ~tcp_v6_check(len, &iph->saddr, &iph->daddr, 0); + } else { + struct iphdr *iph = ip_hdr(skb); + + skb_set_transport_header(skb, nw_off + sizeof(struct iphdr)); + len = skb->len - skb_transport_offset(skb); + th = tcp_hdr(skb); + th->check = ~tcp_v4_check(len, iph->saddr, iph->daddr, 0); + } + + if (inner_mac_off) { /* tunnel */ + struct udphdr *uh = NULL; + __be16 proto = *((__be16 *)(skb->data + outer_ip_off - + ETH_HLEN - 2)); + + if (proto == htons(ETH_P_IP)) { + struct iphdr *iph = (struct iphdr *)skb->data; + + if (iph->protocol == IPPROTO_UDP) + uh = (struct udphdr *)(iph + 1); + } else { + struct ipv6hdr *iph = (struct ipv6hdr *)skb->data; + + if (iph->nexthdr == IPPROTO_UDP) + uh = (struct udphdr *)(iph + 1); + } + if (uh) { + if (uh->check) + skb_shinfo(skb)->gso_type |= + SKB_GSO_UDP_TUNNEL_CSUM; + else + skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL; + } + } +#endif + return skb; +} + #define BNXT_IPV4_HDR_SIZE (sizeof(struct iphdr) + sizeof(struct tcphdr)) #define BNXT_IPV6_HDR_SIZE (sizeof(struct ipv6hdr) + sizeof(struct tcphdr)) -static inline struct sk_buff *bnxt_gro_skb(struct bnxt_tpa_info *tpa_info, - struct rx_tpa_end_cmp *tpa_end, - struct rx_tpa_end_cmp_ext *tpa_end1, +static struct sk_buff *bnxt_gro_func_5730x(struct bnxt_tpa_info *tpa_info, + int payload_off, int tcp_ts, struct sk_buff *skb) { #ifdef CONFIG_INET struct tcphdr *th; - int payload_off, tcp_opt_len = 0; - int len, nw_off; - u16 segs; + int len, nw_off, tcp_opt_len; - segs = TPA_END_TPA_SEGS(tpa_end); - if (segs == 1) - return skb; - - NAPI_GRO_CB(skb)->count = segs; - skb_shinfo(skb)->gso_size = - le32_to_cpu(tpa_end1->rx_tpa_end_cmp_seg_len); - skb_shinfo(skb)->gso_type = tpa_info->gso_type; - payload_off = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) & - RX_TPA_END_CMP_PAYLOAD_OFFSET) >> - RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT; - if (TPA_END_GRO_TS(tpa_end)) + if (tcp_ts) tcp_opt_len = 12; if (tpa_info->gso_type == SKB_GSO_TCPV4) { @@ -1019,6 +1121,32 @@ static inline struct sk_buff *bnxt_gro_skb(struct bnxt_tpa_info *tpa_info, return skb; } +static inline struct sk_buff *bnxt_gro_skb(struct bnxt *bp, + struct bnxt_tpa_info *tpa_info, + struct rx_tpa_end_cmp *tpa_end, + struct rx_tpa_end_cmp_ext *tpa_end1, + struct sk_buff *skb) +{ +#ifdef CONFIG_INET + int payload_off; + u16 segs; + + segs = TPA_END_TPA_SEGS(tpa_end); + if (segs == 1) + return skb; + + NAPI_GRO_CB(skb)->count = segs; + skb_shinfo(skb)->gso_size = + le32_to_cpu(tpa_end1->rx_tpa_end_cmp_seg_len); + skb_shinfo(skb)->gso_type = tpa_info->gso_type; + payload_off = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) & + RX_TPA_END_CMP_PAYLOAD_OFFSET) >> + RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT; + skb = bp->gro_func(tpa_info, payload_off, TPA_END_GRO_TS(tpa_end), skb); +#endif + return skb; +} + static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons, @@ -1112,19 +1240,13 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp, if (tpa_info->hash_type != PKT_HASH_TYPE_NONE) skb_set_hash(skb, tpa_info->rss_hash, tpa_info->hash_type); - if (tpa_info->flags2 & RX_CMP_FLAGS2_META_FORMAT_VLAN) { - netdev_features_t features = skb->dev->features; + if ((tpa_info->flags2 & RX_CMP_FLAGS2_META_FORMAT_VLAN) && + (skb->dev->features & NETIF_F_HW_VLAN_CTAG_RX)) { u16 vlan_proto = tpa_info->metadata >> RX_CMP_FLAGS2_METADATA_TPID_SFT; + u16 vtag = tpa_info->metadata & RX_CMP_FLAGS2_METADATA_VID_MASK; - if (((features & NETIF_F_HW_VLAN_CTAG_RX) && - vlan_proto == ETH_P_8021Q) || - ((features & NETIF_F_HW_VLAN_STAG_RX) && - vlan_proto == ETH_P_8021AD)) { - __vlan_hwaccel_put_tag(skb, htons(vlan_proto), - tpa_info->metadata & - RX_CMP_FLAGS2_METADATA_VID_MASK); - } + __vlan_hwaccel_put_tag(skb, htons(vlan_proto), vtag); } skb_checksum_none_assert(skb); @@ -1135,7 +1257,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp, } if (TPA_END_GRO(tpa_end)) - skb = bnxt_gro_skb(tpa_info, tpa_end, tpa_end1, skb); + skb = bnxt_gro_skb(bp, tpa_info, tpa_end, tpa_end1, skb); return skb; } @@ -1277,19 +1399,14 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons, skb->protocol = eth_type_trans(skb, dev); - if (rxcmp1->rx_cmp_flags2 & - cpu_to_le32(RX_CMP_FLAGS2_META_FORMAT_VLAN)) { - netdev_features_t features = skb->dev->features; + if ((rxcmp1->rx_cmp_flags2 & + cpu_to_le32(RX_CMP_FLAGS2_META_FORMAT_VLAN)) && + (skb->dev->features & NETIF_F_HW_VLAN_CTAG_RX)) { u32 meta_data = le32_to_cpu(rxcmp1->rx_cmp_meta_data); + u16 vtag = meta_data & RX_CMP_FLAGS2_METADATA_VID_MASK; u16 vlan_proto = meta_data >> RX_CMP_FLAGS2_METADATA_TPID_SFT; - if (((features & NETIF_F_HW_VLAN_CTAG_RX) && - vlan_proto == ETH_P_8021Q) || - ((features & NETIF_F_HW_VLAN_STAG_RX) && - vlan_proto == ETH_P_8021AD)) - __vlan_hwaccel_put_tag(skb, htons(vlan_proto), - meta_data & - RX_CMP_FLAGS2_METADATA_VID_MASK); + __vlan_hwaccel_put_tag(skb, htons(vlan_proto), vtag); } skb_checksum_none_assert(skb); @@ -1368,6 +1485,11 @@ static int bnxt_async_event_process(struct bnxt *bp, set_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event); break; } + case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE: + if (BNXT_PF(bp)) + goto async_event_process_exit; + set_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event); + break; default: netdev_err(bp->dev, "unhandled ASYNC event (id 0x%x)\n", event_id); @@ -2272,7 +2394,7 @@ static void bnxt_set_tpa_flags(struct bnxt *bp) bp->flags &= ~BNXT_FLAG_TPA; if (bp->dev->features & NETIF_F_LRO) bp->flags |= BNXT_FLAG_LRO; - if ((bp->dev->features & NETIF_F_GRO) && (bp->pdev->revision > 0)) + if (bp->dev->features & NETIF_F_GRO) bp->flags |= BNXT_FLAG_GRO; } @@ -3287,11 +3409,13 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) unsigned int ring = 0, grp_idx; struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id]; struct hwrm_vnic_cfg_input req = {0}; + u16 def_vlan = 0; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_CFG, -1, -1); /* Only RSS support for now TBD: COS & LB */ req.enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP | - VNIC_CFG_REQ_ENABLES_RSS_RULE); + VNIC_CFG_REQ_ENABLES_RSS_RULE | + VNIC_CFG_REQ_ENABLES_MRU); req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx); req.cos_rule = cpu_to_le16(0xffff); if (vnic->flags & BNXT_VNIC_RSS_FLAG) @@ -3307,7 +3431,11 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) req.mru = cpu_to_le16(bp->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN); - if (bp->flags & BNXT_FLAG_STRIP_VLAN) +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp)) + def_vlan = bp->vf.vlan; +#endif + if ((bp->flags & BNXT_FLAG_STRIP_VLAN) || def_vlan) req.flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE); return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); @@ -3824,7 +3952,7 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_ALLOC, -1, -1); - req.update_period_ms = cpu_to_le32(1000); + req.update_period_ms = cpu_to_le32(bp->stats_coal_ticks / 1000); mutex_lock(&bp->hwrm_cmd_lock); for (i = 0; i < bp->cp_nr_rings; i++) { @@ -3846,6 +3974,39 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) return 0; } +static int bnxt_hwrm_func_qcfg(struct bnxt *bp) +{ + struct hwrm_func_qcfg_input req = {0}; + struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr; + int rc; + + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCFG, -1, -1); + req.fid = cpu_to_le16(0xffff); + mutex_lock(&bp->hwrm_cmd_lock); + rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + if (rc) + goto func_qcfg_exit; + +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp)) { + struct bnxt_vf_info *vf = &bp->vf; + + vf->vlan = le16_to_cpu(resp->vlan) & VLAN_VID_MASK; + } +#endif + switch (resp->port_partition_type) { + case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0: + case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5: + case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0: + bp->port_partition_type = resp->port_partition_type; + break; + } + +func_qcfg_exit: + mutex_unlock(&bp->hwrm_cmd_lock); + return rc; +} + int bnxt_hwrm_func_qcaps(struct bnxt *bp) { int rc = 0; @@ -3865,6 +4026,7 @@ int bnxt_hwrm_func_qcaps(struct bnxt *bp) pf->fw_fid = le16_to_cpu(resp->fid); pf->port_id = le16_to_cpu(resp->port_id); + bp->dev->dev_port = pf->port_id; memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN); memcpy(bp->dev->dev_addr, pf->mac_addr, ETH_ALEN); pf->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx); @@ -4000,6 +4162,8 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp) if (resp->hwrm_intf_maj >= 1) bp->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len); + bp->chip_num = le16_to_cpu(resp->chip_num); + hwrm_ver_get_exit: mutex_unlock(&bp->hwrm_cmd_lock); return rc; @@ -4153,6 +4317,16 @@ static int bnxt_alloc_rfs_vnics(struct bnxt *bp) #endif } +/* Allow PF and VF with default VLAN to be in promiscuous mode */ +static bool bnxt_promisc_ok(struct bnxt *bp) +{ +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp) && !bp->vf.vlan) + return false; +#endif + return true; +} + static int bnxt_cfg_rx_mode(struct bnxt *); static bool bnxt_mc_list_updated(struct bnxt *, u32 *); @@ -4218,7 +4392,7 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) vnic->rx_mask = CFA_L2_SET_RX_MASK_REQ_MASK_BCAST; - if ((bp->dev->flags & IFF_PROMISC) && BNXT_PF(bp)) + if ((bp->dev->flags & IFF_PROMISC) && bnxt_promisc_ok(bp)) vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS; if (bp->dev->flags & IFF_ALLMULTI) { @@ -4240,6 +4414,11 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) netdev_warn(bp->dev, "HWRM set coalescing failure rc: %x\n", rc); + if (BNXT_VF(bp)) { + bnxt_hwrm_func_qcfg(bp); + netdev_update_features(bp->dev); + } + return 0; err_out: @@ -4654,6 +4833,7 @@ static int bnxt_hwrm_phy_qcaps(struct bnxt *bp) int rc = 0; struct hwrm_port_phy_qcaps_input req = {0}; struct hwrm_port_phy_qcaps_output *resp = bp->hwrm_cmd_resp_addr; + struct bnxt_link_info *link_info = &bp->link_info; if (bp->hwrm_spec_code < 0x10201) return 0; @@ -4676,6 +4856,8 @@ static int bnxt_hwrm_phy_qcaps(struct bnxt *bp) bp->lpi_tmr_hi = le32_to_cpu(resp->valid_tx_lpi_timer_high) & PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_MASK; } + link_info->support_auto_speeds = + le16_to_cpu(resp->supported_speeds_auto_mode); hwrm_phy_qcaps_exit: mutex_unlock(&bp->hwrm_cmd_lock); @@ -4933,7 +5115,7 @@ static int bnxt_hwrm_shutdown_link(struct bnxt *bp) { struct hwrm_port_phy_cfg_input req = {0}; - if (BNXT_VF(bp)) + if (!BNXT_SINGLE_PF(bp)) return 0; if (pci_num_vf(bp->pdev)) @@ -5083,15 +5265,8 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init) netdev_warn(bp->dev, "failed to update phy settings\n"); } - if (irq_re_init) { -#if defined(CONFIG_VXLAN) || defined(CONFIG_VXLAN_MODULE) - vxlan_get_rx_port(bp->dev); -#endif - if (!bnxt_hwrm_tunnel_dst_port_alloc( - bp, htons(0x17c1), - TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE)) - bp->nge_port_cnt = 1; - } + if (irq_re_init) + udp_tunnel_get_rx_info(bp->dev); set_bit(BNXT_STATE_OPEN, &bp->state); bnxt_enable_int(bp); @@ -5132,12 +5307,19 @@ static int bnxt_open(struct net_device *dev) struct bnxt *bp = netdev_priv(dev); int rc = 0; - rc = bnxt_hwrm_func_reset(bp); - if (rc) { - netdev_err(bp->dev, "hwrm chip reset failure rc: %x\n", - rc); - rc = -1; - return rc; + if (!test_bit(BNXT_STATE_FN_RST_DONE, &bp->state)) { + rc = bnxt_hwrm_func_reset(bp); + if (rc) { + netdev_err(bp->dev, "hwrm chip reset failure rc: %x\n", + rc); + rc = -EBUSY; + return rc; + } + /* Do func_reset during the 1st PF open only to prevent killing + * the VFs when the PF is brought down and up. + */ + if (BNXT_PF(bp)) + set_bit(BNXT_STATE_FN_RST_DONE, &bp->state); } return __bnxt_open_nic(bp, true, true); } @@ -5357,8 +5539,7 @@ static void bnxt_set_rx_mode(struct net_device *dev) CFA_L2_SET_RX_MASK_REQ_MASK_MCAST | CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST); - /* Only allow PF to be in promiscuous mode */ - if ((dev->flags & IFF_PROMISC) && BNXT_PF(bp)) + if ((dev->flags & IFF_PROMISC) && bnxt_promisc_ok(bp)) mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS; uc_update = bnxt_uc_list_updated(bp); @@ -5466,6 +5647,27 @@ static netdev_features_t bnxt_fix_features(struct net_device *dev, if (!bnxt_rfs_capable(bp)) features &= ~NETIF_F_NTUPLE; + + /* Both CTAG and STAG VLAN accelaration on the RX side have to be + * turned on or off together. + */ + if ((features & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX)) != + (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX)) { + if (dev->features & NETIF_F_HW_VLAN_CTAG_RX) + features &= ~(NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_STAG_RX); + else + features |= NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_STAG_RX; + } +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp)) { + if (bp->vf.vlan) { + features &= ~(NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_STAG_RX); + } + } +#endif return features; } @@ -5581,9 +5783,10 @@ static void bnxt_dbg_dump_states(struct bnxt *bp) } } -static void bnxt_reset_task(struct bnxt *bp) +static void bnxt_reset_task(struct bnxt *bp, bool silent) { - bnxt_dbg_dump_states(bp); + if (!silent) + bnxt_dbg_dump_states(bp); if (netif_running(bp->dev)) { bnxt_close_nic(bp, false, false); bnxt_open_nic(bp, false, false); @@ -5634,6 +5837,23 @@ bnxt_restart_timer: mod_timer(&bp->timer, jiffies + bp->current_interval); } +/* Only called from bnxt_sp_task() */ +static void bnxt_reset(struct bnxt *bp, bool silent) +{ + /* bnxt_reset_task() calls bnxt_close_nic() which waits + * for BNXT_STATE_IN_SP_TASK to clear. + * If there is a parallel dev_close(), bnxt_close() may be holding + * rtnl() and waiting for BNXT_STATE_IN_SP_TASK to clear. So we + * must clear BNXT_STATE_IN_SP_TASK before holding rtnl(). + */ + clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + rtnl_lock(); + if (test_bit(BNXT_STATE_OPEN, &bp->state)) + bnxt_reset_task(bp, silent); + set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + rtnl_unlock(); +} + static void bnxt_cfg_ntp_filters(struct bnxt *); static void bnxt_sp_task(struct work_struct *work) @@ -5670,16 +5890,20 @@ static void bnxt_sp_task(struct work_struct *work) bnxt_hwrm_tunnel_dst_port_free( bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN); } - if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) { - /* bnxt_reset_task() calls bnxt_close_nic() which waits - * for BNXT_STATE_IN_SP_TASK to clear. - */ - clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); - rtnl_lock(); - bnxt_reset_task(bp); - set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); - rtnl_unlock(); + if (test_and_clear_bit(BNXT_GENEVE_ADD_PORT_SP_EVENT, &bp->sp_event)) { + bnxt_hwrm_tunnel_dst_port_alloc( + bp, bp->nge_port, + TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE); } + if (test_and_clear_bit(BNXT_GENEVE_DEL_PORT_SP_EVENT, &bp->sp_event)) { + bnxt_hwrm_tunnel_dst_port_free( + bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE); + } + if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) + bnxt_reset(bp, false); + + if (test_and_clear_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event)) + bnxt_reset(bp, true); if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) bnxt_get_port_module_status(bp); @@ -5770,6 +5994,8 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev) bp->tx_coal_ticks_irq = 2; bp->tx_coal_bufs_irq = 2; + bp->stats_coal_ticks = BNXT_DEF_STATS_COAL_TICKS; + init_timer(&bp->timer); bp->timer.data = (unsigned long)bp; bp->timer.function = bnxt_timer; @@ -5835,7 +6061,7 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu) { struct bnxt *bp = netdev_priv(dev); - if (new_mtu < 60 || new_mtu > 9000) + if (new_mtu < 60 || new_mtu > 9500) return -EINVAL; if (netif_running(dev)) @@ -6044,47 +6270,83 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp) #endif /* CONFIG_RFS_ACCEL */ -static void bnxt_add_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void bnxt_udp_tunnel_add(struct net_device *dev, + struct udp_tunnel_info *ti) { struct bnxt *bp = netdev_priv(dev); - if (!netif_running(dev)) + if (ti->sa_family != AF_INET6 && ti->sa_family != AF_INET) return; - if (sa_family != AF_INET6 && sa_family != AF_INET) + if (!netif_running(dev)) return; - if (bp->vxlan_port_cnt && bp->vxlan_port != port) - return; + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (bp->vxlan_port_cnt && bp->vxlan_port != ti->port) + return; - bp->vxlan_port_cnt++; - if (bp->vxlan_port_cnt == 1) { - bp->vxlan_port = port; - set_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event); - schedule_work(&bp->sp_task); + bp->vxlan_port_cnt++; + if (bp->vxlan_port_cnt == 1) { + bp->vxlan_port = ti->port; + set_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event); + schedule_work(&bp->sp_task); + } + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (bp->nge_port_cnt && bp->nge_port != ti->port) + return; + + bp->nge_port_cnt++; + if (bp->nge_port_cnt == 1) { + bp->nge_port = ti->port; + set_bit(BNXT_GENEVE_ADD_PORT_SP_EVENT, &bp->sp_event); + } + break; + default: + return; } + + schedule_work(&bp->sp_task); } -static void bnxt_del_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void bnxt_udp_tunnel_del(struct net_device *dev, + struct udp_tunnel_info *ti) { struct bnxt *bp = netdev_priv(dev); - if (!netif_running(dev)) + if (ti->sa_family != AF_INET6 && ti->sa_family != AF_INET) return; - if (sa_family != AF_INET6 && sa_family != AF_INET) + if (!netif_running(dev)) return; - if (bp->vxlan_port_cnt && bp->vxlan_port == port) { + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (!bp->vxlan_port_cnt || bp->vxlan_port != ti->port) + return; bp->vxlan_port_cnt--; - if (bp->vxlan_port_cnt == 0) { - set_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event); - schedule_work(&bp->sp_task); - } + if (bp->vxlan_port_cnt != 0) + return; + + set_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (!bp->nge_port_cnt || bp->nge_port != ti->port) + return; + bp->nge_port_cnt--; + + if (bp->nge_port_cnt != 0) + return; + + set_bit(BNXT_GENEVE_DEL_PORT_SP_EVENT, &bp->sp_event); + break; + default: + return; } + + schedule_work(&bp->sp_task); } static const struct net_device_ops bnxt_netdev_ops = { @@ -6115,8 +6377,8 @@ static const struct net_device_ops bnxt_netdev_ops = { #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = bnxt_rx_flow_steer, #endif - .ndo_add_vxlan_port = bnxt_add_vxlan_port, - .ndo_del_vxlan_port = bnxt_del_vxlan_port, + .ndo_udp_tunnel_add = bnxt_udp_tunnel_add, + .ndo_udp_tunnel_del = bnxt_udp_tunnel_del, #ifdef CONFIG_NET_RX_BUSY_POLL .ndo_busy_poll = bnxt_busy_poll, #endif @@ -6165,6 +6427,12 @@ static int bnxt_probe_phy(struct bnxt *bp) return rc; } + /* Older firmware does not have supported_auto_speeds, so assume + * that all supported speeds can be autonegotiated. + */ + if (link_info->auto_link_speeds && !link_info->support_auto_speeds) + link_info->support_auto_speeds = link_info->support_speeds; + /*initialize the ethool setting copy with NVM settings */ if (BNXT_AUTO_MODE(link_info->auto_mode)) { link_info->autoneg = BNXT_AUTONEG_SPEED; @@ -6338,7 +6606,13 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto init_err; mutex_init(&bp->hwrm_cmd_lock); - bnxt_hwrm_ver_get(bp); + rc = bnxt_hwrm_ver_get(bp); + if (rc) + goto init_err; + + bp->gro_func = bnxt_gro_func_5730x; + if (BNXT_CHIP_NUM_57X1X(bp->chip_num)) + bp->gro_func = bnxt_gro_func_5731x; rc = bnxt_hwrm_func_drv_rgtr(bp); if (rc) @@ -6361,6 +6635,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto init_err; } + bnxt_hwrm_func_qcfg(bp); + bnxt_set_tpa_flags(bp); bnxt_set_ring_params(bp); if (BNXT_PF(bp)) @@ -6420,6 +6696,7 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state) { struct net_device *netdev = pci_get_drvdata(pdev); + struct bnxt *bp = netdev_priv(netdev); netdev_info(netdev, "PCI I/O error detected\n"); @@ -6434,6 +6711,8 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev, if (netif_running(netdev)) bnxt_close(netdev); + /* So that func_reset will be done during slot_reset */ + clear_bit(BNXT_STATE_FN_RST_DONE, &bp->state); pci_disable_device(pdev); rtnl_unlock(); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 2824d65b2e35..2313e37e6eb5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -11,10 +11,10 @@ #define BNXT_H #define DRV_MODULE_NAME "bnxt_en" -#define DRV_MODULE_VERSION "1.2.0" +#define DRV_MODULE_VERSION "1.3.0" #define DRV_VER_MAJ 1 -#define DRV_VER_MIN 0 +#define DRV_VER_MIN 3 #define DRV_VER_UPD 0 struct tx_bd { @@ -298,13 +298,14 @@ struct rx_tpa_start_cmp_ext { #define RX_TPA_START_CMP_FLAGS2_L4_CS_CALC (0x1 << 1) #define RX_TPA_START_CMP_FLAGS2_T_IP_CS_CALC (0x1 << 2) #define RX_TPA_START_CMP_FLAGS2_T_L4_CS_CALC (0x1 << 3) + #define RX_TPA_START_CMP_FLAGS2_IP_TYPE (0x1 << 8) __le32 rx_tpa_start_cmp_metadata; __le32 rx_tpa_start_cmp_cfa_code_v2; #define RX_TPA_START_CMP_V2 (0x1 << 0) #define RX_TPA_START_CMP_CFA_CODE (0xffff << 16) #define RX_TPA_START_CMPL_CFA_CODE_SHIFT 16 - __le32 rx_tpa_start_cmp_unused5; + __le32 rx_tpa_start_cmp_hdr_info; }; struct rx_tpa_end_cmp { @@ -358,7 +359,8 @@ struct rx_tpa_end_cmp { RX_TPA_END_CMP_FLAGS_PLACEMENT_ANY_GRO) #define TPA_END_GRO_TS(rx_tpa_end) \ - ((rx_tpa_end)->rx_tpa_end_cmp_tsdelta & cpu_to_le32(RX_TPA_END_GRO_TS)) + (!!((rx_tpa_end)->rx_tpa_end_cmp_tsdelta & \ + cpu_to_le32(RX_TPA_END_GRO_TS))) struct rx_tpa_end_cmp_ext { __le32 rx_tpa_end_cmp_dup_acks; @@ -584,6 +586,19 @@ struct bnxt_tpa_info { u32 metadata; enum pkt_hash_types hash_type; u32 rss_hash; + u32 hdr_info; + +#define BNXT_TPA_L4_SIZE(hdr_info) \ + (((hdr_info) & 0xf8000000) ? ((hdr_info) >> 27) : 32) + +#define BNXT_TPA_INNER_L3_OFF(hdr_info) \ + (((hdr_info) >> 18) & 0x1ff) + +#define BNXT_TPA_INNER_L2_OFF(hdr_info) \ + (((hdr_info) >> 9) & 0x1ff) + +#define BNXT_TPA_OUTER_L3_OFF(hdr_info) \ + ((hdr_info) & 0x1ff) }; struct bnxt_rx_ring_info { @@ -739,8 +754,8 @@ struct bnxt_vf_info { struct bnxt_pf_info { #define BNXT_FIRST_PF_FID 1 #define BNXT_FIRST_VF_FID 128 - u32 fw_fid; - u8 port_id; + u16 fw_fid; + u16 port_id; u8 mac_addr[ETH_ALEN]; u16 max_rsscos_ctxs; u16 max_cp_rings; @@ -835,6 +850,7 @@ struct bnxt_link_info { #define BNXT_LINK_SPEED_MSK_25GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_25GB #define BNXT_LINK_SPEED_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB #define BNXT_LINK_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB + u16 support_auto_speeds; u16 lp_auto_link_speeds; u16 force_link_speed; u32 preemphasis; @@ -873,6 +889,44 @@ struct bnxt { void __iomem *bar2; u32 reg_base; + u16 chip_num; +#define CHIP_NUM_57301 0x16c8 +#define CHIP_NUM_57302 0x16c9 +#define CHIP_NUM_57304 0x16ca +#define CHIP_NUM_57402 0x16d0 +#define CHIP_NUM_57404 0x16d1 +#define CHIP_NUM_57406 0x16d2 + +#define CHIP_NUM_57311 0x16ce +#define CHIP_NUM_57312 0x16cf +#define CHIP_NUM_57314 0x16df +#define CHIP_NUM_57412 0x16d6 +#define CHIP_NUM_57414 0x16d7 +#define CHIP_NUM_57416 0x16d8 +#define CHIP_NUM_57417 0x16d9 + +#define BNXT_CHIP_NUM_5730X(chip_num) \ + ((chip_num) >= CHIP_NUM_57301 && \ + (chip_num) <= CHIP_NUM_57304) + +#define BNXT_CHIP_NUM_5740X(chip_num) \ + ((chip_num) >= CHIP_NUM_57402 && \ + (chip_num) <= CHIP_NUM_57406) + +#define BNXT_CHIP_NUM_5731X(chip_num) \ + ((chip_num) == CHIP_NUM_57311 || \ + (chip_num) == CHIP_NUM_57312 || \ + (chip_num) == CHIP_NUM_57314) + +#define BNXT_CHIP_NUM_5741X(chip_num) \ + ((chip_num) >= CHIP_NUM_57412 && \ + (chip_num) <= CHIP_NUM_57417) + +#define BNXT_CHIP_NUM_57X0X(chip_num) \ + (BNXT_CHIP_NUM_5730X(chip_num) || BNXT_CHIP_NUM_5740X(chip_num)) + +#define BNXT_CHIP_NUM_57X1X(chip_num) \ + (BNXT_CHIP_NUM_5731X(chip_num) || BNXT_CHIP_NUM_5741X(chip_num)) struct net_device *dev; struct pci_dev *pdev; @@ -907,12 +961,17 @@ struct bnxt { #define BNXT_PF(bp) (!((bp)->flags & BNXT_FLAG_VF)) #define BNXT_VF(bp) ((bp)->flags & BNXT_FLAG_VF) +#define BNXT_NPAR(bp) ((bp)->port_partition_type) +#define BNXT_SINGLE_PF(bp) (BNXT_PF(bp) && !BNXT_NPAR(bp)) struct bnxt_napi **bnapi; struct bnxt_rx_ring_info *rx_ring; struct bnxt_tx_ring_info *tx_ring; + struct sk_buff * (*gro_func)(struct bnxt_tpa_info *, int, int, + struct sk_buff *); + u32 rx_buf_size; u32 rx_buf_use_size; /* useable size */ u32 rx_ring_size; @@ -959,6 +1018,7 @@ struct bnxt { unsigned long state; #define BNXT_STATE_OPEN 0 #define BNXT_STATE_IN_SP_TASK 1 +#define BNXT_STATE_FN_RST_DONE 2 struct bnxt_irq *irq_tbl; u8 mac_addr[ETH_ALEN]; @@ -991,8 +1051,10 @@ struct bnxt { __be16 vxlan_port; u8 vxlan_port_cnt; __le16 vxlan_fw_dst_port_id; + __be16 nge_port; u8 nge_port_cnt; __le16 nge_fw_dst_port_id; + u8 port_partition_type; u16 rx_coal_ticks; u16 rx_coal_ticks_irq; @@ -1005,6 +1067,11 @@ struct bnxt { #define BNXT_USEC_TO_COAL_TIMER(x) ((x) * 25 / 2) + u32 stats_coal_ticks; +#define BNXT_DEF_STATS_COAL_TICKS 1000000 +#define BNXT_MIN_STATS_COAL_TICKS 250000 +#define BNXT_MAX_STATS_COAL_TICKS 1000000 + struct work_struct sp_task; unsigned long sp_event; #define BNXT_RX_MASK_SP_EVENT 0 @@ -1018,6 +1085,9 @@ struct bnxt { #define BNXT_HWRM_PF_UNLOAD_SP_EVENT 8 #define BNXT_PERIODIC_STATS_SP_EVENT 9 #define BNXT_HWRM_PORT_MODULE_SP_EVENT 10 +#define BNXT_RESET_TASK_SILENT_SP_EVENT 11 +#define BNXT_GENEVE_ADD_PORT_SP_EVENT 12 +#define BNXT_GENEVE_DEL_PORT_SP_EVENT 13 struct bnxt_pf_info pf; #ifdef CONFIG_BNXT_SRIOV diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index a38cb047b540..0f7dd861ab4d 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -56,6 +56,8 @@ static int bnxt_get_coalesce(struct net_device *dev, coal->tx_coalesce_usecs_irq = bp->tx_coal_ticks_irq; coal->tx_max_coalesced_frames_irq = bp->tx_coal_bufs_irq; + coal->stats_block_coalesce_usecs = bp->stats_coal_ticks; + return 0; } @@ -63,6 +65,7 @@ static int bnxt_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { struct bnxt *bp = netdev_priv(dev); + bool update_stats = false; int rc = 0; bp->rx_coal_ticks = coal->rx_coalesce_usecs; @@ -76,8 +79,26 @@ static int bnxt_set_coalesce(struct net_device *dev, bp->tx_coal_ticks_irq = coal->tx_coalesce_usecs_irq; bp->tx_coal_bufs_irq = coal->tx_max_coalesced_frames_irq; - if (netif_running(dev)) - rc = bnxt_hwrm_set_coal(bp); + if (bp->stats_coal_ticks != coal->stats_block_coalesce_usecs) { + u32 stats_ticks = coal->stats_block_coalesce_usecs; + + stats_ticks = clamp_t(u32, stats_ticks, + BNXT_MIN_STATS_COAL_TICKS, + BNXT_MAX_STATS_COAL_TICKS); + stats_ticks = rounddown(stats_ticks, BNXT_MIN_STATS_COAL_TICKS); + bp->stats_coal_ticks = stats_ticks; + update_stats = true; + } + + if (netif_running(dev)) { + if (update_stats) { + rc = bnxt_close_nic(bp, true, false); + if (!rc) + rc = bnxt_open_nic(bp, true, false); + } else { + rc = bnxt_hwrm_set_coal(bp); + } + } return rc; } @@ -628,7 +649,66 @@ u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause) return speed_mask; } -static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info) +#define BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, name)\ +{ \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_100MB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 100baseT_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_1GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 1000baseT_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_10GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 10000baseT_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_25GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 25000baseCR_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_40GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 40000baseCR4_Full);\ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_50GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 50000baseCR2_Full);\ + if ((fw_pause) & BNXT_LINK_PAUSE_RX) { \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + Pause); \ + if (!((fw_pause) & BNXT_LINK_PAUSE_TX)) \ + ethtool_link_ksettings_add_link_mode( \ + lk_ksettings, name, Asym_Pause);\ + } else if ((fw_pause) & BNXT_LINK_PAUSE_TX) { \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + Asym_Pause); \ + } \ +} + +#define BNXT_ETHTOOL_TO_FW_SPDS(fw_speeds, lk_ksettings, name) \ +{ \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 100baseT_Full) || \ + ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 100baseT_Half)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_100MB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 1000baseT_Full) || \ + ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 1000baseT_Half)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_1GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 10000baseT_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_10GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 25000baseCR_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_25GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 40000baseCR4_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_40GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 50000baseCR2_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_50GB; \ +} + +static void bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) { u16 fw_speeds = link_info->auto_link_speeds; u8 fw_pause = 0; @@ -636,10 +716,11 @@ static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info) if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) fw_pause = link_info->auto_pause_setting; - return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause); + BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, advertising); } -static u32 bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info) +static void bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) { u16 fw_speeds = link_info->lp_auto_link_speeds; u8 fw_pause = 0; @@ -647,16 +728,24 @@ static u32 bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info) if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) fw_pause = link_info->lp_pause; - return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause); + BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, + lp_advertising); } -static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info) +static void bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) { u16 fw_speeds = link_info->support_speeds; - u32 supported; - supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0); - return supported | SUPPORTED_Pause | SUPPORTED_Asym_Pause; + BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, 0, lk_ksettings, supported); + + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, Pause); + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + Asym_Pause); + + if (link_info->support_auto_speeds) + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + Autoneg); } u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed) @@ -683,65 +772,62 @@ u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed) } } -static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int bnxt_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *lk_ksettings) { struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; - u16 ethtool_speed; - - cmd->supported = bnxt_fw_to_ethtool_support_spds(link_info); + struct ethtool_link_settings *base = &lk_ksettings->base; + u32 ethtool_speed; - if (link_info->auto_link_speeds) - cmd->supported |= SUPPORTED_Autoneg; + ethtool_link_ksettings_zero_link_mode(lk_ksettings, supported); + bnxt_fw_to_ethtool_support_spds(link_info, lk_ksettings); + ethtool_link_ksettings_zero_link_mode(lk_ksettings, advertising); if (link_info->autoneg) { - cmd->advertising = - bnxt_fw_to_ethtool_advertised_spds(link_info); - cmd->advertising |= ADVERTISED_Autoneg; - cmd->autoneg = AUTONEG_ENABLE; + bnxt_fw_to_ethtool_advertised_spds(link_info, lk_ksettings); + ethtool_link_ksettings_add_link_mode(lk_ksettings, + advertising, Autoneg); + base->autoneg = AUTONEG_ENABLE; if (link_info->phy_link_status == BNXT_LINK_LINK) - cmd->lp_advertising = - bnxt_fw_to_ethtool_lp_adv(link_info); + bnxt_fw_to_ethtool_lp_adv(link_info, lk_ksettings); ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->link_speed); if (!netif_carrier_ok(dev)) - cmd->duplex = DUPLEX_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; else if (link_info->duplex & BNXT_LINK_DUPLEX_FULL) - cmd->duplex = DUPLEX_FULL; + base->duplex = DUPLEX_FULL; else - cmd->duplex = DUPLEX_HALF; + base->duplex = DUPLEX_HALF; } else { - cmd->autoneg = AUTONEG_DISABLE; - cmd->advertising = 0; + base->autoneg = AUTONEG_DISABLE; ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->req_link_speed); - cmd->duplex = DUPLEX_HALF; + base->duplex = DUPLEX_HALF; if (link_info->req_duplex == BNXT_LINK_DUPLEX_FULL) - cmd->duplex = DUPLEX_FULL; + base->duplex = DUPLEX_FULL; } - ethtool_cmd_speed_set(cmd, ethtool_speed); + base->speed = ethtool_speed; - cmd->port = PORT_NONE; + base->port = PORT_NONE; if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP) { - cmd->port = PORT_TP; - cmd->supported |= SUPPORTED_TP; - cmd->advertising |= ADVERTISED_TP; + base->port = PORT_TP; + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + TP); + ethtool_link_ksettings_add_link_mode(lk_ksettings, advertising, + TP); } else { - cmd->supported |= SUPPORTED_FIBRE; - cmd->advertising |= ADVERTISED_FIBRE; + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + FIBRE); + ethtool_link_ksettings_add_link_mode(lk_ksettings, advertising, + FIBRE); if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_DAC) - cmd->port = PORT_DA; + base->port = PORT_DA; else if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_FIBRE) - cmd->port = PORT_FIBRE; + base->port = PORT_FIBRE; } - - if (link_info->transceiver == - PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_INTERNAL) - cmd->transceiver = XCVR_INTERNAL; - else - cmd->transceiver = XCVR_EXTERNAL; - cmd->phy_address = link_info->phy_addr; + base->phy_address = link_info->phy_addr; return 0; } @@ -815,37 +901,25 @@ u16 bnxt_get_fw_auto_link_speeds(u32 advertising) return fw_speed_mask; } -static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int bnxt_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *lk_ksettings) { - int rc = 0; struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; + const struct ethtool_link_settings *base = &lk_ksettings->base; u32 speed, fw_advertising = 0; bool set_pause = false; + int rc = 0; - if (BNXT_VF(bp)) - return rc; - - if (cmd->autoneg == AUTONEG_ENABLE) { - u32 supported_spds = bnxt_fw_to_ethtool_support_spds(link_info); + if (!BNXT_SINGLE_PF(bp)) + return -EOPNOTSUPP; - if (cmd->advertising & ~(supported_spds | ADVERTISED_Autoneg | - ADVERTISED_TP | ADVERTISED_FIBRE)) { - netdev_err(dev, "Unsupported advertising mask (adv: 0x%x)\n", - cmd->advertising); - rc = -EINVAL; - goto set_setting_exit; - } - fw_advertising = bnxt_get_fw_auto_link_speeds(cmd->advertising); - if (fw_advertising & ~link_info->support_speeds) { - netdev_err(dev, "Advertising parameters are not supported! (adv: 0x%x)\n", - cmd->advertising); - rc = -EINVAL; - goto set_setting_exit; - } + if (base->autoneg == AUTONEG_ENABLE) { + BNXT_ETHTOOL_TO_FW_SPDS(fw_advertising, lk_ksettings, + advertising); link_info->autoneg |= BNXT_AUTONEG_SPEED; if (!fw_advertising) - link_info->advertising = link_info->support_speeds; + link_info->advertising = link_info->support_auto_speeds; else link_info->advertising = fw_advertising; /* any change to autoneg will cause link change, therefore the @@ -863,16 +937,12 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) rc = -EINVAL; goto set_setting_exit; } - /* TODO: currently don't support half duplex */ - if (cmd->duplex == DUPLEX_HALF) { + if (base->duplex == DUPLEX_HALF) { netdev_err(dev, "HALF DUPLEX is not supported!\n"); rc = -EINVAL; goto set_setting_exit; } - /* If received a request for an unknown duplex, assume full*/ - if (cmd->duplex == DUPLEX_UNKNOWN) - cmd->duplex = DUPLEX_FULL; - speed = ethtool_cmd_speed(cmd); + speed = base->speed; fw_speed = bnxt_get_fw_speed(dev, speed); if (!fw_speed) { rc = -EINVAL; @@ -911,8 +981,8 @@ static int bnxt_set_pauseparam(struct net_device *dev, struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; - if (BNXT_VF(bp)) - return rc; + if (!BNXT_SINGLE_PF(bp)) + return -EOPNOTSUPP; if (epause->autoneg) { if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) @@ -1010,6 +1080,8 @@ static int bnxt_firmware_reset(struct net_device *dev, case BNX_DIR_TYPE_APE_FW: case BNX_DIR_TYPE_APE_PATCH: req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT; + /* Self-reset APE upon next PCIe reset: */ + req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST; break; case BNX_DIR_TYPE_KONG_FW: case BNX_DIR_TYPE_KONG_PATCH: @@ -1043,9 +1115,27 @@ static int bnxt_flash_firmware(struct net_device *dev, case BNX_DIR_TYPE_BOOTCODE_2: code_type = CODE_BOOT; break; + case BNX_DIR_TYPE_CHIMP_PATCH: + code_type = CODE_CHIMP_PATCH; + break; case BNX_DIR_TYPE_APE_FW: code_type = CODE_MCTP_PASSTHRU; break; + case BNX_DIR_TYPE_APE_PATCH: + code_type = CODE_APE_PATCH; + break; + case BNX_DIR_TYPE_KONG_FW: + code_type = CODE_KONG_FW; + break; + case BNX_DIR_TYPE_KONG_PATCH: + code_type = CODE_KONG_PATCH; + break; + case BNX_DIR_TYPE_BONO_FW: + code_type = CODE_BONO_FW; + break; + case BNX_DIR_TYPE_BONO_PATCH: + code_type = CODE_BONO_PATCH; + break; default: netdev_err(dev, "Unsupported directory entry type: %u\n", dir_type); @@ -1100,6 +1190,8 @@ static bool bnxt_dir_type_is_ape_bin_format(u16 dir_type) case BNX_DIR_TYPE_APE_PATCH: case BNX_DIR_TYPE_KONG_FW: case BNX_DIR_TYPE_KONG_PATCH: + case BNX_DIR_TYPE_BONO_FW: + case BNX_DIR_TYPE_BONO_PATCH: return true; } @@ -1137,7 +1229,8 @@ static int bnxt_flash_firmware_from_file(struct net_device *dev, const struct firmware *fw; int rc; - if (bnxt_dir_type_is_executable(dir_type) == false) + if (dir_type != BNX_DIR_TYPE_UPDATE && + bnxt_dir_type_is_executable(dir_type) == false) return -EINVAL; rc = request_firmware(&fw, filename, &dev->dev); @@ -1433,8 +1526,8 @@ static int bnxt_set_eee(struct net_device *dev, struct ethtool_eee *edata) _bnxt_fw_to_ethtool_adv_spds(link_info->advertising, 0); int rc = 0; - if (BNXT_VF(bp)) - return 0; + if (!BNXT_SINGLE_PF(bp)) + return -EOPNOTSUPP; if (!(bp->flags & BNXT_FLAG_EEE_CAP)) return -EOPNOTSUPP; @@ -1618,8 +1711,8 @@ static int bnxt_get_module_eeprom(struct net_device *dev, } const struct ethtool_ops bnxt_ethtool_ops = { - .get_settings = bnxt_get_settings, - .set_settings = bnxt_set_settings, + .get_link_ksettings = bnxt_get_link_ksettings, + .set_link_ksettings = bnxt_set_link_ksettings, .get_pauseparam = bnxt_get_pauseparam, .set_pauseparam = bnxt_set_pauseparam, .get_drvinfo = bnxt_get_drvinfo, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h index 461675caaacd..82bf44ab811b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h @@ -70,6 +70,7 @@ enum SUPPORTED_CODE { CODE_KONG_PATCH, /* 18 - KONG Patch firmware */ CODE_BONO_FW, /* 19 - BONO firmware */ CODE_BONO_PATCH, /* 20 - BONO Patch firmware */ + CODE_CHIMP_PATCH, /* 21 - ChiMP Patch firmware */ MAX_CODE_TYPE, }; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h index 05e3c49a7677..517567f6d651 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h @@ -105,6 +105,7 @@ struct hwrm_async_event_cmpl { #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED (0x4UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_NOT_ALLOWED (0x5UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE (0x6UL << 0) + #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE (0x7UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_UNLOAD (0x10UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_LOAD (0x11UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD (0x20UL << 0) @@ -484,12 +485,12 @@ struct hwrm_async_event_cmpl_hwrm_error { #define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA1_TIMESTAMP 0x1UL }; -/* HW Resource Manager Specification 1.2.2 */ +/* HW Resource Manager Specification 1.3.0 */ #define HWRM_VERSION_MAJOR 1 -#define HWRM_VERSION_MINOR 2 -#define HWRM_VERSION_UPDATE 2 +#define HWRM_VERSION_MINOR 3 +#define HWRM_VERSION_UPDATE 0 -#define HWRM_VERSION_STR "1.2.2" +#define HWRM_VERSION_STR "1.3.0" /* * Following is the signature for HWRM message field that indicates not * applicable (All F's). Need to cast it the size of the field if needed. @@ -611,6 +612,9 @@ struct cmd_nums { #define HWRM_FWD_RESP (0xd2UL) #define HWRM_FWD_ASYNC_EVENT_CMPL (0xd3UL) #define HWRM_TEMP_MONITOR_QUERY (0xe0UL) + #define HWRM_WOL_FILTER_ALLOC (0xf0UL) + #define HWRM_WOL_FILTER_FREE (0xf1UL) + #define HWRM_WOL_FILTER_QCFG (0xf2UL) #define HWRM_DBG_READ_DIRECT (0xff10UL) #define HWRM_DBG_READ_INDIRECT (0xff11UL) #define HWRM_DBG_WRITE_DIRECT (0xff12UL) @@ -1020,6 +1024,10 @@ struct hwrm_func_qcaps_output { #define FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED 0x1UL #define FUNC_QCAPS_RESP_FLAGS_GLOBAL_MSIX_AUTOMASKING 0x2UL #define FUNC_QCAPS_RESP_FLAGS_PTP_SUPPORTED 0x4UL + #define FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED 0x8UL + #define FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED 0x10UL + #define FUNC_QCAPS_RESP_FLAGS_WOL_MAGICPKT_SUPPORTED 0x20UL + #define FUNC_QCAPS_RESP_FLAGS_WOL_BMP_SUPPORTED 0x40UL u8 mac_address[6]; __le16 max_rsscos_ctx; __le16 max_cmpl_rings; @@ -1066,8 +1074,9 @@ struct hwrm_func_qcfg_output { __le16 fid; __le16 port_id; __le16 vlan; - u8 unused_0; - u8 unused_1; + __le16 flags; + #define FUNC_QCFG_RESP_FLAGS_OOB_WOL_MAGICPKT_ENABLED 0x1UL + #define FUNC_QCFG_RESP_FLAGS_OOB_WOL_BMP_ENABLED 0x2UL u8 mac_address[6]; __le16 pci_id; __le16 alloc_rsscos_ctx; @@ -1086,23 +1095,23 @@ struct hwrm_func_qcfg_output { #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5 (0x3UL << 0) #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0 (0x4UL << 0) #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_UNKNOWN (0xffUL << 0) - u8 unused_2; + u8 unused_0; __le16 dflt_vnic_id; - u8 unused_3; - u8 unused_4; + u8 unused_1; + u8 unused_2; __le32 min_bw; __le32 max_bw; u8 evb_mode; #define FUNC_QCFG_RESP_EVB_MODE_NO_EVB (0x0UL << 0) #define FUNC_QCFG_RESP_EVB_MODE_VEB (0x1UL << 0) #define FUNC_QCFG_RESP_EVB_MODE_VEPA (0x2UL << 0) - u8 unused_5; - __le16 unused_6; + u8 unused_3; + __le16 unused_4; __le32 alloc_mcast_filters; __le32 alloc_hw_ring_grps; + u8 unused_5; + u8 unused_6; u8 unused_7; - u8 unused_8; - u8 unused_9; u8 valid; }; @@ -1410,8 +1419,8 @@ struct hwrm_func_buf_rgtr_input { #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4K (0xcUL << 0) #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_8K (0xdUL << 0) #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_64K (0x10UL << 0) - #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_2M (0x16UL << 0) - #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4M (0x17UL << 0) + #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_2M (0x15UL << 0) + #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4M (0x16UL << 0) #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_1G (0x1eUL << 0) __le16 req_buf_len; __le16 resp_buf_len; @@ -1499,6 +1508,12 @@ struct hwrm_port_phy_cfg_input { #define PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE 0x20UL #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE 0x40UL #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE 0x80UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE 0x100UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE 0x200UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE 0x400UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE 0x800UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE 0x1000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE 0x2000UL __le32 enables; #define PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE 0x1UL #define PORT_PHY_CFG_REQ_ENABLES_AUTO_DUPLEX 0x2UL @@ -1815,13 +1830,22 @@ struct hwrm_port_phy_qcfg_output { #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP (0xcUL << 24) #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFPPLUS (0xdUL << 24) #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP28 (0x11UL << 24) - __le32 unused_1; + __le16 fec_cfg; + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED 0x1UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED 0x2UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED 0x4UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED 0x8UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED 0x10UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED 0x20UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED 0x40UL + u8 unused_1; + u8 unused_2; char phy_vendor_name[16]; char phy_vendor_partnumber[16]; - __le32 unused_2; - u8 unused_3; + __le32 unused_3; u8 unused_4; u8 unused_5; + u8 unused_6; u8 valid; }; @@ -1842,6 +1866,8 @@ struct hwrm_port_mac_cfg_input { #define PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_DISABLE 0x20UL #define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_ENABLE 0x40UL #define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_DISABLE 0x80UL + #define PORT_MAC_CFG_REQ_FLAGS_OOB_WOL_ENABLE 0x100UL + #define PORT_MAC_CFG_REQ_FLAGS_OOB_WOL_DISABLE 0x200UL __le32 enables; #define PORT_MAC_CFG_REQ_ENABLES_IPG 0x1UL #define PORT_MAC_CFG_REQ_ENABLES_LPBK 0x2UL @@ -2127,6 +2153,7 @@ struct hwrm_port_phy_i2c_read_output { u8 valid; }; +/* hwrm_queue_qportcfg */ /* Input (24 bytes) */ struct hwrm_queue_qportcfg_input { __le16 req_type; @@ -2382,7 +2409,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id0_pri_lvl; u8 queue_id0_bw_weight; u8 queue_id1; @@ -2392,7 +2419,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id1_pri_lvl; u8 queue_id1_bw_weight; u8 queue_id2; @@ -2402,7 +2429,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id2_pri_lvl; u8 queue_id2_bw_weight; u8 queue_id3; @@ -2412,7 +2439,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id3_pri_lvl; u8 queue_id3_bw_weight; u8 queue_id4; @@ -2422,7 +2449,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id4_pri_lvl; u8 queue_id4_bw_weight; u8 queue_id5; @@ -2432,7 +2459,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id5_pri_lvl; u8 queue_id5_bw_weight; u8 queue_id6; @@ -2442,7 +2469,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id6_pri_lvl; u8 queue_id6_bw_weight; u8 queue_id7; @@ -2452,7 +2479,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id7_pri_lvl; u8 queue_id7_bw_weight; u8 unused_1[5]; @@ -3150,7 +3177,7 @@ struct hwrm_cfa_l2_filter_cfg_output { }; /* hwrm_cfa_l2_set_rx_mask */ -/* Input (40 bytes) */ +/* Input (56 bytes) */ struct hwrm_cfa_l2_set_rx_mask_input { __le16 req_type; __le16 cmpl_ring; @@ -3165,9 +3192,15 @@ struct hwrm_cfa_l2_set_rx_mask_input { #define CFA_L2_SET_RX_MASK_REQ_MASK_BCAST 0x8UL #define CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS 0x10UL #define CFA_L2_SET_RX_MASK_REQ_MASK_OUTERMOST 0x20UL + #define CFA_L2_SET_RX_MASK_REQ_MASK_VLANONLY 0x40UL + #define CFA_L2_SET_RX_MASK_REQ_MASK_VLAN_NONVLAN 0x80UL + #define CFA_L2_SET_RX_MASK_REQ_MASK_ANYVLAN_NONVLAN 0x100UL __le64 mc_tbl_addr; __le32 num_mc_entries; __le32 unused_0; + __le64 vlan_tag_tbl_addr; + __le32 num_vlan_tags; + __le32 unused_1; }; /* Output (16 bytes) */ diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h index 40a7b0e09612..73f2249555b5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h @@ -13,6 +13,7 @@ enum bnxt_nvm_directory_type { BNX_DIR_TYPE_UNUSED = 0, BNX_DIR_TYPE_PKG_LOG = 1, + BNX_DIR_TYPE_UPDATE = 2, BNX_DIR_TYPE_CHIMP_PATCH = 3, BNX_DIR_TYPE_BOOTCODE = 4, BNX_DIR_TYPE_VPD = 5, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c index 363884dd9e8a..50d2007a2640 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c @@ -143,6 +143,9 @@ int bnxt_set_vf_vlan(struct net_device *dev, int vf_id, u16 vlan_id, u8 qos) u16 vlan_tag; int rc; + if (bp->hwrm_spec_code < 0x10201) + return -ENOTSUPP; + rc = bnxt_vf_ndo_prep(bp, vf_id); if (rc) return rc; diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 541456398dfb..76ed6df0fe53 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -450,34 +450,6 @@ static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv, genet_dma_ring_regs[r]); } -static int bcmgenet_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct bcmgenet_priv *priv = netdev_priv(dev); - - if (!netif_running(dev)) - return -EINVAL; - - if (!priv->phydev) - return -ENODEV; - - return phy_ethtool_gset(priv->phydev, cmd); -} - -static int bcmgenet_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct bcmgenet_priv *priv = netdev_priv(dev); - - if (!netif_running(dev)) - return -EINVAL; - - if (!priv->phydev) - return -ENODEV; - - return phy_ethtool_sset(priv->phydev, cmd); -} - static int bcmgenet_set_rx_csum(struct net_device *dev, netdev_features_t wanted) { @@ -941,7 +913,7 @@ static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_eee *e) e->eee_active = p->eee_active; e->tx_lpi_timer = bcmgenet_umac_readl(priv, UMAC_EEE_LPI_TIMER); - return phy_ethtool_get_eee(priv->phydev, e); + return phy_ethtool_get_eee(dev->phydev, e); } static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) @@ -958,7 +930,7 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) if (!p->eee_enabled) { bcmgenet_eee_enable_set(dev, false); } else { - ret = phy_init_eee(priv->phydev, 0); + ret = phy_init_eee(dev->phydev, 0); if (ret) { netif_err(priv, hw, dev, "EEE initialization failed\n"); return ret; @@ -968,14 +940,12 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) bcmgenet_eee_enable_set(dev, true); } - return phy_ethtool_set_eee(priv->phydev, e); + return phy_ethtool_set_eee(dev->phydev, e); } static int bcmgenet_nway_reset(struct net_device *dev) { - struct bcmgenet_priv *priv = netdev_priv(dev); - - return genphy_restart_aneg(priv->phydev); + return genphy_restart_aneg(dev->phydev); } /* standard ethtool support functions. */ @@ -983,8 +953,6 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { .get_strings = bcmgenet_get_strings, .get_sset_count = bcmgenet_get_sset_count, .get_ethtool_stats = bcmgenet_get_ethtool_stats, - .get_settings = bcmgenet_get_settings, - .set_settings = bcmgenet_set_settings, .get_drvinfo = bcmgenet_get_drvinfo, .get_link = ethtool_op_get_link, .get_msglevel = bcmgenet_get_msglevel, @@ -996,18 +964,21 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { .nway_reset = bcmgenet_nway_reset, .get_coalesce = bcmgenet_get_coalesce, .set_coalesce = bcmgenet_set_coalesce, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* Power down the unimac, based on mode. */ static int bcmgenet_power_down(struct bcmgenet_priv *priv, enum bcmgenet_power_mode mode) { + struct net_device *ndev = priv->dev; int ret = 0; u32 reg; switch (mode) { case GENET_POWER_CABLE_SENSE: - phy_detach(priv->phydev); + phy_detach(ndev->phydev); break; case GENET_POWER_WOL_MAGIC: @@ -1068,7 +1039,6 @@ static void bcmgenet_power_up(struct bcmgenet_priv *priv, /* ioctl handle special commands that are not present in ethtool. */ static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct bcmgenet_priv *priv = netdev_priv(dev); int val = 0; if (!netif_running(dev)) @@ -1078,10 +1048,10 @@ static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: - if (!priv->phydev) + if (!dev->phydev) val = -ENODEV; else - val = phy_mii_ioctl(priv->phydev, rq, cmd); + val = phy_mii_ioctl(dev->phydev, rq, cmd); break; default: @@ -2464,6 +2434,7 @@ static void bcmgenet_irq_task(struct work_struct *work) { struct bcmgenet_priv *priv = container_of( work, struct bcmgenet_priv, bcmgenet_irq_work); + struct net_device *ndev = priv->dev; netif_dbg(priv, intr, priv->dev, "%s\n", __func__); @@ -2476,7 +2447,7 @@ static void bcmgenet_irq_task(struct work_struct *work) /* Link UP/DOWN event */ if (priv->irq0_stat & UMAC_IRQ_LINK_EVENT) { - phy_mac_interrupt(priv->phydev, + phy_mac_interrupt(ndev->phydev, !!(priv->irq0_stat & UMAC_IRQ_LINK_UP)); priv->irq0_stat &= ~UMAC_IRQ_LINK_EVENT; } @@ -2838,7 +2809,7 @@ static void bcmgenet_netif_start(struct net_device *dev) /* Monitor link interrupts now */ bcmgenet_link_intr_enable(priv); - phy_start(priv->phydev); + phy_start(dev->phydev); } static int bcmgenet_open(struct net_device *dev) @@ -2937,7 +2908,7 @@ static void bcmgenet_netif_stop(struct net_device *dev) struct bcmgenet_priv *priv = netdev_priv(dev); netif_tx_stop_all_queues(dev); - phy_stop(priv->phydev); + phy_stop(dev->phydev); bcmgenet_intr_disable(priv); bcmgenet_disable_rx_napi(priv); bcmgenet_disable_tx_napi(priv); @@ -2963,7 +2934,7 @@ static int bcmgenet_close(struct net_device *dev) bcmgenet_netif_stop(dev); /* Really kill the PHY state machine and disconnect from it */ - phy_disconnect(priv->phydev); + phy_disconnect(dev->phydev); /* Disable MAC receive */ umac_enable_set(priv, CMD_RX_EN, false); @@ -3522,7 +3493,7 @@ static int bcmgenet_suspend(struct device *d) bcmgenet_netif_stop(dev); - phy_suspend(priv->phydev); + phy_suspend(dev->phydev); netif_device_detach(dev); @@ -3586,7 +3557,7 @@ static int bcmgenet_resume(struct device *d) if (priv->wolopts) clk_disable_unprepare(priv->clk_wol); - phy_init_hw(priv->phydev); + phy_init_hw(dev->phydev); /* Speed settings must be restored */ bcmgenet_mii_config(priv->dev); @@ -3619,7 +3590,7 @@ static int bcmgenet_resume(struct device *d) netif_device_attach(dev); - phy_resume(priv->phydev); + phy_resume(dev->phydev); if (priv->eee.eee_enabled) bcmgenet_eee_enable_set(dev, true); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index 1e2dc34d331a..0f0868c56f05 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -597,7 +597,6 @@ struct bcmgenet_priv { /* MDIO bus variables */ wait_queue_head_t wq; - struct phy_device *phydev; bool internal_phy; struct device_node *phy_dn; struct device_node *mdio_dn; diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 457c3bc8cfff..e907acd81da9 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -86,7 +86,7 @@ static int bcmgenet_mii_write(struct mii_bus *bus, int phy_id, void bcmgenet_mii_setup(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; u32 reg, cmd_bits = 0; bool status_changed = false; @@ -183,9 +183,9 @@ void bcmgenet_mii_reset(struct net_device *dev) if (GENET_IS_V4(priv)) return; - if (priv->phydev) { - phy_init_hw(priv->phydev); - phy_start_aneg(priv->phydev); + if (dev->phydev) { + phy_init_hw(dev->phydev); + phy_start_aneg(dev->phydev); } } @@ -236,6 +236,7 @@ static void bcmgenet_internal_phy_setup(struct net_device *dev) static void bcmgenet_moca_phy_setup(struct bcmgenet_priv *priv) { + struct net_device *ndev = priv->dev; u32 reg; /* Speed settings are set in bcmgenet_mii_setup() */ @@ -244,14 +245,14 @@ static void bcmgenet_moca_phy_setup(struct bcmgenet_priv *priv) bcmgenet_sys_writel(priv, reg, SYS_PORT_CTRL); if (priv->hw_params->flags & GENET_HAS_MOCA_LINK_DET) - fixed_phy_set_link_update(priv->phydev, + fixed_phy_set_link_update(ndev->phydev, bcmgenet_fixed_phy_link_update); } int bcmgenet_mii_config(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; struct device *kdev = &priv->pdev->dev; const char *phy_name = NULL; u32 id_mode_dis = 0; @@ -302,7 +303,7 @@ int bcmgenet_mii_config(struct net_device *dev) * capabilities, use that knowledge to also configure the * Reverse MII interface correctly. */ - if ((priv->phydev->supported & PHY_BASIC_FEATURES) == + if ((phydev->supported & PHY_BASIC_FEATURES) == PHY_BASIC_FEATURES) port_ctrl = PORT_MODE_EXT_RVMII_25; else @@ -371,7 +372,7 @@ int bcmgenet_mii_probe(struct net_device *dev) return -ENODEV; } } else { - phydev = priv->phydev; + phydev = dev->phydev; phydev->dev_flags = phy_flags; ret = phy_connect_direct(dev, phydev, bcmgenet_mii_setup, @@ -382,8 +383,6 @@ int bcmgenet_mii_probe(struct net_device *dev) } } - priv->phydev = phydev; - /* Configure port multiplexer based on what the probed PHY device since * reading the 'max-speed' property determines the maximum supported * PHY speed which is needed for bcmgenet_mii_config() to configure @@ -391,7 +390,7 @@ int bcmgenet_mii_probe(struct net_device *dev) */ ret = bcmgenet_mii_config(dev); if (ret) { - phy_disconnect(priv->phydev); + phy_disconnect(phydev); return ret; } @@ -401,7 +400,7 @@ int bcmgenet_mii_probe(struct net_device *dev) * Ethernet MAC ISRs */ if (priv->internal_phy) - priv->phydev->irq = PHY_IGNORE_INTERRUPT; + phydev->irq = PHY_IGNORE_INTERRUPT; return 0; } @@ -606,7 +605,6 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv) } - priv->phydev = phydev; priv->phy_interface = pd->phy_interface; return 0; diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index cb07d95e3dd9..89c0cfa9719f 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -304,7 +304,7 @@ static void macb_set_tx_clk(struct clk *clk, int speed, struct net_device *dev) static void macb_handle_link_change(struct net_device *dev) { struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; int status_change = 0; @@ -414,7 +414,6 @@ static int macb_mii_probe(struct net_device *dev) bp->link = 0; bp->speed = 0; bp->duplex = -1; - bp->phy_dev = phydev; return 0; } @@ -1886,7 +1885,7 @@ static int macb_open(struct net_device *dev) netif_carrier_off(dev); /* if the phy is not yet register, retry later*/ - if (!bp->phy_dev) + if (!dev->phydev) return -EAGAIN; /* RX buffers initialization */ @@ -1905,7 +1904,7 @@ static int macb_open(struct net_device *dev) macb_init_hw(bp); /* schedule a link state check */ - phy_start(bp->phy_dev); + phy_start(dev->phydev); netif_tx_start_all_queues(dev); @@ -1920,8 +1919,8 @@ static int macb_close(struct net_device *dev) netif_tx_stop_all_queues(dev); napi_disable(&bp->napi); - if (bp->phy_dev) - phy_stop(bp->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); spin_lock_irqsave(&bp->lock, flags); macb_reset_hw(bp); @@ -2092,28 +2091,6 @@ static struct net_device_stats *macb_get_stats(struct net_device *dev) return nstat; } -static int macb_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int macb_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static int macb_get_regs_len(struct net_device *netdev) { return MACB_GREGS_NBR * sizeof(u32); @@ -2186,19 +2163,17 @@ static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) } static const struct ethtool_ops macb_ethtool_ops = { - .get_settings = macb_get_settings, - .set_settings = macb_set_settings, .get_regs_len = macb_get_regs_len, .get_regs = macb_get_regs, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, .get_wol = macb_get_wol, .set_wol = macb_set_wol, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct ethtool_ops gem_ethtool_ops = { - .get_settings = macb_get_settings, - .set_settings = macb_set_settings, .get_regs_len = macb_get_regs_len, .get_regs = macb_get_regs, .get_link = ethtool_op_get_link, @@ -2206,12 +2181,13 @@ static const struct ethtool_ops gem_ethtool_ops = { .get_ethtool_stats = gem_get_ethtool_stats, .get_strings = gem_get_ethtool_strings, .get_sset_count = gem_get_sset_count, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int macb_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -2570,7 +2546,7 @@ static int at91ether_open(struct net_device *dev) MACB_BIT(HRESP)); /* schedule a link state check */ - phy_start(lp->phy_dev); + phy_start(dev->phydev); netif_start_queue(dev); @@ -3010,7 +2986,7 @@ static int macb_probe(struct platform_device *pdev) if (err) goto err_out_free_netdev; - phydev = bp->phy_dev; + phydev = dev->phydev; netif_carrier_off(dev); @@ -3029,7 +3005,7 @@ static int macb_probe(struct platform_device *pdev) return 0; err_out_unregister_mdio: - phy_disconnect(bp->phy_dev); + phy_disconnect(dev->phydev); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); @@ -3057,8 +3033,8 @@ static int macb_remove(struct platform_device *pdev) if (dev) { bp = netdev_priv(dev); - if (bp->phy_dev) - phy_disconnect(bp->phy_dev); + if (dev->phydev) + phy_disconnect(dev->phydev); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 8a13824ef802..36893d8958d4 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -823,7 +823,6 @@ struct macb { struct macb_or_gem_ops macbgem_ops; struct mii_bus *mii_bus; - struct phy_device *phy_dev; int link; int speed; int duplex; diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c index 8ad7425f89bf..c03d37016a48 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c @@ -19,26 +19,16 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" int lio_cn6xxx_soft_reset(struct octeon_device *oct) { @@ -74,9 +64,9 @@ void lio_cn6xxx_enable_error_reporting(struct octeon_device *oct) u32 val; pci_read_config_dword(oct->pci_dev, CN6XXX_PCIE_DEVCTL, &val); - if (val & 0x000f0000) { + if (val & 0x000c0000) { dev_err(&oct->pci_dev->dev, "PCI-E Link error detected: 0x%08x\n", - val & 0x000f0000); + val & 0x000c0000); } val |= 0xf; /* Enable Link error reporting */ @@ -229,7 +219,7 @@ void lio_cn6xxx_setup_global_output_regs(struct octeon_device *oct) /* / Select Packet count instead of bytes for SLI_PKTi_CNTS[CNT] */ octeon_write_csr(oct, CN6XXX_SLI_PKT_OUT_BMODE, 0); - /* / Select ES,RO,NS setting from register for Output Queue Packet + /* Select ES, RO, NS setting from register for Output Queue Packet * Address */ octeon_write_csr(oct, CN6XXX_SLI_PKT_DPADDR, 0xFFFFFFFF); @@ -367,7 +357,8 @@ void lio_cn6xxx_enable_io_queues(struct octeon_device *oct) void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) { - u32 mask, i, loop = HZ; + int i; + u32 mask, loop = HZ; u32 d32; /* Reset the Enable bits for Input Queues. */ @@ -376,7 +367,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) octeon_write_csr(oct, CN6XXX_SLI_PKT_INSTR_ENB, mask); /* Wait until hardware indicates that the queues are out of reset. */ - mask = oct->io_qmask.iq; + mask = (u32)oct->io_qmask.iq; d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_IQ); while (((d32 & mask) != mask) && loop--) { d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_IQ); @@ -384,8 +375,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) } /* Reset the doorbell register for each Input queue. */ - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; octeon_write_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i), 0xFFFFFFFF); d32 = octeon_read_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i)); @@ -398,7 +389,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) /* Wait until hardware indicates that the queues are out of reset. */ loop = HZ; - mask = oct->io_qmask.oq; + mask = (u32)oct->io_qmask.oq; d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_OQ); while (((d32 & mask) != mask) && loop--) { d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_OQ); @@ -408,8 +399,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) /* Reset the doorbell register for each Output queue. */ /* for (i = 0; i < oct->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; octeon_write_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i), 0xFFFFFFFF); d32 = octeon_read_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i)); @@ -429,16 +420,16 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) void lio_cn6xxx_reinit_regs(struct octeon_device *oct) { - u32 i; + int i; - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; oct->fn_list.setup_iq_regs(oct, i); } - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; oct->fn_list.setup_oq_regs(oct, i); } @@ -450,8 +441,8 @@ void lio_cn6xxx_reinit_regs(struct octeon_device *oct) oct->fn_list.enable_io_queues(oct); /* for (i = 0; i < oct->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; writel(oct->droq[i]->max_count, oct->droq[i]->pkts_credit_reg); } @@ -495,8 +486,7 @@ u32 lio_cn6xxx_bar1_idx_read(struct octeon_device *oct, u32 idx) } u32 -lio_cn6xxx_update_read_index(struct octeon_device *oct __attribute__((unused)), - struct octeon_instr_queue *iq) +lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq) { u32 new_idx = readl(iq->inst_cnt_reg); @@ -547,17 +537,18 @@ static void lio_cn6xxx_get_pcie_qlmport(struct octeon_device *oct) dev_dbg(&oct->pci_dev->dev, "Using PCIE Port %d\n", oct->pcie_port); } -void +static void lio_cn6xxx_process_pcie_error_intr(struct octeon_device *oct, u64 intr64) { dev_err(&oct->pci_dev->dev, "Error Intr: 0x%016llx\n", CVM_CAST64(intr64)); } -int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) +static int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) { struct octeon_droq *droq; - u32 oq_no, pkt_count, droq_time_mask, droq_mask, droq_int_enb; + int oq_no; + u32 pkt_count, droq_time_mask, droq_mask, droq_int_enb; u32 droq_cnt_enb, droq_cnt_mask; droq_cnt_enb = octeon_read_csr(oct, CN6XXX_SLI_PKT_CNT_INT_ENB); @@ -573,12 +564,12 @@ int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) oct->droq_intr = 0; /* for (oq_no = 0; oq_no < oct->num_oqs; oq_no++) { */ - for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES; oq_no++) { - if (!(droq_mask & (1 << oq_no))) + for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct); oq_no++) { + if (!(droq_mask & (1ULL << oq_no))) continue; droq = oct->droq[oq_no]; - pkt_count = octeon_droq_check_hw_for_pkts(oct, droq); + pkt_count = octeon_droq_check_hw_for_pkts(droq); if (pkt_count) { oct->droq_intr |= (1ULL << oq_no); if (droq->ops.poll_mode) { diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h index f77918779355..28c47224221a 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h @@ -82,8 +82,6 @@ void lio_cn6xxx_setup_iq_regs(struct octeon_device *oct, u32 iq_no); void lio_cn6xxx_setup_oq_regs(struct octeon_device *oct, u32 oq_no); void lio_cn6xxx_enable_io_queues(struct octeon_device *oct); void lio_cn6xxx_disable_io_queues(struct octeon_device *oct); -void lio_cn6xxx_process_pcie_error_intr(struct octeon_device *oct, u64 intr64); -int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct); irqreturn_t lio_cn6xxx_process_interrupt_regs(void *dev); void lio_cn6xxx_reinit_regs(struct octeon_device *oct); void lio_cn6xxx_bar1_idx_setup(struct octeon_device *oct, u64 core_addr, @@ -91,8 +89,7 @@ void lio_cn6xxx_bar1_idx_setup(struct octeon_device *oct, u64 core_addr, void lio_cn6xxx_bar1_idx_write(struct octeon_device *oct, u32 idx, u32 mask); u32 lio_cn6xxx_bar1_idx_read(struct octeon_device *oct, u32 idx); u32 -lio_cn6xxx_update_read_index(struct octeon_device *oct __attribute__((unused)), - struct octeon_instr_queue *iq); +lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq); void lio_cn6xxx_enable_interrupt(void *chip); void lio_cn6xxx_disable_interrupt(void *chip); void cn6xxx_get_pcie_qlmport(struct octeon_device *oct); diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c index 8e830d0c0754..29755bc68f12 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c @@ -19,28 +19,17 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" #include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" static void lio_cn68xx_set_dpi_regs(struct octeon_device *oct) { @@ -129,7 +118,7 @@ static inline void lio_cn68xx_vendor_message_fix(struct octeon_device *oct) pci_write_config_dword(oct->pci_dev, CN6XXX_PCIE_FLTMSK, val); } -int lio_is_210nv(struct octeon_device *oct) +static int lio_is_210nv(struct octeon_device *oct) { u64 mio_qlm4_cfg = lio_pci_readq(oct, CN6XXX_MIO_QLM4_CFG); diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h index d4e1c9fb0bf2..ea7bdcce6044 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h @@ -28,6 +28,5 @@ #define __CN68XX_DEVICE_H__ int lio_setup_cn68xx_octeon_device(struct octeon_device *oct); -int lio_is_210nv(struct octeon_device *oct); #endif diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h b/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h index 38cddbd107b6..d45a0f4aaf1f 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h @@ -29,7 +29,6 @@ #ifndef __CN68XX_REGS_H__ #define __CN68XX_REGS_H__ -#include "cn66xx_regs.h" /*###################### REQUEST QUEUE #########################*/ diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 245c063ed4db..289eb8907922 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -19,13 +19,9 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> #include <linux/netdevice.h> #include <linux/net_tstamp.h> -#include <linux/ethtool.h> -#include <linux/dma-mapping.h> #include <linux/pci.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" @@ -36,9 +32,8 @@ #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" + +static int octnet_get_link_stats(struct net_device *netdev); struct oct_mdio_cmd_context { int octeon_id; @@ -71,34 +66,126 @@ enum { INTERFACE_MODE_RXAUI, INTERFACE_MODE_QSGMII, INTERFACE_MODE_AGL, + INTERFACE_MODE_XLAUI, + INTERFACE_MODE_XFI, + INTERFACE_MODE_10G_KR, + INTERFACE_MODE_40G_KR4, + INTERFACE_MODE_MIXED, }; #define ARRAY_LENGTH(a) (sizeof(a) / sizeof((a)[0])) #define OCT_ETHTOOL_REGDUMP_LEN 4096 #define OCT_ETHTOOL_REGSVER 1 +/* statistics of PF */ +static const char oct_stats_strings[][ETH_GSTRING_LEN] = { + "rx_packets", + "tx_packets", + "rx_bytes", + "tx_bytes", + "rx_errors", /*jabber_err+l2_err+frame_err */ + "tx_errors", /*fw_err_pko+fw_err_link+fw_err_drop */ + "rx_dropped", /*st->fromwire.total_rcvd - st->fromwire.fw_total_rcvd + *+st->fromwire.dmac_drop + st->fromwire.fw_err_drop + */ + "tx_dropped", + + "tx_total_sent", + "tx_total_fwd", + "tx_err_pko", + "tx_err_link", + "tx_err_drop", + + "tx_tso", + "tx_tso_packets", + "tx_tso_err", + "tx_vxlan", + + "mac_tx_total_pkts", + "mac_tx_total_bytes", + "mac_tx_mcast_pkts", + "mac_tx_bcast_pkts", + "mac_tx_ctl_packets", /*oct->link_stats.fromhost.ctl_sent */ + "mac_tx_total_collisions", + "mac_tx_one_collision", + "mac_tx_multi_collison", + "mac_tx_max_collision_fail", + "mac_tx_max_deferal_fail", + "mac_tx_fifo_err", + "mac_tx_runts", + + "rx_total_rcvd", + "rx_total_fwd", + "rx_jabber_err", + "rx_l2_err", + "rx_frame_err", + "rx_err_pko", + "rx_err_link", + "rx_err_drop", + + "rx_vxlan", + "rx_vxlan_err", + + "rx_lro_pkts", + "rx_lro_bytes", + "rx_total_lro", + + "rx_lro_aborts", + "rx_lro_aborts_port", + "rx_lro_aborts_seq", + "rx_lro_aborts_tsval", + "rx_lro_aborts_timer", + "rx_fwd_rate", + + "mac_rx_total_rcvd", + "mac_rx_bytes", + "mac_rx_total_bcst", + "mac_rx_total_mcst", + "mac_rx_runts", + "mac_rx_ctl_packets", + "mac_rx_fifo_err", + "mac_rx_dma_drop", + "mac_rx_fcs_err", + + "link_state_changes", +}; + +/* statistics of host tx queue */ static const char oct_iq_stats_strings[][ETH_GSTRING_LEN] = { - "Instr posted", - "Instr processed", - "Instr dropped", - "Bytes Sent", - "Sgentry_sent", - "Inst cntreg", - "Tx done", - "Tx Iq busy", - "Tx dropped", - "Tx bytes", + "packets", /*oct->instr_queue[iq_no]->stats.tx_done*/ + "bytes", /*oct->instr_queue[iq_no]->stats.tx_tot_bytes*/ + "dropped", + "iq_busy", + "sgentry_sent", + + "fw_instr_posted", + "fw_instr_processed", + "fw_instr_dropped", + "fw_bytes_sent", + + "tso", + "vxlan", + "txq_restart", }; +/* statistics of host rx queue */ static const char oct_droq_stats_strings[][ETH_GSTRING_LEN] = { - "OQ Pkts Received", - "OQ Bytes Received", - "Dropped no dispatch", - "Dropped nomem", - "Dropped toomany", - "Stack RX cnt", - "Stack RX Bytes", - "RX dropped", + "packets", /*oct->droq[oq_no]->stats.rx_pkts_received */ + "bytes", /*oct->droq[oq_no]->stats.rx_bytes_received */ + "dropped", /*oct->droq[oq_no]->stats.rx_dropped+ + *oct->droq[oq_no]->stats.dropped_nodispatch+ + *oct->droq[oq_no]->stats.dropped_toomany+ + *oct->droq[oq_no]->stats.dropped_nomem + */ + "dropped_nomem", + "dropped_toomany", + "fw_dropped", + "fw_pkts_received", + "fw_bytes_received", + "fw_dropped_nodispatch", + + "vxlan", + "buffer_alloc_failure", }; #define OCTNIC_NCMD_AUTONEG_ON 0x1 @@ -112,8 +199,9 @@ static int lio_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) linfo = &lio->linfo; - if (linfo->link.s.interface == INTERFACE_MODE_XAUI || - linfo->link.s.interface == INTERFACE_MODE_RXAUI) { + if (linfo->link.s.if_mode == INTERFACE_MODE_XAUI || + linfo->link.s.if_mode == INTERFACE_MODE_RXAUI || + linfo->link.s.if_mode == INTERFACE_MODE_XFI) { ecmd->port = PORT_FIBRE; ecmd->supported = (SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE | @@ -124,10 +212,11 @@ static int lio_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) ecmd->autoneg = AUTONEG_DISABLE; } else { - dev_err(&oct->pci_dev->dev, "Unknown link interface reported\n"); + dev_err(&oct->pci_dev->dev, "Unknown link interface reported %d\n", + linfo->link.s.if_mode); } - if (linfo->link.s.status) { + if (linfo->link.s.link_up) { ethtool_cmd_speed_set(ecmd, linfo->link.s.speed); ecmd->duplex = linfo->link.s.duplex; } else { @@ -222,23 +311,20 @@ static int octnet_gpio_access(struct net_device *netdev, int addr, int val) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int ret = 0; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_GPIO_ACCESS; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = addr; - nctrl.ncmd.s.param3 = val; + nctrl.ncmd.s.param1 = addr; + nctrl.ncmd.s.param2 = val; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 100; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Failed to configure gpio value\n"); return -EINVAL; @@ -253,20 +339,18 @@ static void octnet_mdio_resp_callback(struct octeon_device *oct, u32 status, void *buf) { - struct oct_mdio_cmd_resp *mdio_cmd_rsp; struct oct_mdio_cmd_context *mdio_cmd_ctx; struct octeon_soft_command *sc = (struct octeon_soft_command *)buf; - mdio_cmd_rsp = (struct oct_mdio_cmd_resp *)sc->virtrptr; mdio_cmd_ctx = (struct oct_mdio_cmd_context *)sc->ctxptr; oct = lio_get_device(mdio_cmd_ctx->octeon_id); if (status) { dev_err(&oct->pci_dev->dev, "MIDO instruction failed. Status: %llx\n", CVM_CAST64(status)); - ACCESS_ONCE(mdio_cmd_ctx->cond) = -1; + WRITE_ONCE(mdio_cmd_ctx->cond, -1); } else { - ACCESS_ONCE(mdio_cmd_ctx->cond) = 1; + WRITE_ONCE(mdio_cmd_ctx->cond, 1); } wake_up_interruptible(&mdio_cmd_ctx->wc); } @@ -297,15 +381,16 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) mdio_cmd_rsp = (struct oct_mdio_cmd_resp *)sc->virtrptr; mdio_cmd = (struct oct_mdio_cmd *)sc->virtdptr; - ACCESS_ONCE(mdio_cmd_ctx->cond) = 0; + WRITE_ONCE(mdio_cmd_ctx->cond, 0); mdio_cmd_ctx->octeon_id = lio_get_device_id(oct_dev); mdio_cmd->op = op; mdio_cmd->mdio_addr = loc; if (op) mdio_cmd->value1 = *value; - mdio_cmd->value2 = lio->linfo.ifidx; octeon_swap_8B_data((u64 *)mdio_cmd, sizeof(struct oct_mdio_cmd) / 8); + sc->iq_no = lio->linfo.txpciq[0].s.q_no; + octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, OPCODE_NIC_MDIO45, 0, 0, 0); @@ -317,7 +402,7 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) retval = octeon_send_soft_command(oct_dev, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { dev_err(&oct_dev->pci_dev->dev, "octnet_mdio45_access instruction failed status: %x\n", retval); @@ -335,7 +420,7 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) octeon_swap_8B_data((u64 *)(&mdio_cmd_rsp->resp), sizeof(struct oct_mdio_cmd) / 8); - if (ACCESS_ONCE(mdio_cmd_ctx->cond) == 1) { + if (READ_ONCE(mdio_cmd_ctx->cond) == 1) { if (!op) *value = mdio_cmd_rsp->resp.value1; } else { @@ -379,18 +464,16 @@ static int lio_set_phys_id(struct net_device *netdev, /* Configure Beacon values */ value = LIO68XX_LED_BEACON_CFGON; - ret = - octnet_mdio45_access(lio, 1, - LIO68XX_LED_BEACON_ADDR, - &value); + ret = octnet_mdio45_access(lio, 1, + LIO68XX_LED_BEACON_ADDR, + &value); if (ret) return ret; value = LIO68XX_LED_CTRL_CFGON; - ret = - octnet_mdio45_access(lio, 1, - LIO68XX_LED_CTRL_ADDR, - &value); + ret = octnet_mdio45_access(lio, 1, + LIO68XX_LED_CTRL_ADDR, + &value); if (ret) return ret; } else { @@ -469,7 +552,7 @@ lio_ethtool_get_ringparam(struct net_device *netdev, tx_pending = CFG_GET_NUM_TX_DESCS_NIC_IF(conf6x, lio->ifidx); } - if (lio->mtu > OCTNET_DEFAULT_FRM_SIZE) { + if (lio->mtu > OCTNET_DEFAULT_FRM_SIZE - OCTNET_FRM_HEADER_SIZE) { ering->rx_pending = 0; ering->rx_max_pending = 0; ering->rx_mini_pending = 0; @@ -503,10 +586,10 @@ static void lio_set_msglevel(struct net_device *netdev, u32 msglvl) if ((msglvl ^ lio->msg_enable) & NETIF_MSG_HW) { if (msglvl & NETIF_MSG_HW) liquidio_set_feature(netdev, - OCTNET_CMD_VERBOSE_ENABLE); + OCTNET_CMD_VERBOSE_ENABLE, 0); else liquidio_set_feature(netdev, - OCTNET_CMD_VERBOSE_DISABLE); + OCTNET_CMD_VERBOSE_DISABLE, 0); } lio->msg_enable = msglvl; @@ -518,61 +601,279 @@ lio_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) /* Notes: Not supporting any auto negotiation in these * drivers. Just report pause frame support. */ - pause->tx_pause = 1; - pause->rx_pause = 1; /* TODO: Need to support RX pause frame!!. */ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + + pause->autoneg = 0; + + pause->tx_pause = oct->tx_pause; + pause->rx_pause = oct->rx_pause; } static void lio_get_ethtool_stats(struct net_device *netdev, - struct ethtool_stats *stats, u64 *data) + struct ethtool_stats *stats __attribute__((unused)), + u64 *data) { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; + struct net_device_stats *netstats = &netdev->stats; int i = 0, j; - for (j = 0; j < MAX_OCTEON_INSTR_QUEUES; j++) { - if (!(oct_dev->io_qmask.iq & (1UL << j))) + netdev->netdev_ops->ndo_get_stats(netdev); + octnet_get_link_stats(netdev); + + /*sum of oct->droq[oq_no]->stats->rx_pkts_received */ + data[i++] = CVM_CAST64(netstats->rx_packets); + /*sum of oct->instr_queue[iq_no]->stats.tx_done */ + data[i++] = CVM_CAST64(netstats->tx_packets); + /*sum of oct->droq[oq_no]->stats->rx_bytes_received */ + data[i++] = CVM_CAST64(netstats->rx_bytes); + /*sum of oct->instr_queue[iq_no]->stats.tx_tot_bytes */ + data[i++] = CVM_CAST64(netstats->tx_bytes); + data[i++] = CVM_CAST64(netstats->rx_errors); + data[i++] = CVM_CAST64(netstats->tx_errors); + /*sum of oct->droq[oq_no]->stats->rx_dropped + + *oct->droq[oq_no]->stats->dropped_nodispatch + + *oct->droq[oq_no]->stats->dropped_toomany + + *oct->droq[oq_no]->stats->dropped_nomem + */ + data[i++] = CVM_CAST64(netstats->rx_dropped); + /*sum of oct->instr_queue[iq_no]->stats.tx_dropped */ + data[i++] = CVM_CAST64(netstats->tx_dropped); + + /*data[i++] = CVM_CAST64(stats->multicast); */ + /*data[i++] = CVM_CAST64(stats->collisions); */ + + /* firmware tx stats */ + /*per_core_stats[cvmx_get_core_num()].link_stats[mdata->from_ifidx]. + *fromhost.fw_total_sent + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_total_sent); + /*per_core_stats[i].link_stats[port].fromwire.fw_total_fwd */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_total_fwd); + /*per_core_stats[j].link_stats[i].fromhost.fw_err_pko */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_pko); + /*per_core_stats[j].link_stats[i].fromhost.fw_err_link */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_link); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_err_drop + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_drop); + + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost.fw_tso */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tso); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_tso_fwd + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tso_fwd); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_err_tso + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_tso); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_tx_vxlan + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tx_vxlan); + + /* mac tx statistics */ + /*CVMX_BGXX_CMRX_TX_STAT5 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_pkts_sent); + /*CVMX_BGXX_CMRX_TX_STAT4 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_bytes_sent); + /*CVMX_BGXX_CMRX_TX_STAT15 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.mcast_pkts_sent); + /*CVMX_BGXX_CMRX_TX_STAT14 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.bcast_pkts_sent); + /*CVMX_BGXX_CMRX_TX_STAT17 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.ctl_sent); + /*CVMX_BGXX_CMRX_TX_STAT0 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_collisions); + /*CVMX_BGXX_CMRX_TX_STAT3 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.one_collision_sent); + /*CVMX_BGXX_CMRX_TX_STAT2 */ + data[i++] = + CVM_CAST64(oct_dev->link_stats.fromhost.multi_collision_sent); + /*CVMX_BGXX_CMRX_TX_STAT0 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.max_collision_fail); + /*CVMX_BGXX_CMRX_TX_STAT1 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.max_deferral_fail); + /*CVMX_BGXX_CMRX_TX_STAT16 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fifo_err); + /*CVMX_BGXX_CMRX_TX_STAT6 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.runts); + + /* RX firmware stats */ + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_total_rcvd + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_rcvd); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_total_fwd + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_fwd); + /*per_core_stats[core_id].link_stats[ifidx].fromwire.jabber_err */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.jabber_err); + /*per_core_stats[core_id].link_stats[ifidx].fromwire.l2_err */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.l2_err); + /*per_core_stats[core_id].link_stats[ifidx].fromwire.frame_err */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.frame_err); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_err_pko + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_pko); + /*per_core_stats[j].link_stats[i].fromwire.fw_err_link */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_link); + /*per_core_stats[cvmx_get_core_num()].link_stats[lro_ctx->ifidx]. + *fromwire.fw_err_drop + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_drop); + + /*per_core_stats[cvmx_get_core_num()].link_stats[lro_ctx->ifidx]. + *fromwire.fw_rx_vxlan + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_rx_vxlan); + /*per_core_stats[cvmx_get_core_num()].link_stats[lro_ctx->ifidx]. + *fromwire.fw_rx_vxlan_err + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_rx_vxlan_err); + + /* LRO */ + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_pkts + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_pkts); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_octs + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_octs); + /*per_core_stats[j].link_stats[i].fromwire.fw_total_lro */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_lro); + /*per_core_stats[j].link_stats[i].fromwire.fw_lro_aborts */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_port + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_port); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_seq + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_seq); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_tsval + */ + data[i++] = + CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_tsval); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_timer + */ + /* intrmod: packet forward rate */ + data[i++] = + CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_timer); + /*per_core_stats[j].link_stats[i].fromwire.fw_lro_aborts */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fwd_rate); + + /* mac: link-level stats */ + /*CVMX_BGXX_CMRX_RX_STAT0 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.total_rcvd); + /*CVMX_BGXX_CMRX_RX_STAT1 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.bytes_rcvd); + /*CVMX_PKI_STATX_STAT5 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.total_bcst); + /*CVMX_PKI_STATX_STAT5 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.total_mcst); + /*wqe->word2.err_code or wqe->word2.err_level */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.runts); + /*CVMX_BGXX_CMRX_RX_STAT2 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.ctl_rcvd); + /*CVMX_BGXX_CMRX_RX_STAT6 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fifo_err); + /*CVMX_BGXX_CMRX_RX_STAT4 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.dmac_drop); + /*wqe->word2.err_code or wqe->word2.err_level */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fcs_err); + /*lio->link_changes*/ + data[i++] = CVM_CAST64(lio->link_changes); + + /* TX -- lio_update_stats(lio); */ + for (j = 0; j < MAX_OCTEON_INSTR_QUEUES(oct_dev); j++) { + if (!(oct_dev->io_qmask.iq & (1ULL << j))) continue; + /*packets to network port*/ + /*# of packets tx to network */ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_done); + /*# of bytes tx to network */ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.instr_posted); - data[i++] = - CVM_CAST64( - oct_dev->instr_queue[j]->stats.instr_processed); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_tot_bytes); + /*# of packets dropped */ data[i++] = - CVM_CAST64( - oct_dev->instr_queue[j]->stats.instr_dropped); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_dropped); + /*# of tx fails due to queue full */ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.bytes_sent); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_iq_busy); + /*XXX gather entries sent */ data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.sgentry_sent); + + /*instruction to firmware: data and control */ + /*# of instructions to the queue */ data[i++] = - readl(oct_dev->instr_queue[j]->inst_cnt_reg); - data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_done); - data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_iq_busy); + CVM_CAST64(oct_dev->instr_queue[j]->stats.instr_posted); + /*# of instructions processed */ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]-> + stats.instr_processed); + /*# of instructions could not be processed */ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]-> + stats.instr_dropped); + /*bytes sent through the queue */ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_dropped); + CVM_CAST64(oct_dev->instr_queue[j]->stats.bytes_sent); + + /*tso request*/ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_gso); + /*vxlan request*/ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_vxlan); + /*txq restart*/ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_tot_bytes); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_restart); } - /* for (j = 0; j < oct_dev->num_oqs; j++){ */ - for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES; j++) { - if (!(oct_dev->io_qmask.oq & (1UL << j))) + /* RX */ + /* for (j = 0; j < oct_dev->num_oqs; j++) { */ + for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES(oct_dev); j++) { + if (!(oct_dev->io_qmask.oq & (1ULL << j))) continue; - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.pkts_received); - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.bytes_received); - data[i++] = - CVM_CAST64(oct_dev->droq[j]->stats.dropped_nodispatch); - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_nomem); - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_toomany); + + /*packets send to TCP/IP network stack */ + /*# of packets to network stack */ data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_pkts_received); + /*# of bytes to network stack */ data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_bytes_received); + /*# of packets dropped */ + data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_nomem + + oct_dev->droq[j]->stats.dropped_toomany + + oct_dev->droq[j]->stats.rx_dropped); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.dropped_nomem); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.dropped_toomany); data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_dropped); + + /*control and data path*/ + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.pkts_received); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.bytes_received); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.dropped_nodispatch); + + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.rx_vxlan); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.rx_alloc_failure); } } @@ -581,26 +882,43 @@ static void lio_get_strings(struct net_device *netdev, u32 stringset, u8 *data) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; int num_iq_stats, num_oq_stats, i, j; + int num_stats; - num_iq_stats = ARRAY_SIZE(oct_iq_stats_strings); - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct_dev->io_qmask.iq & (1UL << i))) - continue; - for (j = 0; j < num_iq_stats; j++) { - sprintf(data, "IQ%d %s", i, oct_iq_stats_strings[j]); + switch (stringset) { + case ETH_SS_STATS: + num_stats = ARRAY_SIZE(oct_stats_strings); + for (j = 0; j < num_stats; j++) { + sprintf(data, "%s", oct_stats_strings[j]); data += ETH_GSTRING_LEN; } - } - num_oq_stats = ARRAY_SIZE(oct_droq_stats_strings); - /* for (i = 0; i < oct_dev->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct_dev->io_qmask.oq & (1UL << i))) - continue; - for (j = 0; j < num_oq_stats; j++) { - sprintf(data, "OQ%d %s", i, oct_droq_stats_strings[j]); - data += ETH_GSTRING_LEN; + num_iq_stats = ARRAY_SIZE(oct_iq_stats_strings); + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct_dev); i++) { + if (!(oct_dev->io_qmask.iq & (1ULL << i))) + continue; + for (j = 0; j < num_iq_stats; j++) { + sprintf(data, "tx-%d-%s", i, + oct_iq_stats_strings[j]); + data += ETH_GSTRING_LEN; + } } + + num_oq_stats = ARRAY_SIZE(oct_droq_stats_strings); + /* for (i = 0; i < oct_dev->num_oqs; i++) { */ + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct_dev); i++) { + if (!(oct_dev->io_qmask.oq & (1ULL << i))) + continue; + for (j = 0; j < num_oq_stats; j++) { + sprintf(data, "rx-%d-%s", i, + oct_droq_stats_strings[j]); + data += ETH_GSTRING_LEN; + } + } + break; + + default: + netif_info(lio, drv, lio->netdev, "Unknown Stringset !!\n"); + break; } } @@ -609,8 +927,14 @@ static int lio_get_sset_count(struct net_device *netdev, int sset) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; - return (ARRAY_SIZE(oct_iq_stats_strings) * oct_dev->num_iqs) + - (ARRAY_SIZE(oct_droq_stats_strings) * oct_dev->num_oqs); + switch (sset) { + case ETH_SS_STATS: + return (ARRAY_SIZE(oct_stats_strings) + + ARRAY_SIZE(oct_iq_stats_strings) * oct_dev->num_iqs + + ARRAY_SIZE(oct_droq_stats_strings) * oct_dev->num_oqs); + default: + return -EOPNOTSUPP; + } } static int lio_get_intr_coalesce(struct net_device *netdev, @@ -618,50 +942,49 @@ static int lio_get_intr_coalesce(struct net_device *netdev, { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; - struct octeon_cn6xxx *cn6xxx = (struct octeon_cn6xxx *)oct->chip; struct octeon_instr_queue *iq; struct oct_intrmod_cfg *intrmod_cfg; intrmod_cfg = &oct->intrmod; switch (oct->chip_id) { - /* case OCTEON_CN73XX: Todo */ - /* break; */ case OCTEON_CN68XX: - case OCTEON_CN66XX: - if (!intrmod_cfg->intrmod_enable) { + case OCTEON_CN66XX: { + struct octeon_cn6xxx *cn6xxx = + (struct octeon_cn6xxx *)oct->chip; + + if (!intrmod_cfg->rx_enable) { intr_coal->rx_coalesce_usecs = CFG_GET_OQ_INTR_TIME(cn6xxx->conf); intr_coal->rx_max_coalesced_frames = CFG_GET_OQ_INTR_PKT(cn6xxx->conf); - } else { - intr_coal->use_adaptive_rx_coalesce = - intrmod_cfg->intrmod_enable; - intr_coal->rate_sample_interval = - intrmod_cfg->intrmod_check_intrvl; - intr_coal->pkt_rate_high = - intrmod_cfg->intrmod_maxpkt_ratethr; - intr_coal->pkt_rate_low = - intrmod_cfg->intrmod_minpkt_ratethr; - intr_coal->rx_max_coalesced_frames_high = - intrmod_cfg->intrmod_maxcnt_trigger; - intr_coal->rx_coalesce_usecs_high = - intrmod_cfg->intrmod_maxtmr_trigger; - intr_coal->rx_coalesce_usecs_low = - intrmod_cfg->intrmod_mintmr_trigger; - intr_coal->rx_max_coalesced_frames_low = - intrmod_cfg->intrmod_mincnt_trigger; } - - iq = oct->instr_queue[lio->linfo.txpciq[0]]; + iq = oct->instr_queue[lio->linfo.txpciq[0].s.q_no]; intr_coal->tx_max_coalesced_frames = iq->fill_threshold; break; - + } default: netif_info(lio, drv, lio->netdev, "Unknown Chip !!\n"); return -EINVAL; } - + if (intrmod_cfg->rx_enable) { + intr_coal->use_adaptive_rx_coalesce = + intrmod_cfg->rx_enable; + intr_coal->rate_sample_interval = + intrmod_cfg->check_intrvl; + intr_coal->pkt_rate_high = + intrmod_cfg->maxpkt_ratethr; + intr_coal->pkt_rate_low = + intrmod_cfg->minpkt_ratethr; + intr_coal->rx_max_coalesced_frames_high = + intrmod_cfg->rx_maxcnt_trigger; + intr_coal->rx_coalesce_usecs_high = + intrmod_cfg->rx_maxtmr_trigger; + intr_coal->rx_coalesce_usecs_low = + intrmod_cfg->rx_mintmr_trigger; + intr_coal->rx_max_coalesced_frames_low = + intrmod_cfg->rx_mincnt_trigger; + } return 0; } @@ -681,19 +1004,20 @@ static void octnet_intrmod_callback(struct octeon_device *oct_dev, else dev_info(&oct_dev->pci_dev->dev, "Rx-Adaptive Interrupt moderation enabled:%llx\n", - oct_dev->intrmod.intrmod_enable); + oct_dev->intrmod.rx_enable); octeon_free_soft_command(oct_dev, sc); } /* Configure interrupt moderation parameters */ -static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) +static int octnet_set_intrmod_cfg(struct lio *lio, + struct oct_intrmod_cfg *intr_cfg) { struct octeon_soft_command *sc; struct oct_intrmod_cmd *cmd; struct oct_intrmod_cfg *cfg; int retval; - struct octeon_device *oct_dev = (struct octeon_device *)oct; + struct octeon_device *oct_dev = lio->oct_dev; /* Alloc soft command */ sc = (struct octeon_soft_command *) @@ -714,6 +1038,8 @@ static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) cmd->cfg = cfg; cmd->oct_dev = oct_dev; + sc->iq_no = lio->linfo.txpciq[0].s.q_no; + octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, OPCODE_NIC_INTRMOD_CFG, 0, 0, 0); @@ -722,17 +1048,171 @@ static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) sc->wait_time = 1000; retval = octeon_send_soft_command(oct_dev, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { + octeon_free_soft_command(oct_dev, sc); + return -EINVAL; + } + + return 0; +} + +static void +octnet_nic_stats_callback(struct octeon_device *oct_dev, + u32 status, void *ptr) +{ + struct octeon_soft_command *sc = (struct octeon_soft_command *)ptr; + struct oct_nic_stats_resp *resp = (struct oct_nic_stats_resp *) + sc->virtrptr; + struct oct_nic_stats_ctrl *ctrl = (struct oct_nic_stats_ctrl *) + sc->ctxptr; + struct nic_rx_stats *rsp_rstats = &resp->stats.fromwire; + struct nic_tx_stats *rsp_tstats = &resp->stats.fromhost; + + struct nic_rx_stats *rstats = &oct_dev->link_stats.fromwire; + struct nic_tx_stats *tstats = &oct_dev->link_stats.fromhost; + + if ((status != OCTEON_REQUEST_TIMEOUT) && !resp->status) { + octeon_swap_8B_data((u64 *)&resp->stats, + (sizeof(struct oct_link_stats)) >> 3); + + /* RX link-level stats */ + rstats->total_rcvd = rsp_rstats->total_rcvd; + rstats->bytes_rcvd = rsp_rstats->bytes_rcvd; + rstats->total_bcst = rsp_rstats->total_bcst; + rstats->total_mcst = rsp_rstats->total_mcst; + rstats->runts = rsp_rstats->runts; + rstats->ctl_rcvd = rsp_rstats->ctl_rcvd; + /* Accounts for over/under-run of buffers */ + rstats->fifo_err = rsp_rstats->fifo_err; + rstats->dmac_drop = rsp_rstats->dmac_drop; + rstats->fcs_err = rsp_rstats->fcs_err; + rstats->jabber_err = rsp_rstats->jabber_err; + rstats->l2_err = rsp_rstats->l2_err; + rstats->frame_err = rsp_rstats->frame_err; + + /* RX firmware stats */ + rstats->fw_total_rcvd = rsp_rstats->fw_total_rcvd; + rstats->fw_total_fwd = rsp_rstats->fw_total_fwd; + rstats->fw_err_pko = rsp_rstats->fw_err_pko; + rstats->fw_err_link = rsp_rstats->fw_err_link; + rstats->fw_err_drop = rsp_rstats->fw_err_drop; + rstats->fw_rx_vxlan = rsp_rstats->fw_rx_vxlan; + rstats->fw_rx_vxlan_err = rsp_rstats->fw_rx_vxlan_err; + + /* Number of packets that are LROed */ + rstats->fw_lro_pkts = rsp_rstats->fw_lro_pkts; + /* Number of octets that are LROed */ + rstats->fw_lro_octs = rsp_rstats->fw_lro_octs; + /* Number of LRO packets formed */ + rstats->fw_total_lro = rsp_rstats->fw_total_lro; + /* Number of times lRO of packet aborted */ + rstats->fw_lro_aborts = rsp_rstats->fw_lro_aborts; + rstats->fw_lro_aborts_port = rsp_rstats->fw_lro_aborts_port; + rstats->fw_lro_aborts_seq = rsp_rstats->fw_lro_aborts_seq; + rstats->fw_lro_aborts_tsval = rsp_rstats->fw_lro_aborts_tsval; + rstats->fw_lro_aborts_timer = rsp_rstats->fw_lro_aborts_timer; + /* intrmod: packet forward rate */ + rstats->fwd_rate = rsp_rstats->fwd_rate; + + /* TX link-level stats */ + tstats->total_pkts_sent = rsp_tstats->total_pkts_sent; + tstats->total_bytes_sent = rsp_tstats->total_bytes_sent; + tstats->mcast_pkts_sent = rsp_tstats->mcast_pkts_sent; + tstats->bcast_pkts_sent = rsp_tstats->bcast_pkts_sent; + tstats->ctl_sent = rsp_tstats->ctl_sent; + /* Packets sent after one collision*/ + tstats->one_collision_sent = rsp_tstats->one_collision_sent; + /* Packets sent after multiple collision*/ + tstats->multi_collision_sent = rsp_tstats->multi_collision_sent; + /* Packets not sent due to max collisions */ + tstats->max_collision_fail = rsp_tstats->max_collision_fail; + /* Packets not sent due to max deferrals */ + tstats->max_deferral_fail = rsp_tstats->max_deferral_fail; + /* Accounts for over/under-run of buffers */ + tstats->fifo_err = rsp_tstats->fifo_err; + tstats->runts = rsp_tstats->runts; + /* Total number of collisions detected */ + tstats->total_collisions = rsp_tstats->total_collisions; + + /* firmware stats */ + tstats->fw_total_sent = rsp_tstats->fw_total_sent; + tstats->fw_total_fwd = rsp_tstats->fw_total_fwd; + tstats->fw_err_pko = rsp_tstats->fw_err_pko; + tstats->fw_err_link = rsp_tstats->fw_err_link; + tstats->fw_err_drop = rsp_tstats->fw_err_drop; + tstats->fw_tso = rsp_tstats->fw_tso; + tstats->fw_tso_fwd = rsp_tstats->fw_tso_fwd; + tstats->fw_err_tso = rsp_tstats->fw_err_tso; + tstats->fw_tx_vxlan = rsp_tstats->fw_tx_vxlan; + + resp->status = 1; + } else { + resp->status = -1; + } + complete(&ctrl->complete); +} + +/* Configure interrupt moderation parameters */ +static int octnet_get_link_stats(struct net_device *netdev) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct_dev = lio->oct_dev; + + struct octeon_soft_command *sc; + struct oct_nic_stats_ctrl *ctrl; + struct oct_nic_stats_resp *resp; + + int retval; + + /* Alloc soft command */ + sc = (struct octeon_soft_command *) + octeon_alloc_soft_command(oct_dev, + 0, + sizeof(struct oct_nic_stats_resp), + sizeof(struct octnic_ctrl_pkt)); + + if (!sc) + return -ENOMEM; + + resp = (struct oct_nic_stats_resp *)sc->virtrptr; + memset(resp, 0, sizeof(struct oct_nic_stats_resp)); + + ctrl = (struct oct_nic_stats_ctrl *)sc->ctxptr; + memset(ctrl, 0, sizeof(struct oct_nic_stats_ctrl)); + ctrl->netdev = netdev; + init_completion(&ctrl->complete); + + sc->iq_no = lio->linfo.txpciq[0].s.q_no; + + octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, + OPCODE_NIC_PORT_STATS, 0, 0, 0); + + sc->callback = octnet_nic_stats_callback; + sc->callback_arg = sc; + sc->wait_time = 500; /*in milli seconds*/ + + retval = octeon_send_soft_command(oct_dev, sc); + if (retval == IQ_SEND_FAILED) { + octeon_free_soft_command(oct_dev, sc); + return -EINVAL; + } + + wait_for_completion_timeout(&ctrl->complete, msecs_to_jiffies(1000)); + + if (resp->status != 1) { octeon_free_soft_command(oct_dev, sc); + return -EINVAL; } + octeon_free_soft_command(oct_dev, sc); + return 0; } /* Enable/Disable auto interrupt Moderation */ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce - *intr_coal, int adaptive) + *intr_coal) { int ret = 0; struct octeon_device *oct = lio->oct_dev; @@ -740,59 +1220,73 @@ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce intrmod_cfg = &oct->intrmod; - if (adaptive) { + if (oct->intrmod.rx_enable || oct->intrmod.tx_enable) { if (intr_coal->rate_sample_interval) - intrmod_cfg->intrmod_check_intrvl = + intrmod_cfg->check_intrvl = intr_coal->rate_sample_interval; else - intrmod_cfg->intrmod_check_intrvl = + intrmod_cfg->check_intrvl = LIO_INTRMOD_CHECK_INTERVAL; if (intr_coal->pkt_rate_high) - intrmod_cfg->intrmod_maxpkt_ratethr = + intrmod_cfg->maxpkt_ratethr = intr_coal->pkt_rate_high; else - intrmod_cfg->intrmod_maxpkt_ratethr = + intrmod_cfg->maxpkt_ratethr = LIO_INTRMOD_MAXPKT_RATETHR; if (intr_coal->pkt_rate_low) - intrmod_cfg->intrmod_minpkt_ratethr = + intrmod_cfg->minpkt_ratethr = intr_coal->pkt_rate_low; else - intrmod_cfg->intrmod_minpkt_ratethr = + intrmod_cfg->minpkt_ratethr = LIO_INTRMOD_MINPKT_RATETHR; - + } + if (oct->intrmod.rx_enable) { if (intr_coal->rx_max_coalesced_frames_high) - intrmod_cfg->intrmod_maxcnt_trigger = + intrmod_cfg->rx_maxcnt_trigger = intr_coal->rx_max_coalesced_frames_high; else - intrmod_cfg->intrmod_maxcnt_trigger = - LIO_INTRMOD_MAXCNT_TRIGGER; + intrmod_cfg->rx_maxcnt_trigger = + LIO_INTRMOD_RXMAXCNT_TRIGGER; if (intr_coal->rx_coalesce_usecs_high) - intrmod_cfg->intrmod_maxtmr_trigger = + intrmod_cfg->rx_maxtmr_trigger = intr_coal->rx_coalesce_usecs_high; else - intrmod_cfg->intrmod_maxtmr_trigger = - LIO_INTRMOD_MAXTMR_TRIGGER; + intrmod_cfg->rx_maxtmr_trigger = + LIO_INTRMOD_RXMAXTMR_TRIGGER; if (intr_coal->rx_coalesce_usecs_low) - intrmod_cfg->intrmod_mintmr_trigger = + intrmod_cfg->rx_mintmr_trigger = intr_coal->rx_coalesce_usecs_low; else - intrmod_cfg->intrmod_mintmr_trigger = - LIO_INTRMOD_MINTMR_TRIGGER; + intrmod_cfg->rx_mintmr_trigger = + LIO_INTRMOD_RXMINTMR_TRIGGER; if (intr_coal->rx_max_coalesced_frames_low) - intrmod_cfg->intrmod_mincnt_trigger = + intrmod_cfg->rx_mincnt_trigger = intr_coal->rx_max_coalesced_frames_low; else - intrmod_cfg->intrmod_mincnt_trigger = - LIO_INTRMOD_MINCNT_TRIGGER; + intrmod_cfg->rx_mincnt_trigger = + LIO_INTRMOD_RXMINCNT_TRIGGER; + } + if (oct->intrmod.tx_enable) { + if (intr_coal->tx_max_coalesced_frames_high) + intrmod_cfg->tx_maxcnt_trigger = + intr_coal->tx_max_coalesced_frames_high; + else + intrmod_cfg->tx_maxcnt_trigger = + LIO_INTRMOD_TXMAXCNT_TRIGGER; + if (intr_coal->tx_max_coalesced_frames_low) + intrmod_cfg->tx_mincnt_trigger = + intr_coal->tx_max_coalesced_frames_low; + else + intrmod_cfg->tx_mincnt_trigger = + LIO_INTRMOD_TXMINCNT_TRIGGER; } - intrmod_cfg->intrmod_enable = adaptive; - ret = octnet_set_intrmod_cfg(oct, intrmod_cfg); + ret = octnet_set_intrmod_cfg(lio, intrmod_cfg); return ret; } @@ -800,54 +1294,82 @@ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce static int oct_cfg_rx_intrcnt(struct lio *lio, struct ethtool_coalesce *intr_coal) { - int ret; struct octeon_device *oct = lio->oct_dev; - struct octeon_cn6xxx *cn6xxx = (struct octeon_cn6xxx *)oct->chip; u32 rx_max_coalesced_frames; - if (!intr_coal->rx_max_coalesced_frames) - rx_max_coalesced_frames = CN6XXX_OQ_INTR_PKT; - else - rx_max_coalesced_frames = intr_coal->rx_max_coalesced_frames; - - /* Disable adaptive interrupt modulation */ - ret = oct_cfg_adaptive_intr(lio, intr_coal, 0); - if (ret) - return ret; - /* Config Cnt based interrupt values */ - octeon_write_csr(oct, CN6XXX_SLI_OQ_INT_LEVEL_PKTS, - rx_max_coalesced_frames); - CFG_SET_OQ_INTR_PKT(cn6xxx->conf, rx_max_coalesced_frames); + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: { + struct octeon_cn6xxx *cn6xxx = + (struct octeon_cn6xxx *)oct->chip; + + if (!intr_coal->rx_max_coalesced_frames) + rx_max_coalesced_frames = CN6XXX_OQ_INTR_PKT; + else + rx_max_coalesced_frames = + intr_coal->rx_max_coalesced_frames; + octeon_write_csr(oct, CN6XXX_SLI_OQ_INT_LEVEL_PKTS, + rx_max_coalesced_frames); + CFG_SET_OQ_INTR_PKT(cn6xxx->conf, rx_max_coalesced_frames); + break; + } + default: + return -EINVAL; + } return 0; } static int oct_cfg_rx_intrtime(struct lio *lio, struct ethtool_coalesce *intr_coal) { - int ret; struct octeon_device *oct = lio->oct_dev; - struct octeon_cn6xxx *cn6xxx = (struct octeon_cn6xxx *)oct->chip; u32 time_threshold, rx_coalesce_usecs; - if (!intr_coal->rx_coalesce_usecs) - rx_coalesce_usecs = CN6XXX_OQ_INTR_TIME; - else - rx_coalesce_usecs = intr_coal->rx_coalesce_usecs; + /* Config Time based interrupt values */ + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: { + struct octeon_cn6xxx *cn6xxx = + (struct octeon_cn6xxx *)oct->chip; + if (!intr_coal->rx_coalesce_usecs) + rx_coalesce_usecs = CN6XXX_OQ_INTR_TIME; + else + rx_coalesce_usecs = intr_coal->rx_coalesce_usecs; - /* Disable adaptive interrupt modulation */ - ret = oct_cfg_adaptive_intr(lio, intr_coal, 0); - if (ret) - return ret; + time_threshold = lio_cn6xxx_get_oq_ticks(oct, + rx_coalesce_usecs); + octeon_write_csr(oct, + CN6XXX_SLI_OQ_INT_LEVEL_TIME, + time_threshold); - /* Config Time based interrupt values */ - time_threshold = lio_cn6xxx_get_oq_ticks(oct, rx_coalesce_usecs); - octeon_write_csr(oct, CN6XXX_SLI_OQ_INT_LEVEL_TIME, time_threshold); - CFG_SET_OQ_INTR_TIME(cn6xxx->conf, rx_coalesce_usecs); + CFG_SET_OQ_INTR_TIME(cn6xxx->conf, rx_coalesce_usecs); + break; + } + default: + return -EINVAL; + } return 0; } +static int +oct_cfg_tx_intrcnt(struct lio *lio, struct ethtool_coalesce *intr_coal + __attribute__((unused))) +{ + struct octeon_device *oct = lio->oct_dev; + + /* Config Cnt based interrupt values */ + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: + break; + default: + return -EINVAL; + } + return 0; +} + static int lio_set_intr_coalesce(struct net_device *netdev, struct ethtool_coalesce *intr_coal) { @@ -855,59 +1377,48 @@ static int lio_set_intr_coalesce(struct net_device *netdev, int ret; struct octeon_device *oct = lio->oct_dev; u32 j, q_no; + int db_max, db_min; - if ((intr_coal->tx_max_coalesced_frames >= CN6XXX_DB_MIN) && - (intr_coal->tx_max_coalesced_frames <= CN6XXX_DB_MAX)) { - for (j = 0; j < lio->linfo.num_txpciq; j++) { - q_no = lio->linfo.txpciq[j]; - oct->instr_queue[q_no]->fill_threshold = - intr_coal->tx_max_coalesced_frames; + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: + db_min = CN6XXX_DB_MIN; + db_max = CN6XXX_DB_MAX; + if ((intr_coal->tx_max_coalesced_frames >= db_min) && + (intr_coal->tx_max_coalesced_frames <= db_max)) { + for (j = 0; j < lio->linfo.num_txpciq; j++) { + q_no = lio->linfo.txpciq[j].s.q_no; + oct->instr_queue[q_no]->fill_threshold = + intr_coal->tx_max_coalesced_frames; + } + } else { + dev_err(&oct->pci_dev->dev, + "LIQUIDIO: Invalid tx-frames:%d. Range is min:%d max:%d\n", + intr_coal->tx_max_coalesced_frames, db_min, + db_max); + return -EINVAL; } - } else { - dev_err(&oct->pci_dev->dev, - "LIQUIDIO: Invalid tx-frames:%d. Range is min:%d max:%d\n", - intr_coal->tx_max_coalesced_frames, CN6XXX_DB_MIN, - CN6XXX_DB_MAX); + break; + default: return -EINVAL; } - /* User requested adaptive-rx on */ - if (intr_coal->use_adaptive_rx_coalesce) { - ret = oct_cfg_adaptive_intr(lio, intr_coal, 1); - if (ret) - goto ret_intrmod; - } + oct->intrmod.rx_enable = intr_coal->use_adaptive_rx_coalesce ? 1 : 0; + oct->intrmod.tx_enable = intr_coal->use_adaptive_tx_coalesce ? 1 : 0; - /* User requested adaptive-rx off and rx coalesce */ - if ((intr_coal->rx_coalesce_usecs) && - (!intr_coal->use_adaptive_rx_coalesce)) { + ret = oct_cfg_adaptive_intr(lio, intr_coal); + + if (!intr_coal->use_adaptive_rx_coalesce) { ret = oct_cfg_rx_intrtime(lio, intr_coal); if (ret) goto ret_intrmod; - } - /* User requested adaptive-rx off and rx coalesce */ - if ((intr_coal->rx_max_coalesced_frames) && - (!intr_coal->use_adaptive_rx_coalesce)) { ret = oct_cfg_rx_intrcnt(lio, intr_coal); if (ret) goto ret_intrmod; } - - /* User requested adaptive-rx off, so use default coalesce params */ - if ((!intr_coal->rx_max_coalesced_frames) && - (!intr_coal->use_adaptive_rx_coalesce) && - (!intr_coal->rx_coalesce_usecs)) { - dev_info(&oct->pci_dev->dev, - "Turning off adaptive-rx interrupt moderation\n"); - dev_info(&oct->pci_dev->dev, - "Using RX Coalesce Default values rx_coalesce_usecs:%d rx_max_coalesced_frames:%d\n", - CN6XXX_OQ_INTR_TIME, CN6XXX_OQ_INTR_PKT); - ret = oct_cfg_rx_intrtime(lio, intr_coal); - if (ret) - goto ret_intrmod; - - ret = oct_cfg_rx_intrcnt(lio, intr_coal); + if (!intr_coal->use_adaptive_tx_coalesce) { + ret = oct_cfg_tx_intrcnt(lio, intr_coal); if (ret) goto ret_intrmod; } @@ -923,23 +1434,28 @@ static int lio_get_ts_info(struct net_device *netdev, struct lio *lio = GET_LIO(netdev); info->so_timestamping = +#ifdef PTP_HARDWARE_TIMESTAMPING SOF_TIMESTAMPING_TX_HARDWARE | - SOF_TIMESTAMPING_TX_SOFTWARE | SOF_TIMESTAMPING_RX_HARDWARE | + SOF_TIMESTAMPING_RAW_HARDWARE | + SOF_TIMESTAMPING_TX_SOFTWARE | +#endif SOF_TIMESTAMPING_RX_SOFTWARE | - SOF_TIMESTAMPING_SOFTWARE | SOF_TIMESTAMPING_RAW_HARDWARE; + SOF_TIMESTAMPING_SOFTWARE; if (lio->ptp_clock) info->phc_index = ptp_clock_index(lio->ptp_clock); else info->phc_index = -1; +#ifdef PTP_HARDWARE_TIMESTAMPING info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON); info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) | (1 << HWTSTAMP_FILTER_PTP_V1_L4_EVENT) | (1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) | (1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT); +#endif return 0; } @@ -950,7 +1466,6 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) struct octeon_device *oct = lio->oct_dev; struct oct_link_info *linfo; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int ret = 0; /* get the link info */ @@ -965,12 +1480,14 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) ecmd->duplex != DUPLEX_FULL))) return -EINVAL; - /* Ethtool Support is not provided for XAUI and RXAUI Interfaces + /* Ethtool Support is not provided for XAUI, RXAUI, and XFI Interfaces * as they operate at fixed Speed and Duplex settings */ - if (linfo->link.s.interface == INTERFACE_MODE_XAUI || - linfo->link.s.interface == INTERFACE_MODE_RXAUI) { - dev_info(&oct->pci_dev->dev, "XAUI IFs settings cannot be modified.\n"); + if (linfo->link.s.if_mode == INTERFACE_MODE_XAUI || + linfo->link.s.if_mode == INTERFACE_MODE_RXAUI || + linfo->link.s.if_mode == INTERFACE_MODE_XFI) { + dev_info(&oct->pci_dev->dev, + "Autonegotiation, duplex and speed settings cannot be modified.\n"); return -EINVAL; } @@ -978,9 +1495,9 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_SET_SETTINGS; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 1000; nctrl.netpndev = (u64)netdev; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; /* Passing the parameters sent by ethtool like Speed, Autoneg & Duplex @@ -990,19 +1507,17 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) /* Autoneg ON */ nctrl.ncmd.s.more = OCTNIC_NCMD_PHY_ON | OCTNIC_NCMD_AUTONEG_ON; - nctrl.ncmd.s.param2 = ecmd->advertising; + nctrl.ncmd.s.param1 = ecmd->advertising; } else { /* Autoneg OFF */ nctrl.ncmd.s.more = OCTNIC_NCMD_PHY_ON; - nctrl.ncmd.s.param3 = ecmd->duplex; + nctrl.ncmd.s.param2 = ecmd->duplex; - nctrl.ncmd.s.param2 = ecmd->speed; + nctrl.ncmd.s.param1 = ecmd->speed; } - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Failed to set settings\n"); return -1; @@ -1026,7 +1541,7 @@ static int lio_nway_reset(struct net_device *netdev) } /* Return register dump len. */ -static int lio_get_regs_len(struct net_device *dev) +static int lio_get_regs_len(struct net_device *dev __attribute__((unused))) { return OCT_ETHTOOL_REGDUMP_LEN; } @@ -1170,13 +1685,12 @@ static void lio_get_regs(struct net_device *dev, int len = 0; struct octeon_device *oct = lio->oct_dev; - memset(regbuf, 0, OCT_ETHTOOL_REGDUMP_LEN); regs->version = OCT_ETHTOOL_REGSVER; switch (oct->chip_id) { - /* case OCTEON_CN73XX: Todo */ case OCTEON_CN68XX: case OCTEON_CN66XX: + memset(regbuf, 0, OCT_ETHTOOL_REGDUMP_LEN); len += cn6xxx_read_csr_reg(regbuf + len, oct); len += cn6xxx_read_config_reg(regbuf + len, oct); break; @@ -1186,6 +1700,23 @@ static void lio_get_regs(struct net_device *dev, } } +static u32 lio_get_priv_flags(struct net_device *netdev) +{ + struct lio *lio = GET_LIO(netdev); + + return lio->oct_dev->priv_flags; +} + +static int lio_set_priv_flags(struct net_device *netdev, u32 flags) +{ + struct lio *lio = GET_LIO(netdev); + bool intr_by_tx_bytes = !!(flags & (0x1 << OCT_PRIV_FLAG_TX_BYTES)); + + lio_set_priv_flag(lio->oct_dev, OCT_PRIV_FLAG_TX_BYTES, + intr_by_tx_bytes); + return 0; +} + static const struct ethtool_ops lio_ethtool_ops = { .get_settings = lio_get_settings, .get_link = ethtool_op_get_link, @@ -1207,6 +1738,8 @@ static const struct ethtool_ops lio_ethtool_ops = { .set_settings = lio_set_settings, .get_coalesce = lio_get_intr_coalesce, .set_coalesce = lio_set_intr_coalesce, + .get_priv_flags = lio_get_priv_flags, + .set_priv_flags = lio_set_priv_flags, .get_ts_info = lio_get_ts_info, }; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 8de79ae63231..20d6942edf40 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -20,24 +20,12 @@ * Contact Cavium, Inc. for more information **********************************************************************/ #include <linux/version.h> -#include <linux/module.h> -#include <linux/crc32.h> -#include <linux/dma-mapping.h> #include <linux/pci.h> -#include <linux/pci_ids.h> -#include <linux/ip.h> -#include <net/ip.h> -#include <linux/ipv6.h> #include <linux/net_tstamp.h> #include <linux/if_vlan.h> #include <linux/firmware.h> -#include <linux/ethtool.h> #include <linux/ptp_clock_kernel.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/workqueue.h> -#include <linux/interrupt.h> -#include "octeon_config.h" +#include <net/vxlan.h> #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" @@ -48,7 +36,6 @@ #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" #include "cn68xx_device.h" #include "liquidio_image.h" @@ -72,6 +59,9 @@ MODULE_PARM_DESC(console_bitmask, #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK) +#define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \ + (octeon_dev_ptr->instr_queue[iq_no]->stats.field += count) + static int debug = -1; module_param(debug, int, 0644); MODULE_PARM_DESC(debug, "NETIF_MSG debug bits"); @@ -84,6 +74,8 @@ static int conf_type; module_param(conf_type, int, 0); MODULE_PARM_DESC(conf_type, "select octeon configuration 0 default 1 ovs"); +static int ptp_enable = 1; + /* Bit mask values for lio->ifstate */ #define LIO_IFSTATE_DROQ_OPS 0x01 #define LIO_IFSTATE_REGISTERED 0x02 @@ -166,6 +158,8 @@ struct octnic_gather { * received from the IP layer. */ struct octeon_sg_entry *sg; + + u64 sg_dma_ptr; }; /** This structure is used by NIC driver to store information required @@ -220,8 +214,8 @@ static void octeon_droq_bh(unsigned long pdev) (struct octeon_device_priv *)oct->priv; /* for (q_no = 0; q_no < oct->num_oqs; q_no++) { */ - for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES; q_no++) { - if (!(oct->io_qmask.oq & (1UL << q_no))) + for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES(oct); q_no++) { + if (!(oct->io_qmask.oq & (1ULL << q_no))) continue; reschedule |= octeon_droq_process_packets(oct, oct->droq[q_no], MAX_PACKET_BUDGET); @@ -241,11 +235,10 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct) do { pending_pkts = 0; - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; - pkt_cnt += octeon_droq_check_hw_for_pkts(oct, - oct->droq[i]); + pkt_cnt += octeon_droq_check_hw_for_pkts(oct->droq[i]); } if (pkt_cnt > 0) { pending_pkts += pkt_cnt; @@ -361,7 +354,7 @@ static int wait_for_pending_requests(struct octeon_device *oct) [OCTEON_ORDERED_SC_LIST].pending_req_count); if (pcount) schedule_timeout_uninterruptible(HZ / 10); - else + else break; } @@ -392,10 +385,10 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct) dev_err(&oct->pci_dev->dev, "There were pending requests\n"); /* Force all requests waiting to be fetched by OCTEON to complete. */ - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { struct octeon_instr_queue *iq; - if (!(oct->io_qmask.iq & (1UL << i))) + if (!(oct->io_qmask.iq & (1ULL << i))) continue; iq = oct->instr_queue[i]; @@ -405,7 +398,7 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct) iq->octeon_read_index = iq->host_write_index; iq->stats.instr_processed += atomic_read(&iq->instr_pending); - lio_process_iq_request_list(oct, iq); + lio_process_iq_request_list(oct, iq, 0); spin_unlock_bh(&iq->lock); } } @@ -500,7 +493,8 @@ static pci_ers_result_t liquidio_pcie_error_detected(struct pci_dev *pdev, * \brief mmio handler * @param pdev Pointer to PCI device */ -static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev *pdev) +static pci_ers_result_t liquidio_pcie_mmio_enabled( + struct pci_dev *pdev __attribute__((unused))) { /* We should never hit this since we never ask for a reset for a Fatal * Error. We always return DISCONNECT in io_error above. @@ -516,7 +510,8 @@ static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev *pdev) * Restart the card from scratch, as if from a cold-boot. Implementation * resembles the first-half of the octeon_resume routine. */ -static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev *pdev) +static pci_ers_result_t liquidio_pcie_slot_reset( + struct pci_dev *pdev __attribute__((unused))) { /* We should never hit this since we never ask for a reset for a Fatal * Error. We always return DISCONNECT in io_error above. @@ -533,7 +528,7 @@ static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev *pdev) * its OK to resume normal operation. Implementation resembles the * second-half of the octeon_resume routine. */ -static void liquidio_pcie_resume(struct pci_dev *pdev) +static void liquidio_pcie_resume(struct pci_dev *pdev __attribute__((unused))) { /* Nothing to be done here. */ } @@ -544,7 +539,8 @@ static void liquidio_pcie_resume(struct pci_dev *pdev) * @param pdev Pointer to PCI device * @param state state to suspend to */ -static int liquidio_suspend(struct pci_dev *pdev, pm_message_t state) +static int liquidio_suspend(struct pci_dev *pdev __attribute__((unused)), + pm_message_t state __attribute__((unused))) { return 0; } @@ -553,7 +549,7 @@ static int liquidio_suspend(struct pci_dev *pdev, pm_message_t state) * \brief called when resuming * @param pdev Pointer to PCI device */ -static int liquidio_resume(struct pci_dev *pdev) +static int liquidio_resume(struct pci_dev *pdev __attribute__((unused))) { return 0; } @@ -678,12 +674,24 @@ static inline void txqs_start(struct net_device *netdev) */ static inline void txqs_wake(struct net_device *netdev) { + struct lio *lio = GET_LIO(netdev); + if (netif_is_multiqueue(netdev)) { int i; - for (i = 0; i < netdev->num_tx_queues; i++) - netif_wake_subqueue(netdev, i); + for (i = 0; i < netdev->num_tx_queues; i++) { + int qno = lio->linfo.txpciq[i % + (lio->linfo.num_txpciq)].s.q_no; + + if (__netif_subqueue_stopped(netdev, i)) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, qno, + tx_restart, 1); + netif_wake_subqueue(netdev, i); + } + } } else { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq, + tx_restart, 1); netif_wake_queue(netdev); } } @@ -705,7 +713,7 @@ static void start_txq(struct net_device *netdev) { struct lio *lio = GET_LIO(netdev); - if (lio->linfo.link.s.status) { + if (lio->linfo.link.s.link_up) { txqs_start(netdev); return; } @@ -752,16 +760,23 @@ static inline int check_txq_status(struct lio *lio) /* check each sub-queue state */ for (q = 0; q < numqs; q++) { - iq = lio->linfo.txpciq[q & (lio->linfo.num_txpciq - 1)]; + iq = lio->linfo.txpciq[q % + (lio->linfo.num_txpciq)].s.q_no; if (octnet_iq_is_full(lio->oct_dev, iq)) continue; - wake_q(lio->netdev, q); - ret_val++; + if (__netif_subqueue_stopped(lio->netdev, q)) { + wake_q(lio->netdev, q); + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq, + tx_restart, 1); + ret_val++; + } } } else { if (octnet_iq_is_full(lio->oct_dev, lio->txq)) return 0; wake_q(lio->netdev, lio->txq); + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq, + tx_restart, 1); ret_val = 1; } return ret_val; @@ -787,64 +802,116 @@ static inline struct list_head *list_delete_head(struct list_head *root) } /** - * \brief Delete gather list + * \brief Delete gather lists * @param lio per-network private data */ -static void delete_glist(struct lio *lio) +static void delete_glists(struct lio *lio) { struct octnic_gather *g; + int i; - do { - g = (struct octnic_gather *) - list_delete_head(&lio->glist); - if (g) { - if (g->sg) - kfree((void *)((unsigned long)g->sg - - g->adjust)); - kfree(g); - } - } while (g); + if (!lio->glist) + return; + + for (i = 0; i < lio->linfo.num_txpciq; i++) { + do { + g = (struct octnic_gather *) + list_delete_head(&lio->glist[i]); + if (g) { + if (g->sg) { + dma_unmap_single(&lio->oct_dev-> + pci_dev->dev, + g->sg_dma_ptr, + g->sg_size, + DMA_TO_DEVICE); + kfree((void *)((unsigned long)g->sg - + g->adjust)); + } + kfree(g); + } + } while (g); + } + + kfree((void *)lio->glist); } /** - * \brief Setup gather list + * \brief Setup gather lists * @param lio per-network private data */ -static int setup_glist(struct lio *lio) +static int setup_glists(struct octeon_device *oct, struct lio *lio, int num_iqs) { - int i; + int i, j; struct octnic_gather *g; - INIT_LIST_HEAD(&lio->glist); + lio->glist_lock = kcalloc(num_iqs, sizeof(*lio->glist_lock), + GFP_KERNEL); + if (!lio->glist_lock) + return 1; - for (i = 0; i < lio->tx_qsize; i++) { - g = kzalloc(sizeof(*g), GFP_KERNEL); - if (!g) - break; + lio->glist = kcalloc(num_iqs, sizeof(*lio->glist), + GFP_KERNEL); + if (!lio->glist) { + kfree((void *)lio->glist_lock); + return 1; + } - g->sg_size = - ((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * OCT_SG_ENTRY_SIZE); + for (i = 0; i < num_iqs; i++) { + int numa_node = cpu_to_node(i % num_online_cpus()); - g->sg = kmalloc(g->sg_size + 8, GFP_KERNEL); - if (!g->sg) { - kfree(g); - break; + spin_lock_init(&lio->glist_lock[i]); + + INIT_LIST_HEAD(&lio->glist[i]); + + for (j = 0; j < lio->tx_qsize; j++) { + g = kzalloc_node(sizeof(*g), GFP_KERNEL, + numa_node); + if (!g) + g = kzalloc(sizeof(*g), GFP_KERNEL); + if (!g) + break; + + g->sg_size = ((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * + OCT_SG_ENTRY_SIZE); + + g->sg = kmalloc_node(g->sg_size + 8, + GFP_KERNEL, numa_node); + if (!g->sg) + g->sg = kmalloc(g->sg_size + 8, GFP_KERNEL); + if (!g->sg) { + kfree(g); + break; + } + + /* The gather component should be aligned on 64-bit + * boundary + */ + if (((unsigned long)g->sg) & 7) { + g->adjust = 8 - (((unsigned long)g->sg) & 7); + g->sg = (struct octeon_sg_entry *) + ((unsigned long)g->sg + g->adjust); + } + g->sg_dma_ptr = dma_map_single(&oct->pci_dev->dev, + g->sg, g->sg_size, + DMA_TO_DEVICE); + if (dma_mapping_error(&oct->pci_dev->dev, + g->sg_dma_ptr)) { + kfree((void *)((unsigned long)g->sg - + g->adjust)); + kfree(g); + break; + } + + list_add_tail(&g->list, &lio->glist[i]); } - /* The gather component should be aligned on 64-bit boundary */ - if (((unsigned long)g->sg) & 7) { - g->adjust = 8 - (((unsigned long)g->sg) & 7); - g->sg = (struct octeon_sg_entry *) - ((unsigned long)g->sg + g->adjust); + if (j != lio->tx_qsize) { + delete_glists(lio); + return 1; } - list_add_tail(&g->list, &lio->glist); } - if (i == lio->tx_qsize) - return 0; - - delete_glist(lio); - return 1; + return 0; } /** @@ -858,7 +925,7 @@ static void print_link_info(struct net_device *netdev) if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) { struct oct_link_info *linfo = &lio->linfo; - if (linfo->link.s.status) { + if (linfo->link.s.link_up) { netif_info(lio, link, lio->netdev, "%d Mbps %s Duplex UP\n", linfo->link.s.speed, (linfo->link.s.duplex) ? "Full" : "Half"); @@ -880,13 +947,15 @@ static inline void update_link_status(struct net_device *netdev, union oct_link_status *ls) { struct lio *lio = GET_LIO(netdev); + int changed = (lio->linfo.link.u64 != ls->u64); - if ((lio->intf_open) && (lio->linfo.link.u64 != ls->u64)) { - lio->linfo.link.u64 = ls->u64; + lio->linfo.link.u64 = ls->u64; + if ((lio->intf_open) && (changed)) { print_link_info(netdev); + lio->link_changes++; - if (lio->linfo.link.s.status) { + if (lio->linfo.link.s.link_up) { netif_carrier_on(netdev); /* start_txq(netdev); */ txqs_wake(netdev); @@ -897,6 +966,42 @@ static inline void update_link_status(struct net_device *netdev, } } +/* Runs in interrupt context. */ +static void update_txq_status(struct octeon_device *oct, int iq_num) +{ + struct net_device *netdev; + struct lio *lio; + struct octeon_instr_queue *iq = oct->instr_queue[iq_num]; + + /*octeon_update_iq_read_idx(oct, iq);*/ + + netdev = oct->props[iq->ifidx].netdev; + + /* This is needed because the first IQ does not have + * a netdev associated with it. + */ + if (!netdev) + return; + + lio = GET_LIO(netdev); + if (netif_is_multiqueue(netdev)) { + if (__netif_subqueue_stopped(netdev, iq->q_index) && + lio->linfo.link.s.link_up && + (!octnet_iq_is_full(oct, iq_num))) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq_num, + tx_restart, 1); + netif_wake_subqueue(netdev, iq->q_index); + } else { + if (!octnet_iq_is_full(oct, lio->txq)) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, + lio->txq, + tx_restart, 1); + wake_q(netdev, lio->txq); + } + } + } +} + /** * \brief Droq packet processor sceduler * @param oct octeon device @@ -910,8 +1015,9 @@ void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct) struct octeon_droq *droq; if (oct->int_status & OCT_DEV_INTR_PKT_DATA) { - for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES; oq_no++) { - if (!(oct->droq_intr & (1 << oq_no))) + for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct); + oq_no++) { + if (!(oct->droq_intr & (1ULL << oq_no))) continue; droq = oct->droq[oq_no]; @@ -987,7 +1093,9 @@ static int octeon_setup_interrupt(struct octeon_device *oct) * @param pdev PCI device structure * @param ent unused */ -static int liquidio_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +static int +liquidio_probe(struct pci_dev *pdev, + const struct pci_device_id *ent __attribute__((unused))) { struct octeon_device *oct_dev = NULL; struct handshake *hs; @@ -1022,6 +1130,9 @@ static int liquidio_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return -ENOMEM; } + oct_dev->rx_pause = 1; + oct_dev->tx_pause = 1; + dev_dbg(&oct_dev->pci_dev->dev, "Device is ready\n"); return 0; @@ -1087,19 +1198,13 @@ static void octeon_destroy_resources(struct octeon_device *oct) if (oct->flags & LIO_FLAG_MSI_ENABLED) pci_disable_msi(oct->pci_dev); - /* Soft reset the octeon device before exiting */ - oct->fn_list.soft_reset(oct); - - /* Disable the device, releasing the PCI INT */ - pci_disable_device(oct->pci_dev); - /* fallthrough */ case OCT_DEV_IN_RESET: case OCT_DEV_DROQ_INIT_DONE: /*atomic_set(&oct->status, OCT_DEV_DROQ_INIT_DONE);*/ mdelay(100); - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; octeon_delete_droq(oct, i); } @@ -1126,8 +1231,8 @@ static void octeon_destroy_resources(struct octeon_device *oct) /* fallthrough */ case OCT_DEV_INSTR_QUEUE_INIT_DONE: - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; octeon_delete_instr_queue(oct, i); } @@ -1139,14 +1244,21 @@ static void octeon_destroy_resources(struct octeon_device *oct) /* fallthrough */ case OCT_DEV_PCI_MAP_DONE: + + /* Soft reset the octeon device before exiting */ + oct->fn_list.soft_reset(oct); + octeon_unmap_pci_barx(oct, 0); octeon_unmap_pci_barx(oct, 1); /* fallthrough */ case OCT_DEV_BEGIN_STATE: + /* Disable the device, releasing the PCI INT */ + pci_disable_device(oct->pci_dev); + /* Nothing to be done here either */ break; - } /* end switch(oct->status) */ + } /* end switch (oct->status) */ tasklet_kill(&oct_priv->droq_tasklet); } @@ -1159,18 +1271,15 @@ static void octeon_destroy_resources(struct octeon_device *oct) static void send_rx_ctrl_cmd(struct lio *lio, int start_stop) { struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.s.cmd = OCTNET_CMD_RX_CTL; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = start_stop; + nctrl.ncmd.s.param1 = start_stop; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.netpndev = (u64)lio->netdev; - nparams.resp_order = OCTEON_RESP_NORESPONSE; - - if (octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams) < 0) + if (octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl) < 0) netif_info(lio, rx_err, lio->netdev, "Failed to send RX Control message\n"); } @@ -1186,6 +1295,7 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) { struct net_device *netdev = oct->props[ifidx].netdev; struct lio *lio; + struct napi_struct *napi, *n; if (!netdev) { dev_err(&oct->pci_dev->dev, "%s No netdevice ptr for index %d\n", @@ -1202,13 +1312,22 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) if (atomic_read(&lio->ifstate) & LIO_IFSTATE_RUNNING) txqs_stop(netdev); + if (oct->props[lio->ifidx].napi_enabled == 1) { + list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) + napi_disable(napi); + + oct->props[lio->ifidx].napi_enabled = 0; + } + if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) unregister_netdev(netdev); - delete_glist(lio); + delete_glists(lio); free_netdev(netdev); + oct->props[ifidx].gmxport = -1; + oct->props[ifidx].netdev = NULL; } @@ -1227,10 +1346,15 @@ static int liquidio_stop_nic_module(struct octeon_device *oct) return 1; } + spin_lock_bh(&oct->cmd_resp_wqlock); + oct->cmd_resp_state = OCT_DRV_OFFLINE; + spin_unlock_bh(&oct->cmd_resp_wqlock); + for (i = 0; i < oct->ifcount; i++) { lio = GET_LIO(oct->props[i].netdev); for (j = 0; j < lio->linfo.num_rxpciq; j++) - octeon_unregister_droq_ops(oct, lio->linfo.rxpciq[j]); + octeon_unregister_droq_ops(oct, + lio->linfo.rxpciq[j].s.q_no); } for (i = 0; i < oct->ifcount; i++) @@ -1274,6 +1398,7 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) { u32 dev_id, rev_id; int ret = 1; + char *s; pci_read_config_dword(oct->pci_dev, 0, &dev_id); pci_read_config_dword(oct->pci_dev, 8, &rev_id); @@ -1283,22 +1408,27 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) case OCTEON_CN68XX_PCIID: oct->chip_id = OCTEON_CN68XX; ret = lio_setup_cn68xx_octeon_device(oct); + s = "CN68XX"; break; case OCTEON_CN66XX_PCIID: oct->chip_id = OCTEON_CN66XX; ret = lio_setup_cn66xx_octeon_device(oct); + s = "CN66XX"; break; + default: + s = "?"; dev_err(&oct->pci_dev->dev, "Unknown device found (dev_id: %x)\n", dev_id); } if (!ret) - dev_info(&oct->pci_dev->dev, "CN68XX PASS%d.%d %s\n", + dev_info(&oct->pci_dev->dev, "%s PASS%d.%d %s Version: %s\n", s, OCTEON_MAJOR_REV(oct), OCTEON_MINOR_REV(oct), - octeon_get_conf(oct)->card_name); + octeon_get_conf(oct)->card_name, + LIQUIDIO_VERSION); return ret; } @@ -1326,6 +1456,16 @@ static int octeon_pci_os_setup(struct octeon_device *oct) return 0; } +static inline int skb_iq(struct lio *lio, struct sk_buff *skb) +{ + int q = 0; + + if (netif_is_multiqueue(lio->netdev)) + q = skb->queue_mapping % lio->linfo.num_txpciq; + + return q; +} + /** * \brief Check Tx queue state for a given network buffer * @param lio per-network private data @@ -1337,14 +1477,19 @@ static inline int check_txq_state(struct lio *lio, struct sk_buff *skb) if (netif_is_multiqueue(lio->netdev)) { q = skb->queue_mapping; - iq = lio->linfo.txpciq[(q & (lio->linfo.num_txpciq - 1))]; + iq = lio->linfo.txpciq[(q % (lio->linfo.num_txpciq))].s.q_no; } else { iq = lio->txq; + q = iq; } if (octnet_iq_is_full(lio->oct_dev, iq)) return 0; - wake_q(lio->netdev, q); + + if (__netif_subqueue_stopped(lio->netdev, q)) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq, tx_restart, 1); + wake_q(lio->netdev, q); + } return 1; } @@ -1367,7 +1512,7 @@ static void free_netbuf(void *buf) check_txq_state(lio, skb); - recv_buffer_free((struct sk_buff *)skb); + tx_buffer_free(skb); } /** @@ -1380,7 +1525,7 @@ static void free_netsgbuf(void *buf) struct sk_buff *skb; struct lio *lio; struct octnic_gather *g; - int i, frags; + int i, frags, iq; finfo = (struct octnet_buf_free_info *)buf; skb = finfo->skb; @@ -1402,17 +1547,17 @@ static void free_netsgbuf(void *buf) i++; } - dma_unmap_single(&lio->oct_dev->pci_dev->dev, - finfo->dptr, g->sg_size, - DMA_TO_DEVICE); + dma_sync_single_for_cpu(&lio->oct_dev->pci_dev->dev, + g->sg_dma_ptr, g->sg_size, DMA_TO_DEVICE); - spin_lock(&lio->lock); - list_add_tail(&g->list, &lio->glist); - spin_unlock(&lio->lock); + iq = skb_iq(lio, skb); + spin_lock(&lio->glist_lock[iq]); + list_add_tail(&g->list, &lio->glist[iq]); + spin_unlock(&lio->glist_lock[iq]); check_txq_state(lio, skb); /* mq support: sub-queue state check */ - recv_buffer_free((struct sk_buff *)skb); + tx_buffer_free(skb); } /** @@ -1426,7 +1571,7 @@ static void free_netsgbuf_with_resp(void *buf) struct sk_buff *skb; struct lio *lio; struct octnic_gather *g; - int i, frags; + int i, frags, iq; sc = (struct octeon_soft_command *)buf; skb = (struct sk_buff *)sc->callback_arg; @@ -1450,13 +1595,14 @@ static void free_netsgbuf_with_resp(void *buf) i++; } - dma_unmap_single(&lio->oct_dev->pci_dev->dev, - finfo->dptr, g->sg_size, - DMA_TO_DEVICE); + dma_sync_single_for_cpu(&lio->oct_dev->pci_dev->dev, + g->sg_dma_ptr, g->sg_size, DMA_TO_DEVICE); + + iq = skb_iq(lio, skb); - spin_lock(&lio->lock); - list_add_tail(&g->list, &lio->glist); - spin_unlock(&lio->lock); + spin_lock(&lio->glist_lock[iq]); + list_add_tail(&g->list, &lio->glist[iq]); + spin_unlock(&lio->glist_lock[iq]); /* Don't free the skb yet */ @@ -1569,8 +1715,10 @@ static int liquidio_ptp_settime(struct ptp_clock_info *ptp, * @param rq request * @param on is it on */ -static int liquidio_ptp_enable(struct ptp_clock_info *ptp, - struct ptp_clock_request *rq, int on) +static int +liquidio_ptp_enable(struct ptp_clock_info *ptp __attribute__((unused)), + struct ptp_clock_request *rq __attribute__((unused)), + int on __attribute__((unused))) { return -EOPNOTSUPP; } @@ -1657,6 +1805,7 @@ static int load_firmware(struct octeon_device *oct) if (ret) { dev_err(&oct->pci_dev->dev, "Request firmware failed. Could not find file %s.\n.", fw_name); + release_firmware(fw); return ret; } @@ -1710,7 +1859,7 @@ static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs, * @param buf pointer to resp structure */ static void if_cfg_callback(struct octeon_device *oct, - u32 status, + u32 status __attribute__((unused)), void *buf) { struct octeon_soft_command *sc = (struct octeon_soft_command *)buf; @@ -1724,7 +1873,10 @@ static void if_cfg_callback(struct octeon_device *oct, if (resp->status) dev_err(&oct->pci_dev->dev, "nic if cfg instruction failed. Status: %llx\n", CVM_CAST64(resp->status)); - ACCESS_ONCE(ctx->cond) = 1; + WRITE_ONCE(ctx->cond, 1); + + snprintf(oct->fw_info.liquidio_firmware_version, 32, "%s", + resp->cfg_info.liquidio_firmware_version); /* This barrier is required to be sure that the response has been * written fully before waking up the handler @@ -1741,16 +1893,16 @@ static void if_cfg_callback(struct octeon_device *oct, * @returns selected queue number */ static u16 select_q(struct net_device *dev, struct sk_buff *skb, - void *accel_priv, select_queue_fallback_t fallback) + void *accel_priv __attribute__((unused)), + select_queue_fallback_t fallback __attribute__((unused))) { - int qindex; + u32 qindex = 0; struct lio *lio; lio = GET_LIO(dev); - /* select queue on chosen queue_mapping or core */ - qindex = skb_rx_queue_recorded(skb) ? - skb_get_rx_queue(skb) : smp_processor_id(); - return (u16)(qindex & (lio->linfo.num_txpciq - 1)); + qindex = skb_tx_hash(dev, skb); + + return (u16)(qindex % (lio->linfo.num_txpciq)); } /** Routine to push packets arriving on Octeon interface upto network layer. @@ -1759,26 +1911,28 @@ static u16 select_q(struct net_device *dev, struct sk_buff *skb, * @param len - size of total data received. * @param rh - Control header associated with the packet * @param param - additional control data with the packet + * @param arg - farg registered in droq_ops */ static void -liquidio_push_packet(u32 octeon_id, +liquidio_push_packet(u32 octeon_id __attribute__((unused)), void *skbuff, u32 len, union octeon_rh *rh, - void *param) + void *param, + void *arg) { struct napi_struct *napi = param; - struct octeon_device *oct = lio_get_device(octeon_id); struct sk_buff *skb = (struct sk_buff *)skbuff; struct skb_shared_hwtstamps *shhwtstamps; u64 ns; - struct net_device *netdev = - (struct net_device *)oct->props[rh->r_dh.link].netdev; + u16 vtag = 0; + struct net_device *netdev = (struct net_device *)arg; struct octeon_droq *droq = container_of(param, struct octeon_droq, napi); if (netdev) { int packet_was_received; struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; /* Do not proceed if the interface is not in RUNNING state. */ if (!ifstate_check(lio, LIO_IFSTATE_RUNNING)) { @@ -1789,32 +1943,86 @@ liquidio_push_packet(u32 octeon_id, skb->dev = netdev; - if (rh->r_dh.has_hwtstamp) { - /* timestamp is included from the hardware at the - * beginning of the packet. - */ - if (ifstate_check(lio, - LIO_IFSTATE_RX_TIMESTAMP_ENABLED)) { - /* Nanoseconds are in the first 64-bits - * of the packet. + skb_record_rx_queue(skb, droq->q_no); + if (likely(len > MIN_SKB_SIZE)) { + struct octeon_skb_page_info *pg_info; + unsigned char *va; + + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + if (pg_info->page) { + /* For Paged allocation use the frags */ + va = page_address(pg_info->page) + + pg_info->page_offset; + memcpy(skb->data, va, MIN_SKB_SIZE); + skb_put(skb, MIN_SKB_SIZE); + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, + pg_info->page, + pg_info->page_offset + + MIN_SKB_SIZE, + len - MIN_SKB_SIZE, + LIO_RXBUFFER_SZ); + } + } else { + struct octeon_skb_page_info *pg_info = + ((struct octeon_skb_page_info *)(skb->cb)); + skb_copy_to_linear_data(skb, page_address(pg_info->page) + + pg_info->page_offset, len); + skb_put(skb, len); + put_page(pg_info->page); + } + + if (((oct->chip_id == OCTEON_CN66XX) || + (oct->chip_id == OCTEON_CN68XX)) && + ptp_enable) { + if (rh->r_dh.has_hwtstamp) { + /* timestamp is included from the hardware at + * the beginning of the packet. */ - memcpy(&ns, (skb->data), sizeof(ns)); - shhwtstamps = skb_hwtstamps(skb); - shhwtstamps->hwtstamp = - ns_to_ktime(ns + lio->ptp_adjust); + if (ifstate_check + (lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED)) { + /* Nanoseconds are in the first 64-bits + * of the packet. + */ + memcpy(&ns, (skb->data), sizeof(ns)); + shhwtstamps = skb_hwtstamps(skb); + shhwtstamps->hwtstamp = + ns_to_ktime(ns + + lio->ptp_adjust); + } + skb_pull(skb, sizeof(ns)); } - skb_pull(skb, sizeof(ns)); } skb->protocol = eth_type_trans(skb, skb->dev); - if ((netdev->features & NETIF_F_RXCSUM) && - (rh->r_dh.csum_verified == CNNIC_CSUM_VERIFIED)) + (((rh->r_dh.encap_on) && + (rh->r_dh.csum_verified & CNNIC_TUN_CSUM_VERIFIED)) || + (!(rh->r_dh.encap_on) && + (rh->r_dh.csum_verified & CNNIC_CSUM_VERIFIED)))) /* checksum has already been verified */ skb->ip_summed = CHECKSUM_UNNECESSARY; else skb->ip_summed = CHECKSUM_NONE; + /* Setting Encapsulation field on basis of status received + * from the firmware + */ + if (rh->r_dh.encap_on) { + skb->encapsulation = 1; + skb->csum_level = 1; + droq->stats.rx_vxlan++; + } + + /* inbound VLAN tag */ + if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) && + (rh->r_dh.vlan != 0)) { + u16 vid = rh->r_dh.vlan; + u16 priority = rh->r_dh.priority; + + vtag = priority << 13 | vid; + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vtag); + } + packet_was_received = napi_gro_receive(napi, skb) != GRO_DROP; if (packet_was_received) { @@ -1869,39 +2077,6 @@ static void liquidio_napi_drv_callback(void *arg) } /** - * \brief Main NAPI poll function - * @param droq octeon output queue - * @param budget maximum number of items to process - */ -static int liquidio_napi_do_rx(struct octeon_droq *droq, int budget) -{ - int work_done; - struct lio *lio = GET_LIO(droq->napi.dev); - struct octeon_device *oct = lio->oct_dev; - - work_done = octeon_process_droq_poll_cmd(oct, droq->q_no, - POLL_EVENT_PROCESS_PKTS, - budget); - if (work_done < 0) { - netif_info(lio, rx_err, lio->netdev, - "Receive work_done < 0, rxq:%d\n", droq->q_no); - goto octnet_napi_finish; - } - - if (work_done > budget) - dev_err(&oct->pci_dev->dev, ">>>> %s work_done: %d budget: %d\n", - __func__, work_done, budget); - - return work_done; - -octnet_napi_finish: - napi_complete(&droq->napi); - octeon_process_droq_poll_cmd(oct, droq->q_no, POLL_EVENT_ENABLE_INTR, - 0); - return 0; -} - -/** * \brief Entry point for NAPI polling * @param napi NAPI structure * @param budget maximum number of items to process @@ -1910,35 +2085,57 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget) { struct octeon_droq *droq; int work_done; + int tx_done = 0, iq_no; + struct octeon_instr_queue *iq; + struct octeon_device *oct; droq = container_of(napi, struct octeon_droq, napi); + oct = droq->oct_dev; + iq_no = droq->q_no; + /* Handle Droq descriptors */ + work_done = octeon_process_droq_poll_cmd(oct, droq->q_no, + POLL_EVENT_PROCESS_PKTS, + budget); - work_done = liquidio_napi_do_rx(droq, budget); + /* Flush the instruction queue */ + iq = oct->instr_queue[iq_no]; + if (iq) { + /* Process iq buffers with in the budget limits */ + tx_done = octeon_flush_iq(oct, iq, 1, budget); + /* Update iq read-index rather than waiting for next interrupt. + * Return back if tx_done is false. + */ + update_txq_status(oct, iq_no); + /*tx_done = (iq->flush_index == iq->octeon_read_index);*/ + } else { + dev_err(&oct->pci_dev->dev, "%s: iq (%d) num invalid\n", + __func__, iq_no); + } - if (work_done < budget) { + if ((work_done < budget) && (tx_done)) { napi_complete(napi); octeon_process_droq_poll_cmd(droq->oct_dev, droq->q_no, POLL_EVENT_ENABLE_INTR, 0); return 0; } - return work_done; + return (!tx_done) ? (budget) : (work_done); } /** * \brief Setup input and output queues * @param octeon_dev octeon device - * @param net_device Net device + * @param ifidx Interface Index * * Note: Queues are with respect to the octeon device. Thus * an input queue is for egress packets, and output queues * are for ingress packets. */ static inline int setup_io_queues(struct octeon_device *octeon_dev, - struct net_device *net_device) + int ifidx) { - static int first_time = 1; - static struct octeon_droq_ops droq_ops; + struct octeon_droq_ops droq_ops; + struct net_device *netdev; static int cpu_id; static int cpu_id_modulus; struct octeon_droq *droq; @@ -1947,23 +2144,26 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, struct lio *lio; int num_tx_descs; - lio = GET_LIO(net_device); - if (first_time) { - first_time = 0; - memset(&droq_ops, 0, sizeof(struct octeon_droq_ops)); + netdev = octeon_dev->props[ifidx].netdev; - droq_ops.fptr = liquidio_push_packet; + lio = GET_LIO(netdev); - droq_ops.poll_mode = 1; - droq_ops.napi_fn = liquidio_napi_drv_callback; - cpu_id = 0; - cpu_id_modulus = num_present_cpus(); - } + memset(&droq_ops, 0, sizeof(struct octeon_droq_ops)); + + droq_ops.fptr = liquidio_push_packet; + droq_ops.farg = (void *)netdev; + + droq_ops.poll_mode = 1; + droq_ops.napi_fn = liquidio_napi_drv_callback; + cpu_id = 0; + cpu_id_modulus = num_present_cpus(); /* set up DROQs. */ for (q = 0; q < lio->linfo.num_rxpciq; q++) { - q_no = lio->linfo.rxpciq[q]; - + q_no = lio->linfo.rxpciq[q].s.q_no; + dev_dbg(&octeon_dev->pci_dev->dev, + "setup_io_queues index:%d linfo.rxpciq.s.q_no:%d\n", + q, q_no); retval = octeon_setup_droq(octeon_dev, q_no, CFG_GET_NUM_RX_DESCS_NIC_IF (octeon_get_conf(octeon_dev), @@ -1980,7 +2180,11 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, droq = octeon_dev->droq[q_no]; napi = &droq->napi; - netif_napi_add(net_device, napi, liquidio_napi_poll, 64); + dev_dbg(&octeon_dev->pci_dev->dev, + "netif_napi_add netdev:%llx oct:%llx\n", + (u64)netdev, + (u64)octeon_dev); + netif_napi_add(netdev, napi, liquidio_napi_poll, 64); /* designate a CPU for this droq */ droq->cpu_id = cpu_id; @@ -1996,9 +2200,9 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, num_tx_descs = CFG_GET_NUM_TX_DESCS_NIC_IF(octeon_get_conf (octeon_dev), lio->ifidx); - retval = octeon_setup_iq(octeon_dev, lio->linfo.txpciq[q], - num_tx_descs, - netdev_get_tx_queue(net_device, q)); + retval = octeon_setup_iq(octeon_dev, ifidx, q, + lio->linfo.txpciq[q], num_tx_descs, + netdev_get_tx_queue(netdev, q)); if (retval) { dev_err(&octeon_dev->pci_dev->dev, " %s : Runtime IQ(TxQ) creation failed.\n", @@ -2036,7 +2240,8 @@ static inline void setup_tx_poll_fn(struct net_device *netdev) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; - lio->txq_status_wq.wq = create_workqueue("txq-status"); + lio->txq_status_wq.wq = alloc_workqueue("txq-status", + WQ_MEM_RECLAIM, 0); if (!lio->txq_status_wq.wq) { dev_err(&oct->pci_dev->dev, "unable to create cavium txq status wq\n"); return; @@ -2048,6 +2253,14 @@ static inline void setup_tx_poll_fn(struct net_device *netdev) &lio->txq_status_wq.wk.work, msecs_to_jiffies(1)); } +static inline void cleanup_tx_poll_fn(struct net_device *netdev) +{ + struct lio *lio = GET_LIO(netdev); + + cancel_delayed_work_sync(&lio->txq_status_wq.wk.work); + destroy_workqueue(lio->txq_status_wq.wq); +} + /** * \brief Net device open for LiquidIO * @param netdev network device @@ -2058,17 +2271,22 @@ static int liquidio_open(struct net_device *netdev) struct octeon_device *oct = lio->oct_dev; struct napi_struct *napi, *n; - list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) - napi_enable(napi); + if (oct->props[lio->ifidx].napi_enabled == 0) { + list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) + napi_enable(napi); + + oct->props[lio->ifidx].napi_enabled = 1; + } oct_ptp_open(netdev); ifstate_set(lio, LIO_IFSTATE_RUNNING); + setup_tx_poll_fn(netdev); + start_txq(netdev); netif_info(lio, ifup, lio->netdev, "Interface Open, ready for traffic\n"); - try_module_get(THIS_MODULE); /* tell Octeon to start forwarding packets to host */ send_rx_ctrl_cmd(lio, 1); @@ -2088,41 +2306,36 @@ static int liquidio_open(struct net_device *netdev) */ static int liquidio_stop(struct net_device *netdev) { - struct napi_struct *napi, *n; struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; - netif_info(lio, ifdown, lio->netdev, "Stopping interface!\n"); + ifstate_reset(lio, LIO_IFSTATE_RUNNING); + + netif_tx_disable(netdev); + /* Inform that netif carrier is down */ + netif_carrier_off(netdev); lio->intf_open = 0; - lio->linfo.link.s.status = 0; + lio->linfo.link.s.link_up = 0; + lio->link_changes++; - netif_carrier_off(netdev); + /* Pause for a moment and wait for Octeon to flush out (to the wire) any + * egress packets that are in-flight. + */ + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(msecs_to_jiffies(100)); - /* tell Octeon to stop forwarding packets to host */ + /* Now it should be safe to tell Octeon that nic interface is down. */ send_rx_ctrl_cmd(lio, 0); - cancel_delayed_work_sync(&lio->txq_status_wq.wk.work); - flush_workqueue(lio->txq_status_wq.wq); - destroy_workqueue(lio->txq_status_wq.wq); + cleanup_tx_poll_fn(netdev); if (lio->ptp_clock) { ptp_clock_unregister(lio->ptp_clock); lio->ptp_clock = NULL; } - ifstate_reset(lio, LIO_IFSTATE_RUNNING); - - /* This is a hack that allows DHCP to continue working. */ - set_bit(__LINK_STATE_START, &lio->netdev->state); - - list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) - napi_disable(napi); - - txqs_stop(netdev); - dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name); - module_put(THIS_MODULE); return 0; } @@ -2133,6 +2346,7 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) struct net_device *netdev = (struct net_device *)nctrl->netpndev; struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; + u8 *mac; switch (nctrl->ncmd.s.cmd) { case OCTNET_CMD_CHANGE_DEVFLAGS: @@ -2140,22 +2354,24 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) break; case OCTNET_CMD_CHANGE_MACADDR: - /* If command is successful, change the MACADDR. */ - netif_info(lio, probe, lio->netdev, " MACAddr changed to 0x%llx\n", - CVM_CAST64(nctrl->udd[0])); - dev_info(&oct->pci_dev->dev, "%s MACAddr changed to 0x%llx\n", - netdev->name, CVM_CAST64(nctrl->udd[0])); - memcpy(netdev->dev_addr, ((u8 *)&nctrl->udd[0]) + 2, ETH_ALEN); + mac = ((u8 *)&nctrl->udd[0]) + 2; + netif_info(lio, probe, lio->netdev, + "%s %2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x\n", + "MACAddr changed to", mac[0], mac[1], + mac[2], mac[3], mac[4], mac[5]); break; case OCTNET_CMD_CHANGE_MTU: /* If command is successful, change the MTU. */ netif_info(lio, probe, lio->netdev, " MTU Changed from %d to %d\n", - netdev->mtu, nctrl->ncmd.s.param2); + netdev->mtu, nctrl->ncmd.s.param1); dev_info(&oct->pci_dev->dev, "%s MTU Changed from %d to %d\n", netdev->name, netdev->mtu, - nctrl->ncmd.s.param2); - netdev->mtu = nctrl->ncmd.s.param2; + nctrl->ncmd.s.param1); + rtnl_lock(); + netdev->mtu = nctrl->ncmd.s.param1; + call_netdevice_notifiers(NETDEV_CHANGEMTU, netdev); + rtnl_unlock(); break; case OCTNET_CMD_GPIO_ACCESS: @@ -2181,11 +2397,79 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) netdev->name); break; + case OCTNET_CMD_ENABLE_VLAN_FILTER: + dev_info(&oct->pci_dev->dev, "%s VLAN filter enabled\n", + netdev->name); + break; + + case OCTNET_CMD_ADD_VLAN_FILTER: + dev_info(&oct->pci_dev->dev, "%s VLAN filter %d added\n", + netdev->name, nctrl->ncmd.s.param1); + break; + + case OCTNET_CMD_DEL_VLAN_FILTER: + dev_info(&oct->pci_dev->dev, "%s VLAN filter %d removed\n", + netdev->name, nctrl->ncmd.s.param1); + break; + case OCTNET_CMD_SET_SETTINGS: dev_info(&oct->pci_dev->dev, "%s settings changed\n", netdev->name); break; + /* Case to handle "OCTNET_CMD_TNL_RX_CSUM_CTL" + * Command passed by NIC driver + */ + case OCTNET_CMD_TNL_RX_CSUM_CTL: + if (nctrl->ncmd.s.param1 == OCTNET_CMD_RXCSUM_ENABLE) { + netif_info(lio, probe, lio->netdev, + "%s RX Checksum Offload Enabled\n", + netdev->name); + } else if (nctrl->ncmd.s.param1 == + OCTNET_CMD_RXCSUM_DISABLE) { + netif_info(lio, probe, lio->netdev, + "%s RX Checksum Offload Disabled\n", + netdev->name); + } + break; + + /* Case to handle "OCTNET_CMD_TNL_TX_CSUM_CTL" + * Command passed by NIC driver + */ + case OCTNET_CMD_TNL_TX_CSUM_CTL: + if (nctrl->ncmd.s.param1 == OCTNET_CMD_TXCSUM_ENABLE) { + netif_info(lio, probe, lio->netdev, + "%s TX Checksum Offload Enabled\n", + netdev->name); + } else if (nctrl->ncmd.s.param1 == + OCTNET_CMD_TXCSUM_DISABLE) { + netif_info(lio, probe, lio->netdev, + "%s TX Checksum Offload Disabled\n", + netdev->name); + } + break; + + /* Case to handle "OCTNET_CMD_VXLAN_PORT_CONFIG" + * Command passed by NIC driver + */ + case OCTNET_CMD_VXLAN_PORT_CONFIG: + if (nctrl->ncmd.s.more == OCTNET_CMD_VXLAN_PORT_ADD) { + netif_info(lio, probe, lio->netdev, + "%s VxLAN Destination UDP PORT:%d ADDED\n", + netdev->name, + nctrl->ncmd.s.param1); + } else if (nctrl->ncmd.s.more == + OCTNET_CMD_VXLAN_PORT_DEL) { + netif_info(lio, probe, lio->netdev, + "%s VxLAN Destination UDP PORT:%d DELETED\n", + netdev->name, + nctrl->ncmd.s.param1); + } + break; + + case OCTNET_CMD_SET_FLOW_CTL: + netif_info(lio, probe, lio->netdev, "Set RX/TX flow control parameters\n"); + break; default: dev_err(&oct->pci_dev->dev, "%s Unknown cmd %d\n", __func__, @@ -2235,10 +2519,9 @@ static void liquidio_set_mcast_list(struct net_device *netdev) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; struct netdev_hw_addr *ha; u64 *mc; - int ret, i; + int ret; int mc_count = min(netdev_mc_count(netdev), MAX_OCTEON_MULTICAST_ADDR); memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); @@ -2246,15 +2529,14 @@ static void liquidio_set_mcast_list(struct net_device *netdev) /* Create a ctrl pkt command to be sent to core app. */ nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_SET_MULTI_LIST; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = get_new_flags(netdev); - nctrl.ncmd.s.param3 = mc_count; + nctrl.ncmd.s.param1 = get_new_flags(netdev); + nctrl.ncmd.s.param2 = mc_count; nctrl.ncmd.s.more = mc_count; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; /* copy all the addresses into the udd */ - i = 0; mc = &nctrl.udd[0]; netdev_for_each_mc_addr(ha, netdev) { *mc = 0; @@ -2270,9 +2552,7 @@ static void liquidio_set_mcast_list(struct net_device *netdev) */ nctrl.wait_time = 0; - nparams.resp_order = OCTEON_RESP_NORESPONSE; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "DEVFLAGS change failed in core (ret: 0x%x)\n", ret); @@ -2290,19 +2570,17 @@ static int liquidio_set_mac(struct net_device *netdev, void *p) struct octeon_device *oct = lio->oct_dev; struct sockaddr *addr = (struct sockaddr *)p; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; - if ((!is_valid_ether_addr(addr->sa_data)) || - (ifstate_check(lio, LIO_IFSTATE_RUNNING))) + if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MACADDR; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = 0; + nctrl.ncmd.s.param1 = 0; nctrl.ncmd.s.more = 1; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; nctrl.wait_time = 100; @@ -2311,9 +2589,7 @@ static int liquidio_set_mac(struct net_device *netdev, void *p) /* The MAC Address is presented in network byte order. */ memcpy((u8 *)&nctrl.udd[0] + 2, addr->sa_data, ETH_ALEN); - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "MAC Address change failed\n"); return -ENOMEM; @@ -2341,7 +2617,7 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev) oct = lio->oct_dev; for (i = 0; i < lio->linfo.num_txpciq; i++) { - iq_no = lio->linfo.txpciq[i]; + iq_no = lio->linfo.txpciq[i].s.q_no; iq_stats = &oct->instr_queue[iq_no]->stats; pkts += iq_stats->tx_done; drop += iq_stats->tx_dropped; @@ -2357,7 +2633,7 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev) bytes = 0; for (i = 0; i < lio->linfo.num_rxpciq; i++) { - oq_no = lio->linfo.rxpciq[i]; + oq_no = lio->linfo.rxpciq[i].s.q_no; oq_stats = &oct->droq[oq_no]->stats; pkts += oq_stats->rx_pkts_received; drop += (oq_stats->rx_dropped + @@ -2383,19 +2659,16 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; - int max_frm_size = new_mtu + OCTNET_FRM_HEADER_SIZE; int ret = 0; - /* Limit the MTU to make sure the ethernet packets are between 64 bytes - * and 65535 bytes + /* Limit the MTU to make sure the ethernet packets are between 68 bytes + * and 16000 bytes */ - if ((max_frm_size < OCTNET_MIN_FRM_SIZE) || - (max_frm_size > OCTNET_MAX_FRM_SIZE)) { + if ((new_mtu < LIO_MIN_MTU_SIZE) || + (new_mtu > LIO_MAX_MTU_SIZE)) { dev_err(&oct->pci_dev->dev, "Invalid MTU: %d\n", new_mtu); dev_err(&oct->pci_dev->dev, "Valid range %d and %d\n", - (OCTNET_MIN_FRM_SIZE - OCTNET_FRM_HEADER_SIZE), - (OCTNET_MAX_FRM_SIZE - OCTNET_FRM_HEADER_SIZE)); + LIO_MIN_MTU_SIZE, LIO_MAX_MTU_SIZE); return -EINVAL; } @@ -2403,15 +2676,13 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MTU; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = new_mtu; + nctrl.ncmd.s.param1 = new_mtu; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 100; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Failed to set MTU\n"); return -1; @@ -2428,7 +2699,7 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) * @param ifr interface request * @param cmd command */ -static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) { struct hwtstamp_config conf; struct lio *lio = GET_LIO(netdev); @@ -2489,7 +2760,7 @@ static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { switch (cmd) { case SIOCSHWTSTAMP: - return hwtstamp_ioctl(netdev, ifr, cmd); + return hwtstamp_ioctl(netdev, ifr); default: return -EOPNOTSUPP; } @@ -2536,7 +2807,7 @@ static void handle_timestamp(struct octeon_device *oct, } octeon_free_soft_command(oct, sc); - recv_buffer_free(skb); + tx_buffer_free(skb); } /* \brief Send a data packet that will be timestamped @@ -2551,10 +2822,9 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, { int retval; struct octeon_soft_command *sc; - struct octeon_instr_ih *ih; - struct octeon_instr_rdp *rdp; struct lio *lio; int ring_doorbell; + u32 len; lio = finfo->lio; @@ -2576,14 +2846,13 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, sc->callback_arg = finfo->skb; sc->iq_no = ndata->q_no; - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + len = (u32)((struct octeon_instr_ih2 *)(&sc->cmd.cmd2.ih2))->dlengsz; ring_doorbell = !xmit_more; retval = octeon_send_command(oct, sc->iq_no, ring_doorbell, &sc->cmd, - sc, ih->dlengsz, ndata->reqtype); + sc, len, ndata->reqtype); - if (retval) { + if (retval == IQ_SEND_FAILED) { dev_err(&oct->pci_dev->dev, "timestamp data packet failed status: %x\n", retval); octeon_free_soft_command(oct, sc); @@ -2594,68 +2863,6 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, return retval; } -static inline int is_ipv4(struct sk_buff *skb) -{ - return (skb->protocol == htons(ETH_P_IP)) && - (ip_hdr(skb)->version == 4); -} - -static inline int is_vlan(struct sk_buff *skb) -{ - return skb->protocol == htons(ETH_P_8021Q); -} - -static inline int is_ip_fragmented(struct sk_buff *skb) -{ - /* The Don't fragment and Reserved flag fields are ignored. - * IP is fragmented if - * - the More fragments bit is set (indicating this IP is a fragment - * with more to follow; the current offset could be 0 ). - * - ths offset field is non-zero. - */ - return (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) ? 1 : 0; -} - -static inline int is_ipv6(struct sk_buff *skb) -{ - return (skb->protocol == htons(ETH_P_IPV6)) && - (ipv6_hdr(skb)->version == 6); -} - -static inline int is_with_extn_hdr(struct sk_buff *skb) -{ - return (ipv6_hdr(skb)->nexthdr != IPPROTO_TCP) && - (ipv6_hdr(skb)->nexthdr != IPPROTO_UDP); -} - -static inline int is_tcpudp(struct sk_buff *skb) -{ - return (ip_hdr(skb)->protocol == IPPROTO_TCP) || - (ip_hdr(skb)->protocol == IPPROTO_UDP); -} - -static inline u32 get_ipv4_5tuple_tag(struct sk_buff *skb) -{ - u32 tag; - struct iphdr *iphdr = ip_hdr(skb); - - tag = crc32(0, &iphdr->protocol, 1); - tag = crc32(tag, (u8 *)&iphdr->saddr, 8); - tag = crc32(tag, skb_transport_header(skb), 4); - return tag; -} - -static inline u32 get_ipv6_5tuple_tag(struct sk_buff *skb) -{ - u32 tag; - struct ipv6hdr *ipv6hdr = ipv6_hdr(skb); - - tag = crc32(0, &ipv6hdr->nexthdr, 1); - tag = crc32(tag, (u8 *)&ipv6hdr->saddr, 32); - tag = crc32(tag, skb_transport_header(skb), 4); - return tag; -} - /** \brief Transmit networks packets to the Octeon interface * @param skbuff skbuff struct to be passed to network layer. * @param netdev pointer to network device @@ -2670,18 +2877,22 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) struct octnic_data_pkt ndata; struct octeon_device *oct; struct oct_iq_stats *stats; - int cpu = 0, status = 0; + struct octeon_instr_irh *irh; + union tx_info *tx_info; + int status = 0; int q_idx = 0, iq_no = 0; - int xmit_more; + int xmit_more, j; + u64 dptr = 0; u32 tag = 0; lio = GET_LIO(netdev); oct = lio->oct_dev; if (netif_is_multiqueue(netdev)) { - cpu = skb->queue_mapping; - q_idx = (cpu & (lio->linfo.num_txpciq - 1)); - iq_no = lio->linfo.txpciq[q_idx]; + q_idx = skb->queue_mapping; + q_idx = (q_idx % (lio->linfo.num_txpciq)); + tag = q_idx; + iq_no = lio->linfo.txpciq[q_idx].s.q_no; } else { iq_no = lio->txq; } @@ -2692,11 +2903,11 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) * transmitted. */ if (!(atomic_read(&lio->ifstate) & LIO_IFSTATE_RUNNING) || - (!lio->linfo.link.s.status) || + (!lio->linfo.link.s.link_up) || (skb->len <= 0)) { netif_info(lio, tx_err, lio->netdev, "Transmit failed link_status : %d\n", - lio->linfo.link.s.status); + lio->linfo.link.s.link_up); goto lio_xmit_failed; } @@ -2728,62 +2939,25 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) /* defer sending if queue is full */ stats->tx_iq_busy++; netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n", - ndata.q_no); + lio->txq); return NETDEV_TX_BUSY; } } /* pr_info(" XMIT - valid Qs: %d, 1st Q no: %d, cpu: %d, q_no:%d\n", - * lio->linfo.num_txpciq, lio->txq, cpu, ndata.q_no ); + * lio->linfo.num_txpciq, lio->txq, cpu, ndata.q_no); */ ndata.datasize = skb->len; cmdsetup.u64 = 0; - cmdsetup.s.ifidx = lio->linfo.ifidx; + cmdsetup.s.iq_no = iq_no; if (skb->ip_summed == CHECKSUM_PARTIAL) { - if (is_ipv4(skb) && !is_ip_fragmented(skb) && is_tcpudp(skb)) { - tag = get_ipv4_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = sizeof(struct ethhdr) + 1; - - if (ip_hdr(skb)->ihl > 5) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV4OPTS; - - } else if (is_ipv6(skb)) { - tag = get_ipv6_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = sizeof(struct ethhdr) + 1; - - if (is_with_extn_hdr(skb)) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV6EXTHDR; - - } else if (is_vlan(skb)) { - if (vlan_eth_hdr(skb)->h_vlan_encapsulated_proto - == htons(ETH_P_IP) && - !is_ip_fragmented(skb) && is_tcpudp(skb)) { - tag = get_ipv4_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = - sizeof(struct vlan_ethhdr) + 1; - - if (ip_hdr(skb)->ihl > 5) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV4OPTS; - - } else if (vlan_eth_hdr(skb)->h_vlan_encapsulated_proto - == htons(ETH_P_IPV6)) { - tag = get_ipv6_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = - sizeof(struct vlan_ethhdr) + 1; - - if (is_with_extn_hdr(skb)) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV6EXTHDR; - } + if (skb->encapsulation) { + cmdsetup.s.tnl_csum = 1; + stats->tx_vxlan++; + } else { + cmdsetup.s.transport_csum = 1; } } if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) { @@ -2793,20 +2967,21 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) if (skb_shinfo(skb)->nr_frags == 0) { cmdsetup.s.u.datasize = skb->len; - octnet_prepare_pci_cmd(&ndata.cmd, &cmdsetup, tag); + octnet_prepare_pci_cmd(oct, &ndata.cmd, &cmdsetup, tag); + /* Offload checksum calculation for TCP/UDP packets */ - ndata.cmd.dptr = dma_map_single(&oct->pci_dev->dev, - skb->data, - skb->len, - DMA_TO_DEVICE); - if (dma_mapping_error(&oct->pci_dev->dev, ndata.cmd.dptr)) { + dptr = dma_map_single(&oct->pci_dev->dev, + skb->data, + skb->len, + DMA_TO_DEVICE); + if (dma_mapping_error(&oct->pci_dev->dev, dptr)) { dev_err(&oct->pci_dev->dev, "%s DMA mapping error 1\n", __func__); return NETDEV_TX_BUSY; } - finfo->dptr = ndata.cmd.dptr; - + ndata.cmd.cmd2.dptr = dptr; + finfo->dptr = dptr; ndata.reqtype = REQTYPE_NORESP_NET; } else { @@ -2814,9 +2989,10 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) struct skb_frag_struct *frag; struct octnic_gather *g; - spin_lock(&lio->lock); - g = (struct octnic_gather *)list_delete_head(&lio->glist); - spin_unlock(&lio->lock); + spin_lock(&lio->glist_lock[q_idx]); + g = (struct octnic_gather *) + list_delete_head(&lio->glist[q_idx]); + spin_unlock(&lio->glist_lock[q_idx]); if (!g) { netif_info(lio, tx_err, lio->netdev, @@ -2826,7 +3002,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) cmdsetup.s.gather = 1; cmdsetup.s.u.gatherptrs = (skb_shinfo(skb)->nr_frags + 1); - octnet_prepare_pci_cmd(&ndata.cmd, &cmdsetup, tag); + octnet_prepare_pci_cmd(oct, &ndata.cmd, &cmdsetup, tag); memset(g->sg, 0, g->sg_size); @@ -2853,36 +3029,52 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) frag->size, DMA_TO_DEVICE); + if (dma_mapping_error(&oct->pci_dev->dev, + g->sg[i >> 2].ptr[i & 3])) { + dma_unmap_single(&oct->pci_dev->dev, + g->sg[0].ptr[0], + skb->len - skb->data_len, + DMA_TO_DEVICE); + for (j = 1; j < i; j++) { + frag = &skb_shinfo(skb)->frags[j - 1]; + dma_unmap_page(&oct->pci_dev->dev, + g->sg[j >> 2].ptr[j & 3], + frag->size, + DMA_TO_DEVICE); + } + dev_err(&oct->pci_dev->dev, "%s DMA mapping error 3\n", + __func__); + return NETDEV_TX_BUSY; + } + add_sg_size(&g->sg[(i >> 2)], frag->size, (i & 3)); i++; } - ndata.cmd.dptr = dma_map_single(&oct->pci_dev->dev, - g->sg, g->sg_size, - DMA_TO_DEVICE); - if (dma_mapping_error(&oct->pci_dev->dev, ndata.cmd.dptr)) { - dev_err(&oct->pci_dev->dev, "%s DMA mapping error 3\n", - __func__); - dma_unmap_single(&oct->pci_dev->dev, g->sg[0].ptr[0], - skb->len - skb->data_len, - DMA_TO_DEVICE); - return NETDEV_TX_BUSY; - } + dma_sync_single_for_device(&oct->pci_dev->dev, g->sg_dma_ptr, + g->sg_size, DMA_TO_DEVICE); + dptr = g->sg_dma_ptr; - finfo->dptr = ndata.cmd.dptr; + ndata.cmd.cmd2.dptr = dptr; + finfo->dptr = dptr; finfo->g = g; ndata.reqtype = REQTYPE_NORESP_NET_SG; } - if (skb_shinfo(skb)->gso_size) { - struct octeon_instr_irh *irh = - (struct octeon_instr_irh *)&ndata.cmd.irh; - union tx_info *tx_info = (union tx_info *)&ndata.cmd.ossp[0]; + irh = (struct octeon_instr_irh *)&ndata.cmd.cmd2.irh; + tx_info = (union tx_info *)&ndata.cmd.cmd2.ossp[0]; - irh->len = 1; /* to indicate that ossp[0] contains tx_info */ + if (skb_shinfo(skb)->gso_size) { tx_info->s.gso_size = skb_shinfo(skb)->gso_size; tx_info->s.gso_segs = skb_shinfo(skb)->gso_segs; + stats->tx_gso++; + } + + /* HW insert VLAN tag */ + if (skb_vlan_tag_present(skb)) { + irh->priority = skb_vlan_tag_get(skb) >> 13; + irh->vlan = skb_vlan_tag_get(skb) & 0xfff; } xmit_more = skb->xmit_more; @@ -2901,7 +3093,10 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) netif_trans_update(netdev); - stats->tx_done++; + if (skb_shinfo(skb)->gso_size) + stats->tx_done += skb_shinfo(skb)->gso_segs; + else + stats->tx_done++; stats->tx_tot_bytes += skb->len; return NETDEV_TX_OK; @@ -2910,9 +3105,10 @@ lio_xmit_failed: stats->tx_dropped++; netif_info(lio, tx_err, lio->netdev, "IQ%d Transmit dropped:%llu\n", iq_no, stats->tx_dropped); - dma_unmap_single(&oct->pci_dev->dev, ndata.cmd.dptr, - ndata.datasize, DMA_TO_DEVICE); - recv_buffer_free(skb); + if (dptr) + dma_unmap_single(&oct->pci_dev->dev, dptr, + ndata.datasize, DMA_TO_DEVICE); + tx_buffer_free(skb); return NETDEV_TX_OK; } @@ -2932,27 +3128,145 @@ static void liquidio_tx_timeout(struct net_device *netdev) txqs_wake(netdev); } -int liquidio_set_feature(struct net_device *netdev, int cmd) +static int liquidio_vlan_rx_add_vid(struct net_device *netdev, + __be16 proto __attribute__((unused)), + u16 vid) { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int ret = 0; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.u64 = 0; - nctrl.ncmd.s.cmd = cmd; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = OCTNIC_LROIPV4 | OCTNIC_LROIPV6; + nctrl.ncmd.s.cmd = OCTNET_CMD_ADD_VLAN_FILTER; + nctrl.ncmd.s.param1 = vid; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n", + ret); + } + + return ret; +} + +static int liquidio_vlan_rx_kill_vid(struct net_device *netdev, + __be16 proto __attribute__((unused)), + u16 vid) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = OCTNET_CMD_DEL_VLAN_FILTER; + nctrl.ncmd.s.param1 = vid; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n", + ret); + } + return ret; +} + +/** Sending command to enable/disable RX checksum offload + * @param netdev pointer to network device + * @param command OCTNET_CMD_TNL_RX_CSUM_CTL + * @param rx_cmd_bit OCTNET_CMD_RXCSUM_ENABLE/ + * OCTNET_CMD_RXCSUM_DISABLE + * @returns SUCCESS or FAILURE + */ +int liquidio_set_rxcsum_command(struct net_device *netdev, int command, + u8 rx_cmd) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = command; + nctrl.ncmd.s.param1 = rx_cmd; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 100; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - nparams.resp_order = OCTEON_RESP_NORESPONSE; + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, + "DEVFLAGS RXCSUM change failed in core(ret:0x%x)\n", + ret); + } + return ret; +} + +/** Sending command to add/delete VxLAN UDP port to firmware + * @param netdev pointer to network device + * @param command OCTNET_CMD_VXLAN_PORT_CONFIG + * @param vxlan_port VxLAN port to be added or deleted + * @param vxlan_cmd_bit OCTNET_CMD_VXLAN_PORT_ADD, + * OCTNET_CMD_VXLAN_PORT_DEL + * @returns SUCCESS or FAILURE + */ +static int liquidio_vxlan_port_command(struct net_device *netdev, int command, + u16 vxlan_port, u8 vxlan_cmd_bit) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = command; + nctrl.ncmd.s.more = vxlan_cmd_bit; + nctrl.ncmd.s.param1 = vxlan_port; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, + "VxLAN port add/delete failed in core (ret:0x%x)\n", + ret); + } + return ret; +} + +int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = cmd; + nctrl.ncmd.s.param1 = param1; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Feature change failed in core (ret: 0x%x)\n", ret); @@ -3008,14 +3322,55 @@ static int liquidio_set_features(struct net_device *netdev, return 0; if ((features & NETIF_F_LRO) && (lio->dev_capability & NETIF_F_LRO)) - liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE); + liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); else if (!(features & NETIF_F_LRO) && (lio->dev_capability & NETIF_F_LRO)) - liquidio_set_feature(netdev, OCTNET_CMD_LRO_DISABLE); + liquidio_set_feature(netdev, OCTNET_CMD_LRO_DISABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); + + /* Sending command to firmware to enable/disable RX checksum + * offload settings using ethtool + */ + if (!(netdev->features & NETIF_F_RXCSUM) && + (lio->enc_dev_capability & NETIF_F_RXCSUM) && + (features & NETIF_F_RXCSUM)) + liquidio_set_rxcsum_command(netdev, + OCTNET_CMD_TNL_RX_CSUM_CTL, + OCTNET_CMD_RXCSUM_ENABLE); + else if ((netdev->features & NETIF_F_RXCSUM) && + (lio->enc_dev_capability & NETIF_F_RXCSUM) && + !(features & NETIF_F_RXCSUM)) + liquidio_set_rxcsum_command(netdev, OCTNET_CMD_TNL_RX_CSUM_CTL, + OCTNET_CMD_RXCSUM_DISABLE); return 0; } +static void liquidio_add_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) +{ + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + liquidio_vxlan_port_command(netdev, + OCTNET_CMD_VXLAN_PORT_CONFIG, + htons(ti->port), + OCTNET_CMD_VXLAN_PORT_ADD); +} + +static void liquidio_del_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) +{ + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + liquidio_vxlan_port_command(netdev, + OCTNET_CMD_VXLAN_PORT_CONFIG, + htons(ti->port), + OCTNET_CMD_VXLAN_PORT_DEL); +} + static struct net_device_ops lionetdevops = { .ndo_open = liquidio_open, .ndo_stop = liquidio_stop, @@ -3024,10 +3379,15 @@ static struct net_device_ops lionetdevops = { .ndo_set_mac_address = liquidio_set_mac, .ndo_set_rx_mode = liquidio_set_mcast_list, .ndo_tx_timeout = liquidio_tx_timeout, + + .ndo_vlan_rx_add_vid = liquidio_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = liquidio_vlan_rx_kill_vid, .ndo_change_mtu = liquidio_change_mtu, .ndo_do_ioctl = liquidio_ioctl, .ndo_fix_features = liquidio_fix_features, .ndo_set_features = liquidio_set_features, + .ndo_udp_tunnel_add = liquidio_add_vxlan_port, + .ndo_udp_tunnel_del = liquidio_del_vxlan_port, }; /** \brief Entry point for the liquidio module @@ -3082,24 +3442,27 @@ static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf) { struct octeon_device *oct = (struct octeon_device *)buf; struct octeon_recv_pkt *recv_pkt = recv_info->recv_pkt; - int ifidx = 0; + int gmxport = 0; union oct_link_status *ls; int i; - if ((recv_pkt->buffer_size[0] != sizeof(*ls)) || - (recv_pkt->rh.r_nic_info.ifidx > oct->ifcount)) { + if (recv_pkt->buffer_size[0] != sizeof(*ls)) { dev_err(&oct->pci_dev->dev, "Malformed NIC_INFO, len=%d, ifidx=%d\n", recv_pkt->buffer_size[0], - recv_pkt->rh.r_nic_info.ifidx); + recv_pkt->rh.r_nic_info.gmxport); goto nic_info_err; } - ifidx = recv_pkt->rh.r_nic_info.ifidx; + gmxport = recv_pkt->rh.r_nic_info.gmxport; ls = (union oct_link_status *)get_rbd(recv_pkt->buffer_ptr[0]); octeon_swap_8B_data((u64 *)ls, (sizeof(union oct_link_status)) >> 3); - - update_link_status(oct->props[ifidx].netdev, ls); + for (i = 0; i < oct->ifcount; i++) { + if (oct->props[i].gmxport == gmxport) { + update_link_status(oct->props[i].netdev, ls); + break; + } + } nic_info_err: for (i = 0; i < recv_pkt->buffer_count; i++) @@ -3125,13 +3488,12 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) struct liquidio_if_cfg_context *ctx; struct liquidio_if_cfg_resp *resp; struct octdev_props *props; - int retval, num_iqueues, num_oqueues, q_no; - u64 q_mask; - int num_cpus = num_online_cpus(); + int retval, num_iqueues, num_oqueues; union oct_nic_if_cfg if_cfg; unsigned int base_queue; unsigned int gmx_port_id; u32 resp_size, ctx_size; + u32 ifidx_or_pfnum; /* This is to handle link status changes */ octeon_register_dispatch_fn(octeon_dev, OPCODE_NIC, @@ -3167,14 +3529,12 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) CFG_GET_BASE_QUE_NIC_IF(octeon_get_conf(octeon_dev), i); gmx_port_id = CFG_GET_GMXID_NIC_IF(octeon_get_conf(octeon_dev), i); - if (num_iqueues > num_cpus) - num_iqueues = num_cpus; - if (num_oqueues > num_cpus) - num_oqueues = num_cpus; + ifidx_or_pfnum = i; + dev_dbg(&octeon_dev->pci_dev->dev, "requesting config for interface %d, iqs %d, oqs %d\n", - i, num_iqueues, num_oqueues); - ACCESS_ONCE(ctx->cond) = 0; + ifidx_or_pfnum, num_iqueues, num_oqueues); + WRITE_ONCE(ctx->cond, 0); ctx->octeon_id = lio_get_device_id(octeon_dev); init_waitqueue_head(&ctx->wc); @@ -3183,16 +3543,19 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) if_cfg.s.num_oqueues = num_oqueues; if_cfg.s.base_queue = base_queue; if_cfg.s.gmx_port_id = gmx_port_id; + + sc->iq_no = 0; + octeon_prepare_soft_command(octeon_dev, sc, OPCODE_NIC, - OPCODE_NIC_IF_CFG, i, + OPCODE_NIC_IF_CFG, 0, if_cfg.u64, 0); sc->callback = if_cfg_callback; sc->callback_arg = sc; - sc->wait_time = 1000; + sc->wait_time = 3000; retval = octeon_send_soft_command(octeon_dev, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { dev_err(&octeon_dev->pci_dev->dev, "iq/oq config failed status: %x\n", retval); @@ -3234,8 +3597,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) goto setup_nic_dev_fail; } - props = &octeon_dev->props[i]; - props->netdev = netdev; + SET_NETDEV_DEV(netdev, &octeon_dev->pci_dev->dev); if (num_iqueues > 1) lionetdevops.ndo_select_queue = select_q; @@ -3249,23 +3611,21 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) memset(lio, 0, sizeof(struct lio)); - lio->linfo.ifidx = resp->cfg_info.ifidx; - lio->ifidx = resp->cfg_info.ifidx; + lio->ifidx = ifidx_or_pfnum; + + props = &octeon_dev->props[i]; + props->gmxport = resp->cfg_info.linfo.gmxport; + props->netdev = netdev; lio->linfo.num_rxpciq = num_oqueues; lio->linfo.num_txpciq = num_iqueues; - q_mask = resp->cfg_info.oqmask; - /* q_mask is 0-based and already verified mask is nonzero */ for (j = 0; j < num_oqueues; j++) { - q_no = __ffs64(q_mask); - q_mask &= (~(1UL << q_no)); - lio->linfo.rxpciq[j] = q_no; + lio->linfo.rxpciq[j].u64 = + resp->cfg_info.linfo.rxpciq[j].u64; } - q_mask = resp->cfg_info.iqmask; for (j = 0; j < num_iqueues; j++) { - q_no = __ffs64(q_mask); - q_mask &= (~(1UL << q_no)); - lio->linfo.txpciq[j] = q_no; + lio->linfo.txpciq[j].u64 = + resp->cfg_info.linfo.txpciq[j].u64; } lio->linfo.hw_addr = resp->cfg_info.linfo.hw_addr; lio->linfo.gmxport = resp->cfg_info.linfo.gmxport; @@ -3274,16 +3634,41 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) lio->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); lio->dev_capability = NETIF_F_HIGHDMA - | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM - | NETIF_F_SG | NETIF_F_RXCSUM - | NETIF_F_TSO | NETIF_F_TSO6 - | NETIF_F_LRO; + | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM + | NETIF_F_SG | NETIF_F_RXCSUM + | NETIF_F_GRO + | NETIF_F_TSO | NETIF_F_TSO6 + | NETIF_F_LRO; netif_set_gso_max_size(netdev, OCTNIC_GSO_MAX_SIZE); - netdev->features = lio->dev_capability; + /* Copy of transmit encapsulation capabilities: + * TSO, TSO6, Checksums for this device + */ + lio->enc_dev_capability = NETIF_F_IP_CSUM + | NETIF_F_IPV6_CSUM + | NETIF_F_GSO_UDP_TUNNEL + | NETIF_F_HW_CSUM | NETIF_F_SG + | NETIF_F_RXCSUM + | NETIF_F_TSO | NETIF_F_TSO6 + | NETIF_F_LRO; + + netdev->hw_enc_features = (lio->enc_dev_capability & + ~NETIF_F_LRO); + + lio->dev_capability |= NETIF_F_GSO_UDP_TUNNEL; + netdev->vlan_features = lio->dev_capability; + /* Add any unchangeable hw features */ + lio->dev_capability |= NETIF_F_HW_VLAN_CTAG_FILTER | + NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_CTAG_TX; + + netdev->features = (lio->dev_capability & ~NETIF_F_LRO); netdev->hw_features = lio->dev_capability; + /*HW_VLAN_RX and HW_VLAN_FILTER is always on*/ + netdev->hw_features = netdev->hw_features & + ~NETIF_F_HW_VLAN_CTAG_RX; /* Point to the properties for octeon device to which this * interface belongs. @@ -3291,7 +3676,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) lio->oct_dev = octeon_dev; lio->octprops = props; lio->netdev = netdev; - spin_lock_init(&lio->lock); dev_dbg(&octeon_dev->pci_dev->dev, "if%d gmx: %d hw_addr: 0x%llx\n", i, @@ -3306,23 +3690,22 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) ether_addr_copy(netdev->dev_addr, mac); - if (setup_io_queues(octeon_dev, netdev)) { + /* By default all interfaces on a single Octeon uses the same + * tx and rx queues + */ + lio->txq = lio->linfo.txpciq[0].s.q_no; + lio->rxq = lio->linfo.rxpciq[0].s.q_no; + if (setup_io_queues(octeon_dev, i)) { dev_err(&octeon_dev->pci_dev->dev, "I/O queues creation failed\n"); goto setup_nic_dev_fail; } ifstate_set(lio, LIO_IFSTATE_DROQ_OPS); - /* By default all interfaces on a single Octeon uses the same - * tx and rx queues - */ - lio->txq = lio->linfo.txpciq[0]; - lio->rxq = lio->linfo.rxpciq[0]; - lio->tx_qsize = octeon_get_tx_qsize(octeon_dev, lio->txq); lio->rx_qsize = octeon_get_rx_qsize(octeon_dev, lio->rxq); - if (setup_glist(lio)) { + if (setup_glists(octeon_dev, lio, num_iqueues)) { dev_err(&octeon_dev->pci_dev->dev, "Gather list allocation failed\n"); goto setup_nic_dev_fail; @@ -3330,11 +3713,17 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) /* Register ethtool support */ liquidio_set_ethtool_ops(netdev); + octeon_dev->priv_flags = 0x0; - liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE); + if (netdev->features & NETIF_F_LRO) + liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); + + liquidio_set_feature(netdev, OCTNET_CMD_ENABLE_VLAN_FILTER, 0); if ((debug != -1) && (debug & NETIF_MSG_HW)) - liquidio_set_feature(netdev, OCTNET_CMD_VERBOSE_ENABLE); + liquidio_set_feature(netdev, + OCTNET_CMD_VERBOSE_ENABLE, 0); /* Register the network device with the OS */ if (register_netdev(netdev)) { @@ -3346,16 +3735,19 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) "Setup NIC ifidx:%d mac:%02x%02x%02x%02x%02x%02x\n", i, mac[0], mac[1], mac[2], mac[3], mac[4], mac[5]); netif_carrier_off(netdev); - - if (lio->linfo.link.s.status) { - netif_carrier_on(netdev); - start_txq(netdev); - } else { - netif_carrier_off(netdev); - } + lio->link_changes++; ifstate_set(lio, LIO_IFSTATE_REGISTERED); + /* Sending command to firmware to enable Rx checksum offload + * by default at the time of setup of Liquidio driver for + * this device + */ + liquidio_set_rxcsum_command(netdev, OCTNET_CMD_TNL_RX_CSUM_CTL, + OCTNET_CMD_RXCSUM_ENABLE); + liquidio_set_feature(netdev, OCTNET_CMD_TNL_TX_CSUM_CTL, + OCTNET_CMD_TXCSUM_ENABLE); + dev_dbg(&octeon_dev->pci_dev->dev, "NIC ifidx:%d Setup successful\n", i); @@ -3386,7 +3778,7 @@ setup_nic_dev_fail: static int liquidio_init_nic_module(struct octeon_device *oct) { struct oct_intrmod_cfg *intrmod_cfg; - int retval = 0; + int i, retval = 0; int num_nic_ports = CFG_GET_NUM_NIC_PORTS(octeon_get_conf(oct)); dev_dbg(&oct->pci_dev->dev, "Initializing network interfaces\n"); @@ -3400,6 +3792,9 @@ static int liquidio_init_nic_module(struct octeon_device *oct) memset(oct->props, 0, sizeof(struct octdev_props) * num_nic_ports); + for (i = 0; i < MAX_OCTEON_LINKS; i++) + oct->props[i].gmxport = -1; + retval = setup_nic_devices(oct); if (retval) { dev_err(&oct->pci_dev->dev, "Setup NIC devices failed\n"); @@ -3410,15 +3805,19 @@ static int liquidio_init_nic_module(struct octeon_device *oct) /* Initialize interrupt moderation params */ intrmod_cfg = &((struct octeon_device *)oct)->intrmod; - intrmod_cfg->intrmod_enable = 1; - intrmod_cfg->intrmod_check_intrvl = LIO_INTRMOD_CHECK_INTERVAL; - intrmod_cfg->intrmod_maxpkt_ratethr = LIO_INTRMOD_MAXPKT_RATETHR; - intrmod_cfg->intrmod_minpkt_ratethr = LIO_INTRMOD_MINPKT_RATETHR; - intrmod_cfg->intrmod_maxcnt_trigger = LIO_INTRMOD_MAXCNT_TRIGGER; - intrmod_cfg->intrmod_maxtmr_trigger = LIO_INTRMOD_MAXTMR_TRIGGER; - intrmod_cfg->intrmod_mintmr_trigger = LIO_INTRMOD_MINTMR_TRIGGER; - intrmod_cfg->intrmod_mincnt_trigger = LIO_INTRMOD_MINCNT_TRIGGER; - + intrmod_cfg->rx_enable = 1; + intrmod_cfg->check_intrvl = LIO_INTRMOD_CHECK_INTERVAL; + intrmod_cfg->maxpkt_ratethr = LIO_INTRMOD_MAXPKT_RATETHR; + intrmod_cfg->minpkt_ratethr = LIO_INTRMOD_MINPKT_RATETHR; + intrmod_cfg->rx_maxcnt_trigger = LIO_INTRMOD_RXMAXCNT_TRIGGER; + intrmod_cfg->rx_maxtmr_trigger = LIO_INTRMOD_RXMAXTMR_TRIGGER; + intrmod_cfg->rx_mintmr_trigger = LIO_INTRMOD_RXMINTMR_TRIGGER; + intrmod_cfg->rx_mincnt_trigger = LIO_INTRMOD_RXMINCNT_TRIGGER; + intrmod_cfg->tx_enable = 1; + intrmod_cfg->tx_maxcnt_trigger = LIO_INTRMOD_TXMAXCNT_TRIGGER; + intrmod_cfg->tx_mincnt_trigger = LIO_INTRMOD_TXMINCNT_TRIGGER; + intrmod_cfg->rx_frames = CFG_GET_OQ_INTR_PKT(octeon_get_conf(oct)); + intrmod_cfg->rx_usecs = CFG_GET_OQ_INTR_TIME(octeon_get_conf(oct)); dev_dbg(&oct->pci_dev->dev, "Network interfaces ready\n"); return retval; @@ -3481,6 +3880,7 @@ static void nic_starter(struct work_struct *work) static int octeon_device_init(struct octeon_device *octeon_dev) { int j, ret; + char bootcmd[] = "\n"; struct octeon_device_priv *oct_priv = (struct octeon_device_priv *)octeon_dev->priv; atomic_set(&octeon_dev->status, OCT_DEV_BEGIN_STATE); @@ -3558,6 +3958,7 @@ static int octeon_device_init(struct octeon_device *octeon_dev) /* Release any previously allocated queues */ for (j = 0; j < octeon_dev->num_oqs; j++) octeon_delete_droq(octeon_dev, j); + return 1; } atomic_set(&octeon_dev->status, OCT_DEV_DROQ_INIT_DONE); @@ -3580,7 +3981,8 @@ static int octeon_device_init(struct octeon_device *octeon_dev) /* Setup the interrupt handler and record the INT SUM register address */ - octeon_setup_interrupt(octeon_dev); + if (octeon_setup_interrupt(octeon_dev)) + return 1; /* Enable Octeon device interrupts */ octeon_dev->fn_list.enable_interrupt(octeon_dev->chip); @@ -3592,14 +3994,19 @@ static int octeon_device_init(struct octeon_device *octeon_dev) dev_dbg(&octeon_dev->pci_dev->dev, "Waiting for DDR initialization...\n"); - if (ddr_timeout == 0) { - dev_info(&octeon_dev->pci_dev->dev, - "WAITING. Set ddr_timeout to non-zero value to proceed with initialization.\n"); - } + if (ddr_timeout == 0) + dev_info(&octeon_dev->pci_dev->dev, "WAITING. Set ddr_timeout to non-zero value to proceed with initialization.\n"); schedule_timeout_uninterruptible(HZ * LIO_RESET_SECS); /* Wait for the octeon to initialize DDR after the soft-reset. */ + while (ddr_timeout == 0) { + set_current_state(TASK_INTERRUPTIBLE); + if (schedule_timeout(HZ / 10)) { + /* user probably pressed Control-C */ + return 1; + } + } ret = octeon_wait_for_ddr_init(octeon_dev, &ddr_timeout); if (ret) { dev_err(&octeon_dev->pci_dev->dev, @@ -3613,6 +4020,9 @@ static int octeon_device_init(struct octeon_device *octeon_dev) return 1; } + /* Divert uboot to take commands from host instead. */ + ret = octeon_console_send_cmd(octeon_dev, bootcmd, 50); + dev_dbg(&octeon_dev->pci_dev->dev, "Initializing consoles\n"); ret = octeon_init_consoles(octeon_dev); if (ret) { diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 0ac347ccc8ba..199a8b9c7dc5 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -30,10 +30,10 @@ #include "octeon_config.h" -#define LIQUIDIO_VERSION "1.1.9" -#define LIQUIDIO_MAJOR_VERSION 1 -#define LIQUIDIO_MINOR_VERSION 1 -#define LIQUIDIO_MICRO_VERSION 9 +#define LIQUIDIO_BASE_VERSION "1.4" +#define LIQUIDIO_MICRO_VERSION ".1" +#define LIQUIDIO_PACKAGE "" +#define LIQUIDIO_VERSION "1.4.1" #define CONTROL_IQ 0 /** Tag types used by Octeon cores in its work. */ @@ -174,9 +174,11 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry, /*------------------------- End Scatter/Gather ---------------------------*/ #define OCTNET_FRM_PTP_HEADER_SIZE 8 -#define OCTNET_FRM_HEADER_SIZE 30 /* PTP timestamp + VLAN + Ethernet */ -#define OCTNET_MIN_FRM_SIZE (64 + OCTNET_FRM_PTP_HEADER_SIZE) +#define OCTNET_FRM_HEADER_SIZE 22 /* VLAN + Ethernet */ + +#define OCTNET_MIN_FRM_SIZE 64 + #define OCTNET_MAX_FRM_SIZE (16000 + OCTNET_FRM_HEADER_SIZE) #define OCTNET_DEFAULT_FRM_SIZE (1500 + OCTNET_FRM_HEADER_SIZE) @@ -212,6 +214,17 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry, #define OCTNET_CMD_VERBOSE_ENABLE 0x14 #define OCTNET_CMD_VERBOSE_DISABLE 0x15 +#define OCTNET_CMD_ENABLE_VLAN_FILTER 0x16 +#define OCTNET_CMD_ADD_VLAN_FILTER 0x17 +#define OCTNET_CMD_DEL_VLAN_FILTER 0x18 +#define OCTNET_CMD_VXLAN_PORT_CONFIG 0x19 +#define OCTNET_CMD_VXLAN_PORT_ADD 0x0 +#define OCTNET_CMD_VXLAN_PORT_DEL 0x1 +#define OCTNET_CMD_RXCSUM_ENABLE 0x0 +#define OCTNET_CMD_RXCSUM_DISABLE 0x1 +#define OCTNET_CMD_TXCSUM_ENABLE 0x0 +#define OCTNET_CMD_TXCSUM_DISABLE 0x1 + /* RX(packets coming from wire) Checksum verification flags */ /* TCP/UDP csum */ #define CNNIC_L4SUM_VERIFIED 0x1 @@ -258,19 +271,19 @@ union octnet_cmd { u64 more:6; /* How many udd words follow the command */ - u64 param1:29; + u64 reserved:29; - u64 param2:16; + u64 param1:16; - u64 param3:8; + u64 param2:8; #else - u64 param3:8; + u64 param2:8; - u64 param2:16; + u64 param1:16; - u64 param1:29; + u64 reserved:29; u64 more:6; @@ -283,8 +296,140 @@ union octnet_cmd { #define OCTNET_CMD_SIZE (sizeof(union octnet_cmd)) +/* Instruction Header(DPI) - for OCTEON-III models */ +struct octeon_instr_ih3 { +#ifdef __BIG_ENDIAN_BITFIELD + + /** Reserved3 */ + u64 reserved3:1; + + /** Gather indicator 1=gather*/ + u64 gather:1; + + /** Data length OR no. of entries in gather list */ + u64 dlengsz:14; + + /** Front Data size */ + u64 fsz:6; + + /** Reserved2 */ + u64 reserved2:4; + + /** PKI port kind - PKIND */ + u64 pkind:6; + + /** Reserved1 */ + u64 reserved1:32; + +#else + /** Reserved1 */ + u64 reserved1:32; + + /** PKI port kind - PKIND */ + u64 pkind:6; + + /** Reserved2 */ + u64 reserved2:4; + + /** Front Data size */ + u64 fsz:6; + + /** Data length OR no. of entries in gather list */ + u64 dlengsz:14; + + /** Gather indicator 1=gather*/ + u64 gather:1; + + /** Reserved3 */ + u64 reserved3:1; + +#endif +}; + +/* Optional PKI Instruction Header(PKI IH) - for OCTEON-III models */ +/** BIG ENDIAN format. */ +struct octeon_instr_pki_ih3 { +#ifdef __BIG_ENDIAN_BITFIELD + + /** Wider bit */ + u64 w:1; + + /** Raw mode indicator 1 = RAW */ + u64 raw:1; + + /** Use Tag */ + u64 utag:1; + + /** Use QPG */ + u64 uqpg:1; + + /** Reserved2 */ + u64 reserved2:1; + + /** Parse Mode */ + u64 pm:3; + + /** Skip Length */ + u64 sl:8; + + /** Use Tag Type */ + u64 utt:1; + + /** Tag type */ + u64 tagtype:2; + + /** Reserved1 */ + u64 reserved1:2; + + /** QPG Value */ + u64 qpg:11; + + /** Tag Value */ + u64 tag:32; + +#else + + /** Tag Value */ + u64 tag:32; + + /** QPG Value */ + u64 qpg:11; + + /** Reserved1 */ + u64 reserved1:2; + + /** Tag type */ + u64 tagtype:2; + + /** Use Tag Type */ + u64 utt:1; + + /** Skip Length */ + u64 sl:8; + + /** Parse Mode */ + u64 pm:3; + + /** Reserved2 */ + u64 reserved2:1; + + /** Use QPG */ + u64 uqpg:1; + + /** Use Tag */ + u64 utag:1; + + /** Raw mode indicator 1 = RAW */ + u64 raw:1; + + /** Wider bit */ + u64 w:1; +#endif + +}; + /** Instruction Header */ -struct octeon_instr_ih { +struct octeon_instr_ih2 { #ifdef __BIG_ENDIAN_BITFIELD /** Raw mode indicator 1 = RAW */ u64 raw:1; @@ -348,15 +493,15 @@ struct octeon_instr_irh { u64 opcode:4; u64 rflag:1; u64 subcode:7; - u64 len:3; - u64 rid:13; - u64 reserved:4; + u64 vlan:12; + u64 priority:3; + u64 reserved:5; u64 ossp:32; /* opcode/subcode specific parameters */ #else u64 ossp:32; /* opcode/subcode specific parameters */ - u64 reserved:4; - u64 rid:13; - u64 len:3; + u64 reserved:5; + u64 priority:3; + u64 vlan:12; u64 subcode:7; u64 rflag:1; u64 opcode:4; @@ -383,75 +528,77 @@ union octeon_rh { struct { u64 opcode:4; u64 subcode:8; - u64 len:3; /** additional 64-bit words */ - u64 rid:13; /** request id in response to pkt sent by host */ - u64 reserved:4; - u64 ossp:32; /** opcode/subcode specific parameters */ + u64 len:3; /** additional 64-bit words */ + u64 reserved:17; + u64 ossp:32; /** opcode/subcode specific parameters */ } r; struct { u64 opcode:4; u64 subcode:8; - u64 len:3; /** additional 64-bit words */ - u64 rid:13; /** request id in response to pkt sent by host */ - u64 extra:24; - u64 link:8; + u64 len:3; /** additional 64-bit words */ + u64 extra:28; + u64 vlan:12; + u64 priority:3; u64 csum_verified:3; /** checksum verified. */ u64 has_hwtstamp:1; /** Has hardware timestamp. 1 = yes. */ + u64 encap_on:1; + u64 has_hash:1; /** Has hash (rth or rss). 1 = yes. */ } r_dh; struct { u64 opcode:4; u64 subcode:8; - u64 len:3; /** additional 64-bit words */ - u64 rid:13; /** request id in response to pkt sent by host */ + u64 len:3; /** additional 64-bit words */ + u64 reserved:11; u64 num_gmx_ports:8; - u64 max_nic_ports:8; + u64 max_nic_ports:10; u64 app_cap_flags:4; - u64 app_mode:16; + u64 app_mode:8; + u64 pkind:8; } r_core_drv_init; struct { u64 opcode:4; u64 subcode:8; u64 len:3; /** additional 64-bit words */ - u64 rid:13; - u64 reserved:4; + u64 reserved:8; u64 extra:25; - u64 ifidx:7; + u64 gmxport:16; } r_nic_info; #else u64 u64; struct { u64 ossp:32; /** opcode/subcode specific parameters */ - u64 reserved:4; - u64 rid:13; /** req id in response to pkt sent by host */ + u64 reserved:17; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; } r; struct { + u64 has_hash:1; /** Has hash (rth or rss). 1 = yes. */ + u64 encap_on:1; u64 has_hwtstamp:1; /** 1 = has hwtstamp */ u64 csum_verified:3; /** checksum verified. */ - u64 link:8; - u64 extra:24; - u64 rid:13; /** req id in response to pkt sent by host */ + u64 priority:3; + u64 vlan:12; + u64 extra:28; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; } r_dh; struct { - u64 app_mode:16; + u64 pkind:8; + u64 app_mode:8; u64 app_cap_flags:4; - u64 max_nic_ports:8; + u64 max_nic_ports:10; u64 num_gmx_ports:8; - u64 rid:13; + u64 reserved:11; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; } r_core_drv_init; struct { - u64 ifidx:7; + u64 gmxport:16; u64 extra:25; - u64 reserved:4; - u64 rid:13; + u64 reserved:8; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; @@ -461,30 +608,25 @@ union octeon_rh { #define OCT_RH_SIZE (sizeof(union octeon_rh)) -#define OCT_PKT_PARAM_IPV4OPTS 1 -#define OCT_PKT_PARAM_IPV6EXTHDR 2 - union octnic_packet_params { u32 u32; struct { #ifdef __BIG_ENDIAN_BITFIELD - u32 reserved:6; + u32 reserved:24; + u32 ip_csum:1; /* Perform IP header checksum(s) */ + /* Perform Outer transport header checksum */ + u32 transport_csum:1; + /* Find tunnel, and perform transport csum. */ u32 tnl_csum:1; - u32 ip_csum:1; - u32 ipv4opts_ipv6exthdr:2; - u32 ipsec_ops:4; - u32 tsflag:1; - u32 csoffset:9; - u32 ifidx:8; + u32 tsflag:1; /* Timestamp this packet */ + u32 ipsec_ops:4; /* IPsec operation */ #else - u32 ifidx:8; - u32 csoffset:9; - u32 tsflag:1; u32 ipsec_ops:4; - u32 ipv4opts_ipv6exthdr:2; - u32 ip_csum:1; + u32 tsflag:1; u32 tnl_csum:1; - u32 reserved:6; + u32 transport_csum:1; + u32 ip_csum:1; + u32 reserved:24; #endif } s; }; @@ -496,56 +638,96 @@ union oct_link_status { struct { #ifdef __BIG_ENDIAN_BITFIELD u64 duplex:8; - u64 status:8; u64 mtu:16; u64 speed:16; + u64 link_up:1; u64 autoneg:1; - u64 interface:4; + u64 if_mode:5; u64 pause:1; - u64 reserved:10; + u64 flashing:1; + u64 reserved:15; #else - u64 reserved:10; + u64 reserved:15; + u64 flashing:1; u64 pause:1; - u64 interface:4; + u64 if_mode:5; u64 autoneg:1; + u64 link_up:1; u64 speed:16; u64 mtu:16; - u64 status:8; u64 duplex:8; #endif } s; }; +/** The txpciq info passed to host from the firmware */ + +union oct_txpciq { + u64 u64; + + struct { +#ifdef __BIG_ENDIAN_BITFIELD + u64 q_no:8; + u64 port:8; + u64 pkind:6; + u64 use_qpg:1; + u64 qpg:11; + u64 reserved:30; +#else + u64 reserved:30; + u64 qpg:11; + u64 use_qpg:1; + u64 pkind:6; + u64 port:8; + u64 q_no:8; +#endif + } s; +}; + +/** The rxpciq info passed to host from the firmware */ + +union oct_rxpciq { + u64 u64; + + struct { +#ifdef __BIG_ENDIAN_BITFIELD + u64 q_no:8; + u64 reserved:56; +#else + u64 reserved:56; + u64 q_no:8; +#endif + } s; +}; + /** Information for a OCTEON ethernet interface shared between core & host. */ struct oct_link_info { union oct_link_status link; u64 hw_addr; #ifdef __BIG_ENDIAN_BITFIELD - u16 gmxport; - u8 rsvd[3]; - u8 num_txpciq; - u8 num_rxpciq; - u8 ifidx; + u64 gmxport:16; + u64 rsvd:32; + u64 num_txpciq:8; + u64 num_rxpciq:8; #else - u8 ifidx; - u8 num_rxpciq; - u8 num_txpciq; - u8 rsvd[3]; - u16 gmxport; + u64 num_rxpciq:8; + u64 num_txpciq:8; + u64 rsvd:32; + u64 gmxport:16; #endif - u8 txpciq[MAX_IOQS_PER_NICIF]; - u8 rxpciq[MAX_IOQS_PER_NICIF]; + union oct_txpciq txpciq[MAX_IOQS_PER_NICIF]; + union oct_rxpciq rxpciq[MAX_IOQS_PER_NICIF]; }; #define OCT_LINK_INFO_SIZE (sizeof(struct oct_link_info)) struct liquidio_if_cfg_info { - u64 ifidx; u64 iqmask; /** mask for IQs enabled for the port */ u64 oqmask; /** mask for OQs enabled for the port */ struct oct_link_info linfo; /** initial link information */ + char liquidio_firmware_version[32]; }; /** Stats for each NIC port in RX direction. */ @@ -570,10 +752,18 @@ struct nic_rx_stats { u64 fw_err_pko; u64 fw_err_link; u64 fw_err_drop; + u64 fw_rx_vxlan; + u64 fw_rx_vxlan_err; + + /* LRO */ u64 fw_lro_pkts; /* Number of packets that are LROed */ u64 fw_lro_octs; /* Number of octets that are LROed */ u64 fw_total_lro; /* Number of LRO packets formed */ u64 fw_lro_aborts; /* Number of times lRO of packet aborted */ + u64 fw_lro_aborts_port; + u64 fw_lro_aborts_seq; + u64 fw_lro_aborts_tsval; + u64 fw_lro_aborts_timer; /* intrmod: packet forward rate */ u64 fwd_rate; }; @@ -597,9 +787,14 @@ struct nic_tx_stats { /* firmware stats */ u64 fw_total_sent; u64 fw_total_fwd; + u64 fw_total_fwd_bytes; u64 fw_err_pko; u64 fw_err_link; u64 fw_err_drop; + u64 fw_err_tso; + u64 fw_tso; /* number of tso requests */ + u64 fw_tso_fwd; /* number of packets segmented in tso */ + u64 fw_tx_vxlan; }; struct oct_link_stats { @@ -630,23 +825,44 @@ struct oct_mdio_cmd { #define OCT_LINK_STATS_SIZE (sizeof(struct oct_link_stats)) +/* intrmod: max. packet rate threshold */ +#define LIO_INTRMOD_MAXPKT_RATETHR 196608 +/* intrmod: min. packet rate threshold */ +#define LIO_INTRMOD_MINPKT_RATETHR 9216 +/* intrmod: max. packets to trigger interrupt */ +#define LIO_INTRMOD_RXMAXCNT_TRIGGER 384 +/* intrmod: min. packets to trigger interrupt */ +#define LIO_INTRMOD_RXMINCNT_TRIGGER 1 +/* intrmod: max. time to trigger interrupt */ +#define LIO_INTRMOD_RXMAXTMR_TRIGGER 128 +/* 66xx:intrmod: min. time to trigger interrupt + * (value of 1 is optimum for TCP_RR) + */ +#define LIO_INTRMOD_RXMINTMR_TRIGGER 1 + +/* intrmod: max. packets to trigger interrupt */ +#define LIO_INTRMOD_TXMAXCNT_TRIGGER 64 +/* intrmod: min. packets to trigger interrupt */ +#define LIO_INTRMOD_TXMINCNT_TRIGGER 0 + +/* intrmod: poll interval in seconds */ #define LIO_INTRMOD_CHECK_INTERVAL 1 -#define LIO_INTRMOD_MAXPKT_RATETHR 196608 /* max pkt rate threshold */ -#define LIO_INTRMOD_MINPKT_RATETHR 9216 /* min pkt rate threshold */ -#define LIO_INTRMOD_MAXCNT_TRIGGER 384 /* max pkts to trigger interrupt */ -#define LIO_INTRMOD_MINCNT_TRIGGER 1 /* min pkts to trigger interrupt */ -#define LIO_INTRMOD_MAXTMR_TRIGGER 128 /* max time to trigger interrupt */ -#define LIO_INTRMOD_MINTMR_TRIGGER 32 /* min time to trigger interrupt */ struct oct_intrmod_cfg { - u64 intrmod_enable; - u64 intrmod_check_intrvl; - u64 intrmod_maxpkt_ratethr; - u64 intrmod_minpkt_ratethr; - u64 intrmod_maxcnt_trigger; - u64 intrmod_maxtmr_trigger; - u64 intrmod_mincnt_trigger; - u64 intrmod_mintmr_trigger; + u64 rx_enable; + u64 tx_enable; + u64 check_intrvl; + u64 maxpkt_ratethr; + u64 minpkt_ratethr; + u64 rx_maxcnt_trigger; + u64 rx_mincnt_trigger; + u64 rx_maxtmr_trigger; + u64 rx_mintmr_trigger; + u64 tx_mincnt_trigger; + u64 tx_maxcnt_trigger; + u64 rx_frames; + u64 tx_frames; + u64 rx_usecs; }; #define BASE_QUEUE_NOT_REQUESTED 65535 @@ -659,9 +875,9 @@ union oct_nic_if_cfg { u64 num_iqueues:16; u64 num_oqueues:16; u64 gmx_port_id:8; - u64 reserved:8; + u64 vf_id:8; #else - u64 reserved:8; + u64 vf_id:8; u64 gmx_port_id:8; u64 num_oqueues:16; u64 num_iqueues:16; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h index 62a8dd5cd3dc..b3396e3a8bab 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h @@ -37,7 +37,7 @@ /* Maximum octeon devices defined as MAX_OCTEON_NICIF to support * multiple(<= MAX_OCTEON_NICIF) Miniports */ -#define MAX_OCTEON_NICIF 32 +#define MAX_OCTEON_NICIF 128 #define MAX_OCTEON_DEVICES MAX_OCTEON_NICIF #define MAX_OCTEON_LINKS MAX_OCTEON_NICIF #define MAX_OCTEON_MULTICAST_ADDR 32 @@ -135,7 +135,7 @@ #define CFG_GET_IS_SLI_BP_ON(cfg) ((cfg)->misc.enable_sli_oq_bp) /* Max IOQs per OCTEON Link */ -#define MAX_IOQS_PER_NICIF 32 +#define MAX_IOQS_PER_NICIF 64 enum lio_card_type { LIO_210SV = 0, /* Two port, 66xx */ @@ -226,7 +226,7 @@ struct octeon_oq_config { */ u64 refill_threshold:16; - /** If set, the Output queue uses info-pointer mode. (Default: 1 ) */ + /** If set, the Output queue uses info-pointer mode. (Default: 1) */ u64 info_ptr:32; /* Max number of OQs available */ @@ -236,7 +236,7 @@ struct octeon_oq_config { /* Max number of OQs available */ u64 max_oqs:8; - /** If set, the Output queue uses info-pointer mode. (Default: 1 ) */ + /** If set, the Output queue uses info-pointer mode. (Default: 1) */ u64 info_ptr:32; /** The number of buffers that were consumed during packet processing by @@ -416,9 +416,11 @@ struct octeon_config { #define DISPATCH_LIST_SIZE BIT(OPCODE_MASK_BITS) /* Maximum number of Octeon Instruction (command) queues */ -#define MAX_OCTEON_INSTR_QUEUES CN6XXX_MAX_INPUT_QUEUES +#define MAX_OCTEON_INSTR_QUEUES(oct) CN6XXX_MAX_INPUT_QUEUES +/* Maximum number of Octeon Output queues */ +#define MAX_OCTEON_OUTPUT_QUEUES(oct) CN6XXX_MAX_OUTPUT_QUEUES -/* Maximum number of Octeon Instruction (command) queues */ -#define MAX_OCTEON_OUTPUT_QUEUES CN6XXX_MAX_OUTPUT_QUEUES +#define MAX_POSSIBLE_OCTEON_INSTR_QUEUES CN6XXX_MAX_INPUT_QUEUES +#define MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES CN6XXX_MAX_OUTPUT_QUEUES #endif /* __OCTEON_CONFIG_H__ */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c index 466147e409c9..bbb50ea66f16 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c @@ -23,27 +23,14 @@ /** * @file octeon_console.c */ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" #include "octeon_mem_ops.h" static void octeon_remote_lock(void); @@ -51,6 +38,8 @@ static void octeon_remote_unlock(void); static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct, const char *name, u32 flags); +static int octeon_console_read(struct octeon_device *oct, u32 console_num, + char *buffer, u32 buf_size); #define MIN(a, b) min((a), (b)) #define CAST_ULL(v) ((u64)(v)) @@ -170,8 +159,8 @@ struct octeon_pci_console_desc { offsetof(struct cvmx_bootmem_desc, field), \ SIZEOF_FIELD(struct cvmx_bootmem_desc, field)) -#define __cvmx_bootmem_lock(flags) -#define __cvmx_bootmem_unlock(flags) +#define __cvmx_bootmem_lock(flags) (flags = flags) +#define __cvmx_bootmem_unlock(flags) (flags = flags) /** * This macro returns a member of the @@ -234,7 +223,7 @@ static void CVMX_BOOTMEM_NAMED_GET_NAME(struct octeon_device *oct, u32 len) { addr += offsetof(struct cvmx_bootmem_named_block_desc, name); - octeon_pci_read_core_mem(oct, addr, str, len); + octeon_pci_read_core_mem(oct, addr, (u8 *)str, len); str[len] = 0; } @@ -323,6 +312,9 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct, if (name && named_size) { char *name_tmp = kmalloc(name_length + 1, GFP_KERNEL); + if (!name_tmp) + break; + CVMX_BOOTMEM_NAMED_GET_NAME(oct, named_addr, name_tmp, name_length); @@ -383,7 +375,7 @@ static void octeon_remote_unlock(void) int octeon_console_send_cmd(struct octeon_device *oct, char *cmd_str, u32 wait_hundredths) { - u32 len = strlen(cmd_str); + u32 len = (u32)strlen(cmd_str); dev_dbg(&oct->pci_dev->dev, "sending \"%s\" to bootloader\n", cmd_str); @@ -440,8 +432,7 @@ int octeon_wait_for_bootloader(struct octeon_device *oct, } static void octeon_console_handle_result(struct octeon_device *oct, - size_t console_num, - char *buffer, s32 bytes_read) + size_t console_num) { struct octeon_console *console; @@ -492,7 +483,7 @@ static void check_console(struct work_struct *work) struct octeon_console *console; struct cavium_wk *wk = (struct cavium_wk *)work; struct octeon_device *oct = (struct octeon_device *)wk->ctxptr; - size_t console_num = wk->ctxul; + u32 console_num = (u32)wk->ctxul; u32 delay; console = &oct->console[console_num]; @@ -505,20 +496,17 @@ static void check_console(struct work_struct *work) */ bytes_read = octeon_console_read(oct, console_num, console_buffer, - sizeof(console_buffer) - 1, 0); + sizeof(console_buffer) - 1); if (bytes_read > 0) { total_read += bytes_read; - if (console->waiting) { - octeon_console_handle_result(oct, console_num, - console_buffer, - bytes_read); - } + if (console->waiting) + octeon_console_handle_result(oct, console_num); if (octeon_console_debug_enabled(console_num)) { output_console_line(oct, console, console_num, console_buffer, bytes_read); } } else if (bytes_read < 0) { - dev_err(&oct->pci_dev->dev, "Error reading console %lu, ret=%d\n", + dev_err(&oct->pci_dev->dev, "Error reading console %u, ret=%d\n", console_num, bytes_read); } @@ -530,7 +518,7 @@ static void check_console(struct work_struct *work) */ if (octeon_console_debug_enabled(console_num) && (total_read == 0) && (console->leftover[0])) { - dev_info(&oct->pci_dev->dev, "%lu: %s\n", + dev_info(&oct->pci_dev->dev, "%u: %s\n", console_num, console->leftover); console->leftover[0] = '\0'; } @@ -675,8 +663,8 @@ static inline int octeon_console_avail_bytes(u32 buffer_size, octeon_console_free_bytes(buffer_size, wr_idx, rd_idx); } -int octeon_console_read(struct octeon_device *oct, u32 console_num, - char *buffer, u32 buf_size, u32 flags) +static int octeon_console_read(struct octeon_device *oct, u32 console_num, + char *buffer, u32 buf_size) { int bytes_to_read; u32 rd_idx, wr_idx; @@ -712,7 +700,7 @@ int octeon_console_read(struct octeon_device *oct, u32 console_num, bytes_to_read = console->buffer_size - rd_idx; octeon_pci_read_core_mem(oct, console->output_base_addr + rd_idx, - buffer, bytes_to_read); + (u8 *)buffer, bytes_to_read); octeon_write_device_mem32(oct, console->addr + offsetof(struct octeon_pci_console, output_read_index), diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 8e23e3fad662..0eb504a4379a 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -19,28 +19,19 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> #include <linux/pci.h> #include <linux/crc32.h> -#include <linux/kthread.h> #include <linux/netdevice.h> #include <linux/vmalloc.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" #include "liquidio_image.h" #include "octeon_mem_ops.h" @@ -449,10 +440,10 @@ static struct octeon_config_ptr { }; static char oct_dev_state_str[OCT_DEV_STATES + 1][32] = { - "BEGIN", "PCI-MAP-DONE", "DISPATCH-INIT-DONE", + "BEGIN", "PCI-MAP-DONE", "DISPATCH-INIT-DONE", "IQ-INIT-DONE", "SCBUFF-POOL-INIT-DONE", "RESPLIST-INIT-DONE", "DROQ-INIT-DONE", "IO-QUEUES-INIT-DONE", "CONSOLE-INIT-DONE", - "HOST-READY", "CORE-READY", "RUNNING", "IN-RESET", + "HOST-READY", "CORE-READY", "RUNNING", "IN-RESET", "INVALID" }; @@ -550,17 +541,19 @@ static char *get_oct_app_string(u32 app_mode) return oct_dev_app_str[CVM_DRV_INVALID_APP - CVM_DRV_APP_START]; } +u8 fbuf[4 * 1024 * 1024]; + int octeon_download_firmware(struct octeon_device *oct, const u8 *data, size_t size) { int ret = 0; - u8 *p; - u8 *buffer; + u8 *p = fbuf; u32 crc32_result; u64 load_addr; u32 image_len; struct octeon_firmware_file_header *h; - u32 i; + u32 i, rem, base_len = strlen(LIQUIDIO_BASE_VERSION); + char *base; if (size < sizeof(struct octeon_firmware_file_header)) { dev_err(&oct->pci_dev->dev, "Firmware file too small (%d < %d).\n", @@ -576,19 +569,26 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data, return -EINVAL; } - crc32_result = - crc32(~0, data, - sizeof(struct octeon_firmware_file_header) - - sizeof(u32)) ^ ~0U; + crc32_result = crc32((unsigned int)~0, data, + sizeof(struct octeon_firmware_file_header) - + sizeof(u32)) ^ ~0U; if (crc32_result != be32_to_cpu(h->crc32)) { dev_err(&oct->pci_dev->dev, "Firmware CRC mismatch (0x%08x != 0x%08x).\n", crc32_result, be32_to_cpu(h->crc32)); return -EINVAL; } - if (memcmp(LIQUIDIO_VERSION, h->version, strlen(LIQUIDIO_VERSION))) { - dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s, got %s.\n", - LIQUIDIO_VERSION, h->version); + if (strncmp(LIQUIDIO_PACKAGE, h->version, strlen(LIQUIDIO_PACKAGE))) { + dev_err(&oct->pci_dev->dev, "Unmatched firmware package type. Expected %s, got %s.\n", + LIQUIDIO_PACKAGE, h->version); + return -EINVAL; + } + + base = h->version + strlen(LIQUIDIO_PACKAGE); + ret = memcmp(LIQUIDIO_BASE_VERSION, base, base_len); + if (ret) { + dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s.x, got %s.\n", + LIQUIDIO_BASE_VERSION, base); return -EINVAL; } @@ -602,58 +602,58 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data, snprintf(oct->fw_info.liquidio_firmware_version, 32, "LIQUIDIO: %s", h->version); - buffer = kmemdup(data, size, GFP_KERNEL); - if (!buffer) - return -ENOMEM; - - p = buffer + sizeof(struct octeon_firmware_file_header); + data += sizeof(struct octeon_firmware_file_header); + dev_info(&oct->pci_dev->dev, "%s: Loading %d images\n", __func__, + be32_to_cpu(h->num_images)); /* load all images */ for (i = 0; i < be32_to_cpu(h->num_images); i++) { load_addr = be64_to_cpu(h->desc[i].addr); image_len = be32_to_cpu(h->desc[i].len); - /* validate the image */ - crc32_result = crc32(~0, p, image_len) ^ ~0U; - if (crc32_result != be32_to_cpu(h->desc[i].crc32)) { - dev_err(&oct->pci_dev->dev, - "Firmware CRC mismatch in image %d (0x%08x != 0x%08x).\n", - i, crc32_result, - be32_to_cpu(h->desc[i].crc32)); - ret = -EINVAL; - goto done_downloading; - } + dev_info(&oct->pci_dev->dev, "Loading firmware %d at %llx\n", + image_len, load_addr); - /* download the image */ - octeon_pci_write_core_mem(oct, load_addr, p, image_len); + /* Write in 4MB chunks*/ + rem = image_len; - p += image_len; - dev_dbg(&oct->pci_dev->dev, - "Downloaded image %d (%d bytes) to address 0x%016llx\n", - i, image_len, load_addr); + while (rem) { + if (rem < (4 * 1024 * 1024)) + size = rem; + else + size = 4 * 1024 * 1024; + + memcpy(p, data, size); + + /* download the image */ + octeon_pci_write_core_mem(oct, load_addr, p, (u32)size); + + data += size; + rem -= (u32)size; + load_addr += size; + } } + dev_info(&oct->pci_dev->dev, "Writing boot command: %s\n", + h->bootcmd); /* Invoke the bootcmd */ ret = octeon_console_send_cmd(oct, h->bootcmd, 50); -done_downloading: - kfree(buffer); - - return ret; + return 0; } void octeon_free_device_mem(struct octeon_device *oct) { - u32 i; + int i; - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - /* could check mask as well */ - vfree(oct->droq[i]); + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (oct->io_qmask.oq & (1ULL << i)) + vfree(oct->droq[i]); } - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - /* could check mask as well */ - vfree(oct->instr_queue[i]); + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (oct->io_qmask.iq & (1ULL << i)) + vfree(oct->instr_queue[i]); } i = oct->octeon_id; @@ -735,55 +735,61 @@ struct octeon_device *octeon_allocate_device(u32 pci_id, octeon_device[oct_idx] = oct; oct->octeon_id = oct_idx; - snprintf((oct->device_name), sizeof(oct->device_name), + snprintf(oct->device_name, sizeof(oct->device_name), "LiquidIO%d", (oct->octeon_id)); return oct; } +/* this function is only for setting up the first queue */ int octeon_setup_instr_queues(struct octeon_device *oct) { - u32 i, num_iqs = 0; u32 num_descs = 0; + u32 iq_no = 0; + union oct_txpciq txpciq; + int numa_node = cpu_to_node(iq_no % num_online_cpus()); /* this causes queue 0 to be default queue */ - if (OCTEON_CN6XXX(oct)) { - num_iqs = 1; + if (OCTEON_CN6XXX(oct)) num_descs = CFG_GET_NUM_DEF_TX_DESCS(CHIP_FIELD(oct, cn6xxx, conf)); - } oct->num_iqs = 0; - for (i = 0; i < num_iqs; i++) { - oct->instr_queue[i] = + oct->instr_queue[0] = vmalloc_node(sizeof(*oct->instr_queue[0]), + numa_node); + if (!oct->instr_queue[0]) + oct->instr_queue[0] = vmalloc(sizeof(struct octeon_instr_queue)); - if (!oct->instr_queue[i]) - return 1; - - memset(oct->instr_queue[i], 0, - sizeof(struct octeon_instr_queue)); - - oct->instr_queue[i]->app_ctx = (void *)(size_t)i; - if (octeon_init_instr_queue(oct, i, num_descs)) - return 1; - - oct->num_iqs++; + if (!oct->instr_queue[0]) + return 1; + memset(oct->instr_queue[0], 0, sizeof(struct octeon_instr_queue)); + oct->instr_queue[0]->q_index = 0; + oct->instr_queue[0]->app_ctx = (void *)(size_t)0; + oct->instr_queue[0]->ifidx = 0; + txpciq.u64 = 0; + txpciq.s.q_no = iq_no; + txpciq.s.use_qpg = 0; + txpciq.s.qpg = 0; + if (octeon_init_instr_queue(oct, txpciq, num_descs)) { + /* prevent memory leak */ + vfree(oct->instr_queue[0]); + return 1; } + oct->num_iqs++; return 0; } int octeon_setup_output_queues(struct octeon_device *oct) { - u32 i, num_oqs = 0; u32 num_descs = 0; u32 desc_size = 0; + u32 oq_no = 0; + int numa_node = cpu_to_node(oq_no % num_online_cpus()); /* this causes queue 0 to be default queue */ if (OCTEON_CN6XXX(oct)) { - /* CFG_GET_OQ_MAX_BASE_Q(CHIP_FIELD(oct, cn6xxx, conf)); */ - num_oqs = 1; num_descs = CFG_GET_NUM_DEF_RX_DESCS(CHIP_FIELD(oct, cn6xxx, conf)); desc_size = @@ -791,19 +797,15 @@ int octeon_setup_output_queues(struct octeon_device *oct) } oct->num_oqs = 0; + oct->droq[0] = vmalloc_node(sizeof(*oct->droq[0]), numa_node); + if (!oct->droq[0]) + oct->droq[0] = vmalloc(sizeof(*oct->droq[0])); + if (!oct->droq[0]) + return 1; - for (i = 0; i < num_oqs; i++) { - oct->droq[i] = vmalloc(sizeof(*oct->droq[i])); - if (!oct->droq[i]) - return 1; - - memset(oct->droq[i], 0, sizeof(struct octeon_droq)); - - if (octeon_init_droq(oct, i, num_descs, desc_size, NULL)) - return 1; - - oct->num_oqs++; - } + if (octeon_init_droq(oct, oq_no, num_descs, desc_size, NULL)) + return 1; + oct->num_oqs++; return 0; } @@ -1005,79 +1007,6 @@ octeon_register_dispatch_fn(struct octeon_device *oct, return 0; } -/* octeon_unregister_dispatch_fn - * Parameters: - * oct - octeon device - * opcode - driver should unregister the function for this opcode - * subcode - driver should unregister the function for this subcode - * Description: - * Unregister the function set for this opcode+subcode. - * Returns: - * Success: 0 - * Failure: 1 - * Locks: - * No locks are held. - */ -int -octeon_unregister_dispatch_fn(struct octeon_device *oct, u16 opcode, - u16 subcode) -{ - int retval = 0; - u32 idx; - struct list_head *dispatch, *dfree = NULL, *tmp2; - u16 combined_opcode = OPCODE_SUBCODE(opcode, subcode); - - idx = combined_opcode & OCTEON_OPCODE_MASK; - - spin_lock_bh(&oct->dispatch.lock); - - if (oct->dispatch.count == 0) { - spin_unlock_bh(&oct->dispatch.lock); - dev_err(&oct->pci_dev->dev, - "No dispatch functions registered for this device\n"); - return 1; - } - - if (oct->dispatch.dlist[idx].opcode == combined_opcode) { - dispatch = &oct->dispatch.dlist[idx].list; - if (dispatch->next != dispatch) { - dispatch = dispatch->next; - oct->dispatch.dlist[idx].opcode = - ((struct octeon_dispatch *)dispatch)->opcode; - oct->dispatch.dlist[idx].dispatch_fn = - ((struct octeon_dispatch *) - dispatch)->dispatch_fn; - oct->dispatch.dlist[idx].arg = - ((struct octeon_dispatch *)dispatch)->arg; - list_del(dispatch); - dfree = dispatch; - } else { - oct->dispatch.dlist[idx].opcode = 0; - oct->dispatch.dlist[idx].dispatch_fn = NULL; - oct->dispatch.dlist[idx].arg = NULL; - } - } else { - retval = 1; - list_for_each_safe(dispatch, tmp2, - &(oct->dispatch.dlist[idx]. - list)) { - if (((struct octeon_dispatch *)dispatch)->opcode == - combined_opcode) { - list_del(dispatch); - dfree = dispatch; - retval = 0; - } - } - } - - if (!retval) - oct->dispatch.count--; - - spin_unlock_bh(&oct->dispatch.lock); - vfree(dfree); - return retval; -} - int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf) { u32 i; @@ -1152,8 +1081,8 @@ core_drv_init_err: int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no) { - if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES) && - (oct->io_qmask.iq & (1UL << q_no))) + if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES(oct)) && + (oct->io_qmask.iq & (1ULL << q_no))) return oct->instr_queue[q_no]->max_count; return -1; @@ -1161,8 +1090,8 @@ int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no) int octeon_get_rx_qsize(struct octeon_device *oct, u32 q_no) { - if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES) && - (oct->io_qmask.oq & (1UL << q_no))) + if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES(oct)) && + (oct->io_qmask.oq & (1ULL << q_no))) return oct->droq[q_no]->max_count; return -1; } @@ -1253,10 +1182,10 @@ void lio_pci_writeq(struct octeon_device *oct, int octeon_mem_access_ok(struct octeon_device *oct) { u64 access_okay = 0; + u64 lmc0_reset_ctl; /* Check to make sure a DDR interface is enabled */ - u64 lmc0_reset_ctl = lio_pci_readq(oct, CN6XXX_LMC0_RESET_CTL); - + lmc0_reset_ctl = lio_pci_readq(oct, CN6XXX_LMC0_RESET_CTL); access_okay = (lmc0_reset_ctl & CN6XXX_LMC0_RESET_CTL_DDR3RST_MASK); return access_okay ? 0 : 1; @@ -1270,9 +1199,6 @@ int octeon_wait_for_ddr_init(struct octeon_device *oct, u32 *timeout) if (!timeout) return ret; - while (*timeout == 0) - schedule_timeout_uninterruptible(HZ / 10); - for (ms = 0; (ret != 0) && ((*timeout == 0) || (ms <= *timeout)); ms += HZ / 10) { ret = octeon_mem_access_ok(oct); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index 36e1f85df8c4..01edfb404346 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -152,9 +152,9 @@ struct octeon_mmio { #define MAX_OCTEON_MAPS 32 struct octeon_io_enable { - u32 iq; - u32 oq; - u32 iq64B; + u64 iq; + u64 oq; + u64 iq64B; }; struct octeon_reg_list { @@ -204,8 +204,7 @@ struct octeon_fn_list { void (*bar1_idx_setup)(struct octeon_device *, u64, u32, int); void (*bar1_idx_write)(struct octeon_device *, u32, u32); u32 (*bar1_idx_read)(struct octeon_device *, u32); - u32 (*update_iq_read_idx)(struct octeon_device *, - struct octeon_instr_queue *); + u32 (*update_iq_read_idx)(struct octeon_instr_queue *); void (*enable_oq_pkt_time_intr)(struct octeon_device *, u32); void (*disable_oq_pkt_time_intr)(struct octeon_device *, u32); @@ -222,7 +221,7 @@ struct octeon_fn_list { /* Structure for named memory blocks * Number of descriptors - * available can be changed without affecting compatiblity, + * available can be changed without affecting compatibility, * but name length changes require a bump in the bootmem * descriptor version * Note: This structure must be naturally 64 bit aligned, as a single @@ -255,7 +254,7 @@ struct oct_fw_info { struct cavium_wk { struct delayed_work work; void *ctxptr; - size_t ctxul; + u64 ctxul; }; struct cavium_wq { @@ -267,6 +266,8 @@ struct octdev_props { /* Each interface in the Octeon device has a network * device pointer (used for OS specific calls). */ + int napi_enabled; + int gmxport; struct net_device *netdev; }; @@ -324,7 +325,8 @@ struct octeon_device { struct octeon_sc_buffer_pool sc_buf_pool; /** The input instruction queues */ - struct octeon_instr_queue *instr_queue[MAX_OCTEON_INSTR_QUEUES]; + struct octeon_instr_queue *instr_queue + [MAX_POSSIBLE_OCTEON_INSTR_QUEUES]; /** The doubly-linked list of instruction response */ struct octeon_response_list response_list[MAX_RESPONSE_LISTS]; @@ -332,7 +334,7 @@ struct octeon_device { u32 num_oqs; /** The DROQ output queues */ - struct octeon_droq *droq[MAX_OCTEON_OUTPUT_QUEUES]; + struct octeon_droq *droq[MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES]; struct octeon_io_enable io_qmask; @@ -381,15 +383,29 @@ struct octeon_device { struct cavium_wq dma_comp_wq; - struct cavium_wq check_db_wq[MAX_OCTEON_INSTR_QUEUES]; + /** Lock for dma response list */ + spinlock_t cmd_resp_wqlock; + u32 cmd_resp_state; + + struct cavium_wq check_db_wq[MAX_POSSIBLE_OCTEON_INSTR_QUEUES]; struct cavium_wk nic_poll_work; struct cavium_wk console_poll_work[MAX_OCTEON_MAPS]; void *priv; + + int rx_pause; + int tx_pause; + + struct oct_link_stats link_stats; /*stastics from firmware*/ + + /* private flags to control driver-specific features through ethtool */ + u32 priv_flags; }; +#define OCT_DRV_ONLINE 1 +#define OCT_DRV_OFFLINE 2 #define OCTEON_CN6XXX(oct) ((oct->chip_id == OCTEON_CN66XX) || \ (oct->chip_id == OCTEON_CN68XX)) #define CHIP_FIELD(oct, TYPE, field) \ @@ -569,8 +585,7 @@ int octeon_add_console(struct octeon_device *oct, u32 console_num); int octeon_console_write(struct octeon_device *oct, u32 console_num, char *buffer, u32 write_request_size, u32 flags); int octeon_console_write_avail(struct octeon_device *oct, u32 console_num); -int octeon_console_read(struct octeon_device *oct, u32 console_num, - char *buffer, u32 buf_size, u32 flags); + int octeon_console_read_avail(struct octeon_device *oct, u32 console_num); /** Removes all attached consoles. */ @@ -646,4 +661,17 @@ void *oct_get_config_info(struct octeon_device *oct, u16 card_type); */ struct octeon_config *octeon_get_conf(struct octeon_device *oct); +/* LiquidIO driver pivate flags */ +enum { + OCT_PRIV_FLAG_TX_BYTES = 0, /* Tx interrupts by pending byte count */ +}; + +static inline void lio_set_priv_flag(struct octeon_device *octdev, u32 flag, + u32 val) +{ + if (val) + octdev->priv_flags |= (0x1 << flag); + else + octdev->priv_flags &= ~(0x1 << flag); +} #endif diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 174072b3740b..e0afe4c1fd01 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -19,30 +19,18 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> #include <linux/vmalloc.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" - -/* #define CAVIUM_ONLY_PERF_MODE */ #define CVM_MIN(d1, d2) (((d1) < (d2)) ? (d1) : (d2)) #define CVM_MAX(d1, d2) (((d1) > (d2)) ? (d1) : (d2)) @@ -104,8 +92,12 @@ static inline void *octeon_get_dispatch_arg(struct octeon_device *octeon_dev, return fn_arg; } -u32 octeon_droq_check_hw_for_pkts(struct octeon_device *oct, - struct octeon_droq *droq) +/** Check for packets on Droq. This function should be called with + * lock held. + * @param droq - Droq on which count is checked. + * @return Returns packet count. + */ +u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq) { u32 pkt_count = 0; @@ -151,22 +143,26 @@ octeon_droq_destroy_ring_buffers(struct octeon_device *oct, struct octeon_droq *droq) { u32 i; + struct octeon_skb_page_info *pg_info; for (i = 0; i < droq->max_count; i++) { - if (droq->recv_buf_list[i].buffer) { - if (droq->desc_ring) { - lio_unmap_ring_info(oct->pci_dev, - (u64)droq-> - desc_ring[i].info_ptr, - OCT_DROQ_INFO_SIZE); - lio_unmap_ring(oct->pci_dev, - (u64)droq->desc_ring[i]. - buffer_ptr, - droq->buffer_size); - } - recv_buffer_free(droq->recv_buf_list[i].buffer); - droq->recv_buf_list[i].buffer = NULL; - } + pg_info = &droq->recv_buf_list[i].pg_info; + + if (pg_info->dma) + lio_unmap_ring(oct->pci_dev, + (u64)pg_info->dma); + pg_info->dma = 0; + + if (pg_info->page) + recv_buffer_destroy(droq->recv_buf_list[i].buffer, + pg_info); + + if (droq->desc_ring && droq->desc_ring[i].info_ptr) + lio_unmap_ring_info(oct->pci_dev, + (u64)droq-> + desc_ring[i].info_ptr, + OCT_DROQ_INFO_SIZE); + droq->recv_buf_list[i].buffer = NULL; } octeon_droq_reset_indices(droq); @@ -181,25 +177,23 @@ octeon_droq_setup_ring_buffers(struct octeon_device *oct, struct octeon_droq_desc *desc_ring = droq->desc_ring; for (i = 0; i < droq->max_count; i++) { - buf = recv_buffer_alloc(oct, droq->q_no, droq->buffer_size); + buf = recv_buffer_alloc(oct, &droq->recv_buf_list[i].pg_info); if (!buf) { dev_err(&oct->pci_dev->dev, "%s buffer alloc failed\n", __func__); + droq->stats.rx_alloc_failure++; return -ENOMEM; } droq->recv_buf_list[i].buffer = buf; droq->recv_buf_list[i].data = get_rbd(buf); - droq->info_list[i].length = 0; /* map ring buffers into memory */ desc_ring[i].info_ptr = lio_map_ring_info(droq, i); desc_ring[i].buffer_ptr = - lio_map_ring(oct->pci_dev, - droq->recv_buf_list[i].buffer, - droq->buffer_size); + lio_map_ring(droq->recv_buf_list[i].buffer); } octeon_droq_reset_indices(droq); @@ -242,6 +236,8 @@ int octeon_init_droq(struct octeon_device *oct, struct octeon_droq *droq; u32 desc_ring_size = 0, c_num_descs = 0, c_buf_size = 0; u32 c_pkts_per_intr = 0, c_refill_threshold = 0; + int orig_node = dev_to_node(&oct->pci_dev->dev); + int numa_node = cpu_to_node(q_no % num_online_cpus()); dev_dbg(&oct->pci_dev->dev, "%s[%d]\n", __func__, q_no); @@ -261,15 +257,23 @@ int octeon_init_droq(struct octeon_device *oct, struct octeon_config *conf6x = CHIP_FIELD(oct, cn6xxx, conf); c_pkts_per_intr = (u32)CFG_GET_OQ_PKTS_PER_INTR(conf6x); - c_refill_threshold = (u32)CFG_GET_OQ_REFILL_THRESHOLD(conf6x); + c_refill_threshold = + (u32)CFG_GET_OQ_REFILL_THRESHOLD(conf6x); + } else { + return 1; } droq->max_count = c_num_descs; droq->buffer_size = c_buf_size; desc_ring_size = droq->max_count * OCT_DROQ_DESC_SIZE; + set_dev_node(&oct->pci_dev->dev, numa_node); droq->desc_ring = lio_dma_alloc(oct, desc_ring_size, (dma_addr_t *)&droq->desc_ring_dma); + set_dev_node(&oct->pci_dev->dev, orig_node); + if (!droq->desc_ring) + droq->desc_ring = lio_dma_alloc(oct, desc_ring_size, + (dma_addr_t *)&droq->desc_ring_dma); if (!droq->desc_ring) { dev_err(&oct->pci_dev->dev, @@ -283,12 +287,11 @@ int octeon_init_droq(struct octeon_device *oct, droq->max_count); droq->info_list = - cnnic_alloc_aligned_dma(oct->pci_dev, - (droq->max_count * OCT_DROQ_INFO_SIZE), - &droq->info_alloc_size, - &droq->info_base_addr, - &droq->info_list_dma); - + cnnic_numa_alloc_aligned_dma((droq->max_count * + OCT_DROQ_INFO_SIZE), + &droq->info_alloc_size, + &droq->info_base_addr, + numa_node); if (!droq->info_list) { dev_err(&oct->pci_dev->dev, "Cannot allocate memory for info list.\n"); lio_dma_free(oct, (droq->max_count * OCT_DROQ_DESC_SIZE), @@ -297,7 +300,12 @@ int octeon_init_droq(struct octeon_device *oct, } droq->recv_buf_list = (struct octeon_recv_buffer *) - vmalloc(droq->max_count * + vmalloc_node(droq->max_count * + OCT_DROQ_RECVBUF_SIZE, + numa_node); + if (!droq->recv_buf_list) + droq->recv_buf_list = (struct octeon_recv_buffer *) + vmalloc(droq->max_count * OCT_DROQ_RECVBUF_SIZE); if (!droq->recv_buf_list) { dev_err(&oct->pci_dev->dev, "Output queue recv buf list alloc failed\n"); @@ -320,7 +328,7 @@ int octeon_init_droq(struct octeon_device *oct, /* For 56xx Pass1, this function won't be called, so no checks. */ oct->fn_list.setup_oq_regs(oct, q_no); - oct->io_qmask.oq |= (1 << q_no); + oct->io_qmask.oq |= (1ULL << q_no); return 0; @@ -358,6 +366,7 @@ static inline struct octeon_recv_info *octeon_create_recv_info( struct octeon_recv_pkt *recv_pkt; struct octeon_recv_info *recv_info; u32 i, bytes_left; + struct octeon_skb_page_info *pg_info; info = &droq->info_list[idx]; @@ -375,9 +384,14 @@ static inline struct octeon_recv_info *octeon_create_recv_info( bytes_left = (u32)info->length; while (buf_cnt) { - lio_unmap_ring(octeon_dev->pci_dev, - (u64)droq->desc_ring[idx].buffer_ptr, - droq->buffer_size); + { + pg_info = &droq->recv_buf_list[idx].pg_info; + + lio_unmap_ring(octeon_dev->pci_dev, + (u64)pg_info->dma); + pg_info->page = NULL; + pg_info->dma = 0; + } recv_pkt->buffer_size[i] = (bytes_left >= @@ -449,6 +463,7 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq) void *buf = NULL; u8 *data; u32 desc_refilled = 0; + struct octeon_skb_page_info *pg_info; desc_ring = droq->desc_ring; @@ -458,13 +473,22 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq) * the buffer, else allocate. */ if (!droq->recv_buf_list[droq->refill_idx].buffer) { - buf = recv_buffer_alloc(octeon_dev, droq->q_no, - droq->buffer_size); + pg_info = + &droq->recv_buf_list[droq->refill_idx].pg_info; + /* Either recycle the existing pages or go for + * new page alloc + */ + if (pg_info->page) + buf = recv_buffer_reuse(octeon_dev, pg_info); + else + buf = recv_buffer_alloc(octeon_dev, pg_info); /* If a buffer could not be allocated, no point in * continuing */ - if (!buf) + if (!buf) { + droq->stats.rx_alloc_failure++; break; + } droq->recv_buf_list[droq->refill_idx].buffer = buf; data = get_rbd(buf); @@ -476,11 +500,8 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq) droq->recv_buf_list[droq->refill_idx].data = data; desc_ring[droq->refill_idx].buffer_ptr = - lio_map_ring(octeon_dev->pci_dev, - droq->recv_buf_list[droq-> - refill_idx].buffer, - droq->buffer_size); - + lio_map_ring(droq->recv_buf_list[droq-> + refill_idx].buffer); /* Reset any previous values in the length field. */ droq->info_list[droq->refill_idx].length = 0; @@ -539,7 +560,9 @@ octeon_droq_dispatch_pkt(struct octeon_device *oct, droq->stats.dropped_nomem++; } } else { - dev_err(&oct->pci_dev->dev, "DROQ: No dispatch function\n"); + dev_err(&oct->pci_dev->dev, "DROQ: No dispatch function (opcode %u/%u)\n", + (unsigned int)rh->r.opcode, + (unsigned int)rh->r.subcode); droq->stats.dropped_nodispatch++; } /* else (dispatch_fn ... */ @@ -586,6 +609,8 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, for (pkt = 0; pkt < pkt_count; pkt++) { u32 pkt_len = 0; struct sk_buff *nicbuf = NULL; + struct octeon_skb_page_info *pg_info; + void *buf; info = &droq->info_list[droq->read_idx]; octeon_swap_8B_data((u64 *)info, 2); @@ -605,7 +630,6 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, rh = &info->rh; total_len += (u32)info->length; - if (OPCODE_SLOW_PATH(rh)) { u32 buf_cnt; @@ -614,50 +638,45 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, droq->refill_count += buf_cnt; } else { if (info->length <= droq->buffer_size) { - lio_unmap_ring(oct->pci_dev, - (u64)droq->desc_ring[ - droq->read_idx].buffer_ptr, - droq->buffer_size); pkt_len = (u32)info->length; nicbuf = droq->recv_buf_list[ droq->read_idx].buffer; + pg_info = &droq->recv_buf_list[ + droq->read_idx].pg_info; + if (recv_buffer_recycle(oct, pg_info)) + pg_info->page = NULL; droq->recv_buf_list[droq->read_idx].buffer = NULL; + INCR_INDEX_BY1(droq->read_idx, droq->max_count); - skb_put(nicbuf, pkt_len); droq->refill_count++; } else { - nicbuf = octeon_fast_packet_alloc(oct, droq, - droq->q_no, - (u32) + nicbuf = octeon_fast_packet_alloc((u32) info->length); pkt_len = 0; /* nicbuf allocation can fail. We'll handle it * inside the loop. */ while (pkt_len < info->length) { - int cpy_len; + int cpy_len, idx = droq->read_idx; - cpy_len = ((pkt_len + - droq->buffer_size) > - info->length) ? + cpy_len = ((pkt_len + droq->buffer_size) + > info->length) ? ((u32)info->length - pkt_len) : droq->buffer_size; if (nicbuf) { - lio_unmap_ring(oct->pci_dev, - (u64) - droq->desc_ring - [droq->read_idx]. - buffer_ptr, - droq-> - buffer_size); octeon_fast_packet_next(droq, nicbuf, cpy_len, - droq-> - read_idx - ); + idx); + buf = droq->recv_buf_list[idx]. + buffer; + recv_buffer_fast_free(buf); + droq->recv_buf_list[idx].buffer + = NULL; + } else { + droq->stats.rx_alloc_failure++; } pkt_len += cpy_len; @@ -668,12 +687,14 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, } if (nicbuf) { - if (droq->ops.fptr) + if (droq->ops.fptr) { droq->ops.fptr(oct->octeon_id, - nicbuf, pkt_len, - rh, &droq->napi); - else + nicbuf, pkt_len, + rh, &droq->napi, + droq->ops.farg); + } else { recv_buffer_free(nicbuf); + } } } @@ -681,16 +702,16 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, int desc_refilled = octeon_droq_refill(oct, droq); /* Flush the droq descriptor data to memory to be sure - * that when we update the credits the data in memory - * is accurate. - */ + * that when we update the credits the data in memory + * is accurate. + */ wmb(); writel((desc_refilled), droq->pkts_credit_reg); /* make sure mmio write completes */ mmiowb(); } - } /* for ( each packet )... */ + } /* for (each packet)... */ /* Increment refill_count by the number of buffers processed. */ droq->stats.pkts_received += pkt; @@ -721,7 +742,7 @@ octeon_droq_process_packets(struct octeon_device *oct, if (pkt_count > budget) pkt_count = budget; - /* Grab the lock */ + /* Grab the droq lock */ spin_lock(&droq->lock); pkts_processed = octeon_droq_fast_process_packets(oct, droq, pkt_count); @@ -783,7 +804,7 @@ octeon_droq_process_poll_pkts(struct octeon_device *oct, total_pkts_processed += pkts_processed; - octeon_droq_check_hw_for_pkts(oct, droq); + octeon_droq_check_hw_for_pkts(droq); } spin_unlock(&droq->lock); @@ -807,18 +828,6 @@ octeon_process_droq_poll_cmd(struct octeon_device *oct, u32 q_no, int cmd, u32 arg) { struct octeon_droq *droq; - struct octeon_config *oct_cfg = NULL; - - oct_cfg = octeon_get_conf(oct); - - if (!oct_cfg) - return -EINVAL; - - if (q_no >= CFG_GET_OQ_MAX_Q(oct_cfg)) { - dev_err(&oct->pci_dev->dev, "%s: droq id (%d) exceeds MAX (%d)\n", - __func__, q_no, (oct->num_oqs - 1)); - return -EINVAL; - } droq = oct->droq[q_no]; @@ -937,6 +946,7 @@ int octeon_unregister_droq_ops(struct octeon_device *oct, u32 q_no) spin_lock_irqsave(&droq->lock, flags); droq->ops.fptr = NULL; + droq->ops.farg = NULL; droq->ops.drop_on_max = 0; spin_unlock_irqrestore(&droq->lock, flags); @@ -949,6 +959,7 @@ int octeon_create_droq(struct octeon_device *oct, u32 desc_size, void *app_ctx) { struct octeon_droq *droq; + int numa_node = cpu_to_node(q_no % num_online_cpus()); if (oct->droq[q_no]) { dev_dbg(&oct->pci_dev->dev, "Droq already in use. Cannot create droq %d again\n", @@ -957,7 +968,9 @@ int octeon_create_droq(struct octeon_device *oct, } /* Allocate the DS for the new droq. */ - droq = vmalloc(sizeof(*droq)); + droq = vmalloc_node(sizeof(*droq), numa_node); + if (!droq) + droq = vmalloc(sizeof(*droq)); if (!droq) goto create_droq_fail; memset(droq, 0, sizeof(struct octeon_droq)); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h index 7940ccee12d9..5a6fb9113bbd 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h @@ -65,6 +65,17 @@ struct octeon_droq_info { #define OCT_DROQ_INFO_SIZE (sizeof(struct octeon_droq_info)) +struct octeon_skb_page_info { + /* DMA address for the page */ + dma_addr_t dma; + + /* Page for the rx dma **/ + struct page *page; + + /** which offset into page */ + unsigned int page_offset; +}; + /** Pointer to data buffer. * Driver keeps a pointer to the data buffer that it made available to * the Octeon device. Since the descriptor ring keeps physical (bus) @@ -77,6 +88,9 @@ struct octeon_recv_buffer { /** Data in the packet buffer. */ u8 *data; + + /** pg_info **/ + struct octeon_skb_page_info pg_info; }; #define OCT_DROQ_RECVBUF_SIZE (sizeof(struct octeon_recv_buffer)) @@ -106,6 +120,13 @@ struct oct_droq_stats { /** Num of Packets dropped due to receive path failures. */ u64 rx_dropped; + + /** Num of vxlan packets received; */ + u64 rx_vxlan; + + /** Num of failures of recv_buffer_alloc() */ + u64 rx_alloc_failure; + }; #define POLL_EVENT_INTR_ARRIVED 1 @@ -213,7 +234,8 @@ struct octeon_droq_ops { * data in the buffer. The receive header gives the port * number to the caller. Function pointer is set by caller. */ - void (*fptr)(u32, void *, u32, union octeon_rh *, void *); + void (*fptr)(u32, void *, u32, union octeon_rh *, void *, void *); + void *farg; /* This function will be called by the driver for all NAPI related * events. The first param is the octeon id. The second param is the @@ -394,24 +416,9 @@ int octeon_register_dispatch_fn(struct octeon_device *oct, u16 subcode, octeon_dispatch_fn_t fn, void *fn_arg); -/** Remove registration for an opcode/subcode. This will delete the mapping for - * an opcode/subcode. The dispatch function will be unregistered and will no - * longer be called if a packet with the opcode/subcode arrives in the driver - * output queues. - * @param oct - the octeon device to unregister from. - * @param opcode - the opcode to be unregistered. - * @param subcode - the subcode to be unregistered. - * - * @return Success: 0; Failure: 1 - */ -int octeon_unregister_dispatch_fn(struct octeon_device *oct, - u16 opcode, - u16 subcode); - void octeon_droq_print_stats(void); -u32 octeon_droq_check_hw_for_pkts(struct octeon_device *oct, - struct octeon_droq *droq); +u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq); int octeon_create_droq(struct octeon_device *oct, u32 q_no, u32 num_descs, u32 desc_size, void *app_ctx); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 592fe49b589d..ff4b1d6f007b 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -65,6 +65,11 @@ struct oct_iq_stats { u64 tx_iq_busy;/**< Numof times this iq was found to be full. */ u64 tx_dropped;/**< Numof pkts dropped dueto xmitpath errors. */ u64 tx_tot_bytes;/**< Total count of bytes sento to network. */ + u64 tx_gso; /* count of tso */ + u64 tx_vxlan; /* tunnel */ + u64 tx_dmamap_fail; + u64 tx_restart; + /*u64 tx_timeout_count;*/ }; #define OCT_IQ_STATS_SIZE (sizeof(struct oct_iq_stats)) @@ -75,18 +80,26 @@ struct oct_iq_stats { * a Octeon device has one such structure to represent it. */ struct octeon_instr_queue { + struct octeon_device *oct_dev; + /** A spinlock to protect access to the input ring. */ spinlock_t lock; + /** A spinlock to protect while posting on the ring. */ + spinlock_t post_lock; + + /** A spinlock to protect access to the input ring.*/ + spinlock_t iq_flush_running_lock; + /** Flag that indicates if the queue uses 64 byte commands. */ u32 iqcmd_64B:1; - /** Queue Number. */ - u32 iq_no:5; + /** Queue info. */ + union oct_txpciq txpciq; u32 rsvd:17; - /* Controls the periodic flushing of iq */ + /* Controls whether extra flushing of IQ is done on Tx */ u32 do_auto_flush:1; u32 status:8; @@ -147,6 +160,13 @@ struct octeon_instr_queue { /** Application context */ void *app_ctx; + + /* network stack queue index */ + int q_index; + + /*os ifidx associated with this queue */ + int ifidx; + }; /*---------------------- INSTRUCTION FORMAT ----------------------------*/ @@ -176,12 +196,12 @@ struct octeon_instr_32B { /** 64-byte instruction format. * Format of instruction for a 64-byte mode input queue. */ -struct octeon_instr_64B { +struct octeon_instr2_64B { /** Pointer where the input data is available. */ u64 dptr; /** Instruction Header. */ - u64 ih; + u64 ih2; /** Input Request Header. */ u64 irh; @@ -198,14 +218,44 @@ struct octeon_instr_64B { u64 rptr; u64 reserved; +}; + +struct octeon_instr3_64B { + /** Pointer where the input data is available. */ + u64 dptr; + + /** Instruction Header. */ + u64 ih3; + + /** Instruction Header. */ + u64 pki_ih3; + + /** Input Request Header. */ + u64 irh; + /** opcode/subcode specific parameters */ + u64 ossp[2]; + + /** Return Data Parameters */ + u64 rdp; + + /** Pointer where the response for a RAW mode packet will be written + * by Octeon. + */ + u64 rptr; + +}; + +union octeon_instr_64B { + struct octeon_instr2_64B cmd2; + struct octeon_instr3_64B cmd3; }; -#define OCT_64B_INSTR_SIZE (sizeof(struct octeon_instr_64B)) +#define OCT_64B_INSTR_SIZE (sizeof(union octeon_instr_64B)) /** The size of each buffer in soft command buffer pool */ -#define SOFT_COMMAND_BUFFER_SIZE 1024 +#define SOFT_COMMAND_BUFFER_SIZE 1536 struct octeon_soft_command { /** Soft command buffer info. */ @@ -214,7 +264,8 @@ struct octeon_soft_command { u32 size; /** Command and return status */ - struct octeon_instr_64B cmd; + union octeon_instr_64B cmd; + #define COMPLETION_WORD_INIT 0xffffffffffffffffULL u64 *status_word; @@ -242,7 +293,7 @@ struct octeon_soft_command { /** Maximum number of buffers to allocate into soft command buffer pool */ -#define MAX_SOFT_COMMAND_BUFFERS 16 +#define MAX_SOFT_COMMAND_BUFFERS 256 /** Head of a soft command buffer pool. */ @@ -268,14 +319,15 @@ void octeon_free_soft_command(struct octeon_device *oct, /** * octeon_init_instr_queue() * @param octeon_dev - pointer to the octeon device structure. - * @param iq_no - queue to be initialized (0 <= q_no <= 3). + * @param txpciq - queue to be initialized (0 <= q_no <= 3). * * Called at driver init time for each input queue. iq_conf has the * configuration parameters for the queue. * * @return Success: 0 Failure: 1 */ -int octeon_init_instr_queue(struct octeon_device *octeon_dev, u32 iq_no, +int octeon_init_instr_queue(struct octeon_device *octeon_dev, + union oct_txpciq txpciq, u32 num_descs); /** @@ -298,7 +350,7 @@ octeon_register_reqtype_free_fn(struct octeon_device *oct, int reqtype, int lio_process_iq_request_list(struct octeon_device *oct, - struct octeon_instr_queue *iq); + struct octeon_instr_queue *iq, u32 napi_budget); int octeon_send_command(struct octeon_device *oct, u32 iq_no, u32 force_db, void *cmd, void *buf, @@ -313,7 +365,10 @@ void octeon_prepare_soft_command(struct octeon_device *oct, int octeon_send_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc); -int octeon_setup_iq(struct octeon_device *oct, u32 iq_no, - u32 num_descs, void *app_ctx); - +int octeon_setup_iq(struct octeon_device *oct, int ifidx, + int q_index, union oct_txpciq iq_no, u32 num_descs, + void *app_ctx); +int +octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, + u32 pending_thresh, u32 napi_budget); #endif /* __OCTEON_IQ_H__ */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h index cbd081981180..bc14e4c27332 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h @@ -126,22 +126,27 @@ static inline int octeon_map_pci_barx(struct octeon_device *oct, } static inline void * -cnnic_alloc_aligned_dma(struct pci_dev *pci_dev, - u32 size, - u32 *alloc_size, - size_t *orig_ptr, - size_t *dma_addr __attribute__((unused))) +cnnic_numa_alloc_aligned_dma(u32 size, + u32 *alloc_size, + size_t *orig_ptr, + int numa_node) { int retries = 0; void *ptr = NULL; #define OCTEON_MAX_ALLOC_RETRIES 1 do { - ptr = - (void *)__get_free_pages(GFP_KERNEL, - get_order(size)); + struct page *page = NULL; + + page = alloc_pages_node(numa_node, + GFP_KERNEL, + get_order(size)); + if (!page) + page = alloc_pages(GFP_KERNEL, + get_order(size)); + ptr = (void *)page_address(page); if ((unsigned long)ptr & 0x07) { - free_pages((unsigned long)ptr, get_order(size)); + __free_pages(page, get_order(size)); ptr = NULL; /* Increment the size required if the first * attempt failed. @@ -169,7 +174,7 @@ sleep_cond(wait_queue_head_t *wait_queue, int *condition) init_waitqueue_entry(&we, current); add_wait_queue(wait_queue, &we); - while (!(ACCESS_ONCE(*condition))) { + while (!(READ_ONCE(*condition))) { set_current_state(TASK_INTERRUPTIBLE); if (signal_pending(current)) goto out; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c index 5aecef870377..95a4bbedf557 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c @@ -19,43 +19,29 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" -#include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" #define MEMOPS_IDX MAX_BAR1_MAP_INDEX +#ifdef __BIG_ENDIAN_BITFIELD static inline void -octeon_toggle_bar1_swapmode(struct octeon_device *oct __attribute__((unused)), - u32 idx __attribute__((unused))) +octeon_toggle_bar1_swapmode(struct octeon_device *oct, u32 idx) { -#ifdef __BIG_ENDIAN_BITFIELD u32 mask; mask = oct->fn_list.bar1_idx_read(oct, idx); mask = (mask & 0x2) ? (mask & ~2) : (mask | 2); oct->fn_list.bar1_idx_write(oct, idx, mask); -#endif } +#else +#define octeon_toggle_bar1_swapmode(oct, idx) (oct = oct) +#endif static void octeon_pci_fastwrite(struct octeon_device *oct, u8 __iomem *mapped_addr, diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index b3abe5818fd3..fb820dc7fcb7 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -30,6 +30,20 @@ #include <linux/dma-mapping.h> #include <linux/ptp_clock_kernel.h> +#define LIO_MAX_MTU_SIZE (OCTNET_MAX_FRM_SIZE - OCTNET_FRM_HEADER_SIZE) +#define LIO_MIN_MTU_SIZE 68 + +struct oct_nic_stats_resp { + u64 rh; + struct oct_link_stats stats; + u64 status; +}; + +struct oct_nic_stats_ctrl { + struct completion complete; + struct net_device *netdev; +}; + /** LiquidIO per-interface network private data */ struct lio { /** State of the interface. Rx/Tx happens only in the RUNNING state. */ @@ -48,11 +62,11 @@ struct lio { */ int rxq; - /** Guards the glist */ - spinlock_t lock; + /** Guards each glist */ + spinlock_t *glist_lock; - /** Linked list of gather components */ - struct list_head glist; + /** Array of gather component linked lists */ + struct list_head *glist; /** Pointer to the NIC properties for the Octeon device this network * interface is associated with. @@ -67,6 +81,9 @@ struct lio { /** Link information sent by the core application for this interface. */ struct oct_link_info linfo; + /** counter of link changes */ + u64 link_changes; + /** Size of Tx queue for this octeon device. */ u32 tx_qsize; @@ -82,6 +99,12 @@ struct lio { /** Copy of Interface capabilities: TSO, TSO6, LRO, Chescksums . */ u64 dev_capability; + /* Copy of transmit encapsulation capabilities: + * TSO, TSO6, Checksums for this device for Kernel + * 3.10.0 onwards + */ + u64 enc_dev_capability; + /** Copy of beacaon reg in phy */ u32 phy_beacon_val; @@ -101,7 +124,6 @@ struct lio { /* work queue for txq status */ struct cavium_wq txq_status_wq; - }; #define LIO_SIZE (sizeof(struct lio)) @@ -111,8 +133,9 @@ struct lio { * \brief Enable or disable feature * @param netdev pointer to network device * @param cmd Command that just requires acknowledgment + * @param param1 Parameter to command */ -int liquidio_set_feature(struct net_device *netdev, int cmd); +int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1); /** * \brief Link control command completion callback @@ -131,14 +154,30 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr); */ void liquidio_set_ethtool_ops(struct net_device *netdev); -static inline void -*recv_buffer_alloc(struct octeon_device *oct __attribute__((unused)), - u32 q_no __attribute__((unused)), u32 size) -{ #define SKB_ADJ_MASK 0x3F #define SKB_ADJ (SKB_ADJ_MASK + 1) - struct sk_buff *skb = dev_alloc_skb(size + SKB_ADJ); +#define MIN_SKB_SIZE 256 /* 8 bytes and more - 8 bytes for PTP */ +#define LIO_RXBUFFER_SZ 2048 + +static inline void +*recv_buffer_alloc(struct octeon_device *oct, + struct octeon_skb_page_info *pg_info) +{ + struct page *page; + struct sk_buff *skb; + struct octeon_skb_page_info *skb_pg_info; + + page = alloc_page(GFP_ATOMIC | __GFP_COLD); + if (unlikely(!page)) + return NULL; + + skb = dev_alloc_skb(MIN_SKB_SIZE + SKB_ADJ); + if (unlikely(!skb)) { + __free_page(page); + pg_info->page = NULL; + return NULL; + } if ((unsigned long)skb->data & SKB_ADJ_MASK) { u32 r = SKB_ADJ - ((unsigned long)skb->data & SKB_ADJ_MASK); @@ -146,11 +185,151 @@ static inline void skb_reserve(skb, r); } + skb_pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + /* Get DMA info */ + pg_info->dma = dma_map_page(&oct->pci_dev->dev, page, 0, + PAGE_SIZE, DMA_FROM_DEVICE); + + /* Mapping failed!! */ + if (dma_mapping_error(&oct->pci_dev->dev, pg_info->dma)) { + __free_page(page); + dev_kfree_skb_any((struct sk_buff *)skb); + pg_info->page = NULL; + return NULL; + } + + pg_info->page = page; + pg_info->page_offset = 0; + skb_pg_info->page = page; + skb_pg_info->page_offset = 0; + skb_pg_info->dma = pg_info->dma; + return (void *)skb; } +static inline void +*recv_buffer_fast_alloc(u32 size) +{ + struct sk_buff *skb; + struct octeon_skb_page_info *skb_pg_info; + + skb = dev_alloc_skb(size + SKB_ADJ); + if (unlikely(!skb)) + return NULL; + + if ((unsigned long)skb->data & SKB_ADJ_MASK) { + u32 r = SKB_ADJ - ((unsigned long)skb->data & SKB_ADJ_MASK); + + skb_reserve(skb, r); + } + + skb_pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + skb_pg_info->page = NULL; + skb_pg_info->page_offset = 0; + skb_pg_info->dma = 0; + + return skb; +} + +static inline int +recv_buffer_recycle(struct octeon_device *oct, void *buf) +{ + struct octeon_skb_page_info *pg_info = buf; + + if (!pg_info->page) { + dev_err(&oct->pci_dev->dev, "%s: pg_info->page NULL\n", + __func__); + return -ENOMEM; + } + + if (unlikely(page_count(pg_info->page) != 1) || + unlikely(page_to_nid(pg_info->page) != numa_node_id())) { + dma_unmap_page(&oct->pci_dev->dev, + pg_info->dma, (PAGE_SIZE << 0), + DMA_FROM_DEVICE); + pg_info->dma = 0; + pg_info->page = NULL; + pg_info->page_offset = 0; + return -ENOMEM; + } + + /* Flip to other half of the buffer */ + if (pg_info->page_offset == 0) + pg_info->page_offset = LIO_RXBUFFER_SZ; + else + pg_info->page_offset = 0; + page_ref_inc(pg_info->page); + + return 0; +} + +static inline void +*recv_buffer_reuse(struct octeon_device *oct, void *buf) +{ + struct octeon_skb_page_info *pg_info = buf, *skb_pg_info; + struct sk_buff *skb; + + skb = dev_alloc_skb(MIN_SKB_SIZE + SKB_ADJ); + if (unlikely(!skb)) { + dma_unmap_page(&oct->pci_dev->dev, + pg_info->dma, (PAGE_SIZE << 0), + DMA_FROM_DEVICE); + return NULL; + } + + if ((unsigned long)skb->data & SKB_ADJ_MASK) { + u32 r = SKB_ADJ - ((unsigned long)skb->data & SKB_ADJ_MASK); + + skb_reserve(skb, r); + } + + skb_pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + skb_pg_info->page = pg_info->page; + skb_pg_info->page_offset = pg_info->page_offset; + skb_pg_info->dma = pg_info->dma; + + return skb; +} + +static inline void +recv_buffer_destroy(void *buffer, struct octeon_skb_page_info *pg_info) +{ + struct sk_buff *skb = (struct sk_buff *)buffer; + + put_page(pg_info->page); + pg_info->dma = 0; + pg_info->page = NULL; + pg_info->page_offset = 0; + + if (skb) + dev_kfree_skb_any(skb); +} + static inline void recv_buffer_free(void *buffer) { + struct sk_buff *skb = (struct sk_buff *)buffer; + struct octeon_skb_page_info *pg_info; + + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + + if (pg_info->page) { + put_page(pg_info->page); + pg_info->dma = 0; + pg_info->page = NULL; + pg_info->page_offset = 0; + } + + dev_kfree_skb_any((struct sk_buff *)buffer); +} + +static inline void +recv_buffer_fast_free(void *buffer) +{ + dev_kfree_skb_any((struct sk_buff *)buffer); +} + +static inline void tx_buffer_free(void *buffer) +{ dev_kfree_skb_any((struct sk_buff *)buffer); } @@ -159,7 +338,17 @@ static inline void recv_buffer_free(void *buffer) #define lio_dma_free(oct, size, virt_addr, dma_addr) \ dma_free_coherent(&oct->pci_dev->dev, size, virt_addr, dma_addr) -#define get_rbd(ptr) (((struct sk_buff *)(ptr))->data) +static inline +void *get_rbd(struct sk_buff *skb) +{ + struct octeon_skb_page_info *pg_info; + unsigned char *va; + + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + va = page_address(pg_info->page) + pg_info->page_offset; + + return va; +} static inline u64 lio_map_ring_info(struct octeon_droq *droq, u32 i) @@ -170,7 +359,7 @@ lio_map_ring_info(struct octeon_droq *droq, u32 i) dma_addr = dma_map_single(&oct->pci_dev->dev, &droq->info_list[i], OCT_DROQ_INFO_SIZE, DMA_FROM_DEVICE); - BUG_ON(dma_mapping_error(&oct->pci_dev->dev, dma_addr)); + WARN_ON(dma_mapping_error(&oct->pci_dev->dev, dma_addr)); return (u64)dma_addr; } @@ -183,33 +372,44 @@ lio_unmap_ring_info(struct pci_dev *pci_dev, } static inline u64 -lio_map_ring(struct pci_dev *pci_dev, - void *buf, u32 size) +lio_map_ring(void *buf) { dma_addr_t dma_addr; - dma_addr = dma_map_single(&pci_dev->dev, get_rbd(buf), size, - DMA_FROM_DEVICE); + struct sk_buff *skb = (struct sk_buff *)buf; + struct octeon_skb_page_info *pg_info; - BUG_ON(dma_mapping_error(&pci_dev->dev, dma_addr)); + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + if (!pg_info->page) { + pr_err("%s: pg_info->page NULL\n", __func__); + WARN_ON(1); + } + + /* Get DMA info */ + dma_addr = pg_info->dma; + if (!pg_info->dma) { + pr_err("%s: ERROR it should be already available\n", + __func__); + WARN_ON(1); + } + dma_addr += pg_info->page_offset; return (u64)dma_addr; } static inline void lio_unmap_ring(struct pci_dev *pci_dev, - u64 buf_ptr, u32 size) + u64 buf_ptr) + { - dma_unmap_single(&pci_dev->dev, - buf_ptr, size, - DMA_FROM_DEVICE); + dma_unmap_page(&pci_dev->dev, + buf_ptr, (PAGE_SIZE << 0), + DMA_FROM_DEVICE); } -static inline void *octeon_fast_packet_alloc(struct octeon_device *oct, - struct octeon_droq *droq, - u32 q_no, u32 size) +static inline void *octeon_fast_packet_alloc(u32 size) { - return recv_buffer_alloc(oct, q_no, size); + return recv_buffer_fast_alloc(size); } static inline void octeon_fast_packet_next(struct octeon_droq *droq, diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c index 1a0191549cb3..166727be928f 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c @@ -19,14 +19,9 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> #include <linux/interrupt.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" @@ -34,21 +29,14 @@ #include "octeon_device.h" #include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" void * octeon_alloc_soft_command_resp(struct octeon_device *oct, - struct octeon_instr_64B *cmd, - size_t rdatasize) + union octeon_instr_64B *cmd, + u32 rdatasize) { struct octeon_soft_command *sc; - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; struct octeon_instr_rdp *rdp; @@ -59,24 +47,25 @@ octeon_alloc_soft_command_resp(struct octeon_device *oct, return NULL; /* Copy existing command structure into the soft command */ - memcpy(&sc->cmd, cmd, sizeof(struct octeon_instr_64B)); + memcpy(&sc->cmd, cmd, sizeof(union octeon_instr_64B)); /* Add in the response related fields. Opcode and Param are already * there. */ - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - ih->fsz = 40; /* irh + ossp[0] + ossp[1] + rdp + rptr = 40 bytes */ + ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; + ih2->fsz = 40; /* irh + ossp[0] + ossp[1] + rdp + rptr = 40 bytes */ - irh = (struct octeon_instr_irh *)&sc->cmd.irh; irh->rflag = 1; /* a response is required */ - irh->len = 4; /* means four 64-bit words immediately follow irh */ - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; rdp->pcie_port = oct->pcie_port; rdp->rlen = rdatasize; *sc->status_word = COMPLETION_WORD_INIT; + sc->cmd.cmd2.rptr = sc->dmarptr; + sc->wait_time = 1000; sc->timeout = jiffies + sc->wait_time; @@ -119,12 +108,11 @@ static void octnet_link_ctrl_callback(struct octeon_device *oct, static inline struct octeon_soft_command *octnic_alloc_ctrl_pkt_sc(struct octeon_device *oct, - struct octnic_ctrl_pkt *nctrl, - struct octnic_ctrl_params nparams) + struct octnic_ctrl_pkt *nctrl) { struct octeon_soft_command *sc = NULL; u8 *data; - size_t rdatasize; + u32 rdatasize; u32 uddsize = 0, datasize = 0; uddsize = (u32)(nctrl->ncmd.s.more * 8); @@ -143,7 +131,7 @@ static inline struct octeon_soft_command data = (u8 *)sc->virtdptr; - memcpy(data, &nctrl->ncmd, OCTNET_CMD_SIZE); + memcpy(data, &nctrl->ncmd, OCTNET_CMD_SIZE); octeon_swap_8B_data((u64 *)data, (OCTNET_CMD_SIZE >> 3)); @@ -152,6 +140,8 @@ static inline struct octeon_soft_command memcpy(data + OCTNET_CMD_SIZE, nctrl->udd, uddsize); } + sc->iq_no = (u32)nctrl->iq_no; + octeon_prepare_soft_command(oct, sc, OPCODE_NIC, OPCODE_NIC_CMD, 0, 0, 0); @@ -164,26 +154,41 @@ static inline struct octeon_soft_command int octnet_send_nic_ctrl_pkt(struct octeon_device *oct, - struct octnic_ctrl_pkt *nctrl, - struct octnic_ctrl_params nparams) + struct octnic_ctrl_pkt *nctrl) { int retval; struct octeon_soft_command *sc = NULL; - sc = octnic_alloc_ctrl_pkt_sc(oct, nctrl, nparams); + spin_lock_bh(&oct->cmd_resp_wqlock); + /* Allow only rx ctrl command to stop traffic on the chip + * during offline operations + */ + if ((oct->cmd_resp_state == OCT_DRV_OFFLINE) && + (nctrl->ncmd.s.cmd != OCTNET_CMD_RX_CTL)) { + spin_unlock_bh(&oct->cmd_resp_wqlock); + dev_err(&oct->pci_dev->dev, + "%s cmd:%d not processed since driver offline\n", + __func__, nctrl->ncmd.s.cmd); + return -1; + } + + sc = octnic_alloc_ctrl_pkt_sc(oct, nctrl); if (!sc) { dev_err(&oct->pci_dev->dev, "%s soft command alloc failed\n", __func__); + spin_unlock_bh(&oct->cmd_resp_wqlock); return -1; } retval = octeon_send_soft_command(oct, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { octeon_free_soft_command(oct, sc); - dev_err(&oct->pci_dev->dev, "%s soft command send failed status: %x\n", - __func__, retval); + dev_err(&oct->pci_dev->dev, "%s soft command:%d send failed status: %x\n", + __func__, nctrl->ncmd.s.cmd, retval); + spin_unlock_bh(&oct->cmd_resp_wqlock); return -1; } + spin_unlock_bh(&oct->cmd_resp_wqlock); return retval; } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h index 0238857c8105..b71a2bbe4bee 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h @@ -52,6 +52,9 @@ struct octnic_ctrl_pkt { /** Additional data that may be needed by some commands. */ u64 udd[MAX_NCTRL_UDD]; + /** Input queue to use to send this command. */ + u64 iq_no; + /** Time to wait for Octeon software to respond to this control command. * If wait_time is 0, OSI assumes no response is expected. */ @@ -82,7 +85,7 @@ struct octnic_data_pkt { u32 datasize; /** Command to be passed to the Octeon device software. */ - struct octeon_instr_64B cmd; + union octeon_instr_64B cmd; /** Input queue to use to send this command. */ u32 q_no; @@ -94,15 +97,14 @@ struct octnic_data_pkt { */ union octnic_cmd_setup { struct { - u32 ifidx:8; - u32 cksum_offset:7; + u32 iq_no:8; u32 gather:1; u32 timestamp:1; - u32 ipv4opts_ipv6exthdr:2; u32 ip_csum:1; + u32 transport_csum:1; u32 tnl_csum:1; + u32 rsvd:19; - u32 rsvd:11; union { u32 datasize; u32 gatherptrs; @@ -113,79 +115,146 @@ union octnic_cmd_setup { }; -struct octnic_ctrl_params { - u32 resp_order; -}; - static inline int octnet_iq_is_full(struct octeon_device *oct, u32 q_no) { return ((u32)atomic_read(&oct->instr_queue[q_no]->instr_pending) >= (oct->instr_queue[q_no]->max_count - 2)); } -/** Utility function to prepare a 64B NIC instruction based on a setup command - * @param cmd - pointer to instruction to be filled in. - * @param setup - pointer to the setup structure - * @param q_no - which queue for back pressure - * - * Assumes the cmd instruction is pre-allocated, but no fields are filled in. - */ static inline void -octnet_prepare_pci_cmd(struct octeon_instr_64B *cmd, - union octnic_cmd_setup *setup, u32 tag) +octnet_prepare_pci_cmd_o2(struct octeon_device *oct, + union octeon_instr_64B *cmd, + union octnic_cmd_setup *setup, u32 tag) { - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; union octnic_packet_params packet_params; + int port; - memset(cmd, 0, sizeof(struct octeon_instr_64B)); + memset(cmd, 0, sizeof(union octeon_instr_64B)); - ih = (struct octeon_instr_ih *)&cmd->ih; + ih2 = (struct octeon_instr_ih2 *)&cmd->cmd2.ih2; /* assume that rflag is cleared so therefore front data will only have - * irh and ossp[1] and ossp[2] for a total of 24 bytes + * irh and ossp[0], ossp[1] for a total of 32 bytes */ - ih->fsz = 24; + ih2->fsz = 24; + + ih2->tagtype = ORDERED_TAG; + ih2->grp = DEFAULT_POW_GRP; - ih->tagtype = ORDERED_TAG; - ih->grp = DEFAULT_POW_GRP; + port = (int)oct->instr_queue[setup->s.iq_no]->txpciq.s.port; if (tag) - ih->tag = tag; + ih2->tag = tag; else - ih->tag = LIO_DATA(setup->s.ifidx); + ih2->tag = LIO_DATA(port); - ih->raw = 1; - ih->qos = (setup->s.ifidx & 3) + 4; /* map qos based on interface */ + ih2->raw = 1; + ih2->qos = (port & 3) + 4; /* map qos based on interface */ if (!setup->s.gather) { - ih->dlengsz = setup->s.u.datasize; + ih2->dlengsz = setup->s.u.datasize; } else { - ih->gather = 1; - ih->dlengsz = setup->s.u.gatherptrs; + ih2->gather = 1; + ih2->dlengsz = setup->s.u.gatherptrs; } - irh = (struct octeon_instr_irh *)&cmd->irh; + irh = (struct octeon_instr_irh *)&cmd->cmd2.irh; irh->opcode = OPCODE_NIC; irh->subcode = OPCODE_NIC_NW_DATA; packet_params.u32 = 0; - if (setup->s.cksum_offset) { - packet_params.s.csoffset = setup->s.cksum_offset; - packet_params.s.ipv4opts_ipv6exthdr = - setup->s.ipv4opts_ipv6exthdr; + packet_params.s.ip_csum = setup->s.ip_csum; + packet_params.s.transport_csum = setup->s.transport_csum; + packet_params.s.tnl_csum = setup->s.tnl_csum; + packet_params.s.tsflag = setup->s.timestamp; + + irh->ossp = packet_params.u32; +} + +static inline void +octnet_prepare_pci_cmd_o3(struct octeon_device *oct, + union octeon_instr_64B *cmd, + union octnic_cmd_setup *setup, u32 tag) +{ + struct octeon_instr_irh *irh; + struct octeon_instr_ih3 *ih3; + struct octeon_instr_pki_ih3 *pki_ih3; + union octnic_packet_params packet_params; + int port; + + memset(cmd, 0, sizeof(union octeon_instr_64B)); + + ih3 = (struct octeon_instr_ih3 *)&cmd->cmd3.ih3; + pki_ih3 = (struct octeon_instr_pki_ih3 *)&cmd->cmd3.pki_ih3; + + /* assume that rflag is cleared so therefore front data will only have + * irh and ossp[1] and ossp[2] for a total of 24 bytes + */ + ih3->pkind = oct->instr_queue[setup->s.iq_no]->txpciq.s.pkind; + /*PKI IH*/ + ih3->fsz = 24 + 8; + + if (!setup->s.gather) { + ih3->dlengsz = setup->s.u.datasize; + } else { + ih3->gather = 1; + ih3->dlengsz = setup->s.u.gatherptrs; } + pki_ih3->w = 1; + pki_ih3->raw = 1; + pki_ih3->utag = 1; + pki_ih3->utt = 1; + pki_ih3->uqpg = oct->instr_queue[setup->s.iq_no]->txpciq.s.use_qpg; + + port = (int)oct->instr_queue[setup->s.iq_no]->txpciq.s.port; + + if (tag) + pki_ih3->tag = tag; + else + pki_ih3->tag = LIO_DATA(port); + + pki_ih3->tagtype = ORDERED_TAG; + pki_ih3->qpg = oct->instr_queue[setup->s.iq_no]->txpciq.s.qpg; + pki_ih3->pm = 0x7; /*0x7 - meant for Parse nothing, uninterpreted*/ + pki_ih3->sl = 8; /* sl will be sizeof(pki_ih3)*/ + + irh = (struct octeon_instr_irh *)&cmd->cmd3.irh; + + irh->opcode = OPCODE_NIC; + irh->subcode = OPCODE_NIC_NW_DATA; + + packet_params.u32 = 0; + packet_params.s.ip_csum = setup->s.ip_csum; + packet_params.s.transport_csum = setup->s.transport_csum; packet_params.s.tnl_csum = setup->s.tnl_csum; - packet_params.s.ifidx = setup->s.ifidx; packet_params.s.tsflag = setup->s.timestamp; irh->ossp = packet_params.u32; } +/** Utility function to prepare a 64B NIC instruction based on a setup command + * @param cmd - pointer to instruction to be filled in. + * @param setup - pointer to the setup structure + * @param q_no - which queue for back pressure + * + * Assumes the cmd instruction is pre-allocated, but no fields are filled in. + */ +static inline void +octnet_prepare_pci_cmd(struct octeon_device *oct, union octeon_instr_64B *cmd, + union octnic_cmd_setup *setup, u32 tag) +{ + if (OCTEON_CN6XXX(oct)) + octnet_prepare_pci_cmd_o2(oct, cmd, setup, tag); + else + octnet_prepare_pci_cmd_o3(oct, cmd, setup, tag); +} + /** Allocate and a soft command with space for a response immediately following * the commnad. * @param oct - octeon device pointer @@ -198,8 +267,8 @@ octnet_prepare_pci_cmd(struct octeon_instr_64B *cmd, */ void * octeon_alloc_soft_command_resp(struct octeon_device *oct, - struct octeon_instr_64B *cmd, - size_t rdatasize); + union octeon_instr_64B *cmd, + u32 rdatasize); /** Send a NIC data packet to the device * @param oct - octeon device pointer @@ -214,14 +283,11 @@ int octnet_send_nic_data_pkt(struct octeon_device *oct, /** Send a NIC control packet to the device * @param oct - octeon device pointer * @param nctrl - control structure with command, timout, and callback info - * @param nparams - response control structure - * * @returns IQ_FAILED if it failed to add to the input queue. IQ_STOP if it the * queue should be stopped, and IQ_SEND_OK if it sent okay. */ int octnet_send_nic_ctrl_pkt(struct octeon_device *oct, - struct octnic_ctrl_pkt *nctrl, - struct octnic_ctrl_params nparams); + struct octnic_ctrl_pkt *nctrl); #endif diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index a2a24652c8f3..d32492f185ff 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -19,28 +19,17 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> #include <linux/vmalloc.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" #include "octeon_network.h" -#include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" #define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \ (octeon_dev_ptr->instr_queue[iq_no]->stats.field += count) @@ -51,7 +40,7 @@ struct iq_post_status { }; static void check_db_timeout(struct work_struct *work); -static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no); +static void __check_db_timeout(struct octeon_device *oct, u64 iq_no); static void (*reqtype_free_fn[MAX_OCTEON_DEVICES][REQTYPE_LAST + 1]) (void *); @@ -69,12 +58,16 @@ static inline int IQ_INSTR_MODE_64B(struct octeon_device *oct, int iq_no) /* Return 0 on success, 1 on failure */ int octeon_init_instr_queue(struct octeon_device *oct, - u32 iq_no, u32 num_descs) + union oct_txpciq txpciq, + u32 num_descs) { struct octeon_instr_queue *iq; struct octeon_iq_config *conf = NULL; + u32 iq_no = (u32)txpciq.s.q_no; u32 q_size; struct cavium_wq *db_wq; + int orig_node = dev_to_node(&oct->pci_dev->dev); + int numa_node = cpu_to_node(iq_no % num_online_cpus()); if (OCTEON_CN6XXX(oct)) conf = &(CFG_GET_IQ_CFG(CHIP_FIELD(oct, cn6xxx, conf))); @@ -95,9 +88,15 @@ int octeon_init_instr_queue(struct octeon_device *oct, q_size = (u32)conf->instr_type * num_descs; iq = oct->instr_queue[iq_no]; + iq->oct_dev = oct; + set_dev_node(&oct->pci_dev->dev, numa_node); iq->base_addr = lio_dma_alloc(oct, q_size, (dma_addr_t *)&iq->base_addr_dma); + set_dev_node(&oct->pci_dev->dev, orig_node); + if (!iq->base_addr) + iq->base_addr = lio_dma_alloc(oct, q_size, + (dma_addr_t *)&iq->base_addr_dma); if (!iq->base_addr) { dev_err(&oct->pci_dev->dev, "Cannot allocate memory for instr queue %d\n", iq_no); @@ -109,7 +108,11 @@ int octeon_init_instr_queue(struct octeon_device *oct, /* Initialize a list to holds requests that have been posted to Octeon * but has yet to be fetched by octeon */ - iq->request_list = vmalloc(sizeof(*iq->request_list) * num_descs); + iq->request_list = vmalloc_node((sizeof(*iq->request_list) * num_descs), + numa_node); + if (!iq->request_list) + iq->request_list = vmalloc(sizeof(*iq->request_list) * + num_descs); if (!iq->request_list) { lio_dma_free(oct, q_size, iq->base_addr, iq->base_addr_dma); dev_err(&oct->pci_dev->dev, "Alloc failed for IQ[%d] nr free list\n", @@ -122,7 +125,7 @@ int octeon_init_instr_queue(struct octeon_device *oct, dev_dbg(&oct->pci_dev->dev, "IQ[%d]: base: %p basedma: %llx count: %d\n", iq_no, iq->base_addr, iq->base_addr_dma, iq->max_count); - iq->iq_no = iq_no; + iq->txpciq.u64 = txpciq.u64; iq->fill_threshold = (u32)conf->db_min; iq->fill_cnt = 0; iq->host_write_index = 0; @@ -135,8 +138,11 @@ int octeon_init_instr_queue(struct octeon_device *oct, /* Initialize the spinlock for this instruction queue */ spin_lock_init(&iq->lock); + spin_lock_init(&iq->post_lock); - oct->io_qmask.iq |= (1 << iq_no); + spin_lock_init(&iq->iq_flush_running_lock); + + oct->io_qmask.iq |= (1ULL << iq_no); /* Set the 32B/64B mode for each input queue */ oct->io_qmask.iq64B |= ((conf->instr_type == 64) << iq_no); @@ -144,7 +150,9 @@ int octeon_init_instr_queue(struct octeon_device *oct, oct->fn_list.setup_iq_regs(oct, iq_no); - oct->check_db_wq[iq_no].wq = create_workqueue("check_iq_db"); + oct->check_db_wq[iq_no].wq = alloc_workqueue("check_iq_db", + WQ_MEM_RECLAIM, + 0); if (!oct->check_db_wq[iq_no].wq) { lio_dma_free(oct, q_size, iq->base_addr, iq->base_addr_dma); dev_err(&oct->pci_dev->dev, "check db wq create failed for iq %d\n", @@ -168,7 +176,6 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no) struct octeon_instr_queue *iq = oct->instr_queue[iq_no]; cancel_delayed_work_sync(&oct->check_db_wq[iq_no].wk.work); - flush_workqueue(oct->check_db_wq[iq_no].wq); destroy_workqueue(oct->check_db_wq[iq_no].wq); if (OCTEON_CN6XXX(oct)) @@ -188,26 +195,38 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no) /* Return 0 on success, 1 on failure */ int octeon_setup_iq(struct octeon_device *oct, - u32 iq_no, + int ifidx, + int q_index, + union oct_txpciq txpciq, u32 num_descs, void *app_ctx) { + u32 iq_no = (u32)txpciq.s.q_no; + int numa_node = cpu_to_node(iq_no % num_online_cpus()); + if (oct->instr_queue[iq_no]) { dev_dbg(&oct->pci_dev->dev, "IQ is in use. Cannot create the IQ: %d again\n", iq_no); + oct->instr_queue[iq_no]->txpciq.u64 = txpciq.u64; oct->instr_queue[iq_no]->app_ctx = app_ctx; return 0; } oct->instr_queue[iq_no] = - vmalloc(sizeof(struct octeon_instr_queue)); + vmalloc_node(sizeof(struct octeon_instr_queue), numa_node); + if (!oct->instr_queue[iq_no]) + oct->instr_queue[iq_no] = + vmalloc(sizeof(struct octeon_instr_queue)); if (!oct->instr_queue[iq_no]) return 1; memset(oct->instr_queue[iq_no], 0, sizeof(struct octeon_instr_queue)); + oct->instr_queue[iq_no]->q_index = q_index; oct->instr_queue[iq_no]->app_ctx = app_ctx; - if (octeon_init_instr_queue(oct, iq_no, num_descs)) { + oct->instr_queue[iq_no]->ifidx = ifidx; + + if (octeon_init_instr_queue(oct, txpciq, num_descs)) { vfree(oct->instr_queue[iq_no]); oct->instr_queue[iq_no] = NULL; return 1; @@ -226,8 +245,8 @@ int lio_wait_for_instr_fetch(struct octeon_device *oct) instr_cnt = 0; /*for (i = 0; i < oct->num_iqs; i++) {*/ - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; pending = atomic_read(&oct-> @@ -271,40 +290,8 @@ static inline void __copy_cmd_into_iq(struct octeon_instr_queue *iq, memcpy(iqptr, cmd, cmdsize); } -static inline int -__post_command(struct octeon_device *octeon_dev __attribute__((unused)), - struct octeon_instr_queue *iq, - u32 force_db __attribute__((unused)), u8 *cmd) -{ - u32 index = -1; - - /* This ensures that the read index does not wrap around to the same - * position if queue gets full before Octeon could fetch any instr. - */ - if (atomic_read(&iq->instr_pending) >= (s32)(iq->max_count - 1)) - return -1; - - __copy_cmd_into_iq(iq, cmd); - - /* "index" is returned, host_write_index is modified. */ - index = iq->host_write_index; - INCR_INDEX_BY1(iq->host_write_index, iq->max_count); - iq->fill_cnt++; - - /* Flush the command into memory. We need to be sure the data is in - * memory before indicating that the instruction is pending. - */ - wmb(); - - atomic_inc(&iq->instr_pending); - - return index; -} - static inline struct iq_post_status -__post_command2(struct octeon_device *octeon_dev __attribute__((unused)), - struct octeon_instr_queue *iq, - u32 force_db __attribute__((unused)), u8 *cmd) +__post_command2(struct octeon_instr_queue *iq, u8 *cmd) { struct iq_post_status st; @@ -362,17 +349,19 @@ __add_to_request_list(struct octeon_instr_queue *iq, iq->request_list[idx].reqtype = reqtype; } +/* Can only run in process context */ int lio_process_iq_request_list(struct octeon_device *oct, - struct octeon_instr_queue *iq) + struct octeon_instr_queue *iq, u32 napi_budget) { int reqtype; void *buf; u32 old = iq->flush_index; u32 inst_count = 0; - unsigned pkts_compl = 0, bytes_compl = 0; + unsigned int pkts_compl = 0, bytes_compl = 0; struct octeon_soft_command *sc; struct octeon_instr_irh *irh; + unsigned long flags; while (old != iq->octeon_read_index) { reqtype = iq->request_list[old].reqtype; @@ -394,7 +383,7 @@ lio_process_iq_request_list(struct octeon_device *oct, case REQTYPE_SOFT_COMMAND: sc = buf; - irh = (struct octeon_instr_irh *)&sc->cmd.irh; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; if (irh->rflag) { /* We're expecting a response from Octeon. * It's up to lio_process_ordered_list() to @@ -402,17 +391,22 @@ lio_process_iq_request_list(struct octeon_device *oct, * command response list because we expect * a response from Octeon. */ - spin_lock_bh(&oct->response_list - [OCTEON_ORDERED_SC_LIST].lock); + spin_lock_irqsave + (&oct->response_list + [OCTEON_ORDERED_SC_LIST].lock, + flags); atomic_inc(&oct->response_list [OCTEON_ORDERED_SC_LIST]. pending_req_count); list_add_tail(&sc->node, &oct->response_list [OCTEON_ORDERED_SC_LIST].head); - spin_unlock_bh(&oct->response_list - [OCTEON_ORDERED_SC_LIST].lock); + spin_unlock_irqrestore + (&oct->response_list + [OCTEON_ORDERED_SC_LIST].lock, + flags); } else { if (sc->callback) { + /* This callback must not sleep */ sc->callback(oct, OCTEON_REQUEST_DONE, sc->callback_arg); } @@ -430,6 +424,9 @@ lio_process_iq_request_list(struct octeon_device *oct, skip_this: inst_count++; INCR_INDEX_BY1(old, iq->max_count); + + if ((napi_budget) && (inst_count >= napi_budget)) + break; } if (bytes_compl) octeon_report_tx_completion_to_bql(iq->app_ctx, pkts_compl, @@ -439,38 +436,63 @@ lio_process_iq_request_list(struct octeon_device *oct, return inst_count; } -static inline void -update_iq_indices(struct octeon_device *oct, struct octeon_instr_queue *iq) +/* Can only be called from process context */ +int +octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, + u32 pending_thresh, u32 napi_budget) { u32 inst_processed = 0; + u32 tot_inst_processed = 0; + int tx_done = 1; - /* Calculate how many commands Octeon has read and move the read index - * accordingly. - */ - iq->octeon_read_index = oct->fn_list.update_iq_read_idx(oct, iq); + if (!spin_trylock(&iq->iq_flush_running_lock)) + return tx_done; - /* Move the NORESPONSE requests to the per-device completion list. */ - if (iq->flush_index != iq->octeon_read_index) - inst_processed = lio_process_iq_request_list(oct, iq); + spin_lock_bh(&iq->lock); - if (inst_processed) { - atomic_sub(inst_processed, &iq->instr_pending); - iq->stats.instr_processed += inst_processed; - } -} + iq->octeon_read_index = oct->fn_list.update_iq_read_idx(iq); -static void -octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, - u32 pending_thresh) -{ if (atomic_read(&iq->instr_pending) >= (s32)pending_thresh) { - spin_lock_bh(&iq->lock); - update_iq_indices(oct, iq); - spin_unlock_bh(&iq->lock); + do { + /* Process any outstanding IQ packets. */ + if (iq->flush_index == iq->octeon_read_index) + break; + + if (napi_budget) + inst_processed = lio_process_iq_request_list + (oct, iq, + napi_budget - tot_inst_processed); + else + inst_processed = + lio_process_iq_request_list(oct, iq, 0); + + if (inst_processed) { + atomic_sub(inst_processed, &iq->instr_pending); + iq->stats.instr_processed += inst_processed; + } + + tot_inst_processed += inst_processed; + inst_processed = 0; + + } while (tot_inst_processed < napi_budget); + + if (napi_budget && (tot_inst_processed >= napi_budget)) + tx_done = 0; } + + iq->last_db_time = jiffies; + + spin_unlock_bh(&iq->lock); + + spin_unlock(&iq->iq_flush_running_lock); + + return tx_done; } -static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no) +/* Process instruction queue after timeout. + * This routine gets called from a workqueue or when removing the module. + */ +static void __check_db_timeout(struct octeon_device *oct, u64 iq_no) { struct octeon_instr_queue *iq; u64 next_time; @@ -481,24 +503,17 @@ static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no) if (!iq) return; + /* return immediately, if no work pending */ + if (!atomic_read(&iq->instr_pending)) + return; /* If jiffies - last_db_time < db_timeout do nothing */ next_time = iq->last_db_time + iq->db_timeout; if (!time_after(jiffies, (unsigned long)next_time)) return; iq->last_db_time = jiffies; - /* Get the lock and prevent tasklets. This routine gets called from - * the poll thread. Instructions can now be posted in tasklet context - */ - spin_lock_bh(&iq->lock); - if (iq->fill_cnt != 0) - ring_doorbell(oct, iq); - - spin_unlock_bh(&iq->lock); - /* Flush the instruction queue */ - if (iq->do_auto_flush) - octeon_flush_iq(oct, iq, 1); + octeon_flush_iq(oct, iq, 1, 0); } /* Called by the Poll thread at regular intervals to check the instruction @@ -508,11 +523,12 @@ static void check_db_timeout(struct work_struct *work) { struct cavium_wk *wk = (struct cavium_wk *)work; struct octeon_device *oct = (struct octeon_device *)wk->ctxptr; - unsigned long iq_no = wk->ctxul; + u64 iq_no = wk->ctxul; struct cavium_wq *db_wq = &oct->check_db_wq[iq_no]; + u32 delay = 10; __check_db_timeout(oct, iq_no); - queue_delayed_work(db_wq->wq, &db_wq->wk.work, msecs_to_jiffies(1)); + queue_delayed_work(db_wq->wq, &db_wq->wk.work, msecs_to_jiffies(delay)); } int @@ -523,9 +539,12 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no, struct iq_post_status st; struct octeon_instr_queue *iq = oct->instr_queue[iq_no]; - spin_lock_bh(&iq->lock); + /* Get the lock and prevent other tasks and tx interrupt handler from + * running. + */ + spin_lock_bh(&iq->post_lock); - st = __post_command2(oct, iq, force_db, cmd); + st = __post_command2(iq, cmd); if (st.status != IQ_SEND_FAILED) { octeon_report_sent_bytes_to_bql(buf, reqtype); @@ -533,16 +552,19 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no, INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, bytes_sent, datasize); INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_posted, 1); - if (iq->fill_cnt >= iq->fill_threshold || force_db) + if (force_db) ring_doorbell(oct, iq); } else { INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_dropped, 1); } - spin_unlock_bh(&iq->lock); + spin_unlock_bh(&iq->post_lock); - if (iq->do_auto_flush) - octeon_flush_iq(oct, iq, 2); + /* This is only done here to expedite packets being flushed + * for cases where there are no IQ completion interrupts. + */ + /*if (iq->do_auto_flush)*/ + /* octeon_flush_iq(oct, iq, 2, 0);*/ return st.status; } @@ -557,82 +579,75 @@ octeon_prepare_soft_command(struct octeon_device *oct, u64 ossp1) { struct octeon_config *oct_cfg; - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; struct octeon_instr_rdp *rdp; - BUG_ON(opcode > 15); - BUG_ON(subcode > 127); + WARN_ON(opcode > 15); + WARN_ON(subcode > 127); oct_cfg = octeon_get_conf(oct); - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - ih->tagtype = ATOMIC_TAG; - ih->tag = LIO_CONTROL; - ih->raw = 1; - ih->grp = CFG_GET_CTRL_Q_GRP(oct_cfg); + ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; + ih2->tagtype = ATOMIC_TAG; + ih2->tag = LIO_CONTROL; + ih2->raw = 1; + ih2->grp = CFG_GET_CTRL_Q_GRP(oct_cfg); if (sc->datasize) { - ih->dlengsz = sc->datasize; - ih->rs = 1; + ih2->dlengsz = sc->datasize; + ih2->rs = 1; } - irh = (struct octeon_instr_irh *)&sc->cmd.irh; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; irh->opcode = opcode; irh->subcode = subcode; /* opcode/subcode specific parameters (ossp) */ irh->ossp = irh_ossp; - sc->cmd.ossp[0] = ossp0; - sc->cmd.ossp[1] = ossp1; + sc->cmd.cmd2.ossp[0] = ossp0; + sc->cmd.cmd2.ossp[1] = ossp1; if (sc->rdatasize) { - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; rdp->pcie_port = oct->pcie_port; rdp->rlen = sc->rdatasize; irh->rflag = 1; - irh->len = 4; - ih->fsz = 40; /* irh+ossp[0]+ossp[1]+rdp+rptr = 40 bytes */ + ih2->fsz = 40; /* irh+ossp[0]+ossp[1]+rdp+rptr = 40 bytes */ } else { irh->rflag = 0; - irh->len = 2; - ih->fsz = 24; /* irh + ossp[0] + ossp[1] = 24 bytes */ + ih2->fsz = 24; /* irh + ossp[0] + ossp[1] = 24 bytes */ } - - while (!(oct->io_qmask.iq & (1 << sc->iq_no))) - sc->iq_no++; } int octeon_send_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc) { - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; - struct octeon_instr_rdp *rdp; + u32 len; - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - if (ih->dlengsz) { - BUG_ON(!sc->dmadptr); - sc->cmd.dptr = sc->dmadptr; + ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; + if (ih2->dlengsz) { + WARN_ON(!sc->dmadptr); + sc->cmd.cmd2.dptr = sc->dmadptr; } - - irh = (struct octeon_instr_irh *)&sc->cmd.irh; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; if (irh->rflag) { - BUG_ON(!sc->dmarptr); - BUG_ON(!sc->status_word); + WARN_ON(!sc->dmarptr); + WARN_ON(!sc->status_word); *sc->status_word = COMPLETION_WORD_INIT; - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; - - sc->cmd.rptr = sc->dmarptr; + sc->cmd.cmd2.rptr = sc->dmarptr; } + len = (u32)ih2->dlengsz; if (sc->wait_time) sc->timeout = jiffies + sc->wait_time; - return octeon_send_command(oct, sc->iq_no, 1, &sc->cmd, sc, - (u32)ih->dlengsz, REQTYPE_SOFT_COMMAND); + return (octeon_send_command(oct, sc->iq_no, 1, &sc->cmd, sc, + len, REQTYPE_SOFT_COMMAND)); } int octeon_setup_sc_buffer_pool(struct octeon_device *oct) @@ -667,7 +682,7 @@ int octeon_free_sc_buffer_pool(struct octeon_device *oct) struct list_head *tmp, *tmp2; struct octeon_soft_command *sc; - spin_lock(&oct->sc_buf_pool.lock); + spin_lock_bh(&oct->sc_buf_pool.lock); list_for_each_safe(tmp, tmp2, &oct->sc_buf_pool.head) { list_del(tmp); @@ -679,7 +694,7 @@ int octeon_free_sc_buffer_pool(struct octeon_device *oct) INIT_LIST_HEAD(&oct->sc_buf_pool.head); - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); return 0; } @@ -695,13 +710,13 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc = NULL; struct list_head *tmp; - BUG_ON((offset + datasize + rdatasize + ctxsize) > + WARN_ON((offset + datasize + rdatasize + ctxsize) > SOFT_COMMAND_BUFFER_SIZE); - spin_lock(&oct->sc_buf_pool.lock); + spin_lock_bh(&oct->sc_buf_pool.lock); if (list_empty(&oct->sc_buf_pool.head)) { - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); return NULL; } @@ -712,7 +727,7 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, atomic_inc(&oct->sc_buf_pool.alloc_buf_count); - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); sc = (struct octeon_soft_command *)tmp; @@ -742,7 +757,7 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, offset = (offset + datasize + 127) & 0xffffff80; if (rdatasize) { - BUG_ON(rdatasize < 16); + WARN_ON(rdatasize < 16); sc->virtrptr = (u8 *)sc + offset; sc->dmarptr = dma_addr + offset; sc->rdatasize = rdatasize; @@ -755,11 +770,11 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, void octeon_free_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc) { - spin_lock(&oct->sc_buf_pool.lock); + spin_lock_bh(&oct->sc_buf_pool.lock); list_add_tail(&sc->node, &oct->sc_buf_pool.head); atomic_dec(&oct->sc_buf_pool.alloc_buf_count); - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); } diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c index 091f537a946e..709049e36627 100644 --- a/drivers/net/ethernet/cavium/liquidio/response_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c @@ -19,28 +19,14 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include <linux/version.h> -#include <linux/types.h> -#include <linux/list.h> -#include <linux/interrupt.h> -#include <linux/dma-mapping.h> #include <linux/pci.h> -#include <linux/kthread.h> #include <linux/netdevice.h> -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" static void oct_poll_req_completion(struct work_struct *work); @@ -54,8 +40,9 @@ int octeon_setup_response_list(struct octeon_device *oct) spin_lock_init(&oct->response_list[i].lock); atomic_set(&oct->response_list[i].pending_req_count, 0); } + spin_lock_init(&oct->cmd_resp_wqlock); - oct->dma_comp_wq.wq = create_workqueue("dma-comp"); + oct->dma_comp_wq.wq = alloc_workqueue("dma-comp", WQ_MEM_RECLAIM, 0); if (!oct->dma_comp_wq.wq) { dev_err(&oct->pci_dev->dev, "failed to create wq thread\n"); return -ENOMEM; @@ -64,7 +51,8 @@ int octeon_setup_response_list(struct octeon_device *oct) cwq = &oct->dma_comp_wq; INIT_DELAYED_WORK(&cwq->wk.work, oct_poll_req_completion); cwq->wk.ctxptr = oct; - queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(100)); + oct->cmd_resp_state = OCT_DRV_ONLINE; + queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(50)); return ret; } @@ -72,7 +60,6 @@ int octeon_setup_response_list(struct octeon_device *oct) void octeon_delete_response_list(struct octeon_device *oct) { cancel_delayed_work_sync(&oct->dma_comp_wq.wk.work); - flush_workqueue(oct->dma_comp_wq.wq); destroy_workqueue(oct->dma_comp_wq.wq); } @@ -86,6 +73,7 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev, u32 status; u64 status64; struct octeon_instr_rdp *rdp; + u64 rptr; ordered_sc_list = &octeon_dev->response_list[OCTEON_ORDERED_SC_LIST]; @@ -103,7 +91,8 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev, sc = (struct octeon_soft_command *)ordered_sc_list-> head.next; - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; + rptr = sc->cmd.cmd2.rptr; status = OCTEON_REQUEST_PENDING; @@ -111,7 +100,7 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev, * to where rptr is pointing to */ dma_sync_single_for_cpu(&octeon_dev->pci_dev->dev, - sc->cmd.rptr, rdp->rlen, + rptr, rdp->rlen, DMA_FROM_DEVICE); status64 = *sc->status_word; @@ -173,6 +162,5 @@ static void oct_poll_req_completion(struct work_struct *work) struct cavium_wq *cwq = &oct->dma_comp_wq; lio_process_ordered_list(oct, 0); - - queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(100)); + queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(50)); } diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index 388cd799d9ed..e8bc15bcde70 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -146,7 +146,6 @@ struct octeon_mgmt { struct device *dev; struct napi_struct napi; struct tasklet_struct tx_clean_tasklet; - struct phy_device *phydev; struct device_node *phy_np; resource_size_t mix_phys; resource_size_t mix_size; @@ -787,14 +786,12 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev, static int octeon_mgmt_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd) { - struct octeon_mgmt *p = netdev_priv(netdev); - switch (cmd) { case SIOCSHWTSTAMP: return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd); default: - if (p->phydev) - return phy_mii_ioctl(p->phydev, rq, cmd); + if (netdev->phydev) + return phy_mii_ioctl(netdev->phydev, rq, cmd); return -EINVAL; } } @@ -836,16 +833,18 @@ static void octeon_mgmt_enable_link(struct octeon_mgmt *p) static void octeon_mgmt_update_link(struct octeon_mgmt *p) { + struct net_device *ndev = p->netdev; + struct phy_device *phydev = ndev->phydev; union cvmx_agl_gmx_prtx_cfg prtx_cfg; prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG); - if (!p->phydev->link) + if (!phydev->link) prtx_cfg.s.duplex = 1; else - prtx_cfg.s.duplex = p->phydev->duplex; + prtx_cfg.s.duplex = phydev->duplex; - switch (p->phydev->speed) { + switch (phydev->speed) { case 10: prtx_cfg.s.speed = 0; prtx_cfg.s.slottime = 0; @@ -871,7 +870,7 @@ static void octeon_mgmt_update_link(struct octeon_mgmt *p) prtx_cfg.s.speed_msb = 0; /* Only matters for half-duplex */ prtx_cfg.s.slottime = 1; - prtx_cfg.s.burst = p->phydev->duplex; + prtx_cfg.s.burst = phydev->duplex; } break; case 0: /* No link */ @@ -894,9 +893,9 @@ static void octeon_mgmt_update_link(struct octeon_mgmt *p) /* MII (both speeds) and RGMII 1000 speed. */ agl_clk.s.clk_cnt = 1; if (prtx_ctl.s.mode == 0) { /* RGMII mode */ - if (p->phydev->speed == 10) + if (phydev->speed == 10) agl_clk.s.clk_cnt = 50; - else if (p->phydev->speed == 100) + else if (phydev->speed == 100) agl_clk.s.clk_cnt = 5; } cvmx_write_csr(p->agl + AGL_GMX_TX_CLK, agl_clk.u64); @@ -906,39 +905,40 @@ static void octeon_mgmt_update_link(struct octeon_mgmt *p) static void octeon_mgmt_adjust_link(struct net_device *netdev) { struct octeon_mgmt *p = netdev_priv(netdev); + struct phy_device *phydev = netdev->phydev; unsigned long flags; int link_changed = 0; - if (!p->phydev) + if (!phydev) return; spin_lock_irqsave(&p->lock, flags); - if (!p->phydev->link && p->last_link) + if (!phydev->link && p->last_link) link_changed = -1; - if (p->phydev->link - && (p->last_duplex != p->phydev->duplex - || p->last_link != p->phydev->link - || p->last_speed != p->phydev->speed)) { + if (phydev->link && + (p->last_duplex != phydev->duplex || + p->last_link != phydev->link || + p->last_speed != phydev->speed)) { octeon_mgmt_disable_link(p); link_changed = 1; octeon_mgmt_update_link(p); octeon_mgmt_enable_link(p); } - p->last_link = p->phydev->link; - p->last_speed = p->phydev->speed; - p->last_duplex = p->phydev->duplex; + p->last_link = phydev->link; + p->last_speed = phydev->speed; + p->last_duplex = phydev->duplex; spin_unlock_irqrestore(&p->lock, flags); if (link_changed != 0) { if (link_changed > 0) { pr_info("%s: Link is up - %d/%s\n", netdev->name, - p->phydev->speed, - DUPLEX_FULL == p->phydev->duplex ? + phydev->speed, + phydev->duplex == DUPLEX_FULL ? "Full" : "Half"); } else { pr_info("%s: Link is down\n", netdev->name); @@ -949,6 +949,7 @@ static void octeon_mgmt_adjust_link(struct net_device *netdev) static int octeon_mgmt_init_phy(struct net_device *netdev) { struct octeon_mgmt *p = netdev_priv(netdev); + struct phy_device *phydev = NULL; if (octeon_is_simulation() || p->phy_np == NULL) { /* No PHYs in the simulator. */ @@ -956,11 +957,11 @@ static int octeon_mgmt_init_phy(struct net_device *netdev) return 0; } - p->phydev = of_phy_connect(netdev, p->phy_np, - octeon_mgmt_adjust_link, 0, - PHY_INTERFACE_MODE_MII); + phydev = of_phy_connect(netdev, p->phy_np, + octeon_mgmt_adjust_link, 0, + PHY_INTERFACE_MODE_MII); - if (!p->phydev) + if (!phydev) return -ENODEV; return 0; @@ -1080,9 +1081,9 @@ static int octeon_mgmt_open(struct net_device *netdev) } /* Set the mode of the interface, RGMII/MII. */ - if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && p->phydev) { + if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && netdev->phydev) { union cvmx_agl_prtx_ctl agl_prtx_ctl; - int rgmii_mode = (p->phydev->supported & + int rgmii_mode = (netdev->phydev->supported & (SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full)) != 0; agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl); @@ -1205,7 +1206,7 @@ static int octeon_mgmt_open(struct net_device *netdev) /* Configure the port duplex, speed and enables */ octeon_mgmt_disable_link(p); - if (p->phydev) + if (netdev->phydev) octeon_mgmt_update_link(p); octeon_mgmt_enable_link(p); @@ -1214,9 +1215,9 @@ static int octeon_mgmt_open(struct net_device *netdev) /* PHY is not present in simulator. The carrier is enabled * while initializing the phy for simulator, leave it enabled. */ - if (p->phydev) { + if (netdev->phydev) { netif_carrier_off(netdev); - phy_start_aneg(p->phydev); + phy_start_aneg(netdev->phydev); } netif_wake_queue(netdev); @@ -1244,9 +1245,8 @@ static int octeon_mgmt_stop(struct net_device *netdev) napi_disable(&p->napi); netif_stop_queue(netdev); - if (p->phydev) - phy_disconnect(p->phydev); - p->phydev = NULL; + if (netdev->phydev) + phy_disconnect(netdev->phydev); netif_carrier_off(netdev); @@ -1346,50 +1346,23 @@ static void octeon_mgmt_get_drvinfo(struct net_device *netdev, strlcpy(info->bus_info, "N/A", sizeof(info->bus_info)); } -static int octeon_mgmt_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - struct octeon_mgmt *p = netdev_priv(netdev); - - if (p->phydev) - return phy_ethtool_gset(p->phydev, cmd); - - return -EOPNOTSUPP; -} - -static int octeon_mgmt_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - struct octeon_mgmt *p = netdev_priv(netdev); - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (p->phydev) - return phy_ethtool_sset(p->phydev, cmd); - - return -EOPNOTSUPP; -} - static int octeon_mgmt_nway_reset(struct net_device *dev) { - struct octeon_mgmt *p = netdev_priv(dev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (p->phydev) - return phy_start_aneg(p->phydev); + if (dev->phydev) + return phy_start_aneg(dev->phydev); return -EOPNOTSUPP; } static const struct ethtool_ops octeon_mgmt_ethtool_ops = { .get_drvinfo = octeon_mgmt_get_drvinfo, - .get_settings = octeon_mgmt_get_settings, - .set_settings = octeon_mgmt_set_settings, .nway_reset = octeon_mgmt_nway_reset, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops octeon_mgmt_ops = { diff --git a/drivers/net/ethernet/cavium/thunder/nic_main.c b/drivers/net/ethernet/cavium/thunder/nic_main.c index 95f17f8cadac..16ed20357c5c 100644 --- a/drivers/net/ethernet/cavium/thunder/nic_main.c +++ b/drivers/net/ethernet/cavium/thunder/nic_main.c @@ -499,6 +499,7 @@ static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, u32 rr_quantum; u8 sq_idx = sq->sq_num; u8 pqs_vnic; + int svf; if (sq->sqs_mode) pqs_vnic = nic->pqs_vf[vnic]; @@ -511,10 +512,19 @@ static void nic_tx_channel_cfg(struct nicpf *nic, u8 vnic, /* 24 bytes for FCS, IPG and preamble */ rr_quantum = ((NIC_HW_MAX_FRS + 24) / 4); - tl4 = (lmac * NIC_TL4_PER_LMAC) + (bgx * NIC_TL4_PER_BGX); + if (!sq->sqs_mode) { + tl4 = (lmac * NIC_TL4_PER_LMAC) + (bgx * NIC_TL4_PER_BGX); + } else { + for (svf = 0; svf < MAX_SQS_PER_VF; svf++) { + if (nic->vf_sqs[pqs_vnic][svf] == vnic) + break; + } + tl4 = (MAX_LMAC_PER_BGX * NIC_TL4_PER_LMAC); + tl4 += (lmac * NIC_TL4_PER_LMAC * MAX_SQS_PER_VF); + tl4 += (svf * NIC_TL4_PER_LMAC); + tl4 += (bgx * NIC_TL4_PER_BGX); + } tl4 += sq_idx; - if (sq->sqs_mode) - tl4 += vnic * 8; tl3 = tl4 / (NIC_MAX_TL4 / NIC_MAX_TL3); nic_reg_write(nic, NIC_PF_QSET_0_127_SQ_0_7_CFG2 | diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c index 3ed21988626b..63a39ac97d53 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c @@ -551,7 +551,9 @@ static int bgx_xaui_check_link(struct lmac *lmac) } /* Clear rcvflt bit (latching high) and read it back */ - bgx_reg_modify(bgx, lmacid, BGX_SPUX_STATUS2, SPU_STATUS2_RCVFLT); + if (bgx_reg_read(bgx, lmacid, BGX_SPUX_STATUS2) & SPU_STATUS2_RCVFLT) + bgx_reg_modify(bgx, lmacid, + BGX_SPUX_STATUS2, SPU_STATUS2_RCVFLT); if (bgx_reg_read(bgx, lmacid, BGX_SPUX_STATUS2) & SPU_STATUS2_RCVFLT) { dev_err(&bgx->pdev->dev, "Receive fault, retry training\n"); if (bgx->use_training) { @@ -570,13 +572,6 @@ static int bgx_xaui_check_link(struct lmac *lmac) return -1; } - /* Wait for MAC RX to be ready */ - if (bgx_poll_reg(bgx, lmacid, BGX_SMUX_RX_CTL, - SMU_RX_CTL_STATUS, true)) { - dev_err(&bgx->pdev->dev, "SMU RX link not okay\n"); - return -1; - } - /* Wait for BGX RX to be idle */ if (bgx_poll_reg(bgx, lmacid, BGX_SMUX_CTL, SMU_CTL_RX_IDLE, false)) { dev_err(&bgx->pdev->dev, "SMU RX not idle\n"); @@ -589,29 +584,30 @@ static int bgx_xaui_check_link(struct lmac *lmac) return -1; } - if (bgx_reg_read(bgx, lmacid, BGX_SPUX_STATUS2) & SPU_STATUS2_RCVFLT) { - dev_err(&bgx->pdev->dev, "Receive fault\n"); - return -1; - } - - /* Receive link is latching low. Force it high and verify it */ - bgx_reg_modify(bgx, lmacid, BGX_SPUX_STATUS1, SPU_STATUS1_RCV_LNK); - if (bgx_poll_reg(bgx, lmacid, BGX_SPUX_STATUS1, - SPU_STATUS1_RCV_LNK, false)) { - dev_err(&bgx->pdev->dev, "SPU receive link down\n"); - return -1; - } - + /* Clear receive packet disable */ cfg = bgx_reg_read(bgx, lmacid, BGX_SPUX_MISC_CONTROL); cfg &= ~SPU_MISC_CTL_RX_DIS; bgx_reg_write(bgx, lmacid, BGX_SPUX_MISC_CONTROL, cfg); - return 0; + + /* Check for MAC RX faults */ + cfg = bgx_reg_read(bgx, lmacid, BGX_SMUX_RX_CTL); + /* 0 - Link is okay, 1 - Local fault, 2 - Remote fault */ + cfg &= SMU_RX_CTL_STATUS; + if (!cfg) + return 0; + + /* Rx local/remote fault seen. + * Do lmac reinit to see if condition recovers + */ + bgx_lmac_xaui_init(bgx, lmacid, bgx->lmac_type); + + return -1; } static void bgx_poll_for_link(struct work_struct *work) { struct lmac *lmac; - u64 link; + u64 spu_link, smu_link; lmac = container_of(work, struct lmac, dwork.work); @@ -621,8 +617,11 @@ static void bgx_poll_for_link(struct work_struct *work) bgx_poll_reg(lmac->bgx, lmac->lmacid, BGX_SPUX_STATUS1, SPU_STATUS1_RCV_LNK, false); - link = bgx_reg_read(lmac->bgx, lmac->lmacid, BGX_SPUX_STATUS1); - if (link & SPU_STATUS1_RCV_LNK) { + spu_link = bgx_reg_read(lmac->bgx, lmac->lmacid, BGX_SPUX_STATUS1); + smu_link = bgx_reg_read(lmac->bgx, lmac->lmacid, BGX_SMUX_RX_CTL); + + if ((spu_link & SPU_STATUS1_RCV_LNK) && + !(smu_link & SMU_RX_CTL_STATUS)) { lmac->link_up = 1; if (lmac->bgx->lmac_type == BGX_MODE_XLAUI) lmac->last_speed = 40000; @@ -636,9 +635,15 @@ static void bgx_poll_for_link(struct work_struct *work) } if (lmac->last_link != lmac->link_up) { + if (lmac->link_up) { + if (bgx_xaui_check_link(lmac)) { + /* Errors, clear link_up state */ + lmac->link_up = 0; + lmac->last_speed = SPEED_UNKNOWN; + lmac->last_duplex = DUPLEX_UNKNOWN; + } + } lmac->last_link = lmac->link_up; - if (lmac->link_up) - bgx_xaui_check_link(lmac); } queue_delayed_work(lmac->check_link, &lmac->dwork, HZ * 2); @@ -710,7 +715,7 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid) static void bgx_lmac_disable(struct bgx *bgx, u8 lmacid) { struct lmac *lmac; - u64 cmrx_cfg; + u64 cfg; lmac = &bgx->lmac[lmacid]; if (lmac->check_link) { @@ -719,9 +724,33 @@ static void bgx_lmac_disable(struct bgx *bgx, u8 lmacid) destroy_workqueue(lmac->check_link); } - cmrx_cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); - cmrx_cfg &= ~(1 << 15); - bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cmrx_cfg); + /* Disable packet reception */ + cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); + cfg &= ~CMR_PKT_RX_EN; + bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg); + + /* Give chance for Rx/Tx FIFO to get drained */ + bgx_poll_reg(bgx, lmacid, BGX_CMRX_RX_FIFO_LEN, (u64)0x1FFF, true); + bgx_poll_reg(bgx, lmacid, BGX_CMRX_TX_FIFO_LEN, (u64)0x3FFF, true); + + /* Disable packet transmission */ + cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); + cfg &= ~CMR_PKT_TX_EN; + bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg); + + /* Disable serdes lanes */ + if (!lmac->is_sgmii) + bgx_reg_modify(bgx, lmacid, + BGX_SPUX_CONTROL1, SPU_CTL_LOW_POWER); + else + bgx_reg_modify(bgx, lmacid, + BGX_GMP_PCS_MRX_CTL, PCS_MRX_CTL_PWR_DN); + + /* Disable LMAC */ + cfg = bgx_reg_read(bgx, lmacid, BGX_CMRX_CFG); + cfg &= ~CMR_EN; + bgx_reg_write(bgx, lmacid, BGX_CMRX_CFG, cfg); + bgx_flush_dmac_addrs(bgx, lmacid); if ((bgx->lmac_type != BGX_MODE_XFI) && diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h index 149e179363a1..42010d2e5ddf 100644 --- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.h +++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.h @@ -41,6 +41,7 @@ #define BGX_CMRX_RX_STAT10 0xC0 #define BGX_CMRX_RX_BP_DROP 0xC8 #define BGX_CMRX_RX_DMAC_CTL 0x0E8 +#define BGX_CMRX_RX_FIFO_LEN 0x108 #define BGX_CMR_RX_DMACX_CAM 0x200 #define RX_DMACX_CAM_EN BIT_ULL(48) #define RX_DMACX_CAM_LMACID(x) (x << 49) @@ -50,6 +51,7 @@ #define BGX_CMR_CHAN_MSK_AND 0x450 #define BGX_CMR_BIST_STATUS 0x460 #define BGX_CMR_RX_LMACS 0x468 +#define BGX_CMRX_TX_FIFO_LEN 0x518 #define BGX_CMRX_TX_STAT0 0x600 #define BGX_CMRX_TX_STAT1 0x608 #define BGX_CMRX_TX_STAT2 0x610 diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 477db477b133..c45de49dc963 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -64,6 +64,7 @@ #include <net/bonding.h> #include <net/addrconf.h> #include <asm/uaccess.h> +#include <linux/crash_dump.h> #include "cxgb4.h" #include "t4_regs.h" @@ -206,7 +207,7 @@ static int rx_dma_offset = 2; static unsigned int num_vf[NUM_OF_PF_WITH_SRIOV]; module_param_array(num_vf, uint, NULL, 0644); -MODULE_PARM_DESC(num_vf, "number of VFs for each of PFs 0-3"); +MODULE_PARM_DESC(num_vf, "number of VFs for each of PFs 0-3, deprecated parameter - please use the pci sysfs interface."); #endif /* TX Queue select used to determine what algorithm to use for selecting TX @@ -460,11 +461,8 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok) struct port_info *pi = netdev_priv(dev); struct adapter *adapter = pi->adapter; - if (!(dev->flags & IFF_PROMISC)) { - __dev_uc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); - if (!(dev->flags & IFF_ALLMULTI)) - __dev_mc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); - } + __dev_uc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); + __dev_mc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); return t4_set_rxmode(adapter, adapter->mbox, pi->viid, mtu, (dev->flags & IFF_PROMISC) ? 1 : 0, @@ -3735,7 +3733,8 @@ static int adap_init0(struct adapter *adap) return ret; /* Contact FW, advertising Master capability */ - ret = t4_fw_hello(adap, adap->mbox, adap->mbox, MASTER_MAY, &state); + ret = t4_fw_hello(adap, adap->mbox, adap->mbox, + is_kdump_kernel() ? MASTER_MUST : MASTER_MAY, &state); if (ret < 0) { dev_err(adap->pdev_dev, "could not connect to FW, error %d\n", ret); @@ -4366,6 +4365,11 @@ static void cfg_queues(struct adapter *adap) if (q10g > netif_get_num_default_rss_queues()) q10g = netif_get_num_default_rss_queues(); + /* Reduce memory usage in kdump environment, disable all offload. + */ + if (is_kdump_kernel()) + adap->params.offload = 0; + for_each_port(adap, i) { struct port_info *pi = adap2pinfo(adap, i); @@ -4829,6 +4833,60 @@ static int get_chip_type(struct pci_dev *pdev, u32 pl_rev) return -EINVAL; } +#ifdef CONFIG_PCI_IOV +static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) +{ + int err = 0; + int current_vfs = pci_num_vf(pdev); + u32 pcie_fw; + void __iomem *regs; + + regs = pci_ioremap_bar(pdev, 0); + if (!regs) { + dev_err(&pdev->dev, "cannot map device registers\n"); + return -ENOMEM; + } + + pcie_fw = readl(regs + PCIE_FW_A); + iounmap(regs); + /* Check if cxgb4 is the MASTER and fw is initialized */ + if (!(pcie_fw & PCIE_FW_INIT_F) || + !(pcie_fw & PCIE_FW_MASTER_VLD_F) || + PCIE_FW_MASTER_G(pcie_fw) != 4) { + dev_warn(&pdev->dev, + "cxgb4 driver needs to be MASTER to support SRIOV\n"); + return -EOPNOTSUPP; + } + + /* If any of the VF's is already assigned to Guest OS, then + * SRIOV for the same cannot be modified + */ + if (current_vfs && pci_vfs_assigned(pdev)) { + dev_err(&pdev->dev, + "Cannot modify SR-IOV while VFs are assigned\n"); + num_vfs = current_vfs; + return num_vfs; + } + + /* Disable SRIOV when zero is passed. + * One needs to disable SRIOV before modifying it, else + * stack throws the below warning: + * " 'n' VFs already enabled. Disable before enabling 'm' VFs." + */ + if (!num_vfs) { + pci_disable_sriov(pdev); + return num_vfs; + } + + if (num_vfs != current_vfs) { + err = pci_enable_sriov(pdev, num_vfs); + if (err) + return err; + } + return num_vfs; +} +#endif + static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { int func, i, err, s_qpp, qpp, num_seg; @@ -5162,11 +5220,16 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) sriov: #ifdef CONFIG_PCI_IOV - if (func < ARRAY_SIZE(num_vf) && num_vf[func] > 0) + if (func < ARRAY_SIZE(num_vf) && num_vf[func] > 0) { + dev_warn(&pdev->dev, + "Enabling SR-IOV VFs using the num_vf module " + "parameter is deprecated - please use the pci sysfs " + "interface instead.\n"); if (pci_enable_sriov(pdev, num_vf[func]) == 0) dev_info(&pdev->dev, "instantiated %u virtual functions\n", num_vf[func]); + } #endif return 0; @@ -5259,6 +5322,9 @@ static struct pci_driver cxgb4_driver = { .probe = init_one, .remove = remove_one, .shutdown = remove_one, +#ifdef CONFIG_PCI_IOV + .sriov_configure = cxgb4_iov_configure, +#endif .err_handler = &cxgb4_eeh, }; diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h index a2cdfc1261dc..50812a1d67bd 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h @@ -144,6 +144,7 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN CH_PCI_ID_TABLE_FENTRY(0x5015), /* T502-bt */ CH_PCI_ID_TABLE_FENTRY(0x5016), /* T580-OCP-SO */ CH_PCI_ID_TABLE_FENTRY(0x5017), /* T520-OCP-SO */ + CH_PCI_ID_TABLE_FENTRY(0x5018), /* T540-BT */ CH_PCI_ID_TABLE_FENTRY(0x5080), /* Custom T540-cr */ CH_PCI_ID_TABLE_FENTRY(0x5081), /* Custom T540-LL-cr */ CH_PCI_ID_TABLE_FENTRY(0x5082), /* Custom T504-cr */ diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h index 734dd776c22f..109bc630408b 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h +++ b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h @@ -353,6 +353,10 @@ struct hash_mac_addr { u8 addr[ETH_ALEN]; }; +struct mbox_list { + struct list_head list; +}; + /* * Per-"adapter" (Virtual Function) information. */ @@ -387,6 +391,10 @@ struct adapter { /* various locks */ spinlock_t stats_lock; + /* lock for mailbox cmd list */ + spinlock_t mbox_lock; + struct mbox_list mlist; + /* support for mailbox command/reply logging */ #define T4VF_OS_LOG_MBOX_CMDS 256 struct mbox_cmd_log *mbox_log; diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 04fc6f6d1e25..9f5526478d2f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -937,12 +937,8 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok) { struct port_info *pi = netdev_priv(dev); - if (!(dev->flags & IFF_PROMISC)) { - __dev_uc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync); - if (!(dev->flags & IFF_ALLMULTI)) - __dev_mc_sync(dev, cxgb4vf_mac_sync, - cxgb4vf_mac_unsync); - } + __dev_uc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync); + __dev_mc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync); return t4vf_set_rxmode(pi->adapter, pi->viid, -1, (dev->flags & IFF_PROMISC) != 0, (dev->flags & IFF_ALLMULTI) != 0, @@ -2778,6 +2774,8 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev, * Initialize SMP data synchronization resources. */ spin_lock_init(&adapter->stats_lock); + spin_lock_init(&adapter->mbox_lock); + INIT_LIST_HEAD(&adapter->mlist.list); /* * Map our I/O registers in BAR0. diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c index 955ff7c61f1b..61bfe86da86d 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c @@ -139,6 +139,7 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, u32 mbox_ctl = T4VF_CIM_BASE_ADDR + CIM_VF_EXT_MAILBOX_CTRL; u32 cmd_op = FW_CMD_OP_G(be32_to_cpu(((struct fw_cmd_hdr *)cmd)->hi)); __be64 cmd_rpl[MBOX_LEN / 8]; + struct mbox_list entry; /* In T6, mailbox size is changed to 128 bytes to avoid * invalidating the entire prefetch buffer. @@ -156,6 +157,51 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, size > NUM_CIM_VF_MAILBOX_DATA_INSTANCES * 4) return -EINVAL; + /* Queue ourselves onto the mailbox access list. When our entry is at + * the front of the list, we have rights to access the mailbox. So we + * wait [for a while] till we're at the front [or bail out with an + * EBUSY] ... + */ + spin_lock(&adapter->mbox_lock); + list_add_tail(&entry.list, &adapter->mlist.list); + spin_unlock(&adapter->mbox_lock); + + delay_idx = 0; + ms = delay[0]; + + for (i = 0; ; i += ms) { + /* If we've waited too long, return a busy indication. This + * really ought to be based on our initial position in the + * mailbox access list but this is a start. We very rearely + * contend on access to the mailbox ... + */ + if (i > FW_CMD_MAX_TIMEOUT) { + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); + ret = -EBUSY; + t4vf_record_mbox(adapter, cmd, size, access, ret); + return ret; + } + + /* If we're at the head, break out and start the mailbox + * protocol. + */ + if (list_first_entry(&adapter->mlist.list, struct mbox_list, + list) == &entry) + break; + + /* Delay for a bit before checking again ... */ + if (sleep_ok) { + ms = delay[delay_idx]; /* last element may repeat */ + if (delay_idx < ARRAY_SIZE(delay) - 1) + delay_idx++; + msleep(ms); + } else { + mdelay(ms); + } + } + /* * Loop trying to get ownership of the mailbox. Return an error * if we can't gain ownership. @@ -164,6 +210,9 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, for (i = 0; v == MBOX_OWNER_NONE && i < 3; i++) v = MBOWNER_G(t4_read_reg(adapter, mbox_ctl)); if (v != MBOX_OWNER_DRV) { + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); ret = (v == MBOX_OWNER_FW) ? -EBUSY : -ETIMEDOUT; t4vf_record_mbox(adapter, cmd, size, access, ret); return ret; @@ -248,6 +297,9 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, if (cmd_op != FW_VI_STATS_CMD) t4vf_record_mbox(adapter, cmd_rpl, size, access, execute); + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); return -FW_CMD_RETVAL_G(v); } } @@ -255,6 +307,9 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, /* We timed out. Return the error ... */ ret = -ETIMEDOUT; t4vf_record_mbox(adapter, cmd, size, access, ret); + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); return ret; } diff --git a/drivers/net/ethernet/cirrus/cs89x0.c b/drivers/net/ethernet/cirrus/cs89x0.c index 60383040d6c6..c363b58552e9 100644 --- a/drivers/net/ethernet/cirrus/cs89x0.c +++ b/drivers/net/ethernet/cirrus/cs89x0.c @@ -53,6 +53,8 @@ #include <linux/errno.h> #include <linux/netdevice.h> #include <linux/etherdevice.h> +#include <linux/of.h> +#include <linux/of_device.h> #include <linux/platform_device.h> #include <linux/kernel.h> #include <linux/types.h> @@ -1895,9 +1897,17 @@ static int cs89x0_platform_remove(struct platform_device *pdev) return 0; } +static const struct __maybe_unused of_device_id cs89x0_match[] = { + { .compatible = "cirrus,cs8900", }, + { .compatible = "cirrus,cs8920", }, + { }, +}; +MODULE_DEVICE_TABLE(of, cs89x0_match); + static struct platform_driver cs89x0_driver = { .driver = { - .name = DRV_NAME, + .name = DRV_NAME, + .of_match_table = of_match_ptr(cs89x0_match), }, .remove = cs89x0_platform_remove, }; diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c index f44a39c40642..fd3980cc1e34 100644 --- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c +++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c @@ -103,25 +103,29 @@ static void enic_intr_coal_set_rx(struct enic *enic, u32 timer) } } -static int enic_get_settings(struct net_device *netdev, - struct ethtool_cmd *ecmd) +static int enic_get_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *ecmd) { struct enic *enic = netdev_priv(netdev); + struct ethtool_link_settings *base = &ecmd->base; - ecmd->supported = (SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE); - ecmd->advertising = (ADVERTISED_10000baseT_Full | ADVERTISED_FIBRE); - ecmd->port = PORT_FIBRE; - ecmd->transceiver = XCVR_EXTERNAL; + ethtool_link_ksettings_add_link_mode(ecmd, supported, + 10000baseT_Full); + ethtool_link_ksettings_add_link_mode(ecmd, supported, FIBRE); + ethtool_link_ksettings_add_link_mode(ecmd, advertising, + 10000baseT_Full); + ethtool_link_ksettings_add_link_mode(ecmd, advertising, FIBRE); + base->port = PORT_FIBRE; if (netif_carrier_ok(netdev)) { - ethtool_cmd_speed_set(ecmd, vnic_dev_port_speed(enic->vdev)); - ecmd->duplex = DUPLEX_FULL; + base->speed = vnic_dev_port_speed(enic->vdev); + base->duplex = DUPLEX_FULL; } else { - ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN); - ecmd->duplex = DUPLEX_UNKNOWN; + base->speed = SPEED_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; } - ecmd->autoneg = AUTONEG_DISABLE; + base->autoneg = AUTONEG_DISABLE; return 0; } @@ -500,7 +504,6 @@ static int enic_set_rxfh(struct net_device *netdev, const u32 *indir, } static const struct ethtool_ops enic_ethtool_ops = { - .get_settings = enic_get_settings, .get_drvinfo = enic_get_drvinfo, .get_msglevel = enic_get_msglevel, .set_msglevel = enic_set_msglevel, @@ -516,6 +519,7 @@ static const struct ethtool_ops enic_ethtool_ops = { .get_rxfh_key_size = enic_get_rxfh_key_size, .get_rxfh = enic_get_rxfh, .set_rxfh = enic_set_rxfh, + .get_link_ksettings = enic_get_ksettings, }; void enic_set_ethtool_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index b69a9eacc531..c3b64cdd0dec 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -173,7 +173,7 @@ static int dnet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, static void dnet_handle_link_change(struct net_device *dev) { struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; u32 mode_reg, ctl_reg; @@ -295,7 +295,6 @@ static int dnet_mii_probe(struct net_device *dev) bp->link = 0; bp->speed = 0; bp->duplex = -1; - bp->phy_dev = phydev; return 0; } @@ -629,16 +628,16 @@ static int dnet_open(struct net_device *dev) struct dnet *bp = netdev_priv(dev); /* if the phy is not yet register, retry later */ - if (!bp->phy_dev) + if (!dev->phydev) return -EAGAIN; napi_enable(&bp->napi); dnet_init_hw(bp); - phy_start_aneg(bp->phy_dev); + phy_start_aneg(dev->phydev); /* schedule a link state check */ - phy_start(bp->phy_dev); + phy_start(dev->phydev); netif_start_queue(dev); @@ -652,8 +651,8 @@ static int dnet_close(struct net_device *dev) netif_stop_queue(dev); napi_disable(&bp->napi); - if (bp->phy_dev) - phy_stop(bp->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); dnet_reset_hw(bp); netif_carrier_off(dev); @@ -731,32 +730,9 @@ static struct net_device_stats *dnet_get_stats(struct net_device *dev) return nstat; } -static int dnet_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int dnet_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static int dnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -776,11 +752,11 @@ static void dnet_get_drvinfo(struct net_device *dev, } static const struct ethtool_ops dnet_ethtool_ops = { - .get_settings = dnet_get_settings, - .set_settings = dnet_set_settings, .get_drvinfo = dnet_get_drvinfo, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops dnet_netdev_ops = { @@ -875,7 +851,7 @@ static int dnet_probe(struct platform_device *pdev) (bp->capabilities & DNET_HAS_IRQ) ? "" : "no ", (bp->capabilities & DNET_HAS_GIGABIT) ? "" : "no ", (bp->capabilities & DNET_HAS_DMA) ? "" : "no "); - phydev = bp->phy_dev; + phydev = dev->phydev; phy_attached_info(phydev); return 0; @@ -899,8 +875,8 @@ static int dnet_remove(struct platform_device *pdev) if (dev) { bp = netdev_priv(dev); - if (bp->phy_dev) - phy_disconnect(bp->phy_dev); + if (dev->phydev) + phy_disconnect(dev->phydev); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); unregister_netdev(dev); diff --git a/drivers/net/ethernet/dnet.h b/drivers/net/ethernet/dnet.h index 37f5b30fa78b..d985080bbd5d 100644 --- a/drivers/net/ethernet/dnet.h +++ b/drivers/net/ethernet/dnet.h @@ -216,7 +216,6 @@ struct dnet { /* PHY stuff */ struct mii_bus *mii_bus; - struct phy_device *phy_dev; unsigned int link; unsigned int speed; unsigned int duplex; diff --git a/drivers/net/ethernet/emulex/benet/Kconfig b/drivers/net/ethernet/emulex/benet/Kconfig index 7108563260ae..b4853ec9de8d 100644 --- a/drivers/net/ethernet/emulex/benet/Kconfig +++ b/drivers/net/ethernet/emulex/benet/Kconfig @@ -13,11 +13,3 @@ config BE2NET_HWMON ---help--- Say Y here if you want to expose thermal sensor data on be2net network adapter. - -config BE2NET_VXLAN - bool "VXLAN offload support on be2net driver" - default y - depends on BE2NET && VXLAN && !(BE2NET=y && VXLAN=m) - ---help--- - Say Y here if you want to enable VXLAN offload support on - be2net driver. diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index fe3763df3f13..4555e041ef69 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -97,7 +97,8 @@ * SURF/DPDK */ -#define MAX_RSS_IFACES 15 +#define MAX_PORT_RSS_TABLES 15 +#define MAX_NIC_FUNCS 16 #define MAX_RX_QS 32 #define MAX_EVT_QS 32 #define MAX_TX_QS 32 @@ -442,8 +443,20 @@ struct be_resources { u16 max_iface_count; u16 max_mcc_count; u16 max_evt_qs; + u16 max_nic_evt_qs; /* NIC's share of evt qs */ u32 if_cap_flags; u32 vf_if_cap_flags; /* VF if capability flags */ + u32 flags; + /* Calculated PF Pool's share of RSS Tables. This is not enforced by + * the FW, but is a self-imposed driver limitation. + */ + u16 max_rss_tables; +}; + +/* These are port-wide values */ +struct be_port_resources { + u16 max_vfs; + u16 nic_pfs; }; #define be_is_os2bmc_enabled(adapter) (adapter->flags & BE_FLAGS_OS2BMC) @@ -513,7 +526,8 @@ struct be_adapter { spinlock_t mcc_lock; /* For serializing mcc cmds to BE card */ spinlock_t mcc_cq_lock; - u16 cfg_num_qs; /* configured via set-channels */ + u16 cfg_num_rx_irqs; /* configured via set-channels */ + u16 cfg_num_tx_irqs; /* configured via set-channels */ u16 num_evt_qs; u16 num_msix_vec; struct be_eq_obj eq_obj[MAX_EVT_QS]; @@ -632,16 +646,42 @@ struct be_adapter { #define be_max_txqs(adapter) (adapter->res.max_tx_qs) #define be_max_prio_txqs(adapter) (adapter->res.max_prio_tx_qs) #define be_max_rxqs(adapter) (adapter->res.max_rx_qs) -#define be_max_eqs(adapter) (adapter->res.max_evt_qs) +/* Max number of EQs available for the function (NIC + RoCE (if enabled)) */ +#define be_max_func_eqs(adapter) (adapter->res.max_evt_qs) +/* Max number of EQs available avaialble only for NIC */ +#define be_max_nic_eqs(adapter) (adapter->res.max_nic_evt_qs) #define be_if_cap_flags(adapter) (adapter->res.if_cap_flags) - -static inline u16 be_max_qs(struct be_adapter *adapter) +#define be_max_pf_pool_rss_tables(adapter) \ + (adapter->pool_res.max_rss_tables) +/* Max irqs avaialble for NIC */ +#define be_max_irqs(adapter) \ + (min_t(u16, be_max_nic_eqs(adapter), num_online_cpus())) + +/* Max irqs *needed* for RX queues */ +static inline u16 be_max_rx_irqs(struct be_adapter *adapter) { - /* If no RSS, need atleast the one def RXQ */ + /* If no RSS, need atleast one irq for def-RXQ */ u16 num = max_t(u16, be_max_rss(adapter), 1); - num = min(num, be_max_eqs(adapter)); - return min_t(u16, num, num_online_cpus()); + return min_t(u16, num, be_max_irqs(adapter)); +} + +/* Max irqs *needed* for TX queues */ +static inline u16 be_max_tx_irqs(struct be_adapter *adapter) +{ + return min_t(u16, be_max_txqs(adapter), be_max_irqs(adapter)); +} + +/* Max irqs *needed* for combined queues */ +static inline u16 be_max_qp_irqs(struct be_adapter *adapter) +{ + return min(be_max_tx_irqs(adapter), be_max_rx_irqs(adapter)); +} + +/* Max irqs *needed* for RX and TX queues together */ +static inline u16 be_max_any_irqs(struct be_adapter *adapter) +{ + return max(be_max_tx_irqs(adapter), be_max_rx_irqs(adapter)); } /* Is BE in pvid_tagging mode */ diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index 22402db275f2..2cc11756859f 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -87,6 +87,11 @@ static struct be_cmd_priv_map cmd_priv_map[] = { CMD_SUBSYSTEM_LOWLEVEL, BE_PRIV_DEVCFG | BE_PRIV_DEVSEC }, + { + OPCODE_COMMON_SET_HSW_CONFIG, + CMD_SUBSYSTEM_COMMON, + BE_PRIV_DEVCFG | BE_PRIV_VHADM + }, }; static bool be_cmd_allowed(struct be_adapter *adapter, u8 opcode, u8 subsystem) @@ -3850,6 +3855,10 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, void *ctxt; int status; + if (!be_cmd_allowed(adapter, OPCODE_COMMON_SET_HSW_CONFIG, + CMD_SUBSYSTEM_COMMON)) + return -EPERM; + spin_lock_bh(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); @@ -3871,7 +3880,7 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, AMAP_SET_BITS(struct amap_set_hsw_context, pvid_valid, ctxt, 1); AMAP_SET_BITS(struct amap_set_hsw_context, pvid, ctxt, pvid); } - if (!BEx_chip(adapter) && hsw_mode) { + if (hsw_mode) { AMAP_SET_BITS(struct amap_set_hsw_context, interface_id, ctxt, adapter->hba_port_num); AMAP_SET_BITS(struct amap_set_hsw_context, pport, ctxt, 1); @@ -4023,7 +4032,10 @@ int be_cmd_get_acpi_wol_cap(struct be_adapter *adapter) resp = (struct be_cmd_resp_acpi_wol_magic_config_v1 *)cmd.va; adapter->wol_cap = resp->wol_settings; - if (adapter->wol_cap & BE_WOL_CAP) + + /* Non-zero macaddr indicates WOL is enabled */ + if (adapter->wol_cap & BE_WOL_CAP && + !is_zero_ether_addr(resp->magic_mac)) adapter->wol_en = true; } err: @@ -4360,9 +4372,35 @@ err: return status; } +/* This routine returns a list of all the NIC PF_nums in the adapter */ +u16 be_get_nic_pf_num_list(u8 *buf, u32 desc_count, u16 *nic_pf_nums) +{ + struct be_res_desc_hdr *hdr = (struct be_res_desc_hdr *)buf; + struct be_pcie_res_desc *pcie = NULL; + int i; + u16 nic_pf_count = 0; + + for (i = 0; i < desc_count; i++) { + if (hdr->desc_type == PCIE_RESOURCE_DESC_TYPE_V0 || + hdr->desc_type == PCIE_RESOURCE_DESC_TYPE_V1) { + pcie = (struct be_pcie_res_desc *)hdr; + if (pcie->pf_state && (pcie->pf_type == MISSION_NIC || + pcie->pf_type == MISSION_RDMA)) { + nic_pf_nums[nic_pf_count++] = pcie->pf_num; + } + } + + hdr->desc_len = hdr->desc_len ? : RESOURCE_DESC_SIZE_V0; + hdr = (void *)hdr + hdr->desc_len; + } + return nic_pf_count; +} + /* Will use MBOX only if MCCQ has not been created */ int be_cmd_get_profile_config(struct be_adapter *adapter, - struct be_resources *res, u8 query, u8 domain) + struct be_resources *res, + struct be_port_resources *port_res, + u8 profile_type, u8 query, u8 domain) { struct be_cmd_resp_get_profile_config *resp; struct be_cmd_req_get_profile_config *req; @@ -4389,7 +4427,7 @@ int be_cmd_get_profile_config(struct be_adapter *adapter, if (!lancer_chip(adapter)) req->hdr.version = 1; - req->type = ACTIVE_PROFILE_TYPE; + req->type = profile_type; req->hdr.domain = domain; /* When QUERY_MODIFIABLE_FIELDS_TYPE bit is set, cmd returns the @@ -4406,6 +4444,28 @@ int be_cmd_get_profile_config(struct be_adapter *adapter, resp = cmd.va; desc_count = le16_to_cpu(resp->desc_count); + if (port_res) { + u16 nic_pf_cnt = 0, i; + u16 nic_pf_num_list[MAX_NIC_FUNCS]; + + nic_pf_cnt = be_get_nic_pf_num_list(resp->func_param, + desc_count, + nic_pf_num_list); + + for (i = 0; i < nic_pf_cnt; i++) { + nic = be_get_func_nic_desc(resp->func_param, desc_count, + nic_pf_num_list[i]); + if (nic->link_param == adapter->port_num) { + port_res->nic_pfs++; + pcie = be_get_pcie_desc(resp->func_param, + desc_count, + nic_pf_num_list[i]); + port_res->max_vfs += le16_to_cpu(pcie->num_vfs); + } + } + return status; + } + pcie = be_get_pcie_desc(resp->func_param, desc_count, adapter->pf_num); if (pcie) @@ -4465,7 +4525,7 @@ static int be_cmd_set_profile_config(struct be_adapter *adapter, void *desc, } /* Mark all fields invalid */ -static void be_reset_nic_desc(struct be_nic_res_desc *nic) +void be_reset_nic_desc(struct be_nic_res_desc *nic) { memset(nic, 0, sizeof(*nic)); nic->unicast_mac_count = 0xFFFF; @@ -4534,73 +4594,9 @@ int be_cmd_config_qos(struct be_adapter *adapter, u32 max_rate, u16 link_speed, 1, version, domain); } -static void be_fill_vf_res_template(struct be_adapter *adapter, - struct be_resources pool_res, - u16 num_vfs, u16 num_vf_qs, - struct be_nic_res_desc *nic_vft) -{ - u32 vf_if_cap_flags = pool_res.vf_if_cap_flags; - struct be_resources res_mod = {0}; - - /* Resource with fields set to all '1's by GET_PROFILE_CONFIG cmd, - * which are modifiable using SET_PROFILE_CONFIG cmd. - */ - be_cmd_get_profile_config(adapter, &res_mod, RESOURCE_MODIFIABLE, 0); - - /* If RSS IFACE capability flags are modifiable for a VF, set the - * capability flag as valid and set RSS and DEFQ_RSS IFACE flags if - * more than 1 RSSQ is available for a VF. - * Otherwise, provision only 1 queue pair for VF. - */ - if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_RSS) { - nic_vft->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); - if (num_vf_qs > 1) { - vf_if_cap_flags |= BE_IF_FLAGS_RSS; - if (pool_res.if_cap_flags & BE_IF_FLAGS_DEFQ_RSS) - vf_if_cap_flags |= BE_IF_FLAGS_DEFQ_RSS; - } else { - vf_if_cap_flags &= ~(BE_IF_FLAGS_RSS | - BE_IF_FLAGS_DEFQ_RSS); - } - } else { - num_vf_qs = 1; - } - - if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { - nic_vft->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); - vf_if_cap_flags &= ~BE_IF_FLAGS_VLAN_PROMISCUOUS; - } - - nic_vft->cap_flags = cpu_to_le32(vf_if_cap_flags); - nic_vft->rq_count = cpu_to_le16(num_vf_qs); - nic_vft->txq_count = cpu_to_le16(num_vf_qs); - nic_vft->rssq_count = cpu_to_le16(num_vf_qs); - nic_vft->cq_count = cpu_to_le16(pool_res.max_cq_count / - (num_vfs + 1)); - - /* Distribute unicast MACs, VLANs, IFACE count and MCCQ count equally - * among the PF and it's VFs, if the fields are changeable - */ - if (res_mod.max_uc_mac == FIELD_MODIFIABLE) - nic_vft->unicast_mac_count = cpu_to_le16(pool_res.max_uc_mac / - (num_vfs + 1)); - - if (res_mod.max_vlans == FIELD_MODIFIABLE) - nic_vft->vlan_count = cpu_to_le16(pool_res.max_vlans / - (num_vfs + 1)); - - if (res_mod.max_iface_count == FIELD_MODIFIABLE) - nic_vft->iface_count = cpu_to_le16(pool_res.max_iface_count / - (num_vfs + 1)); - - if (res_mod.max_mcc_count == FIELD_MODIFIABLE) - nic_vft->mcc_count = cpu_to_le16(pool_res.max_mcc_count / - (num_vfs + 1)); -} - int be_cmd_set_sriov_config(struct be_adapter *adapter, struct be_resources pool_res, u16 num_vfs, - u16 num_vf_qs) + struct be_resources *vft_res) { struct { struct be_pcie_res_desc pcie; @@ -4620,12 +4616,26 @@ int be_cmd_set_sriov_config(struct be_adapter *adapter, be_reset_nic_desc(&desc.nic_vft); desc.nic_vft.hdr.desc_type = NIC_RESOURCE_DESC_TYPE_V1; desc.nic_vft.hdr.desc_len = RESOURCE_DESC_SIZE_V1; - desc.nic_vft.flags = BIT(VFT_SHIFT) | BIT(IMM_SHIFT) | BIT(NOSV_SHIFT); + desc.nic_vft.flags = vft_res->flags | BIT(VFT_SHIFT) | + BIT(IMM_SHIFT) | BIT(NOSV_SHIFT); desc.nic_vft.pf_num = adapter->pdev->devfn; desc.nic_vft.vf_num = 0; - - be_fill_vf_res_template(adapter, pool_res, num_vfs, num_vf_qs, - &desc.nic_vft); + desc.nic_vft.cap_flags = cpu_to_le32(vft_res->vf_if_cap_flags); + desc.nic_vft.rq_count = cpu_to_le16(vft_res->max_rx_qs); + desc.nic_vft.txq_count = cpu_to_le16(vft_res->max_tx_qs); + desc.nic_vft.rssq_count = cpu_to_le16(vft_res->max_rss_qs); + desc.nic_vft.cq_count = cpu_to_le16(vft_res->max_cq_count); + + if (vft_res->max_uc_mac) + desc.nic_vft.unicast_mac_count = + cpu_to_le16(vft_res->max_uc_mac); + if (vft_res->max_vlans) + desc.nic_vft.vlan_count = cpu_to_le16(vft_res->max_vlans); + if (vft_res->max_iface_count) + desc.nic_vft.iface_count = + cpu_to_le16(vft_res->max_iface_count); + if (vft_res->max_mcc_count) + desc.nic_vft.mcc_count = cpu_to_le16(vft_res->max_mcc_count); return be_cmd_set_profile_config(adapter, &desc, 2 * RESOURCE_DESC_SIZE_V1, 2, 1, 0); diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index d8540ae95e5a..0d6be224a787 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -1556,7 +1556,9 @@ struct be_cmd_resp_acpi_wol_magic_config_v1 { u8 rsvd0[2]; u8 wol_settings; u8 rsvd1[5]; - u32 rsvd2[295]; + u32 rsvd2[288]; + u8 magic_mac[6]; + u8 rsvd3[22]; } __packed; #define BE_GET_WOL_CAP 2 @@ -2128,6 +2130,9 @@ struct be_cmd_req_set_ext_fat_caps { #define IMM_SHIFT 6 /* Immediate */ #define NOSV_SHIFT 7 /* No save */ +#define MISSION_NIC 1 +#define MISSION_RDMA 8 + struct be_res_desc_hdr { u8 desc_type; u8 desc_len; @@ -2244,6 +2249,7 @@ struct be_cmd_req_get_profile_config { struct be_cmd_req_hdr hdr; u8 rsvd; #define ACTIVE_PROFILE_TYPE 0x2 +#define SAVED_PROFILE_TYPE 0x0 #define QUERY_MODIFIABLE_FIELDS_TYPE BIT(3) u8 type; u16 rsvd1; @@ -2449,7 +2455,9 @@ int be_cmd_query_port_name(struct be_adapter *adapter); int be_cmd_get_func_config(struct be_adapter *adapter, struct be_resources *res); int be_cmd_get_profile_config(struct be_adapter *adapter, - struct be_resources *res, u8 query, u8 domain); + struct be_resources *res, + struct be_port_resources *port_res, + u8 profile_type, u8 query, u8 domain); int be_cmd_get_active_profile(struct be_adapter *adapter, u16 *profile); int be_cmd_get_if_id(struct be_adapter *adapter, struct be_vf_cfg *vf_cfg, int vf_num); @@ -2461,4 +2469,4 @@ int be_cmd_set_vxlan_port(struct be_adapter *adapter, __be16 port); int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op); int be_cmd_set_sriov_config(struct be_adapter *adapter, struct be_resources res, u16 num_vfs, - u16 num_vf_qs); + struct be_resources *vft_res); diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index 2ff691636dac..50e7be5da50c 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -793,6 +793,11 @@ static void be_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) static int be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) { struct be_adapter *adapter = netdev_priv(netdev); + struct device *dev = &adapter->pdev->dev; + struct be_dma_mem cmd; + u8 mac[ETH_ALEN]; + bool enable; + int status; if (wol->wolopts & ~WAKE_MAGIC) return -EOPNOTSUPP; @@ -802,12 +807,32 @@ static int be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) return -EOPNOTSUPP; } - if (wol->wolopts & WAKE_MAGIC) - adapter->wol_en = true; - else - adapter->wol_en = false; + cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config); + cmd.va = dma_zalloc_coherent(dev, cmd.size, &cmd.dma, GFP_KERNEL); + if (!cmd.va) + return -ENOMEM; - return 0; + eth_zero_addr(mac); + + enable = wol->wolopts & WAKE_MAGIC; + if (enable) + ether_addr_copy(mac, adapter->netdev->dev_addr); + + status = be_cmd_enable_magic_wol(adapter, mac, &cmd); + if (status) { + dev_err(dev, "Could not set Wake-on-lan mac address\n"); + status = be_cmd_status(status); + goto err; + } + + pci_enable_wake(adapter->pdev, PCI_D3hot, enable); + pci_enable_wake(adapter->pdev, PCI_D3cold, enable); + + adapter->wol_en = enable ? true : false; + +err: + dma_free_coherent(dev, cmd.size, cmd.va, cmd.dma); + return status; } static int be_test_ddr_dma(struct be_adapter *adapter) @@ -1171,9 +1196,17 @@ static void be_get_channels(struct net_device *netdev, struct ethtool_channels *ch) { struct be_adapter *adapter = netdev_priv(netdev); + u16 num_rx_irqs = max_t(u16, adapter->num_rss_qs, 1); - ch->combined_count = adapter->num_evt_qs; - ch->max_combined = be_max_qs(adapter); + /* num_tx_qs is always same as the number of irqs used for TX */ + ch->combined_count = min(adapter->num_tx_qs, num_rx_irqs); + ch->rx_count = num_rx_irqs - ch->combined_count; + ch->tx_count = adapter->num_tx_qs - ch->combined_count; + + ch->max_combined = be_max_qp_irqs(adapter); + /* The user must create atleast one combined channel */ + ch->max_rx = be_max_rx_irqs(adapter) - 1; + ch->max_tx = be_max_tx_irqs(adapter) - 1; } static int be_set_channels(struct net_device *netdev, @@ -1182,11 +1215,22 @@ static int be_set_channels(struct net_device *netdev, struct be_adapter *adapter = netdev_priv(netdev); int status; - if (ch->rx_count || ch->tx_count || ch->other_count || - !ch->combined_count || ch->combined_count > be_max_qs(adapter)) + /* we support either only combined channels or a combination of + * combined and either RX-only or TX-only channels. + */ + if (ch->other_count || !ch->combined_count || + (ch->rx_count && ch->tx_count)) + return -EINVAL; + + if (ch->combined_count > be_max_qp_irqs(adapter) || + (ch->rx_count && + (ch->rx_count + ch->combined_count) > be_max_rx_irqs(adapter)) || + (ch->tx_count && + (ch->tx_count + ch->combined_count) > be_max_tx_irqs(adapter))) return -EINVAL; - adapter->cfg_num_qs = ch->combined_count; + adapter->cfg_num_rx_irqs = ch->combined_count + ch->rx_count; + adapter->cfg_num_tx_irqs = ch->combined_count + ch->tx_count; status = be_update_queues(adapter); return be_cmd_status(status); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index ed98ef1ecac3..1f16e73f6d0c 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -2620,8 +2620,10 @@ static int be_evt_queues_create(struct be_adapter *adapter) struct be_aic_obj *aic; int i, rc; + /* need enough EQs to service both RX and TX queues */ adapter->num_evt_qs = min_t(u16, num_irqs(adapter), - adapter->cfg_num_qs); + max(adapter->cfg_num_rx_irqs, + adapter->cfg_num_tx_irqs)); for_all_evt_queues(adapter, eqo, i) { int numa_node = dev_to_node(&adapter->pdev->dev); @@ -2726,7 +2728,7 @@ static int be_tx_qs_create(struct be_adapter *adapter) struct be_eq_obj *eqo; int status, i; - adapter->num_tx_qs = min(adapter->num_evt_qs, be_max_txqs(adapter)); + adapter->num_tx_qs = min(adapter->num_evt_qs, adapter->cfg_num_tx_irqs); for_all_tx_queues(adapter, txo, i) { cq = &txo->cq; @@ -2784,11 +2786,11 @@ static int be_rx_cqs_create(struct be_adapter *adapter) struct be_rx_obj *rxo; int rc, i; - /* We can create as many RSS rings as there are EQs. */ - adapter->num_rss_qs = adapter->num_evt_qs; + adapter->num_rss_qs = + min(adapter->num_evt_qs, adapter->cfg_num_rx_irqs); /* We'll use RSS only if atleast 2 RSS rings are supported. */ - if (adapter->num_rss_qs <= 1) + if (adapter->num_rss_qs < 2) adapter->num_rss_qs = 0; adapter->num_rx_qs = adapter->num_rss_qs + adapter->need_def_rxq; @@ -3249,18 +3251,23 @@ static void be_msix_disable(struct be_adapter *adapter) static int be_msix_enable(struct be_adapter *adapter) { - int i, num_vec; + unsigned int i, max_roce_eqs; struct device *dev = &adapter->pdev->dev; + int num_vec; - /* If RoCE is supported, program the max number of NIC vectors that - * may be configured via set-channels, along with vectors needed for - * RoCe. Else, just program the number we'll use initially. + /* If RoCE is supported, program the max number of vectors that + * could be used for NIC and RoCE, else, just program the number + * we'll use initially. */ - if (be_roce_supported(adapter)) - num_vec = min_t(int, 2 * be_max_eqs(adapter), - 2 * num_online_cpus()); - else - num_vec = adapter->cfg_num_qs; + if (be_roce_supported(adapter)) { + max_roce_eqs = + be_max_func_eqs(adapter) - be_max_nic_eqs(adapter); + max_roce_eqs = min(max_roce_eqs, num_online_cpus()); + num_vec = be_max_any_irqs(adapter) + max_roce_eqs; + } else { + num_vec = max(adapter->cfg_num_rx_irqs, + adapter->cfg_num_tx_irqs); + } for (i = 0; i < num_vec; i++) adapter->msix_entries[i].entry = i; @@ -3625,10 +3632,8 @@ static int be_open(struct net_device *netdev) be_link_status_update(adapter, link_status); netif_tx_start_all_queues(netdev); -#ifdef CONFIG_BE2NET_VXLAN if (skyhawk_chip(adapter)) - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); return 0; err: @@ -3636,40 +3641,6 @@ err: return -EIO; } -static int be_setup_wol(struct be_adapter *adapter, bool enable) -{ - struct device *dev = &adapter->pdev->dev; - struct be_dma_mem cmd; - u8 mac[ETH_ALEN]; - int status; - - eth_zero_addr(mac); - - cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config); - cmd.va = dma_zalloc_coherent(dev, cmd.size, &cmd.dma, GFP_KERNEL); - if (!cmd.va) - return -ENOMEM; - - if (enable) { - status = pci_write_config_dword(adapter->pdev, - PCICFG_PM_CONTROL_OFFSET, - PCICFG_PM_CONTROL_MASK); - if (status) { - dev_err(dev, "Could not enable Wake-on-lan\n"); - goto err; - } - } else { - ether_addr_copy(mac, adapter->netdev->dev_addr); - } - - status = be_cmd_enable_magic_wol(adapter, mac, &cmd); - pci_enable_wake(adapter->pdev, PCI_D3hot, enable); - pci_enable_wake(adapter->pdev, PCI_D3cold, enable); -err: - dma_free_coherent(dev, cmd.size, cmd.va, cmd.dma); - return status; -} - static void be_vf_eth_addr_generate(struct be_adapter *adapter, u8 *mac) { u32 addr; @@ -3759,6 +3730,11 @@ static void be_vf_clear(struct be_adapter *adapter) be_cmd_if_destroy(adapter, vf_cfg->if_handle, vf + 1); } + + if (BE3_chip(adapter)) + be_cmd_set_hsw_config(adapter, 0, 0, + adapter->if_handle, + PORT_FWD_TYPE_PASSTHRU, 0); done: kfree(adapter->vf_cfg); adapter->num_vfs = 0; @@ -3789,7 +3765,6 @@ static void be_cancel_err_detection(struct be_adapter *adapter) } } -#ifdef CONFIG_BE2NET_VXLAN static void be_disable_vxlan_offloads(struct be_adapter *adapter) { struct net_device *netdev = adapter->netdev; @@ -3808,37 +3783,87 @@ static void be_disable_vxlan_offloads(struct be_adapter *adapter) netdev->hw_features &= ~(NETIF_F_GSO_UDP_TUNNEL); netdev->features &= ~(NETIF_F_GSO_UDP_TUNNEL); } -#endif -static u16 be_calculate_vf_qs(struct be_adapter *adapter, u16 num_vfs) +static void be_calculate_vf_res(struct be_adapter *adapter, u16 num_vfs, + struct be_resources *vft_res) { struct be_resources res = adapter->pool_res; + u32 vf_if_cap_flags = res.vf_if_cap_flags; + struct be_resources res_mod = {0}; u16 num_vf_qs = 1; - /* Distribute the queue resources among the PF and it's VFs - * Do not distribute queue resources in multi-channel configuration. - */ - if (num_vfs && !be_is_mc(adapter)) { - /* Divide the qpairs evenly among the VFs and the PF, capped - * at VF-EQ-count. Any remainder qpairs belong to the PF. - */ + /* Distribute the queue resources among the PF and it's VFs */ + if (num_vfs) { + /* Divide the rx queues evenly among the VFs and the PF, capped + * at VF-EQ-count. Any remainder queues belong to the PF. + */ num_vf_qs = min(SH_VF_MAX_NIC_EQS, res.max_rss_qs / (num_vfs + 1)); - /* Skyhawk-R chip supports only MAX_RSS_IFACES RSS capable - * interfaces per port. Provide RSS on VFs, only if number - * of VFs requested is less than MAX_RSS_IFACES limit. + /* Skyhawk-R chip supports only MAX_PORT_RSS_TABLES + * RSS Tables per port. Provide RSS on VFs, only if number of + * VFs requested is less than it's PF Pool's RSS Tables limit. */ - if (num_vfs >= MAX_RSS_IFACES) + if (num_vfs >= be_max_pf_pool_rss_tables(adapter)) num_vf_qs = 1; } - return num_vf_qs; + + /* Resource with fields set to all '1's by GET_PROFILE_CONFIG cmd, + * which are modifiable using SET_PROFILE_CONFIG cmd. + */ + be_cmd_get_profile_config(adapter, &res_mod, NULL, ACTIVE_PROFILE_TYPE, + RESOURCE_MODIFIABLE, 0); + + /* If RSS IFACE capability flags are modifiable for a VF, set the + * capability flag as valid and set RSS and DEFQ_RSS IFACE flags if + * more than 1 RSSQ is available for a VF. + * Otherwise, provision only 1 queue pair for VF. + */ + if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_RSS) { + vft_res->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); + if (num_vf_qs > 1) { + vf_if_cap_flags |= BE_IF_FLAGS_RSS; + if (res.if_cap_flags & BE_IF_FLAGS_DEFQ_RSS) + vf_if_cap_flags |= BE_IF_FLAGS_DEFQ_RSS; + } else { + vf_if_cap_flags &= ~(BE_IF_FLAGS_RSS | + BE_IF_FLAGS_DEFQ_RSS); + } + } else { + num_vf_qs = 1; + } + + if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { + vft_res->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); + vf_if_cap_flags &= ~BE_IF_FLAGS_VLAN_PROMISCUOUS; + } + + vft_res->vf_if_cap_flags = vf_if_cap_flags; + vft_res->max_rx_qs = num_vf_qs; + vft_res->max_rss_qs = num_vf_qs; + vft_res->max_tx_qs = res.max_tx_qs / (num_vfs + 1); + vft_res->max_cq_count = res.max_cq_count / (num_vfs + 1); + + /* Distribute unicast MACs, VLANs, IFACE count and MCCQ count equally + * among the PF and it's VFs, if the fields are changeable + */ + if (res_mod.max_uc_mac == FIELD_MODIFIABLE) + vft_res->max_uc_mac = res.max_uc_mac / (num_vfs + 1); + + if (res_mod.max_vlans == FIELD_MODIFIABLE) + vft_res->max_vlans = res.max_vlans / (num_vfs + 1); + + if (res_mod.max_iface_count == FIELD_MODIFIABLE) + vft_res->max_iface_count = res.max_iface_count / (num_vfs + 1); + + if (res_mod.max_mcc_count == FIELD_MODIFIABLE) + vft_res->max_mcc_count = res.max_mcc_count / (num_vfs + 1); } static int be_clear(struct be_adapter *adapter) { struct pci_dev *pdev = adapter->pdev; - u16 num_vf_qs; + struct be_resources vft_res = {0}; be_cancel_worker(adapter); @@ -3850,16 +3875,15 @@ static int be_clear(struct be_adapter *adapter) */ if (skyhawk_chip(adapter) && be_physfn(adapter) && !pci_vfs_assigned(pdev)) { - num_vf_qs = be_calculate_vf_qs(adapter, - pci_sriov_get_totalvfs(pdev)); + be_calculate_vf_res(adapter, + pci_sriov_get_totalvfs(pdev), + &vft_res); be_cmd_set_sriov_config(adapter, adapter->pool_res, pci_sriov_get_totalvfs(pdev), - num_vf_qs); + &vft_res); } -#ifdef CONFIG_BE2NET_VXLAN be_disable_vxlan_offloads(adapter); -#endif kfree(adapter->pmac_id); adapter->pmac_id = NULL; @@ -3884,7 +3908,8 @@ static int be_vfs_if_create(struct be_adapter *adapter) for_all_vfs(adapter, vf_cfg, vf) { if (!BE3_chip(adapter)) { - status = be_cmd_get_profile_config(adapter, &res, + status = be_cmd_get_profile_config(adapter, &res, NULL, + ACTIVE_PROFILE_TYPE, RESOURCE_LIMITS, vf + 1); if (!status) { @@ -4000,6 +4025,15 @@ static int be_vf_setup(struct be_adapter *adapter) } } + if (BE3_chip(adapter)) { + /* On BE3, enable VEB only when SRIOV is enabled */ + status = be_cmd_set_hsw_config(adapter, 0, 0, + adapter->if_handle, + PORT_FWD_TYPE_VEB, 0); + if (status) + goto err; + } + adapter->flags |= BE_FLAGS_SRIOV_ENABLED; return 0; err: @@ -4069,8 +4103,9 @@ static void BEx_get_resources(struct be_adapter *adapter, /* On a SuperNIC profile, the driver needs to use the * GET_PROFILE_CONFIG cmd to query the per-function TXQ limits */ - be_cmd_get_profile_config(adapter, &super_nic_res, - RESOURCE_LIMITS, 0); + be_cmd_get_profile_config(adapter, &super_nic_res, NULL, + ACTIVE_PROFILE_TYPE, RESOURCE_LIMITS, + 0); /* Some old versions of BE3 FW don't report max_tx_qs value */ res->max_tx_qs = super_nic_res.max_tx_qs ? : BE3_MAX_TX_QS; } else { @@ -4109,12 +4144,38 @@ static void be_setup_init(struct be_adapter *adapter) adapter->cmd_privileges = MIN_PRIVILEGES; } +/* HW supports only MAX_PORT_RSS_TABLES RSS Policy Tables per port. + * However, this HW limitation is not exposed to the host via any SLI cmd. + * As a result, in the case of SRIOV and in particular multi-partition configs + * the driver needs to calcuate a proportional share of RSS Tables per PF-pool + * for distribution between the VFs. This self-imposed limit will determine the + * no: of VFs for which RSS can be enabled. + */ +void be_calculate_pf_pool_rss_tables(struct be_adapter *adapter) +{ + struct be_port_resources port_res = {0}; + u8 rss_tables_on_port; + u16 max_vfs = be_max_vfs(adapter); + + be_cmd_get_profile_config(adapter, NULL, &port_res, SAVED_PROFILE_TYPE, + RESOURCE_LIMITS, 0); + + rss_tables_on_port = MAX_PORT_RSS_TABLES - port_res.nic_pfs; + + /* Each PF Pool's RSS Tables limit = + * PF's Max VFs / Total_Max_VFs on Port * RSS Tables on Port + */ + adapter->pool_res.max_rss_tables = + max_vfs * rss_tables_on_port / port_res.max_vfs; +} + static int be_get_sriov_config(struct be_adapter *adapter) { struct be_resources res = {0}; int max_vfs, old_vfs; - be_cmd_get_profile_config(adapter, &res, RESOURCE_LIMITS, 0); + be_cmd_get_profile_config(adapter, &res, NULL, ACTIVE_PROFILE_TYPE, + RESOURCE_LIMITS, 0); /* Some old versions of BE3 FW don't report max_vfs value */ if (BE3_chip(adapter) && !res.max_vfs) { @@ -4138,13 +4199,19 @@ static int be_get_sriov_config(struct be_adapter *adapter) adapter->num_vfs = old_vfs; } + if (skyhawk_chip(adapter) && be_max_vfs(adapter) && !old_vfs) { + be_calculate_pf_pool_rss_tables(adapter); + dev_info(&adapter->pdev->dev, + "RSS can be enabled for all VFs if num_vfs <= %d\n", + be_max_pf_pool_rss_tables(adapter)); + } return 0; } static void be_alloc_sriov_res(struct be_adapter *adapter) { int old_vfs = pci_num_vf(adapter->pdev); - u16 num_vf_qs; + struct be_resources vft_res = {0}; int status; be_get_sriov_config(adapter); @@ -4158,9 +4225,9 @@ static void be_alloc_sriov_res(struct be_adapter *adapter) * Also, this is done by FW in Lancer chip. */ if (skyhawk_chip(adapter) && be_max_vfs(adapter) && !old_vfs) { - num_vf_qs = be_calculate_vf_qs(adapter, 0); + be_calculate_vf_res(adapter, 0, &vft_res); status = be_cmd_set_sriov_config(adapter, adapter->pool_res, 0, - num_vf_qs); + &vft_res); if (status) dev_err(&adapter->pdev->dev, "Failed to optimize SRIOV resources\n"); @@ -4173,16 +4240,13 @@ static int be_get_resources(struct be_adapter *adapter) struct be_resources res = {0}; int status; - if (BEx_chip(adapter)) { - BEx_get_resources(adapter, &res); - adapter->res = res; - } - /* For Lancer, SH etc read per-function resource limits from FW. * GET_FUNC_CONFIG returns per function guaranteed limits. * GET_PROFILE_CONFIG returns PCI-E related limits PF-pool limits */ - if (!BEx_chip(adapter)) { + if (BEx_chip(adapter)) { + BEx_get_resources(adapter, &res); + } else { status = be_cmd_get_func_config(adapter, &res); if (status) return status; @@ -4191,13 +4255,13 @@ static int be_get_resources(struct be_adapter *adapter) if (res.max_rss_qs && res.max_rss_qs == res.max_rx_qs && !(res.if_cap_flags & BE_IF_FLAGS_DEFQ_RSS)) res.max_rss_qs -= 1; - - /* If RoCE may be enabled stash away half the EQs for RoCE */ - if (be_roce_supported(adapter)) - res.max_evt_qs /= 2; - adapter->res = res; } + /* If RoCE is supported stash away half the EQs for RoCE */ + res.max_nic_evt_qs = be_roce_supported(adapter) ? + res.max_evt_qs / 2 : res.max_evt_qs; + adapter->res = res; + /* If FW supports RSS default queue, then skip creating non-RSS * queue for non-IP traffic. */ @@ -4206,15 +4270,17 @@ static int be_get_resources(struct be_adapter *adapter) dev_info(dev, "Max: txqs %d, rxqs %d, rss %d, eqs %d, vfs %d\n", be_max_txqs(adapter), be_max_rxqs(adapter), - be_max_rss(adapter), be_max_eqs(adapter), + be_max_rss(adapter), be_max_nic_eqs(adapter), be_max_vfs(adapter)); dev_info(dev, "Max: uc-macs %d, mc-macs %d, vlans %d\n", be_max_uc(adapter), be_max_mc(adapter), be_max_vlans(adapter)); - /* Sanitize cfg_num_qs based on HW and platform limits */ - adapter->cfg_num_qs = min_t(u16, netif_get_num_default_rss_queues(), - be_max_qs(adapter)); + /* Ensure RX and TX queues are created in pairs at init time */ + adapter->cfg_num_rx_irqs = + min_t(u16, netif_get_num_default_rss_queues(), + be_max_qp_irqs(adapter)); + adapter->cfg_num_tx_irqs = adapter->cfg_num_rx_irqs; return 0; } @@ -4241,6 +4307,8 @@ static int be_get_config(struct be_adapter *adapter) } be_cmd_get_acpi_wol_cap(adapter); + pci_enable_wake(adapter->pdev, PCI_D3hot, adapter->wol_en); + pci_enable_wake(adapter->pdev, PCI_D3cold, adapter->wol_en); be_cmd_query_port_name(adapter); @@ -4251,15 +4319,6 @@ static int be_get_config(struct be_adapter *adapter) "Using profile 0x%x\n", profile_id); } - status = be_get_resources(adapter); - if (status) - return status; - - adapter->pmac_id = kcalloc(be_max_uc(adapter), - sizeof(*adapter->pmac_id), GFP_KERNEL); - if (!adapter->pmac_id) - return -ENOMEM; - return 0; } @@ -4334,7 +4393,7 @@ static int be_if_create(struct be_adapter *adapter) u32 cap_flags = be_if_cap_flags(adapter); int status; - if (adapter->cfg_num_qs == 1) + if (adapter->cfg_num_rx_irqs == 1) cap_flags &= ~(BE_IF_FLAGS_DEFQ_RSS | BE_IF_FLAGS_RSS); en_flags &= cap_flags; @@ -4460,13 +4519,22 @@ static int be_setup(struct be_adapter *adapter) return status; } + status = be_get_config(adapter); + if (status) + goto err; + if (!BE2_chip(adapter) && be_physfn(adapter)) be_alloc_sriov_res(adapter); - status = be_get_config(adapter); + status = be_get_resources(adapter); if (status) goto err; + adapter->pmac_id = kcalloc(be_max_uc(adapter), + sizeof(*adapter->pmac_id), GFP_KERNEL); + if (!adapter->pmac_id) + return -ENOMEM; + status = be_msix_enable(adapter); if (status) goto err; @@ -4511,6 +4579,15 @@ static int be_setup(struct be_adapter *adapter) be_cmd_set_logical_link_config(adapter, IFLA_VF_LINK_STATE_AUTO, 0); + /* BE3 EVB echoes broadcast/multicast packets back to PF's vport + * confusing a linux bridge or OVS that it might be connected to. + * Set the EVB to PASSTHRU mode which effectively disables the EVB + * when SRIOV is not enabled. + */ + if (BE3_chip(adapter)) + be_cmd_set_hsw_config(adapter, 0, 0, adapter->if_handle, + PORT_FWD_TYPE_PASSTHRU, 0); + if (adapter->num_vfs) be_vf_setup(adapter); @@ -4651,7 +4728,6 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, 0, 0, nlflags, filter_mask, NULL); } -#ifdef CONFIG_BE2NET_VXLAN /* VxLAN offload Notes: * * The stack defines tunnel offload flags (hw_enc_features) for IP and doesn't @@ -4666,13 +4742,17 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, * adds more than one port, disable offloads and don't re-enable them again * until after all the tunnels are removed. */ -static void be_add_vxlan_port(struct net_device *netdev, sa_family_t sa_family, - __be16 port) +static void be_add_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct be_adapter *adapter = netdev_priv(netdev); struct device *dev = &adapter->pdev->dev; + __be16 port = ti->port; int status; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) return; @@ -4720,10 +4800,14 @@ err: be_disable_vxlan_offloads(adapter); } -static void be_del_vxlan_port(struct net_device *netdev, sa_family_t sa_family, - __be16 port) +static void be_del_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct be_adapter *adapter = netdev_priv(netdev); + __be16 port = ti->port; + + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) return; @@ -4785,7 +4869,6 @@ static netdev_features_t be_features_check(struct sk_buff *skb, return features; } -#endif static int be_get_phys_port_id(struct net_device *dev, struct netdev_phys_item_id *ppid) @@ -4833,11 +4916,9 @@ static const struct net_device_ops be_netdev_ops = { #ifdef CONFIG_NET_RX_BUSY_POLL .ndo_busy_poll = be_busy_poll, #endif -#ifdef CONFIG_BE2NET_VXLAN - .ndo_add_vxlan_port = be_add_vxlan_port, - .ndo_del_vxlan_port = be_del_vxlan_port, + .ndo_udp_tunnel_add = be_add_vxlan_port, + .ndo_udp_tunnel_del = be_del_vxlan_port, .ndo_features_check = be_features_check, -#endif .ndo_get_phys_port_id = be_get_phys_port_id, }; @@ -5410,9 +5491,6 @@ static int be_suspend(struct pci_dev *pdev, pm_message_t state) { struct be_adapter *adapter = pci_get_drvdata(pdev); - if (adapter->wol_en) - be_setup_wol(adapter, true); - be_intr_set(adapter, false); be_cancel_err_detection(adapter); @@ -5441,9 +5519,6 @@ static int be_pci_resume(struct pci_dev *pdev) be_schedule_err_detection(adapter, ERR_DETECTION_DELAY); - if (adapter->wol_en) - be_setup_wol(adapter, false); - return 0; } @@ -5552,7 +5627,7 @@ err: static int be_pci_sriov_configure(struct pci_dev *pdev, int num_vfs) { struct be_adapter *adapter = pci_get_drvdata(pdev); - u16 num_vf_qs; + struct be_resources vft_res = {0}; int status; if (!num_vfs) @@ -5575,9 +5650,10 @@ static int be_pci_sriov_configure(struct pci_dev *pdev, int num_vfs) * Also, this is done by FW in Lancer chip. */ if (skyhawk_chip(adapter) && !pci_num_vf(pdev)) { - num_vf_qs = be_calculate_vf_qs(adapter, adapter->num_vfs); + be_calculate_vf_res(adapter, adapter->num_vfs, + &vft_res); status = be_cmd_set_sriov_config(adapter, adapter->pool_res, - adapter->num_vfs, num_vf_qs); + adapter->num_vfs, &vft_res); if (status) dev_err(&pdev->dev, "Failed to optimize SR-IOV resources\n"); diff --git a/drivers/net/ethernet/emulex/benet/be_roce.c b/drivers/net/ethernet/emulex/benet/be_roce.c index 4089156a7f5e..2b62841c4c63 100644 --- a/drivers/net/ethernet/emulex/benet/be_roce.c +++ b/drivers/net/ethernet/emulex/benet/be_roce.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_roce.h b/drivers/net/ethernet/emulex/benet/be_roce.h index fde609789483..e51719a7307f 100644 --- a/drivers/net/ethernet/emulex/benet/be_roce.h +++ b/drivers/net/ethernet/emulex/benet/be_roce.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 41b010645100..4edb98c3c6c7 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -1195,7 +1195,7 @@ static int ethoc_probe(struct platform_device *pdev) priv->mdio = mdiobus_alloc(); if (!priv->mdio) { ret = -ENOMEM; - goto free; + goto free2; } priv->mdio->name = "ethoc-mdio"; @@ -1208,7 +1208,7 @@ static int ethoc_probe(struct platform_device *pdev) ret = mdiobus_register(priv->mdio); if (ret) { dev_err(&netdev->dev, "failed to register MDIO bus\n"); - goto free; + goto free2; } ret = ethoc_mdio_probe(netdev); @@ -1241,9 +1241,10 @@ error2: error: mdiobus_unregister(priv->mdio); mdiobus_free(priv->mdio); -free: +free2: if (priv->clk) clk_disable_unprepare(priv->clk); +free: free_netdev(netdev); out: return ret; diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index f58f9ea51639..92fd5c0bf4df 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -442,6 +442,8 @@ struct bufdesc_ex { #define FEC_QUIRK_SINGLE_MDIO (1 << 11) /* Controller supports RACC register */ #define FEC_QUIRK_HAS_RACC (1 << 12) +/* Controller supports interrupt coalesc */ +#define FEC_QUIRK_HAS_COALESCE (1 << 13) struct bufdesc_prop { int qid; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 3c0255e98535..4040003a74f9 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -111,7 +111,13 @@ static struct platform_device_id fec_devtype[] = { FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM | FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB | FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE | - FEC_QUIRK_HAS_RACC, + FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE, + }, { + .name = "imx6ul-fec", + .driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT | + FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM | + FEC_QUIRK_HAS_VLAN | FEC_QUIRK_BUG_CAPTURE | + FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE, }, { /* sentinel */ } @@ -125,6 +131,7 @@ enum imx_fec_type { IMX6Q_FEC, MVF600_FEC, IMX6SX_FEC, + IMX6UL_FEC, }; static const struct of_device_id fec_dt_ids[] = { @@ -134,6 +141,7 @@ static const struct of_device_id fec_dt_ids[] = { { .compatible = "fsl,imx6q-fec", .data = &fec_devtype[IMX6Q_FEC], }, { .compatible = "fsl,mvf600-fec", .data = &fec_devtype[MVF600_FEC], }, { .compatible = "fsl,imx6sx-fec", .data = &fec_devtype[IMX6SX_FEC], }, + { .compatible = "fsl,imx6ul-fec", .data = &fec_devtype[IMX6UL_FEC], }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, fec_dt_ids); @@ -2358,9 +2366,6 @@ static void fec_enet_itr_coal_set(struct net_device *ndev) struct fec_enet_private *fep = netdev_priv(ndev); int rx_itr, tx_itr; - if (!(fep->quirks & FEC_QUIRK_HAS_AVB)) - return; - /* Must be greater than zero to avoid unpredictable behavior */ if (!fep->rx_time_itr || !fep->rx_pkts_itr || !fep->tx_time_itr || !fep->tx_pkts_itr) @@ -2383,10 +2388,12 @@ static void fec_enet_itr_coal_set(struct net_device *ndev) writel(tx_itr, fep->hwp + FEC_TXIC0); writel(rx_itr, fep->hwp + FEC_RXIC0); - writel(tx_itr, fep->hwp + FEC_TXIC1); - writel(rx_itr, fep->hwp + FEC_RXIC1); - writel(tx_itr, fep->hwp + FEC_TXIC2); - writel(rx_itr, fep->hwp + FEC_RXIC2); + if (fep->quirks & FEC_QUIRK_HAS_AVB) { + writel(tx_itr, fep->hwp + FEC_TXIC1); + writel(rx_itr, fep->hwp + FEC_RXIC1); + writel(tx_itr, fep->hwp + FEC_TXIC2); + writel(rx_itr, fep->hwp + FEC_RXIC2); + } } static int @@ -2394,7 +2401,7 @@ fec_enet_get_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec) { struct fec_enet_private *fep = netdev_priv(ndev); - if (!(fep->quirks & FEC_QUIRK_HAS_AVB)) + if (!(fep->quirks & FEC_QUIRK_HAS_COALESCE)) return -EOPNOTSUPP; ec->rx_coalesce_usecs = fep->rx_time_itr; @@ -2412,28 +2419,28 @@ fec_enet_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec) struct fec_enet_private *fep = netdev_priv(ndev); unsigned int cycle; - if (!(fep->quirks & FEC_QUIRK_HAS_AVB)) + if (!(fep->quirks & FEC_QUIRK_HAS_COALESCE)) return -EOPNOTSUPP; if (ec->rx_max_coalesced_frames > 255) { - pr_err("Rx coalesced frames exceed hardware limiation"); + pr_err("Rx coalesced frames exceed hardware limitation\n"); return -EINVAL; } if (ec->tx_max_coalesced_frames > 255) { - pr_err("Tx coalesced frame exceed hardware limiation"); + pr_err("Tx coalesced frame exceed hardware limitation\n"); return -EINVAL; } cycle = fec_enet_us_to_itr_clock(ndev, fep->rx_time_itr); if (cycle > 0xFFFF) { - pr_err("Rx coalesed usec exceeed hardware limiation"); + pr_err("Rx coalesced usec exceed hardware limitation\n"); return -EINVAL; } cycle = fec_enet_us_to_itr_clock(ndev, fep->tx_time_itr); if (cycle > 0xFFFF) { - pr_err("Rx coalesed usec exceeed hardware limiation"); + pr_err("Rx coalesced usec exceed hardware limitation\n"); return -EINVAL; } @@ -3191,7 +3198,12 @@ static void fec_reset_phy(struct platform_device *pdev) dev_err(&pdev->dev, "failed to get phy-reset-gpios: %d\n", err); return; } - msleep(msec); + + if (msec > 20) + msleep(msec); + else + usleep_range(msec * 1000, msec * 1000 + 1000); + gpio_set_value_cansleep(phy_reset, !active_high); } #else /* CONFIG_OF */ diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 7615e0668acb..d20935dc8399 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -2275,7 +2275,7 @@ static inline void gfar_tx_checksum(struct sk_buff *skb, struct txfcb *fcb, fcb->flags = flags; } -void inline gfar_tx_vlan(struct sk_buff *skb, struct txfcb *fcb) +static inline void gfar_tx_vlan(struct sk_buff *skb, struct txfcb *fcb) { fcb->flags |= TXFCB_VLN; fcb->vlctl = cpu_to_be16(skb_vlan_tag_get(skb)); @@ -2440,7 +2440,8 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) tx_queue->tx_ring_size); if (likely(!nr_frags)) { - lstatus |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT); + if (likely(!do_tstamp)) + lstatus |= BD_LFLAG(TXBD_LAST | TXBD_INTERRUPT); } else { u32 lstatus_start = lstatus; diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig index 4ccc032633c4..2e2566230e27 100644 --- a/drivers/net/ethernet/hisilicon/Kconfig +++ b/drivers/net/ethernet/hisilicon/Kconfig @@ -5,7 +5,7 @@ config NET_VENDOR_HISILICON bool "Hisilicon devices" default y - depends on OF && HAS_DMA + depends on (OF || ACPI) && HAS_DMA depends on ARM || ARM64 || COMPILE_TEST ---help--- If you have a network (Ethernet) card belonging to this class, say Y. diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c index b9f2ea59308a..275618bb4646 100644 --- a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c +++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c @@ -218,7 +218,6 @@ struct hix5hd2_priv { struct device *dev; struct net_device *netdev; - struct phy_device *phy; struct device_node *phy_node; phy_interface_t phy_mode; @@ -402,7 +401,7 @@ static int hix5hd2_net_set_mac_address(struct net_device *dev, void *p) static void hix5hd2_adjust_link(struct net_device *dev) { struct hix5hd2_priv *priv = netdev_priv(dev); - struct phy_device *phy = priv->phy; + struct phy_device *phy = dev->phydev; if ((priv->speed != phy->speed) || (priv->duplex != phy->duplex)) { hix5hd2_config_port(dev, phy->speed, phy->duplex); @@ -679,6 +678,7 @@ static void hix5hd2_free_dma_desc_rings(struct hix5hd2_priv *priv) static int hix5hd2_net_open(struct net_device *dev) { struct hix5hd2_priv *priv = netdev_priv(dev); + struct phy_device *phy; int ret; ret = clk_prepare_enable(priv->clk); @@ -687,12 +687,12 @@ static int hix5hd2_net_open(struct net_device *dev) return ret; } - priv->phy = of_phy_connect(dev, priv->phy_node, - &hix5hd2_adjust_link, 0, priv->phy_mode); - if (!priv->phy) + phy = of_phy_connect(dev, priv->phy_node, + &hix5hd2_adjust_link, 0, priv->phy_mode); + if (!phy) return -ENODEV; - phy_start(priv->phy); + phy_start(phy); hix5hd2_hw_init(priv); hix5hd2_rx_refill(priv); @@ -716,9 +716,9 @@ static int hix5hd2_net_close(struct net_device *dev) netif_stop_queue(dev); hix5hd2_free_dma_desc_rings(priv); - if (priv->phy) { - phy_stop(priv->phy); - phy_disconnect(priv->phy); + if (dev->phydev) { + phy_stop(dev->phydev); + phy_disconnect(dev->phydev); } clk_disable_unprepare(priv->clk); @@ -750,32 +750,10 @@ static const struct net_device_ops hix5hd2_netdev_ops = { .ndo_set_mac_address = hix5hd2_net_set_mac_address, }; -static int hix5hd2_get_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) -{ - struct hix5hd2_priv *priv = netdev_priv(net_dev); - - if (!priv->phy) - return -ENODEV; - - return phy_ethtool_gset(priv->phy, cmd); -} - -static int hix5hd2_set_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) -{ - struct hix5hd2_priv *priv = netdev_priv(net_dev); - - if (!priv->phy) - return -ENODEV; - - return phy_ethtool_sset(priv->phy, cmd); -} - static struct ethtool_ops hix5hd2_ethtools_ops = { .get_link = ethtool_op_get_link, - .get_settings = hix5hd2_get_settings, - .set_settings = hix5hd2_set_settings, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int hix5hd2_mdio_wait_ready(struct mii_bus *bus) diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c index 5d3047cc8380..c54c6fac0d1d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.c +++ b/drivers/net/ethernet/hisilicon/hns/hnae.c @@ -400,7 +400,6 @@ int hnae_ae_register(struct hnae_ae_dev *hdev, struct module *owner) if (!hdev->ops || !hdev->ops->get_handle || !hdev->ops->toggle_ring_irq || - !hdev->ops->toggle_queue_status || !hdev->ops->get_status || !hdev->ops->adjust_link) return -EINVAL; diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h index 529cb1341270..e093cbf26c8c 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.h +++ b/drivers/net/ethernet/hisilicon/hns/hnae.h @@ -363,6 +363,14 @@ enum hnae_port_type { HNAE_PORT_DEBUG }; +/* mac media type */ +enum hnae_media_type { + HNAE_MEDIA_TYPE_UNKNOWN = 0, + HNAE_MEDIA_TYPE_FIBER, + HNAE_MEDIA_TYPE_COPPER, + HNAE_MEDIA_TYPE_BACKPLANE, +}; + /* This struct defines the operation on the handle. * * get_handle(): (mandatory) @@ -454,7 +462,6 @@ struct hnae_ae_ops { int (*get_info)(struct hnae_handle *handle, u8 *auto_neg, u16 *speed, u8 *duplex); void (*toggle_ring_irq)(struct hnae_ring *ring, u32 val); - void (*toggle_queue_status)(struct hnae_queue *queue, u32 val); void (*adjust_link)(struct hnae_handle *handle, int speed, int duplex); int (*set_loopback)(struct hnae_handle *handle, enum hnae_loop loop_mode, int en); @@ -473,6 +480,11 @@ struct hnae_ae_ops { int (*set_coalesce_usecs)(struct hnae_handle *handle, u32 timeout); int (*set_coalesce_frames)(struct hnae_handle *handle, u32 coalesce_frames); + void (*get_coalesce_range)(struct hnae_handle *handle, + u32 *tx_frames_low, u32 *rx_frames_low, + u32 *tx_frames_high, u32 *rx_frames_high, + u32 *tx_usecs_low, u32 *rx_usecs_low, + u32 *tx_usecs_high, u32 *rx_usecs_high); void (*set_promisc_mode)(struct hnae_handle *handle, u32 en); int (*get_mac_addr)(struct hnae_handle *handle, void **p); int (*set_mac_addr)(struct hnae_handle *handle, void *p); @@ -521,6 +533,7 @@ struct hnae_handle { u32 eport_id; u32 dport_id; /* v2 tx bd should fill the dport_id */ enum hnae_port_type port_type; + enum hnae_media_type media_type; struct list_head node; /* list to hnae_ae_dev->handle_list */ struct hnae_buf_ops *bops; /* operation for the buffer */ struct hnae_queue **qs; /* array base of all queues */ diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index d37b778b4a1b..e28d960997af 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -134,6 +134,7 @@ struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev, ae_handle->phy_dev = vf_cb->mac_cb->phy_dev; ae_handle->if_support = vf_cb->mac_cb->if_support; ae_handle->port_type = vf_cb->mac_cb->mac_type; + ae_handle->media_type = vf_cb->mac_cb->media_type; ae_handle->dport_id = port_id; return ae_handle; @@ -247,12 +248,21 @@ static void hns_ae_set_tso_stats(struct hnae_handle *handle, int enable) static int hns_ae_start(struct hnae_handle *handle) { int ret; + int k; struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle); ret = hns_mac_vm_config_bc_en(mac_cb, 0, true); if (ret) return ret; + for (k = 0; k < handle->q_num; k++) { + if (AE_IS_VER1(mac_cb->dsaf_dev->dsaf_ver)) + hns_rcb_int_clr_hw(handle->qs[k], + RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); + else + hns_rcbv2_int_clr_hw(handle->qs[k], + RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); + } hns_ae_ring_enable_all(handle, 1); msleep(100); @@ -313,18 +323,6 @@ static void hns_aev2_toggle_ring_irq(struct hnae_ring *ring, u32 mask) hns_rcbv2_int_ctrl_hw(ring->q, flag, mask); } -static void hns_ae_toggle_queue_status(struct hnae_queue *queue, u32 val) -{ - struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(queue->dev); - - if (AE_IS_VER1(dsaf_dev->dsaf_ver)) - hns_rcb_int_clr_hw(queue, RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); - else - hns_rcbv2_int_clr_hw(queue, RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); - - hns_rcb_start(queue, val); -} - static int hns_ae_get_link_status(struct hnae_handle *handle) { u32 link_status; @@ -465,6 +463,30 @@ static int hns_ae_set_coalesce_frames(struct hnae_handle *handle, ring_pair->port_id_in_comm, coalesce_frames); } +static void hns_ae_get_coalesce_range(struct hnae_handle *handle, + u32 *tx_frames_low, u32 *rx_frames_low, + u32 *tx_frames_high, u32 *rx_frames_high, + u32 *tx_usecs_low, u32 *rx_usecs_low, + u32 *tx_usecs_high, u32 *rx_usecs_high) +{ + struct dsaf_device *dsaf_dev; + + dsaf_dev = hns_ae_get_dsaf_dev(handle->dev); + + *tx_frames_low = HNS_RCB_MIN_COALESCED_FRAMES; + *rx_frames_low = HNS_RCB_MIN_COALESCED_FRAMES; + *tx_frames_high = + (dsaf_dev->desc_num - 1 > HNS_RCB_MAX_COALESCED_FRAMES) ? + HNS_RCB_MAX_COALESCED_FRAMES : dsaf_dev->desc_num - 1; + *rx_frames_high = + (dsaf_dev->desc_num - 1 > HNS_RCB_MAX_COALESCED_FRAMES) ? + HNS_RCB_MAX_COALESCED_FRAMES : dsaf_dev->desc_num - 1; + *tx_usecs_low = 0; + *rx_usecs_low = 0; + *tx_usecs_high = HNS_RCB_MAX_COALESCED_USECS; + *rx_usecs_high = HNS_RCB_MAX_COALESCED_USECS; +} + void hns_ae_update_stats(struct hnae_handle *handle, struct net_device_stats *net_stats) { @@ -587,6 +609,7 @@ void hns_ae_get_strings(struct hnae_handle *handle, int idx; struct hns_mac_cb *mac_cb; struct hns_ppe_cb *ppe_cb; + struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(handle->dev); u8 *p = data; struct hnae_vf_cb *vf_cb; @@ -609,13 +632,14 @@ void hns_ae_get_strings(struct hnae_handle *handle, p += ETH_GSTRING_LEN * hns_mac_get_sset_count(mac_cb, stringset); if (mac_cb->mac_type == HNAE_PORT_SERVICE) - hns_dsaf_get_strings(stringset, p, port); + hns_dsaf_get_strings(stringset, p, port, dsaf_dev); } int hns_ae_get_sset_count(struct hnae_handle *handle, int stringset) { u32 sset_count = 0; struct hns_mac_cb *mac_cb; + struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(handle->dev); assert(handle); @@ -626,7 +650,7 @@ int hns_ae_get_sset_count(struct hnae_handle *handle, int stringset) sset_count += hns_mac_get_sset_count(mac_cb, stringset); if (mac_cb->mac_type == HNAE_PORT_SERVICE) - sset_count += hns_dsaf_get_sset_count(stringset); + sset_count += hns_dsaf_get_sset_count(dsaf_dev, stringset); return sset_count; } @@ -782,7 +806,6 @@ static struct hnae_ae_ops hns_dsaf_ops = { .stop = hns_ae_stop, .reset = hns_ae_reset, .toggle_ring_irq = hns_ae_toggle_ring_irq, - .toggle_queue_status = hns_ae_toggle_queue_status, .get_status = hns_ae_get_link_status, .get_info = hns_ae_get_mac_info, .adjust_link = hns_ae_adjust_link, @@ -796,6 +819,7 @@ static struct hnae_ae_ops hns_dsaf_ops = { .get_rx_max_coalesced_frames = hns_ae_get_rx_max_coalesced_frames, .set_coalesce_usecs = hns_ae_set_coalesce_usecs, .set_coalesce_frames = hns_ae_set_coalesce_frames, + .get_coalesce_range = hns_ae_get_coalesce_range, .set_promisc_mode = hns_ae_set_promisc_mode, .set_mac_addr = hns_ae_set_mac_address, .set_mc_addr = hns_ae_set_multicast_one, diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index c526558e6367..3fb87e233c49 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -56,20 +56,6 @@ static const enum mac_mode g_mac_mode_1000[] = { [PHY_INTERFACE_MODE_RTBI] = MAC_MODE_RTBI_1000 }; -static enum mac_mode hns_mac_dev_to_enet_if(const struct hns_mac_cb *mac_cb) -{ - switch (mac_cb->max_speed) { - case MAC_SPEED_100: - return g_mac_mode_100[mac_cb->phy_if]; - case MAC_SPEED_1000: - return g_mac_mode_1000[mac_cb->phy_if]; - case MAC_SPEED_10000: - return MAC_MODE_XGMII_10000; - default: - return MAC_MODE_MII_100; - } -} - static enum mac_mode hns_get_enet_interface(const struct hns_mac_cb *mac_cb) { switch (mac_cb->max_speed) { @@ -134,7 +120,6 @@ void hns_mac_adjust_link(struct hns_mac_cb *mac_cb, int speed, int duplex) mac_cb->speed = speed; mac_cb->half_duplex = !duplex; - mac_ctrl_drv->mac_mode = hns_mac_dev_to_enet_if(mac_cb); if (mac_ctrl_drv->adjust_link) { ret = mac_ctrl_drv->adjust_link(mac_ctrl_drv, @@ -748,6 +733,18 @@ static void hns_mac_register_phy(struct hns_mac_cb *mac_cb) mac_cb->mac_id, addr); } +#define MAC_MEDIA_TYPE_MAX_LEN 16 + +static const struct { + enum hnae_media_type value; + const char *name; +} media_type_defs[] = { + {HNAE_MEDIA_TYPE_UNKNOWN, "unknown" }, + {HNAE_MEDIA_TYPE_FIBER, "fiber" }, + {HNAE_MEDIA_TYPE_COPPER, "copper" }, + {HNAE_MEDIA_TYPE_BACKPLANE, "backplane" }, +}; + /** *hns_mac_get_info - get mac information from device node *@mac_cb: mac device @@ -759,10 +756,13 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) struct device_node *np; struct regmap *syscon; struct of_phandle_args cpld_args; + const char *media_type; + u32 i; u32 ret; mac_cb->link = false; mac_cb->half_duplex = false; + mac_cb->media_type = HNAE_MEDIA_TYPE_UNKNOWN; mac_cb->speed = mac_phy_to_speed[mac_cb->phy_if]; mac_cb->max_speed = mac_cb->speed; @@ -864,6 +864,17 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) mac_cb->mac_id); } + if (!fwnode_property_read_string(mac_cb->fw_port, "media-type", + &media_type)) { + for (i = 0; i < ARRAY_SIZE(media_type_defs); i++) { + if (!strncmp(media_type_defs[i].name, media_type, + MAC_MEDIA_TYPE_MAX_LEN)) { + mac_cb->media_type = media_type_defs[i].value; + break; + } + } + } + return 0; } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h index 05a6e8f7a419..4cbdf14f5c16 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h @@ -335,6 +335,7 @@ struct hns_mac_cb { u64 txpkt_for_led; u64 rxpkt_for_led; enum hnae_port_type mac_type; + enum hnae_media_type media_type; phy_interface_t phy_if; enum hnae_loop loop_mode; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index ac03c4a53e08..2ef4277d00b3 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -114,9 +114,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) dsaf_dev->sc_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->sc_base) { + if (IS_ERR(dsaf_dev->sc_base)) { dev_err(dsaf_dev->dev, "subctrl can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->sc_base); } res = platform_get_resource(pdev, IORESOURCE_MEM, @@ -128,9 +128,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) dsaf_dev->sds_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->sds_base) { + if (IS_ERR(dsaf_dev->sds_base)) { dev_err(dsaf_dev->dev, "serdes-ctrl can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->sds_base); } } else { dsaf_dev->sub_ctrl = syscon; @@ -146,9 +146,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) } } dsaf_dev->ppe_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->ppe_base) { + if (IS_ERR(dsaf_dev->ppe_base)) { dev_err(dsaf_dev->dev, "ppe-base resource can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->ppe_base); } dsaf_dev->ppe_paddr = res->start; @@ -165,9 +165,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) } } dsaf_dev->io_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->io_base) { + if (IS_ERR(dsaf_dev->io_base)) { dev_err(dsaf_dev->dev, "dsaf-base resource can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->io_base); } } @@ -176,7 +176,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) desc_num > HNS_DSAF_MAX_DESC_CNT) { dev_err(dsaf_dev->dev, "get desc-num(%d) fail, ret=%d!\n", desc_num, ret); - goto unmap_base_addr; + return -EINVAL; } dsaf_dev->desc_num = desc_num; @@ -192,7 +192,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) if (ret < 0) { dev_err(dsaf_dev->dev, "get buf-size fail, ret=%d!\r\n", ret); - goto unmap_base_addr; + return ret; } dsaf_dev->buf_size = buf_size; @@ -200,7 +200,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) if (dsaf_dev->buf_size_type < 0) { dev_err(dsaf_dev->dev, "buf_size(%d) is wrong!\n", buf_size); - goto unmap_base_addr; + return -EINVAL; } dsaf_dev->misc_op = hns_misc_op_get(dsaf_dev); @@ -213,32 +213,6 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) dev_err(dsaf_dev->dev, "set mask to 64bit fail!\n"); return 0; - -unmap_base_addr: - if (dsaf_dev->io_base) - iounmap(dsaf_dev->io_base); - if (dsaf_dev->ppe_base) - iounmap(dsaf_dev->ppe_base); - if (dsaf_dev->sds_base) - iounmap(dsaf_dev->sds_base); - if (dsaf_dev->sc_base) - iounmap(dsaf_dev->sc_base); - return ret; -} - -static void hns_dsaf_free_cfg(struct dsaf_device *dsaf_dev) -{ - if (dsaf_dev->io_base) - iounmap(dsaf_dev->io_base); - - if (dsaf_dev->ppe_base) - iounmap(dsaf_dev->ppe_base); - - if (dsaf_dev->sds_base) - iounmap(dsaf_dev->sds_base); - - if (dsaf_dev->sc_base) - iounmap(dsaf_dev->sc_base); } /** @@ -542,10 +516,10 @@ static void hns_dsafv2_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev) o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG3_SET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG3_SET_BUF_NUM_NO_PFC_S, 110); + DSAFV2_SBM_CFG3_SET_BUF_NUM_NO_PFC_S, 48); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG3_RESET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG3_RESET_BUF_NUM_NO_PFC_S, 160); + DSAFV2_SBM_CFG3_RESET_BUF_NUM_NO_PFC_S, 80); dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); /* for no enable pfc mode */ @@ -553,29 +527,39 @@ static void hns_dsafv2_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev) o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG4_SET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG4_SET_BUF_NUM_NO_PFC_S, 128); + DSAFV2_SBM_CFG4_SET_BUF_NUM_NO_PFC_S, 192); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_S, 192); + DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_S, 240); dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); } /* PPE */ - reg = DSAF_SBM_BP_CFG_2_PPE_REG_0_REG + 0x80 * i; - o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_SET_BUF_NUM_M, - DSAFV2_SBM_CFG2_SET_BUF_NUM_S, 10); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_RESET_BUF_NUM_M, - DSAFV2_SBM_CFG2_RESET_BUF_NUM_S, 12); - dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); + for (i = 0; i < DSAFV2_SBM_PPE_CHN; i++) { + reg = DSAF_SBM_BP_CFG_2_PPE_REG_0_REG + 0x80 * i; + o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_M, + DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_S, 2); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_M, + DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_S, 3); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_M, + DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_S, 52); + dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); + } + /* RoCEE */ for (i = 0; i < DASFV2_ROCEE_CRD_NUM; i++) { reg = DSAFV2_SBM_BP_CFG_2_ROCEE_REG_0_REG + 0x80 * i; o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_SET_BUF_NUM_M, - DSAFV2_SBM_CFG2_SET_BUF_NUM_S, 2); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_RESET_BUF_NUM_M, - DSAFV2_SBM_CFG2_RESET_BUF_NUM_S, 4); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_M, + DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_S, 2); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_M, + DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_S, 4); dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); } } @@ -886,6 +870,8 @@ static void hns_dsaf_single_line_tbl_cfg( struct dsaf_device *dsaf_dev, u32 address, struct dsaf_tbl_line_cfg *ptbl_line) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_line_addr_cfg(dsaf_dev, address); @@ -894,6 +880,8 @@ static void hns_dsaf_single_line_tbl_cfg( /*Write Plus*/ hns_dsaf_tbl_line_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -907,6 +895,8 @@ static void hns_dsaf_tcam_uc_cfg( struct dsaf_tbl_tcam_data *ptbl_tcam_data, struct dsaf_tbl_tcam_ucast_cfg *ptbl_tcam_ucast) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); /*Write Tcam Data*/ @@ -915,6 +905,8 @@ static void hns_dsaf_tcam_uc_cfg( hns_dsaf_tbl_tcam_ucast_cfg(dsaf_dev, ptbl_tcam_ucast); /*Write Plus*/ hns_dsaf_tbl_tcam_data_ucast_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -929,6 +921,8 @@ static void hns_dsaf_tcam_mc_cfg( struct dsaf_tbl_tcam_data *ptbl_tcam_data, struct dsaf_tbl_tcam_mcast_cfg *ptbl_tcam_mcast) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); /*Write Tcam Data*/ @@ -937,6 +931,8 @@ static void hns_dsaf_tcam_mc_cfg( hns_dsaf_tbl_tcam_mcast_cfg(dsaf_dev, ptbl_tcam_mcast); /*Write Plus*/ hns_dsaf_tbl_tcam_data_mcast_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -946,6 +942,8 @@ static void hns_dsaf_tcam_mc_cfg( */ static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); @@ -958,6 +956,8 @@ static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address) /*Write Plus*/ hns_dsaf_tbl_tcam_mcast_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -975,6 +975,8 @@ static void hns_dsaf_tcam_uc_get( u32 tcam_read_data0; u32 tcam_read_data4; + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); @@ -983,9 +985,9 @@ static void hns_dsaf_tcam_uc_get( /*read tcam data*/ ptbl_tcam_data->tbl_tcam_data_high - = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); - ptbl_tcam_data->tbl_tcam_data_low = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_HIGH_0_REG); + ptbl_tcam_data->tbl_tcam_data_low + = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); /*read tcam mcast*/ tcam_read_data0 = dsaf_read_dev(dsaf_dev, @@ -1007,6 +1009,8 @@ static void hns_dsaf_tcam_uc_get( DSAF_TBL_UCAST_CFG1_OUT_PORT_S); ptbl_tcam_ucast->tbl_ucast_dvc = dsaf_get_bit(tcam_read_data0, DSAF_TBL_UCAST_CFG1_DVC_S); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -1023,6 +1027,8 @@ static void hns_dsaf_tcam_mc_get( { u32 data_tmp; + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); @@ -1031,9 +1037,9 @@ static void hns_dsaf_tcam_mc_get( /*read tcam data*/ ptbl_tcam_data->tbl_tcam_data_high = - dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); - ptbl_tcam_data->tbl_tcam_data_low = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_HIGH_0_REG); + ptbl_tcam_data->tbl_tcam_data_low = + dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); /*read tcam mcast*/ ptbl_tcam_mcast->tbl_mcast_port_msk[0] = @@ -1053,6 +1059,8 @@ static void hns_dsaf_tcam_mc_get( ptbl_tcam_mcast->tbl_mcast_port_msk[4] = dsaf_get_field(data_tmp, DSAF_TBL_MCAST_CFG4_VM128_112_M, DSAF_TBL_MCAST_CFG4_VM128_112_S); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -1114,10 +1122,10 @@ int hns_dsaf_set_rx_mac_pause_en(struct dsaf_device *dsaf_dev, int mac_id, u32 en) { if (AE_IS_VER1(dsaf_dev->dsaf_ver)) { - if (!en) + if (!en) { dev_err(dsaf_dev->dev, "dsafv1 can't close rx_pause!\n"); - - return -EINVAL; + return -EINVAL; + } } dsaf_set_dev_bit(dsaf_dev, DSAF_PAUSE_CFG_REG + mac_id * 4, @@ -1377,6 +1385,7 @@ static int hns_dsaf_init(struct dsaf_device *dsaf_dev) if (HNS_DSAF_IS_DEBUG(dsaf_dev)) return 0; + spin_lock_init(&dsaf_dev->tcam_lock); ret = hns_dsaf_init_hw(dsaf_dev); if (ret) return ret; @@ -2122,11 +2131,24 @@ void hns_dsaf_fix_mac_mode(struct hns_mac_cb *mac_cb) hns_dsaf_port_work_rate_cfg(dsaf_dev, mac_id, mode); } +static u32 hns_dsaf_get_inode_prio_reg(int index) +{ + int base_index, offset; + u32 base_addr = DSAF_INODE_IN_PRIO_PAUSE_BASE_REG; + + base_index = (index + 1) / DSAF_REG_PER_ZONE; + offset = (index + 1) % DSAF_REG_PER_ZONE; + + return base_addr + DSAF_INODE_IN_PRIO_PAUSE_BASE_OFFSET * base_index + + DSAF_INODE_IN_PRIO_PAUSE_OFFSET * offset; +} + void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num) { struct dsaf_hw_stats *hw_stats = &dsaf_dev->hw_stats[node_num]; bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); + int i; u32 reg_tmp; hw_stats->pad_drop += dsaf_read_dev(dsaf_dev, @@ -2161,6 +2183,18 @@ void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num) hw_stats->stp_drop += dsaf_read_dev(dsaf_dev, DSAF_INODE_IN_DATA_STP_DISC_0_REG + 0x80 * (u64)node_num); + /* pfc pause frame statistics stored in dsaf inode*/ + if ((node_num < DSAF_SERVICE_NW_NUM) && !is_ver1) { + for (i = 0; i < DSAF_PRIO_NR; i++) { + reg_tmp = hns_dsaf_get_inode_prio_reg(i); + hw_stats->rx_pfc[i] += dsaf_read_dev(dsaf_dev, + reg_tmp + 0x4 * (u64)node_num); + hw_stats->tx_pfc[i] += dsaf_read_dev(dsaf_dev, + DSAF_XOD_XGE_PFC_PRIO_CNT_BASE_REG + + DSAF_XOD_XGE_PFC_PRIO_CNT_OFFSET * i + + 0xF0 * (u64)node_num); + } + } hw_stats->tx_pkts += dsaf_read_dev(dsaf_dev, DSAF_XOD_RCVPKT_CNT_0_REG + 0x90 * (u64)node_num); } @@ -2498,38 +2532,53 @@ void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data) p[i] = 0xdddddddd; } -static char *hns_dsaf_get_node_stats_strings(char *data, int node) +static char *hns_dsaf_get_node_stats_strings(char *data, int node, + struct dsaf_device *dsaf_dev) { char *buff = data; + int i; + bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); snprintf(buff, ETH_GSTRING_LEN, "innod%d_pad_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_manage_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pkt_id", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pause_frame", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_release_buf_num", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_sbm_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_crc_false_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_bp_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_lookup_rslt_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_local_rslt_fail_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_vlan_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_stp_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; + if (node < DSAF_SERVICE_NW_NUM && !is_ver1) { + for (i = 0; i < DSAF_PRIO_NR; i++) { + snprintf(buff + 0 * ETH_GSTRING_LEN * DSAF_PRIO_NR, + ETH_GSTRING_LEN, "inod%d_pfc_prio%d_pkts", + node, i); + snprintf(buff + 1 * ETH_GSTRING_LEN * DSAF_PRIO_NR, + ETH_GSTRING_LEN, "onod%d_pfc_prio%d_pkts", + node, i); + buff += ETH_GSTRING_LEN; + } + buff += 1 * DSAF_PRIO_NR * ETH_GSTRING_LEN; + } snprintf(buff, ETH_GSTRING_LEN, "onnod%d_tx_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; return buff; } @@ -2538,7 +2587,9 @@ static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data, int node_num) { u64 *p = data; + int i; struct dsaf_hw_stats *hw_stats = &ddev->hw_stats[node_num]; + bool is_ver1 = AE_IS_VER1(ddev->dsaf_ver); p[0] = hw_stats->pad_drop; p[1] = hw_stats->man_pkts; @@ -2553,8 +2604,16 @@ static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data, p[10] = hw_stats->local_addr_false; p[11] = hw_stats->vlan_drop; p[12] = hw_stats->stp_drop; - p[13] = hw_stats->tx_pkts; + if (node_num < DSAF_SERVICE_NW_NUM && !is_ver1) { + for (i = 0; i < DSAF_PRIO_NR; i++) { + p[13 + i + 0 * DSAF_PRIO_NR] = hw_stats->rx_pfc[i]; + p[13 + i + 1 * DSAF_PRIO_NR] = hw_stats->tx_pfc[i]; + } + p[29] = hw_stats->tx_pkts; + return &p[30]; + } + p[13] = hw_stats->tx_pkts; return &p[14]; } @@ -2582,11 +2641,16 @@ void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port) *@stringset: type of values in data *return dsaf string name count */ -int hns_dsaf_get_sset_count(int stringset) +int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset) { - if (stringset == ETH_SS_STATS) - return DSAF_STATIC_NUM; + bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); + if (stringset == ETH_SS_STATS) { + if (is_ver1) + return DSAF_STATIC_NUM; + else + return DSAF_V2_STATIC_NUM; + } return 0; } @@ -2596,7 +2660,8 @@ int hns_dsaf_get_sset_count(int stringset) *@data:strings name value *@port:port index */ -void hns_dsaf_get_strings(int stringset, u8 *data, int port) +void hns_dsaf_get_strings(int stringset, u8 *data, int port, + struct dsaf_device *dsaf_dev) { char *buff = (char *)data; int node = port; @@ -2605,11 +2670,11 @@ void hns_dsaf_get_strings(int stringset, u8 *data, int port) return; /* for ge/xge node info */ - buff = hns_dsaf_get_node_stats_strings(buff, node); + buff = hns_dsaf_get_node_stats_strings(buff, node, dsaf_dev); /* for ppe node info */ node = port + DSAF_PPE_INODE_BASE; - (void)hns_dsaf_get_node_stats_strings(buff, node); + (void)hns_dsaf_get_node_stats_strings(buff, node, dsaf_dev); } /** @@ -2645,7 +2710,7 @@ static int hns_dsaf_probe(struct platform_device *pdev) ret = hns_dsaf_init(dsaf_dev); if (ret) - goto free_cfg; + goto free_dev; ret = hns_mac_init(dsaf_dev); if (ret) @@ -2670,9 +2735,6 @@ uninit_mac: uninit_dsaf: hns_dsaf_free(dsaf_dev); -free_cfg: - hns_dsaf_free_cfg(dsaf_dev); - free_dev: hns_dsaf_free_dev(dsaf_dev); @@ -2695,8 +2757,6 @@ static int hns_dsaf_remove(struct platform_device *pdev) hns_dsaf_free(dsaf_dev); - hns_dsaf_free_cfg(dsaf_dev); - hns_dsaf_free_dev(dsaf_dev); return 0; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h index 2e55b3cc4515..1daf018d9071 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h @@ -39,6 +39,9 @@ struct hns_mac_cb; #define DSAF_DUMP_REGS_NUM 504 #define DSAF_STATIC_NUM 28 +#define DSAF_V2_STATIC_NUM 44 +#define DSAF_PRIO_NR 8 +#define DSAF_REG_PER_ZONE 3 #define DSAF_STATS_READ(p, offset) (*((u64 *)((u8 *)(p) + (offset)))) #define HNS_DSAF_IS_DEBUG(dev) (dev->dsaf_mode == DSAF_MODE_DISABLE_SP) @@ -176,6 +179,8 @@ struct dsaf_hw_stats { u64 local_addr_false; u64 vlan_drop; u64 stp_drop; + u64 rx_pfc[DSAF_PRIO_NR]; + u64 tx_pfc[DSAF_PRIO_NR]; u64 tx_pkts; }; @@ -317,6 +322,8 @@ struct dsaf_device { struct dsaf_hw_stats hw_stats[DSAF_NODE_NUM]; struct dsaf_int_stat int_stat; + /* make sure tcam table config spinlock */ + spinlock_t tcam_lock; }; static inline void *hns_dsaf_dev_priv(const struct dsaf_device *dsaf_dev) @@ -417,9 +424,10 @@ void hns_dsaf_ae_uninit(struct dsaf_device *dsaf_dev); void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 inode_num); -int hns_dsaf_get_sset_count(int stringset); +int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset); void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port); -void hns_dsaf_get_strings(int stringset, u8 *data, int port); +void hns_dsaf_get_strings(int stringset, u8 *data, int port, + struct dsaf_device *dsaf_dev); void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data); int hns_dsaf_get_regs_count(void); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index 96cb628a7c5f..611b67b6f450 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -86,9 +86,10 @@ static void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status, mac_cb->cpld_led_value = value; } } else { - dsaf_write_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg, - CPLD_LED_DEFAULT_VALUE); - mac_cb->cpld_led_value = CPLD_LED_DEFAULT_VALUE; + value = (mac_cb->cpld_led_value) & (0x1 << DSAF_LED_ANCHOR_B); + dsaf_write_syscon(mac_cb->cpld_ctrl, + mac_cb->cpld_ctrl_reg, value); + mac_cb->cpld_led_value = value; } } @@ -114,7 +115,7 @@ static int cpld_set_led_id(struct hns_mac_cb *mac_cb, CPLD_LED_ON_VALUE); dsaf_write_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg, mac_cb->cpld_led_value); - return 2; + break; case HNAE_LED_INACTIVE: dsaf_set_bit(mac_cb->cpld_led_value, DSAF_LED_ANCHOR_B, CPLD_LED_DEFAULT_VALUE); @@ -122,7 +123,8 @@ static int cpld_set_led_id(struct hns_mac_cb *mac_cb, mac_cb->cpld_led_value); break; default: - break; + dev_err(mac_cb->dev, "invalid led state: %d!", status); + return -EINVAL; } return 0; @@ -251,10 +253,9 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, reg_val_1 = 0x1 << port; port_rst_off = dsaf_dev->mac_cb[port]->port_rst_off; /* there is difference between V1 and V2 in register.*/ - if (AE_IS_VER1(dsaf_dev->dsaf_ver)) - reg_val_2 = 0x1041041 << port_rst_off; - else - reg_val_2 = 0x2082082 << port_rst_off; + reg_val_2 = AE_IS_VER1(dsaf_dev->dsaf_ver) ? + 0x1041041 : 0x2082082; + reg_val_2 <<= port_rst_off; if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, @@ -270,8 +271,11 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, reg_val_1); } } else { - reg_val_1 = 0x15540 << dsaf_dev->reset_offset; - reg_val_2 = 0x100 << dsaf_dev->reset_offset; + reg_val_1 = 0x15540; + reg_val_2 = AE_IS_VER1(dsaf_dev->dsaf_ver) ? 0x100 : 0x40; + + reg_val_1 <<= dsaf_dev->reset_offset; + reg_val_2 <<= dsaf_dev->reset_offset; if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, @@ -431,11 +435,6 @@ int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt) */ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) { - /* port 0-3 hilink4 base is serdes_vaddr + 0x00280000 - * port 4-7 hilink3 base is serdes_vaddr + 0x00200000 - */ - u8 *base_addr = (u8 *)mac_cb->serdes_vaddr + - (mac_cb->mac_id <= 3 ? 0x00280000 : 0x00200000); const u8 lane_id[] = { 0, /* mac 0 -> lane 0 */ 1, /* mac 1 -> lane 1 */ @@ -461,11 +460,30 @@ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) } if (mac_cb->serdes_ctrl) { - u32 origin = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset); + u32 origin; + + if (!AE_IS_VER1(mac_cb->dsaf_dev->dsaf_ver)) { +#define HILINK_ACCESS_SEL_CFG 0x40008 + /* hilink4 & hilink3 use the same xge training and + * xge u adaptor. There is a hilink access sel cfg + * register to select which one to be configed + */ + if ((!HNS_DSAF_IS_DEBUG(mac_cb->dsaf_dev)) && + (mac_cb->mac_id <= 3)) + dsaf_write_syscon(mac_cb->serdes_ctrl, + HILINK_ACCESS_SEL_CFG, 0); + else + dsaf_write_syscon(mac_cb->serdes_ctrl, + HILINK_ACCESS_SEL_CFG, 3); + } + + origin = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset); dsaf_set_field(origin, 1ull << 10, 10, en); dsaf_write_syscon(mac_cb->serdes_ctrl, reg_offset, origin); } else { + u8 *base_addr = (u8 *)mac_cb->serdes_vaddr + + (mac_cb->mac_id <= 3 ? 0x00280000 : 0x00200000); dsaf_set_reg_field(base_addr, reg_offset, 1ull << 10, 10, en); } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c index 3ce24097fcf4..ef1107777c08 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c @@ -540,7 +540,7 @@ int hns_rcb_set_coalesce_usecs( } if (timeout > HNS_RCB_MAX_COALESCED_USECS) { dev_err(rcb_common->dsaf_dev->dev, - "error: not support coalesce %dus!\n", timeout); + "error: coalesce_usecs setting supports 0~1023us\n"); return -EINVAL; } hns_rcb_set_port_timeout(rcb_common, port_idx, timeout); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h index bd54dac82ee0..99b4e1ba0a94 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h @@ -40,7 +40,7 @@ struct rcb_common_cb; #define HNS_RCB_DEF_COALESCED_FRAMES 50 #define HNS_RCB_CLK_FREQ_MHZ 350 #define HNS_RCB_MAX_COALESCED_USECS 0x3ff -#define HNS_RCB_DEF_COALESCED_USECS 3 +#define HNS_RCB_DEF_COALESCED_USECS 50 #define HNS_RCB_COMMON_ENDIAN 1 diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h index 7c3b5103d151..235f74444b1d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h @@ -32,7 +32,7 @@ #define DSAFV2_SBM_NUM 8 #define DSAFV2_SBM_XGE_CHN 6 #define DSAFV2_SBM_PPE_CHN 1 -#define DASFV2_ROCEE_CRD_NUM 8 +#define DASFV2_ROCEE_CRD_NUM 1 #define DSAF_VOQ_NUM DSAF_NODE_NUM #define DSAF_INODE_NUM DSAF_NODE_NUM @@ -166,6 +166,9 @@ #define DSAF_INODE_GE_FC_EN_0_REG 0x1B00 #define DSAF_INODE_VC0_IN_PKT_NUM_0_REG 0x1B50 #define DSAF_INODE_VC1_IN_PKT_NUM_0_REG 0x1C00 +#define DSAF_INODE_IN_PRIO_PAUSE_BASE_REG 0x1C00 +#define DSAF_INODE_IN_PRIO_PAUSE_BASE_OFFSET 0x100 +#define DSAF_INODE_IN_PRIO_PAUSE_OFFSET 0x50 #define DSAF_SBM_CFG_REG_0_REG 0x2000 #define DSAF_SBM_BP_CFG_0_XGE_REG_0_REG 0x2004 @@ -175,7 +178,7 @@ #define DSAF_SBM_BP_CFG_2_XGE_REG_0_REG 0x200C #define DSAF_SBM_BP_CFG_2_PPE_REG_0_REG 0x230C #define DSAF_SBM_BP_CFG_2_ROCEE_REG_0_REG 0x260C -#define DSAFV2_SBM_BP_CFG_2_ROCEE_REG_0_REG 0x238C +#define DSAFV2_SBM_BP_CFG_2_ROCEE_REG_0_REG 0x238C #define DSAF_SBM_FREE_CNT_0_0_REG 0x2010 #define DSAF_SBM_FREE_CNT_1_0_REG 0x2014 #define DSAF_SBM_BP_CNT_0_0_REG 0x2018 @@ -232,6 +235,8 @@ #define DSAF_XOD_ROCEE_RCVIN0_CNT_0_REG 0x3074 #define DSAF_XOD_ROCEE_RCVIN1_CNT_0_REG 0x3078 #define DSAF_XOD_FIFO_STATUS_0_REG 0x307C +#define DSAF_XOD_XGE_PFC_PRIO_CNT_BASE_REG 0x3A00 +#define DSAF_XOD_XGE_PFC_PRIO_CNT_OFFSET 0x4 #define DSAF_VOQ_ECC_INVERT_EN_0_REG 0x4004 #define DSAF_VOQ_SRAM_PKT_NUM_0_REG 0x4008 @@ -791,6 +796,18 @@ #define DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_S 9 #define DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_M (((1ULL << 9) - 1) << 9) +#define DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_S 0 +#define DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_M (((1ULL << 8) - 1) << 0) +#define DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_S 8 +#define DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_M (((1ULL << 8) - 1) << 8) + +#define DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_S (0) +#define DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_M (((1ULL << 6) - 1) << 0) +#define DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_S (6) +#define DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_M (((1ULL << 6) - 1) << 6) +#define DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_S (12) +#define DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_M (((1ULL << 6) - 1) << 12) + #define DSAF_TBL_TCAM_ADDR_S 0 #define DSAF_TBL_TCAM_ADDR_M ((1ULL << 9) - 1) diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index ad742a6f5879..d7e1f8c7ae92 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -600,6 +600,7 @@ static int hns_nic_poll_rx_skb(struct hns_nic_ring_data *ring_data, ring->stats.sw_err_cnt++; return -ENOMEM; } + skb_reset_mac_header(skb); prefetchw(skb->data); length = le16_to_cpu(desc->rx.pkt_len); @@ -761,16 +762,16 @@ static int hns_nic_rx_poll_one(struct hns_nic_ring_data *ring_data, recv_pkts = 0, recv_bds = 0, clean_count = 0; recv: while (recv_pkts < budget && recv_bds < num) { - /* reuse or realloc buffers*/ + /* reuse or realloc buffers */ if (clean_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) { hns_nic_alloc_rx_buffers(ring_data, clean_count); clean_count = 0; } - /* poll one pkg*/ + /* poll one pkt */ err = hns_nic_poll_rx_skb(ring_data, &skb, &bnum); if (unlikely(!skb)) /* this fault cannot be repaired */ - break; + goto out; recv_bds += bnum; clean_count += bnum; @@ -796,6 +797,7 @@ recv: } } +out: /* make all data has been write before submit */ if (clean_count > 0) hns_nic_alloc_rx_buffers(ring_data, clean_count); @@ -990,8 +992,26 @@ static void hns_nic_adjust_link(struct net_device *ndev) { struct hns_nic_priv *priv = netdev_priv(ndev); struct hnae_handle *h = priv->ae_handle; + int state = 1; + + if (priv->phy) { + h->dev->ops->adjust_link(h, ndev->phydev->speed, + ndev->phydev->duplex); + state = priv->phy->link; + } + state = state && h->dev->ops->get_status(h); - h->dev->ops->adjust_link(h, ndev->phydev->speed, ndev->phydev->duplex); + if (state != priv->link) { + if (state) { + netif_carrier_on(ndev); + netif_tx_wake_all_queues(ndev); + netdev_info(ndev, "link up\n"); + } else { + netif_carrier_off(ndev); + netdev_info(ndev, "link down\n"); + } + priv->link = state; + } } /** @@ -1181,7 +1201,7 @@ static int hns_nic_net_up(struct net_device *ndev) { struct hns_nic_priv *priv = netdev_priv(ndev); struct hnae_handle *h = priv->ae_handle; - int i, j, k; + int i, j; int ret; ret = hns_nic_init_irq(priv); @@ -1196,9 +1216,6 @@ static int hns_nic_net_up(struct net_device *ndev) goto out_has_some_queues; } - for (k = 0; k < h->q_num; k++) - h->dev->ops->toggle_queue_status(h->qs[k], 1); - ret = h->dev->ops->set_mac_addr(h, ndev->dev_addr); if (ret) goto out_set_mac_addr_err; @@ -1218,8 +1235,6 @@ static int hns_nic_net_up(struct net_device *ndev) out_start_err: netif_stop_queue(ndev); out_set_mac_addr_err: - for (k = 0; k < h->q_num; k++) - h->dev->ops->toggle_queue_status(h->qs[k], 0); out_has_some_queues: for (j = i - 1; j >= 0; j--) hns_nic_ring_close(ndev, j); @@ -1426,7 +1441,6 @@ static int hns_nic_set_features(struct net_device *netdev, netdev_features_t features) { struct hns_nic_priv *priv = netdev_priv(netdev); - struct hnae_handle *h = priv->ae_handle; switch (priv->enet_ver) { case AE_VERSION_1: @@ -1439,11 +1453,9 @@ static int hns_nic_set_features(struct net_device *netdev, priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tso; /* The chip only support 7*4096 */ netif_set_gso_max_size(netdev, 7 * 4096); - h->dev->ops->set_tso_stats(h, 1); } else { priv->ops.fill_desc = fill_v2_desc; priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx; - h->dev->ops->set_tso_stats(h, 0); } break; } @@ -1576,27 +1588,14 @@ static void hns_nic_update_link_status(struct net_device *netdev) struct hns_nic_priv *priv = netdev_priv(netdev); struct hnae_handle *h = priv->ae_handle; - int state = 1; - if (priv->phy) { - if (!genphy_update_link(priv->phy)) - state = priv->phy->link; - else - state = 0; - } - state = state && h->dev->ops->get_status(h); + if (h->phy_dev) { + if (h->phy_if != PHY_INTERFACE_MODE_XGMII) + return; - if (state != priv->link) { - if (state) { - netif_carrier_on(netdev); - netif_tx_wake_all_queues(netdev); - netdev_info(netdev, "link up\n"); - } else { - netif_carrier_off(netdev); - netdev_info(netdev, "link down\n"); - } - priv->link = state; + (void)genphy_read_status(h->phy_dev); } + hns_nic_adjust_link(netdev); } /* for dumping key regs*/ @@ -1632,7 +1631,7 @@ static void hns_nic_dump(struct hns_nic_priv *priv) } } -/* for resetting suntask*/ +/* for resetting subtask */ static void hns_nic_reset_subtask(struct hns_nic_priv *priv) { enum hnae_port_type type = priv->ae_handle->port_type; @@ -1802,11 +1801,14 @@ static void hns_nic_set_priv_ops(struct net_device *netdev) priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tso; /* This chip only support 7*4096 */ netif_set_gso_max_size(netdev, 7 * 4096); - h->dev->ops->set_tso_stats(h, 1); } else { priv->ops.fill_desc = fill_v2_desc; priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx; } + /* enable tso when init + * control tso on/off through TSE bit in bd + */ + h->dev->ops->set_tso_stats(h, 1); } } @@ -1971,7 +1973,7 @@ static int hns_nic_dev_probe(struct platform_device *pdev) if (!dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64))) dev_dbg(dev, "set mask to 64bit\n"); else - dev_err(dev, "set mask to 32bit fail!\n"); + dev_err(dev, "set mask to 64bit fail!\n"); /* carrier off reporting is important to ethtool even BEFORE open */ netif_carrier_off(ndev); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index a809f52f6c16..ab33487a5321 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -49,7 +49,7 @@ static u32 hns_nic_get_link(struct net_device *net_dev) h = priv->ae_handle; if (priv->phy) { - if (!genphy_update_link(priv->phy)) + if (!genphy_read_status(priv->phy)) link_stat = priv->phy->link; else link_stat = 0; @@ -165,13 +165,21 @@ static int hns_nic_get_settings(struct net_device *net_dev, cmd->advertising |= ADVERTISED_10000baseKR_Full; } - if (h->port_type == HNAE_PORT_SERVICE) { + switch (h->media_type) { + case HNAE_MEDIA_TYPE_FIBER: cmd->port = PORT_FIBRE; - cmd->supported |= SUPPORTED_Pause; - } else { + break; + case HNAE_MEDIA_TYPE_COPPER: cmd->port = PORT_TP; + break; + case HNAE_MEDIA_TYPE_UNKNOWN: + default: + break; } + if (!(AE_IS_VER1(priv->enet_ver) && h->port_type == HNAE_PORT_DEBUG)) + cmd->supported |= SUPPORTED_Pause; + cmd->transceiver = XCVR_EXTERNAL; cmd->mdio_support = (ETH_MDIO_SUPPORTS_C45 | ETH_MDIO_SUPPORTS_C22); hns_get_mdix_mode(net_dev, cmd); @@ -242,6 +250,7 @@ static const char hns_nic_test_strs[][ETH_GSTRING_LEN] = { static int hns_nic_config_phy_loopback(struct phy_device *phy_dev, u8 en) { #define COPPER_CONTROL_REG 0 +#define PHY_POWER_DOWN BIT(11) #define PHY_LOOP_BACK BIT(14) u16 val = 0; @@ -252,33 +261,40 @@ static int hns_nic_config_phy_loopback(struct phy_device *phy_dev, u8 en) /* speed : 1000M */ phy_write(phy_dev, HNS_PHY_PAGE_REG, 2); phy_write(phy_dev, 21, 0x1046); + + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0); /* Force Master */ phy_write(phy_dev, 9, 0x1F00); + /* Soft-reset */ phy_write(phy_dev, 0, 0x9140); /* If autoneg disabled,two soft-reset operations */ phy_write(phy_dev, 0, 0x9140); - phy_write(phy_dev, 22, 0xFA); + + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0xFA); /* Default is 0x0400 */ phy_write(phy_dev, 1, 0x418); /* Force 1000M Link, Default is 0x0200 */ phy_write(phy_dev, 7, 0x20C); - phy_write(phy_dev, 22, 0); + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0); - /* Enable MAC loop-back */ + /* Enable PHY loop-back */ val = phy_read(phy_dev, COPPER_CONTROL_REG); val |= PHY_LOOP_BACK; + val &= ~PHY_POWER_DOWN; phy_write(phy_dev, COPPER_CONTROL_REG, val); } else { - phy_write(phy_dev, 22, 0xFA); + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0xFA); phy_write(phy_dev, 1, 0x400); phy_write(phy_dev, 7, 0x200); - phy_write(phy_dev, 22, 0); + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0); + phy_write(phy_dev, 9, 0xF00); val = phy_read(phy_dev, COPPER_CONTROL_REG); val &= ~PHY_LOOP_BACK; + val |= PHY_POWER_DOWN; phy_write(phy_dev, COPPER_CONTROL_REG, val); } return 0; @@ -339,28 +355,16 @@ static int __lb_up(struct net_device *ndev, hns_nic_net_reset(ndev); - if (priv->phy) { - phy_disconnect(priv->phy); - msleep(100); - - ret = hns_nic_init_phy(ndev, h); - if (ret) - return ret; - } - ret = __lb_setup(ndev, loop_mode); if (ret) return ret; - msleep(100); + msleep(200); ret = h->dev->ops->start ? h->dev->ops->start(h) : 0; if (ret) return ret; - if (priv->phy) - phy_start(priv->phy); - /* link adjust duplex*/ if (priv->ae_handle->phy_if != PHY_INTERFACE_MODE_XGMII) speed = 1000; @@ -561,9 +565,6 @@ static int __lb_down(struct net_device *ndev) __func__, ret); - if (priv->phy) - phy_stop(priv->phy); - if (h->dev->ops->stop) h->dev->ops->stop(h); @@ -758,6 +759,16 @@ static int hns_get_coalesce(struct net_device *net_dev, &ec->tx_max_coalesced_frames, &ec->rx_max_coalesced_frames); + ops->get_coalesce_range(priv->ae_handle, + &ec->tx_max_coalesced_frames_low, + &ec->rx_max_coalesced_frames_low, + &ec->tx_max_coalesced_frames_high, + &ec->rx_max_coalesced_frames_high, + &ec->tx_coalesce_usecs_low, + &ec->rx_coalesce_usecs_low, + &ec->tx_coalesce_usecs_high, + &ec->rx_coalesce_usecs_high); + return 0; } diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c index 761a32fceceb..33f4c483af0f 100644 --- a/drivers/net/ethernet/hisilicon/hns_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns_mdio.c @@ -37,9 +37,19 @@ #define MDIO_TIMEOUT 1000000 +struct hns_mdio_sc_reg { + u16 mdio_clk_en; + u16 mdio_clk_dis; + u16 mdio_reset_req; + u16 mdio_reset_dreq; + u16 mdio_clk_st; + u16 mdio_reset_st; +}; + struct hns_mdio_device { void *vbase; /* mdio reg base address */ struct regmap *subctrl_vbase; + struct hns_mdio_sc_reg sc_reg; }; /* mdio reg */ @@ -93,7 +103,6 @@ enum mdio_c45_op_seq { #define MDIO_SC_CLK_DIS 0x33C #define MDIO_SC_RESET_REQ 0xA38 #define MDIO_SC_RESET_DREQ 0xA3C -#define MDIO_SC_CTRL 0x2010 #define MDIO_SC_CLK_ST 0x531C #define MDIO_SC_RESET_ST 0x5A1C @@ -353,6 +362,7 @@ static int hns_mdio_read(struct mii_bus *bus, int phy_id, int regnum) static int hns_mdio_reset(struct mii_bus *bus) { struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv; + const struct hns_mdio_sc_reg *sc_reg; int ret; if (dev_of_node(bus->parent)) { @@ -361,9 +371,10 @@ static int hns_mdio_reset(struct mii_bus *bus) return -ENODEV; } + sc_reg = &mdio_dev->sc_reg; /* 1. reset req, and read reset st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_REQ, 0x1, - MDIO_SC_RESET_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_reset_req, + 0x1, sc_reg->mdio_reset_st, 0x1, MDIO_CHECK_SET_ST); if (ret) { dev_err(&bus->dev, "MDIO reset fail\n"); @@ -371,8 +382,8 @@ static int hns_mdio_reset(struct mii_bus *bus) } /* 2. dis clk, and read clk st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_DIS, - 0x1, MDIO_SC_CLK_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_clk_dis, + 0x1, sc_reg->mdio_clk_st, 0x1, MDIO_CHECK_CLR_ST); if (ret) { dev_err(&bus->dev, "MDIO dis clk fail\n"); @@ -380,8 +391,8 @@ static int hns_mdio_reset(struct mii_bus *bus) } /* 3. reset dreq, and read reset st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_DREQ, 0x1, - MDIO_SC_RESET_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_reset_dreq, + 0x1, sc_reg->mdio_reset_st, 0x1, MDIO_CHECK_CLR_ST); if (ret) { dev_err(&bus->dev, "MDIO dis clk fail\n"); @@ -389,8 +400,8 @@ static int hns_mdio_reset(struct mii_bus *bus) } /* 4. en clk, and read clk st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_EN, - 0x1, MDIO_SC_CLK_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_clk_en, + 0x1, sc_reg->mdio_clk_st, 0x1, MDIO_CHECK_SET_ST); if (ret) dev_err(&bus->dev, "MDIO en clk fail\n"); @@ -458,13 +469,54 @@ static int hns_mdio_probe(struct platform_device *pdev) snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s-%s", "Mii", dev_name(&pdev->dev)); if (dev_of_node(&pdev->dev)) { - mdio_dev->subctrl_vbase = syscon_node_to_regmap( - of_parse_phandle(pdev->dev.of_node, - "subctrl-vbase", 0)); - if (IS_ERR(mdio_dev->subctrl_vbase)) { - dev_warn(&pdev->dev, "no syscon hisilicon,peri-c-subctrl\n"); + struct of_phandle_args reg_args; + + ret = of_parse_phandle_with_fixed_args(pdev->dev.of_node, + "subctrl-vbase", + 4, + 0, + ®_args); + if (!ret) { + mdio_dev->subctrl_vbase = + syscon_node_to_regmap(reg_args.np); + if (IS_ERR(mdio_dev->subctrl_vbase)) { + dev_warn(&pdev->dev, "syscon_node_to_regmap error\n"); + mdio_dev->subctrl_vbase = NULL; + } else { + if (reg_args.args_count == 4) { + mdio_dev->sc_reg.mdio_clk_en = + (u16)reg_args.args[0]; + mdio_dev->sc_reg.mdio_clk_dis = + (u16)reg_args.args[0] + 4; + mdio_dev->sc_reg.mdio_reset_req = + (u16)reg_args.args[1]; + mdio_dev->sc_reg.mdio_reset_dreq = + (u16)reg_args.args[1] + 4; + mdio_dev->sc_reg.mdio_clk_st = + (u16)reg_args.args[2]; + mdio_dev->sc_reg.mdio_reset_st = + (u16)reg_args.args[3]; + } else { + /* for compatible */ + mdio_dev->sc_reg.mdio_clk_en = + MDIO_SC_CLK_EN; + mdio_dev->sc_reg.mdio_clk_dis = + MDIO_SC_CLK_DIS; + mdio_dev->sc_reg.mdio_reset_req = + MDIO_SC_RESET_REQ; + mdio_dev->sc_reg.mdio_reset_dreq = + MDIO_SC_RESET_DREQ; + mdio_dev->sc_reg.mdio_clk_st = + MDIO_SC_CLK_ST; + mdio_dev->sc_reg.mdio_reset_st = + MDIO_SC_RESET_ST; + } + } + } else { + dev_warn(&pdev->dev, "find syscon ret = %#x\n", ret); mdio_dev->subctrl_vbase = NULL; } + ret = of_mdiobus_register(new_bus, pdev->dev.of_node); } else if (is_acpi_node(pdev->dev.fwnode)) { /* Clear all the IRQ properties */ diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 864cb21351a4..ecdb6854a898 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -2121,7 +2121,7 @@ static void handle_error_info_rsp(union ibmvnic_crq *crq, struct ibmvnic_adapter *adapter) { struct device *dev = &adapter->vdev->dev; - struct ibmvnic_error_buff *error_buff; + struct ibmvnic_error_buff *error_buff, *tmp; unsigned long flags; bool found = false; int i; @@ -2133,7 +2133,7 @@ static void handle_error_info_rsp(union ibmvnic_crq *crq, } spin_lock_irqsave(&adapter->error_list_lock, flags); - list_for_each_entry(error_buff, &adapter->errors, list) + list_for_each_entry_safe(error_buff, tmp, &adapter->errors, list) if (error_buff->error_id == crq->request_error_rsp.error_id) { found = true; list_del(&error_buff->list); @@ -3141,14 +3141,14 @@ static void handle_request_ras_comp_num_rsp(union ibmvnic_crq *crq, static void ibmvnic_free_inflight(struct ibmvnic_adapter *adapter) { - struct ibmvnic_inflight_cmd *inflight_cmd; + struct ibmvnic_inflight_cmd *inflight_cmd, *tmp1; struct device *dev = &adapter->vdev->dev; - struct ibmvnic_error_buff *error_buff; + struct ibmvnic_error_buff *error_buff, *tmp2; unsigned long flags; unsigned long flags2; spin_lock_irqsave(&adapter->inflight_lock, flags); - list_for_each_entry(inflight_cmd, &adapter->inflight, list) { + list_for_each_entry_safe(inflight_cmd, tmp1, &adapter->inflight, list) { switch (inflight_cmd->crq.generic.cmd) { case LOGIN: dma_unmap_single(dev, adapter->login_buf_token, @@ -3165,8 +3165,8 @@ static void ibmvnic_free_inflight(struct ibmvnic_adapter *adapter) break; case REQUEST_ERROR_INFO: spin_lock_irqsave(&adapter->error_list_lock, flags2); - list_for_each_entry(error_buff, &adapter->errors, - list) { + list_for_each_entry_safe(error_buff, tmp2, + &adapter->errors, list) { dma_unmap_single(dev, error_buff->dma, error_buff->len, DMA_FROM_DEVICE); diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 714bd1014ddb..c0e17433f623 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -167,17 +167,6 @@ config IXGBE To compile this driver as a module, choose M here. The module will be called ixgbe. -config IXGBE_VXLAN - bool "Virtual eXtensible Local Area Network Support" - default n - depends on IXGBE && VXLAN && !(IXGBE=y && VXLAN=m) - ---help--- - This allows one to create VXLAN virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. VXLAN is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use Virtual eXtensible Local Area Network - (VXLAN) in the driver. - config IXGBE_HWMON bool "Intel(R) 10GbE PCI Express adapters HWMON support" default y @@ -236,27 +225,6 @@ config I40E To compile this driver as a module, choose M here. The module will be called i40e. -config I40E_VXLAN - bool "Virtual eXtensible Local Area Network Support" - default n - depends on I40E && VXLAN && !(I40E=y && VXLAN=m) - ---help--- - This allows one to create VXLAN virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. VXLAN is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use Virtual eXtensible Local Area Network - (VXLAN) in the driver. - -config I40E_GENEVE - bool "Generic Network Virtualization Encapsulation (GENEVE) Support" - depends on I40E && GENEVE && !(I40E=y && GENEVE=m) - default n - ---help--- - This allows one to create GENEVE virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. GENEVE is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use GENEVE in the driver. - config I40E_DCB bool "Data Center Bridging (DCB) Support" default n @@ -307,15 +275,4 @@ config FM10K To compile this driver as a module, choose M here. The module will be called fm10k. MSI-X interrupt support is required -config FM10K_VXLAN - bool "Virtual eXtensible Local Area Network Support" - default n - depends on FM10K && VXLAN && !(FM10K=y && VXLAN=m) - ---help--- - This allows one to create VXLAN virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. VXLAN is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use Virtual eXtensible Local Area Network - (VXLAN) in the driver. - endif # NET_VENDOR_INTEL diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 75e60897b7e7..8294c9a10bd2 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -154,6 +154,16 @@ void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val) writel(val, hw->hw_addr + reg); } +static bool e1000e_vlan_used(struct e1000_adapter *adapter) +{ + u16 vid; + + for_each_set_bit(vid, adapter->active_vlans, VLAN_N_VID) + return true; + + return false; +} + /** * e1000_regdump - register printout routine * @hw: pointer to the HW structure @@ -2789,7 +2799,7 @@ static void e1000e_vlan_filter_enable(struct e1000_adapter *adapter) } /** - * e1000e_vlan_strip_enable - helper to disable HW VLAN stripping + * e1000e_vlan_strip_disable - helper to disable HW VLAN stripping * @adapter: board private structure to initialize **/ static void e1000e_vlan_strip_disable(struct e1000_adapter *adapter) @@ -3443,7 +3453,8 @@ static void e1000e_set_rx_mode(struct net_device *netdev) ew32(RCTL, rctl); - if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX) + if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX || + e1000e_vlan_used(adapter)) e1000e_vlan_strip_enable(adapter); else e1000e_vlan_strip_disable(adapter); @@ -4352,7 +4363,8 @@ static cycle_t e1000e_cyclecounter_read(const struct cyclecounter *cc) time_delta = systim_next - systim; temp = time_delta; - rem = do_div(temp, incvalue); + /* VMWare users have seen incvalue of zero, don't div / 0 */ + rem = incvalue ? do_div(temp, incvalue) : (time_delta != 0); systim = systim_next; diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h index fcf106e545c5..e98b86bf0ca1 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k.h @@ -406,7 +406,7 @@ static inline u16 fm10k_desc_unused(struct fm10k_ring *ring) (&(((union fm10k_rx_desc *)((R)->desc))[i])) #define FM10K_MAX_TXD_PWR 14 -#define FM10K_MAX_DATA_PER_TXD BIT(FM10K_MAX_TXD_PWR) +#define FM10K_MAX_DATA_PER_TXD (1u << FM10K_MAX_TXD_PWR) /* Tx Descriptors needed, worst case */ #define TXD_USE_COUNT(S) DIV_ROUND_UP((S), FM10K_MAX_DATA_PER_TXD) diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c index 9c0d87503977..9b5195435c87 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c @@ -983,9 +983,10 @@ void fm10k_write_reta(struct fm10k_intfc *interface, const u32 *indir) /* generate a new table if we weren't given one */ for (j = 0; j < 4; j++) { if (indir) - n = indir[i + j]; + n = indir[4 * i + j]; else - n = ethtool_rxfh_indir_default(i + j, rss_i); + n = ethtool_rxfh_indir_default(4 * i + j, + rss_i); table[j] = n; } diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index 0e166e9c90c8..a9ccc1eb3ea4 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -56,7 +56,7 @@ static int __init fm10k_init_module(void) pr_info("%s\n", fm10k_copyright); /* create driver workqueue */ - fm10k_workqueue = create_workqueue("fm10k"); + fm10k_workqueue = alloc_workqueue("fm10k", WQ_MEM_RECLAIM, 0); fm10k_dbg_init(); @@ -77,7 +77,6 @@ static void __exit fm10k_exit_module(void) fm10k_dbg_exit(); /* destroy driver workqueue */ - flush_workqueue(fm10k_workqueue); destroy_workqueue(fm10k_workqueue); } module_exit(fm10k_exit_module); @@ -272,7 +271,7 @@ static bool fm10k_add_rx_frag(struct fm10k_rx_buffer *rx_buffer, #if (PAGE_SIZE < 8192) unsigned int truesize = FM10K_RX_BUFSZ; #else - unsigned int truesize = SKB_DATA_ALIGN(size); + unsigned int truesize = ALIGN(size, 512); #endif unsigned int pull_len; diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c index 2a08d3f5b6df..d00cb193da9a 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c @@ -20,9 +20,7 @@ #include "fm10k.h" #include <linux/vmalloc.h> -#ifdef CONFIG_FM10K_VXLAN -#include <net/vxlan.h> -#endif /* CONFIG_FM10K_VXLAN */ +#include <net/udp_tunnel.h> /** * fm10k_setup_tx_resources - allocate Tx resources (Descriptors) @@ -436,6 +434,7 @@ static void fm10k_restore_vxlan_port(struct fm10k_intfc *interface) * @netdev: network interface device structure * @sa_family: Address family of new port * @port: port number used for VXLAN + * @type: Enumerated value specifying udp encapsulation type * * This function is called when a new VXLAN interface has added a new port * number to the range that is currently in use for VXLAN. The new port @@ -444,18 +443,21 @@ static void fm10k_restore_vxlan_port(struct fm10k_intfc *interface) * is always used as the VXLAN port number for offloads. **/ static void fm10k_add_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) { + struct udp_tunnel_info *ti) +{ struct fm10k_intfc *interface = netdev_priv(dev); struct fm10k_vxlan_port *vxlan_port; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; /* only the PF supports configuring tunnels */ if (interface->hw.mac.type != fm10k_mac_pf) return; /* existing ports are pulled out so our new entry is always last */ fm10k_vxlan_port_for_each(vxlan_port, interface) { - if ((vxlan_port->port == port) && - (vxlan_port->sa_family == sa_family)) { + if ((vxlan_port->port == ti->port) && + (vxlan_port->sa_family == ti->sa_family)) { list_del(&vxlan_port->list); goto insert_tail; } @@ -465,8 +467,8 @@ static void fm10k_add_vxlan_port(struct net_device *dev, vxlan_port = kmalloc(sizeof(*vxlan_port), GFP_ATOMIC); if (!vxlan_port) return; - vxlan_port->port = port; - vxlan_port->sa_family = sa_family; + vxlan_port->port = ti->port; + vxlan_port->sa_family = ti->sa_family; insert_tail: /* add new port value to list */ @@ -480,6 +482,7 @@ insert_tail: * @netdev: network interface device structure * @sa_family: Address family of freed port * @port: port number used for VXLAN + * @type: Enumerated value specifying udp encapsulation type * * This function is called when a new VXLAN interface has freed a port * number from the range that is currently in use for VXLAN. The freed @@ -487,17 +490,20 @@ insert_tail: * the port number for offloads. **/ static void fm10k_del_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) { + struct udp_tunnel_info *ti) +{ struct fm10k_intfc *interface = netdev_priv(dev); struct fm10k_vxlan_port *vxlan_port; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; if (interface->hw.mac.type != fm10k_mac_pf) return; /* find the port in the list and free it */ fm10k_vxlan_port_for_each(vxlan_port, interface) { - if ((vxlan_port->port == port) && - (vxlan_port->sa_family == sa_family)) { + if ((vxlan_port->port == ti->port) && + (vxlan_port->sa_family == ti->sa_family)) { list_del(&vxlan_port->list); kfree(vxlan_port); break; @@ -553,10 +559,8 @@ int fm10k_open(struct net_device *netdev) if (err) goto err_set_queues; -#ifdef CONFIG_FM10K_VXLAN /* update VXLAN port configuration */ - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); fm10k_up(interface); @@ -1375,8 +1379,8 @@ static const struct net_device_ops fm10k_netdev_ops = { .ndo_set_vf_vlan = fm10k_ndo_set_vf_vlan, .ndo_set_vf_rate = fm10k_ndo_set_vf_bw, .ndo_get_vf_config = fm10k_ndo_get_vf_config, - .ndo_add_vxlan_port = fm10k_add_vxlan_port, - .ndo_del_vxlan_port = fm10k_del_vxlan_port, + .ndo_udp_tunnel_add = fm10k_add_vxlan_port, + .ndo_udp_tunnel_del = fm10k_del_vxlan_port, .ndo_dfwd_add_station = fm10k_dfwd_add_station, .ndo_dfwd_del_station = fm10k_dfwd_del_station, #ifdef CONFIG_NET_POLL_CONTROLLER diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index 9c44739da5e2..e83fc8afb30f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -283,6 +283,7 @@ struct i40e_pf { #endif /* I40E_FCOE */ u16 num_lan_qps; /* num lan queues this PF has set up */ u16 num_lan_msix; /* num queue vectors for the base PF vsi */ + u16 num_fdsb_msix; /* num queue vectors for sideband Fdir */ u16 num_iwarp_msix; /* num of iwarp vectors for this PF */ int iwarp_base_vector; int queues_left; /* queues left unclaimed */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index 422b41d61c9a..e447dc435464 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -1967,6 +1967,62 @@ aq_add_vsi_exit: } /** + * i40e_aq_set_default_vsi + * @hw: pointer to the hw struct + * @seid: vsi number + * @cmd_details: pointer to command details structure or NULL + **/ +i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw, + u16 seid, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_set_vsi_promiscuous_modes *cmd = + (struct i40e_aqc_set_vsi_promiscuous_modes *) + &desc.params.raw; + i40e_status status; + + i40e_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_set_vsi_promiscuous_modes); + + cmd->promiscuous_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT); + cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT); + cmd->seid = cpu_to_le16(seid); + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + +/** + * i40e_aq_clear_default_vsi + * @hw: pointer to the hw struct + * @seid: vsi number + * @cmd_details: pointer to command details structure or NULL + **/ +i40e_status i40e_aq_clear_default_vsi(struct i40e_hw *hw, + u16 seid, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_set_vsi_promiscuous_modes *cmd = + (struct i40e_aqc_set_vsi_promiscuous_modes *) + &desc.params.raw; + i40e_status status; + + i40e_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_set_vsi_promiscuous_modes); + + cmd->promiscuous_flags = cpu_to_le16(0); + cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT); + cmd->seid = cpu_to_le16(seid); + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + +/** * i40e_aq_set_vsi_unicast_promiscuous * @hw: pointer to the hw struct * @seid: vsi number diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 5e8d84ff7d5f..4962e855fbd3 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -313,8 +313,7 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, *advertising |= ADVERTISED_Autoneg | ADVERTISED_40000baseCR4_Full; } - if ((phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) && - !(phy_types & I40E_CAP_PHY_TYPE_1000BASE_T)) { + if (phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) { *supported |= SUPPORTED_Autoneg | SUPPORTED_100baseT_Full; *advertising |= ADVERTISED_Autoneg | @@ -663,6 +662,7 @@ static int i40e_set_settings(struct net_device *netdev, if (hw->phy.media_type != I40E_MEDIA_TYPE_BASET && hw->phy.media_type != I40E_MEDIA_TYPE_FIBER && hw->phy.media_type != I40E_MEDIA_TYPE_BACKPLANE && + hw->phy.media_type != I40E_MEDIA_TYPE_DA && hw->phy.link_info.link_info & I40E_AQ_LINK_UP) return -EOPNOTSUPP; diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 5ea22008d721..2b1140563a64 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -31,12 +31,7 @@ /* Local includes */ #include "i40e.h" #include "i40e_diag.h" -#if IS_ENABLED(CONFIG_VXLAN) -#include <net/vxlan.h> -#endif -#if IS_ENABLED(CONFIG_GENEVE) -#include <net/geneve.h> -#endif +#include <net/udp_tunnel.h> const char i40e_driver_name[] = "i40e"; static const char i40e_driver_string[] = @@ -45,8 +40,8 @@ static const char i40e_driver_string[] = #define DRV_KERN "-k" #define DRV_VERSION_MAJOR 1 -#define DRV_VERSION_MINOR 5 -#define DRV_VERSION_BUILD 16 +#define DRV_VERSION_MINOR 6 +#define DRV_VERSION_BUILD 4 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) DRV_KERN @@ -1584,14 +1579,8 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi, vsi->tc_config.numtc = numtc; vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1; /* Number of queues per enabled TC */ - /* In MFP case we can have a much lower count of MSIx - * vectors available and so we need to lower the used - * q count. - */ - if (pf->flags & I40E_FLAG_MSIX_ENABLED) - qcount = min_t(int, vsi->alloc_queue_pairs, pf->num_lan_msix); - else - qcount = vsi->alloc_queue_pairs; + qcount = vsi->alloc_queue_pairs; + num_tc_qps = qcount / numtc; num_tc_qps = min_t(int, num_tc_qps, i40e_pf_get_max_q_per_tc(pf)); @@ -1845,8 +1834,10 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) { struct list_head tmp_del_list, tmp_add_list; struct i40e_mac_filter *f, *ftmp, *fclone; + struct i40e_hw *hw = &vsi->back->hw; bool promisc_forced_on = false; bool add_happened = false; + char vsi_name[16] = "PF"; int filter_list_len = 0; u32 changed_flags = 0; i40e_status aq_ret = 0; @@ -1874,6 +1865,11 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) INIT_LIST_HEAD(&tmp_del_list); INIT_LIST_HEAD(&tmp_add_list); + if (vsi->type == I40E_VSI_SRIOV) + snprintf(vsi_name, sizeof(vsi_name) - 1, "VF %d", vsi->vf_id); + else if (vsi->type != I40E_VSI_MAIN) + snprintf(vsi_name, sizeof(vsi_name) - 1, "vsi %d", vsi->seid); + if (vsi->flags & I40E_VSI_FLAG_FILTER_CHANGED) { vsi->flags &= ~I40E_VSI_FLAG_FILTER_CHANGED; @@ -1925,7 +1921,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (!list_empty(&tmp_del_list)) { int del_list_size; - filter_list_len = pf->hw.aq.asq_buf_size / + filter_list_len = hw->aq.asq_buf_size / sizeof(struct i40e_aqc_remove_macvlan_element_data); del_list_size = filter_list_len * sizeof(struct i40e_aqc_remove_macvlan_element_data); @@ -1957,21 +1953,21 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* flush a full buffer */ if (num_del == filter_list_len) { - aq_ret = i40e_aq_remove_macvlan(&pf->hw, - vsi->seid, - del_list, - num_del, - NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_ret = + i40e_aq_remove_macvlan(hw, vsi->seid, + del_list, + num_del, NULL); + aq_err = hw->aq.asq_last_status; num_del = 0; memset(del_list, 0, del_list_size); if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) { retval = -EIO; dev_err(&pf->pdev->dev, - "ignoring delete macvlan error, err %s, aq_err %s while flushing a full buffer\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, aq_err)); + "ignoring delete macvlan error on %s, err %s, aq_err %s while flushing a full buffer\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, aq_err)); } } /* Release memory for MAC filter entries which were @@ -1982,17 +1978,17 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } if (num_del) { - aq_ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid, - del_list, num_del, - NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_ret = i40e_aq_remove_macvlan(hw, vsi->seid, del_list, + num_del, NULL); + aq_err = hw->aq.asq_last_status; num_del = 0; if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) dev_info(&pf->pdev->dev, - "ignoring delete macvlan error, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, aq_err)); + "ignoring delete macvlan error on %s, err %s aq_err %s\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, aq_err)); } kfree(del_list); @@ -2003,7 +1999,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) int add_list_size; /* do all the adds now */ - filter_list_len = pf->hw.aq.asq_buf_size / + filter_list_len = hw->aq.asq_buf_size / sizeof(struct i40e_aqc_add_macvlan_element_data), add_list_size = filter_list_len * sizeof(struct i40e_aqc_add_macvlan_element_data); @@ -2038,10 +2034,10 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* flush a full buffer */ if (num_add == filter_list_len) { - aq_ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid, + aq_ret = i40e_aq_add_macvlan(hw, vsi->seid, add_list, num_add, NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_err = hw->aq.asq_last_status; num_add = 0; if (aq_ret) @@ -2056,9 +2052,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } if (num_add) { - aq_ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid, + aq_ret = i40e_aq_add_macvlan(hw, vsi->seid, add_list, num_add, NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_err = hw->aq.asq_last_status; num_add = 0; } kfree(add_list); @@ -2067,16 +2063,18 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (add_happened && aq_ret && aq_err != I40E_AQ_RC_EINVAL) { retval = i40e_aq_rc_to_posix(aq_ret, aq_err); dev_info(&pf->pdev->dev, - "add filter failed, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, aq_err)); - if ((pf->hw.aq.asq_last_status == I40E_AQ_RC_ENOSPC) && + "add filter failed on %s, err %s aq_err %s\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, aq_err)); + if ((hw->aq.asq_last_status == I40E_AQ_RC_ENOSPC) && !test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state)) { promisc_forced_on = true; set_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state); - dev_info(&pf->pdev->dev, "promiscuous mode forced on\n"); + dev_info(&pf->pdev->dev, "promiscuous mode forced on %s\n", + vsi_name); } } } @@ -2098,12 +2096,12 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) NULL); if (aq_ret) { retval = i40e_aq_rc_to_posix(aq_ret, - pf->hw.aq.asq_last_status); + hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set multi promisc failed, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, - pf->hw.aq.asq_last_status)); + "set multi promisc failed on %s, err %s aq_err %s\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, hw->aq.asq_last_status)); } } if ((changed_flags & IFF_PROMISC) || promisc_forced_on) { @@ -2122,33 +2120,58 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) */ if (pf->cur_promisc != cur_promisc) { pf->cur_promisc = cur_promisc; - set_bit(__I40E_PF_RESET_REQUESTED, &pf->state); + if (cur_promisc) + aq_ret = + i40e_aq_set_default_vsi(hw, + vsi->seid, + NULL); + else + aq_ret = + i40e_aq_clear_default_vsi(hw, + vsi->seid, + NULL); + if (aq_ret) { + retval = i40e_aq_rc_to_posix(aq_ret, + hw->aq.asq_last_status); + dev_info(&pf->pdev->dev, + "Set default VSI failed on %s, err %s, aq_err %s\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); + } } } else { aq_ret = i40e_aq_set_vsi_unicast_promiscuous( - &vsi->back->hw, + hw, vsi->seid, cur_promisc, NULL, true); if (aq_ret) { retval = i40e_aq_rc_to_posix(aq_ret, - pf->hw.aq.asq_last_status); + hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set unicast promisc failed, err %d, aq_err %d\n", - aq_ret, pf->hw.aq.asq_last_status); + "set unicast promisc failed on %s, err %s, aq_err %s\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); } aq_ret = i40e_aq_set_vsi_multicast_promiscuous( - &vsi->back->hw, + hw, vsi->seid, cur_promisc, NULL); if (aq_ret) { retval = i40e_aq_rc_to_posix(aq_ret, - pf->hw.aq.asq_last_status); + hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set multicast promisc failed, err %d, aq_err %d\n", - aq_ret, pf->hw.aq.asq_last_status); + "set multicast promisc failed on %s, err %s, aq_err %s\n", + vsi_name, + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); } } aq_ret = i40e_aq_set_vsi_broadcast(&vsi->back->hw, @@ -2159,9 +2182,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) pf->hw.aq.asq_last_status); dev_info(&pf->pdev->dev, "set brdcast promisc failed, err %s, aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, - pf->hw.aq.asq_last_status)); + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); } } out: @@ -3952,6 +3975,7 @@ static void i40e_vsi_free_irq(struct i40e_vsi *vsi) /* clear the affinity_mask in the IRQ descriptor */ irq_set_affinity_hint(pf->msix_entries[vector].vector, NULL); + synchronize_irq(pf->msix_entries[vector].vector); free_irq(pf->msix_entries[vector].vector, vsi->q_vectors[i]); @@ -4958,7 +4982,6 @@ static void i40e_dcb_reconfigure(struct i40e_pf *pf) if (pf->vsi[v]->netdev) i40e_dcbnl_set_all(pf->vsi[v]); } - i40e_notify_client_of_l2_param_changes(pf->vsi[v]); } } @@ -5183,12 +5206,6 @@ static void i40e_vsi_reinit_locked(struct i40e_vsi *vsi) usleep_range(1000, 2000); i40e_down(vsi); - /* Give a VF some time to respond to the reset. The - * two second wait is based upon the watchdog cycle in - * the VF driver. - */ - if (vsi->type == I40E_VSI_SRIOV) - msleep(2000); i40e_up(vsi); clear_bit(__I40E_CONFIG_BUSY, &pf->state); } @@ -5231,6 +5248,9 @@ void i40e_down(struct i40e_vsi *vsi) i40e_clean_tx_ring(vsi->tx_rings[i]); i40e_clean_rx_ring(vsi->rx_rings[i]); } + + i40e_notify_client_of_netdev_close(vsi, false); + } /** @@ -5342,14 +5362,7 @@ int i40e_open(struct net_device *netdev) TCP_FLAG_CWR) >> 16); wr32(&pf->hw, I40E_GLLAN_TSOMSK_L, be32_to_cpu(TCP_FLAG_CWR) >> 16); -#ifdef CONFIG_I40E_VXLAN - vxlan_get_rx_port(netdev); -#endif -#ifdef CONFIG_I40E_GENEVE - if (pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE) - geneve_get_rx_port(netdev); -#endif - + udp_tunnel_get_rx_info(netdev); i40e_notify_client_of_netdev_open(vsi); return 0; @@ -5716,6 +5729,8 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf, i40e_service_event_schedule(pf); } else { i40e_pf_unquiesce_all_vsi(pf); + /* Notify the client for the DCB changes */ + i40e_notify_client_of_l2_param_changes(pf->vsi[pf->lan_vsi]); } exit: @@ -5940,7 +5955,6 @@ static void i40e_fdir_flush_and_replay(struct i40e_pf *pf) if (I40E_DEBUG_FD & pf->hw.debug_mask) dev_info(&pf->pdev->dev, "FD Filter table flushed and FD-SB replayed.\n"); } - } /** @@ -7057,7 +7071,6 @@ static void i40e_handle_mdd_event(struct i40e_pf *pf) **/ static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf) { -#if IS_ENABLED(CONFIG_VXLAN) || IS_ENABLED(CONFIG_GENEVE) struct i40e_hw *hw = &pf->hw; i40e_status ret; __be16 port; @@ -7092,7 +7105,6 @@ static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf) } } } -#endif } /** @@ -7174,7 +7186,7 @@ static int i40e_set_num_rings_in_vsi(struct i40e_vsi *vsi) vsi->alloc_queue_pairs = 1; vsi->num_desc = ALIGN(I40E_FDIR_RING_COUNT, I40E_REQ_DESCRIPTOR_MULTIPLE); - vsi->num_q_vectors = 1; + vsi->num_q_vectors = pf->num_fdsb_msix; break; case I40E_VSI_VMDQ2: @@ -7558,9 +7570,11 @@ static int i40e_init_msix(struct i40e_pf *pf) /* reserve one vector for sideband flow director */ if (pf->flags & I40E_FLAG_FD_SB_ENABLED) { if (vectors_left) { + pf->num_fdsb_msix = 1; v_budget++; vectors_left--; } else { + pf->num_fdsb_msix = 0; pf->flags &= ~I40E_FLAG_FD_SB_ENABLED; } } @@ -8579,7 +8593,9 @@ bool i40e_set_ntuple(struct i40e_pf *pf, netdev_features_t features) /* Enable filters and mark for reset */ if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED)) need_reset = true; - pf->flags |= I40E_FLAG_FD_SB_ENABLED; + /* enable FD_SB only if there is MSI-X vector */ + if (pf->num_fdsb_msix > 0) + pf->flags |= I40E_FLAG_FD_SB_ENABLED; } else { /* turn off filters, mark for reset and clear SW filter list */ if (pf->flags & I40E_FLAG_FD_SB_ENABLED) { @@ -8628,7 +8644,6 @@ static int i40e_set_features(struct net_device *netdev, return 0; } -#if IS_ENABLED(CONFIG_VXLAN) || IS_ENABLED(CONFIG_GENEVE) /** * i40e_get_udp_port_idx - Lookup a possibly offloaded for Rx UDP port * @pf: board private structure @@ -8648,21 +8663,18 @@ static u8 i40e_get_udp_port_idx(struct i40e_pf *pf, __be16 port) return i; } -#endif - -#if IS_ENABLED(CONFIG_VXLAN) /** - * i40e_add_vxlan_port - Get notifications about VXLAN ports that come up + * i40e_udp_tunnel_add - Get notifications about UDP tunnel ports that come up * @netdev: This physical port's netdev - * @sa_family: Socket Family that VXLAN is notifying us about - * @port: New UDP port number that VXLAN started listening to + * @ti: Tunnel endpoint information **/ -static void i40e_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void i40e_udp_tunnel_add(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_vsi *vsi = np->vsi; struct i40e_pf *pf = vsi->back; + __be16 port = ti->port; u8 next_idx; u8 idx; @@ -8670,7 +8682,7 @@ static void i40e_add_vxlan_port(struct net_device *netdev, /* Check if port already exists */ if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "vxlan port %d already offloaded\n", + netdev_info(netdev, "port %d already offloaded\n", ntohs(port)); return; } @@ -8679,131 +8691,75 @@ static void i40e_add_vxlan_port(struct net_device *netdev, next_idx = i40e_get_udp_port_idx(pf, 0); if (next_idx == I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "maximum number of vxlan UDP ports reached, not adding port %d\n", - ntohs(port)); - return; - } - - /* New port: add it and mark its index in the bitmap */ - pf->udp_ports[next_idx].index = port; - pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_VXLAN; - pf->pending_udp_bitmap |= BIT_ULL(next_idx); - pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; -} - -/** - * i40e_del_vxlan_port - Get notifications about VXLAN ports that go away - * @netdev: This physical port's netdev - * @sa_family: Socket Family that VXLAN is notifying us about - * @port: UDP port number that VXLAN stopped listening to - **/ -static void i40e_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct i40e_netdev_priv *np = netdev_priv(netdev); - struct i40e_vsi *vsi = np->vsi; - struct i40e_pf *pf = vsi->back; - u8 idx; - - idx = i40e_get_udp_port_idx(pf, port); - - /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - /* if port exists, set it to 0 (mark for deletion) - * and make it pending - */ - pf->udp_ports[idx].index = 0; - pf->pending_udp_bitmap |= BIT_ULL(idx); - pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; - } else { - netdev_warn(netdev, "vxlan port %d was not found, not deleting\n", - ntohs(port)); - } -} -#endif - -#if IS_ENABLED(CONFIG_GENEVE) -/** - * i40e_add_geneve_port - Get notifications about GENEVE ports that come up - * @netdev: This physical port's netdev - * @sa_family: Socket Family that GENEVE is notifying us about - * @port: New UDP port number that GENEVE started listening to - **/ -static void i40e_add_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct i40e_netdev_priv *np = netdev_priv(netdev); - struct i40e_vsi *vsi = np->vsi; - struct i40e_pf *pf = vsi->back; - u8 next_idx; - u8 idx; - - if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)) - return; - - idx = i40e_get_udp_port_idx(pf, port); - - /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "udp port %d already offloaded\n", + netdev_info(netdev, "maximum number of offloaded UDP ports reached, not adding port %d\n", ntohs(port)); return; } - /* Now check if there is space to add the new port */ - next_idx = i40e_get_udp_port_idx(pf, 0); - - if (next_idx == I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "maximum number of UDP ports reached, not adding port %d\n", - ntohs(port)); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_VXLAN; + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)) + return; + pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_NGE; + break; + default: return; } /* New port: add it and mark its index in the bitmap */ pf->udp_ports[next_idx].index = port; - pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_NGE; pf->pending_udp_bitmap |= BIT_ULL(next_idx); pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; - - dev_info(&pf->pdev->dev, "adding geneve port %d\n", ntohs(port)); } /** - * i40e_del_geneve_port - Get notifications about GENEVE ports that go away + * i40e_udp_tunnel_del - Get notifications about UDP tunnel ports that go away * @netdev: This physical port's netdev - * @sa_family: Socket Family that GENEVE is notifying us about - * @port: UDP port number that GENEVE stopped listening to + * @ti: Tunnel endpoint information **/ -static void i40e_del_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void i40e_udp_tunnel_del(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_vsi *vsi = np->vsi; struct i40e_pf *pf = vsi->back; + __be16 port = ti->port; u8 idx; - if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)) - return; - idx = i40e_get_udp_port_idx(pf, port); /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - /* if port exists, set it to 0 (mark for deletion) - * and make it pending - */ - pf->udp_ports[idx].index = 0; - pf->pending_udp_bitmap |= BIT_ULL(idx); - pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; + if (idx >= I40E_MAX_PF_UDP_OFFLOAD_PORTS) + goto not_found; - dev_info(&pf->pdev->dev, "deleting geneve port %d\n", - ntohs(port)); - } else { - netdev_warn(netdev, "geneve port %d was not found, not deleting\n", - ntohs(port)); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (pf->udp_ports[idx].type != I40E_AQC_TUNNEL_TYPE_VXLAN) + goto not_found; + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (pf->udp_ports[idx].type != I40E_AQC_TUNNEL_TYPE_NGE) + goto not_found; + break; + default: + goto not_found; } + + /* if port exists, set it to 0 (mark for deletion) + * and make it pending + */ + pf->udp_ports[idx].index = 0; + pf->pending_udp_bitmap |= BIT_ULL(idx); + pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; + + return; +not_found: + netdev_warn(netdev, "UDP port %d was not found, not deleting\n", + ntohs(port)); } -#endif static int i40e_get_phys_port_id(struct net_device *netdev, struct netdev_phys_item_id *ppid) @@ -9033,14 +8989,8 @@ static const struct net_device_ops i40e_netdev_ops = { .ndo_set_vf_link_state = i40e_ndo_set_vf_link_state, .ndo_set_vf_spoofchk = i40e_ndo_set_vf_spoofchk, .ndo_set_vf_trust = i40e_ndo_set_vf_trust, -#if IS_ENABLED(CONFIG_VXLAN) - .ndo_add_vxlan_port = i40e_add_vxlan_port, - .ndo_del_vxlan_port = i40e_del_vxlan_port, -#endif -#if IS_ENABLED(CONFIG_GENEVE) - .ndo_add_geneve_port = i40e_add_geneve_port, - .ndo_del_geneve_port = i40e_del_geneve_port, -#endif + .ndo_udp_tunnel_add = i40e_udp_tunnel_add, + .ndo_udp_tunnel_del = i40e_udp_tunnel_del, .ndo_get_phys_port_id = i40e_get_phys_port_id, .ndo_fdb_add = i40e_ndo_fdb_add, .ndo_features_check = i40e_features_check, @@ -10133,14 +10083,14 @@ void i40e_veb_release(struct i40e_veb *veb) static int i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi) { struct i40e_pf *pf = veb->pf; - bool is_default = veb->pf->cur_promisc; bool enable_stats = !!(pf->flags & I40E_FLAG_VEB_STATS_ENABLED); int ret; - /* get a VEB from the hardware */ ret = i40e_aq_add_veb(&pf->hw, veb->uplink_seid, vsi->seid, - veb->enabled_tc, is_default, + veb->enabled_tc, false, &veb->seid, enable_stats, NULL); + + /* get a VEB from the hardware */ if (ret) { dev_info(&pf->pdev->dev, "couldn't add VEB, err %s aq_err %s\n", @@ -10689,12 +10639,8 @@ static void i40e_print_features(struct i40e_pf *pf) } if (pf->flags & I40E_FLAG_DCB_CAPABLE) i += snprintf(&buf[i], REMAIN(i), " DCB"); -#if IS_ENABLED(CONFIG_VXLAN) i += snprintf(&buf[i], REMAIN(i), " VxLAN"); -#endif -#if IS_ENABLED(CONFIG_GENEVE) i += snprintf(&buf[i], REMAIN(i), " Geneve"); -#endif if (pf->flags & I40E_FLAG_PTP) i += snprintf(&buf[i], REMAIN(i), " PTP"); #ifdef I40E_FCOE @@ -11525,6 +11471,7 @@ static int i40e_suspend(struct pci_dev *pdev, pm_message_t state) { struct i40e_pf *pf = pci_get_drvdata(pdev); struct i40e_hw *hw = &pf->hw; + int retval = 0; set_bit(__I40E_SUSPENDED, &pf->state); set_bit(__I40E_DOWN, &pf->state); @@ -11536,10 +11483,16 @@ static int i40e_suspend(struct pci_dev *pdev, pm_message_t state) wr32(hw, I40E_PFPM_APM, (pf->wol_en ? I40E_PFPM_APM_APME_MASK : 0)); wr32(hw, I40E_PFPM_WUFC, (pf->wol_en ? I40E_PFPM_WUFC_MAG_MASK : 0)); + i40e_stop_misc_vector(pf); + + retval = pci_save_state(pdev); + if (retval) + return retval; + pci_wake_from_d3(pdev, pf->wol_en); pci_set_power_state(pdev, PCI_D3hot); - return 0; + return retval; } /** diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h index 80403c6ee7f0..4660c5abc855 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h +++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h @@ -98,6 +98,8 @@ i40e_status i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags, struct i40e_asq_cmd_details *cmd_details); i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw, u16 vsi_id, struct i40e_asq_cmd_details *cmd_details); +i40e_status i40e_aq_clear_default_vsi(struct i40e_hw *hw, u16 vsi_id, + struct i40e_asq_cmd_details *cmd_details); enum i40e_status_code i40e_aq_get_phy_capabilities(struct i40e_hw *hw, bool qualified_modules, bool report_init, struct i40e_aq_get_phy_abilities_resp *abilities, diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 1fcafcfa8f14..6fcbf764f32b 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -665,6 +665,8 @@ static int i40e_alloc_vsi_res(struct i40e_vf *vf, enum i40e_vsi_type type) goto error_alloc_vsi_res; } if (type == I40E_VSI_SRIOV) { + u64 hena = i40e_pf_get_default_rss_hena(pf); + vf->lan_vsi_idx = vsi->idx; vf->lan_vsi_id = vsi->id; /* If the port VLAN has been configured and then the @@ -687,6 +689,10 @@ static int i40e_alloc_vsi_res(struct i40e_vf *vf, enum i40e_vsi_type type) vf->default_lan_addr.addr, vf->vf_id); } spin_unlock_bh(&vsi->mac_filter_list_lock); + i40e_write_rx_ctl(&pf->hw, I40E_VFQF_HENA1(0, vf->vf_id), + (u32)hena); + i40e_write_rx_ctl(&pf->hw, I40E_VFQF_HENA1(1, vf->vf_id), + (u32)(hena >> 32)); } /* program mac filter */ diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index 16c552952860..eac057b88055 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -37,8 +37,8 @@ static const char i40evf_driver_string[] = #define DRV_KERN "-k" #define DRV_VERSION_MAJOR 1 -#define DRV_VERSION_MINOR 5 -#define DRV_VERSION_BUILD 10 +#define DRV_VERSION_MINOR 6 +#define DRV_VERSION_BUILD 4 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) \ @@ -825,7 +825,7 @@ i40evf_mac_filter *i40evf_add_filter(struct i40evf_adapter *adapter, ether_addr_copy(f->macaddr, macaddr); - list_add(&f->list, &adapter->mac_filter_list); + list_add_tail(&f->list, &adapter->mac_filter_list); f->add = true; adapter->aq_required |= I40EVF_FLAG_AQ_ADD_MAC_FILTER; } diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c index f13445691507..d76c221d4c8a 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c @@ -434,6 +434,8 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter) ether_addr_copy(veal->list[i].addr, f->macaddr); i++; f->add = false; + if (i == count) + break; } } if (!more) @@ -497,6 +499,8 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter) i++; list_del(&f->list); kfree(f); + if (i == count) + break; } } if (!more) @@ -560,6 +564,8 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter) vvfl->vlan_id[i] = f->vlan; i++; f->add = false; + if (i == count) + break; } } if (!more) @@ -623,6 +629,8 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter) i++; list_del(&f->list); kfree(f); + if (i == count) + break; } } if (!more) diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h index b9609afa5ca3..5387b3a96489 100644 --- a/drivers/net/ethernet/intel/igb/igb.h +++ b/drivers/net/ethernet/intel/igb/igb.h @@ -445,6 +445,7 @@ struct igb_adapter { unsigned long ptp_tx_start; unsigned long last_rx_ptp_check; unsigned long last_rx_timestamp; + unsigned int ptp_flags; spinlock_t tmreg_lock; struct cyclecounter cc; struct timecounter tc; @@ -474,12 +475,15 @@ struct igb_adapter { u16 eee_advert; }; +/* flags controlling PTP/1588 function */ +#define IGB_PTP_ENABLED BIT(0) +#define IGB_PTP_OVERFLOW_CHECK BIT(1) + #define IGB_FLAG_HAS_MSI BIT(0) #define IGB_FLAG_DCA_ENABLED BIT(1) #define IGB_FLAG_QUAD_PORT_A BIT(2) #define IGB_FLAG_QUEUE_PAIRS BIT(3) #define IGB_FLAG_DMAC BIT(4) -#define IGB_FLAG_PTP BIT(5) #define IGB_FLAG_RSS_FIELD_IPV4_UDP BIT(6) #define IGB_FLAG_RSS_FIELD_IPV6_UDP BIT(7) #define IGB_FLAG_WOL_SUPPORTED BIT(8) @@ -546,6 +550,7 @@ void igb_set_fw_version(struct igb_adapter *); void igb_ptp_init(struct igb_adapter *adapter); void igb_ptp_stop(struct igb_adapter *adapter); void igb_ptp_reset(struct igb_adapter *adapter); +void igb_ptp_suspend(struct igb_adapter *adapter); void igb_ptp_rx_hang(struct igb_adapter *adapter); void igb_ptp_rx_rgtstamp(struct igb_q_vector *q_vector, struct sk_buff *skb); void igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, unsigned char *va, diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index ef3d642f5ff2..9bcba42abb91 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -2027,7 +2027,8 @@ void igb_reset(struct igb_adapter *adapter) wr32(E1000_VET, ETHERNET_IEEE_VLAN_TYPE); /* Re-enable PTP, where applicable. */ - igb_ptp_reset(adapter); + if (adapter->ptp_flags & IGB_PTP_ENABLED) + igb_ptp_reset(adapter); igb_get_phy_info(hw); } @@ -6855,12 +6856,12 @@ static bool igb_can_reuse_rx_page(struct igb_rx_buffer *rx_buffer, **/ static bool igb_add_rx_frag(struct igb_ring *rx_ring, struct igb_rx_buffer *rx_buffer, + unsigned int size, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { struct page *page = rx_buffer->page; unsigned char *va = page_address(page) + rx_buffer->page_offset; - unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); #if (PAGE_SIZE < 8192) unsigned int truesize = IGB_RX_BUFSZ; #else @@ -6912,6 +6913,7 @@ static struct sk_buff *igb_fetch_rx_buffer(struct igb_ring *rx_ring, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { + unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); struct igb_rx_buffer *rx_buffer; struct page *page; @@ -6947,11 +6949,11 @@ static struct sk_buff *igb_fetch_rx_buffer(struct igb_ring *rx_ring, dma_sync_single_range_for_cpu(rx_ring->dev, rx_buffer->dma, rx_buffer->page_offset, - IGB_RX_BUFSZ, + size, DMA_FROM_DEVICE); /* pull page into skb */ - if (igb_add_rx_frag(rx_ring, rx_buffer, rx_desc, skb)) { + if (igb_add_rx_frag(rx_ring, rx_buffer, size, rx_desc, skb)) { /* hand second half of page back to the ring */ igb_reuse_rx_page(rx_ring, rx_buffer); } else { @@ -7527,6 +7529,8 @@ static int __igb_shutdown(struct pci_dev *pdev, bool *enable_wake, if (netif_running(netdev)) __igb_close(netdev, true); + igb_ptp_suspend(adapter); + igb_clear_interrupt_scheme(adapter); #ifdef CONFIG_PM diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index f097c5a8ab93..e61b647f5f2a 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -684,6 +684,7 @@ void igb_ptp_rx_hang(struct igb_adapter *adapter) u32 tsyncrxctl = rd32(E1000_TSYNCRXCTL); unsigned long rx_event; + /* Other hardware uses per-packet timestamps */ if (hw->mac.type != e1000_82576) return; @@ -1042,6 +1043,13 @@ int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr) -EFAULT : 0; } +/** + * igb_ptp_init - Initialize PTP functionality + * @adapter: Board private structure + * + * This function is called at device probe to initialize the PTP + * functionality. + */ void igb_ptp_init(struct igb_adapter *adapter) { struct e1000_hw *hw = &adapter->hw; @@ -1064,8 +1072,7 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->cc.mask = CYCLECOUNTER_MASK(64); adapter->cc.mult = 1; adapter->cc.shift = IGB_82576_TSYNC_SHIFT; - /* Dial the nominal frequency. */ - wr32(E1000_TIMINCA, INCPERIOD_82576 | INCVALUE_82576); + adapter->ptp_flags |= IGB_PTP_OVERFLOW_CHECK; break; case e1000_82580: case e1000_i354: @@ -1084,8 +1091,7 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->cc.mask = CYCLECOUNTER_MASK(IGB_NBITS_82580); adapter->cc.mult = 1; adapter->cc.shift = 0; - /* Enable the timer functions by clearing bit 31. */ - wr32(E1000_TSAUXC, 0x0); + adapter->ptp_flags |= IGB_PTP_OVERFLOW_CHECK; break; case e1000_i210: case e1000_i211: @@ -1110,44 +1116,24 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->ptp_caps.settime64 = igb_ptp_settime_i210; adapter->ptp_caps.enable = igb_ptp_feature_enable_i210; adapter->ptp_caps.verify = igb_ptp_verify_pin; - /* Enable the timer functions by clearing bit 31. */ - wr32(E1000_TSAUXC, 0x0); break; default: adapter->ptp_clock = NULL; return; } - wrfl(); - spin_lock_init(&adapter->tmreg_lock); INIT_WORK(&adapter->ptp_tx_work, igb_ptp_tx_work); - /* Initialize the clock and overflow work for devices that need it. */ - if ((hw->mac.type == e1000_i210) || (hw->mac.type == e1000_i211)) { - struct timespec64 ts = ktime_to_timespec64(ktime_get_real()); - - igb_ptp_settime_i210(&adapter->ptp_caps, &ts); - } else { - timecounter_init(&adapter->tc, &adapter->cc, - ktime_to_ns(ktime_get_real())); - + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) INIT_DELAYED_WORK(&adapter->ptp_overflow_work, igb_ptp_overflow_check); - schedule_delayed_work(&adapter->ptp_overflow_work, - IGB_SYSTIM_OVERFLOW_PERIOD); - } - - /* Initialize the time sync interrupts for devices that support it. */ - if (hw->mac.type >= e1000_82580) { - wr32(E1000_TSIM, TSYNC_INTERRUPTS); - wr32(E1000_IMS, E1000_IMS_TS); - } - adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE; adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF; + igb_ptp_reset(adapter); + adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps, &adapter->pdev->dev); if (IS_ERR(adapter->ptp_clock)) { @@ -1156,32 +1142,24 @@ void igb_ptp_init(struct igb_adapter *adapter) } else { dev_info(&adapter->pdev->dev, "added PHC on %s\n", adapter->netdev->name); - adapter->flags |= IGB_FLAG_PTP; + adapter->ptp_flags |= IGB_PTP_ENABLED; } } /** - * igb_ptp_stop - Disable PTP device and stop the overflow check. - * @adapter: Board private structure. + * igb_ptp_suspend - Disable PTP work items and prepare for suspend + * @adapter: Board private structure * - * This function stops the PTP support and cancels the delayed work. - **/ -void igb_ptp_stop(struct igb_adapter *adapter) + * This function stops the overflow check work and PTP Tx timestamp work, and + * will prepare the device for OS suspend. + */ +void igb_ptp_suspend(struct igb_adapter *adapter) { - switch (adapter->hw.mac.type) { - case e1000_82576: - case e1000_82580: - case e1000_i354: - case e1000_i350: - cancel_delayed_work_sync(&adapter->ptp_overflow_work); - break; - case e1000_i210: - case e1000_i211: - /* No delayed work to cancel. */ - break; - default: + if (!(adapter->ptp_flags & IGB_PTP_ENABLED)) return; - } + + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) + cancel_delayed_work_sync(&adapter->ptp_overflow_work); cancel_work_sync(&adapter->ptp_tx_work); if (adapter->ptp_tx_skb) { @@ -1189,12 +1167,23 @@ void igb_ptp_stop(struct igb_adapter *adapter) adapter->ptp_tx_skb = NULL; clear_bit_unlock(__IGB_PTP_TX_IN_PROGRESS, &adapter->state); } +} + +/** + * igb_ptp_stop - Disable PTP device and stop the overflow check. + * @adapter: Board private structure. + * + * This function stops the PTP support and cancels the delayed work. + **/ +void igb_ptp_stop(struct igb_adapter *adapter) +{ + igb_ptp_suspend(adapter); if (adapter->ptp_clock) { ptp_clock_unregister(adapter->ptp_clock); dev_info(&adapter->pdev->dev, "removed PHC on %s\n", adapter->netdev->name); - adapter->flags &= ~IGB_FLAG_PTP; + adapter->ptp_flags &= ~IGB_PTP_ENABLED; } } @@ -1209,9 +1198,6 @@ void igb_ptp_reset(struct igb_adapter *adapter) struct e1000_hw *hw = &adapter->hw; unsigned long flags; - if (!(adapter->flags & IGB_FLAG_PTP)) - return; - /* reset the tstamp_config */ igb_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config); @@ -1248,4 +1234,10 @@ void igb_ptp_reset(struct igb_adapter *adapter) } out: spin_unlock_irqrestore(&adapter->tmreg_lock, flags); + + wrfl(); + + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) + schedule_delayed_work(&adapter->ptp_overflow_work, + IGB_SYSTIM_OVERFLOW_PERIOD); } diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 59b771b9b354..8a8450788124 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -2991,10 +2991,15 @@ static int ixgbe_get_ts_info(struct net_device *dev, { struct ixgbe_adapter *adapter = netdev_priv(dev); + /* we always support timestamping disabled */ + info->rx_filters = BIT(HWTSTAMP_FILTER_NONE); + switch (adapter->hw.mac.type) { case ixgbe_mac_X550: case ixgbe_mac_X550EM_x: case ixgbe_mac_x550em_a: + info->rx_filters |= BIT(HWTSTAMP_FILTER_ALL); + /* fallthrough */ case ixgbe_mac_X540: case ixgbe_mac_82599EB: info->so_timestamping = @@ -3014,8 +3019,7 @@ static int ixgbe_get_ts_info(struct net_device *dev, BIT(HWTSTAMP_TX_OFF) | BIT(HWTSTAMP_TX_ON); - info->rx_filters = - BIT(HWTSTAMP_FILTER_NONE) | + info->rx_filters |= BIT(HWTSTAMP_FILTER_PTP_V1_L4_SYNC) | BIT(HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ) | BIT(HWTSTAMP_FILTER_PTP_V2_EVENT); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 088c47cf27d9..fd5a761c68f3 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -50,7 +50,7 @@ #include <linux/if_bridge.h> #include <linux/prefetch.h> #include <scsi/fc/fc_fcoe.h> -#include <net/vxlan.h> +#include <net/udp_tunnel.h> #include <net/pkt_cls.h> #include <net/tc_act/tc_gact.h> #include <net/tc_act/tc_mirred.h> @@ -5722,9 +5722,7 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter) #ifdef CONFIG_IXGBE_DCA adapter->flags &= ~IXGBE_FLAG_DCA_CAPABLE; #endif -#ifdef CONFIG_IXGBE_VXLAN adapter->flags |= IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE; -#endif break; default: break; @@ -6158,9 +6156,7 @@ int ixgbe_open(struct net_device *netdev) ixgbe_up_complete(adapter); ixgbe_clear_vxlan_port(adapter); -#ifdef CONFIG_IXGBE_VXLAN - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); return 0; @@ -7262,14 +7258,12 @@ static void ixgbe_service_task(struct work_struct *work) ixgbe_service_event_complete(adapter); return; } -#ifdef CONFIG_IXGBE_VXLAN - rtnl_lock(); if (adapter->flags2 & IXGBE_FLAG2_VXLAN_REREG_NEEDED) { + rtnl_lock(); adapter->flags2 &= ~IXGBE_FLAG2_VXLAN_REREG_NEEDED; - vxlan_get_rx_port(adapter->netdev); + udp_tunnel_get_rx_info(adapter->netdev); + rtnl_unlock(); } - rtnl_unlock(); -#endif /* CONFIG_IXGBE_VXLAN */ ixgbe_reset_subtask(adapter); ixgbe_phy_interrupt_subtask(adapter); ixgbe_sfp_detection_subtask(adapter); @@ -7697,7 +7691,6 @@ static void ixgbe_atr(struct ixgbe_ring *ring, /* snag network header to get L4 type and address */ skb = first->skb; hdr.network = skb_network_header(skb); -#ifdef CONFIG_IXGBE_VXLAN if (skb->encapsulation && first->protocol == htons(ETH_P_IP) && hdr.ipv4->protocol != IPPROTO_UDP) { @@ -7708,7 +7701,6 @@ static void ixgbe_atr(struct ixgbe_ring *ring, udp_hdr(skb)->dest == adapter->vxlan_port) hdr.network = skb_inner_network_header(skb); } -#endif /* CONFIG_IXGBE_VXLAN */ /* Currently only IPv4/IPv6 with TCP is supported */ switch (hdr.ipv4->version) { @@ -8308,14 +8300,53 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc) static int ixgbe_delete_clsu32(struct ixgbe_adapter *adapter, struct tc_cls_u32_offload *cls) { + u32 hdl = cls->knode.handle; u32 uhtid = TC_U32_USERHTID(cls->knode.handle); - u32 loc; - int err; + u32 loc = cls->knode.handle & 0xfffff; + int err = 0, i, j; + struct ixgbe_jump_table *jump = NULL; + + if (loc > IXGBE_MAX_HW_ENTRIES) + return -EINVAL; if ((uhtid != 0x800) && (uhtid >= IXGBE_MAX_LINK_HANDLE)) return -EINVAL; - loc = cls->knode.handle & 0xfffff; + /* Clear this filter in the link data it is associated with */ + if (uhtid != 0x800) { + jump = adapter->jump_tables[uhtid]; + if (!jump) + return -EINVAL; + if (!test_bit(loc - 1, jump->child_loc_map)) + return -EINVAL; + clear_bit(loc - 1, jump->child_loc_map); + } + + /* Check if the filter being deleted is a link */ + for (i = 1; i < IXGBE_MAX_LINK_HANDLE; i++) { + jump = adapter->jump_tables[i]; + if (jump && jump->link_hdl == hdl) { + /* Delete filters in the hardware in the child hash + * table associated with this link + */ + for (j = 0; j < IXGBE_MAX_HW_ENTRIES; j++) { + if (!test_bit(j, jump->child_loc_map)) + continue; + spin_lock(&adapter->fdir_perfect_lock); + err = ixgbe_update_ethtool_fdir_entry(adapter, + NULL, + j + 1); + spin_unlock(&adapter->fdir_perfect_lock); + clear_bit(j, jump->child_loc_map); + } + /* Remove resources for this link */ + kfree(jump->input); + kfree(jump->mask); + kfree(jump); + adapter->jump_tables[i] = NULL; + return err; + } + } spin_lock(&adapter->fdir_perfect_lock); err = ixgbe_update_ethtool_fdir_entry(adapter, NULL, loc); @@ -8549,6 +8580,18 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, if (!test_bit(link_uhtid - 1, &adapter->tables)) return err; + /* Multiple filters as links to the same hash table are not + * supported. To add a new filter with the same next header + * but different match/jump conditions, create a new hash table + * and link to it. + */ + if (adapter->jump_tables[link_uhtid] && + (adapter->jump_tables[link_uhtid])->link_hdl) { + e_err(drv, "Link filter exists for link: %x\n", + link_uhtid); + return err; + } + for (i = 0; nexthdr[i].jump; i++) { if (nexthdr[i].o != cls->knode.sel->offoff || nexthdr[i].s != cls->knode.sel->offshift || @@ -8570,6 +8613,8 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, } jump->input = input; jump->mask = mask; + jump->link_hdl = cls->knode.handle; + err = ixgbe_clsu32_build_input(input, mask, cls, field_ptr, &nexthdr[i]); if (!err) { @@ -8597,6 +8642,20 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, if ((adapter->jump_tables[uhtid])->mask) memcpy(mask, (adapter->jump_tables[uhtid])->mask, sizeof(*mask)); + + /* Lookup in all child hash tables if this location is already + * filled with a filter + */ + for (i = 1; i < IXGBE_MAX_LINK_HANDLE; i++) { + struct ixgbe_jump_table *link = adapter->jump_tables[i]; + + if (link && (test_bit(loc - 1, link->child_loc_map))) { + e_err(drv, "Filter exists in location: %x\n", + loc); + err = -EINVAL; + goto err_out; + } + } } err = ixgbe_clsu32_build_input(input, mask, cls, field_ptr, NULL); if (err) @@ -8628,6 +8687,9 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, ixgbe_update_ethtool_fdir_entry(adapter, input, input->sw_idx); spin_unlock(&adapter->fdir_perfect_lock); + if ((uhtid != 0x800) && (adapter->jump_tables[uhtid])) + set_bit(loc - 1, (adapter->jump_tables[uhtid])->child_loc_map); + kfree(mask); return err; err_out_w_lock: @@ -8770,14 +8832,12 @@ static int ixgbe_set_features(struct net_device *netdev, netdev->features = features; -#ifdef CONFIG_IXGBE_VXLAN if ((adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) { if (features & NETIF_F_RXCSUM) adapter->flags2 |= IXGBE_FLAG2_VXLAN_REREG_NEEDED; else ixgbe_clear_vxlan_port(adapter); } -#endif /* CONFIG_IXGBE_VXLAN */ if (need_reset) ixgbe_do_reset(netdev); @@ -8788,23 +8848,27 @@ static int ixgbe_set_features(struct net_device *netdev, return 0; } -#ifdef CONFIG_IXGBE_VXLAN /** * ixgbe_add_vxlan_port - Get notifications about VXLAN ports that come up * @dev: The port's netdev * @sa_family: Socket Family that VXLAN is notifiying us about * @port: New UDP port number that VXLAN started listening to + * @type: Enumerated type specifying UDP tunnel type **/ -static void ixgbe_add_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void ixgbe_add_vxlan_port(struct net_device *dev, + struct udp_tunnel_info *ti) { struct ixgbe_adapter *adapter = netdev_priv(dev); struct ixgbe_hw *hw = &adapter->hw; + __be16 port = ti->port; - if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) return; - if (sa_family == AF_INET6) + if (ti->sa_family != AF_INET) + return; + + if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) return; if (adapter->vxlan_port == port) @@ -8826,28 +8890,31 @@ static void ixgbe_add_vxlan_port(struct net_device *dev, sa_family_t sa_family, * @dev: The port's netdev * @sa_family: Socket Family that VXLAN is notifying us about * @port: UDP port number that VXLAN stopped listening to + * @type: Enumerated type specifying UDP tunnel type **/ -static void ixgbe_del_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void ixgbe_del_vxlan_port(struct net_device *dev, + struct udp_tunnel_info *ti) { struct ixgbe_adapter *adapter = netdev_priv(dev); - if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) return; - if (sa_family == AF_INET6) + if (ti->sa_family != AF_INET) return; - if (adapter->vxlan_port != port) { + if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) + return; + + if (adapter->vxlan_port != ti->port) { netdev_info(dev, "Port %d was not found, not deleting\n", - ntohs(port)); + ntohs(ti->port)); return; } ixgbe_clear_vxlan_port(adapter); adapter->flags2 |= IXGBE_FLAG2_VXLAN_REREG_NEEDED; } -#endif /* CONFIG_IXGBE_VXLAN */ static int ixgbe_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], struct net_device *dev, @@ -9160,10 +9227,8 @@ static const struct net_device_ops ixgbe_netdev_ops = { .ndo_bridge_getlink = ixgbe_ndo_bridge_getlink, .ndo_dfwd_add_station = ixgbe_fwd_add, .ndo_dfwd_del_station = ixgbe_fwd_del, -#ifdef CONFIG_IXGBE_VXLAN - .ndo_add_vxlan_port = ixgbe_add_vxlan_port, - .ndo_del_vxlan_port = ixgbe_del_vxlan_port, -#endif /* CONFIG_IXGBE_VXLAN */ + .ndo_udp_tunnel_add = ixgbe_add_vxlan_port, + .ndo_udp_tunnel_del = ixgbe_del_vxlan_port, .ndo_features_check = ixgbe_features_check, }; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h index a8bed3d887f7..538a1c5475b6 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h @@ -42,8 +42,12 @@ struct ixgbe_jump_table { struct ixgbe_mat_field *mat; struct ixgbe_fdir_filter *input; union ixgbe_atr_input *mask; + u32 link_hdl; + unsigned long child_loc_map[32]; }; +#define IXGBE_MAX_HW_ENTRIES 2045 + static inline int ixgbe_mat_prgm_sip(struct ixgbe_fdir_filter *input, union ixgbe_atr_input *mask, u32 val, u32 m) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c index c5caacdd193d..8618599dfd6f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c @@ -954,6 +954,7 @@ static int ixgbe_set_vf_macvlan_msg(struct ixgbe_adapter *adapter, struct ixgbe_hw *hw = &adapter->hw; hw->mac.ops.set_mac_anti_spoofing(hw, false, vf); + hw->mac.ops.set_vlan_anti_spoofing(hw, false, vf); } } diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index dc82b1b19574..91e09d68b7e2 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -102,7 +102,6 @@ struct ltq_etop_priv { struct resource *res; struct mii_bus *mii_bus; - struct phy_device *phydev; struct ltq_etop_chan ch[MAX_DMA_CHAN]; int tx_free[MAX_DMA_CHAN >> 1]; @@ -305,34 +304,16 @@ ltq_etop_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) } static int -ltq_etop_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct ltq_etop_priv *priv = netdev_priv(dev); - - return phy_ethtool_gset(priv->phydev, cmd); -} - -static int -ltq_etop_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct ltq_etop_priv *priv = netdev_priv(dev); - - return phy_ethtool_sset(priv->phydev, cmd); -} - -static int ltq_etop_nway_reset(struct net_device *dev) { - struct ltq_etop_priv *priv = netdev_priv(dev); - - return phy_start_aneg(priv->phydev); + return phy_start_aneg(dev->phydev); } static const struct ethtool_ops ltq_etop_ethtool_ops = { .get_drvinfo = ltq_etop_get_drvinfo, - .get_settings = ltq_etop_get_settings, - .set_settings = ltq_etop_set_settings, .nway_reset = ltq_etop_nway_reset, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int @@ -401,7 +382,6 @@ ltq_etop_mdio_probe(struct net_device *dev) | SUPPORTED_TP); phydev->advertising = phydev->supported; - priv->phydev = phydev; phy_attached_info(phydev); return 0; @@ -411,7 +391,6 @@ static int ltq_etop_mdio_init(struct net_device *dev) { struct ltq_etop_priv *priv = netdev_priv(dev); - int i; int err; priv->mii_bus = mdiobus_alloc(); @@ -451,7 +430,7 @@ ltq_etop_mdio_cleanup(struct net_device *dev) { struct ltq_etop_priv *priv = netdev_priv(dev); - phy_disconnect(priv->phydev); + phy_disconnect(dev->phydev); mdiobus_unregister(priv->mii_bus); mdiobus_free(priv->mii_bus); } @@ -470,7 +449,7 @@ ltq_etop_open(struct net_device *dev) ltq_dma_open(&ch->dma); napi_enable(&ch->napi); } - phy_start(priv->phydev); + phy_start(dev->phydev); netif_tx_start_all_queues(dev); return 0; } @@ -482,7 +461,7 @@ ltq_etop_stop(struct net_device *dev) int i; netif_tx_stop_all_queues(dev); - phy_stop(priv->phydev); + phy_stop(dev->phydev); for (i = 0; i < MAX_DMA_CHAN; i++) { struct ltq_etop_chan *ch = &priv->ch[i]; @@ -557,10 +536,8 @@ ltq_etop_change_mtu(struct net_device *dev, int new_mtu) static int ltq_etop_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct ltq_etop_priv *priv = netdev_priv(dev); - /* TODO: mii-toll reports "No MII transceiver present!." ?!*/ - return phy_mii_ioctl(priv->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static int diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c index 868a957f24bb..0b047178cda1 100644 --- a/drivers/net/ethernet/marvell/mvpp2.c +++ b/drivers/net/ethernet/marvell/mvpp2.c @@ -699,7 +699,6 @@ struct mvpp2_port { u16 rx_ring_size; struct mvpp2_pcpu_stats __percpu *stats; - struct phy_device *phy_dev; phy_interface_t phy_interface; struct device_node *phy_node; unsigned int link; @@ -4850,7 +4849,7 @@ static irqreturn_t mvpp2_isr(int irq, void *dev_id) static void mvpp2_link_event(struct net_device *dev) { struct mvpp2_port *port = netdev_priv(dev); - struct phy_device *phydev = port->phy_dev; + struct phy_device *phydev = dev->phydev; int status_change = 0; u32 val; @@ -5416,6 +5415,8 @@ static int mvpp2_poll(struct napi_struct *napi, int budget) /* Set hw internals when starting port */ static void mvpp2_start_dev(struct mvpp2_port *port) { + struct net_device *ndev = port->dev; + mvpp2_gmac_max_rx_size_set(port); mvpp2_txp_max_tx_size_set(port); @@ -5425,13 +5426,15 @@ static void mvpp2_start_dev(struct mvpp2_port *port) mvpp2_interrupts_enable(port); mvpp2_port_enable(port); - phy_start(port->phy_dev); + phy_start(ndev->phydev); netif_tx_start_all_queues(port->dev); } /* Set hw internals when stopping port */ static void mvpp2_stop_dev(struct mvpp2_port *port) { + struct net_device *ndev = port->dev; + /* Stop new packets from arriving to RXQs */ mvpp2_ingress_disable(port); @@ -5447,7 +5450,7 @@ static void mvpp2_stop_dev(struct mvpp2_port *port) mvpp2_egress_disable(port); mvpp2_port_disable(port); - phy_stop(port->phy_dev); + phy_stop(ndev->phydev); } /* Return positive if MTU is valid */ @@ -5535,7 +5538,6 @@ static int mvpp2_phy_connect(struct mvpp2_port *port) phy_dev->supported &= PHY_GBIT_FEATURES; phy_dev->advertising = phy_dev->supported; - port->phy_dev = phy_dev; port->link = 0; port->duplex = 0; port->speed = 0; @@ -5545,8 +5547,9 @@ static int mvpp2_phy_connect(struct mvpp2_port *port) static void mvpp2_phy_disconnect(struct mvpp2_port *port) { - phy_disconnect(port->phy_dev); - port->phy_dev = NULL; + struct net_device *ndev = port->dev; + + phy_disconnect(ndev->phydev); } static int mvpp2_open(struct net_device *dev) @@ -5796,13 +5799,12 @@ mvpp2_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct mvpp2_port *port = netdev_priv(dev); int ret; - if (!port->phy_dev) + if (!dev->phydev) return -ENOTSUPP; - ret = phy_mii_ioctl(port->phy_dev, ifr, cmd); + ret = phy_mii_ioctl(dev->phydev, ifr, cmd); if (!ret) mvpp2_link_event(dev); @@ -5811,28 +5813,6 @@ static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) /* Ethtool methods */ -/* Get settings (phy address, speed) for ethtools */ -static int mvpp2_ethtool_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct mvpp2_port *port = netdev_priv(dev); - - if (!port->phy_dev) - return -ENODEV; - return phy_ethtool_gset(port->phy_dev, cmd); -} - -/* Set settings (phy address, speed) for ethtools */ -static int mvpp2_ethtool_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct mvpp2_port *port = netdev_priv(dev); - - if (!port->phy_dev) - return -ENODEV; - return phy_ethtool_sset(port->phy_dev, cmd); -} - /* Set interrupt coalescing for ethtools */ static int mvpp2_ethtool_set_coalesce(struct net_device *dev, struct ethtool_coalesce *c) @@ -5967,13 +5947,13 @@ static const struct net_device_ops mvpp2_netdev_ops = { static const struct ethtool_ops mvpp2_eth_tool_ops = { .get_link = ethtool_op_get_link, - .get_settings = mvpp2_ethtool_get_settings, - .set_settings = mvpp2_ethtool_set_settings, .set_coalesce = mvpp2_ethtool_set_coalesce, .get_coalesce = mvpp2_ethtool_get_coalesce, .get_drvinfo = mvpp2_ethtool_get_drvinfo, .get_ringparam = mvpp2_ethtool_get_ringparam, .set_ringparam = mvpp2_ethtool_set_ringparam, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* Driver initialization */ diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index c984462fad2a..682797ea5e11 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -133,6 +133,8 @@ static int mtk_mdio_read(struct mii_bus *bus, int phy_addr, int phy_reg) static void mtk_phy_link_adjust(struct net_device *dev) { struct mtk_mac *mac = netdev_priv(dev); + u16 lcl_adv = 0, rmt_adv = 0; + u8 flowctrl; u32 mcr = MAC_MCR_MAX_RX_1536 | MAC_MCR_IPG_CFG | MAC_MCR_FORCE_MODE | MAC_MCR_TX_EN | MAC_MCR_RX_EN | MAC_MCR_BACKOFF_EN | @@ -150,11 +152,30 @@ static void mtk_phy_link_adjust(struct net_device *dev) if (mac->phy_dev->link) mcr |= MAC_MCR_FORCE_LINK; - if (mac->phy_dev->duplex) + if (mac->phy_dev->duplex) { mcr |= MAC_MCR_FORCE_DPX; - if (mac->phy_dev->pause) - mcr |= MAC_MCR_FORCE_RX_FC | MAC_MCR_FORCE_TX_FC; + if (mac->phy_dev->pause) + rmt_adv = LPA_PAUSE_CAP; + if (mac->phy_dev->asym_pause) + rmt_adv |= LPA_PAUSE_ASYM; + + if (mac->phy_dev->advertising & ADVERTISED_Pause) + lcl_adv |= ADVERTISE_PAUSE_CAP; + if (mac->phy_dev->advertising & ADVERTISED_Asym_Pause) + lcl_adv |= ADVERTISE_PAUSE_ASYM; + + flowctrl = mii_resolve_flowctrl_fdx(lcl_adv, rmt_adv); + + if (flowctrl & FLOW_CTRL_TX) + mcr |= MAC_MCR_FORCE_TX_FC; + if (flowctrl & FLOW_CTRL_RX) + mcr |= MAC_MCR_FORCE_RX_FC; + + netif_dbg(mac->hw, link, dev, "rx pause %s, tx pause %s\n", + flowctrl & FLOW_CTRL_RX ? "enabled" : "disabled", + flowctrl & FLOW_CTRL_TX ? "enabled" : "disabled"); + } mtk_w32(mac->hw, mcr, MTK_MAC_MCR(mac->id)); @@ -208,10 +229,16 @@ static int mtk_phy_connect(struct mtk_mac *mac) u32 val, ge_mode; np = of_parse_phandle(mac->of_node, "phy-handle", 0); + if (!np && of_phy_is_fixed_link(mac->of_node)) + if (!of_phy_register_fixed_link(mac->of_node)) + np = of_node_get(mac->of_node); if (!np) return -ENODEV; switch (of_get_phy_mode(np)) { + case PHY_INTERFACE_MODE_RGMII_TXID: + case PHY_INTERFACE_MODE_RGMII_RXID: + case PHY_INTERFACE_MODE_RGMII_ID: case PHY_INTERFACE_MODE_RGMII: ge_mode = 0; break; @@ -236,7 +263,8 @@ static int mtk_phy_connect(struct mtk_mac *mac) mac->phy_dev->autoneg = AUTONEG_ENABLE; mac->phy_dev->speed = 0; mac->phy_dev->duplex = 0; - mac->phy_dev->supported &= PHY_BASIC_FEATURES; + mac->phy_dev->supported &= PHY_GBIT_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause; mac->phy_dev->advertising = mac->phy_dev->supported | ADVERTISED_Autoneg; phy_start_aneg(mac->phy_dev); @@ -280,7 +308,7 @@ static int mtk_mdio_init(struct mtk_eth *eth) return 0; err_free_bus: - kfree(eth->mii_bus); + mdiobus_free(eth->mii_bus); err_put_node: of_node_put(mii_np); @@ -295,27 +323,29 @@ static void mtk_mdio_cleanup(struct mtk_eth *eth) mdiobus_unregister(eth->mii_bus); of_node_put(eth->mii_bus->dev.of_node); - kfree(eth->mii_bus); + mdiobus_free(eth->mii_bus); } static inline void mtk_irq_disable(struct mtk_eth *eth, u32 mask) { + unsigned long flags; u32 val; + spin_lock_irqsave(ð->irq_lock, flags); val = mtk_r32(eth, MTK_QDMA_INT_MASK); mtk_w32(eth, val & ~mask, MTK_QDMA_INT_MASK); - /* flush write */ - mtk_r32(eth, MTK_QDMA_INT_MASK); + spin_unlock_irqrestore(ð->irq_lock, flags); } static inline void mtk_irq_enable(struct mtk_eth *eth, u32 mask) { + unsigned long flags; u32 val; + spin_lock_irqsave(ð->irq_lock, flags); val = mtk_r32(eth, MTK_QDMA_INT_MASK); mtk_w32(eth, val | mask, MTK_QDMA_INT_MASK); - /* flush write */ - mtk_r32(eth, MTK_QDMA_INT_MASK); + spin_unlock_irqrestore(ð->irq_lock, flags); } static int mtk_set_mac_address(struct net_device *dev, void *p) @@ -453,20 +483,23 @@ static inline void mtk_rx_get_desc(struct mtk_rx_dma *rxd, /* the qdma core needs scratch memory to be setup */ static int mtk_init_fq_dma(struct mtk_eth *eth) { - dma_addr_t phy_ring_head, phy_ring_tail; + dma_addr_t phy_ring_tail; int cnt = MTK_DMA_SIZE; dma_addr_t dma_addr; int i; eth->scratch_ring = dma_alloc_coherent(eth->dev, cnt * sizeof(struct mtk_tx_dma), - &phy_ring_head, + ð->phy_scratch_ring, GFP_ATOMIC | __GFP_ZERO); if (unlikely(!eth->scratch_ring)) return -ENOMEM; eth->scratch_head = kcalloc(cnt, MTK_QDMA_PAGE_SIZE, GFP_KERNEL); + if (unlikely(!eth->scratch_head)) + return -ENOMEM; + dma_addr = dma_map_single(eth->dev, eth->scratch_head, cnt * MTK_QDMA_PAGE_SIZE, DMA_FROM_DEVICE); @@ -474,19 +507,19 @@ static int mtk_init_fq_dma(struct mtk_eth *eth) return -ENOMEM; memset(eth->scratch_ring, 0x0, sizeof(struct mtk_tx_dma) * cnt); - phy_ring_tail = phy_ring_head + + phy_ring_tail = eth->phy_scratch_ring + (sizeof(struct mtk_tx_dma) * (cnt - 1)); for (i = 0; i < cnt; i++) { eth->scratch_ring[i].txd1 = (dma_addr + (i * MTK_QDMA_PAGE_SIZE)); if (i < cnt - 1) - eth->scratch_ring[i].txd2 = (phy_ring_head + + eth->scratch_ring[i].txd2 = (eth->phy_scratch_ring + ((i + 1) * sizeof(struct mtk_tx_dma))); eth->scratch_ring[i].txd3 = TX_DMA_SDL(MTK_QDMA_PAGE_SIZE); } - mtk_w32(eth, phy_ring_head, MTK_QDMA_FQ_HEAD); + mtk_w32(eth, eth->phy_scratch_ring, MTK_QDMA_FQ_HEAD); mtk_w32(eth, phy_ring_tail, MTK_QDMA_FQ_TAIL); mtk_w32(eth, (cnt << 16) | cnt, MTK_QDMA_FQ_CNT); mtk_w32(eth, MTK_QDMA_PAGE_SIZE << 16, MTK_QDMA_FQ_BLEN); @@ -643,7 +676,7 @@ static int mtk_tx_map(struct sk_buff *skb, struct net_device *dev, err_dma: do { - tx_buf = mtk_desc_to_tx_buf(ring, txd); + tx_buf = mtk_desc_to_tx_buf(ring, itxd); /* unmap dma */ mtk_tx_unmap(&dev->dev, tx_buf); @@ -673,6 +706,20 @@ static inline int mtk_cal_txd_req(struct sk_buff *skb) return nfrags; } +static int mtk_queue_stopped(struct mtk_eth *eth) +{ + int i; + + for (i = 0; i < MTK_MAC_COUNT; i++) { + if (!eth->netdev[i]) + continue; + if (netif_queue_stopped(eth->netdev[i])) + return 1; + } + + return 0; +} + static void mtk_wake_queue(struct mtk_eth *eth) { int i; @@ -738,12 +785,9 @@ static int mtk_start_xmit(struct sk_buff *skb, struct net_device *dev) if (mtk_tx_map(skb, dev, tx_num, ring, gso) < 0) goto drop; - if (unlikely(atomic_read(&ring->free_count) <= ring->thresh)) { + if (unlikely(atomic_read(&ring->free_count) <= ring->thresh)) mtk_stop_queue(eth); - if (unlikely(atomic_read(&ring->free_count) > - ring->thresh)) - mtk_wake_queue(eth); - } + spin_unlock_irqrestore(ð->page_lock, flags); return NETDEV_TX_OK; @@ -756,7 +800,7 @@ drop: } static int mtk_poll_rx(struct napi_struct *napi, int budget, - struct mtk_eth *eth, u32 rx_intr) + struct mtk_eth *eth) { struct mtk_rx_ring *ring = ð->rx_ring; int idx = ring->calc_idx; @@ -798,6 +842,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget, DMA_FROM_DEVICE); if (unlikely(dma_mapping_error(&netdev->dev, dma_addr))) { skb_free_frag(new_data); + netdev->stats.rx_dropped++; goto release_desc; } @@ -805,6 +850,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget, skb = build_skb(data, ring->frag_size); if (unlikely(!skb)) { put_page(virt_to_head_page(new_data)); + netdev->stats.rx_dropped++; goto release_desc; } skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN); @@ -842,22 +888,22 @@ release_desc: } if (done < budget) - mtk_w32(eth, rx_intr, MTK_QMTK_INT_STATUS); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS); return done; } -static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) +static int mtk_poll_tx(struct mtk_eth *eth, int budget) { struct mtk_tx_ring *ring = ð->tx_ring; struct mtk_tx_dma *desc; struct sk_buff *skb; struct mtk_tx_buf *tx_buf; - int total = 0, done[MTK_MAX_DEVS]; + unsigned int done[MTK_MAX_DEVS]; unsigned int bytes[MTK_MAX_DEVS]; u32 cpu, dma; static int condition; - int i; + int total = 0, i; memset(done, 0, sizeof(done)); memset(bytes, 0, sizeof(bytes)); @@ -893,7 +939,6 @@ static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) } mtk_tx_unmap(eth->dev, tx_buf); - ring->last_free->txd2 = next_cpu; ring->last_free = desc; atomic_inc(&ring->free_count); @@ -909,64 +954,82 @@ static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) total += done[i]; } - /* read hw index again make sure no new tx packet */ - if (cpu != dma || cpu != mtk_r32(eth, MTK_QTX_DRX_PTR)) - *tx_again = true; - else - mtk_w32(eth, MTK_TX_DONE_INT, MTK_QMTK_INT_STATUS); - - if (!total) - return 0; - - if (atomic_read(&ring->free_count) > ring->thresh) + if (mtk_queue_stopped(eth) && + (atomic_read(&ring->free_count) > ring->thresh)) mtk_wake_queue(eth); return total; } -static int mtk_poll(struct napi_struct *napi, int budget) +static void mtk_handle_status_irq(struct mtk_eth *eth) { - struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi); - u32 status, status2, mask, tx_intr, rx_intr, status_intr; - int tx_done, rx_done; - bool tx_again = false; + u32 status2 = mtk_r32(eth, MTK_INT_STATUS2); - status = mtk_r32(eth, MTK_QMTK_INT_STATUS); - status2 = mtk_r32(eth, MTK_INT_STATUS2); - tx_intr = MTK_TX_DONE_INT; - rx_intr = MTK_RX_DONE_INT; - status_intr = (MTK_GDM1_AF | MTK_GDM2_AF); - tx_done = 0; - rx_done = 0; - tx_again = 0; + if (unlikely(status2 & (MTK_GDM1_AF | MTK_GDM2_AF))) { + mtk_stats_update(eth); + mtk_w32(eth, (MTK_GDM1_AF | MTK_GDM2_AF), + MTK_INT_STATUS2); + } +} - if (status & tx_intr) - tx_done = mtk_poll_tx(eth, budget, &tx_again); +static int mtk_napi_tx(struct napi_struct *napi, int budget) +{ + struct mtk_eth *eth = container_of(napi, struct mtk_eth, tx_napi); + u32 status, mask; + int tx_done = 0; - if (status & rx_intr) - rx_done = mtk_poll_rx(napi, budget, eth, rx_intr); + mtk_handle_status_irq(eth); + mtk_w32(eth, MTK_TX_DONE_INT, MTK_QMTK_INT_STATUS); + tx_done = mtk_poll_tx(eth, budget); - if (unlikely(status2 & status_intr)) { - mtk_stats_update(eth); - mtk_w32(eth, status_intr, MTK_INT_STATUS2); + if (unlikely(netif_msg_intr(eth))) { + status = mtk_r32(eth, MTK_QMTK_INT_STATUS); + mask = mtk_r32(eth, MTK_QDMA_INT_MASK); + dev_info(eth->dev, + "done tx %d, intr 0x%08x/0x%x\n", + tx_done, status, mask); } + if (tx_done == budget) + return budget; + + status = mtk_r32(eth, MTK_QMTK_INT_STATUS); + if (status & MTK_TX_DONE_INT) + return budget; + + napi_complete(napi); + mtk_irq_enable(eth, MTK_TX_DONE_INT); + + return tx_done; +} + +static int mtk_napi_rx(struct napi_struct *napi, int budget) +{ + struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi); + u32 status, mask; + int rx_done = 0; + + mtk_handle_status_irq(eth); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS); + rx_done = mtk_poll_rx(napi, budget, eth); + if (unlikely(netif_msg_intr(eth))) { + status = mtk_r32(eth, MTK_QMTK_INT_STATUS); mask = mtk_r32(eth, MTK_QDMA_INT_MASK); - netdev_info(eth->netdev[0], - "done tx %d, rx %d, intr 0x%08x/0x%x\n", - tx_done, rx_done, status, mask); + dev_info(eth->dev, + "done rx %d, intr 0x%08x/0x%x\n", + rx_done, status, mask); } - if (tx_again || rx_done == budget) + if (rx_done == budget) return budget; status = mtk_r32(eth, MTK_QMTK_INT_STATUS); - if (status & (tx_intr | rx_intr)) + if (status & MTK_RX_DONE_INT) return budget; napi_complete(napi); - mtk_irq_enable(eth, tx_intr | rx_intr); + mtk_irq_enable(eth, MTK_RX_DONE_INT); return rx_done; } @@ -999,9 +1062,8 @@ static int mtk_tx_alloc(struct mtk_eth *eth) atomic_set(&ring->free_count, MTK_DMA_SIZE - 2); ring->next_free = &ring->dma[0]; - ring->last_free = &ring->dma[MTK_DMA_SIZE - 2]; - ring->thresh = max((unsigned long)MTK_DMA_SIZE >> 2, - MAX_SKB_FRAGS); + ring->last_free = &ring->dma[MTK_DMA_SIZE - 1]; + ring->thresh = MAX_SKB_FRAGS; /* make sure that all changes to the dma ring are flushed before we * continue @@ -1179,6 +1241,14 @@ static void mtk_dma_free(struct mtk_eth *eth) for (i = 0; i < MTK_MAC_COUNT; i++) if (eth->netdev[i]) netdev_reset_queue(eth->netdev[i]); + if (eth->scratch_ring) { + dma_free_coherent(eth->dev, + MTK_DMA_SIZE * sizeof(struct mtk_tx_dma), + eth->scratch_ring, + eth->phy_scratch_ring); + eth->scratch_ring = NULL; + eth->phy_scratch_ring = 0; + } mtk_tx_clean(eth); mtk_rx_clean(eth); kfree(eth->scratch_head); @@ -1195,22 +1265,26 @@ static void mtk_tx_timeout(struct net_device *dev) schedule_work(ð->pending_work); } -static irqreturn_t mtk_handle_irq(int irq, void *_eth) +static irqreturn_t mtk_handle_irq_rx(int irq, void *_eth) { struct mtk_eth *eth = _eth; - u32 status; - status = mtk_r32(eth, MTK_QMTK_INT_STATUS); - if (unlikely(!status)) - return IRQ_NONE; + if (likely(napi_schedule_prep(ð->rx_napi))) { + __napi_schedule(ð->rx_napi); + mtk_irq_disable(eth, MTK_RX_DONE_INT); + } - if (likely(status & (MTK_RX_DONE_INT | MTK_TX_DONE_INT))) { - if (likely(napi_schedule_prep(ð->rx_napi))) - __napi_schedule(ð->rx_napi); - } else { - mtk_w32(eth, status, MTK_QMTK_INT_STATUS); + return IRQ_HANDLED; +} + +static irqreturn_t mtk_handle_irq_tx(int irq, void *_eth) +{ + struct mtk_eth *eth = _eth; + + if (likely(napi_schedule_prep(ð->tx_napi))) { + __napi_schedule(ð->tx_napi); + mtk_irq_disable(eth, MTK_TX_DONE_INT); } - mtk_irq_disable(eth, (MTK_RX_DONE_INT | MTK_TX_DONE_INT)); return IRQ_HANDLED; } @@ -1223,7 +1297,7 @@ static void mtk_poll_controller(struct net_device *dev) u32 int_mask = MTK_TX_DONE_INT | MTK_RX_DONE_INT; mtk_irq_disable(eth, int_mask); - mtk_handle_irq(dev->irq, dev); + mtk_handle_irq_rx(eth->irq[2], dev); mtk_irq_enable(eth, int_mask); } #endif @@ -1241,7 +1315,7 @@ static int mtk_start_dma(struct mtk_eth *eth) mtk_w32(eth, MTK_TX_WB_DDONE | MTK_RX_DMA_EN | MTK_TX_DMA_EN | MTK_RX_2B_OFFSET | MTK_DMA_SIZE_16DWORDS | - MTK_RX_BT_32DWORDS, + MTK_RX_BT_32DWORDS | MTK_NDP_CO_PRO, MTK_QDMA_GLO_CFG); return 0; @@ -1259,6 +1333,7 @@ static int mtk_open(struct net_device *dev) if (err) return err; + napi_enable(ð->tx_napi); napi_enable(ð->rx_napi); mtk_irq_enable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT); } @@ -1307,6 +1382,7 @@ static int mtk_stop(struct net_device *dev) return 0; mtk_irq_disable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT); + napi_disable(ð->tx_napi); napi_disable(ð->rx_napi); mtk_stop_dma(eth, MTK_QDMA_GLO_CFG); @@ -1344,7 +1420,11 @@ static int __init mtk_hw_init(struct mtk_eth *eth) /* Enable RX VLan Offloading */ mtk_w32(eth, 1, MTK_CDMP_EG_CTRL); - err = devm_request_irq(eth->dev, eth->irq, mtk_handle_irq, 0, + err = devm_request_irq(eth->dev, eth->irq[1], mtk_handle_irq_tx, 0, + dev_name(eth->dev), eth); + if (err) + return err; + err = devm_request_irq(eth->dev, eth->irq[2], mtk_handle_irq_rx, 0, dev_name(eth->dev), eth); if (err) return err; @@ -1355,12 +1435,16 @@ static int __init mtk_hw_init(struct mtk_eth *eth) /* disable delay and normal interrupt */ mtk_w32(eth, 0, MTK_QDMA_DELAY_INT); - mtk_irq_disable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT); + mtk_irq_disable(eth, ~0); mtk_w32(eth, RST_GL_PSE, MTK_RST_GL); mtk_w32(eth, 0, MTK_RST_GL); /* FE int grouping */ - mtk_w32(eth, 0, MTK_FE_INT_GRP); + mtk_w32(eth, MTK_TX_DONE_INT, MTK_PDMA_INT_GRP1); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_PDMA_INT_GRP2); + mtk_w32(eth, MTK_TX_DONE_INT, MTK_QDMA_INT_GRP1); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_QDMA_INT_GRP2); + mtk_w32(eth, 0x21021000, MTK_FE_INT_GRP); for (i = 0; i < 2; i++) { u32 val = mtk_r32(eth, MTK_GDMA_FWD_CFG(i)); @@ -1408,7 +1492,8 @@ static void mtk_uninit(struct net_device *dev) phy_disconnect(mac->phy_dev); mtk_mdio_cleanup(eth); mtk_irq_disable(eth, ~0); - free_irq(dev->irq, dev); + free_irq(eth->irq[1], dev); + free_irq(eth->irq[2], dev); } static int mtk_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) @@ -1669,7 +1754,7 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np) mac->hw_stats->reg_offset = id * MTK_STAT_OFFSET; SET_NETDEV_DEV(eth->netdev[id], eth->dev); - eth->netdev[id]->watchdog_timeo = HZ; + eth->netdev[id]->watchdog_timeo = 5 * HZ; eth->netdev[id]->netdev_ops = &mtk_netdev_ops; eth->netdev[id]->base_addr = (unsigned long)eth->base; eth->netdev[id]->vlan_features = MTK_HW_FEATURES & @@ -1682,10 +1767,10 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np) dev_err(eth->dev, "error bringing up device\n"); goto free_netdev; } - eth->netdev[id]->irq = eth->irq; + eth->netdev[id]->irq = eth->irq[0]; netif_info(eth, probe, eth->netdev[id], "mediatek frame engine at 0x%08lx, irq %d\n", - eth->netdev[id]->base_addr, eth->netdev[id]->irq); + eth->netdev[id]->base_addr, eth->irq[0]); return 0; @@ -1702,6 +1787,7 @@ static int mtk_probe(struct platform_device *pdev) struct mtk_soc_data *soc; struct mtk_eth *eth; int err; + int i; match = of_match_device(of_mtk_match, &pdev->dev); soc = (struct mtk_soc_data *)match->data; @@ -1715,6 +1801,7 @@ static int mtk_probe(struct platform_device *pdev) return PTR_ERR(eth->base); spin_lock_init(ð->page_lock); + spin_lock_init(ð->irq_lock); eth->ethsys = syscon_regmap_lookup_by_phandle(pdev->dev.of_node, "mediatek,ethsys"); @@ -1736,10 +1823,12 @@ static int mtk_probe(struct platform_device *pdev) return PTR_ERR(eth->rstc); } - eth->irq = platform_get_irq(pdev, 0); - if (eth->irq < 0) { - dev_err(&pdev->dev, "no IRQ resource found\n"); - return -ENXIO; + for (i = 0; i < 3; i++) { + eth->irq[i] = platform_get_irq(pdev, i); + if (eth->irq[i] < 0) { + dev_err(&pdev->dev, "no IRQ%d resource found\n", i); + return -ENXIO; + } } eth->clk_ethif = devm_clk_get(&pdev->dev, "ethif"); @@ -1780,7 +1869,9 @@ static int mtk_probe(struct platform_device *pdev) * for NAPI to work */ init_dummy_netdev(ð->dummy_dev); - netif_napi_add(ð->dummy_dev, ð->rx_napi, mtk_poll, + netif_napi_add(ð->dummy_dev, ð->tx_napi, mtk_napi_tx, + MTK_NAPI_WEIGHT); + netif_napi_add(ð->dummy_dev, ð->rx_napi, mtk_napi_rx, MTK_NAPI_WEIGHT); platform_set_drvdata(pdev, eth); @@ -1801,6 +1892,7 @@ static int mtk_remove(struct platform_device *pdev) clk_disable_unprepare(eth->clk_gp1); clk_disable_unprepare(eth->clk_gp2); + netif_napi_del(ð->tx_napi); netif_napi_del(ð->rx_napi); mtk_cleanup(eth); platform_set_drvdata(pdev, NULL); diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h index eed626d56ea4..f82e3acb947b 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -68,6 +68,10 @@ /* Unicast Filter MAC Address Register - High */ #define MTK_GDMA_MAC_ADRH(x) (0x50C + (x * 0x1000)) +/* PDMA Interrupt grouping registers */ +#define MTK_PDMA_INT_GRP1 0xa50 +#define MTK_PDMA_INT_GRP2 0xa54 + /* QDMA TX Queue Configuration Registers */ #define MTK_QTX_CFG(x) (0x1800 + (x * 0x10)) #define QDMA_RES_THRES 4 @@ -91,6 +95,7 @@ #define MTK_QDMA_GLO_CFG 0x1A04 #define MTK_RX_2B_OFFSET BIT(31) #define MTK_RX_BT_32DWORDS (3 << 11) +#define MTK_NDP_CO_PRO BIT(10) #define MTK_TX_WB_DDONE BIT(6) #define MTK_DMA_SIZE_16DWORDS (2 << 4) #define MTK_RX_DMA_BUSY BIT(3) @@ -124,6 +129,11 @@ #define MTK_TX_DONE_INT (MTK_TX_DONE_INT0 | MTK_TX_DONE_INT1 | \ MTK_TX_DONE_INT2 | MTK_TX_DONE_INT3) +/* QDMA Interrupt grouping registers */ +#define MTK_QDMA_INT_GRP1 0x1a20 +#define MTK_QDMA_INT_GRP2 0x1a24 +#define MTK_RLS_DONE_INT BIT(0) + /* QDMA Interrupt Status Register */ #define MTK_QDMA_INT_MASK 0x1A1C @@ -355,8 +365,10 @@ struct mtk_rx_ring { * @dma_refcnt: track how many netdevs are using the DMA engine * @tx_ring: Pointer to the memore holding info about the TX ring * @rx_ring: Pointer to the memore holding info about the RX ring - * @rx_napi: The NAPI struct + * @tx_napi: The TX NAPI struct + * @rx_napi: The RX NAPI struct * @scratch_ring: Newer SoCs need memory for a second HW managed TX ring + * @phy_scratch_ring: physical address of scratch_ring * @scratch_head: The scratch memory that scratch_ring points to. * @clk_ethif: The ethif clock * @clk_esw: The switch clock @@ -371,10 +383,11 @@ struct mtk_eth { void __iomem *base; struct reset_control *rstc; spinlock_t page_lock; + spinlock_t irq_lock; struct net_device dummy_dev; struct net_device *netdev[MTK_MAX_DEVS]; struct mtk_mac *mac[MTK_MAX_DEVS]; - int irq; + int irq[3]; u32 msg_enable; unsigned long sysclk; struct regmap *ethsys; @@ -382,8 +395,10 @@ struct mtk_eth { atomic_t dma_refcnt; struct mtk_tx_ring tx_ring; struct mtk_rx_ring rx_ring; + struct napi_struct tx_napi; struct napi_struct rx_napi; struct mtk_tx_dma *scratch_ring; + dma_addr_t phy_scratch_ring; void *scratch_head; struct clk *clk_ethif; struct clk *clk_esw; diff --git a/drivers/net/ethernet/mellanox/mlx4/Kconfig b/drivers/net/ethernet/mellanox/mlx4/Kconfig index 9ca3734ebb6b..5098e7f21987 100644 --- a/drivers/net/ethernet/mellanox/mlx4/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx4/Kconfig @@ -24,13 +24,6 @@ config MLX4_EN_DCB If unsure, set to Y -config MLX4_EN_VXLAN - bool "VXLAN offloads Support" - default y - depends on MLX4_EN && VXLAN && !(MLX4_EN=y && VXLAN=m) - ---help--- - Say Y here if you want to use VXLAN offloads in the driver. - config MLX4_CORE tristate depends on PCI diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index e94ca1c3fc7c..f04a423ff79d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -2597,7 +2597,6 @@ int mlx4_cmd_use_events(struct mlx4_dev *dev) priv->cmd.free_head = 0; sema_init(&priv->cmd.event_sem, priv->cmd.max_cmds); - spin_lock_init(&priv->cmd.context_lock); for (priv->cmd.token_mask = 1; priv->cmd.token_mask < priv->cmd.max_cmds; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index f01918c63f28..99c6bbdff501 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -37,6 +37,11 @@ #include "mlx4_en.h" #include "fw_qos.h" +enum { + MLX4_CEE_STATE_DOWN = 0, + MLX4_CEE_STATE_UP = 1, +}; + /* Definitions for QCN */ @@ -80,13 +85,202 @@ struct mlx4_congestion_control_mb_prio_802_1_qau_statistics { __be32 reserved3[4]; }; +static u8 mlx4_en_dcbnl_getcap(struct net_device *dev, int capid, u8 *cap) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + switch (capid) { + case DCB_CAP_ATTR_PFC: + *cap = true; + break; + case DCB_CAP_ATTR_DCBX: + *cap = priv->cee_params.dcbx_cap; + break; + case DCB_CAP_ATTR_PFC_TCS: + *cap = 1 << mlx4_max_tc(priv->mdev->dev); + break; + default: + *cap = false; + break; + } + + return 0; +} + +static u8 mlx4_en_dcbnl_getpfcstate(struct net_device *netdev) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + return priv->cee_params.dcb_cfg.pfc_state; +} + +static void mlx4_en_dcbnl_setpfcstate(struct net_device *netdev, u8 state) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + priv->cee_params.dcb_cfg.pfc_state = state; +} + +static void mlx4_en_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority, + u8 *setting) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + *setting = priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc; +} + +static void mlx4_en_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority, + u8 setting) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc = setting; + priv->cee_params.dcb_cfg.pfc_state = true; +} + +static int mlx4_en_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + if (!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) + return -EINVAL; + + if (tcid == DCB_NUMTCS_ATTR_PFC) + *num = mlx4_max_tc(priv->mdev->dev); + else + *num = 0; + + return 0; +} + +static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct mlx4_en_dev *mdev = priv->mdev; + struct mlx4_en_cee_config *dcb_cfg = &priv->cee_params.dcb_cfg; + int err = 0; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return -EINVAL; + + if (dcb_cfg->pfc_state) { + int tc; + + priv->prof->rx_pause = 0; + priv->prof->tx_pause = 0; + for (tc = 0; tc < CEE_DCBX_MAX_PRIO; tc++) { + u8 tc_mask = 1 << tc; + + switch (dcb_cfg->tc_config[tc].dcb_pfc) { + case pfc_disabled: + priv->prof->tx_ppp &= ~tc_mask; + priv->prof->rx_ppp &= ~tc_mask; + break; + case pfc_enabled_full: + priv->prof->tx_ppp |= tc_mask; + priv->prof->rx_ppp |= tc_mask; + break; + case pfc_enabled_tx: + priv->prof->tx_ppp |= tc_mask; + priv->prof->rx_ppp &= ~tc_mask; + break; + case pfc_enabled_rx: + priv->prof->tx_ppp &= ~tc_mask; + priv->prof->rx_ppp |= tc_mask; + break; + default: + break; + } + } + en_dbg(DRV, priv, "Set pfc on\n"); + } else { + priv->prof->rx_pause = 1; + priv->prof->tx_pause = 1; + en_dbg(DRV, priv, "Set pfc off\n"); + } + + err = mlx4_SET_PORT_general(mdev->dev, priv->port, + priv->rx_skb_size + ETH_FCS_LEN, + priv->prof->tx_pause, + priv->prof->tx_ppp, + priv->prof->rx_pause, + priv->prof->rx_ppp); + if (err) + en_err(priv, "Failed setting pause params\n"); + return err; +} + +static u8 mlx4_en_dcbnl_get_state(struct net_device *dev) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + if (priv->flags & MLX4_EN_FLAG_DCB_ENABLED) + return MLX4_CEE_STATE_UP; + + return MLX4_CEE_STATE_DOWN; +} + +static u8 mlx4_en_dcbnl_set_state(struct net_device *dev, u8 state) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + int num_tcs = 0; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return 1; + + if (!!(state) == !!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) + return 0; + + if (state) { + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + num_tcs = IEEE_8021QAZ_MAX_TCS; + } else { + priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; + } + + return mlx4_en_setup_tc(dev, num_tcs); +} + +/* On success returns a non-zero 802.1p user priority bitmap + * otherwise returns 0 as the invalid user priority bitmap to + * indicate an error. + */ +static int mlx4_en_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct dcb_app app = { + .selector = idtype, + .protocol = id, + }; + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return 0; + + return dcb_getapp(netdev, &app); +} + +static int mlx4_en_dcbnl_setapp(struct net_device *netdev, u8 idtype, + u16 id, u8 up) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct dcb_app app; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return -EINVAL; + + memset(&app, 0, sizeof(struct dcb_app)); + app.selector = idtype; + app.protocol = id; + app.priority = up; + + return dcb_setapp(netdev, &app); +} + static int mlx4_en_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets) { struct mlx4_en_priv *priv = netdev_priv(dev); struct ieee_ets *my_ets = &priv->ets; - /* No IEEE PFC settings available */ if (!my_ets) return -EINVAL; @@ -237,18 +431,51 @@ static int mlx4_en_dcbnl_ieee_setpfc(struct net_device *dev, static u8 mlx4_en_dcbnl_getdcbx(struct net_device *dev) { - return DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE; + struct mlx4_en_priv *priv = netdev_priv(dev); + + return priv->cee_params.dcbx_cap; } static u8 mlx4_en_dcbnl_setdcbx(struct net_device *dev, u8 mode) { + struct mlx4_en_priv *priv = netdev_priv(dev); + struct ieee_ets ets = {0}; + struct ieee_pfc pfc = {0}; + + if (mode == priv->cee_params.dcbx_cap) + return 0; + if ((mode & DCB_CAP_DCBX_LLD_MANAGED) || - (mode & DCB_CAP_DCBX_VER_CEE) || - !(mode & DCB_CAP_DCBX_VER_IEEE) || + ((mode & DCB_CAP_DCBX_VER_IEEE) && + (mode & DCB_CAP_DCBX_VER_CEE)) || !(mode & DCB_CAP_DCBX_HOST)) - return 1; + goto err; + + priv->cee_params.dcbx_cap = mode; + + ets.ets_cap = IEEE_8021QAZ_MAX_TCS; + pfc.pfc_cap = IEEE_8021QAZ_MAX_TCS; + + if (mode & DCB_CAP_DCBX_VER_IEEE) { + if (mlx4_en_dcbnl_ieee_setets(dev, &ets)) + goto err; + if (mlx4_en_dcbnl_ieee_setpfc(dev, &pfc)) + goto err; + } else if (mode & DCB_CAP_DCBX_VER_CEE) { + if (mlx4_en_dcbnl_set_all(dev)) + goto err; + } else { + if (mlx4_en_dcbnl_ieee_setets(dev, &ets)) + goto err; + if (mlx4_en_dcbnl_ieee_setpfc(dev, &pfc)) + goto err; + if (mlx4_en_setup_tc(dev, 0)) + goto err; + } return 0; +err: + return 1; } #define MLX4_RATELIMIT_UNITS_IN_KB 100000 /* rate-limit HW unit in Kbps */ @@ -463,24 +690,46 @@ static int mlx4_en_dcbnl_ieee_getqcnstats(struct net_device *dev, } const struct dcbnl_rtnl_ops mlx4_en_dcbnl_ops = { - .ieee_getets = mlx4_en_dcbnl_ieee_getets, - .ieee_setets = mlx4_en_dcbnl_ieee_setets, - .ieee_getmaxrate = mlx4_en_dcbnl_ieee_getmaxrate, - .ieee_setmaxrate = mlx4_en_dcbnl_ieee_setmaxrate, - .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, - .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + .ieee_getets = mlx4_en_dcbnl_ieee_getets, + .ieee_setets = mlx4_en_dcbnl_ieee_setets, + .ieee_getmaxrate = mlx4_en_dcbnl_ieee_getmaxrate, + .ieee_setmaxrate = mlx4_en_dcbnl_ieee_setmaxrate, + .ieee_getqcn = mlx4_en_dcbnl_ieee_getqcn, + .ieee_setqcn = mlx4_en_dcbnl_ieee_setqcn, + .ieee_getqcnstats = mlx4_en_dcbnl_ieee_getqcnstats, + .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, + .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + + .getstate = mlx4_en_dcbnl_get_state, + .setstate = mlx4_en_dcbnl_set_state, + .getpfccfg = mlx4_en_dcbnl_get_pfc_cfg, + .setpfccfg = mlx4_en_dcbnl_set_pfc_cfg, + .setall = mlx4_en_dcbnl_set_all, + .getcap = mlx4_en_dcbnl_getcap, + .getnumtcs = mlx4_en_dcbnl_getnumtcs, + .getpfcstate = mlx4_en_dcbnl_getpfcstate, + .setpfcstate = mlx4_en_dcbnl_setpfcstate, + .getapp = mlx4_en_dcbnl_getapp, + .setapp = mlx4_en_dcbnl_setapp, .getdcbx = mlx4_en_dcbnl_getdcbx, .setdcbx = mlx4_en_dcbnl_setdcbx, - .ieee_getqcn = mlx4_en_dcbnl_ieee_getqcn, - .ieee_setqcn = mlx4_en_dcbnl_ieee_setqcn, - .ieee_getqcnstats = mlx4_en_dcbnl_ieee_getqcnstats, }; const struct dcbnl_rtnl_ops mlx4_en_dcbnl_pfc_ops = { .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + .setstate = mlx4_en_dcbnl_set_state, + .getpfccfg = mlx4_en_dcbnl_get_pfc_cfg, + .setpfccfg = mlx4_en_dcbnl_set_pfc_cfg, + .setall = mlx4_en_dcbnl_set_all, + .getnumtcs = mlx4_en_dcbnl_getnumtcs, + .getpfcstate = mlx4_en_dcbnl_getpfcstate, + .setpfcstate = mlx4_en_dcbnl_setpfcstate, + .getapp = mlx4_en_dcbnl_getapp, + .setapp = mlx4_en_dcbnl_setapp, + .getdcbx = mlx4_en_dcbnl_getdcbx, .setdcbx = mlx4_en_dcbnl_setdcbx, }; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index fc95affaf76b..51a2e8252b82 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1107,7 +1107,7 @@ static u32 mlx4_en_get_rxfh_indir_size(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); - return priv->rx_ring_num; + return rounddown_pow_of_two(priv->rx_ring_num); } static u32 mlx4_en_get_rxfh_key_size(struct net_device *netdev) @@ -1141,19 +1141,17 @@ static int mlx4_en_get_rxfh(struct net_device *dev, u32 *ring_index, u8 *key, u8 *hfunc) { struct mlx4_en_priv *priv = netdev_priv(dev); - struct mlx4_en_rss_map *rss_map = &priv->rss_map; - int rss_rings; - size_t n = priv->rx_ring_num; + u32 n = mlx4_en_get_rxfh_indir_size(dev); + u32 i, rss_rings; int err = 0; - rss_rings = priv->prof->rss_rings ?: priv->rx_ring_num; - rss_rings = 1 << ilog2(rss_rings); + rss_rings = priv->prof->rss_rings ?: n; + rss_rings = rounddown_pow_of_two(rss_rings); - while (n--) { + for (i = 0; i < n; i++) { if (!ring_index) break; - ring_index[n] = rss_map->qps[n % rss_rings].qpn - - rss_map->base_qpn; + ring_index[i] = i % rss_rings; } if (key) memcpy(key, priv->rss_key, MLX4_EN_RSS_KEY_SIZE); @@ -1166,6 +1164,7 @@ static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index, const u8 *key, const u8 hfunc) { struct mlx4_en_priv *priv = netdev_priv(dev); + u32 n = mlx4_en_get_rxfh_indir_size(dev); struct mlx4_en_dev *mdev = priv->mdev; int port_up = 0; int err = 0; @@ -1175,18 +1174,18 @@ static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index, /* Calculate RSS table size and make sure flows are spread evenly * between rings */ - for (i = 0; i < priv->rx_ring_num; i++) { + for (i = 0; i < n; i++) { if (!ring_index) - continue; + break; if (i > 0 && !ring_index[i] && !rss_rings) rss_rings = i; - if (ring_index[i] != (i % (rss_rings ?: priv->rx_ring_num))) + if (ring_index[i] != (i % (rss_rings ?: n))) return -EINVAL; } if (!rss_rings) - rss_rings = priv->rx_ring_num; + rss_rings = n; /* RSS table size must be an order of 2 */ if (!is_power_of_2(rss_rings)) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 19ceced6736c..6083775dae16 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -67,6 +67,17 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up) offset += priv->num_tx_rings_p_up; } +#ifdef CONFIG_MLX4_EN_DCB + if (!mlx4_is_slave(priv->mdev->dev)) { + if (up) { + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + } else { + priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; + priv->cee_params.dcb_cfg.pfc_state = false; + } + } +#endif /* CONFIG_MLX4_EN_DCB */ + return 0; } @@ -406,14 +417,18 @@ static int mlx4_en_vlan_rx_add_vid(struct net_device *dev, mutex_lock(&mdev->state_lock); if (mdev->device_up && priv->port_up) { err = mlx4_SET_VLAN_FLTR(mdev->dev, priv); - if (err) + if (err) { en_err(priv, "Failed configuring VLAN filter\n"); + goto out; + } } - if (mlx4_register_vlan(mdev->dev, priv->port, vid, &idx)) - en_dbg(HW, priv, "failed adding vlan %d\n", vid); - mutex_unlock(&mdev->state_lock); + err = mlx4_register_vlan(mdev->dev, priv->port, vid, &idx); + if (err) + en_dbg(HW, priv, "Failed adding vlan %d\n", vid); - return 0; +out: + mutex_unlock(&mdev->state_lock); + return err; } static int mlx4_en_vlan_rx_kill_vid(struct net_device *dev, @@ -421,7 +436,7 @@ static int mlx4_en_vlan_rx_kill_vid(struct net_device *dev, { struct mlx4_en_priv *priv = netdev_priv(dev); struct mlx4_en_dev *mdev = priv->mdev; - int err; + int err = 0; en_dbg(HW, priv, "Killing VID:%d\n", vid); @@ -438,7 +453,7 @@ static int mlx4_en_vlan_rx_kill_vid(struct net_device *dev, } mutex_unlock(&mdev->state_lock); - return 0; + return err; } static void mlx4_en_u64_to_mac(unsigned char dst_mac[ETH_ALEN + 2], u64 src_mac) @@ -1197,8 +1212,8 @@ static void mlx4_en_netpoll(struct net_device *dev) struct mlx4_en_cq *cq; int i; - for (i = 0; i < priv->rx_ring_num; i++) { - cq = priv->rx_cq[i]; + for (i = 0; i < priv->tx_ring_num; i++) { + cq = priv->tx_cq[i]; napi_schedule(&cq->napi); } } @@ -1692,10 +1707,9 @@ int mlx4_en_start_port(struct net_device *dev) /* Schedule multicast task to populate multicast list */ queue_work(mdev->workqueue, &priv->rx_mode_task); -#ifdef CONFIG_MLX4_EN_VXLAN if (priv->mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) - vxlan_get_rx_port(dev); -#endif + udp_tunnel_get_rx_info(dev); + priv->port_up = true; netif_tx_start_all_queues(dev); netif_device_attach(dev); @@ -2032,11 +2046,20 @@ err: return -ENOMEM; } +static void mlx4_en_shutdown(struct net_device *dev) +{ + rtnl_lock(); + netif_device_detach(dev); + mlx4_en_close(dev); + rtnl_unlock(); +} void mlx4_en_destroy_netdev(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); struct mlx4_en_dev *mdev = priv->mdev; + bool shutdown = mdev->dev->persist->interface_state & + MLX4_INTERFACE_STATE_SHUTDOWN; en_dbg(DRV, priv, "Destroying netdev on port:%d\n", priv->port); @@ -2044,7 +2067,10 @@ void mlx4_en_destroy_netdev(struct net_device *dev) if (priv->registered) { devlink_port_type_clear(mlx4_get_devlink_port(mdev->dev, priv->port)); - unregister_netdev(dev); + if (shutdown) + mlx4_en_shutdown(dev); + else + unregister_netdev(dev); } if (priv->allocated) @@ -2069,7 +2095,8 @@ void mlx4_en_destroy_netdev(struct net_device *dev) kfree(priv->tx_ring); kfree(priv->tx_cq); - free_netdev(dev); + if (!shutdown) + free_netdev(dev); } static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu) @@ -2342,7 +2369,6 @@ static int mlx4_en_get_phys_port_id(struct net_device *dev, return 0; } -#ifdef CONFIG_MLX4_EN_VXLAN static void mlx4_en_add_vxlan_offloads(struct work_struct *work) { int ret; @@ -2392,15 +2418,19 @@ static void mlx4_en_del_vxlan_offloads(struct work_struct *work) } static void mlx4_en_add_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx4_en_priv *priv = netdev_priv(dev); + __be16 port = ti->port; __be16 current_port; - if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + if (ti->sa_family != AF_INET) return; - if (sa_family == AF_INET6) + if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) return; current_port = priv->vxlan_port; @@ -2415,15 +2445,19 @@ static void mlx4_en_add_vxlan_port(struct net_device *dev, } static void mlx4_en_del_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx4_en_priv *priv = netdev_priv(dev); + __be16 port = ti->port; __be16 current_port; - if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + if (ti->sa_family != AF_INET) return; - if (sa_family == AF_INET6) + if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) return; current_port = priv->vxlan_port; @@ -2447,13 +2481,17 @@ static netdev_features_t mlx4_en_features_check(struct sk_buff *skb, * strip that feature if this is an IPv6 encapsulated frame. */ if (skb->encapsulation && - (skb->ip_summed == CHECKSUM_PARTIAL) && - (ip_hdr(skb)->version != 4)) - features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); + (skb->ip_summed == CHECKSUM_PARTIAL)) { + struct mlx4_en_priv *priv = netdev_priv(dev); + + if (!priv->vxlan_port || + (ip_hdr(skb)->version != 4) || + (udp_hdr(skb)->dest != priv->vxlan_port)) + features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); + } return features; } -#endif static int mlx4_en_set_tx_maxrate(struct net_device *dev, int queue_index, u32 maxrate) { @@ -2506,11 +2544,9 @@ static const struct net_device_ops mlx4_netdev_ops = { .ndo_rx_flow_steer = mlx4_en_filter_rfs, #endif .ndo_get_phys_port_id = mlx4_en_get_phys_port_id, -#ifdef CONFIG_MLX4_EN_VXLAN - .ndo_add_vxlan_port = mlx4_en_add_vxlan_port, - .ndo_del_vxlan_port = mlx4_en_del_vxlan_port, + .ndo_udp_tunnel_add = mlx4_en_add_vxlan_port, + .ndo_udp_tunnel_del = mlx4_en_del_vxlan_port, .ndo_features_check = mlx4_en_features_check, -#endif .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, }; @@ -2544,11 +2580,9 @@ static const struct net_device_ops mlx4_netdev_ops_master = { .ndo_rx_flow_steer = mlx4_en_filter_rfs, #endif .ndo_get_phys_port_id = mlx4_en_get_phys_port_id, -#ifdef CONFIG_MLX4_EN_VXLAN - .ndo_add_vxlan_port = mlx4_en_add_vxlan_port, - .ndo_del_vxlan_port = mlx4_en_del_vxlan_port, + .ndo_udp_tunnel_add = mlx4_en_add_vxlan_port, + .ndo_udp_tunnel_del = mlx4_en_del_vxlan_port, .ndo_features_check = mlx4_en_features_check, -#endif .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, }; @@ -2814,6 +2848,9 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, struct mlx4_en_priv *priv; int i; int err; +#ifdef CONFIG_MLX4_EN_DCB + struct tc_configuration *tc; +#endif dev = alloc_etherdev_mqs(sizeof(struct mlx4_en_priv), MAX_TX_RINGS, MAX_RX_RINGS); @@ -2839,10 +2876,8 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, INIT_WORK(&priv->linkstate_task, mlx4_en_linkstate); INIT_DELAYED_WORK(&priv->stats_task, mlx4_en_do_get_stats); INIT_DELAYED_WORK(&priv->service_task, mlx4_en_service_task); -#ifdef CONFIG_MLX4_EN_VXLAN INIT_WORK(&priv->vxlan_add_task, mlx4_en_add_vxlan_offloads); INIT_WORK(&priv->vxlan_del_task, mlx4_en_del_vxlan_offloads); -#endif #ifdef CONFIG_RFS_ACCEL INIT_LIST_HEAD(&priv->filters); spin_lock_init(&priv->filters_lock); @@ -2882,6 +2917,17 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, priv->msg_enable = MLX4_EN_MSG_LEVEL; #ifdef CONFIG_MLX4_EN_DCB if (!mlx4_is_slave(priv->mdev->dev)) { + priv->cee_params.dcbx_cap = DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_HOST | + DCB_CAP_DCBX_VER_IEEE; + priv->flags |= MLX4_EN_DCB_ENABLED; + priv->cee_params.dcb_cfg.pfc_state = false; + + for (i = 0; i < MLX4_EN_NUM_UP; i++) { + tc = &priv->cee_params.dcb_cfg.tc_config[i]; + tc->dcb_pfc = pfc_disabled; + } + if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) { dev->dcbnl_ops = &mlx4_en_dcbnl_ops; } else { diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index e97094598b2d..f4497cf4d06d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -1128,6 +1128,7 @@ int mlx4_QUERY_PORT(struct mlx4_dev *dev, int port, struct mlx4_port_cap *port_c port_cap->max_pkeys = 1 << (field & 0xf); MLX4_GET(field, outbox, QUERY_PORT_MAX_VL_OFFSET); port_cap->max_vl = field & 0xf; + port_cap->max_tc_eth = field >> 4; MLX4_GET(field, outbox, QUERY_PORT_MAX_MACVLAN_OFFSET); port_cap->log_max_macs = field & 0xf; port_cap->log_max_vlans = field >> 4; diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h index 7ea258af636a..cdbd76f10ced 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw.h @@ -53,6 +53,7 @@ struct mlx4_port_cap { int ib_mtu; int max_port_width; int max_vl; + int max_tc_eth; int max_gids; int max_pkeys; u64 def_mac; diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c index dec77d6f0ac9..7ae1cdad9bf0 100644 --- a/drivers/net/ethernet/mellanox/mlx4/intf.c +++ b/drivers/net/ethernet/mellanox/mlx4/intf.c @@ -147,7 +147,7 @@ int mlx4_do_bond(struct mlx4_dev *dev, bool enable) if (enable) { dev->flags |= MLX4_FLAG_BONDED; } else { - ret = mlx4_virt2phy_port_map(dev, 1, 2); + ret = mlx4_virt2phy_port_map(dev, 1, 2); if (ret) { mlx4_err(dev, "Fail to reset port map\n"); return ret; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 12c77a70abdb..75dd2e3d3059 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -292,6 +292,7 @@ static int _mlx4_dev_port(struct mlx4_dev *dev, int port, dev->caps.pkey_table_len[port] = port_cap->max_pkeys; dev->caps.port_width_cap[port] = port_cap->max_port_width; dev->caps.eth_mtu_cap[port] = port_cap->eth_mtu; + dev->caps.max_tc_eth = port_cap->max_tc_eth; dev->caps.def_mac[port] = port_cap->def_mac; dev->caps.supported_type[port] = port_cap->supported_port_types; dev->caps.suggested_type[port] = port_cap->suggested_type; @@ -2599,7 +2600,7 @@ static int mlx4_setup_hca(struct mlx4_dev *dev) err = mlx4_init_uar_table(dev); if (err) { mlx4_err(dev, "Failed to initialize user access region table, aborting\n"); - return err; + return err; } err = mlx4_uar_alloc(dev, &priv->driver_uar); @@ -3222,6 +3223,7 @@ static int mlx4_load_one(struct pci_dev *pdev, int pci_dev_data, INIT_LIST_HEAD(&priv->pgdir_list); mutex_init(&priv->pgdir_mutex); + spin_lock_init(&priv->cmd.context_lock); INIT_LIST_HEAD(&priv->bf_list); mutex_init(&priv->bf_mutex); @@ -4134,8 +4136,11 @@ static void mlx4_shutdown(struct pci_dev *pdev) mlx4_info(persist->dev, "mlx4_shutdown was called\n"); mutex_lock(&persist->interface_state_mutex); - if (persist->interface_state & MLX4_INTERFACE_STATE_UP) + if (persist->interface_state & MLX4_INTERFACE_STATE_UP) { + /* Notify mlx4 clients that the kernel is being shut down */ + persist->interface_state |= MLX4_INTERFACE_STATE_SHUTDOWN; mlx4_unload_one(pdev); + } mutex_unlock(&persist->interface_state_mutex); } diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c index f2d0920018a5..94b891c118c1 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c @@ -618,8 +618,8 @@ static int remove_promisc_qp(struct mlx4_dev *dev, u8 port, err = mlx4_READ_ENTRY(dev, entry->index, mailbox); - if (err) - goto out_mailbox; + if (err) + goto out_mailbox; members_count = be32_to_cpu(mgm->members_count) & 0xffffff; @@ -657,8 +657,8 @@ static int remove_promisc_qp(struct mlx4_dev *dev, u8 port, err = mlx4_WRITE_ENTRY(dev, entry->index, mailbox); - if (err) - goto out_mailbox; + if (err) + goto out_mailbox; } } } diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 467d47ed2c39..d39bf594abe4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -448,6 +448,27 @@ struct mlx4_en_frag_info { #define MLX4_EN_TC_ETS 7 +enum dcb_pfc_type { + pfc_disabled = 0, + pfc_enabled_full, + pfc_enabled_tx, + pfc_enabled_rx +}; + +struct tc_configuration { + enum dcb_pfc_type dcb_pfc; +}; + +struct mlx4_en_cee_config { + bool pfc_state; + struct tc_configuration tc_config[MLX4_EN_NUM_UP]; +}; + +struct mlx4_en_cee_params { + u8 dcbx_cap; + struct mlx4_en_cee_config dcb_cfg; +}; + #endif struct ethtool_flow_id { @@ -467,6 +488,9 @@ enum { MLX4_EN_FLAG_RX_FILTER_NEEDED = (1 << 3), MLX4_EN_FLAG_FORCE_PROMISC = (1 << 4), MLX4_EN_FLAG_RX_CSUM_NON_TCP_UDP = (1 << 5), +#ifdef CONFIG_MLX4_EN_DCB + MLX4_EN_FLAG_DCB_ENABLED = (1 << 6), +#endif }; #define PORT_BEACON_MAX_LIMIT (65535) @@ -545,10 +569,8 @@ struct mlx4_en_priv { struct work_struct linkstate_task; struct delayed_work stats_task; struct delayed_work service_task; -#ifdef CONFIG_MLX4_EN_VXLAN struct work_struct vxlan_add_task; struct work_struct vxlan_del_task; -#endif struct mlx4_en_perf_stats pstats; struct mlx4_en_pkt_stats pkstats; struct mlx4_en_counter_stats pf_stats; @@ -570,9 +592,11 @@ struct mlx4_en_priv { u32 counter_index; #ifdef CONFIG_MLX4_EN_DCB +#define MLX4_EN_DCB_ENABLED 0x3 struct ieee_ets ets; u16 maxrate[IEEE_8021QAZ_MAX_TCS]; enum dcbnl_cndd_states cndd_state[IEEE_8021QAZ_MAX_TCS]; + struct mlx4_en_cee_params cee_params; #endif #ifdef CONFIG_RFS_ACCEL spinlock_t filters_lock; diff --git a/drivers/net/ethernet/mellanox/mlx4/mr.c b/drivers/net/ethernet/mellanox/mlx4/mr.c index 93195191f45b..395b5463cfd9 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mr.c +++ b/drivers/net/ethernet/mellanox/mlx4/mr.c @@ -248,7 +248,7 @@ static void mlx4_free_mtt_range(struct mlx4_dev *dev, u32 offset, int order) offset, order); return; } - __mlx4_free_mtt_range(dev, offset, order); + __mlx4_free_mtt_range(dev, offset, order); } void mlx4_mtt_cleanup(struct mlx4_dev *dev, struct mlx4_mtt *mtt) diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index 087b23b320cb..3d2095e5c61c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -52,6 +52,7 @@ #define MLX4_FLAG_V_IGNORE_FCS_MASK 0x2 #define MLX4_IGNORE_FCS_MASK 0x1 +#define MLNX4_TX_MAX_NUMBER 8 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table) { @@ -2015,3 +2016,14 @@ out: return ret; } EXPORT_SYMBOL(mlx4_get_module_info); + +int mlx4_max_tc(struct mlx4_dev *dev) +{ + u8 num_tc = dev->caps.max_tc_eth; + + if (!num_tc) + num_tc = MLNX4_TX_MAX_NUMBER; + + return num_tc; +} +EXPORT_SYMBOL(mlx4_max_tc); diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index cd9b2b28df88..8b81114bdc72 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -2372,16 +2372,15 @@ static int mpt_free_res(struct mlx4_dev *dev, int slave, int op, int cmd, __mlx4_mpt_release(dev, index); break; case RES_OP_MAP_ICM: - index = get_param_l(&in_param); - id = index & mpt_mask(dev); - err = mr_res_start_move_to(dev, slave, id, - RES_MPT_RESERVED, &mpt); - if (err) - return err; - - __mlx4_mpt_free_icm(dev, mpt->key); - res_end_move(dev, slave, RES_MPT, id); + index = get_param_l(&in_param); + id = index & mpt_mask(dev); + err = mr_res_start_move_to(dev, slave, id, + RES_MPT_RESERVED, &mpt); + if (err) return err; + + __mlx4_mpt_free_icm(dev, mpt->key); + res_end_move(dev, slave, RES_MPT, id); break; default: err = -EINVAL; @@ -4253,9 +4252,8 @@ int mlx4_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave, (1ULL << MLX4_UPD_QP_PATH_MASK_ETH_SRC_CHECK_MC_LB)) && !(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_UPDATE_QP_SRC_CHECK_LB)) { - mlx4_warn(dev, - "Src check LB for slave %d isn't supported\n", - slave); + mlx4_warn(dev, "Src check LB for slave %d isn't supported\n", + slave); return -ENOTSUPP; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig index 1cf722eba607..aae46884bf93 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig @@ -4,6 +4,7 @@ config MLX5_CORE tristate "Mellanox Technologies ConnectX-4 and Connect-IB core driver" + depends on MAY_USE_DEVLINK depends on PCI default n ---help--- diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 9ea7b583096a..05cc1effc13c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -1,11 +1,13 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ - health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ - mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o fs_counters.o + health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ + mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \ + fs_counters.o rl.o -mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \ - en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ - en_txrx.o en_clock.o vxlan.o en_tc.o en_arfs.o +mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ + en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \ + en_rx.o en_rx_am.o en_txrx.o en_clock.o vxlan.o \ + en_tc.o en_arfs.o en_rep.o en_fs_ethtool.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index dcd2df6518de..0b4986268cc9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -545,6 +545,7 @@ const char *mlx5_command_str(int command) MLX5_COMMAND_STR_CASE(ALLOC_FLOW_COUNTER); MLX5_COMMAND_STR_CASE(DEALLOC_FLOW_COUNTER); MLX5_COMMAND_STR_CASE(QUERY_FLOW_COUNTER); + MLX5_COMMAND_STR_CASE(MODIFY_FLOW_TABLE); default: return "unknown command opcode"; } } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index e8a6c3325b39..1365cdc81838 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -44,6 +44,7 @@ #include <linux/mlx5/vport.h> #include <linux/mlx5/transobj.h> #include <linux/rhashtable.h> +#include <net/switchdev.h> #include "wq.h" #include "mlx5_core.h" #include "en_stats.h" @@ -79,6 +80,7 @@ #define MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ (64 * 1024) #define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC 0x10 +#define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE 0x3 #define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS 0x20 #define MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC 0x10 #define MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS 0x20 @@ -88,6 +90,7 @@ #define MLX5E_LOG_INDIR_RQT_SIZE 0x7 #define MLX5E_INDIR_RQT_SIZE BIT(MLX5E_LOG_INDIR_RQT_SIZE) #define MLX5E_MAX_NUM_CHANNELS (MLX5E_INDIR_RQT_SIZE >> 1) +#define MLX5E_MAX_NUM_SQS (MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC) #define MLX5E_TX_CQ_POLL_BUDGET 128 #define MLX5E_UPDATE_STATS_INTERVAL 200 /* msecs */ #define MLX5E_SQ_BF_BUDGET 16 @@ -143,11 +146,32 @@ struct mlx5e_umr_wqe { struct mlx5_wqe_data_seg data; }; +static const char mlx5e_priv_flags[][ETH_GSTRING_LEN] = { + "rx_cqe_moder", +}; + +enum mlx5e_priv_flag { + MLX5E_PFLAG_RX_CQE_BASED_MODER = (1 << 0), +}; + +#define MLX5E_SET_PRIV_FLAG(priv, pflag, enable) \ + do { \ + if (enable) \ + priv->pflags |= pflag; \ + else \ + priv->pflags &= ~pflag; \ + } while (0) + #ifdef CONFIG_MLX5_CORE_EN_DCB #define MLX5E_MAX_BW_ALLOC 100 /* Max percentage of BW allocation */ #define MLX5E_MIN_BW_ALLOC 1 /* Min percentage of BW allocation */ #endif +struct mlx5e_cq_moder { + u16 usec; + u16 pkts; +}; + struct mlx5e_params { u8 log_sq_size; u8 rq_wq_type; @@ -156,12 +180,11 @@ struct mlx5e_params { u8 log_rq_size; u16 num_channels; u8 num_tc; + u8 rx_cq_period_mode; bool rx_cqe_compress_admin; bool rx_cqe_compress; - u16 rx_cq_moderation_usec; - u16 rx_cq_moderation_pkts; - u16 tx_cq_moderation_usec; - u16 tx_cq_moderation_pkts; + struct mlx5e_cq_moder rx_cq_moderation; + struct mlx5e_cq_moder tx_cq_moderation; u16 min_rx_wqes; bool lro_en; u32 lro_wqe_sz; @@ -173,6 +196,7 @@ struct mlx5e_params { #ifdef CONFIG_MLX5_CORE_EN_DCB struct ieee_ets ets; #endif + bool rx_am_enabled; }; struct mlx5e_tstamp { @@ -191,6 +215,7 @@ struct mlx5e_tstamp { enum { MLX5E_RQ_STATE_POST_WQES_ENABLE, MLX5E_RQ_STATE_UMR_WQE_IN_PROGRESS, + MLX5E_RQ_STATE_AM, }; struct mlx5e_cq { @@ -198,6 +223,7 @@ struct mlx5e_cq { struct mlx5_cqwq wq; /* data path - accessed per napi poll */ + u16 event_ctr; struct napi_struct *napi; struct mlx5_core_cq mcq; struct mlx5e_channel *channel; @@ -225,6 +251,30 @@ struct mlx5e_dma_info { dma_addr_t addr; }; +struct mlx5e_rx_am_stats { + int ppms; /* packets per msec */ + int epms; /* events per msec */ +}; + +struct mlx5e_rx_am_sample { + ktime_t time; + unsigned int pkt_ctr; + u16 event_ctr; +}; + +struct mlx5e_rx_am { /* Adaptive Moderation */ + u8 state; + struct mlx5e_rx_am_stats prev_stats; + struct mlx5e_rx_am_sample start_sample; + struct work_struct work; + u8 profile_ix; + u8 mode; + u8 tune_state; + u8 steps_right; + u8 steps_left; + u8 tired; +}; + struct mlx5e_rq { /* data path */ struct mlx5_wq_ll wq; @@ -245,6 +295,8 @@ struct mlx5e_rq { unsigned long state; int ix; + struct mlx5e_rx_am am; /* Adaptive Moderation */ + /* control */ struct mlx5_wq_ctrl wq_ctrl; u8 wq_type; @@ -354,6 +406,7 @@ struct mlx5e_sq { struct mlx5e_channel *channel; int tc; struct mlx5e_ico_wqe_info *ico_wqe_info; + u32 rate_limit; } ____cacheline_aligned_in_smp; static inline bool mlx5e_sq_has_room_for(struct mlx5e_sq *sq, u16 n) @@ -401,7 +454,7 @@ enum mlx5e_traffic_types { }; enum { - MLX5E_STATE_ASYNC_EVENTS_ENABLE, + MLX5E_STATE_ASYNC_EVENTS_ENABLED, MLX5E_STATE_OPENED, MLX5E_STATE_DESTROYING, }; @@ -491,8 +544,24 @@ enum { MLX5E_ARFS_FT_LEVEL }; +struct mlx5e_ethtool_table { + struct mlx5_flow_table *ft; + int num_rules; +}; + +#define ETHTOOL_NUM_L3_L4_FTS 7 +#define ETHTOOL_NUM_L2_FTS 4 + +struct mlx5e_ethtool_steering { + struct mlx5e_ethtool_table l3_l4_ft[ETHTOOL_NUM_L3_L4_FTS]; + struct mlx5e_ethtool_table l2_ft[ETHTOOL_NUM_L2_FTS]; + struct list_head rules; + int tot_num_rules; +}; + struct mlx5e_flow_steering { struct mlx5_flow_namespace *ns; + struct mlx5e_ethtool_steering ethtool; struct mlx5e_tc_table tc; struct mlx5e_vlan_table vlan; struct mlx5e_l2_table l2; @@ -500,9 +569,15 @@ struct mlx5e_flow_steering { struct mlx5e_arfs_tables arfs; }; -struct mlx5e_direct_tir { - u32 tirn; +struct mlx5e_rqt { u32 rqtn; + bool enabled; +}; + +struct mlx5e_tir { + u32 tirn; + struct mlx5e_rqt rqt; + struct list_head list; }; enum { @@ -510,6 +585,22 @@ enum { MLX5E_NIC_PRIO }; +struct mlx5e_profile { + void (*init)(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, void *ppriv); + void (*cleanup)(struct mlx5e_priv *priv); + int (*init_rx)(struct mlx5e_priv *priv); + void (*cleanup_rx)(struct mlx5e_priv *priv); + int (*init_tx)(struct mlx5e_priv *priv); + void (*cleanup_tx)(struct mlx5e_priv *priv); + void (*enable)(struct mlx5e_priv *priv); + void (*disable)(struct mlx5e_priv *priv); + void (*update_stats)(struct mlx5e_priv *priv); + int (*max_nch)(struct mlx5_core_dev *mdev); + int max_tc; +}; + struct mlx5e_priv { /* priv data path fields - start */ struct mlx5e_sq **txq_to_sq_map; @@ -518,18 +609,15 @@ struct mlx5e_priv { unsigned long state; struct mutex state_lock; /* Protects Interface state */ - struct mlx5_uar cq_uar; - u32 pdn; - u32 tdn; - struct mlx5_core_mkey mkey; struct mlx5_core_mkey umr_mkey; struct mlx5e_rq drop_rq; struct mlx5e_channel **channel; u32 tisn[MLX5E_MAX_NUM_TC]; - u32 indir_rqtn; - u32 indir_tirn[MLX5E_NUM_INDIR_TIRS]; - struct mlx5e_direct_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; + struct mlx5e_rqt indir_rqt; + struct mlx5e_tir indir_tir[MLX5E_NUM_INDIR_TIRS]; + struct mlx5e_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; + u32 tx_rates[MLX5E_MAX_NUM_SQS]; struct mlx5e_flow_steering fs; struct mlx5e_vxlan_db vxlan; @@ -540,11 +628,14 @@ struct mlx5e_priv { struct work_struct set_rx_mode_work; struct delayed_work update_stats_work; + u32 pflags; struct mlx5_core_dev *mdev; struct net_device *netdev; struct mlx5e_stats stats; struct mlx5e_tstamp tstamp; u16 q_counter; + const struct mlx5e_profile *profile; + void *ppriv; }; enum mlx5e_link_mode { @@ -562,6 +653,7 @@ enum mlx5e_link_mode { MLX5E_10GBASE_ER = 14, MLX5E_40GBASE_SR4 = 15, MLX5E_40GBASE_LR4 = 16, + MLX5E_50GBASE_SR2 = 18, MLX5E_100GBASE_CR4 = 20, MLX5E_100GBASE_SR4 = 21, MLX5E_100GBASE_KR4 = 22, @@ -579,6 +671,9 @@ enum mlx5e_link_mode { #define MLX5E_PROT_MASK(link_mode) (1 << link_mode) + +void mlx5e_build_ptys2ethtool_map(void); + void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw); u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, void *accel_priv, select_queue_fallback_t fallback); @@ -612,12 +707,26 @@ void mlx5e_free_rx_fragmented_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi); struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq); +void mlx5e_rx_am(struct mlx5e_rq *rq); +void mlx5e_rx_am_work(struct work_struct *work); +struct mlx5e_cq_moder mlx5e_am_get_def_profile(u8 rx_cq_period_mode); + void mlx5e_update_stats(struct mlx5e_priv *priv); int mlx5e_create_flow_steering(struct mlx5e_priv *priv); void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv); void mlx5e_init_l2_addr(struct mlx5e_priv *priv); void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft); +int mlx5e_ethtool_get_flow(struct mlx5e_priv *priv, struct ethtool_rxnfc *info, + int location); +int mlx5e_ethtool_get_all_flows(struct mlx5e_priv *priv, + struct ethtool_rxnfc *info, u32 *rule_locs); +int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs); +int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, + int location); +void mlx5e_ethtool_init_steering(struct mlx5e_priv *priv); +void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv); void mlx5e_set_rx_mode_work(struct work_struct *work); void mlx5e_fill_hwstamp(struct mlx5e_tstamp *clock, u64 timestamp, @@ -647,6 +756,9 @@ void mlx5e_build_default_indir_rqt(struct mlx5_core_dev *mdev, int num_channels); int mlx5e_get_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed); +void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, + u8 cq_period_mode); + static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq, struct mlx5_wqe_ctrl_seg *ctrl, int bf_sz) { @@ -723,5 +835,39 @@ int mlx5e_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, #endif u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev); +int mlx5e_create_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir, u32 *in, int inlen); +void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir); +int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev); +void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); +int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5_core_dev *mdev); + +struct mlx5_eswitch_rep; +int mlx5e_vport_rep_load(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +void mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +int mlx5e_nic_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep); +void mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv); +void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv); +int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr); + +int mlx5e_create_direct_rqts(struct mlx5e_priv *priv); +void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt); +int mlx5e_create_direct_tirs(struct mlx5e_priv *priv); +void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv); +int mlx5e_create_tises(struct mlx5e_priv *priv); +void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv); +int mlx5e_close(struct net_device *netdev); +int mlx5e_open(struct net_device *netdev); +void mlx5e_update_stats_work(struct work_struct *work); +void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile, void *ppriv); +void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv); +struct rtnl_link_stats64 * +mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats); #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index 3515e78ba68f..a8cb38789774 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -93,14 +93,14 @@ static enum mlx5e_traffic_types arfs_get_tt(enum arfs_type type) static int arfs_disable(struct mlx5e_priv *priv) { struct mlx5_flow_destination dest; - u32 *tirn = priv->indir_tirn; + struct mlx5e_tir *tir = priv->indir_tir; int err = 0; int tt; int i; dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; for (i = 0; i < ARFS_NUM_TYPES; i++) { - dest.tir_num = tirn[i]; + dest.tir_num = tir[i].tirn; tt = arfs_get_tt(i); /* Modify ttc rules destination to bypass the aRFS tables*/ err = mlx5_modify_rule_destination(priv->fs.ttc.rules[tt], @@ -175,15 +175,12 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, { struct arfs_table *arfs_t = &priv->fs.arfs.arfs_tables[type]; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; - u32 *tirn = priv->indir_tirn; - u32 *match_criteria; - u32 *match_value; + struct mlx5e_tir *tir = priv->indir_tir; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto out; @@ -192,24 +189,23 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; switch (type) { case ARFS_IPV4_TCP: - dest.tir_num = tirn[MLX5E_TT_IPV4_TCP]; + dest.tir_num = tir[MLX5E_TT_IPV4_TCP].tirn; break; case ARFS_IPV4_UDP: - dest.tir_num = tirn[MLX5E_TT_IPV4_UDP]; + dest.tir_num = tir[MLX5E_TT_IPV4_UDP].tirn; break; case ARFS_IPV6_TCP: - dest.tir_num = tirn[MLX5E_TT_IPV6_TCP]; + dest.tir_num = tir[MLX5E_TT_IPV6_TCP].tirn; break; case ARFS_IPV6_UDP: - dest.tir_num = tirn[MLX5E_TT_IPV6_UDP]; + dest.tir_num = tir[MLX5E_TT_IPV6_UDP].tirn; break; default: err = -EINVAL; goto out; } - arfs_t->default_rule = mlx5_add_flow_rule(arfs_t->ft.t, match_criteria_enable, - match_criteria, match_value, + arfs_t->default_rule = mlx5_add_flow_rule(arfs_t->ft.t, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); @@ -220,8 +216,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, __func__, type); } out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -475,23 +470,20 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, struct mlx5_flow_rule *rule = NULL; struct mlx5_flow_destination dest; struct arfs_table *arfs_table; - u8 match_criteria_enable = 0; + struct mlx5_flow_spec *spec; struct mlx5_flow_table *ft; - u32 *match_criteria; - u32 *match_value; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto out; } - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype); - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ethertype, ntohs(tuple->etype)); arfs_table = arfs_get_table(arfs, tuple->ip_proto, tuple->etype); if (!arfs_table) { @@ -501,59 +493,58 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, ft = arfs_table->ft.t; if (tuple->ip_proto == IPPROTO_TCP) { - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.tcp_dport); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.tcp_sport); - MLX5_SET(fte_match_param, match_value, outer_headers.tcp_dport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.tcp_dport, ntohs(tuple->dst_port)); - MLX5_SET(fte_match_param, match_value, outer_headers.tcp_sport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.tcp_sport, ntohs(tuple->src_port)); } else { - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.udp_dport); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.udp_sport); - MLX5_SET(fte_match_param, match_value, outer_headers.udp_dport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_dport, ntohs(tuple->dst_port)); - MLX5_SET(fte_match_param, match_value, outer_headers.udp_sport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_sport, ntohs(tuple->src_port)); } if (tuple->etype == htons(ETH_P_IP)) { - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), &tuple->src_ipv4, 4); - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), &tuple->dst_ipv4, 4); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4); } else { - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), &tuple->src_ipv6, 16); - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), &tuple->dst_ipv6, 16); - memset(MLX5_ADDR_OF(fte_match_param, match_criteria, + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), 0xff, 16); - memset(MLX5_ADDR_OF(fte_match_param, match_criteria, + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), 0xff, 16); } dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; dest.tir_num = priv->direct_tir[arfs_rule->rxq].tirn; - rule = mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); if (IS_ERR(rule)) { @@ -563,8 +554,7 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, } out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err ? ERR_PTR(err) : rule; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c new file mode 100644 index 000000000000..673043ccd76c --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -0,0 +1,160 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +/* mlx5e global resources should be placed in this file. + * Global resources are common to all the netdevices crated on the same nic. + */ + +int mlx5e_create_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir, u32 *in, int inlen) +{ + int err; + + err = mlx5_core_create_tir(mdev, in, inlen, &tir->tirn); + if (err) + return err; + + list_add(&tir->list, &mdev->mlx5e_res.td.tirs_list); + + return 0; +} + +void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir) +{ + mlx5_core_destroy_tir(mdev, tir->tirn); + list_del(&tir->list); +} + +static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, + struct mlx5_core_mkey *mkey) +{ + struct mlx5_create_mkey_mbox_in *in; + int err; + + in = mlx5_vzalloc(sizeof(*in)); + if (!in) + return -ENOMEM; + + in->seg.flags = MLX5_PERM_LOCAL_WRITE | + MLX5_PERM_LOCAL_READ | + MLX5_ACCESS_MODE_PA; + in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); + in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); + + err = mlx5_core_create_mkey(mdev, mkey, in, sizeof(*in), NULL, NULL, + NULL); + + kvfree(in); + + return err; +} + +int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev) +{ + struct mlx5e_resources *res = &mdev->mlx5e_res; + int err; + + err = mlx5_alloc_map_uar(mdev, &res->cq_uar, false); + if (err) { + mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err); + return err; + } + + err = mlx5_core_alloc_pd(mdev, &res->pdn); + if (err) { + mlx5_core_err(mdev, "alloc pd failed, %d\n", err); + goto err_unmap_free_uar; + } + + err = mlx5_core_alloc_transport_domain(mdev, &res->td.tdn); + if (err) { + mlx5_core_err(mdev, "alloc td failed, %d\n", err); + goto err_dealloc_pd; + } + + err = mlx5e_create_mkey(mdev, res->pdn, &res->mkey); + if (err) { + mlx5_core_err(mdev, "create mkey failed, %d\n", err); + goto err_dealloc_transport_domain; + } + + INIT_LIST_HEAD(&mdev->mlx5e_res.td.tirs_list); + + return 0; + +err_dealloc_transport_domain: + mlx5_core_dealloc_transport_domain(mdev, res->td.tdn); +err_dealloc_pd: + mlx5_core_dealloc_pd(mdev, res->pdn); +err_unmap_free_uar: + mlx5_unmap_free_uar(mdev, &res->cq_uar); + + return err; +} + +void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) +{ + struct mlx5e_resources *res = &mdev->mlx5e_res; + + mlx5_core_destroy_mkey(mdev, &res->mkey); + mlx5_core_dealloc_transport_domain(mdev, res->td.tdn); + mlx5_core_dealloc_pd(mdev, res->pdn); + mlx5_unmap_free_uar(mdev, &res->cq_uar); +} + +int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5_core_dev *mdev) +{ + struct mlx5e_tir *tir; + void *in; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(modify_tir_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); + + list_for_each_entry(tir, &mdev->mlx5e_res.td.tirs_list, list) { + err = mlx5_core_modify_tir(mdev, tir->tirn, in, inlen); + if (err) + return err; + } + + kvfree(in); + + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index b2db180ae2a5..e6883132b555 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -191,7 +191,6 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5_core_dev *mdev = priv->mdev; - enum mlx5_port_status ps; u8 curr_pfc_en; int ret; @@ -200,14 +199,8 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, if (pfc->pfc_en == curr_pfc_en) return 0; - mlx5_query_port_admin_status(mdev, &ps); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); - ret = mlx5_set_port_pfc(mdev, pfc->pfc_en, pfc->pfc_en); - - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_UP); + mlx5_toggle_port_link(mdev); return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index fc7dcc03b1de..4a3757e60441 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -48,123 +48,85 @@ static void mlx5e_get_drvinfo(struct net_device *dev, sizeof(drvinfo->bus_info)); } -static const struct { - u32 supported; - u32 advertised; +struct ptys2ethtool_config { + __ETHTOOL_DECLARE_LINK_MODE_MASK(supported); + __ETHTOOL_DECLARE_LINK_MODE_MASK(advertised); u32 speed; -} ptys2ethtool_table[MLX5E_LINK_MODES_NUMBER] = { - [MLX5E_1000BASE_CX_SGMII] = { - .supported = SUPPORTED_1000baseKX_Full, - .advertised = ADVERTISED_1000baseKX_Full, - .speed = 1000, - }, - [MLX5E_1000BASE_KX] = { - .supported = SUPPORTED_1000baseKX_Full, - .advertised = ADVERTISED_1000baseKX_Full, - .speed = 1000, - }, - [MLX5E_10GBASE_CX4] = { - .supported = SUPPORTED_10000baseKX4_Full, - .advertised = ADVERTISED_10000baseKX4_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_KX4] = { - .supported = SUPPORTED_10000baseKX4_Full, - .advertised = ADVERTISED_10000baseKX4_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_KR] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_20GBASE_KR2] = { - .supported = SUPPORTED_20000baseKR2_Full, - .advertised = ADVERTISED_20000baseKR2_Full, - .speed = 20000, - }, - [MLX5E_40GBASE_CR4] = { - .supported = SUPPORTED_40000baseCR4_Full, - .advertised = ADVERTISED_40000baseCR4_Full, - .speed = 40000, - }, - [MLX5E_40GBASE_KR4] = { - .supported = SUPPORTED_40000baseKR4_Full, - .advertised = ADVERTISED_40000baseKR4_Full, - .speed = 40000, - }, - [MLX5E_56GBASE_R4] = { - .supported = SUPPORTED_56000baseKR4_Full, - .advertised = ADVERTISED_56000baseKR4_Full, - .speed = 56000, - }, - [MLX5E_10GBASE_CR] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_SR] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_ER] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_40GBASE_SR4] = { - .supported = SUPPORTED_40000baseSR4_Full, - .advertised = ADVERTISED_40000baseSR4_Full, - .speed = 40000, - }, - [MLX5E_40GBASE_LR4] = { - .supported = SUPPORTED_40000baseLR4_Full, - .advertised = ADVERTISED_40000baseLR4_Full, - .speed = 40000, - }, - [MLX5E_100GBASE_CR4] = { - .speed = 100000, - }, - [MLX5E_100GBASE_SR4] = { - .speed = 100000, - }, - [MLX5E_100GBASE_KR4] = { - .speed = 100000, - }, - [MLX5E_100GBASE_LR4] = { - .speed = 100000, - }, - [MLX5E_100BASE_TX] = { - .speed = 100, - }, - [MLX5E_1000BASE_T] = { - .supported = SUPPORTED_1000baseT_Full, - .advertised = ADVERTISED_1000baseT_Full, - .speed = 1000, - }, - [MLX5E_10GBASE_T] = { - .supported = SUPPORTED_10000baseT_Full, - .advertised = ADVERTISED_10000baseT_Full, - .speed = 1000, - }, - [MLX5E_25GBASE_CR] = { - .speed = 25000, - }, - [MLX5E_25GBASE_KR] = { - .speed = 25000, - }, - [MLX5E_25GBASE_SR] = { - .speed = 25000, - }, - [MLX5E_50GBASE_CR2] = { - .speed = 50000, - }, - [MLX5E_50GBASE_KR2] = { - .speed = 50000, - }, }; +static struct ptys2ethtool_config ptys2ethtool_table[MLX5E_LINK_MODES_NUMBER]; + +#define MLX5_BUILD_PTYS2ETHTOOL_CONFIG(reg_, speed_, ...) \ + ({ \ + struct ptys2ethtool_config *cfg; \ + const unsigned int modes[] = { __VA_ARGS__ }; \ + unsigned int i; \ + cfg = &ptys2ethtool_table[reg_]; \ + cfg->speed = speed_; \ + bitmap_zero(cfg->supported, \ + __ETHTOOL_LINK_MODE_MASK_NBITS); \ + bitmap_zero(cfg->advertised, \ + __ETHTOOL_LINK_MODE_MASK_NBITS); \ + for (i = 0 ; i < ARRAY_SIZE(modes) ; ++i) { \ + __set_bit(modes[i], cfg->supported); \ + __set_bit(modes[i], cfg->advertised); \ + } \ + }) + +void mlx5e_build_ptys2ethtool_map(void) +{ + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_1000BASE_CX_SGMII, SPEED_1000, + ETHTOOL_LINK_MODE_1000baseKX_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_1000BASE_KX, SPEED_1000, + ETHTOOL_LINK_MODE_1000baseKX_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_CX4, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_KX4, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_KR, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_20GBASE_KR2, SPEED_20000, + ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_CR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_KR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_56GBASE_R4, SPEED_56000, + ETHTOOL_LINK_MODE_56000baseKR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_CR, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_SR, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_ER, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_SR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_LR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_50GBASE_SR2, SPEED_50000, + ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_CR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_SR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_KR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_LR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_T, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseT_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_25GBASE_CR, SPEED_25000, + ETHTOOL_LINK_MODE_25000baseCR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_25GBASE_KR, SPEED_25000, + ETHTOOL_LINK_MODE_25000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_25GBASE_SR, SPEED_25000, + ETHTOOL_LINK_MODE_25000baseSR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_50GBASE_CR2, SPEED_50000, + ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_50GBASE_KR2, SPEED_50000, + ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT); +} + static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -177,6 +139,18 @@ static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) return err ? 0 : pfc_en_tx | pfc_en_rx; } +static bool mlx5e_query_global_pause_combined(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + u32 rx_pause; + u32 tx_pause; + int err; + + err = mlx5_query_port_pause(mdev, &rx_pause, &tx_pause); + + return err ? false : rx_pause | tx_pause; +} + #define MLX5E_NUM_Q_CNTRS(priv) (NUM_Q_COUNTERS * (!!priv->q_counter)) #define MLX5E_NUM_RQ_STATS(priv) \ (NUM_RQ_STATS * priv->params.num_channels * \ @@ -184,7 +158,9 @@ static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) #define MLX5E_NUM_SQ_STATS(priv) \ (NUM_SQ_STATS * priv->params.num_channels * priv->params.num_tc * \ test_bit(MLX5E_STATE_OPENED, &priv->state)) -#define MLX5E_NUM_PFC_COUNTERS(priv) hweight8(mlx5e_query_pfc_combined(priv)) +#define MLX5E_NUM_PFC_COUNTERS(priv) \ + ((mlx5e_query_global_pause_combined(priv) + hweight8(mlx5e_query_pfc_combined(priv))) * \ + NUM_PPORT_PER_PRIO_PFC_COUNTERS) static int mlx5e_get_sset_count(struct net_device *dev, int sset) { @@ -198,6 +174,8 @@ static int mlx5e_get_sset_count(struct net_device *dev, int sset) MLX5E_NUM_RQ_STATS(priv) + MLX5E_NUM_SQ_STATS(priv) + MLX5E_NUM_PFC_COUNTERS(priv); + case ETH_SS_PRIV_FLAGS: + return ARRAY_SIZE(mlx5e_priv_flags); /* fallthrough */ default: return -EOPNOTSUPP; @@ -211,42 +189,51 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data) /* SW counters */ for (i = 0; i < NUM_SW_COUNTERS; i++) - strcpy(data + (idx++) * ETH_GSTRING_LEN, sw_stats_desc[i].name); + strcpy(data + (idx++) * ETH_GSTRING_LEN, sw_stats_desc[i].format); /* Q counters */ for (i = 0; i < MLX5E_NUM_Q_CNTRS(priv); i++) - strcpy(data + (idx++) * ETH_GSTRING_LEN, q_stats_desc[i].name); + strcpy(data + (idx++) * ETH_GSTRING_LEN, q_stats_desc[i].format); /* VPORT counters */ for (i = 0; i < NUM_VPORT_COUNTERS; i++) strcpy(data + (idx++) * ETH_GSTRING_LEN, - vport_stats_desc[i].name); + vport_stats_desc[i].format); /* PPORT counters */ for (i = 0; i < NUM_PPORT_802_3_COUNTERS; i++) strcpy(data + (idx++) * ETH_GSTRING_LEN, - pport_802_3_stats_desc[i].name); + pport_802_3_stats_desc[i].format); for (i = 0; i < NUM_PPORT_2863_COUNTERS; i++) strcpy(data + (idx++) * ETH_GSTRING_LEN, - pport_2863_stats_desc[i].name); + pport_2863_stats_desc[i].format); for (i = 0; i < NUM_PPORT_2819_COUNTERS; i++) strcpy(data + (idx++) * ETH_GSTRING_LEN, - pport_2819_stats_desc[i].name); + pport_2819_stats_desc[i].format); for (prio = 0; prio < NUM_PPORT_PRIO; prio++) { for (i = 0; i < NUM_PPORT_PER_PRIO_TRAFFIC_COUNTERS; i++) - sprintf(data + (idx++) * ETH_GSTRING_LEN, "prio%d_%s", - prio, - pport_per_prio_traffic_stats_desc[i].name); + sprintf(data + (idx++) * ETH_GSTRING_LEN, + pport_per_prio_traffic_stats_desc[i].format, prio); } pfc_combined = mlx5e_query_pfc_combined(priv); for_each_set_bit(prio, &pfc_combined, NUM_PPORT_PRIO) { for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { - sprintf(data + (idx++) * ETH_GSTRING_LEN, "prio%d_%s", - prio, pport_per_prio_pfc_stats_desc[i].name); + char pfc_string[ETH_GSTRING_LEN]; + + snprintf(pfc_string, sizeof(pfc_string), "prio%d", prio); + sprintf(data + (idx++) * ETH_GSTRING_LEN, + pport_per_prio_pfc_stats_desc[i].format, pfc_string); + } + } + + if (mlx5e_query_global_pause_combined(priv)) { + for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { + sprintf(data + (idx++) * ETH_GSTRING_LEN, + pport_per_prio_pfc_stats_desc[i].format, "global"); } } @@ -256,25 +243,27 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data) /* per channel counters */ for (i = 0; i < priv->params.num_channels; i++) for (j = 0; j < NUM_RQ_STATS; j++) - sprintf(data + (idx++) * ETH_GSTRING_LEN, "rx%d_%s", i, - rq_stats_desc[j].name); + sprintf(data + (idx++) * ETH_GSTRING_LEN, + rq_stats_desc[j].format, i); for (tc = 0; tc < priv->params.num_tc; tc++) for (i = 0; i < priv->params.num_channels; i++) for (j = 0; j < NUM_SQ_STATS; j++) sprintf(data + (idx++) * ETH_GSTRING_LEN, - "tx%d_%s", - priv->channeltc_to_txq_map[i][tc], - sq_stats_desc[j].name); + sq_stats_desc[j].format, + priv->channeltc_to_txq_map[i][tc]); } static void mlx5e_get_strings(struct net_device *dev, uint32_t stringset, uint8_t *data) { struct mlx5e_priv *priv = netdev_priv(dev); + int i; switch (stringset) { case ETH_SS_PRIV_FLAGS: + for (i = 0; i < ARRAY_SIZE(mlx5e_priv_flags); i++) + strcpy(data + i * ETH_GSTRING_LEN, mlx5e_priv_flags[i]); break; case ETH_SS_TEST: @@ -339,6 +328,13 @@ static void mlx5e_get_ethtool_stats(struct net_device *dev, } } + if (mlx5e_query_global_pause_combined(priv)) { + for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { + data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.per_prio_counters[0], + pport_per_prio_pfc_stats_desc, 0); + } + } + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) return; @@ -519,10 +515,11 @@ static int mlx5e_get_coalesce(struct net_device *netdev, if (!MLX5_CAP_GEN(priv->mdev, cq_moderation)) return -ENOTSUPP; - coal->rx_coalesce_usecs = priv->params.rx_cq_moderation_usec; - coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation_pkts; - coal->tx_coalesce_usecs = priv->params.tx_cq_moderation_usec; - coal->tx_max_coalesced_frames = priv->params.tx_cq_moderation_pkts; + coal->rx_coalesce_usecs = priv->params.rx_cq_moderation.usec; + coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation.pkts; + coal->tx_coalesce_usecs = priv->params.tx_cq_moderation.usec; + coal->tx_max_coalesced_frames = priv->params.tx_cq_moderation.pkts; + coal->use_adaptive_rx_coalesce = priv->params.rx_am_enabled; return 0; } @@ -533,6 +530,10 @@ static int mlx5e_set_coalesce(struct net_device *netdev, struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_channel *c; + bool restart = + !!coal->use_adaptive_rx_coalesce != priv->params.rx_am_enabled; + bool was_opened; + int err = 0; int tc; int i; @@ -540,12 +541,19 @@ static int mlx5e_set_coalesce(struct net_device *netdev, return -ENOTSUPP; mutex_lock(&priv->state_lock); - priv->params.tx_cq_moderation_usec = coal->tx_coalesce_usecs; - priv->params.tx_cq_moderation_pkts = coal->tx_max_coalesced_frames; - priv->params.rx_cq_moderation_usec = coal->rx_coalesce_usecs; - priv->params.rx_cq_moderation_pkts = coal->rx_max_coalesced_frames; - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + was_opened = test_bit(MLX5E_STATE_OPENED, &priv->state); + if (was_opened && restart) { + mlx5e_close_locked(netdev); + priv->params.rx_am_enabled = !!coal->use_adaptive_rx_coalesce; + } + + priv->params.tx_cq_moderation.usec = coal->tx_coalesce_usecs; + priv->params.tx_cq_moderation.pkts = coal->tx_max_coalesced_frames; + priv->params.rx_cq_moderation.usec = coal->rx_coalesce_usecs; + priv->params.rx_cq_moderation.pkts = coal->rx_max_coalesced_frames; + + if (!was_opened || restart) goto out; for (i = 0; i < priv->params.num_channels; ++i) { @@ -564,35 +572,37 @@ static int mlx5e_set_coalesce(struct net_device *netdev, } out: + if (was_opened && restart) + err = mlx5e_open_locked(netdev); + mutex_unlock(&priv->state_lock); - return 0; + return err; } -static u32 ptys2ethtool_supported_link(u32 eth_proto_cap) +static void ptys2ethtool_supported_link(unsigned long *supported_modes, + u32 eth_proto_cap) { - int i; - u32 supported_modes = 0; + int proto; - for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { - if (eth_proto_cap & MLX5E_PROT_MASK(i)) - supported_modes |= ptys2ethtool_table[i].supported; - } - return supported_modes; + for_each_set_bit(proto, (unsigned long *)ð_proto_cap, MLX5E_LINK_MODES_NUMBER) + bitmap_or(supported_modes, supported_modes, + ptys2ethtool_table[proto].supported, + __ETHTOOL_LINK_MODE_MASK_NBITS); } -static u32 ptys2ethtool_adver_link(u32 eth_proto_cap) +static void ptys2ethtool_adver_link(unsigned long *advertising_modes, + u32 eth_proto_cap) { - int i; - u32 advertising_modes = 0; + int proto; - for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { - if (eth_proto_cap & MLX5E_PROT_MASK(i)) - advertising_modes |= ptys2ethtool_table[i].advertised; - } - return advertising_modes; + for_each_set_bit(proto, (unsigned long *)ð_proto_cap, MLX5E_LINK_MODES_NUMBER) + bitmap_or(advertising_modes, advertising_modes, + ptys2ethtool_table[proto].advertised, + __ETHTOOL_LINK_MODE_MASK_NBITS); } -static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) +static void ptys2ethtool_supported_port(struct ethtool_link_ksettings *link_ksettings, + u32 eth_proto_cap) { if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_10GBASE_CR) | MLX5E_PROT_MASK(MLX5E_10GBASE_SR) @@ -600,7 +610,7 @@ static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) | MLX5E_PROT_MASK(MLX5E_40GBASE_SR4) | MLX5E_PROT_MASK(MLX5E_100GBASE_SR4) | MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII))) { - return SUPPORTED_FIBRE; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, FIBRE); } if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_100GBASE_KR4) @@ -608,9 +618,8 @@ static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) | MLX5E_PROT_MASK(MLX5E_10GBASE_KR) | MLX5E_PROT_MASK(MLX5E_10GBASE_KX4) | MLX5E_PROT_MASK(MLX5E_1000BASE_KX))) { - return SUPPORTED_Backplane; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Backplane); } - return 0; } int mlx5e_get_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed) @@ -634,7 +643,7 @@ int mlx5e_get_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed) static void get_speed_duplex(struct net_device *netdev, u32 eth_proto_oper, - struct ethtool_cmd *cmd) + struct ethtool_link_ksettings *link_ksettings) { int i; u32 speed = SPEED_UNKNOWN; @@ -651,23 +660,32 @@ static void get_speed_duplex(struct net_device *netdev, } } out: - ethtool_cmd_speed_set(cmd, speed); - cmd->duplex = duplex; + link_ksettings->base.speed = speed; + link_ksettings->base.duplex = duplex; } -static void get_supported(u32 eth_proto_cap, u32 *supported) +static void get_supported(u32 eth_proto_cap, + struct ethtool_link_ksettings *link_ksettings) { - *supported |= ptys2ethtool_supported_port(eth_proto_cap); - *supported |= ptys2ethtool_supported_link(eth_proto_cap); - *supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause; + unsigned long *supported = link_ksettings->link_modes.supported; + + ptys2ethtool_supported_port(link_ksettings, eth_proto_cap); + ptys2ethtool_supported_link(supported, eth_proto_cap); + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Pause); + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Asym_Pause); } static void get_advertising(u32 eth_proto_cap, u8 tx_pause, - u8 rx_pause, u32 *advertising) + u8 rx_pause, + struct ethtool_link_ksettings *link_ksettings) { - *advertising |= ptys2ethtool_adver_link(eth_proto_cap); - *advertising |= tx_pause ? ADVERTISED_Pause : 0; - *advertising |= (tx_pause ^ rx_pause) ? ADVERTISED_Asym_Pause : 0; + unsigned long *advertising = link_ksettings->link_modes.advertising; + + ptys2ethtool_adver_link(advertising, eth_proto_cap); + if (tx_pause) + ethtool_link_ksettings_add_link_mode(link_ksettings, advertising, Pause); + if (tx_pause ^ rx_pause) + ethtool_link_ksettings_add_link_mode(link_ksettings, advertising, Asym_Pause); } static u8 get_connector_port(u32 eth_proto) @@ -695,13 +713,16 @@ static u8 get_connector_port(u32 eth_proto) return PORT_OTHER; } -static void get_lp_advertising(u32 eth_proto_lp, u32 *lp_advertising) +static void get_lp_advertising(u32 eth_proto_lp, + struct ethtool_link_ksettings *link_ksettings) { - *lp_advertising = ptys2ethtool_adver_link(eth_proto_lp); + unsigned long *lp_advertising = link_ksettings->link_modes.lp_advertising; + + ptys2ethtool_adver_link(lp_advertising, eth_proto_lp); } -static int mlx5e_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) +static int mlx5e_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *link_ksettings) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -710,6 +731,8 @@ static int mlx5e_get_settings(struct net_device *netdev, u32 eth_proto_admin; u32 eth_proto_lp; u32 eth_proto_oper; + u8 an_disable_admin; + u8 an_status; int err; err = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_EN, 1); @@ -720,35 +743,49 @@ static int mlx5e_get_settings(struct net_device *netdev, goto err_query_ptys; } - eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); - eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); - eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); - eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); + eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); + eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); + eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); + an_status = MLX5_GET(ptys_reg, out, an_status); - cmd->supported = 0; - cmd->advertising = 0; + ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); + ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); - get_supported(eth_proto_cap, &cmd->supported); - get_advertising(eth_proto_admin, 0, 0, &cmd->advertising); - get_speed_duplex(netdev, eth_proto_oper, cmd); + get_supported(eth_proto_cap, link_ksettings); + get_advertising(eth_proto_admin, 0, 0, link_ksettings); + get_speed_duplex(netdev, eth_proto_oper, link_ksettings); eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap; - cmd->port = get_connector_port(eth_proto_oper); - get_lp_advertising(eth_proto_lp, &cmd->lp_advertising); + link_ksettings->base.port = get_connector_port(eth_proto_oper); + get_lp_advertising(eth_proto_lp, link_ksettings); - cmd->transceiver = XCVR_INTERNAL; + if (an_status == MLX5_AN_COMPLETE) + ethtool_link_ksettings_add_link_mode(link_ksettings, + lp_advertising, Autoneg); + + link_ksettings->base.autoneg = an_disable_admin ? AUTONEG_DISABLE : + AUTONEG_ENABLE; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, + Autoneg); + if (!an_disable_admin) + ethtool_link_ksettings_add_link_mode(link_ksettings, + advertising, Autoneg); err_query_ptys: return err; } -static u32 mlx5e_ethtool2ptys_adver_link(u32 link_modes) +static u32 mlx5e_ethtool2ptys_adver_link(const unsigned long *link_modes) { u32 i, ptys_modes = 0; for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { - if (ptys2ethtool_table[i].advertised & link_modes) + if (bitmap_intersects(ptys2ethtool_table[i].advertised, + link_modes, + __ETHTOOL_LINK_MODE_MASK_NBITS)) ptys_modes |= MLX5E_PROT_MASK(i); } @@ -767,21 +804,25 @@ static u32 mlx5e_ethtool2ptys_speed_link(u32 speed) return speed_links; } -static int mlx5e_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) +static int mlx5e_set_link_ksettings(struct net_device *netdev, + const struct ethtool_link_ksettings *link_ksettings) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; + u32 eth_proto_cap, eth_proto_admin; + bool an_changes = false; + u8 an_disable_admin; + u8 an_disable_cap; + bool an_disable; u32 link_modes; + u8 an_status; u32 speed; - u32 eth_proto_cap, eth_proto_admin; - enum mlx5_port_status ps; int err; - speed = ethtool_cmd_speed(cmd); + speed = link_ksettings->base.speed; - link_modes = cmd->autoneg == AUTONEG_ENABLE ? - mlx5e_ethtool2ptys_adver_link(cmd->advertising) : + link_modes = link_ksettings->base.autoneg == AUTONEG_ENABLE ? + mlx5e_ethtool2ptys_adver_link(link_ksettings->link_modes.advertising) : mlx5e_ethtool2ptys_speed_link(speed); err = mlx5_query_port_proto_cap(mdev, ð_proto_cap, MLX5_PTYS_EN); @@ -806,15 +847,18 @@ static int mlx5e_set_settings(struct net_device *netdev, goto out; } - if (link_modes == eth_proto_admin) + mlx5_query_port_autoneg(mdev, MLX5_PTYS_EN, &an_status, + &an_disable_cap, &an_disable_admin); + + an_disable = link_ksettings->base.autoneg == AUTONEG_DISABLE; + an_changes = ((!an_disable && an_disable_admin) || + (an_disable && !an_disable_admin)); + + if (!an_changes && link_modes == eth_proto_admin) goto out; - mlx5_query_port_admin_status(mdev, &ps); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); - mlx5_set_port_proto(mdev, link_modes, MLX5_PTYS_EN); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_UP); + mlx5_set_port_ptys(mdev, an_disable, link_modes, MLX5_PTYS_EN); + mlx5_toggle_port_link(mdev); out: return err; @@ -861,7 +905,7 @@ static void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen) mlx5e_build_tir_ctx_hash(tirc, priv); for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5_core_modify_tir(mdev, priv->indir_tirn[i], in, inlen); + mlx5_core_modify_tir(mdev, priv->indir_tir[i].tirn, in, inlen); } static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, @@ -883,7 +927,7 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, mutex_lock(&priv->state_lock); if (indir) { - u32 rqtn = priv->indir_rqtn; + u32 rqtn = priv->indir_rqt.rqtn; memcpy(priv->params.indirection_rqt, indir, sizeof(priv->params.indirection_rqt)); @@ -916,6 +960,15 @@ static int mlx5e_get_rxnfc(struct net_device *netdev, case ETHTOOL_GRXRINGS: info->data = priv->params.num_channels; break; + case ETHTOOL_GRXCLSRLCNT: + info->rule_cnt = priv->fs.ethtool.tot_num_rules; + break; + case ETHTOOL_GRXCLSRULE: + err = mlx5e_ethtool_get_flow(priv, info, info->fs.location); + break; + case ETHTOOL_GRXCLSRLALL: + err = mlx5e_ethtool_get_all_flows(priv, info, rule_locs); + break; default: err = -EOPNOTSUPP; break; @@ -1272,6 +1325,107 @@ static int mlx5e_get_module_eeprom(struct net_device *netdev, return 0; } +typedef int (*mlx5e_pflag_handler)(struct net_device *netdev, bool enable); + +static int set_pflag_rx_cqe_based_moder(struct net_device *netdev, bool enable) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + bool rx_mode_changed; + u8 rx_cq_period_mode; + int err = 0; + bool reset; + + rx_cq_period_mode = enable ? + MLX5_CQ_PERIOD_MODE_START_FROM_CQE : + MLX5_CQ_PERIOD_MODE_START_FROM_EQE; + rx_mode_changed = rx_cq_period_mode != priv->params.rx_cq_period_mode; + + if (rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE && + !MLX5_CAP_GEN(mdev, cq_period_start_from_cqe)) + return -ENOTSUPP; + + if (!rx_mode_changed) + return 0; + + reset = test_bit(MLX5E_STATE_OPENED, &priv->state); + if (reset) + mlx5e_close_locked(netdev); + + mlx5e_set_rx_cq_mode_params(&priv->params, rx_cq_period_mode); + + if (reset) + err = mlx5e_open_locked(netdev); + + return err; +} + +static int mlx5e_handle_pflag(struct net_device *netdev, + u32 wanted_flags, + enum mlx5e_priv_flag flag, + mlx5e_pflag_handler pflag_handler) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + bool enable = !!(wanted_flags & flag); + u32 changes = wanted_flags ^ priv->pflags; + int err; + + if (!(changes & flag)) + return 0; + + err = pflag_handler(netdev, enable); + if (err) { + netdev_err(netdev, "%s private flag 0x%x failed err %d\n", + enable ? "Enable" : "Disable", flag, err); + return err; + } + + MLX5E_SET_PRIV_FLAG(priv, flag, enable); + return 0; +} + +static int mlx5e_set_priv_flags(struct net_device *netdev, u32 pflags) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + int err; + + mutex_lock(&priv->state_lock); + + err = mlx5e_handle_pflag(netdev, pflags, + MLX5E_PFLAG_RX_CQE_BASED_MODER, + set_pflag_rx_cqe_based_moder); + + mutex_unlock(&priv->state_lock); + return err ? -EINVAL : 0; +} + +static u32 mlx5e_get_priv_flags(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + return priv->pflags; +} + +static int mlx5e_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +{ + int err = 0; + struct mlx5e_priv *priv = netdev_priv(dev); + + switch (cmd->cmd) { + case ETHTOOL_SRXCLSRLINS: + err = mlx5e_ethtool_flow_replace(priv, &cmd->fs); + break; + case ETHTOOL_SRXCLSRLDEL: + err = mlx5e_ethtool_flow_remove(priv, cmd->fs.location); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + const struct ethtool_ops mlx5e_ethtool_ops = { .get_drvinfo = mlx5e_get_drvinfo, .get_link = ethtool_op_get_link, @@ -1284,13 +1438,14 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .set_channels = mlx5e_set_channels, .get_coalesce = mlx5e_get_coalesce, .set_coalesce = mlx5e_set_coalesce, - .get_settings = mlx5e_get_settings, - .set_settings = mlx5e_set_settings, + .get_link_ksettings = mlx5e_get_link_ksettings, + .set_link_ksettings = mlx5e_set_link_ksettings, .get_rxfh_key_size = mlx5e_get_rxfh_key_size, .get_rxfh_indir_size = mlx5e_get_rxfh_indir_size, .get_rxfh = mlx5e_get_rxfh, .set_rxfh = mlx5e_set_rxfh, .get_rxnfc = mlx5e_get_rxnfc, + .set_rxnfc = mlx5e_set_rxnfc, .get_tunable = mlx5e_get_tunable, .set_tunable = mlx5e_set_tunable, .get_pauseparam = mlx5e_get_pauseparam, @@ -1301,4 +1456,6 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .set_wol = mlx5e_set_wol, .get_module_info = mlx5e_get_module_info, .get_module_eeprom = mlx5e_get_module_eeprom, + .get_priv_flags = mlx5e_get_priv_flags, + .set_priv_flags = mlx5e_set_priv_flags }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index b32740092854..1587a9fd5724 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -156,19 +156,18 @@ enum mlx5e_vlan_rule_type { static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, enum mlx5e_vlan_rule_type rule_type, - u16 vid, u32 *mc, u32 *mv) + u16 vid, struct mlx5_flow_spec *spec) { struct mlx5_flow_table *ft = priv->fs.vlan.ft.t; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; struct mlx5_flow_rule **rule_p; int err = 0; dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest.ft = priv->fs.l2.ft.t; - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); switch (rule_type) { case MLX5E_VLAN_RULE_TYPE_UNTAGGED: @@ -176,17 +175,19 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, break; case MLX5E_VLAN_RULE_TYPE_ANY_VID: rule_p = &priv->fs.vlan.any_vlan_rule; - MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.vlan_tag, 1); break; default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */ rule_p = &priv->fs.vlan.active_vlans_rule[vid]; - MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); - MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid); - MLX5_SET(fte_match_param, mv, outer_headers.first_vid, vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.vlan_tag, 1); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, + vid); break; } - *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + *rule_p = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); @@ -203,27 +204,21 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv, enum mlx5e_vlan_rule_type rule_type, u16 vid) { - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_vlan_rule_out; + return -ENOMEM; } if (rule_type == MLX5E_VLAN_RULE_TYPE_MATCH_VID) mlx5e_vport_context_update_vlans(priv); - err = __mlx5e_add_vlan_rule(priv, rule_type, vid, match_criteria, - match_value); + err = __mlx5e_add_vlan_rule(priv, rule_type, vid, spec); -add_vlan_rule_out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -598,32 +593,27 @@ static struct mlx5_flow_rule *mlx5e_generate_ttc_rule(struct mlx5e_priv *priv, u8 proto) { struct mlx5_flow_rule *rule; - u8 match_criteria_enable = 0; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto out; + return ERR_PTR(-ENOMEM); } if (proto) { - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.ip_protocol); - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, proto); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_protocol); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_protocol, proto); } if (etype) { - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.ethertype); - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, etype); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ethertype, etype); } - rule = mlx5_add_flow_rule(ft, match_criteria_enable, - match_criteria, match_value, + rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, dest); @@ -631,9 +621,8 @@ static struct mlx5_flow_rule *mlx5e_generate_ttc_rule(struct mlx5e_priv *priv, err = PTR_ERR(rule); netdev_err(priv->netdev, "%s: add rule failed\n", __func__); } -out: - kvfree(match_criteria); - kvfree(match_value); + + kvfree(spec); return err ? ERR_PTR(err) : rule; } @@ -655,7 +644,7 @@ static int mlx5e_generate_ttc_table_rules(struct mlx5e_priv *priv) if (tt == MLX5E_TT_ANY) dest.tir_num = priv->direct_tir[0].tirn; else - dest.tir_num = priv->indir_tirn[tt]; + dest.tir_num = priv->indir_tir[tt].tirn; rules[tt] = mlx5e_generate_ttc_rule(priv, ft, &dest, ttc_rules[tt].etype, ttc_rules[tt].proto); @@ -792,24 +781,20 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, { struct mlx5_flow_table *ft = priv->fs.l2.ft.t; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; u8 *mc_dmac; u8 *mv_dmac; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_l2_rule_out; + return -ENOMEM; } - mc_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, + mc_dmac = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dmac_47_16); - mv_dmac = MLX5_ADDR_OF(fte_match_param, match_value, + mv_dmac = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dmac_47_16); dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; @@ -817,13 +802,13 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, switch (type) { case MLX5E_FULLMATCH: - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; eth_broadcast_addr(mc_dmac); ether_addr_copy(mv_dmac, ai->addr); break; case MLX5E_ALLMULTI: - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; mc_dmac[0] = 0x01; mv_dmac[0] = 0x01; break; @@ -832,8 +817,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, break; } - ai->rule = mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, + ai->rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); if (IS_ERR(ai->rule)) { @@ -843,9 +827,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, ai->rule = NULL; } -add_l2_rule_out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -1102,6 +1084,8 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv) goto err_destroy_l2_table; } + mlx5e_ethtool_init_steering(priv); + return 0; err_destroy_l2_table: @@ -1121,4 +1105,5 @@ void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv) mlx5e_destroy_l2_table(priv); mlx5e_destroy_ttc_table(priv); mlx5e_arfs_destroy_tables(priv); + mlx5e_ethtool_cleanup_steering(priv); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c new file mode 100644 index 000000000000..d17c24227900 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -0,0 +1,586 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/mlx5/fs.h> +#include "en.h" + +struct mlx5e_ethtool_rule { + struct list_head list; + struct ethtool_rx_flow_spec flow_spec; + struct mlx5_flow_rule *rule; + struct mlx5e_ethtool_table *eth_ft; +}; + +static void put_flow_table(struct mlx5e_ethtool_table *eth_ft) +{ + if (!--eth_ft->num_rules) { + mlx5_destroy_flow_table(eth_ft->ft); + eth_ft->ft = NULL; + } +} + +#define MLX5E_ETHTOOL_L3_L4_PRIO 0 +#define MLX5E_ETHTOOL_L2_PRIO (MLX5E_ETHTOOL_L3_L4_PRIO + ETHTOOL_NUM_L3_L4_FTS) +#define MLX5E_ETHTOOL_NUM_ENTRIES 64000 +#define MLX5E_ETHTOOL_NUM_GROUPS 10 +static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs, + int num_tuples) +{ + struct mlx5e_ethtool_table *eth_ft; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + int max_tuples; + int table_size; + int prio; + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case TCP_V4_FLOW: + case UDP_V4_FLOW: + max_tuples = ETHTOOL_NUM_L3_L4_FTS; + prio = MLX5E_ETHTOOL_L3_L4_PRIO + (max_tuples - num_tuples); + eth_ft = &priv->fs.ethtool.l3_l4_ft[prio]; + break; + case IP_USER_FLOW: + max_tuples = ETHTOOL_NUM_L3_L4_FTS; + prio = MLX5E_ETHTOOL_L3_L4_PRIO + (max_tuples - num_tuples); + eth_ft = &priv->fs.ethtool.l3_l4_ft[prio]; + break; + case ETHER_FLOW: + max_tuples = ETHTOOL_NUM_L2_FTS; + prio = max_tuples - num_tuples; + eth_ft = &priv->fs.ethtool.l2_ft[prio]; + prio += MLX5E_ETHTOOL_L2_PRIO; + break; + default: + return ERR_PTR(-EINVAL); + } + + eth_ft->num_rules++; + if (eth_ft->ft) + return eth_ft; + + ns = mlx5_get_flow_namespace(priv->mdev, + MLX5_FLOW_NAMESPACE_ETHTOOL); + if (!ns) + return ERR_PTR(-ENOTSUPP); + + table_size = min_t(u32, BIT(MLX5_CAP_FLOWTABLE(priv->mdev, + flow_table_properties_nic_receive.log_max_ft_size)), + MLX5E_ETHTOOL_NUM_ENTRIES); + ft = mlx5_create_auto_grouped_flow_table(ns, prio, + table_size, + MLX5E_ETHTOOL_NUM_GROUPS, 0); + if (IS_ERR(ft)) + return (void *)ft; + + eth_ft->ft = ft; + return eth_ft; +} + +static void mask_spec(u8 *mask, u8 *val, size_t size) +{ + unsigned int i; + + for (i = 0; i < size; i++, mask++, val++) + *((u8 *)val) = *((u8 *)mask) & *((u8 *)val); +} + +static void set_ips(void *outer_headers_v, void *outer_headers_c, __be32 ip4src_m, + __be32 ip4src_v, __be32 ip4dst_m, __be32 ip4dst_v) +{ + if (ip4src_m) { + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + &ip4src_v, sizeof(ip4src_v)); + memset(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + 0xff, sizeof(ip4src_m)); + } + if (ip4dst_m) { + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + &ip4dst_v, sizeof(ip4dst_v)); + memset(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + 0xff, sizeof(ip4dst_m)); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + ethertype, ETH_P_IP); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + ethertype, 0xffff); +} + +static int set_flow_attrs(u32 *match_c, u32 *match_v, + struct ethtool_rx_flow_spec *fs) +{ + void *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers); + void *outer_headers_v = MLX5_ADDR_OF(fte_match_param, match_v, + outer_headers); + u32 flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT); + struct ethtool_tcpip4_spec *l4_mask; + struct ethtool_tcpip4_spec *l4_val; + struct ethtool_usrip4_spec *l3_mask; + struct ethtool_usrip4_spec *l3_val; + struct ethhdr *eth_val; + struct ethhdr *eth_mask; + + switch (flow_type) { + case TCP_V4_FLOW: + l4_mask = &fs->m_u.tcp_ip4_spec; + l4_val = &fs->h_u.tcp_ip4_spec; + set_ips(outer_headers_v, outer_headers_c, l4_mask->ip4src, + l4_val->ip4src, l4_mask->ip4dst, l4_val->ip4dst); + + if (l4_mask->psrc) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_sport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_sport, + ntohs(l4_val->psrc)); + } + if (l4_mask->pdst) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_dport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_dport, + ntohs(l4_val->pdst)); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + IPPROTO_TCP); + break; + case UDP_V4_FLOW: + l4_mask = &fs->m_u.tcp_ip4_spec; + l4_val = &fs->h_u.tcp_ip4_spec; + set_ips(outer_headers_v, outer_headers_c, l4_mask->ip4src, + l4_val->ip4src, l4_mask->ip4dst, l4_val->ip4dst); + + if (l4_mask->psrc) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_sport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_sport, + ntohs(l4_val->psrc)); + } + if (l4_mask->pdst) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_dport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_dport, + ntohs(l4_val->pdst)); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + IPPROTO_UDP); + break; + case IP_USER_FLOW: + l3_mask = &fs->m_u.usr_ip4_spec; + l3_val = &fs->h_u.usr_ip4_spec; + set_ips(outer_headers_v, outer_headers_c, l3_mask->ip4src, + l3_val->ip4src, l3_mask->ip4dst, l3_val->ip4dst); + break; + case ETHER_FLOW: + eth_mask = &fs->m_u.ether_spec; + eth_val = &fs->h_u.ether_spec; + + mask_spec((u8 *)eth_mask, (u8 *)eth_val, sizeof(*eth_mask)); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, smac_47_16), + eth_mask->h_source); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, smac_47_16), + eth_val->h_source); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, dmac_47_16), + eth_mask->h_dest); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, dmac_47_16), + eth_val->h_dest); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ethertype, + ntohs(eth_mask->h_proto)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ethertype, + ntohs(eth_val->h_proto)); + break; + default: + return -EINVAL; + } + + if ((fs->flow_type & FLOW_EXT) && + (fs->m_ext.vlan_tci & cpu_to_be16(VLAN_VID_MASK))) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + first_vid, 0xfff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + first_vid, ntohs(fs->h_ext.vlan_tci)); + } + if (fs->flow_type & FLOW_MAC_EXT && + !is_zero_ether_addr(fs->m_ext.h_dest)) { + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, dmac_47_16), + fs->m_ext.h_dest); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, dmac_47_16), + fs->h_ext.h_dest); + } + + return 0; +} + +static void add_rule_to_list(struct mlx5e_priv *priv, + struct mlx5e_ethtool_rule *rule) +{ + struct mlx5e_ethtool_rule *iter; + struct list_head *head = &priv->fs.ethtool.rules; + + list_for_each_entry(iter, &priv->fs.ethtool.rules, list) { + if (iter->flow_spec.location > rule->flow_spec.location) + break; + head = &iter->list; + } + priv->fs.ethtool.tot_num_rules++; + list_add(&rule->list, head); +} + +static bool outer_header_zero(u32 *match_criteria) +{ + int size = MLX5_ST_SZ_BYTES(fte_match_param); + char *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers); + + return outer_headers_c[0] == 0 && !memcmp(outer_headers_c, + outer_headers_c + 1, + size - 1); +} + +static struct mlx5_flow_rule *add_ethtool_flow_rule(struct mlx5e_priv *priv, + struct mlx5_flow_table *ft, + struct ethtool_rx_flow_spec *fs) +{ + struct mlx5_flow_destination *dst = NULL; + struct mlx5_flow_spec *spec; + struct mlx5_flow_rule *rule; + int err = 0; + u32 action; + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) + return ERR_PTR(-ENOMEM); + err = set_flow_attrs(spec->match_criteria, spec->match_value, + fs); + if (err) + goto free; + + if (fs->ring_cookie == RX_CLS_FLOW_DISC) { + action = MLX5_FLOW_CONTEXT_ACTION_DROP; + } else { + dst = kzalloc(sizeof(*dst), GFP_KERNEL); + if (!dst) { + err = -ENOMEM; + goto free; + } + + dst->type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dst->tir_num = priv->direct_tir[fs->ring_cookie].tirn; + action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + } + + spec->match_criteria_enable = (!outer_header_zero(spec->match_criteria)); + rule = mlx5_add_flow_rule(ft, spec, action, + MLX5_FS_DEFAULT_FLOW_TAG, dst); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + netdev_err(priv->netdev, "%s: failed to add ethtool steering rule: %d\n", + __func__, err); + goto free; + } +free: + kvfree(spec); + kfree(dst); + return err ? ERR_PTR(err) : rule; +} + +static void del_ethtool_rule(struct mlx5e_priv *priv, + struct mlx5e_ethtool_rule *eth_rule) +{ + if (eth_rule->rule) + mlx5_del_flow_rule(eth_rule->rule); + list_del(ð_rule->list); + priv->fs.ethtool.tot_num_rules--; + put_flow_table(eth_rule->eth_ft); + kfree(eth_rule); +} + +static struct mlx5e_ethtool_rule *find_ethtool_rule(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *iter; + + list_for_each_entry(iter, &priv->fs.ethtool.rules, list) { + if (iter->flow_spec.location == location) + return iter; + } + return NULL; +} + +static struct mlx5e_ethtool_rule *get_ethtool_rule(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + + eth_rule = find_ethtool_rule(priv, location); + if (eth_rule) + del_ethtool_rule(priv, eth_rule); + + eth_rule = kzalloc(sizeof(*eth_rule), GFP_KERNEL); + if (!eth_rule) + return ERR_PTR(-ENOMEM); + + add_rule_to_list(priv, eth_rule); + return eth_rule; +} + +#define MAX_NUM_OF_ETHTOOL_RULES BIT(10) + +#define all_ones(field) (field == (__force typeof(field))-1) +#define all_zeros_or_all_ones(field) \ + ((field) == 0 || (field) == (__force typeof(field))-1) + +static int validate_flow(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs) +{ + struct ethtool_tcpip4_spec *l4_mask; + struct ethtool_usrip4_spec *l3_mask; + struct ethhdr *eth_mask; + int num_tuples = 0; + + if (fs->location >= MAX_NUM_OF_ETHTOOL_RULES) + return -EINVAL; + + if (fs->ring_cookie >= priv->params.num_channels && + fs->ring_cookie != RX_CLS_FLOW_DISC) + return -EINVAL; + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case ETHER_FLOW: + eth_mask = &fs->m_u.ether_spec; + if (!is_zero_ether_addr(eth_mask->h_dest)) + num_tuples++; + if (!is_zero_ether_addr(eth_mask->h_source)) + num_tuples++; + if (eth_mask->h_proto) + num_tuples++; + break; + case TCP_V4_FLOW: + case UDP_V4_FLOW: + if (fs->m_u.tcp_ip4_spec.tos) + return -EINVAL; + l4_mask = &fs->m_u.tcp_ip4_spec; + if (l4_mask->ip4src) { + if (!all_ones(l4_mask->ip4src)) + return -EINVAL; + num_tuples++; + } + if (l4_mask->ip4dst) { + if (!all_ones(l4_mask->ip4dst)) + return -EINVAL; + num_tuples++; + } + if (l4_mask->psrc) { + if (!all_ones(l4_mask->psrc)) + return -EINVAL; + num_tuples++; + } + if (l4_mask->pdst) { + if (!all_ones(l4_mask->pdst)) + return -EINVAL; + num_tuples++; + } + /* Flow is TCP/UDP */ + num_tuples++; + break; + case IP_USER_FLOW: + l3_mask = &fs->m_u.usr_ip4_spec; + if (l3_mask->l4_4_bytes || l3_mask->tos || l3_mask->proto || + fs->h_u.usr_ip4_spec.ip_ver != ETH_RX_NFC_IP4) + return -EINVAL; + if (l3_mask->ip4src) { + if (!all_ones(l3_mask->ip4src)) + return -EINVAL; + num_tuples++; + } + if (l3_mask->ip4dst) { + if (!all_ones(l3_mask->ip4dst)) + return -EINVAL; + num_tuples++; + } + /* Flow is IPv4 */ + num_tuples++; + break; + default: + return -EINVAL; + } + if ((fs->flow_type & FLOW_EXT)) { + if (fs->m_ext.vlan_etype || + (fs->m_ext.vlan_tci != cpu_to_be16(VLAN_VID_MASK))) + return -EINVAL; + + if (fs->m_ext.vlan_tci) { + if (be16_to_cpu(fs->h_ext.vlan_tci) >= VLAN_N_VID) + return -EINVAL; + } + num_tuples++; + } + + if (fs->flow_type & FLOW_MAC_EXT && + !is_zero_ether_addr(fs->m_ext.h_dest)) + num_tuples++; + + return num_tuples; +} + +int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs) +{ + struct mlx5e_ethtool_table *eth_ft; + struct mlx5e_ethtool_rule *eth_rule; + struct mlx5_flow_rule *rule; + int num_tuples; + int err; + + num_tuples = validate_flow(priv, fs); + if (num_tuples <= 0) { + netdev_warn(priv->netdev, "%s: flow is not valid\n", __func__); + return -EINVAL; + } + + eth_ft = get_flow_table(priv, fs, num_tuples); + if (IS_ERR(eth_ft)) + return PTR_ERR(eth_ft); + + eth_rule = get_ethtool_rule(priv, fs->location); + if (IS_ERR(eth_rule)) { + put_flow_table(eth_ft); + return PTR_ERR(eth_rule); + } + + eth_rule->flow_spec = *fs; + eth_rule->eth_ft = eth_ft; + if (!eth_ft->ft) { + err = -EINVAL; + goto del_ethtool_rule; + } + rule = add_ethtool_flow_rule(priv, eth_ft->ft, fs); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + goto del_ethtool_rule; + } + + eth_rule->rule = rule; + + return 0; + +del_ethtool_rule: + del_ethtool_rule(priv, eth_rule); + + return err; +} + +int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + int err = 0; + + if (location >= MAX_NUM_OF_ETHTOOL_RULES) + return -ENOSPC; + + eth_rule = find_ethtool_rule(priv, location); + if (!eth_rule) { + err = -ENOENT; + goto out; + } + + del_ethtool_rule(priv, eth_rule); +out: + return err; +} + +int mlx5e_ethtool_get_flow(struct mlx5e_priv *priv, struct ethtool_rxnfc *info, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + + if (location < 0 || location >= MAX_NUM_OF_ETHTOOL_RULES) + return -EINVAL; + + list_for_each_entry(eth_rule, &priv->fs.ethtool.rules, list) { + if (eth_rule->flow_spec.location == location) { + info->fs = eth_rule->flow_spec; + return 0; + } + } + + return -ENOENT; +} + +int mlx5e_ethtool_get_all_flows(struct mlx5e_priv *priv, struct ethtool_rxnfc *info, + u32 *rule_locs) +{ + int location = 0; + int idx = 0; + int err = 0; + + while ((!err || err == -ENOENT) && idx < info->rule_cnt) { + err = mlx5e_ethtool_get_flow(priv, info, location); + if (!err) + rule_locs[idx++] = location; + location++; + } + return err; +} + +void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv) +{ + struct mlx5e_ethtool_rule *iter; + struct mlx5e_ethtool_rule *temp; + + list_for_each_entry_safe(iter, temp, &priv->fs.ethtool.rules, list) + del_ethtool_rule(priv, iter); +} + +void mlx5e_ethtool_init_steering(struct mlx5e_priv *priv) +{ + INIT_LIST_HEAD(&priv->fs.ethtool.rules); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index fd4392999eee..96ec53a6a595 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -40,8 +40,9 @@ #include "vxlan.h" struct mlx5e_rq_param { - u32 rqc[MLX5_ST_SZ_DW(rqc)]; - struct mlx5_wq_param wq; + u32 rqc[MLX5_ST_SZ_DW(rqc)]; + struct mlx5_wq_param wq; + bool am_enabled; }; struct mlx5e_sq_param { @@ -55,6 +56,7 @@ struct mlx5e_cq_param { u32 cqc[MLX5_ST_SZ_DW(cqc)]; struct mlx5_wq_param wq; u16 eq_ix; + u8 cq_period_mode; }; struct mlx5e_channel_param { @@ -105,11 +107,11 @@ static void mlx5e_update_sw_counters(struct mlx5e_priv *priv) s->rx_packets += rq_stats->packets; s->rx_bytes += rq_stats->bytes; - s->lro_packets += rq_stats->lro_packets; - s->lro_bytes += rq_stats->lro_bytes; + s->rx_lro_packets += rq_stats->lro_packets; + s->rx_lro_bytes += rq_stats->lro_bytes; s->rx_csum_none += rq_stats->csum_none; - s->rx_csum_sw += rq_stats->csum_sw; - s->rx_csum_inner += rq_stats->csum_inner; + s->rx_csum_complete += rq_stats->csum_complete; + s->rx_csum_unnecessary_inner += rq_stats->csum_unnecessary_inner; s->rx_wqe_err += rq_stats->wqe_err; s->rx_mpwqe_filler += rq_stats->mpwqe_filler; s->rx_mpwqe_frag += rq_stats->mpwqe_frag; @@ -122,24 +124,23 @@ static void mlx5e_update_sw_counters(struct mlx5e_priv *priv) s->tx_packets += sq_stats->packets; s->tx_bytes += sq_stats->bytes; - s->tso_packets += sq_stats->tso_packets; - s->tso_bytes += sq_stats->tso_bytes; - s->tso_inner_packets += sq_stats->tso_inner_packets; - s->tso_inner_bytes += sq_stats->tso_inner_bytes; + s->tx_tso_packets += sq_stats->tso_packets; + s->tx_tso_bytes += sq_stats->tso_bytes; + s->tx_tso_inner_packets += sq_stats->tso_inner_packets; + s->tx_tso_inner_bytes += sq_stats->tso_inner_bytes; s->tx_queue_stopped += sq_stats->stopped; s->tx_queue_wake += sq_stats->wake; s->tx_queue_dropped += sq_stats->dropped; - s->tx_csum_inner += sq_stats->csum_offload_inner; - tx_offload_none += sq_stats->csum_offload_none; + s->tx_csum_partial_inner += sq_stats->csum_partial_inner; + tx_offload_none += sq_stats->csum_none; } } /* Update calculated offload counters */ - s->tx_csum_offload = s->tx_packets - tx_offload_none - s->tx_csum_inner; - s->rx_csum_good = s->rx_packets - s->rx_csum_none - - s->rx_csum_sw; + s->tx_csum_partial = s->tx_packets - tx_offload_none - s->tx_csum_partial_inner; + s->rx_csum_unnecessary = s->rx_packets - s->rx_csum_none - s->rx_csum_complete; - s->link_down_events = MLX5_GET(ppcnt_reg, + s->link_down_events_phy = MLX5_GET(ppcnt_reg, priv->stats.pport.phy_counters, counter_set.phys_layer_cntrs.link_down_events); } @@ -225,14 +226,14 @@ void mlx5e_update_stats(struct mlx5e_priv *priv) mlx5e_update_sw_counters(priv); } -static void mlx5e_update_stats_work(struct work_struct *work) +void mlx5e_update_stats_work(struct work_struct *work) { struct delayed_work *dwork = to_delayed_work(work); struct mlx5e_priv *priv = container_of(dwork, struct mlx5e_priv, update_stats_work); mutex_lock(&priv->state_lock); if (test_bit(MLX5E_STATE_OPENED, &priv->state)) { - mlx5e_update_stats(priv); + priv->profile->update_stats(priv); queue_delayed_work(priv->wq, dwork, msecs_to_jiffies(MLX5E_UPDATE_STATS_INTERVAL)); } @@ -244,7 +245,7 @@ static void mlx5e_async_event(struct mlx5_core_dev *mdev, void *vpriv, { struct mlx5e_priv *priv = vpriv; - if (!test_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state)) + if (!test_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLED, &priv->state)) return; switch (event) { @@ -260,12 +261,12 @@ static void mlx5e_async_event(struct mlx5_core_dev *mdev, void *vpriv, static void mlx5e_enable_async_events(struct mlx5e_priv *priv) { - set_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state); + set_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLED, &priv->state); } static void mlx5e_disable_async_events(struct mlx5e_priv *priv) { - clear_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLE, &priv->state); + clear_bit(MLX5E_STATE_ASYNC_EVENTS_ENABLED, &priv->state); synchronize_irq(mlx5_get_msix_vec(priv->mdev, MLX5_EQ_VEC_ASYNC)); } @@ -336,6 +337,9 @@ static int mlx5e_create_rq(struct mlx5e_channel *c, wqe->data.byte_count = cpu_to_be32(byte_count); } + INIT_WORK(&rq->am.work, mlx5e_rx_am_work); + rq->am.mode = priv->params.rx_cq_period_mode; + rq->wq_type = priv->params.rq_wq_type; rq->pdev = c->pdev; rq->netdev = c->netdev; @@ -508,6 +512,9 @@ static int mlx5e_open_rq(struct mlx5e_channel *c, if (err) goto err_disable_rq; + if (param->am_enabled) + set_bit(MLX5E_RQ_STATE_AM, &c->rq.state); + set_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state); sq->ico_wqe_info[pi].opcode = MLX5_OPCODE_NOP; @@ -536,6 +543,8 @@ static void mlx5e_close_rq(struct mlx5e_rq *rq) /* avoid destroying rq before mlx5e_poll_rx_cq() is done with it */ napi_synchronize(&rq->channel->napi); + cancel_work_sync(&rq->am.work); + mlx5e_disable_rq(rq); mlx5e_destroy_rq(rq); } @@ -580,7 +589,7 @@ static int mlx5e_create_sq(struct mlx5e_channel *c, void *sqc_wq = MLX5_ADDR_OF(sqc, sqc, wq); int err; - err = mlx5_alloc_map_uar(mdev, &sq->uar, true); + err = mlx5_alloc_map_uar(mdev, &sq->uar, !!MLX5_CAP_GEN(mdev, bf)); if (err) return err; @@ -702,7 +711,8 @@ static int mlx5e_enable_sq(struct mlx5e_sq *sq, struct mlx5e_sq_param *param) return err; } -static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, int next_state) +static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, + int next_state, bool update_rl, int rl_index) { struct mlx5e_channel *c = sq->channel; struct mlx5e_priv *priv = c->priv; @@ -722,6 +732,10 @@ static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, int next_state) MLX5_SET(modify_sq_in, in, sq_state, curr_state); MLX5_SET(sqc, sqc, state, next_state); + if (update_rl && next_state == MLX5_SQC_STATE_RDY) { + MLX5_SET64(modify_sq_in, in, modify_bitmask, 1); + MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, rl_index); + } err = mlx5_core_modify_sq(mdev, sq->sqn, in, inlen); @@ -737,6 +751,8 @@ static void mlx5e_disable_sq(struct mlx5e_sq *sq) struct mlx5_core_dev *mdev = priv->mdev; mlx5_core_destroy_sq(mdev, sq->sqn); + if (sq->rate_limit) + mlx5_rl_remove_rate(mdev, sq->rate_limit); } static int mlx5e_open_sq(struct mlx5e_channel *c, @@ -754,7 +770,8 @@ static int mlx5e_open_sq(struct mlx5e_channel *c, if (err) goto err_destroy_sq; - err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RST, MLX5_SQC_STATE_RDY); + err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RST, MLX5_SQC_STATE_RDY, + false, 0); if (err) goto err_disable_sq; @@ -793,7 +810,8 @@ static void mlx5e_close_sq(struct mlx5e_sq *sq) if (mlx5e_sq_has_room_for(sq, 1)) mlx5e_send_nop(sq, true); - mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, MLX5_SQC_STATE_ERR); + mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, MLX5_SQC_STATE_ERR, + false, 0); } while (sq->cc != sq->pc) /* wait till sq is empty */ @@ -840,7 +858,7 @@ static int mlx5e_create_cq(struct mlx5e_channel *c, mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; mcq->irqn = irqn; - mcq->uar = &priv->cq_uar; + mcq->uar = &mdev->mlx5e_res.cq_uar; for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) { struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i); @@ -887,6 +905,7 @@ static int mlx5e_enable_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used); + MLX5_SET(cqc, cqc, cq_period_mode, param->cq_period_mode); MLX5_SET(cqc, cqc, c_eqn, eqn); MLX5_SET(cqc, cqc, uar_page, mcq->uar->index); MLX5_SET(cqc, cqc, log_page_size, cq->wq_ctrl.buf.page_shift - @@ -916,8 +935,7 @@ static void mlx5e_disable_cq(struct mlx5e_cq *cq) static int mlx5e_open_cq(struct mlx5e_channel *c, struct mlx5e_cq_param *param, struct mlx5e_cq *cq, - u16 moderation_usecs, - u16 moderation_frames) + struct mlx5e_cq_moder moderation) { int err; struct mlx5e_priv *priv = c->priv; @@ -933,8 +951,8 @@ static int mlx5e_open_cq(struct mlx5e_channel *c, if (MLX5_CAP_GEN(mdev, cq_moderation)) mlx5_core_modify_cq_moderation(mdev, &cq->mcq, - moderation_usecs, - moderation_frames); + moderation.usec, + moderation.pkts); return 0; err_destroy_cq: @@ -963,8 +981,7 @@ static int mlx5e_open_tx_cqs(struct mlx5e_channel *c, for (tc = 0; tc < c->num_tc; tc++) { err = mlx5e_open_cq(c, &cparam->tx_cq, &c->sq[tc].cq, - priv->params.tx_cq_moderation_usec, - priv->params.tx_cq_moderation_pkts); + priv->params.tx_cq_moderation); if (err) goto err_close_tx_cqs; } @@ -1019,19 +1036,96 @@ static void mlx5e_build_channeltc_to_txq_map(struct mlx5e_priv *priv, int ix) { int i; - for (i = 0; i < MLX5E_MAX_NUM_TC; i++) + for (i = 0; i < priv->profile->max_tc; i++) priv->channeltc_to_txq_map[ix][i] = ix + i * priv->params.num_channels; } +static int mlx5e_set_sq_maxrate(struct net_device *dev, + struct mlx5e_sq *sq, u32 rate) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_core_dev *mdev = priv->mdev; + u16 rl_index = 0; + int err; + + if (rate == sq->rate_limit) + /* nothing to do */ + return 0; + + if (sq->rate_limit) + /* remove current rl index to free space to next ones */ + mlx5_rl_remove_rate(mdev, sq->rate_limit); + + sq->rate_limit = 0; + + if (rate) { + err = mlx5_rl_add_rate(mdev, rate, &rl_index); + if (err) { + netdev_err(dev, "Failed configuring rate %u: %d\n", + rate, err); + return err; + } + } + + err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, + MLX5_SQC_STATE_RDY, true, rl_index); + if (err) { + netdev_err(dev, "Failed configuring rate %u: %d\n", + rate, err); + /* remove the rate from the table */ + if (rate) + mlx5_rl_remove_rate(mdev, rate); + return err; + } + + sq->rate_limit = rate; + return 0; +} + +static int mlx5e_set_tx_maxrate(struct net_device *dev, int index, u32 rate) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_sq *sq = priv->txq_to_sq_map[index]; + int err = 0; + + if (!mlx5_rl_is_supported(mdev)) { + netdev_err(dev, "Rate limiting is not supported on this device\n"); + return -EINVAL; + } + + /* rate is given in Mb/sec, HW config is in Kb/sec */ + rate = rate << 10; + + /* Check whether rate in valid range, 0 is always valid */ + if (rate && !mlx5_rl_is_in_range(mdev, rate)) { + netdev_err(dev, "TX rate %u, is not in range\n", rate); + return -ERANGE; + } + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + err = mlx5e_set_sq_maxrate(dev, sq, rate); + if (!err) + priv->tx_rates[index] = rate; + mutex_unlock(&priv->state_lock); + + return err; +} + static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_channel_param *cparam, struct mlx5e_channel **cp) { + struct mlx5e_cq_moder icosq_cq_moder = {0, 0}; struct net_device *netdev = priv->netdev; + struct mlx5e_cq_moder rx_cq_profile; int cpu = mlx5e_get_cpu(priv, ix); struct mlx5e_channel *c; + struct mlx5e_sq *sq; int err; + int i; c = kzalloc_node(sizeof(*c), GFP_KERNEL, cpu_to_node(cpu)); if (!c) @@ -1042,14 +1136,19 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->cpu = cpu; c->pdev = &priv->mdev->pdev->dev; c->netdev = priv->netdev; - c->mkey_be = cpu_to_be32(priv->mkey.key); + c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.mkey.key); c->num_tc = priv->params.num_tc; + if (priv->params.rx_am_enabled) + rx_cq_profile = mlx5e_am_get_def_profile(priv->params.rx_cq_period_mode); + else + rx_cq_profile = priv->params.rx_cq_moderation; + mlx5e_build_channeltc_to_txq_map(priv, ix); netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); - err = mlx5e_open_cq(c, &cparam->icosq_cq, &c->icosq.cq, 0, 0); + err = mlx5e_open_cq(c, &cparam->icosq_cq, &c->icosq.cq, icosq_cq_moder); if (err) goto err_napi_del; @@ -1058,8 +1157,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, goto err_close_icosq_cq; err = mlx5e_open_cq(c, &cparam->rx_cq, &c->rq.cq, - priv->params.rx_cq_moderation_usec, - priv->params.rx_cq_moderation_pkts); + rx_cq_profile); if (err) goto err_close_tx_cqs; @@ -1073,6 +1171,16 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, if (err) goto err_close_icosq; + for (i = 0; i < priv->params.num_tc; i++) { + u32 txq_ix = priv->channeltc_to_txq_map[ix][i]; + + if (priv->tx_rates[txq_ix]) { + sq = priv->txq_to_sq_map[txq_ix]; + mlx5e_set_sq_maxrate(priv->netdev, sq, + priv->tx_rates[txq_ix]); + } + } + err = mlx5e_open_rq(c, &cparam->rq, &c->rq); if (err) goto err_close_sqs; @@ -1144,11 +1252,13 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv, MLX5_SET(wq, wq, end_padding_mode, MLX5_WQ_END_PAD_MODE_ALIGN); MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe))); MLX5_SET(wq, wq, log_wq_sz, priv->params.log_rq_size); - MLX5_SET(wq, wq, pd, priv->pdn); + MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); MLX5_SET(rqc, rqc, counter_set_id, priv->q_counter); param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); param->wq.linear = 1; + + param->am_enabled = priv->params.rx_am_enabled; } static void mlx5e_build_drop_rq_param(struct mlx5e_rq_param *param) @@ -1167,7 +1277,7 @@ static void mlx5e_build_sq_param_common(struct mlx5e_priv *priv, void *wq = MLX5_ADDR_OF(sqc, sqc, wq); MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB)); - MLX5_SET(wq, wq, pd, priv->pdn); + MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); } @@ -1189,7 +1299,7 @@ static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, { void *cqc = param->cqc; - MLX5_SET(cqc, cqc, uar_page, priv->cq_uar.index); + MLX5_SET(cqc, cqc, uar_page, priv->mdev->mlx5e_res.cq_uar.index); } static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, @@ -1214,6 +1324,8 @@ static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, } mlx5e_build_common_cq_param(priv, param); + + param->cq_period_mode = priv->params.rx_cq_period_mode; } static void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, @@ -1224,6 +1336,8 @@ static void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, MLX5_SET(cqc, cqc, log_cq_size, priv->params.log_sq_size); mlx5e_build_common_cq_param(priv, param); + + param->cq_period_mode = MLX5_CQ_PERIOD_MODE_START_FROM_EQE; } static void mlx5e_build_ico_cq_param(struct mlx5e_priv *priv, @@ -1235,6 +1349,8 @@ static void mlx5e_build_ico_cq_param(struct mlx5e_priv *priv, MLX5_SET(cqc, cqc, log_cq_size, log_wq_size); mlx5e_build_common_cq_param(priv, param); + + param->cq_period_mode = MLX5_CQ_PERIOD_MODE_START_FROM_EQE; } static void mlx5e_build_icosq_param(struct mlx5e_priv *priv, @@ -1370,7 +1486,8 @@ static void mlx5e_fill_direct_rqt_rqn(struct mlx5e_priv *priv, void *rqtc, MLX5_SET(rqtc, rqtc, rq_num[0], rqn); } -static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, int ix, u32 *rqtn) +static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, + int ix, struct mlx5e_rqt *rqt) { struct mlx5_core_dev *mdev = priv->mdev; void *rqtc; @@ -1393,34 +1510,36 @@ static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, int ix, u32 *rqtn) else mlx5e_fill_direct_rqt_rqn(priv, rqtc, ix); - err = mlx5_core_create_rqt(mdev, in, inlen, rqtn); + err = mlx5_core_create_rqt(mdev, in, inlen, &rqt->rqtn); + if (!err) + rqt->enabled = true; kvfree(in); return err; } -static void mlx5e_destroy_rqt(struct mlx5e_priv *priv, u32 rqtn) +void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt) { - mlx5_core_destroy_rqt(priv->mdev, rqtn); + rqt->enabled = false; + mlx5_core_destroy_rqt(priv->mdev, rqt->rqtn); } -static int mlx5e_create_rqts(struct mlx5e_priv *priv) +static int mlx5e_create_indirect_rqts(struct mlx5e_priv *priv) { - int nch = mlx5e_get_max_num_channels(priv->mdev); - u32 *rqtn; + struct mlx5e_rqt *rqt = &priv->indir_rqt; + + return mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqt); +} + +int mlx5e_create_direct_rqts(struct mlx5e_priv *priv) +{ + struct mlx5e_rqt *rqt; int err; int ix; - /* Indirect RQT */ - rqtn = &priv->indir_rqtn; - err = mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqtn); - if (err) - return err; - - /* Direct RQTs */ - for (ix = 0; ix < nch; ix++) { - rqtn = &priv->direct_tir[ix].rqtn; - err = mlx5e_create_rqt(priv, 1 /*size */, ix, rqtn); + for (ix = 0; ix < priv->profile->max_nch(priv->mdev); ix++) { + rqt = &priv->direct_tir[ix].rqt; + err = mlx5e_create_rqt(priv, 1 /*size */, ix, rqt); if (err) goto err_destroy_rqts; } @@ -1429,24 +1548,11 @@ static int mlx5e_create_rqts(struct mlx5e_priv *priv) err_destroy_rqts: for (ix--; ix >= 0; ix--) - mlx5e_destroy_rqt(priv, priv->direct_tir[ix].rqtn); - - mlx5e_destroy_rqt(priv, priv->indir_rqtn); + mlx5e_destroy_rqt(priv, &priv->direct_tir[ix].rqt); return err; } -static void mlx5e_destroy_rqts(struct mlx5e_priv *priv) -{ - int nch = mlx5e_get_max_num_channels(priv->mdev); - int i; - - for (i = 0; i < nch; i++) - mlx5e_destroy_rqt(priv, priv->direct_tir[i].rqtn); - - mlx5e_destroy_rqt(priv, priv->indir_rqtn); -} - int mlx5e_redirect_rqt(struct mlx5e_priv *priv, u32 rqtn, int sz, int ix) { struct mlx5_core_dev *mdev = priv->mdev; @@ -1482,10 +1588,15 @@ static void mlx5e_redirect_rqts(struct mlx5e_priv *priv) u32 rqtn; int ix; - rqtn = priv->indir_rqtn; - mlx5e_redirect_rqt(priv, rqtn, MLX5E_INDIR_RQT_SIZE, 0); + if (priv->indir_rqt.enabled) { + rqtn = priv->indir_rqt.rqtn; + mlx5e_redirect_rqt(priv, rqtn, MLX5E_INDIR_RQT_SIZE, 0); + } + for (ix = 0; ix < priv->params.num_channels; ix++) { - rqtn = priv->direct_tir[ix].rqtn; + if (!priv->direct_tir[ix].rqt.enabled) + continue; + rqtn = priv->direct_tir[ix].rqt.rqtn; mlx5e_redirect_rqt(priv, rqtn, 1, ix); } } @@ -1545,13 +1656,13 @@ static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv) mlx5e_build_tir_ctx_lro(tirc, priv); for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { - err = mlx5_core_modify_tir(mdev, priv->indir_tirn[tt], in, + err = mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in, inlen); if (err) goto free_in; } - for (ix = 0; ix < mlx5e_get_max_num_channels(mdev); ix++) { + for (ix = 0; ix < priv->profile->max_nch(priv->mdev); ix++) { err = mlx5_core_modify_tir(mdev, priv->direct_tir[ix].tirn, in, inlen); if (err) @@ -1564,40 +1675,6 @@ free_in: return err; } -static int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5e_priv *priv) -{ - void *in; - int inlen; - int err; - int i; - - inlen = MLX5_ST_SZ_BYTES(modify_tir_in); - in = mlx5_vzalloc(inlen); - if (!in) - return -ENOMEM; - - MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); - - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) { - err = mlx5_core_modify_tir(priv->mdev, priv->indir_tirn[i], in, - inlen); - if (err) - return err; - } - - for (i = 0; i < priv->params.num_channels; i++) { - err = mlx5_core_modify_tir(priv->mdev, - priv->direct_tir[i].tirn, in, - inlen); - if (err) - return err; - } - - kvfree(in); - - return 0; -} - static int mlx5e_set_mtu(struct mlx5e_priv *priv, u16 mtu) { struct mlx5_core_dev *mdev = priv->mdev; @@ -1666,6 +1743,7 @@ static void mlx5e_netdev_set_tcs(struct net_device *netdev) int mlx5e_open_locked(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; int num_txqs; int err; @@ -1688,7 +1766,7 @@ int mlx5e_open_locked(struct net_device *netdev) goto err_clear_state_opened_flag; } - err = mlx5e_refresh_tirs_self_loopback_enable(priv); + err = mlx5e_refresh_tirs_self_loopback_enable(priv->mdev); if (err) { netdev_err(netdev, "%s: mlx5e_refresh_tirs_self_loopback_enable failed, %d\n", __func__, err); @@ -1701,9 +1779,14 @@ int mlx5e_open_locked(struct net_device *netdev) #ifdef CONFIG_RFS_ACCEL priv->netdev->rx_cpu_rmap = priv->mdev->rmap; #endif + if (priv->profile->update_stats) + queue_delayed_work(priv->wq, &priv->update_stats_work, 0); - queue_delayed_work(priv->wq, &priv->update_stats_work, 0); - + if (MLX5_CAP_GEN(mdev, vport_group_manager)) { + err = mlx5e_add_sqs_fwd_rules(priv); + if (err) + goto err_close_channels; + } return 0; err_close_channels: @@ -1713,7 +1796,7 @@ err_clear_state_opened_flag: return err; } -static int mlx5e_open(struct net_device *netdev) +int mlx5e_open(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); int err; @@ -1728,6 +1811,7 @@ static int mlx5e_open(struct net_device *netdev) int mlx5e_close_locked(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; /* May already be CLOSED in case a previous configuration operation * (e.g RX/TX queue size change) that involves close&open failed. @@ -1737,6 +1821,9 @@ int mlx5e_close_locked(struct net_device *netdev) clear_bit(MLX5E_STATE_OPENED, &priv->state); + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + mlx5e_remove_sqs_fwd_rules(priv); + mlx5e_timestamp_cleanup(priv); netif_carrier_off(priv->netdev); mlx5e_redirect_rqts(priv); @@ -1745,7 +1832,7 @@ int mlx5e_close_locked(struct net_device *netdev) return 0; } -static int mlx5e_close(struct net_device *netdev) +int mlx5e_close(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); int err; @@ -1804,7 +1891,7 @@ static int mlx5e_create_drop_cq(struct mlx5e_priv *priv, mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; mcq->irqn = irqn; - mcq->uar = &priv->cq_uar; + mcq->uar = &mdev->mlx5e_res.cq_uar; cq->priv = priv; @@ -1870,7 +1957,7 @@ static int mlx5e_create_tis(struct mlx5e_priv *priv, int tc) memset(in, 0, sizeof(in)); MLX5_SET(tisc, tisc, prio, tc << 1); - MLX5_SET(tisc, tisc, transport_domain, priv->tdn); + MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.td.tdn); return mlx5_core_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]); } @@ -1880,12 +1967,12 @@ static void mlx5e_destroy_tis(struct mlx5e_priv *priv, int tc) mlx5_core_destroy_tis(priv->mdev, priv->tisn[tc]); } -static int mlx5e_create_tises(struct mlx5e_priv *priv) +int mlx5e_create_tises(struct mlx5e_priv *priv) { int err; int tc; - for (tc = 0; tc < MLX5E_MAX_NUM_TC; tc++) { + for (tc = 0; tc < priv->profile->max_tc; tc++) { err = mlx5e_create_tis(priv, tc); if (err) goto err_close_tises; @@ -1900,11 +1987,11 @@ err_close_tises: return err; } -static void mlx5e_destroy_tises(struct mlx5e_priv *priv) +void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv) { int tc; - for (tc = 0; tc < MLX5E_MAX_NUM_TC; tc++) + for (tc = 0; tc < priv->profile->max_tc; tc++) mlx5e_destroy_tis(priv, tc); } @@ -1913,7 +2000,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, { void *hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); - MLX5_SET(tirc, tirc, transport_domain, priv->tdn); + MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); #define MLX5_HASH_IP (MLX5_HASH_FIELD_SEL_SRC_IP |\ MLX5_HASH_FIELD_SEL_DST_IP) @@ -1930,7 +2017,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, mlx5e_build_tir_ctx_lro(tirc, priv); MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT); - MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqtn); + MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqt.rqtn); mlx5e_build_tir_ctx_hash(tirc, priv); switch (tt) { @@ -2020,7 +2107,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, u32 rqtn) { - MLX5_SET(tirc, tirc, transport_domain, priv->tdn); + MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); mlx5e_build_tir_ctx_lro(tirc, priv); @@ -2029,15 +2116,13 @@ static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_INVERTED_XOR8); } -static int mlx5e_create_tirs(struct mlx5e_priv *priv) +static int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv) { - int nch = mlx5e_get_max_num_channels(priv->mdev); + struct mlx5e_tir *tir; void *tirc; int inlen; - u32 *tirn; int err; u32 *in; - int ix; int tt; inlen = MLX5_ST_SZ_BYTES(create_tir_in); @@ -2045,25 +2130,51 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) if (!in) return -ENOMEM; - /* indirect tirs */ for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { memset(in, 0, inlen); - tirn = &priv->indir_tirn[tt]; + tir = &priv->indir_tir[tt]; tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); mlx5e_build_indir_tir_ctx(priv, tirc, tt); - err = mlx5_core_create_tir(priv->mdev, in, inlen, tirn); + err = mlx5e_create_tir(priv->mdev, tir, in, inlen); if (err) goto err_destroy_tirs; } - /* direct tirs */ + kvfree(in); + + return 0; + +err_destroy_tirs: + for (tt--; tt >= 0; tt--) + mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[tt]); + + kvfree(in); + + return err; +} + +int mlx5e_create_direct_tirs(struct mlx5e_priv *priv) +{ + int nch = priv->profile->max_nch(priv->mdev); + struct mlx5e_tir *tir; + void *tirc; + int inlen; + int err; + u32 *in; + int ix; + + inlen = MLX5_ST_SZ_BYTES(create_tir_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + for (ix = 0; ix < nch; ix++) { memset(in, 0, inlen); - tirn = &priv->direct_tir[ix].tirn; + tir = &priv->direct_tir[ix]; tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); mlx5e_build_direct_tir_ctx(priv, tirc, - priv->direct_tir[ix].rqtn); - err = mlx5_core_create_tir(priv->mdev, in, inlen, tirn); + priv->direct_tir[ix].rqt.rqtn); + err = mlx5e_create_tir(priv->mdev, tir, in, inlen); if (err) goto err_destroy_ch_tirs; } @@ -2074,27 +2185,28 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) err_destroy_ch_tirs: for (ix--; ix >= 0; ix--) - mlx5_core_destroy_tir(priv->mdev, priv->direct_tir[ix].tirn); - -err_destroy_tirs: - for (tt--; tt >= 0; tt--) - mlx5_core_destroy_tir(priv->mdev, priv->indir_tirn[tt]); + mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[ix]); kvfree(in); return err; } -static void mlx5e_destroy_tirs(struct mlx5e_priv *priv) +static void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv) { - int nch = mlx5e_get_max_num_channels(priv->mdev); int i; - for (i = 0; i < nch; i++) - mlx5_core_destroy_tir(priv->mdev, priv->direct_tir[i].tirn); - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5_core_destroy_tir(priv->mdev, priv->indir_tirn[i]); + mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]); +} + +void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv) +{ + int nch = priv->profile->max_nch(priv->mdev); + int i; + + for (i = 0; i < nch; i++) + mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[i]); } int mlx5e_modify_rqs_vsd(struct mlx5e_priv *priv, bool vsd) @@ -2168,7 +2280,7 @@ mqprio: return mlx5e_setup_tc(dev, tc->tc); } -static struct rtnl_link_stats64 * +struct rtnl_link_stats64 * mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) { struct mlx5e_priv *priv = netdev_priv(dev); @@ -2520,25 +2632,31 @@ static int mlx5e_get_vf_stats(struct net_device *dev, } static void mlx5e_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx5e_priv *priv = netdev_priv(netdev); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (!mlx5e_vxlan_allowed(priv->mdev)) return; - mlx5e_vxlan_queue_work(priv, sa_family, be16_to_cpu(port), 1); + mlx5e_vxlan_queue_work(priv, ti->sa_family, be16_to_cpu(ti->port), 1); } static void mlx5e_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx5e_priv *priv = netdev_priv(netdev); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (!mlx5e_vxlan_allowed(priv->mdev)) return; - mlx5e_vxlan_queue_work(priv, sa_family, be16_to_cpu(port), 0); + mlx5e_vxlan_queue_work(priv, ti->sa_family, be16_to_cpu(ti->port), 0); } static netdev_features_t mlx5e_vxlan_features_check(struct mlx5e_priv *priv, @@ -2605,6 +2723,7 @@ static const struct net_device_ops mlx5e_netdev_ops_basic = { .ndo_set_features = mlx5e_set_features, .ndo_change_mtu = mlx5e_change_mtu, .ndo_do_ioctl = mlx5e_ioctl, + .ndo_set_tx_maxrate = mlx5e_set_tx_maxrate, #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = mlx5e_rx_flow_steer, #endif @@ -2624,8 +2743,9 @@ static const struct net_device_ops mlx5e_netdev_ops_sriov = { .ndo_set_features = mlx5e_set_features, .ndo_change_mtu = mlx5e_change_mtu, .ndo_do_ioctl = mlx5e_ioctl, - .ndo_add_vxlan_port = mlx5e_add_vxlan_port, - .ndo_del_vxlan_port = mlx5e_del_vxlan_port, + .ndo_udp_tunnel_add = mlx5e_add_vxlan_port, + .ndo_udp_tunnel_del = mlx5e_del_vxlan_port, + .ndo_set_tx_maxrate = mlx5e_set_tx_maxrate, .ndo_features_check = mlx5e_features_check, #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = mlx5e_rx_flow_steer, @@ -2754,13 +2874,31 @@ static bool cqe_compress_heuristic(u32 link_speed, u32 pci_bw) (pci_bw < 40000) && (pci_bw < link_speed)); } -static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, - struct net_device *netdev, - int num_channels) +void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) +{ + params->rx_cq_period_mode = cq_period_mode; + + params->rx_cq_moderation.pkts = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS; + params->rx_cq_moderation.usec = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC; + + if (cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE) + params->rx_cq_moderation.usec = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE; +} + +static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, + void *ppriv) { struct mlx5e_priv *priv = netdev_priv(netdev); u32 link_speed = 0; u32 pci_bw = 0; + u8 cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? + MLX5_CQ_PERIOD_MODE_START_FROM_CQE : + MLX5_CQ_PERIOD_MODE_START_FROM_EQE; priv->params.log_sq_size = MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE; @@ -2806,13 +2944,13 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, priv->params.min_rx_wqes = mlx5_min_rx_wqes(priv->params.rq_wq_type, BIT(priv->params.log_rq_size)); - priv->params.rx_cq_moderation_usec = - MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC; - priv->params.rx_cq_moderation_pkts = - MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS; - priv->params.tx_cq_moderation_usec = + + priv->params.rx_am_enabled = MLX5_CAP_GEN(mdev, cq_moderation); + mlx5e_set_rx_cq_mode_params(&priv->params, cq_period_mode); + + priv->params.tx_cq_moderation.usec = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC; - priv->params.tx_cq_moderation_pkts = + priv->params.tx_cq_moderation.pkts = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS; priv->params.tx_max_inline = mlx5e_get_max_inline_cap(mdev); priv->params.num_tc = 1; @@ -2822,14 +2960,20 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, sizeof(priv->params.toeplitz_hash_key)); mlx5e_build_default_indir_rqt(mdev, priv->params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, num_channels); + MLX5E_INDIR_RQT_SIZE, profile->max_nch(mdev)); priv->params.lro_wqe_sz = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + /* Initialize pflags */ + MLX5E_SET_PRIV_FLAG(priv, MLX5E_PFLAG_RX_CQE_BASED_MODER, + priv->params.rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE); + priv->mdev = mdev; priv->netdev = netdev; - priv->params.num_channels = num_channels; + priv->params.num_channels = profile->max_nch(mdev); + priv->profile = profile; + priv->ppriv = ppriv; #ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_ets_init(priv); @@ -2854,7 +2998,11 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev) } } -static void mlx5e_build_netdev(struct net_device *netdev) +static const struct switchdev_ops mlx5e_switchdev_ops = { + .switchdev_port_attr_get = mlx5e_attr_get, +}; + +static void mlx5e_build_nic_netdev(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -2935,31 +3083,11 @@ static void mlx5e_build_netdev(struct net_device *netdev) netdev->priv_flags |= IFF_UNICAST_FLT; mlx5e_set_netdev_dev_addr(netdev); -} - -static int mlx5e_create_mkey(struct mlx5e_priv *priv, u32 pdn, - struct mlx5_core_mkey *mkey) -{ - struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5_create_mkey_mbox_in *in; - int err; - - in = mlx5_vzalloc(sizeof(*in)); - if (!in) - return -ENOMEM; - in->seg.flags = MLX5_PERM_LOCAL_WRITE | - MLX5_PERM_LOCAL_READ | - MLX5_ACCESS_MODE_PA; - in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); - in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - - err = mlx5_core_create_mkey(mdev, mkey, in, sizeof(*in), NULL, NULL, - NULL); - - kvfree(in); - - return err; +#ifdef CONFIG_NET_SWITCHDEV + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + netdev->switchdev_ops = &mlx5e_switchdev_ops; +#endif } static void mlx5e_create_q_counter(struct mlx5e_priv *priv) @@ -2989,7 +3117,7 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) struct mlx5_mkey_seg *mkc; int inlen = sizeof(*in); u64 npages = - mlx5e_get_max_num_channels(mdev) * MLX5_CHANNEL_MAX_NUM_MTTS; + priv->profile->max_nch(mdev) * MLX5_CHANNEL_MAX_NUM_MTTS; int err; in = mlx5_vzalloc(inlen); @@ -3004,7 +3132,7 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) MLX5_ACCESS_MODE_MTT; mkc->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - mkc->flags_pd = cpu_to_be32(priv->pdn); + mkc->flags_pd = cpu_to_be32(mdev->mlx5e_res.pdn); mkc->len = cpu_to_be64(npages << PAGE_SHIFT); mkc->xlt_oct_size = cpu_to_be32(mlx5e_get_mtt_octw(npages)); mkc->log2_page_size = PAGE_SHIFT; @@ -3017,160 +3145,233 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) return err; } -static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) +static void mlx5e_nic_init(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, + void *ppriv) { - struct net_device *netdev; - struct mlx5e_priv *priv; - int nch = mlx5e_get_max_num_channels(mdev); - int err; - - if (mlx5e_check_required_hca_cap(mdev)) - return NULL; + struct mlx5e_priv *priv = netdev_priv(netdev); - netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), - nch * MLX5E_MAX_NUM_TC, - nch); - if (!netdev) { - mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n"); - return NULL; - } + mlx5e_build_nic_netdev_priv(mdev, netdev, profile, ppriv); + mlx5e_build_nic_netdev(netdev); + mlx5e_vxlan_init(priv); +} - mlx5e_build_netdev_priv(mdev, netdev, nch); - mlx5e_build_netdev(netdev); +static void mlx5e_nic_cleanup(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_eswitch *esw = mdev->priv.eswitch; - netif_carrier_off(netdev); + mlx5e_vxlan_cleanup(priv); - priv = netdev_priv(netdev); + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + mlx5_eswitch_unregister_vport_rep(esw, 0); +} - priv->wq = create_singlethread_workqueue("mlx5e"); - if (!priv->wq) - goto err_free_netdev; +static int mlx5e_init_nic_rx(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + int err; + int i; - err = mlx5_alloc_map_uar(mdev, &priv->cq_uar, false); + err = mlx5e_create_indirect_rqts(priv); if (err) { - mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err); - goto err_destroy_wq; + mlx5_core_warn(mdev, "create indirect rqts failed, %d\n", err); + return err; } - err = mlx5_core_alloc_pd(mdev, &priv->pdn); + err = mlx5e_create_direct_rqts(priv); if (err) { - mlx5_core_err(mdev, "alloc pd failed, %d\n", err); - goto err_unmap_free_uar; + mlx5_core_warn(mdev, "create direct rqts failed, %d\n", err); + goto err_destroy_indirect_rqts; } - err = mlx5_core_alloc_transport_domain(mdev, &priv->tdn); + err = mlx5e_create_indirect_tirs(priv); if (err) { - mlx5_core_err(mdev, "alloc td failed, %d\n", err); - goto err_dealloc_pd; + mlx5_core_warn(mdev, "create indirect tirs failed, %d\n", err); + goto err_destroy_direct_rqts; } - err = mlx5e_create_mkey(priv, priv->pdn, &priv->mkey); + err = mlx5e_create_direct_tirs(priv); if (err) { - mlx5_core_err(mdev, "create mkey failed, %d\n", err); - goto err_dealloc_transport_domain; + mlx5_core_warn(mdev, "create direct tirs failed, %d\n", err); + goto err_destroy_indirect_tirs; } - err = mlx5e_create_umr_mkey(priv); + err = mlx5e_create_flow_steering(priv); if (err) { - mlx5_core_err(mdev, "create umr mkey failed, %d\n", err); - goto err_destroy_mkey; + mlx5_core_warn(mdev, "create flow steering failed, %d\n", err); + goto err_destroy_direct_tirs; } + err = mlx5e_tc_init(priv); + if (err) + goto err_destroy_flow_steering; + + return 0; + +err_destroy_flow_steering: + mlx5e_destroy_flow_steering(priv); +err_destroy_direct_tirs: + mlx5e_destroy_direct_tirs(priv); +err_destroy_indirect_tirs: + mlx5e_destroy_indirect_tirs(priv); +err_destroy_direct_rqts: + for (i = 0; i < priv->profile->max_nch(mdev); i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); +err_destroy_indirect_rqts: + mlx5e_destroy_rqt(priv, &priv->indir_rqt); + return err; +} + +static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv) +{ + int i; + + mlx5e_tc_cleanup(priv); + mlx5e_destroy_flow_steering(priv); + mlx5e_destroy_direct_tirs(priv); + mlx5e_destroy_indirect_tirs(priv); + for (i = 0; i < priv->profile->max_nch(priv->mdev); i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); + mlx5e_destroy_rqt(priv, &priv->indir_rqt); +} + +static int mlx5e_init_nic_tx(struct mlx5e_priv *priv) +{ + int err; + err = mlx5e_create_tises(priv); if (err) { - mlx5_core_warn(mdev, "create tises failed, %d\n", err); - goto err_destroy_umr_mkey; + mlx5_core_warn(priv->mdev, "create tises failed, %d\n", err); + return err; } - err = mlx5e_open_drop_rq(priv); - if (err) { - mlx5_core_err(mdev, "open drop rq failed, %d\n", err); - goto err_destroy_tises; +#ifdef CONFIG_MLX5_CORE_EN_DCB + mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets); +#endif + return 0; +} + +static void mlx5e_nic_enable(struct mlx5e_priv *priv) +{ + struct net_device *netdev = priv->netdev; + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_eswitch *esw = mdev->priv.eswitch; + struct mlx5_eswitch_rep rep; + + if (mlx5e_vxlan_allowed(mdev)) { + rtnl_lock(); + udp_tunnel_get_rx_info(netdev); + rtnl_unlock(); } - err = mlx5e_create_rqts(priv); - if (err) { - mlx5_core_warn(mdev, "create rqts failed, %d\n", err); - goto err_close_drop_rq; + mlx5e_enable_async_events(priv); + queue_work(priv->wq, &priv->set_rx_mode_work); + + if (MLX5_CAP_GEN(mdev, vport_group_manager)) { + rep.load = mlx5e_nic_rep_load; + rep.unload = mlx5e_nic_rep_unload; + rep.vport = 0; + rep.priv_data = priv; + mlx5_eswitch_register_vport_rep(esw, &rep); } +} - err = mlx5e_create_tirs(priv); - if (err) { - mlx5_core_warn(mdev, "create tirs failed, %d\n", err); - goto err_destroy_rqts; +static void mlx5e_nic_disable(struct mlx5e_priv *priv) +{ + queue_work(priv->wq, &priv->set_rx_mode_work); + mlx5e_disable_async_events(priv); +} + +static const struct mlx5e_profile mlx5e_nic_profile = { + .init = mlx5e_nic_init, + .cleanup = mlx5e_nic_cleanup, + .init_rx = mlx5e_init_nic_rx, + .cleanup_rx = mlx5e_cleanup_nic_rx, + .init_tx = mlx5e_init_nic_tx, + .cleanup_tx = mlx5e_cleanup_nic_tx, + .enable = mlx5e_nic_enable, + .disable = mlx5e_nic_disable, + .update_stats = mlx5e_update_stats, + .max_nch = mlx5e_get_max_num_channels, + .max_tc = MLX5E_MAX_NUM_TC, +}; + +void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile, void *ppriv) +{ + struct net_device *netdev; + struct mlx5e_priv *priv; + int nch = profile->max_nch(mdev); + int err; + + netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), + nch * profile->max_tc, + nch); + if (!netdev) { + mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n"); + return NULL; } - err = mlx5e_create_flow_steering(priv); + profile->init(mdev, netdev, profile, ppriv); + + netif_carrier_off(netdev); + + priv = netdev_priv(netdev); + + priv->wq = create_singlethread_workqueue("mlx5e"); + if (!priv->wq) + goto err_free_netdev; + + err = mlx5e_create_umr_mkey(priv); if (err) { - mlx5_core_warn(mdev, "create flow steering failed, %d\n", err); - goto err_destroy_tirs; + mlx5_core_err(mdev, "create umr mkey failed, %d\n", err); + goto err_destroy_wq; } - mlx5e_create_q_counter(priv); - - mlx5e_init_l2_addr(priv); + err = profile->init_tx(priv); + if (err) + goto err_destroy_umr_mkey; - mlx5e_vxlan_init(priv); + err = mlx5e_open_drop_rq(priv); + if (err) { + mlx5_core_err(mdev, "open drop rq failed, %d\n", err); + goto err_cleanup_tx; + } - err = mlx5e_tc_init(priv); + err = profile->init_rx(priv); if (err) - goto err_dealloc_q_counters; + goto err_close_drop_rq; -#ifdef CONFIG_MLX5_CORE_EN_DCB - mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets); -#endif + mlx5e_create_q_counter(priv); + + mlx5e_init_l2_addr(priv); err = register_netdev(netdev); if (err) { mlx5_core_err(mdev, "register_netdev failed, %d\n", err); - goto err_tc_cleanup; - } - - if (mlx5e_vxlan_allowed(mdev)) { - rtnl_lock(); - vxlan_get_rx_port(netdev); - rtnl_unlock(); + goto err_dealloc_q_counters; } - mlx5e_enable_async_events(priv); - queue_work(priv->wq, &priv->set_rx_mode_work); + if (profile->enable) + profile->enable(priv); return priv; -err_tc_cleanup: - mlx5e_tc_cleanup(priv); - err_dealloc_q_counters: mlx5e_destroy_q_counter(priv); - mlx5e_destroy_flow_steering(priv); - -err_destroy_tirs: - mlx5e_destroy_tirs(priv); - -err_destroy_rqts: - mlx5e_destroy_rqts(priv); + profile->cleanup_rx(priv); err_close_drop_rq: mlx5e_close_drop_rq(priv); -err_destroy_tises: - mlx5e_destroy_tises(priv); +err_cleanup_tx: + profile->cleanup_tx(priv); err_destroy_umr_mkey: mlx5_core_destroy_mkey(mdev, &priv->umr_mkey); -err_destroy_mkey: - mlx5_core_destroy_mkey(mdev, &priv->mkey); - -err_dealloc_transport_domain: - mlx5_core_dealloc_transport_domain(mdev, priv->tdn); - -err_dealloc_pd: - mlx5_core_dealloc_pd(mdev, priv->pdn); - -err_unmap_free_uar: - mlx5_unmap_free_uar(mdev, &priv->cq_uar); - err_destroy_wq: destroy_workqueue(priv->wq); @@ -3180,46 +3381,96 @@ err_free_netdev: return NULL; } -static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) +static void mlx5e_register_vport_rep(struct mlx5_core_dev *mdev) { - struct mlx5e_priv *priv = vpriv; + struct mlx5_eswitch *esw = mdev->priv.eswitch; + int total_vfs = MLX5_TOTAL_VPORTS(mdev); + int vport; + + if (!MLX5_CAP_GEN(mdev, vport_group_manager)) + return; + + for (vport = 1; vport < total_vfs; vport++) { + struct mlx5_eswitch_rep rep; + + rep.load = mlx5e_vport_rep_load; + rep.unload = mlx5e_vport_rep_unload; + rep.vport = vport; + mlx5_eswitch_register_vport_rep(esw, &rep); + } +} + +static void *mlx5e_add(struct mlx5_core_dev *mdev) +{ + struct mlx5_eswitch *esw = mdev->priv.eswitch; + void *ppriv = NULL; + void *ret; + + if (mlx5e_check_required_hca_cap(mdev)) + return NULL; + + if (mlx5e_create_mdev_resources(mdev)) + return NULL; + + mlx5e_register_vport_rep(mdev); + + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + ppriv = &esw->offloads.vport_reps[0]; + + ret = mlx5e_create_netdev(mdev, &mlx5e_nic_profile, ppriv); + if (!ret) { + mlx5e_destroy_mdev_resources(mdev); + return NULL; + } + return ret; +} + +void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv) +{ + const struct mlx5e_profile *profile = priv->profile; struct net_device *netdev = priv->netdev; set_bit(MLX5E_STATE_DESTROYING, &priv->state); + if (profile->disable) + profile->disable(priv); - queue_work(priv->wq, &priv->set_rx_mode_work); - mlx5e_disable_async_events(priv); flush_workqueue(priv->wq); if (test_bit(MLX5_INTERFACE_STATE_SHUTDOWN, &mdev->intf_state)) { netif_device_detach(netdev); - mutex_lock(&priv->state_lock); - if (test_bit(MLX5E_STATE_OPENED, &priv->state)) - mlx5e_close_locked(netdev); - mutex_unlock(&priv->state_lock); + mlx5e_close(netdev); } else { unregister_netdev(netdev); } - mlx5e_tc_cleanup(priv); - mlx5e_vxlan_cleanup(priv); mlx5e_destroy_q_counter(priv); - mlx5e_destroy_flow_steering(priv); - mlx5e_destroy_tirs(priv); - mlx5e_destroy_rqts(priv); + profile->cleanup_rx(priv); mlx5e_close_drop_rq(priv); - mlx5e_destroy_tises(priv); + profile->cleanup_tx(priv); mlx5_core_destroy_mkey(priv->mdev, &priv->umr_mkey); - mlx5_core_destroy_mkey(priv->mdev, &priv->mkey); - mlx5_core_dealloc_transport_domain(priv->mdev, priv->tdn); - mlx5_core_dealloc_pd(priv->mdev, priv->pdn); - mlx5_unmap_free_uar(priv->mdev, &priv->cq_uar); cancel_delayed_work_sync(&priv->update_stats_work); destroy_workqueue(priv->wq); + if (profile->cleanup) + profile->cleanup(priv); if (!test_bit(MLX5_INTERFACE_STATE_SHUTDOWN, &mdev->intf_state)) free_netdev(netdev); } +static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) +{ + struct mlx5_eswitch *esw = mdev->priv.eswitch; + int total_vfs = MLX5_TOTAL_VPORTS(mdev); + struct mlx5e_priv *priv = vpriv; + int vport; + + mlx5e_destroy_netdev(mdev, priv); + + for (vport = 1; vport < total_vfs; vport++) + mlx5_eswitch_unregister_vport_rep(esw, vport); + + mlx5e_destroy_mdev_resources(mdev); +} + static void *mlx5e_get_netdev(void *vpriv) { struct mlx5e_priv *priv = vpriv; @@ -3228,8 +3479,8 @@ static void *mlx5e_get_netdev(void *vpriv) } static struct mlx5_interface mlx5e_interface = { - .add = mlx5e_create_netdev, - .remove = mlx5e_destroy_netdev, + .add = mlx5e_add, + .remove = mlx5e_remove, .event = mlx5e_async_event, .protocol = MLX5_INTERFACE_PROTOCOL_ETH, .get_dev = mlx5e_get_netdev, @@ -3237,6 +3488,7 @@ static struct mlx5_interface mlx5e_interface = { void mlx5e_init(void) { + mlx5e_build_ptys2ethtool_map(); mlx5_register_interface(&mlx5e_interface); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c new file mode 100644 index 000000000000..5ef02f02a1d5 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -0,0 +1,394 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <generated/utsrelease.h> +#include <linux/mlx5/fs.h> +#include <net/switchdev.h> + +#include "eswitch.h" +#include "en.h" + +static const char mlx5e_rep_driver_name[] = "mlx5e_rep"; + +static void mlx5e_rep_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *drvinfo) +{ + strlcpy(drvinfo->driver, mlx5e_rep_driver_name, + sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version)); +} + +static const struct counter_desc sw_rep_stats_desc[] = { + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_bytes) }, +}; + +#define NUM_VPORT_REP_COUNTERS ARRAY_SIZE(sw_rep_stats_desc) + +static void mlx5e_rep_get_strings(struct net_device *dev, + u32 stringset, uint8_t *data) +{ + int i; + + switch (stringset) { + case ETH_SS_STATS: + for (i = 0; i < NUM_VPORT_REP_COUNTERS; i++) + strcpy(data + (i * ETH_GSTRING_LEN), + sw_rep_stats_desc[i].format); + break; + } +} + +static void mlx5e_update_sw_rep_counters(struct mlx5e_priv *priv) +{ + struct mlx5e_sw_stats *s = &priv->stats.sw; + struct mlx5e_rq_stats *rq_stats; + struct mlx5e_sq_stats *sq_stats; + int i, j; + + memset(s, 0, sizeof(*s)); + for (i = 0; i < priv->params.num_channels; i++) { + rq_stats = &priv->channel[i]->rq.stats; + + s->rx_packets += rq_stats->packets; + s->rx_bytes += rq_stats->bytes; + + for (j = 0; j < priv->params.num_tc; j++) { + sq_stats = &priv->channel[i]->sq[j].stats; + + s->tx_packets += sq_stats->packets; + s->tx_bytes += sq_stats->bytes; + } + } +} + +static void mlx5e_rep_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int i; + + if (!data) + return; + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_update_sw_rep_counters(priv); + mutex_unlock(&priv->state_lock); + + for (i = 0; i < NUM_VPORT_REP_COUNTERS; i++) + data[i] = MLX5E_READ_CTR64_CPU(&priv->stats.sw, + sw_rep_stats_desc, i); +} + +static int mlx5e_rep_get_sset_count(struct net_device *dev, int sset) +{ + switch (sset) { + case ETH_SS_STATS: + return NUM_VPORT_REP_COUNTERS; + default: + return -EOPNOTSUPP; + } +} + +static const struct ethtool_ops mlx5e_rep_ethtool_ops = { + .get_drvinfo = mlx5e_rep_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = mlx5e_rep_get_strings, + .get_sset_count = mlx5e_rep_get_sset_count, + .get_ethtool_stats = mlx5e_rep_get_ethtool_stats, +}; + +int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + u8 mac[ETH_ALEN]; + + if (esw->mode == SRIOV_NONE) + return -EOPNOTSUPP; + + switch (attr->id) { + case SWITCHDEV_ATTR_ID_PORT_PARENT_ID: + mlx5_query_nic_vport_mac_address(priv->mdev, 0, mac); + attr->u.ppid.id_len = ETH_ALEN; + memcpy(&attr->u.ppid.id, &mac, ETH_ALEN); + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + +int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv) + +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + struct mlx5e_channel *c; + int n, tc, err, num_sqs = 0; + u16 *sqs; + + sqs = kcalloc(priv->params.num_channels * priv->params.num_tc, sizeof(u16), GFP_KERNEL); + if (!sqs) + return -ENOMEM; + + for (n = 0; n < priv->params.num_channels; n++) { + c = priv->channel[n]; + for (tc = 0; tc < c->num_tc; tc++) + sqs[num_sqs++] = c->sq[tc].sqn; + } + + err = mlx5_eswitch_sqs2vport_start(esw, rep, sqs, num_sqs); + + kfree(sqs); + return err; +} + +int mlx5e_nic_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_priv *priv = rep->priv_data; + + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + return mlx5e_add_sqs_fwd_rules(priv); + return 0; +} + +void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + + mlx5_eswitch_sqs2vport_stop(esw, rep); +} + +void mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_priv *priv = rep->priv_data; + + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_remove_sqs_fwd_rules(priv); +} + +static int mlx5e_rep_get_phys_port_name(struct net_device *dev, + char *buf, size_t len) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_eswitch_rep *rep = priv->ppriv; + int ret; + + ret = snprintf(buf, len, "%d", rep->vport - 1); + if (ret >= len) + return -EOPNOTSUPP; + + return 0; +} + +static const struct switchdev_ops mlx5e_rep_switchdev_ops = { + .switchdev_port_attr_get = mlx5e_attr_get, +}; + +static const struct net_device_ops mlx5e_netdev_ops_rep = { + .ndo_open = mlx5e_open, + .ndo_stop = mlx5e_close, + .ndo_start_xmit = mlx5e_xmit, + .ndo_get_phys_port_name = mlx5e_rep_get_phys_port_name, + .ndo_get_stats64 = mlx5e_get_stats, +}; + +static void mlx5e_build_rep_netdev_priv(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, + void *ppriv) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + u8 cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? + MLX5_CQ_PERIOD_MODE_START_FROM_CQE : + MLX5_CQ_PERIOD_MODE_START_FROM_EQE; + + priv->params.log_sq_size = + MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE; + priv->params.rq_wq_type = MLX5_WQ_TYPE_LINKED_LIST; + priv->params.log_rq_size = MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE; + + priv->params.min_rx_wqes = mlx5_min_rx_wqes(priv->params.rq_wq_type, + BIT(priv->params.log_rq_size)); + + priv->params.rx_am_enabled = MLX5_CAP_GEN(mdev, cq_moderation); + mlx5e_set_rx_cq_mode_params(&priv->params, cq_period_mode); + + priv->params.tx_max_inline = mlx5e_get_max_inline_cap(mdev); + priv->params.num_tc = 1; + + priv->params.lro_wqe_sz = + MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + + priv->mdev = mdev; + priv->netdev = netdev; + priv->params.num_channels = profile->max_nch(mdev); + priv->profile = profile; + priv->ppriv = ppriv; + + mutex_init(&priv->state_lock); + + INIT_DELAYED_WORK(&priv->update_stats_work, mlx5e_update_stats_work); +} + +static void mlx5e_build_rep_netdev(struct net_device *netdev) +{ + netdev->netdev_ops = &mlx5e_netdev_ops_rep; + + netdev->watchdog_timeo = 15 * HZ; + + netdev->ethtool_ops = &mlx5e_rep_ethtool_ops; + +#ifdef CONFIG_NET_SWITCHDEV + netdev->switchdev_ops = &mlx5e_rep_switchdev_ops; +#endif + + netdev->features |= NETIF_F_VLAN_CHALLENGED; + + eth_hw_addr_random(netdev); +} + +static void mlx5e_init_rep(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, + void *ppriv) +{ + mlx5e_build_rep_netdev_priv(mdev, netdev, profile, ppriv); + mlx5e_build_rep_netdev(netdev); +} + +static int mlx5e_init_rep_rx(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_flow_rule *flow_rule; + int err; + int i; + + err = mlx5e_create_direct_rqts(priv); + if (err) { + mlx5_core_warn(mdev, "create direct rqts failed, %d\n", err); + return err; + } + + err = mlx5e_create_direct_tirs(priv); + if (err) { + mlx5_core_warn(mdev, "create direct tirs failed, %d\n", err); + goto err_destroy_direct_rqts; + } + + flow_rule = mlx5_eswitch_create_vport_rx_rule(esw, + rep->vport, + priv->direct_tir[0].tirn); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + goto err_destroy_direct_tirs; + } + rep->vport_rx_rule = flow_rule; + + return 0; + +err_destroy_direct_tirs: + mlx5e_destroy_direct_tirs(priv); +err_destroy_direct_rqts: + for (i = 0; i < priv->params.num_channels; i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); + return err; +} + +static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch_rep *rep = priv->ppriv; + int i; + + mlx5_del_flow_rule(rep->vport_rx_rule); + mlx5e_destroy_direct_tirs(priv); + for (i = 0; i < priv->params.num_channels; i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); +} + +static int mlx5e_init_rep_tx(struct mlx5e_priv *priv) +{ + int err; + + err = mlx5e_create_tises(priv); + if (err) { + mlx5_core_warn(priv->mdev, "create tises failed, %d\n", err); + return err; + } + return 0; +} + +static int mlx5e_get_rep_max_num_channels(struct mlx5_core_dev *mdev) +{ +#define MLX5E_PORT_REPRESENTOR_NCH 1 + return MLX5E_PORT_REPRESENTOR_NCH; +} + +static struct mlx5e_profile mlx5e_rep_profile = { + .init = mlx5e_init_rep, + .init_rx = mlx5e_init_rep_rx, + .cleanup_rx = mlx5e_cleanup_rep_rx, + .init_tx = mlx5e_init_rep_tx, + .cleanup_tx = mlx5e_cleanup_nic_tx, + .update_stats = mlx5e_update_sw_rep_counters, + .max_nch = mlx5e_get_rep_max_num_channels, + .max_tc = 1, +}; + +int mlx5e_vport_rep_load(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + rep->priv_data = mlx5e_create_netdev(esw->dev, &mlx5e_rep_profile, rep); + if (!rep->priv_data) { + pr_warn("Failed to create representor for vport %d\n", + rep->vport); + return -EINVAL; + } + return 0; +} + +void mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_priv *priv = rep->priv_data; + + mlx5e_destroy_netdev(esw->dev, priv); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index bd947704b59c..022acc2e8922 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -689,7 +689,7 @@ static inline void mlx5e_handle_csum(struct net_device *netdev, if (is_first_ethertype_ip(skb)) { skb->ip_summed = CHECKSUM_COMPLETE; skb->csum = csum_unfold((__force __sum16)cqe->check_sum); - rq->stats.csum_sw++; + rq->stats.csum_complete++; return; } @@ -699,7 +699,7 @@ static inline void mlx5e_handle_csum(struct net_device *netdev, if (cqe_is_tunneled(cqe)) { skb->csum_level = 1; skb->encapsulation = 1; - rq->stats.csum_inner++; + rq->stats.csum_unnecessary_inner++; } return; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c new file mode 100644 index 000000000000..1fffe48a93cc --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c @@ -0,0 +1,335 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +/* Adaptive moderation profiles */ +#define MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE 256 +#define MLX5E_RX_AM_DEF_PROFILE_CQE 1 +#define MLX5E_RX_AM_DEF_PROFILE_EQE 1 +#define MLX5E_PARAMS_AM_NUM_PROFILES 5 + +/* All profiles sizes must be MLX5E_PARAMS_AM_NUM_PROFILES */ +#define MLX5_AM_EQE_PROFILES { \ + {1, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {8, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {64, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {128, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {256, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ +} + +#define MLX5_AM_CQE_PROFILES { \ + {2, 256}, \ + {8, 128}, \ + {16, 64}, \ + {32, 64}, \ + {64, 64} \ +} + +static const struct mlx5e_cq_moder +profile[MLX5_CQ_PERIOD_NUM_MODES][MLX5E_PARAMS_AM_NUM_PROFILES] = { + MLX5_AM_EQE_PROFILES, + MLX5_AM_CQE_PROFILES, +}; + +static inline struct mlx5e_cq_moder mlx5e_am_get_profile(u8 cq_period_mode, int ix) +{ + return profile[cq_period_mode][ix]; +} + +struct mlx5e_cq_moder mlx5e_am_get_def_profile(u8 rx_cq_period_mode) +{ + int default_profile_ix; + + if (rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE) + default_profile_ix = MLX5E_RX_AM_DEF_PROFILE_CQE; + else /* MLX5_CQ_PERIOD_MODE_START_FROM_EQE */ + default_profile_ix = MLX5E_RX_AM_DEF_PROFILE_EQE; + + return profile[rx_cq_period_mode][default_profile_ix]; +} + +/* Adaptive moderation logic */ +enum { + MLX5E_AM_START_MEASURE, + MLX5E_AM_MEASURE_IN_PROGRESS, + MLX5E_AM_APPLY_NEW_PROFILE, +}; + +enum { + MLX5E_AM_PARKING_ON_TOP, + MLX5E_AM_PARKING_TIRED, + MLX5E_AM_GOING_RIGHT, + MLX5E_AM_GOING_LEFT, +}; + +enum { + MLX5E_AM_STATS_WORSE, + MLX5E_AM_STATS_SAME, + MLX5E_AM_STATS_BETTER, +}; + +enum { + MLX5E_AM_STEPPED, + MLX5E_AM_TOO_TIRED, + MLX5E_AM_ON_EDGE, +}; + +static bool mlx5e_am_on_top(struct mlx5e_rx_am *am) +{ + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + case MLX5E_AM_PARKING_TIRED: + WARN_ONCE(true, "mlx5e_am_on_top: PARKING\n"); + return true; + case MLX5E_AM_GOING_RIGHT: + return (am->steps_left > 1) && (am->steps_right == 1); + default: /* MLX5E_AM_GOING_LEFT */ + return (am->steps_right > 1) && (am->steps_left == 1); + } +} + +static void mlx5e_am_turn(struct mlx5e_rx_am *am) +{ + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + case MLX5E_AM_PARKING_TIRED: + WARN_ONCE(true, "mlx5e_am_turn: PARKING\n"); + break; + case MLX5E_AM_GOING_RIGHT: + am->tune_state = MLX5E_AM_GOING_LEFT; + am->steps_left = 0; + break; + case MLX5E_AM_GOING_LEFT: + am->tune_state = MLX5E_AM_GOING_RIGHT; + am->steps_right = 0; + break; + } +} + +static int mlx5e_am_step(struct mlx5e_rx_am *am) +{ + if (am->tired == (MLX5E_PARAMS_AM_NUM_PROFILES * 2)) + return MLX5E_AM_TOO_TIRED; + + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + case MLX5E_AM_PARKING_TIRED: + WARN_ONCE(true, "mlx5e_am_step: PARKING\n"); + break; + case MLX5E_AM_GOING_RIGHT: + if (am->profile_ix == (MLX5E_PARAMS_AM_NUM_PROFILES - 1)) + return MLX5E_AM_ON_EDGE; + am->profile_ix++; + am->steps_right++; + break; + case MLX5E_AM_GOING_LEFT: + if (am->profile_ix == 0) + return MLX5E_AM_ON_EDGE; + am->profile_ix--; + am->steps_left++; + break; + } + + am->tired++; + return MLX5E_AM_STEPPED; +} + +static void mlx5e_am_park_on_top(struct mlx5e_rx_am *am) +{ + am->steps_right = 0; + am->steps_left = 0; + am->tired = 0; + am->tune_state = MLX5E_AM_PARKING_ON_TOP; +} + +static void mlx5e_am_park_tired(struct mlx5e_rx_am *am) +{ + am->steps_right = 0; + am->steps_left = 0; + am->tune_state = MLX5E_AM_PARKING_TIRED; +} + +static void mlx5e_am_exit_parking(struct mlx5e_rx_am *am) +{ + am->tune_state = am->profile_ix ? MLX5E_AM_GOING_LEFT : + MLX5E_AM_GOING_RIGHT; + mlx5e_am_step(am); +} + +static int mlx5e_am_stats_compare(struct mlx5e_rx_am_stats *curr, + struct mlx5e_rx_am_stats *prev) +{ + int diff; + + if (!prev->ppms) + return curr->ppms ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_SAME; + + diff = curr->ppms - prev->ppms; + if (((100 * abs(diff)) / prev->ppms) > 10) /* more than 10% diff */ + return (diff > 0) ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_WORSE; + + if (!prev->epms) + return curr->epms ? MLX5E_AM_STATS_WORSE : + MLX5E_AM_STATS_SAME; + + diff = curr->epms - prev->epms; + if (((100 * abs(diff)) / prev->epms) > 10) /* more than 10% diff */ + return (diff < 0) ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_WORSE; + + return MLX5E_AM_STATS_SAME; +} + +static bool mlx5e_am_decision(struct mlx5e_rx_am_stats *curr_stats, + struct mlx5e_rx_am *am) +{ + int prev_state = am->tune_state; + int prev_ix = am->profile_ix; + int stats_res; + int step_res; + + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + stats_res = mlx5e_am_stats_compare(curr_stats, &am->prev_stats); + if (stats_res != MLX5E_AM_STATS_SAME) + mlx5e_am_exit_parking(am); + break; + + case MLX5E_AM_PARKING_TIRED: + am->tired--; + if (!am->tired) + mlx5e_am_exit_parking(am); + break; + + case MLX5E_AM_GOING_RIGHT: + case MLX5E_AM_GOING_LEFT: + stats_res = mlx5e_am_stats_compare(curr_stats, &am->prev_stats); + if (stats_res != MLX5E_AM_STATS_BETTER) + mlx5e_am_turn(am); + + if (mlx5e_am_on_top(am)) { + mlx5e_am_park_on_top(am); + break; + } + + step_res = mlx5e_am_step(am); + switch (step_res) { + case MLX5E_AM_ON_EDGE: + mlx5e_am_park_on_top(am); + break; + case MLX5E_AM_TOO_TIRED: + mlx5e_am_park_tired(am); + break; + } + + break; + } + + if ((prev_state != MLX5E_AM_PARKING_ON_TOP) || + (am->tune_state != MLX5E_AM_PARKING_ON_TOP)) + am->prev_stats = *curr_stats; + + return am->profile_ix != prev_ix; +} + +static void mlx5e_am_sample(struct mlx5e_rq *rq, + struct mlx5e_rx_am_sample *s) +{ + s->time = ktime_get(); + s->pkt_ctr = rq->stats.packets; + s->event_ctr = rq->cq.event_ctr; +} + +#define MLX5E_AM_NEVENTS 64 + +static void mlx5e_am_calc_stats(struct mlx5e_rx_am_sample *start, + struct mlx5e_rx_am_sample *end, + struct mlx5e_rx_am_stats *curr_stats) +{ + /* u32 holds up to 71 minutes, should be enough */ + u32 delta_us = ktime_us_delta(end->time, start->time); + unsigned int npkts = end->pkt_ctr - start->pkt_ctr; + + if (!delta_us) { + WARN_ONCE(true, "mlx5e_am_calc_stats: delta_us=0\n"); + return; + } + + curr_stats->ppms = (npkts * USEC_PER_MSEC) / delta_us; + curr_stats->epms = (MLX5E_AM_NEVENTS * USEC_PER_MSEC) / delta_us; +} + +void mlx5e_rx_am_work(struct work_struct *work) +{ + struct mlx5e_rx_am *am = container_of(work, struct mlx5e_rx_am, + work); + struct mlx5e_rq *rq = container_of(am, struct mlx5e_rq, am); + struct mlx5e_cq_moder cur_profile = profile[am->mode][am->profile_ix]; + + mlx5_core_modify_cq_moderation(rq->priv->mdev, &rq->cq.mcq, + cur_profile.usec, cur_profile.pkts); + + am->state = MLX5E_AM_START_MEASURE; +} + +void mlx5e_rx_am(struct mlx5e_rq *rq) +{ + struct mlx5e_rx_am *am = &rq->am; + struct mlx5e_rx_am_sample end_sample; + struct mlx5e_rx_am_stats curr_stats; + u16 nevents; + + switch (am->state) { + case MLX5E_AM_MEASURE_IN_PROGRESS: + nevents = rq->cq.event_ctr - am->start_sample.event_ctr; + if (nevents < MLX5E_AM_NEVENTS) + break; + mlx5e_am_sample(rq, &end_sample); + mlx5e_am_calc_stats(&am->start_sample, &end_sample, + &curr_stats); + if (mlx5e_am_decision(&curr_stats, am)) { + am->state = MLX5E_AM_APPLY_NEW_PROFILE; + schedule_work(&am->work); + break; + } + /* fall through */ + case MLX5E_AM_START_MEASURE: + mlx5e_am_sample(rq, &am->start_sample); + am->state = MLX5E_AM_MEASURE_IN_PROGRESS; + break; + case MLX5E_AM_APPLY_NEW_PROFILE: + break; + } +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 83bc32b25849..7b9d8a989b52 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -42,9 +42,11 @@ be64_to_cpu(*(__be32 *)((char *)ptr + dsc[i].offset)) #define MLX5E_DECLARE_STAT(type, fld) #fld, offsetof(type, fld) +#define MLX5E_DECLARE_RX_STAT(type, fld) "rx%d_"#fld, offsetof(type, fld) +#define MLX5E_DECLARE_TX_STAT(type, fld) "tx%d_"#fld, offsetof(type, fld) struct counter_desc { - char name[ETH_GSTRING_LEN]; + char format[ETH_GSTRING_LEN]; int offset; /* Byte offset */ }; @@ -53,18 +55,18 @@ struct mlx5e_sw_stats { u64 rx_bytes; u64 tx_packets; u64 tx_bytes; - u64 tso_packets; - u64 tso_bytes; - u64 tso_inner_packets; - u64 tso_inner_bytes; - u64 lro_packets; - u64 lro_bytes; - u64 rx_csum_good; + u64 tx_tso_packets; + u64 tx_tso_bytes; + u64 tx_tso_inner_packets; + u64 tx_tso_inner_bytes; + u64 rx_lro_packets; + u64 rx_lro_bytes; + u64 rx_csum_unnecessary; u64 rx_csum_none; - u64 rx_csum_sw; - u64 rx_csum_inner; - u64 tx_csum_offload; - u64 tx_csum_inner; + u64 rx_csum_complete; + u64 rx_csum_unnecessary_inner; + u64 tx_csum_partial; + u64 tx_csum_partial_inner; u64 tx_queue_stopped; u64 tx_queue_wake; u64 tx_queue_dropped; @@ -76,7 +78,7 @@ struct mlx5e_sw_stats { u64 rx_cqe_compress_pkts; /* Special handling counters */ - u64 link_down_events; + u64 link_down_events_phy; }; static const struct counter_desc sw_stats_desc[] = { @@ -84,18 +86,18 @@ static const struct counter_desc sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_bytes) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_packets) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tso_packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tso_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tso_inner_packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tso_inner_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, lro_packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, lro_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_good) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_inner_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_inner_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_lro_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_lro_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_unnecessary) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_none) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_sw) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_inner) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_csum_offload) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_csum_inner) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_complete) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_unnecessary_inner) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_csum_partial) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_csum_partial_inner) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_queue_stopped) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_queue_wake) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_queue_dropped) }, @@ -105,7 +107,7 @@ static const struct counter_desc sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_buff_alloc_err) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cqe_compress_blks) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_cqe_compress_pkts) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, link_down_events) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, link_down_events_phy) }, }; struct mlx5e_qcounter_stats { @@ -125,12 +127,6 @@ struct mlx5e_vport_stats { }; static const struct counter_desc vport_stats_desc[] = { - { "rx_vport_error_packets", - VPORT_COUNTER_OFF(received_errors.packets) }, - { "rx_vport_error_bytes", VPORT_COUNTER_OFF(received_errors.octets) }, - { "tx_vport_error_packets", - VPORT_COUNTER_OFF(transmit_errors.packets) }, - { "tx_vport_error_bytes", VPORT_COUNTER_OFF(transmit_errors.octets) }, { "rx_vport_unicast_packets", VPORT_COUNTER_OFF(received_eth_unicast.packets) }, { "rx_vport_unicast_bytes", @@ -155,6 +151,22 @@ static const struct counter_desc vport_stats_desc[] = { VPORT_COUNTER_OFF(transmitted_eth_broadcast.packets) }, { "tx_vport_broadcast_bytes", VPORT_COUNTER_OFF(transmitted_eth_broadcast.octets) }, + { "rx_vport_rdma_unicast_packets", + VPORT_COUNTER_OFF(received_ib_unicast.packets) }, + { "rx_vport_rdma_unicast_bytes", + VPORT_COUNTER_OFF(received_ib_unicast.octets) }, + { "tx_vport_rdma_unicast_packets", + VPORT_COUNTER_OFF(transmitted_ib_unicast.packets) }, + { "tx_vport_rdma_unicast_bytes", + VPORT_COUNTER_OFF(transmitted_ib_unicast.octets) }, + { "rx_vport_rdma_multicast_packets", + VPORT_COUNTER_OFF(received_ib_multicast.packets) }, + { "rx_vport_rdma_multicast_bytes", + VPORT_COUNTER_OFF(received_ib_multicast.octets) }, + { "tx_vport_rdma_multicast_packets", + VPORT_COUNTER_OFF(transmitted_ib_multicast.packets) }, + { "tx_vport_rdma_multicast_bytes", + VPORT_COUNTER_OFF(transmitted_ib_multicast.octets) }, }; #define PPORT_802_3_OFF(c) \ @@ -192,94 +204,69 @@ struct mlx5e_pport_stats { }; static const struct counter_desc pport_802_3_stats_desc[] = { - { "frames_tx", PPORT_802_3_OFF(a_frames_transmitted_ok) }, - { "frames_rx", PPORT_802_3_OFF(a_frames_received_ok) }, - { "check_seq_err", PPORT_802_3_OFF(a_frame_check_sequence_errors) }, - { "alignment_err", PPORT_802_3_OFF(a_alignment_errors) }, - { "octets_tx", PPORT_802_3_OFF(a_octets_transmitted_ok) }, - { "octets_received", PPORT_802_3_OFF(a_octets_received_ok) }, - { "multicast_xmitted", PPORT_802_3_OFF(a_multicast_frames_xmitted_ok) }, - { "broadcast_xmitted", PPORT_802_3_OFF(a_broadcast_frames_xmitted_ok) }, - { "multicast_rx", PPORT_802_3_OFF(a_multicast_frames_received_ok) }, - { "broadcast_rx", PPORT_802_3_OFF(a_broadcast_frames_received_ok) }, - { "in_range_len_errors", PPORT_802_3_OFF(a_in_range_length_errors) }, - { "out_of_range_len", PPORT_802_3_OFF(a_out_of_range_length_field) }, - { "too_long_errors", PPORT_802_3_OFF(a_frame_too_long_errors) }, - { "symbol_err", PPORT_802_3_OFF(a_symbol_error_during_carrier) }, - { "mac_control_tx", PPORT_802_3_OFF(a_mac_control_frames_transmitted) }, - { "mac_control_rx", PPORT_802_3_OFF(a_mac_control_frames_received) }, - { "unsupported_op_rx", - PPORT_802_3_OFF(a_unsupported_opcodes_received) }, - { "pause_ctrl_rx", PPORT_802_3_OFF(a_pause_mac_ctrl_frames_received) }, - { "pause_ctrl_tx", - PPORT_802_3_OFF(a_pause_mac_ctrl_frames_transmitted) }, + { "tx_packets_phy", PPORT_802_3_OFF(a_frames_transmitted_ok) }, + { "rx_packets_phy", PPORT_802_3_OFF(a_frames_received_ok) }, + { "rx_crc_errors_phy", PPORT_802_3_OFF(a_frame_check_sequence_errors) }, + { "tx_bytes_phy", PPORT_802_3_OFF(a_octets_transmitted_ok) }, + { "rx_bytes_phy", PPORT_802_3_OFF(a_octets_received_ok) }, + { "tx_multicast_phy", PPORT_802_3_OFF(a_multicast_frames_xmitted_ok) }, + { "tx_broadcast_phy", PPORT_802_3_OFF(a_broadcast_frames_xmitted_ok) }, + { "rx_multicast_phy", PPORT_802_3_OFF(a_multicast_frames_received_ok) }, + { "rx_broadcast_phy", PPORT_802_3_OFF(a_broadcast_frames_received_ok) }, + { "rx_in_range_len_errors_phy", PPORT_802_3_OFF(a_in_range_length_errors) }, + { "rx_out_of_range_len_phy", PPORT_802_3_OFF(a_out_of_range_length_field) }, + { "rx_oversize_pkts_phy", PPORT_802_3_OFF(a_frame_too_long_errors) }, + { "rx_symbol_err_phy", PPORT_802_3_OFF(a_symbol_error_during_carrier) }, + { "tx_mac_control_phy", PPORT_802_3_OFF(a_mac_control_frames_transmitted) }, + { "rx_mac_control_phy", PPORT_802_3_OFF(a_mac_control_frames_received) }, + { "rx_unsupported_op_phy", PPORT_802_3_OFF(a_unsupported_opcodes_received) }, + { "rx_pause_ctrl_phy", PPORT_802_3_OFF(a_pause_mac_ctrl_frames_received) }, + { "tx_pause_ctrl_phy", PPORT_802_3_OFF(a_pause_mac_ctrl_frames_transmitted) }, }; static const struct counter_desc pport_2863_stats_desc[] = { - { "in_octets", PPORT_2863_OFF(if_in_octets) }, - { "in_ucast_pkts", PPORT_2863_OFF(if_in_ucast_pkts) }, - { "in_discards", PPORT_2863_OFF(if_in_discards) }, - { "in_errors", PPORT_2863_OFF(if_in_errors) }, - { "in_unknown_protos", PPORT_2863_OFF(if_in_unknown_protos) }, - { "out_octets", PPORT_2863_OFF(if_out_octets) }, - { "out_ucast_pkts", PPORT_2863_OFF(if_out_ucast_pkts) }, - { "out_discards", PPORT_2863_OFF(if_out_discards) }, - { "out_errors", PPORT_2863_OFF(if_out_errors) }, - { "in_multicast_pkts", PPORT_2863_OFF(if_in_multicast_pkts) }, - { "in_broadcast_pkts", PPORT_2863_OFF(if_in_broadcast_pkts) }, - { "out_multicast_pkts", PPORT_2863_OFF(if_out_multicast_pkts) }, - { "out_broadcast_pkts", PPORT_2863_OFF(if_out_broadcast_pkts) }, + { "rx_discards_phy", PPORT_2863_OFF(if_in_discards) }, + { "tx_discards_phy", PPORT_2863_OFF(if_out_discards) }, + { "tx_errors_phy", PPORT_2863_OFF(if_out_errors) }, }; static const struct counter_desc pport_2819_stats_desc[] = { - { "drop_events", PPORT_2819_OFF(ether_stats_drop_events) }, - { "octets", PPORT_2819_OFF(ether_stats_octets) }, - { "pkts", PPORT_2819_OFF(ether_stats_pkts) }, - { "broadcast_pkts", PPORT_2819_OFF(ether_stats_broadcast_pkts) }, - { "multicast_pkts", PPORT_2819_OFF(ether_stats_multicast_pkts) }, - { "crc_align_errors", PPORT_2819_OFF(ether_stats_crc_align_errors) }, - { "undersize_pkts", PPORT_2819_OFF(ether_stats_undersize_pkts) }, - { "oversize_pkts", PPORT_2819_OFF(ether_stats_oversize_pkts) }, - { "fragments", PPORT_2819_OFF(ether_stats_fragments) }, - { "jabbers", PPORT_2819_OFF(ether_stats_jabbers) }, - { "collisions", PPORT_2819_OFF(ether_stats_collisions) }, - { "p64octets", PPORT_2819_OFF(ether_stats_pkts64octets) }, - { "p65to127octets", PPORT_2819_OFF(ether_stats_pkts65to127octets) }, - { "p128to255octets", PPORT_2819_OFF(ether_stats_pkts128to255octets) }, - { "p256to511octets", PPORT_2819_OFF(ether_stats_pkts256to511octets) }, - { "p512to1023octets", PPORT_2819_OFF(ether_stats_pkts512to1023octets) }, - { "p1024to1518octets", - PPORT_2819_OFF(ether_stats_pkts1024to1518octets) }, - { "p1519to2047octets", - PPORT_2819_OFF(ether_stats_pkts1519to2047octets) }, - { "p2048to4095octets", - PPORT_2819_OFF(ether_stats_pkts2048to4095octets) }, - { "p4096to8191octets", - PPORT_2819_OFF(ether_stats_pkts4096to8191octets) }, - { "p8192to10239octets", - PPORT_2819_OFF(ether_stats_pkts8192to10239octets) }, + { "rx_undersize_pkts_phy", PPORT_2819_OFF(ether_stats_undersize_pkts) }, + { "rx_fragments_phy", PPORT_2819_OFF(ether_stats_fragments) }, + { "rx_jabbers_phy", PPORT_2819_OFF(ether_stats_jabbers) }, + { "rx_64_bytes_phy", PPORT_2819_OFF(ether_stats_pkts64octets) }, + { "rx_65_to_127_bytes_phy", PPORT_2819_OFF(ether_stats_pkts65to127octets) }, + { "rx_128_to_255_bytes_phy", PPORT_2819_OFF(ether_stats_pkts128to255octets) }, + { "rx_256_to_511_bytes_phy", PPORT_2819_OFF(ether_stats_pkts256to511octets) }, + { "rx_512_to_1023_bytes_phy", PPORT_2819_OFF(ether_stats_pkts512to1023octets) }, + { "rx_1024_to_1518_bytes_phy", PPORT_2819_OFF(ether_stats_pkts1024to1518octets) }, + { "rx_1519_to_2047_bytes_phy", PPORT_2819_OFF(ether_stats_pkts1519to2047octets) }, + { "rx_2048_to_4095_bytes_phy", PPORT_2819_OFF(ether_stats_pkts2048to4095octets) }, + { "rx_4096_to_8191_bytes_phy", PPORT_2819_OFF(ether_stats_pkts4096to8191octets) }, + { "rx_8192_to_10239_bytes_phy", PPORT_2819_OFF(ether_stats_pkts8192to10239octets) }, }; static const struct counter_desc pport_per_prio_traffic_stats_desc[] = { - { "rx_octets", PPORT_PER_PRIO_OFF(rx_octets) }, - { "rx_frames", PPORT_PER_PRIO_OFF(rx_frames) }, - { "tx_octets", PPORT_PER_PRIO_OFF(tx_octets) }, - { "tx_frames", PPORT_PER_PRIO_OFF(tx_frames) }, + { "rx_prio%d_bytes", PPORT_PER_PRIO_OFF(rx_octets) }, + { "rx_prio%d_packets", PPORT_PER_PRIO_OFF(rx_frames) }, + { "tx_prio%d_bytes", PPORT_PER_PRIO_OFF(tx_octets) }, + { "tx_prio%d_packets", PPORT_PER_PRIO_OFF(tx_frames) }, }; static const struct counter_desc pport_per_prio_pfc_stats_desc[] = { - { "rx_pause", PPORT_PER_PRIO_OFF(rx_pause) }, - { "rx_pause_duration", PPORT_PER_PRIO_OFF(rx_pause_duration) }, - { "tx_pause", PPORT_PER_PRIO_OFF(tx_pause) }, - { "tx_pause_duration", PPORT_PER_PRIO_OFF(tx_pause_duration) }, - { "rx_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) }, + /* %s is "global" or "prio{i}" */ + { "rx_%s_pause", PPORT_PER_PRIO_OFF(rx_pause) }, + { "rx_%s_pause_duration", PPORT_PER_PRIO_OFF(rx_pause_duration) }, + { "tx_%s_pause", PPORT_PER_PRIO_OFF(tx_pause) }, + { "tx_%s_pause_duration", PPORT_PER_PRIO_OFF(tx_pause_duration) }, + { "rx_%s_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) }, }; struct mlx5e_rq_stats { u64 packets; u64 bytes; - u64 csum_sw; - u64 csum_inner; + u64 csum_complete; + u64 csum_unnecessary_inner; u64 csum_none; u64 lro_packets; u64 lro_bytes; @@ -292,19 +279,19 @@ struct mlx5e_rq_stats { }; static const struct counter_desc rq_stats_desc[] = { - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, csum_sw) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, csum_inner) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, csum_none) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, lro_packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, lro_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, wqe_err) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, mpwqe_filler) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, mpwqe_frag) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, buff_alloc_err) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, cqe_compress_blks) }, - { MLX5E_DECLARE_STAT(struct mlx5e_rq_stats, cqe_compress_pkts) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, packets) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, bytes) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, csum_complete) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, csum_unnecessary_inner) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, csum_none) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, lro_packets) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, lro_bytes) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, wqe_err) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, mpwqe_filler) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, mpwqe_frag) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, buff_alloc_err) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cqe_compress_blks) }, + { MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, cqe_compress_pkts) }, }; struct mlx5e_sq_stats { @@ -315,28 +302,28 @@ struct mlx5e_sq_stats { u64 tso_bytes; u64 tso_inner_packets; u64 tso_inner_bytes; - u64 csum_offload_inner; + u64 csum_partial_inner; u64 nop; /* less likely accessed in data path */ - u64 csum_offload_none; + u64 csum_none; u64 stopped; u64 wake; u64 dropped; }; static const struct counter_desc sq_stats_desc[] = { - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, tso_packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, tso_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, tso_inner_packets) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, tso_inner_bytes) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, csum_offload_inner) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, nop) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, csum_offload_none) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, stopped) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, wake) }, - { MLX5E_DECLARE_STAT(struct mlx5e_sq_stats, dropped) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, packets) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, bytes) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tso_packets) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tso_bytes) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tso_inner_packets) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, tso_inner_bytes) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, csum_partial_inner) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, nop) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, csum_none) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, stopped) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, wake) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, dropped) }, }; #define NUM_SW_COUNTERS ARRAY_SIZE(sw_stats_desc) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 704c3d30493e..3261e8b1286e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -50,7 +50,7 @@ struct mlx5e_tc_flow { #define MLX5E_TC_TABLE_NUM_GROUPS 4 static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, - u32 *match_c, u32 *match_v, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag) { struct mlx5_core_dev *dev = priv->mdev; @@ -88,8 +88,8 @@ static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, table_created = true; } - rule = mlx5_add_flow_rule(priv->fs.tc.t, MLX5_MATCH_OUTER_HEADERS, - match_c, match_v, + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + rule = mlx5_add_flow_rule(priv->fs.tc.t, spec, action, flow_tag, &dest); @@ -126,12 +126,13 @@ static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, } } -static int parse_cls_flower(struct mlx5e_priv *priv, - u32 *match_c, u32 *match_v, +static int parse_cls_flower(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, struct tc_cls_flower_offload *f) { - void *headers_c = MLX5_ADDR_OF(fte_match_param, match_c, outer_headers); - void *headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers); + void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); + void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); u16 addr_type = 0; u8 ip_proto = 0; @@ -342,12 +343,11 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, struct tc_cls_flower_offload *f) { struct mlx5e_tc_table *tc = &priv->fs.tc; - u32 *match_c; - u32 *match_v; int err = 0; u32 flow_tag; u32 action; struct mlx5e_tc_flow *flow; + struct mlx5_flow_spec *spec; struct mlx5_flow_rule *old = NULL; flow = rhashtable_lookup_fast(&tc->ht, &f->cookie, @@ -357,16 +357,15 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, else flow = kzalloc(sizeof(*flow), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_c || !match_v || !flow) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec || !flow) { err = -ENOMEM; goto err_free; } flow->cookie = f->cookie; - err = parse_cls_flower(priv, match_c, match_v, f); + err = parse_cls_flower(priv, spec, f); if (err < 0) goto err_free; @@ -379,8 +378,7 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, if (err) goto err_free; - flow->rule = mlx5e_tc_add_flow(priv, match_c, match_v, action, - flow_tag); + flow->rule = mlx5e_tc_add_flow(priv, spec, action, flow_tag); if (IS_ERR(flow->rule)) { err = PTR_ERR(flow->rule); goto err_hash_del; @@ -398,8 +396,7 @@ err_free: if (!old) kfree(flow); out: - kfree(match_c); - kfree(match_v); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 229ab16fb8d3..5a750b9cd006 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -192,12 +192,12 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) if (skb->encapsulation) { eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM | MLX5_ETH_WQE_L4_INNER_CSUM; - sq->stats.csum_offload_inner++; + sq->stats.csum_partial_inner++; } else { eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM; } } else - sq->stats.csum_offload_none++; + sq->stats.csum_none++; if (sq->cc != sq->prev_cc) { sq->prev_cc = sq->cc; @@ -317,7 +317,8 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) while ((sq->pc & wq->sz_m1) > sq->edge) mlx5e_send_nop(sq, false); - sq->bf_budget = bf ? sq->bf_budget - 1 : 0; + if (bf) + sq->bf_budget--; sq->stats.packets++; sq->stats.bytes += num_bytes; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index c38781fa567d..64ae2e800daa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -136,6 +136,10 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget) for (i = 0; i < c->num_tc; i++) mlx5e_cq_arm(&c->sq[i].cq); + + if (test_bit(MLX5E_RQ_STATE_AM, &c->rq.state)) + mlx5e_rx_am(&c->rq); + mlx5e_cq_arm(&c->rq.cq); mlx5e_cq_arm(&c->icosq.cq); @@ -146,6 +150,7 @@ void mlx5e_completion_event(struct mlx5_core_cq *mcq) { struct mlx5e_cq *cq = container_of(mcq, struct mlx5e_cq, mcq); + cq->event_ctr++; set_bit(MLX5E_CHANNEL_NAPI_SCHED, &cq->channel->flags); napi_schedule(cq->napi); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index b84a6918a700..f6d667797ee1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -40,17 +40,6 @@ #define UPLINK_VPORT 0xFFFF -#define MLX5_DEBUG_ESWITCH_MASK BIT(3) - -#define esw_info(dev, format, ...) \ - pr_info("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) - -#define esw_warn(dev, format, ...) \ - pr_warn("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) - -#define esw_debug(dev, format, ...) \ - mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__) - enum { MLX5_ACTION_NONE = 0, MLX5_ACTION_ADD = 1, @@ -92,6 +81,9 @@ enum { MC_ADDR_CHANGE | \ PROMISC_CHANGE) +int esw_offloads_init(struct mlx5_eswitch *esw, int nvports); +void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports); + static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport, u32 events_mask) { @@ -337,25 +329,23 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, MLX5_MATCH_OUTER_HEADERS); struct mlx5_flow_rule *flow_rule = NULL; struct mlx5_flow_destination dest; + struct mlx5_flow_spec *spec; void *mv_misc = NULL; void *mc_misc = NULL; u8 *dmac_v = NULL; u8 *dmac_c = NULL; - u32 *match_v; - u32 *match_c; if (rx_rule) match_header |= MLX5_MATCH_MISC_PARAMETERS; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { pr_warn("FDB: Failed to alloc match parameters\n"); - goto out; + return NULL; } - - dmac_v = MLX5_ADDR_OF(fte_match_param, match_v, + dmac_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dmac_47_16); - dmac_c = MLX5_ADDR_OF(fte_match_param, match_c, + dmac_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dmac_47_16); if (match_header & MLX5_MATCH_OUTER_HEADERS) { @@ -364,8 +354,10 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, } if (match_header & MLX5_MATCH_MISC_PARAMETERS) { - mv_misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); - mc_misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + mv_misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, + misc_parameters); + mc_misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + misc_parameters); MLX5_SET(fte_match_set_misc, mv_misc, source_port, UPLINK_VPORT); MLX5_SET_TO_ONES(fte_match_set_misc, mc_misc, source_port); } @@ -376,22 +368,19 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, esw_debug(esw->dev, "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n", dmac_v, dmac_c, vport); + spec->match_criteria_enable = match_header; flow_rule = - mlx5_add_flow_rule(esw->fdb_table.fdb, - match_header, - match_c, - match_v, + mlx5_add_flow_rule(esw->fdb_table.fdb, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); - if (IS_ERR_OR_NULL(flow_rule)) { + if (IS_ERR(flow_rule)) { pr_warn( "FDB: Failed to add flow rule: dmac_v(%pM) dmac_c(%pM) -> vport(%d), err(%ld)\n", dmac_v, dmac_c, vport, PTR_ERR(flow_rule)); flow_rule = NULL; } -out: - kfree(match_v); - kfree(match_c); + + kvfree(spec); return flow_rule; } @@ -428,7 +417,7 @@ esw_fdb_set_vport_promisc_rule(struct mlx5_eswitch *esw, u32 vport) return __esw_fdb_set_vport_rule(esw, vport, true, mac_c, mac_v); } -static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) +static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw, int nvports) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_core_dev *dev = esw->dev; @@ -457,7 +446,7 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) table_size = BIT(MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); fdb = mlx5_create_flow_table(root_ns, 0, table_size, 0); - if (IS_ERR_OR_NULL(fdb)) { + if (IS_ERR(fdb)) { err = PTR_ERR(fdb); esw_warn(dev, "Failed to create FDB Table err %d\n", err); goto out; @@ -474,12 +463,12 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 3); eth_broadcast_addr(dmac); g = mlx5_create_flow_group(fdb, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create flow group err(%d)\n", err); goto out; } - esw->fdb_table.addr_grp = g; + esw->fdb_table.legacy.addr_grp = g; /* Allmulti group : One rule that forwards any mcast traffic */ MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, @@ -489,12 +478,12 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) eth_zero_addr(dmac); dmac[0] = 0x01; g = mlx5_create_flow_group(fdb, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create allmulti flow group err(%d)\n", err); goto out; } - esw->fdb_table.allmulti_grp = g; + esw->fdb_table.legacy.allmulti_grp = g; /* Promiscuous group : * One rule that forward all unmatched traffic from previous groups @@ -506,22 +495,22 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, table_size - 1); MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, table_size - 1); g = mlx5_create_flow_group(fdb, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create promisc flow group err(%d)\n", err); goto out; } - esw->fdb_table.promisc_grp = g; + esw->fdb_table.legacy.promisc_grp = g; out: if (err) { - if (!IS_ERR_OR_NULL(esw->fdb_table.allmulti_grp)) { - mlx5_destroy_flow_group(esw->fdb_table.allmulti_grp); - esw->fdb_table.allmulti_grp = NULL; + if (!IS_ERR_OR_NULL(esw->fdb_table.legacy.allmulti_grp)) { + mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp); + esw->fdb_table.legacy.allmulti_grp = NULL; } - if (!IS_ERR_OR_NULL(esw->fdb_table.addr_grp)) { - mlx5_destroy_flow_group(esw->fdb_table.addr_grp); - esw->fdb_table.addr_grp = NULL; + if (!IS_ERR_OR_NULL(esw->fdb_table.legacy.addr_grp)) { + mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp); + esw->fdb_table.legacy.addr_grp = NULL; } if (!IS_ERR_OR_NULL(esw->fdb_table.fdb)) { mlx5_destroy_flow_table(esw->fdb_table.fdb); @@ -529,24 +518,24 @@ out: } } - kfree(flow_group_in); + kvfree(flow_group_in); return err; } -static void esw_destroy_fdb_table(struct mlx5_eswitch *esw) +static void esw_destroy_legacy_fdb_table(struct mlx5_eswitch *esw) { if (!esw->fdb_table.fdb) return; esw_debug(esw->dev, "Destroy FDB Table\n"); - mlx5_destroy_flow_group(esw->fdb_table.promisc_grp); - mlx5_destroy_flow_group(esw->fdb_table.allmulti_grp); - mlx5_destroy_flow_group(esw->fdb_table.addr_grp); + mlx5_destroy_flow_group(esw->fdb_table.legacy.promisc_grp); + mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp); + mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp); mlx5_destroy_flow_table(esw->fdb_table.fdb); esw->fdb_table.fdb = NULL; - esw->fdb_table.addr_grp = NULL; - esw->fdb_table.allmulti_grp = NULL; - esw->fdb_table.promisc_grp = NULL; + esw->fdb_table.legacy.addr_grp = NULL; + esw->fdb_table.legacy.allmulti_grp = NULL; + esw->fdb_table.legacy.promisc_grp = NULL; } /* E-Switch vport UC/MC lists management */ @@ -578,7 +567,8 @@ static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) if (err) goto abort; - if (esw->fdb_table.fdb) /* SRIOV is enabled: Forward UC MAC to vport */ + /* SRIOV is enabled: Forward UC MAC to vport */ + if (esw->fdb_table.fdb && esw->mode == SRIOV_LEGACY) vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport); esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM index:%d fr(%p)\n", @@ -651,6 +641,7 @@ static void update_allmulti_vports(struct mlx5_eswitch *esw, esw_fdb_set_vport_rule(esw, mac, vport_idx); + iter_vaddr->mc_promisc = true; break; case MLX5_ACTION_DEL: if (!iter_vaddr) @@ -1060,7 +1051,7 @@ static void esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, return; acl = mlx5_create_vport_flow_table(root_ns, 0, table_size, 0, vport->vport); - if (IS_ERR_OR_NULL(acl)) { + if (IS_ERR(acl)) { err = PTR_ERR(acl); esw_warn(dev, "Failed to create E-Switch vport[%d] egress flow Table, err(%d)\n", vport->vport, err); @@ -1075,7 +1066,7 @@ static void esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); vlan_grp = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR_OR_NULL(vlan_grp)) { + if (IS_ERR(vlan_grp)) { err = PTR_ERR(vlan_grp); esw_warn(dev, "Failed to create E-Switch vport[%d] egress allowed vlans flow group, err(%d)\n", vport->vport, err); @@ -1086,7 +1077,7 @@ static void esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); drop_grp = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR_OR_NULL(drop_grp)) { + if (IS_ERR(drop_grp)) { err = PTR_ERR(drop_grp); esw_warn(dev, "Failed to create E-Switch vport[%d] egress drop flow group, err(%d)\n", vport->vport, err); @@ -1097,7 +1088,7 @@ static void esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, vport->egress.drop_grp = drop_grp; vport->egress.allowed_vlans_grp = vlan_grp; out: - kfree(flow_group_in); + kvfree(flow_group_in); if (err && !IS_ERR_OR_NULL(vlan_grp)) mlx5_destroy_flow_group(vlan_grp); if (err && !IS_ERR_OR_NULL(acl)) @@ -1174,7 +1165,7 @@ static void esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw, return; acl = mlx5_create_vport_flow_table(root_ns, 0, table_size, 0, vport->vport); - if (IS_ERR_OR_NULL(acl)) { + if (IS_ERR(acl)) { err = PTR_ERR(acl); esw_warn(dev, "Failed to create E-Switch vport[%d] ingress flow Table, err(%d)\n", vport->vport, err); @@ -1192,7 +1183,7 @@ static void esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); g = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create E-Switch vport[%d] ingress untagged spoofchk flow group, err(%d)\n", vport->vport, err); @@ -1207,7 +1198,7 @@ static void esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); g = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create E-Switch vport[%d] ingress untagged flow group, err(%d)\n", vport->vport, err); @@ -1223,7 +1214,7 @@ static void esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 2); g = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create E-Switch vport[%d] ingress spoofchk flow group, err(%d)\n", vport->vport, err); @@ -1236,7 +1227,7 @@ static void esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw, MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 3); g = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR_OR_NULL(g)) { + if (IS_ERR(g)) { err = PTR_ERR(g); esw_warn(dev, "Failed to create E-Switch vport[%d] ingress drop flow group, err(%d)\n", vport->vport, err); @@ -1259,7 +1250,7 @@ out: mlx5_destroy_flow_table(vport->ingress.acl); } - kfree(flow_group_in); + kvfree(flow_group_in); } static void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw, @@ -1299,9 +1290,8 @@ static void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw, static int esw_vport_ingress_config(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { + struct mlx5_flow_spec *spec; u8 smac[ETH_ALEN]; - u32 *match_v; - u32 *match_c; int err = 0; u8 *smac_v; @@ -1335,9 +1325,8 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, "vport[%d] configure ingress rules, vlan(%d) qos(%d)\n", vport->vport, vport->vlan, vport->qos); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { err = -ENOMEM; esw_warn(esw->dev, "vport[%d] configure ingress rules failed, err(%d)\n", vport->vport, err); @@ -1345,25 +1334,23 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, } if (vport->vlan || vport->qos) - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); if (vport->spoofchk) { - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.smac_47_16); - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.smac_15_0); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_47_16); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_15_0); smac_v = MLX5_ADDR_OF(fte_match_param, - match_v, + spec->match_value, outer_headers.smac_47_16); ether_addr_copy(smac_v, smac); } + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; vport->ingress.allow_rule = - mlx5_add_flow_rule(vport->ingress.acl, - MLX5_MATCH_OUTER_HEADERS, - match_c, - match_v, + mlx5_add_flow_rule(vport->ingress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_ALLOW, 0, NULL); - if (IS_ERR_OR_NULL(vport->ingress.allow_rule)) { + if (IS_ERR(vport->ingress.allow_rule)) { err = PTR_ERR(vport->ingress.allow_rule); pr_warn("vport[%d] configure ingress allow rule, err(%d)\n", vport->vport, err); @@ -1371,16 +1358,12 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, goto out; } - memset(match_c, 0, MLX5_ST_SZ_BYTES(fte_match_param)); - memset(match_v, 0, MLX5_ST_SZ_BYTES(fte_match_param)); + memset(spec, 0, sizeof(*spec)); vport->ingress.drop_rule = - mlx5_add_flow_rule(vport->ingress.acl, - 0, - match_c, - match_v, + mlx5_add_flow_rule(vport->ingress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_DROP, 0, NULL); - if (IS_ERR_OR_NULL(vport->ingress.drop_rule)) { + if (IS_ERR(vport->ingress.drop_rule)) { err = PTR_ERR(vport->ingress.drop_rule); pr_warn("vport[%d] configure ingress drop rule, err(%d)\n", vport->vport, err); @@ -1391,17 +1374,14 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, out: if (err) esw_vport_cleanup_ingress_rules(esw, vport); - - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } static int esw_vport_egress_config(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { - u32 *match_v; - u32 *match_c; + struct mlx5_flow_spec *spec; int err = 0; esw_vport_cleanup_egress_rules(esw, vport); @@ -1417,9 +1397,8 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, "vport[%d] configure egress rules, vlan(%d) qos(%d)\n", vport->vport, vport->vlan, vport->qos); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { err = -ENOMEM; esw_warn(esw->dev, "vport[%d] configure egress rules failed, err(%d)\n", vport->vport, err); @@ -1427,19 +1406,17 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, } /* Allowed vlan rule */ - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_v, outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.first_vid); - MLX5_SET(fte_match_param, match_v, outer_headers.first_vid, vport->vlan); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_value, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, vport->vlan); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; vport->egress.allowed_vlan = - mlx5_add_flow_rule(vport->egress.acl, - MLX5_MATCH_OUTER_HEADERS, - match_c, - match_v, + mlx5_add_flow_rule(vport->egress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_ALLOW, 0, NULL); - if (IS_ERR_OR_NULL(vport->egress.allowed_vlan)) { + if (IS_ERR(vport->egress.allowed_vlan)) { err = PTR_ERR(vport->egress.allowed_vlan); pr_warn("vport[%d] configure egress allowed vlan rule failed, err(%d)\n", vport->vport, err); @@ -1448,24 +1425,19 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, } /* Drop others rule (star rule) */ - memset(match_c, 0, MLX5_ST_SZ_BYTES(fte_match_param)); - memset(match_v, 0, MLX5_ST_SZ_BYTES(fte_match_param)); + memset(spec, 0, sizeof(*spec)); vport->egress.drop_rule = - mlx5_add_flow_rule(vport->egress.acl, - 0, - match_c, - match_v, + mlx5_add_flow_rule(vport->egress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_DROP, 0, NULL); - if (IS_ERR_OR_NULL(vport->egress.drop_rule)) { + if (IS_ERR(vport->egress.drop_rule)) { err = PTR_ERR(vport->egress.drop_rule); pr_warn("vport[%d] configure egress drop rule failed, err(%d)\n", vport->vport, err); vport->egress.drop_rule = NULL; } out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } @@ -1491,14 +1463,11 @@ static void esw_enable_vport(struct mlx5_eswitch *esw, int vport_num, /* Sync with current vport context */ vport->enabled_events = enable_events; - esw_vport_change_handle_locked(vport); - vport->enabled = true; /* only PF is trusted by default */ vport->trusted = (vport_num) ? false : true; - - arm_vport_context_events_cmd(esw->dev, vport_num, enable_events); + esw_vport_change_handle_locked(vport); esw->enabled_vports++; esw_debug(esw->dev, "Enabled VPORT(%d)\n", vport_num); @@ -1542,10 +1511,10 @@ static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num) } /* Public E-Switch API */ -int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs) +int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) { int err; - int i; + int i, enabled_events; if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) @@ -1563,16 +1532,20 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs) if (!MLX5_CAP_ESW_EGRESS_ACL(esw->dev, ft_support)) esw_warn(esw->dev, "E-Switch engress ACL is not supported by FW\n"); - esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d)\n", nvfs); - + esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d) mode (%d)\n", nvfs, mode); + esw->mode = mode; esw_disable_vport(esw, 0); - err = esw_create_fdb_table(esw, nvfs + 1); + if (mode == SRIOV_LEGACY) + err = esw_create_legacy_fdb_table(esw, nvfs + 1); + else + err = esw_offloads_init(esw, nvfs + 1); if (err) goto abort; + enabled_events = (mode == SRIOV_LEGACY) ? SRIOV_VPORT_EVENTS : UC_ADDR_CHANGE; for (i = 0; i <= nvfs; i++) - esw_enable_vport(esw, i, SRIOV_VPORT_EVENTS); + esw_enable_vport(esw, i, enabled_events); esw_info(esw->dev, "SRIOV enabled: active vports(%d)\n", esw->enabled_vports); @@ -1586,16 +1559,18 @@ abort: void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) { struct esw_mc_addr *mc_promisc; + int nvports; int i; if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) return; - esw_info(esw->dev, "disable SRIOV: active vports(%d)\n", - esw->enabled_vports); + esw_info(esw->dev, "disable SRIOV: active vports(%d) mode(%d)\n", + esw->enabled_vports, esw->mode); mc_promisc = esw->mc_promisc; + nvports = esw->enabled_vports; for (i = 0; i < esw->total_vports; i++) esw_disable_vport(esw, i); @@ -1603,8 +1578,12 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) if (mc_promisc && mc_promisc->uplink_rule) mlx5_del_flow_rule(mc_promisc->uplink_rule); - esw_destroy_fdb_table(esw); + if (esw->mode == SRIOV_LEGACY) + esw_destroy_legacy_fdb_table(esw); + else if (esw->mode == SRIOV_OFFLOADS) + esw_offloads_cleanup(esw, nvports); + esw->mode = SRIOV_NONE; /* VPORT 0 (PF) must be enabled back with non-sriov configuration */ esw_enable_vport(esw, 0, UC_ADDR_CHANGE); } @@ -1662,6 +1641,14 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) goto abort; } + esw->offloads.vport_reps = + kzalloc(total_vports * sizeof(struct mlx5_eswitch_rep), + GFP_KERNEL); + if (!esw->offloads.vport_reps) { + err = -ENOMEM; + goto abort; + } + mutex_init(&esw->state_lock); for (vport_num = 0; vport_num < total_vports; vport_num++) { @@ -1675,6 +1662,7 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) esw->total_vports = total_vports; esw->enabled_vports = 0; + esw->mode = SRIOV_NONE; dev->priv.eswitch = esw; esw_enable_vport(esw, 0, UC_ADDR_CHANGE); @@ -1685,6 +1673,7 @@ abort: destroy_workqueue(esw->work_queue); kfree(esw->l2_table.bitmap); kfree(esw->vports); + kfree(esw->offloads.vport_reps); kfree(esw); return err; } @@ -1702,6 +1691,7 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) destroy_workqueue(esw->work_queue); kfree(esw->l2_table.bitmap); kfree(esw->mc_promisc); + kfree(esw->offloads.vport_reps); kfree(esw->vports); kfree(esw); } @@ -1728,11 +1718,24 @@ void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe) (esw && MLX5_CAP_GEN(esw->dev, vport_group_manager) && mlx5_core_is_pf(esw->dev)) #define LEGAL_VPORT(esw, vport) (vport >= 0 && vport < esw->total_vports) +static void node_guid_gen_from_mac(u64 *node_guid, u8 mac[ETH_ALEN]) +{ + ((u8 *)node_guid)[7] = mac[0]; + ((u8 *)node_guid)[6] = mac[1]; + ((u8 *)node_guid)[5] = mac[2]; + ((u8 *)node_guid)[4] = 0xff; + ((u8 *)node_guid)[3] = 0xfe; + ((u8 *)node_guid)[2] = mac[3]; + ((u8 *)node_guid)[1] = mac[4]; + ((u8 *)node_guid)[0] = mac[5]; +} + int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, int vport, u8 mac[ETH_ALEN]) { - int err = 0; struct mlx5_vport *evport; + u64 node_guid; + int err = 0; if (!ESW_ALLOWED(esw)) return -EPERM; @@ -1756,11 +1759,17 @@ int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, return err; } + node_guid_gen_from_mac(&node_guid, mac); + err = mlx5_modify_nic_vport_node_guid(esw->dev, vport, node_guid); + if (err) + mlx5_core_warn(esw->dev, + "Failed to set vport %d node guid, err = %d. RDMA_CM will not function properly for this VF.\n", + vport, err); + mutex_lock(&esw->state_lock); if (evport->enabled) err = esw_vport_ingress_config(esw, evport); mutex_unlock(&esw->state_lock); - return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index fd6800256d4a..7b45e6a6efb8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -35,6 +35,7 @@ #include <linux/if_ether.h> #include <linux/if_link.h> +#include <net/devlink.h> #include <linux/mlx5/device.h> #define MLX5_MAX_UC_PER_VPORT(dev) \ @@ -46,6 +47,8 @@ #define MLX5_L2_ADDR_HASH_SIZE (BIT(BITS_PER_BYTE)) #define MLX5_L2_ADDR_HASH(addr) (addr[5]) +#define FDB_UPLINK_VPORT 0xffff + /* L2 -mac address based- hash helpers */ struct l2addr_node { struct hlist_node hlist; @@ -134,9 +137,48 @@ struct mlx5_l2_table { struct mlx5_eswitch_fdb { void *fdb; - struct mlx5_flow_group *addr_grp; - struct mlx5_flow_group *allmulti_grp; - struct mlx5_flow_group *promisc_grp; + union { + struct legacy_fdb { + struct mlx5_flow_group *addr_grp; + struct mlx5_flow_group *allmulti_grp; + struct mlx5_flow_group *promisc_grp; + } legacy; + + struct offloads_fdb { + struct mlx5_flow_group *send_to_vport_grp; + struct mlx5_flow_group *miss_grp; + struct mlx5_flow_rule *miss_rule; + } offloads; + }; +}; + +enum { + SRIOV_NONE, + SRIOV_LEGACY, + SRIOV_OFFLOADS +}; + +struct mlx5_esw_sq { + struct mlx5_flow_rule *send_to_vport_rule; + struct list_head list; +}; + +struct mlx5_eswitch_rep { + int (*load)(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); + void (*unload)(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); + u16 vport; + struct mlx5_flow_rule *vport_rx_rule; + void *priv_data; + struct list_head vport_sqs_list; + bool valid; +}; + +struct mlx5_esw_offload { + struct mlx5_flow_table *ft_offloads; + struct mlx5_flow_group *vport_rx_group; + struct mlx5_eswitch_rep *vport_reps; }; struct mlx5_eswitch { @@ -153,13 +195,15 @@ struct mlx5_eswitch { */ struct mutex state_lock; struct esw_mc_addr *mc_promisc; + struct mlx5_esw_offload offloads; + int mode; }; /* E-Switch API */ int mlx5_eswitch_init(struct mlx5_core_dev *dev); void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw); void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe); -int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs); +int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode); void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, int vport, u8 mac[ETH_ALEN]); @@ -177,4 +221,30 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, int vport, struct ifla_vf_stats *vf_stats); +struct mlx5_flow_rule * +mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn); + +int mlx5_eswitch_sqs2vport_start(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep, + u16 *sqns_array, int sqns_num); +void mlx5_eswitch_sqs2vport_stop(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); + +int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode); +int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode); +void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, + int vport); + +#define MLX5_DEBUG_ESWITCH_MASK BIT(3) + +#define esw_info(dev, format, ...) \ + pr_info("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) + +#define esw_warn(dev, format, ...) \ + pr_warn("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) + +#define esw_debug(dev, format, ...) \ + mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__) #endif /* __MLX5_ESWITCH_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c new file mode 100644 index 000000000000..1842dfb4636b --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -0,0 +1,561 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/etherdevice.h> +#include <linux/mlx5/driver.h> +#include <linux/mlx5/mlx5_ifc.h> +#include <linux/mlx5/vport.h> +#include <linux/mlx5/fs.h> +#include "mlx5_core.h" +#include "eswitch.h" + +static struct mlx5_flow_rule * +mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn) +{ + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule; + struct mlx5_flow_spec *spec; + void *misc; + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { + esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); + flow_rule = ERR_PTR(-ENOMEM); + goto out; + } + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); + MLX5_SET(fte_match_set_misc, misc, source_sqn, sqn); + MLX5_SET(fte_match_set_misc, misc, source_port, 0x0); /* source vport is 0 */ + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_sqn); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS; + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + dest.vport_num = vport; + + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); + if (IS_ERR(flow_rule)) + esw_warn(esw->dev, "FDB: Failed to add send to vport rule err %ld\n", PTR_ERR(flow_rule)); +out: + kvfree(spec); + return flow_rule; +} + +void mlx5_eswitch_sqs2vport_stop(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5_esw_sq *esw_sq, *tmp; + + if (esw->mode != SRIOV_OFFLOADS) + return; + + list_for_each_entry_safe(esw_sq, tmp, &rep->vport_sqs_list, list) { + mlx5_del_flow_rule(esw_sq->send_to_vport_rule); + list_del(&esw_sq->list); + kfree(esw_sq); + } +} + +int mlx5_eswitch_sqs2vport_start(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep, + u16 *sqns_array, int sqns_num) +{ + struct mlx5_flow_rule *flow_rule; + struct mlx5_esw_sq *esw_sq; + int vport; + int err; + int i; + + if (esw->mode != SRIOV_OFFLOADS) + return 0; + + vport = rep->vport == 0 ? + FDB_UPLINK_VPORT : rep->vport; + + for (i = 0; i < sqns_num; i++) { + esw_sq = kzalloc(sizeof(*esw_sq), GFP_KERNEL); + if (!esw_sq) { + err = -ENOMEM; + goto out_err; + } + + /* Add re-inject rule to the PF/representor sqs */ + flow_rule = mlx5_eswitch_add_send_to_vport_rule(esw, + vport, + sqns_array[i]); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + kfree(esw_sq); + goto out_err; + } + esw_sq->send_to_vport_rule = flow_rule; + list_add(&esw_sq->list, &rep->vport_sqs_list); + } + return 0; + +out_err: + mlx5_eswitch_sqs2vport_stop(esw, rep); + return err; +} + +static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) +{ + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule = NULL; + struct mlx5_flow_spec *spec; + int err = 0; + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { + esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); + err = -ENOMEM; + goto out; + } + + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + dest.vport_num = 0; + + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + esw_warn(esw->dev, "FDB: Failed to add miss flow rule err %d\n", err); + goto out; + } + + esw->fdb_table.offloads.miss_rule = flow_rule; +out: + kvfree(spec); + return err; +} + +#define MAX_PF_SQ 256 + +static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_namespace *root_ns; + struct mlx5_flow_table *fdb = NULL; + struct mlx5_flow_group *g; + u32 *flow_group_in; + void *match_criteria; + int table_size, ix, err = 0; + + flow_group_in = mlx5_vzalloc(inlen); + if (!flow_group_in) + return -ENOMEM; + + root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); + if (!root_ns) { + esw_warn(dev, "Failed to get FDB flow namespace\n"); + goto ns_err; + } + + esw_debug(dev, "Create offloads FDB table, log_max_size(%d)\n", + MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); + + table_size = nvports + MAX_PF_SQ + 1; + fdb = mlx5_create_flow_table(root_ns, 0, table_size, 0); + if (IS_ERR(fdb)) { + err = PTR_ERR(fdb); + esw_warn(dev, "Failed to create FDB Table err %d\n", err); + goto fdb_err; + } + esw->fdb_table.fdb = fdb; + + /* create send-to-vport group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_MISC_PARAMETERS); + + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); + + MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_sqn); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port); + + ix = nvports + MAX_PF_SQ; + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix - 1); + + g = mlx5_create_flow_group(fdb, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "Failed to create send-to-vport flow group err(%d)\n", err); + goto send_vport_err; + } + esw->fdb_table.offloads.send_to_vport_grp = g; + + /* create miss group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, 0); + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ix); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix + 1); + + g = mlx5_create_flow_group(fdb, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "Failed to create miss flow group err(%d)\n", err); + goto miss_err; + } + esw->fdb_table.offloads.miss_grp = g; + + err = esw_add_fdb_miss_rule(esw); + if (err) + goto miss_rule_err; + + return 0; + +miss_rule_err: + mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); +miss_err: + mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); +send_vport_err: + mlx5_destroy_flow_table(fdb); +fdb_err: +ns_err: + kvfree(flow_group_in); + return err; +} + +static void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) +{ + if (!esw->fdb_table.fdb) + return; + + esw_debug(esw->dev, "Destroy offloads FDB Table\n"); + mlx5_del_flow_rule(esw->fdb_table.offloads.miss_rule); + mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); + mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); + + mlx5_destroy_flow_table(esw->fdb_table.fdb); +} + +static int esw_create_offloads_table(struct mlx5_eswitch *esw) +{ + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft_offloads; + struct mlx5_core_dev *dev = esw->dev; + int err = 0; + + ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_OFFLOADS); + if (!ns) { + esw_warn(esw->dev, "Failed to get offloads flow namespace\n"); + return -ENOMEM; + } + + ft_offloads = mlx5_create_flow_table(ns, 0, dev->priv.sriov.num_vfs + 2, 0); + if (IS_ERR(ft_offloads)) { + err = PTR_ERR(ft_offloads); + esw_warn(esw->dev, "Failed to create offloads table, err %d\n", err); + return err; + } + + esw->offloads.ft_offloads = ft_offloads; + return 0; +} + +static void esw_destroy_offloads_table(struct mlx5_eswitch *esw) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + + mlx5_destroy_flow_table(offloads->ft_offloads); +} + +static int esw_create_vport_rx_group(struct mlx5_eswitch *esw) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *g; + struct mlx5_priv *priv = &esw->dev->priv; + u32 *flow_group_in; + void *match_criteria, *misc; + int err = 0; + int nvports = priv->sriov.num_vfs + 2; + + flow_group_in = mlx5_vzalloc(inlen); + if (!flow_group_in) + return -ENOMEM; + + /* create vport rx group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_MISC_PARAMETERS); + + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); + misc = MLX5_ADDR_OF(fte_match_param, match_criteria, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, nvports - 1); + + g = mlx5_create_flow_group(esw->offloads.ft_offloads, flow_group_in); + + if (IS_ERR(g)) { + err = PTR_ERR(g); + mlx5_core_warn(esw->dev, "Failed to create vport rx group err %d\n", err); + goto out; + } + + esw->offloads.vport_rx_group = g; +out: + kfree(flow_group_in); + return err; +} + +static void esw_destroy_vport_rx_group(struct mlx5_eswitch *esw) +{ + mlx5_destroy_flow_group(esw->offloads.vport_rx_group); +} + +struct mlx5_flow_rule * +mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) +{ + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule; + struct mlx5_flow_spec *spec; + void *misc; + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { + esw_warn(esw->dev, "Failed to alloc match parameters\n"); + flow_rule = ERR_PTR(-ENOMEM); + goto out; + } + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); + MLX5_SET(fte_match_set_misc, misc, source_port, vport); + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS; + dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dest.tir_num = tirn; + + flow_rule = mlx5_add_flow_rule(esw->offloads.ft_offloads, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); + if (IS_ERR(flow_rule)) { + esw_warn(esw->dev, "fs offloads: Failed to add vport rx rule err %ld\n", PTR_ERR(flow_rule)); + goto out; + } + +out: + kvfree(spec); + return flow_rule; +} + +static int esw_offloads_start(struct mlx5_eswitch *esw) +{ + int err, num_vfs = esw->dev->priv.sriov.num_vfs; + + if (esw->mode != SRIOV_LEGACY) { + esw_warn(esw->dev, "Can't set offloads mode, SRIOV legacy not enabled\n"); + return -EINVAL; + } + + mlx5_eswitch_disable_sriov(esw); + err = mlx5_eswitch_enable_sriov(esw, num_vfs, SRIOV_OFFLOADS); + if (err) + esw_warn(esw->dev, "Failed set eswitch to offloads, err %d\n", err); + return err; +} + +int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) +{ + struct mlx5_eswitch_rep *rep; + int vport; + int err; + + err = esw_create_offloads_fdb_table(esw, nvports); + if (err) + return err; + + err = esw_create_offloads_table(esw); + if (err) + goto create_ft_err; + + err = esw_create_vport_rx_group(esw); + if (err) + goto create_fg_err; + + for (vport = 0; vport < nvports; vport++) { + rep = &esw->offloads.vport_reps[vport]; + if (!rep->valid) + continue; + + err = rep->load(esw, rep); + if (err) + goto err_reps; + } + return 0; + +err_reps: + for (vport--; vport >= 0; vport--) { + rep = &esw->offloads.vport_reps[vport]; + if (!rep->valid) + continue; + rep->unload(esw, rep); + } + esw_destroy_vport_rx_group(esw); + +create_fg_err: + esw_destroy_offloads_table(esw); + +create_ft_err: + esw_destroy_offloads_fdb_table(esw); + return err; +} + +static int esw_offloads_stop(struct mlx5_eswitch *esw) +{ + int err, num_vfs = esw->dev->priv.sriov.num_vfs; + + mlx5_eswitch_disable_sriov(esw); + err = mlx5_eswitch_enable_sriov(esw, num_vfs, SRIOV_LEGACY); + if (err) + esw_warn(esw->dev, "Failed set eswitch legacy mode. err %d\n", err); + + return err; +} + +void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports) +{ + struct mlx5_eswitch_rep *rep; + int vport; + + for (vport = 0; vport < nvports; vport++) { + rep = &esw->offloads.vport_reps[vport]; + if (!rep->valid) + continue; + rep->unload(esw, rep); + } + + esw_destroy_vport_rx_group(esw); + esw_destroy_offloads_table(esw); + esw_destroy_offloads_fdb_table(esw); +} + +static int mlx5_esw_mode_from_devlink(u16 mode, u16 *mlx5_mode) +{ + switch (mode) { + case DEVLINK_ESWITCH_MODE_LEGACY: + *mlx5_mode = SRIOV_LEGACY; + break; + case DEVLINK_ESWITCH_MODE_SWITCHDEV: + *mlx5_mode = SRIOV_OFFLOADS; + break; + default: + return -EINVAL; + } + + return 0; +} + +int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode) +{ + struct mlx5_core_dev *dev; + u16 cur_mlx5_mode, mlx5_mode = 0; + + dev = devlink_priv(devlink); + + if (!MLX5_CAP_GEN(dev, vport_group_manager)) + return -EOPNOTSUPP; + + cur_mlx5_mode = dev->priv.eswitch->mode; + + if (cur_mlx5_mode == SRIOV_NONE) + return -EOPNOTSUPP; + + if (mlx5_esw_mode_from_devlink(mode, &mlx5_mode)) + return -EINVAL; + + if (cur_mlx5_mode == mlx5_mode) + return 0; + + if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) + return esw_offloads_start(dev->priv.eswitch); + else if (mode == DEVLINK_ESWITCH_MODE_LEGACY) + return esw_offloads_stop(dev->priv.eswitch); + else + return -EINVAL; +} + +int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode) +{ + struct mlx5_core_dev *dev; + + dev = devlink_priv(devlink); + + if (!MLX5_CAP_GEN(dev, vport_group_manager)) + return -EOPNOTSUPP; + + if (dev->priv.eswitch->mode == SRIOV_NONE) + return -EOPNOTSUPP; + + *mode = dev->priv.eswitch->mode; + + return 0; +} + +void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + + memcpy(&offloads->vport_reps[rep->vport], rep, + sizeof(struct mlx5_eswitch_rep)); + + INIT_LIST_HEAD(&offloads->vport_reps[rep->vport].vport_sqs_list); + offloads->vport_reps[rep->vport].valid = true; +} + +void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, + int vport) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + struct mlx5_eswitch_rep *rep; + + rep = &offloads->vport_reps[vport]; + + if (esw->mode == SRIOV_OFFLOADS && esw->vports[vport].enabled) + rep->unload(esw, rep); + + offloads->vport_reps[vport].valid = false; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 8b5f0b2c0d5c..b0a130479085 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -67,13 +67,21 @@ #define FS_REQUIRED_CAPS(...) {.arr_sz = INIT_CAPS_ARRAY_SIZE(__VA_ARGS__), \ .caps = (long[]) {__VA_ARGS__} } +#define FS_CHAINING_CAPS FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), \ + FS_CAP(flow_table_properties_nic_receive.modify_root), \ + FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), \ + FS_CAP(flow_table_properties_nic_receive.flow_table_modify)) + #define LEFTOVERS_NUM_LEVELS 1 #define LEFTOVERS_NUM_PRIOS 1 #define BY_PASS_PRIO_NUM_LEVELS 1 -#define BY_PASS_MIN_LEVEL (KERNEL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ +#define BY_PASS_MIN_LEVEL (ETHTOOL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ LEFTOVERS_NUM_PRIOS) +#define ETHTOOL_PRIO_NUM_LEVELS 1 +#define ETHTOOL_NUM_PRIOS 10 +#define ETHTOOL_MIN_LEVEL (KERNEL_MIN_LEVEL + ETHTOOL_NUM_PRIOS) /* Vlan, mac, ttc, aRFS */ #define KERNEL_NIC_PRIO_NUM_LEVELS 4 #define KERNEL_NIC_NUM_PRIOS 1 @@ -83,6 +91,11 @@ #define ANCHOR_NUM_LEVELS 1 #define ANCHOR_NUM_PRIOS 1 #define ANCHOR_MIN_LEVEL (BY_PASS_MIN_LEVEL + 1) + +#define OFFLOADS_MAX_FT 1 +#define OFFLOADS_NUM_PRIOS 1 +#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1) + struct node_caps { size_t arr_sz; long *caps; @@ -98,24 +111,24 @@ static struct init_tree_node { int num_levels; } root_fs = { .type = FS_TYPE_NAMESPACE, - .ar_size = 4, + .ar_size = 6, .children = (struct init_tree_node[]) { ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, - FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), - FS_CAP(flow_table_properties_nic_receive.modify_root), - FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), - FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + FS_CHAINING_CAPS, ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_NUM_LEVELS))), + ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {}, + ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))), + ADD_PRIO(0, ETHTOOL_MIN_LEVEL, 0, + FS_CHAINING_CAPS, + ADD_NS(ADD_MULTIPLE_PRIO(ETHTOOL_NUM_PRIOS, + ETHTOOL_PRIO_NUM_LEVELS))), ADD_PRIO(0, KERNEL_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(1, 1), ADD_MULTIPLE_PRIO(KERNEL_NIC_NUM_PRIOS, KERNEL_NIC_PRIO_NUM_LEVELS))), ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, - FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), - FS_CAP(flow_table_properties_nic_receive.modify_root), - FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), - FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + FS_CHAINING_CAPS, ADD_NS(ADD_MULTIPLE_PRIO(LEFTOVERS_NUM_PRIOS, LEFTOVERS_NUM_LEVELS))), ADD_PRIO(0, ANCHOR_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(ANCHOR_NUM_PRIOS, ANCHOR_NUM_LEVELS))), @@ -1152,9 +1165,7 @@ static bool dest_is_valid(struct mlx5_flow_destination *dest, static struct mlx5_flow_rule * _mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest) @@ -1168,22 +1179,23 @@ _mlx5_add_flow_rule(struct mlx5_flow_table *ft, nested_lock_ref_node(&ft->node, FS_MUTEX_GRANDPARENT); fs_for_each_fg(g, ft) if (compare_match_criteria(g->mask.match_criteria_enable, - match_criteria_enable, + spec->match_criteria_enable, g->mask.match_criteria, - match_criteria)) { - rule = add_rule_fg(g, match_value, + spec->match_criteria)) { + rule = add_rule_fg(g, spec->match_value, action, flow_tag, dest); if (!IS_ERR(rule) || PTR_ERR(rule) != -ENOSPC) goto unlock; } - g = create_autogroup(ft, match_criteria_enable, match_criteria); + g = create_autogroup(ft, spec->match_criteria_enable, + spec->match_criteria); if (IS_ERR(g)) { rule = (void *)g; goto unlock; } - rule = add_rule_fg(g, match_value, + rule = add_rule_fg(g, spec->match_value, action, flow_tag, dest); if (IS_ERR(rule)) { /* Remove assumes refcount > 0 and autogroup creates a group @@ -1207,9 +1219,7 @@ static bool fwd_next_prio_supported(struct mlx5_flow_table *ft) struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest) @@ -1240,8 +1250,7 @@ mlx5_add_flow_rule(struct mlx5_flow_table *ft, } } - rule = _mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, action, flow_tag, dest); + rule = _mlx5_add_flow_rule(ft, spec, action, flow_tag, dest); if (sw_action == MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO) { if (!IS_ERR_OR_NULL(rule) && @@ -1292,8 +1301,8 @@ static int update_root_ft_destroy(struct mlx5_flow_table *ft) ft->id); return err; } - root->root_ft = new_root_ft; } + root->root_ft = new_root_ft; return 0; } @@ -1359,40 +1368,47 @@ void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, enum mlx5_flow_namespace_type type) { - struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + struct mlx5_flow_steering *steering = dev->priv.steering; + struct mlx5_flow_root_namespace *root_ns; int prio; struct fs_prio *fs_prio; struct mlx5_flow_namespace *ns; - if (!root_ns) + if (!steering) return NULL; switch (type) { case MLX5_FLOW_NAMESPACE_BYPASS: + case MLX5_FLOW_NAMESPACE_OFFLOADS: + case MLX5_FLOW_NAMESPACE_ETHTOOL: case MLX5_FLOW_NAMESPACE_KERNEL: case MLX5_FLOW_NAMESPACE_LEFTOVERS: case MLX5_FLOW_NAMESPACE_ANCHOR: prio = type; break; case MLX5_FLOW_NAMESPACE_FDB: - if (dev->priv.fdb_root_ns) - return &dev->priv.fdb_root_ns->ns; + if (steering->fdb_root_ns) + return &steering->fdb_root_ns->ns; else return NULL; case MLX5_FLOW_NAMESPACE_ESW_EGRESS: - if (dev->priv.esw_egress_root_ns) - return &dev->priv.esw_egress_root_ns->ns; + if (steering->esw_egress_root_ns) + return &steering->esw_egress_root_ns->ns; else return NULL; case MLX5_FLOW_NAMESPACE_ESW_INGRESS: - if (dev->priv.esw_ingress_root_ns) - return &dev->priv.esw_ingress_root_ns->ns; + if (steering->esw_ingress_root_ns) + return &steering->esw_ingress_root_ns->ns; else return NULL; default: return NULL; } + root_ns = steering->root_ns; + if (!root_ns) + return NULL; + fs_prio = find_prio(&root_ns->ns, prio); if (!fs_prio) return NULL; @@ -1478,13 +1494,13 @@ static bool has_required_caps(struct mlx5_core_dev *dev, struct node_caps *caps) return true; } -static int init_root_tree_recursive(struct mlx5_core_dev *dev, +static int init_root_tree_recursive(struct mlx5_flow_steering *steering, struct init_tree_node *init_node, struct fs_node *fs_parent_node, struct init_tree_node *init_parent_node, int prio) { - int max_ft_level = MLX5_CAP_FLOWTABLE(dev, + int max_ft_level = MLX5_CAP_FLOWTABLE(steering->dev, flow_table_properties_nic_receive. max_ft_level); struct mlx5_flow_namespace *fs_ns; @@ -1495,7 +1511,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, if (init_node->type == FS_TYPE_PRIO) { if ((init_node->min_ft_level > max_ft_level) || - !has_required_caps(dev, &init_node->caps)) + !has_required_caps(steering->dev, &init_node->caps)) return 0; fs_get_obj(fs_ns, fs_parent_node); @@ -1516,7 +1532,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, } prio = 0; for (i = 0; i < init_node->ar_size; i++) { - err = init_root_tree_recursive(dev, &init_node->children[i], + err = init_root_tree_recursive(steering, &init_node->children[i], base, init_node, prio); if (err) return err; @@ -1529,7 +1545,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, return 0; } -static int init_root_tree(struct mlx5_core_dev *dev, +static int init_root_tree(struct mlx5_flow_steering *steering, struct init_tree_node *init_node, struct fs_node *fs_parent_node) { @@ -1539,7 +1555,7 @@ static int init_root_tree(struct mlx5_core_dev *dev, fs_get_obj(fs_ns, fs_parent_node); for (i = 0; i < init_node->ar_size; i++) { - err = init_root_tree_recursive(dev, &init_node->children[i], + err = init_root_tree_recursive(steering, &init_node->children[i], &fs_ns->node, init_node, i); if (err) @@ -1548,7 +1564,7 @@ static int init_root_tree(struct mlx5_core_dev *dev, return 0; } -static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev, +static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_flow_steering *steering, enum fs_flow_table_type table_type) { @@ -1560,7 +1576,7 @@ static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev if (!root_ns) return NULL; - root_ns->dev = dev; + root_ns->dev = steering->dev; root_ns->table_type = table_type; ns = &root_ns->ns; @@ -1615,209 +1631,126 @@ static void set_prio_attrs(struct mlx5_flow_root_namespace *root_ns) #define ANCHOR_PRIO 0 #define ANCHOR_SIZE 1 #define ANCHOR_LEVEL 0 -static int create_anchor_flow_table(struct mlx5_core_dev - *dev) +static int create_anchor_flow_table(struct mlx5_flow_steering *steering) { struct mlx5_flow_namespace *ns = NULL; struct mlx5_flow_table *ft; - ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ANCHOR); + ns = mlx5_get_flow_namespace(steering->dev, MLX5_FLOW_NAMESPACE_ANCHOR); if (!ns) return -EINVAL; ft = mlx5_create_flow_table(ns, ANCHOR_PRIO, ANCHOR_SIZE, ANCHOR_LEVEL); if (IS_ERR(ft)) { - mlx5_core_err(dev, "Failed to create last anchor flow table"); + mlx5_core_err(steering->dev, "Failed to create last anchor flow table"); return PTR_ERR(ft); } return 0; } -static int init_root_ns(struct mlx5_core_dev *dev) +static int init_root_ns(struct mlx5_flow_steering *steering) { - dev->priv.root_ns = create_root_ns(dev, FS_FT_NIC_RX); - if (IS_ERR_OR_NULL(dev->priv.root_ns)) + steering->root_ns = create_root_ns(steering, FS_FT_NIC_RX); + if (IS_ERR_OR_NULL(steering->root_ns)) goto cleanup; - if (init_root_tree(dev, &root_fs, &dev->priv.root_ns->ns.node)) + if (init_root_tree(steering, &root_fs, &steering->root_ns->ns.node)) goto cleanup; - set_prio_attrs(dev->priv.root_ns); + set_prio_attrs(steering->root_ns); - if (create_anchor_flow_table(dev)) + if (create_anchor_flow_table(steering)) goto cleanup; return 0; cleanup: - mlx5_cleanup_fs(dev); + mlx5_cleanup_fs(steering->dev); return -ENOMEM; } -static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev, - struct mlx5_flow_root_namespace *root_ns) +static void clean_tree(struct fs_node *node) { - struct fs_node *prio; - - if (!root_ns) - return; + if (node) { + struct fs_node *iter; + struct fs_node *temp; - if (!list_empty(&root_ns->ns.node.children)) { - prio = list_first_entry(&root_ns->ns.node.children, - struct fs_node, - list); - if (tree_remove_node(prio)) - mlx5_core_warn(dev, - "Flow steering priority wasn't destroyed, refcount > 1\n"); + list_for_each_entry_safe(iter, temp, &node->children, list) + clean_tree(iter); + tree_remove_node(node); } - if (tree_remove_node(&root_ns->ns.node)) - mlx5_core_warn(dev, - "Flow steering namespace wasn't destroyed, refcount > 1\n"); - root_ns = NULL; -} - -static void destroy_flow_tables(struct fs_prio *prio) -{ - struct mlx5_flow_table *iter; - struct mlx5_flow_table *tmp; - - fs_for_each_ft_safe(iter, tmp, prio) - mlx5_destroy_flow_table(iter); } -static void cleanup_root_ns(struct mlx5_core_dev *dev) +static void cleanup_root_ns(struct mlx5_flow_root_namespace *root_ns) { - struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; - struct fs_prio *iter_prio; - - if (!MLX5_CAP_GEN(dev, nic_flow_table)) - return; - if (!root_ns) return; - /* stage 1 */ - fs_for_each_prio(iter_prio, &root_ns->ns) { - struct fs_node *node; - struct mlx5_flow_namespace *iter_ns; - - fs_for_each_ns_or_ft(node, iter_prio) { - if (node->type == FS_TYPE_FLOW_TABLE) - continue; - fs_get_obj(iter_ns, node); - while (!list_empty(&iter_ns->node.children)) { - struct fs_prio *obj_iter_prio2; - struct fs_node *iter_prio2 = - list_first_entry(&iter_ns->node.children, - struct fs_node, - list); - - fs_get_obj(obj_iter_prio2, iter_prio2); - destroy_flow_tables(obj_iter_prio2); - if (tree_remove_node(iter_prio2)) { - mlx5_core_warn(dev, - "Priority %d wasn't destroyed, refcount > 1\n", - obj_iter_prio2->prio); - return; - } - } - } - } - - /* stage 2 */ - fs_for_each_prio(iter_prio, &root_ns->ns) { - while (!list_empty(&iter_prio->node.children)) { - struct fs_node *iter_ns = - list_first_entry(&iter_prio->node.children, - struct fs_node, - list); - if (tree_remove_node(iter_ns)) { - mlx5_core_warn(dev, - "Namespace wasn't destroyed, refcount > 1\n"); - return; - } - } - } - - /* stage 3 */ - while (!list_empty(&root_ns->ns.node.children)) { - struct fs_prio *obj_prio_node; - struct fs_node *prio_node = - list_first_entry(&root_ns->ns.node.children, - struct fs_node, - list); - - fs_get_obj(obj_prio_node, prio_node); - if (tree_remove_node(prio_node)) { - mlx5_core_warn(dev, - "Priority %d wasn't destroyed, refcount > 1\n", - obj_prio_node->prio); - return; - } - } - - if (tree_remove_node(&root_ns->ns.node)) { - mlx5_core_warn(dev, - "root namespace wasn't destroyed, refcount > 1\n"); - return; - } - - dev->priv.root_ns = NULL; + clean_tree(&root_ns->ns.node); } void mlx5_cleanup_fs(struct mlx5_core_dev *dev) { - cleanup_root_ns(dev); - cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); - cleanup_single_prio_root_ns(dev, dev->priv.esw_egress_root_ns); - cleanup_single_prio_root_ns(dev, dev->priv.esw_ingress_root_ns); + struct mlx5_flow_steering *steering = dev->priv.steering; + + if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return; + + cleanup_root_ns(steering->root_ns); + cleanup_root_ns(steering->esw_egress_root_ns); + cleanup_root_ns(steering->esw_ingress_root_ns); + cleanup_root_ns(steering->fdb_root_ns); mlx5_cleanup_fc_stats(dev); + kfree(steering); } -static int init_fdb_root_ns(struct mlx5_core_dev *dev) +static int init_fdb_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.fdb_root_ns = create_root_ns(dev, FS_FT_FDB); - if (!dev->priv.fdb_root_ns) + steering->fdb_root_ns = create_root_ns(steering, FS_FT_FDB); + if (!steering->fdb_root_ns) return -ENOMEM; /* Create single prio */ - prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1); + prio = fs_create_prio(&steering->fdb_root_ns->ns, 0, 1); if (IS_ERR(prio)) { - cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); + cleanup_root_ns(steering->fdb_root_ns); + steering->fdb_root_ns = NULL; return PTR_ERR(prio); } else { return 0; } } -static int init_egress_acl_root_ns(struct mlx5_core_dev *dev) +static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.esw_egress_root_ns = create_root_ns(dev, FS_FT_ESW_EGRESS_ACL); - if (!dev->priv.esw_egress_root_ns) + steering->esw_egress_root_ns = create_root_ns(steering, FS_FT_ESW_EGRESS_ACL); + if (!steering->esw_egress_root_ns) return -ENOMEM; /* create 1 prio*/ - prio = fs_create_prio(&dev->priv.esw_egress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(dev)); + prio = fs_create_prio(&steering->esw_egress_root_ns->ns, 0, + MLX5_TOTAL_VPORTS(steering->dev)); if (IS_ERR(prio)) return PTR_ERR(prio); else return 0; } -static int init_ingress_acl_root_ns(struct mlx5_core_dev *dev) +static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.esw_ingress_root_ns = create_root_ns(dev, FS_FT_ESW_INGRESS_ACL); - if (!dev->priv.esw_ingress_root_ns) + steering->esw_ingress_root_ns = create_root_ns(steering, FS_FT_ESW_INGRESS_ACL); + if (!steering->esw_ingress_root_ns) return -ENOMEM; /* create 1 prio*/ - prio = fs_create_prio(&dev->priv.esw_ingress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(dev)); + prio = fs_create_prio(&steering->esw_ingress_root_ns->ns, 0, + MLX5_TOTAL_VPORTS(steering->dev)); if (IS_ERR(prio)) return PTR_ERR(prio); else @@ -1826,31 +1759,45 @@ static int init_ingress_acl_root_ns(struct mlx5_core_dev *dev) int mlx5_init_fs(struct mlx5_core_dev *dev) { + struct mlx5_flow_steering *steering; int err = 0; + if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) + return 0; + err = mlx5_init_fc_stats(dev); if (err) return err; - if (MLX5_CAP_GEN(dev, nic_flow_table)) { - err = init_root_ns(dev); + steering = kzalloc(sizeof(*steering), GFP_KERNEL); + if (!steering) + return -ENOMEM; + steering->dev = dev; + dev->priv.steering = steering; + + if (MLX5_CAP_GEN(dev, nic_flow_table) && + MLX5_CAP_FLOWTABLE_NIC_RX(dev, ft_support)) { + err = init_root_ns(steering); if (err) goto err; } + if (MLX5_CAP_GEN(dev, eswitch_flow_table)) { - err = init_fdb_root_ns(dev); - if (err) - goto err; - } - if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) { - err = init_egress_acl_root_ns(dev); - if (err) - goto err; - } - if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) { - err = init_ingress_acl_root_ns(dev); - if (err) - goto err; + if (MLX5_CAP_ESW_FLOWTABLE_FDB(dev, ft_support)) { + err = init_fdb_root_ns(steering); + if (err) + goto err; + } + if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) { + err = init_egress_acl_root_ns(steering); + if (err) + goto err; + } + if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) { + err = init_ingress_acl_root_ns(steering); + if (err) + goto err; + } } return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index aa41a7314691..d7ba91a1eea2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -55,6 +55,14 @@ enum fs_fte_status { FS_FTE_STATUS_EXISTING = 1UL << 0, }; +struct mlx5_flow_steering { + struct mlx5_core_dev *dev; + struct mlx5_flow_root_namespace *root_ns; + struct mlx5_flow_root_namespace *fdb_root_ns; + struct mlx5_flow_root_namespace *esw_egress_root_ns; + struct mlx5_flow_root_namespace *esw_ingress_root_ns; +}; + struct fs_node { struct list_head list; struct list_head children; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 75c7ae6a5cc4..77fc1aa26114 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -151,6 +151,12 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev) return err; } + if (MLX5_CAP_GEN(dev, qos)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_QOS); + if (err) + return err; + } + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index a19b59348dd6..1fb3c681df97 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -51,6 +51,7 @@ #ifdef CONFIG_RFS_ACCEL #include <linux/cpu_rmap.h> #endif +#include <net/devlink.h> #include "mlx5_core.h" #include "fs_core.h" #ifdef CONFIG_MLX5_CORE_EN @@ -1144,6 +1145,13 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) dev_err(&pdev->dev, "Failed to init flow steering\n"); goto err_fs; } + + err = mlx5_init_rl_table(dev); + if (err) { + dev_err(&pdev->dev, "Failed to init rate limiting\n"); + goto err_rl; + } + #ifdef CONFIG_MLX5_CORE_EN err = mlx5_eswitch_init(dev); if (err) { @@ -1183,6 +1191,8 @@ err_sriov: mlx5_eswitch_cleanup(dev->priv.eswitch); #endif err_reg_dev: + mlx5_cleanup_rl_table(dev); +err_rl: mlx5_cleanup_fs(dev); err_fs: mlx5_cleanup_mkey_table(dev); @@ -1253,6 +1263,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_eswitch_cleanup(dev->priv.eswitch); #endif + mlx5_cleanup_rl_table(dev); mlx5_cleanup_fs(dev); mlx5_cleanup_mkey_table(dev); mlx5_cleanup_srq_table(dev); @@ -1305,19 +1316,28 @@ struct mlx5_core_event_handler { void *data); }; +static const struct devlink_ops mlx5_devlink_ops = { +#ifdef CONFIG_MLX5_CORE_EN + .eswitch_mode_set = mlx5_devlink_eswitch_mode_set, + .eswitch_mode_get = mlx5_devlink_eswitch_mode_get, +#endif +}; static int init_one(struct pci_dev *pdev, const struct pci_device_id *id) { struct mlx5_core_dev *dev; + struct devlink *devlink; struct mlx5_priv *priv; int err; - dev = kzalloc(sizeof(*dev), GFP_KERNEL); - if (!dev) { + devlink = devlink_alloc(&mlx5_devlink_ops, sizeof(*dev)); + if (!devlink) { dev_err(&pdev->dev, "kzalloc failed\n"); return -ENOMEM; } + + dev = devlink_priv(devlink); priv = &dev->priv; priv->pci_dev_data = id->driver_data; @@ -1354,15 +1374,21 @@ static int init_one(struct pci_dev *pdev, goto clean_health; } + err = devlink_register(devlink, &pdev->dev); + if (err) + goto clean_load; + return 0; +clean_load: + mlx5_unload_one(dev, priv); clean_health: mlx5_health_cleanup(dev); close_pci: mlx5_pci_close(dev, priv); clean_dev: pci_set_drvdata(pdev, NULL); - kfree(dev); + devlink_free(devlink); return err; } @@ -1370,8 +1396,10 @@ clean_dev: static void remove_one(struct pci_dev *pdev) { struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct devlink *devlink = priv_to_devlink(dev); struct mlx5_priv *priv = &dev->priv; + devlink_unregister(devlink); if (mlx5_unload_one(dev, priv)) { dev_err(&dev->pdev->dev, "mlx5_unload_one failed\n"); mlx5_health_cleanup(dev); @@ -1380,7 +1408,7 @@ static void remove_one(struct pci_dev *pdev) mlx5_health_cleanup(dev); mlx5_pci_close(dev, priv); pci_set_drvdata(pdev, NULL); - kfree(dev); + devlink_free(devlink); } static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev, @@ -1508,8 +1536,9 @@ static const struct pci_device_id mlx5_core_pci_table[] = { { PCI_VDEVICE(MELLANOX, 0x1014), MLX5_PCI_DEV_IS_VF}, /* ConnectX-4 VF */ { PCI_VDEVICE(MELLANOX, 0x1015) }, /* ConnectX-4LX */ { PCI_VDEVICE(MELLANOX, 0x1016), MLX5_PCI_DEV_IS_VF}, /* ConnectX-4LX VF */ - { PCI_VDEVICE(MELLANOX, 0x1017) }, /* ConnectX-5 */ + { PCI_VDEVICE(MELLANOX, 0x1017) }, /* ConnectX-5, PCIe 3.0 */ { PCI_VDEVICE(MELLANOX, 0x1018), MLX5_PCI_DEV_IS_VF}, /* ConnectX-5 VF */ + { PCI_VDEVICE(MELLANOX, 0x1019) }, /* ConnectX-5, PCIe 4.0 */ { 0, } }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 3e35611b19c3..752c08127138 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -202,15 +202,24 @@ int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, } EXPORT_SYMBOL_GPL(mlx5_query_port_proto_oper); -int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, - int proto_mask) +int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable, + u32 proto_admin, int proto_mask) { - u32 in[MLX5_ST_SZ_DW(ptys_reg)]; u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + u32 in[MLX5_ST_SZ_DW(ptys_reg)]; + u8 an_disable_admin; + u8 an_disable_cap; + u8 an_status; + + mlx5_query_port_autoneg(dev, proto_mask, &an_status, + &an_disable_cap, &an_disable_admin); + if (!an_disable_cap && an_disable) + return -EPERM; memset(in, 0, sizeof(in)); MLX5_SET(ptys_reg, in, local_port, 1); + MLX5_SET(ptys_reg, in, an_disable_admin, an_disable); MLX5_SET(ptys_reg, in, proto_mask, proto_mask); if (proto_mask == MLX5_PTYS_EN) MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin); @@ -220,7 +229,19 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, return mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out), MLX5_REG_PTYS, 0, 1); } -EXPORT_SYMBOL_GPL(mlx5_set_port_proto); +EXPORT_SYMBOL_GPL(mlx5_set_port_ptys); + +/* This function should be used after setting a port register only */ +void mlx5_toggle_port_link(struct mlx5_core_dev *dev) +{ + enum mlx5_port_status ps; + + mlx5_query_port_admin_status(dev, &ps); + mlx5_set_port_admin_status(dev, MLX5_PORT_DOWN); + if (ps == MLX5_PORT_UP) + mlx5_set_port_admin_status(dev, MLX5_PORT_UP); +} +EXPORT_SYMBOL_GPL(mlx5_toggle_port_link); int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status) @@ -518,6 +539,25 @@ int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx) } EXPORT_SYMBOL_GPL(mlx5_query_port_pfc); +void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask, + u8 *an_status, + u8 *an_disable_cap, u8 *an_disable_admin) +{ + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + + *an_status = 0; + *an_disable_cap = 0; + *an_disable_admin = 0; + + if (mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, 1)) + return; + + *an_status = MLX5_GET(ptys_reg, out, an_status); + *an_disable_cap = MLX5_GET(ptys_reg, out, an_disable_cap); + *an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); +} +EXPORT_SYMBOL_GPL(mlx5_query_port_autoneg); + int mlx5_max_tc(struct mlx5_core_dev *mdev) { u8 num_tc = MLX5_CAP_GEN(mdev, max_tc) ? : 8; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c index b720a274220d..b82d65802d96 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c @@ -418,7 +418,7 @@ int mlx5_core_xrcd_alloc(struct mlx5_core_dev *dev, u32 *xrcdn) if (out.hdr.status) err = mlx5_cmd_status_to_err(&out.hdr); else - *xrcdn = be32_to_cpu(out.xrcdn); + *xrcdn = be32_to_cpu(out.xrcdn) & 0xffffff; return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rl.c b/drivers/net/ethernet/mellanox/mlx5/core/rl.c new file mode 100644 index 000000000000..c07c28bd3d55 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/rl.c @@ -0,0 +1,209 @@ +/* + * Copyright (c) 2013-2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/mlx5/driver.h> +#include <linux/mlx5/cmd.h> +#include "mlx5_core.h" + +/* Finds an entry where we can register the given rate + * If the rate already exists, return the entry where it is registered, + * otherwise return the first available entry. + * If the table is full, return NULL + */ +static struct mlx5_rl_entry *find_rl_entry(struct mlx5_rl_table *table, + u32 rate) +{ + struct mlx5_rl_entry *ret_entry = NULL; + bool empty_found = false; + int i; + + for (i = 0; i < table->max_size; i++) { + if (table->rl_entry[i].rate == rate) + return &table->rl_entry[i]; + if (!empty_found && !table->rl_entry[i].rate) { + empty_found = true; + ret_entry = &table->rl_entry[i]; + } + } + + return ret_entry; +} + +static int mlx5_set_rate_limit_cmd(struct mlx5_core_dev *dev, + u32 rate, u16 index) +{ + u32 in[MLX5_ST_SZ_DW(set_rate_limit_in)]; + u32 out[MLX5_ST_SZ_DW(set_rate_limit_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(set_rate_limit_in, in, opcode, + MLX5_CMD_OP_SET_RATE_LIMIT); + MLX5_SET(set_rate_limit_in, in, rate_limit_index, index); + MLX5_SET(set_rate_limit_in, in, rate_limit, rate); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + out, sizeof(out)); +} + +bool mlx5_rl_is_in_range(struct mlx5_core_dev *dev, u32 rate) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + + return (rate <= table->max_rate && rate >= table->min_rate); +} +EXPORT_SYMBOL(mlx5_rl_is_in_range); + +int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u32 rate, u16 *index) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry; + int err = 0; + + mutex_lock(&table->rl_lock); + + if (!rate || !mlx5_rl_is_in_range(dev, rate)) { + mlx5_core_err(dev, "Invalid rate: %u, should be %u to %u\n", + rate, table->min_rate, table->max_rate); + err = -EINVAL; + goto out; + } + + entry = find_rl_entry(table, rate); + if (!entry) { + mlx5_core_err(dev, "Max number of %u rates reached\n", + table->max_size); + err = -ENOSPC; + goto out; + } + if (entry->refcount) { + /* rate already configured */ + entry->refcount++; + } else { + /* new rate limit */ + err = mlx5_set_rate_limit_cmd(dev, rate, entry->index); + if (err) { + mlx5_core_err(dev, "Failed configuring rate: %u (%d)\n", + rate, err); + goto out; + } + entry->rate = rate; + entry->refcount = 1; + } + *index = entry->index; + +out: + mutex_unlock(&table->rl_lock); + return err; +} +EXPORT_SYMBOL(mlx5_rl_add_rate); + +void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, u32 rate) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry = NULL; + + /* 0 is a reserved value for unlimited rate */ + if (rate == 0) + return; + + mutex_lock(&table->rl_lock); + entry = find_rl_entry(table, rate); + if (!entry || !entry->refcount) { + mlx5_core_warn(dev, "Rate %u is not configured\n", rate); + goto out; + } + + entry->refcount--; + if (!entry->refcount) { + /* need to remove rate */ + mlx5_set_rate_limit_cmd(dev, 0, entry->index); + entry->rate = 0; + } + +out: + mutex_unlock(&table->rl_lock); +} +EXPORT_SYMBOL(mlx5_rl_remove_rate); + +int mlx5_init_rl_table(struct mlx5_core_dev *dev) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + int i; + + mutex_init(&table->rl_lock); + if (!MLX5_CAP_GEN(dev, qos) || !MLX5_CAP_QOS(dev, packet_pacing)) { + table->max_size = 0; + return 0; + } + + /* First entry is reserved for unlimited rate */ + table->max_size = MLX5_CAP_QOS(dev, packet_pacing_rate_table_size) - 1; + table->max_rate = MLX5_CAP_QOS(dev, packet_pacing_max_rate); + table->min_rate = MLX5_CAP_QOS(dev, packet_pacing_min_rate); + + table->rl_entry = kcalloc(table->max_size, sizeof(struct mlx5_rl_entry), + GFP_KERNEL); + if (!table->rl_entry) + return -ENOMEM; + + /* The index represents the index in HW rate limit table + * Index 0 is reserved for unlimited rate + */ + for (i = 0; i < table->max_size; i++) + table->rl_entry[i].index = i + 1; + + /* Index 0 is reserved */ + mlx5_core_info(dev, "Rate limit: %u rates are supported, range: %uMbps to %uMbps\n", + table->max_size, + table->min_rate >> 10, + table->max_rate >> 10); + + return 0; +} + +void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + int i; + + /* Clear all configured rates */ + for (i = 0; i < table->max_size; i++) + if (table->rl_entry[i].rate) + mlx5_set_rate_limit_cmd(dev, 0, + table->rl_entry[i].index); + + kfree(dev->priv.rl_table.rl_entry); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c index d6a3f412ba9f..b380a6bc1f85 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c @@ -167,7 +167,7 @@ int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs) mlx5_core_init_vfs(dev, num_vfs); #ifdef CONFIG_MLX5_CORE_EN - mlx5_eswitch_enable_sriov(dev->priv.eswitch, num_vfs); + mlx5_eswitch_enable_sriov(dev->priv.eswitch, num_vfs, SRIOV_LEGACY); #endif return num_vfs; @@ -209,7 +209,8 @@ int mlx5_sriov_init(struct mlx5_core_dev *dev) mlx5_core_init_vfs(dev, cur_vfs); #ifdef CONFIG_MLX5_CORE_EN if (cur_vfs) - mlx5_eswitch_enable_sriov(dev->priv.eswitch, cur_vfs); + mlx5_eswitch_enable_sriov(dev->priv.eswitch, cur_vfs, + SRIOV_LEGACY); #endif enable_vfs(dev, cur_vfs); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index b69dadcfb897..daf44cd4c566 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -508,6 +508,44 @@ int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid) } EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_node_guid); +int mlx5_modify_nic_vport_node_guid(struct mlx5_core_dev *mdev, + u32 vport, u64 node_guid) +{ + int inlen = MLX5_ST_SZ_BYTES(modify_nic_vport_context_in); + void *nic_vport_context; + u8 *guid; + void *in; + int err; + + if (!vport) + return -EINVAL; + if (!MLX5_CAP_GEN(mdev, vport_group_manager)) + return -EACCES; + if (!MLX5_CAP_ESW(mdev, nic_vport_node_guid_modify)) + return -ENOTSUPP; + + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + MLX5_SET(modify_nic_vport_context_in, in, + field_select.node_guid, 1); + MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport); + MLX5_SET(modify_nic_vport_context_in, in, other_vport, !!vport); + + nic_vport_context = MLX5_ADDR_OF(modify_nic_vport_context_in, + in, nic_vport_context); + guid = MLX5_ADDR_OF(nic_vport_context, nic_vport_context, + node_guid); + MLX5_SET64(nic_vport_context, nic_vport_context, node_guid, node_guid); + + err = mlx5_modify_nic_vport_context(mdev, in, inlen); + + kvfree(in); + + return err; +} + int mlx5_query_nic_vport_qkey_viol_cntr(struct mlx5_core_dev *mdev, u16 *qkey_viol_cntr) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c index f2fd1ef16da7..05de77267d58 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vxlan.c @@ -105,6 +105,9 @@ static void mlx5e_vxlan_add_port(struct work_struct *work) struct mlx5e_vxlan *vxlan; int err; + if (mlx5e_vxlan_lookup_port(priv, port)) + goto free_work; + if (mlx5e_vxlan_core_add_port_cmd(priv->mdev, port)) goto free_work; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.c b/drivers/net/ethernet/mellanox/mlx5/core/wq.c index ce21ee5b2357..821a087c7ae2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.c @@ -75,14 +75,14 @@ int mlx5_wq_cyc_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, err = mlx5_db_alloc_node(mdev, &wq_ctrl->db, param->db_numa_node); if (err) { - mlx5_core_warn(mdev, "mlx5_db_alloc() failed, %d\n", err); + mlx5_core_warn(mdev, "mlx5_db_alloc_node() failed, %d\n", err); return err; } err = mlx5_buf_alloc_node(mdev, mlx5_wq_cyc_get_byte_size(wq), &wq_ctrl->buf, param->buf_numa_node); if (err) { - mlx5_core_warn(mdev, "mlx5_buf_alloc() failed, %d\n", err); + mlx5_core_warn(mdev, "mlx5_buf_alloc_node() failed, %d\n", err); goto err_db_free; } @@ -111,14 +111,14 @@ int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, err = mlx5_db_alloc_node(mdev, &wq_ctrl->db, param->db_numa_node); if (err) { - mlx5_core_warn(mdev, "mlx5_db_alloc() failed, %d\n", err); + mlx5_core_warn(mdev, "mlx5_db_alloc_node() failed, %d\n", err); return err; } err = mlx5_buf_alloc_node(mdev, mlx5_cqwq_get_byte_size(wq), &wq_ctrl->buf, param->buf_numa_node); if (err) { - mlx5_core_warn(mdev, "mlx5_buf_alloc() failed, %d\n", err); + mlx5_core_warn(mdev, "mlx5_buf_alloc_node() failed, %d\n", err); goto err_db_free; } @@ -148,13 +148,14 @@ int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, err = mlx5_db_alloc_node(mdev, &wq_ctrl->db, param->db_numa_node); if (err) { - mlx5_core_warn(mdev, "mlx5_db_alloc() failed, %d\n", err); + mlx5_core_warn(mdev, "mlx5_db_alloc_node() failed, %d\n", err); return err; } - err = mlx5_buf_alloc(mdev, mlx5_wq_ll_get_byte_size(wq), &wq_ctrl->buf); + err = mlx5_buf_alloc_node(mdev, mlx5_wq_ll_get_byte_size(wq), + &wq_ctrl->buf, param->buf_numa_node); if (err) { - mlx5_core_warn(mdev, "mlx5_buf_alloc() failed, %d\n", err); + mlx5_core_warn(mdev, "mlx5_buf_alloc_node() failed, %d\n", err); goto err_db_free; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile index 9b5ebf84c051..d20ae1838a64 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/Makefile +++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile @@ -7,5 +7,6 @@ obj-$(CONFIG_MLXSW_SWITCHX2) += mlxsw_switchx2.o mlxsw_switchx2-objs := switchx2.o obj-$(CONFIG_MLXSW_SPECTRUM) += mlxsw_spectrum.o mlxsw_spectrum-objs := spectrum.o spectrum_buffers.o \ - spectrum_switchdev.o + spectrum_switchdev.o spectrum_router.o \ + spectrum_kvdl.o mlxsw_spectrum-$(CONFIG_MLXSW_SPECTRUM_DCB) += spectrum_dcb.o diff --git a/drivers/net/ethernet/mellanox/mlxsw/cmd.h b/drivers/net/ethernet/mellanox/mlxsw/cmd.h index cd63b8263688..f9cd6e3f7709 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/cmd.h +++ b/drivers/net/ethernet/mellanox/mlxsw/cmd.h @@ -607,6 +607,24 @@ MLXSW_ITEM32(cmd_mbox, config_profile, */ MLXSW_ITEM32(cmd_mbox, config_profile, set_ar_sec, 0x0C, 15, 1); +/* cmd_mbox_config_set_kvd_linear_size + * Capability bit. Setting a bit to 1 configures the profile + * according to the mailbox contents. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_linear_size, 0x0C, 24, 1); + +/* cmd_mbox_config_set_kvd_hash_single_size + * Capability bit. Setting a bit to 1 configures the profile + * according to the mailbox contents. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_hash_single_size, 0x0C, 25, 1); + +/* cmd_mbox_config_set_kvd_hash_double_size + * Capability bit. Setting a bit to 1 configures the profile + * according to the mailbox contents. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_hash_double_size, 0x0C, 26, 1); + /* cmd_mbox_config_profile_max_vepa_channels * Maximum number of VEPA channels per port (0 through 16) * 0 - multi-channel VEPA is disabled @@ -733,6 +751,31 @@ MLXSW_ITEM32(cmd_mbox, config_profile, adaptive_routing_group_cap, 0x4C, 0, 16); */ MLXSW_ITEM32(cmd_mbox, config_profile, arn, 0x50, 31, 1); +/* cmd_mbox_config_kvd_linear_size + * KVD Linear Size + * Valid for Spectrum only + * Allowed values are 128*N where N=0 or higher + */ +MLXSW_ITEM32(cmd_mbox, config_profile, kvd_linear_size, 0x54, 0, 24); + +/* cmd_mbox_config_kvd_hash_single_size + * KVD Hash single-entries size + * Valid for Spectrum only + * Allowed values are 128*N where N=0 or higher + * Must be greater or equal to cap_min_kvd_hash_single_size + * Must be smaller or equal to cap_kvd_size - kvd_linear_size + */ +MLXSW_ITEM32(cmd_mbox, config_profile, kvd_hash_single_size, 0x58, 0, 24); + +/* cmd_mbox_config_kvd_hash_double_size + * KVD Hash double-entries size (units of single-size entries) + * Valid for Spectrum only + * Allowed values are 128*N where N=0 or higher + * Must be either 0 or greater or equal to cap_min_kvd_hash_double_size + * Must be smaller or equal to cap_kvd_size - kvd_linear_size + */ +MLXSW_ITEM32(cmd_mbox, config_profile, kvd_hash_double_size, 0x5C, 0, 24); + /* cmd_mbox_config_profile_swid_config_mask * Modify Switch Partition Configuration mask. When set, the configu- * ration value for the Switch Partition are taken from the mailbox. diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index b0a0b01bb4ef..01ae54826d5c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -1736,7 +1736,7 @@ static int __init mlxsw_core_module_init(void) { int err; - mlxsw_wq = create_workqueue(mlxsw_core_driver_name); + mlxsw_wq = alloc_workqueue(mlxsw_core_driver_name, WQ_MEM_RECLAIM, 0); if (!mlxsw_wq) return -ENOMEM; mlxsw_core_dbg_root = debugfs_create_dir(mlxsw_core_driver_name, NULL); diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index 436bc49df6ab..2fe385cce203 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -190,7 +190,8 @@ struct mlxsw_config_profile { used_max_ib_mc:1, used_max_pkey:1, used_ar_sec:1, - used_adaptive_routing_group_cap:1; + used_adaptive_routing_group_cap:1, + used_kvd_sizes:1; u8 max_vepa_channels; u16 max_lag; u16 max_port_per_lag; @@ -211,6 +212,9 @@ struct mlxsw_config_profile { u8 ar_sec; u16 adaptive_routing_group_cap; u8 arn; + u32 kvd_linear_size; + u32 kvd_hash_single_size; + u32 kvd_hash_double_size; struct mlxsw_swid_config swid_config[MLXSW_CONFIG_PROFILE_SWID_COUNT]; }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index 7f4173c8eda3..ddbc9f22278d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -1255,6 +1255,20 @@ static int mlxsw_pci_config_profile(struct mlxsw_pci *mlxsw_pci, char *mbox, mlxsw_cmd_mbox_config_profile_adaptive_routing_group_cap_set( mbox, profile->adaptive_routing_group_cap); } + if (profile->used_kvd_sizes) { + mlxsw_cmd_mbox_config_profile_set_kvd_linear_size_set( + mbox, 1); + mlxsw_cmd_mbox_config_profile_kvd_linear_size_set( + mbox, profile->kvd_linear_size); + mlxsw_cmd_mbox_config_profile_set_kvd_hash_single_size_set( + mbox, 1); + mlxsw_cmd_mbox_config_profile_kvd_hash_single_size_set( + mbox, profile->kvd_hash_single_size); + mlxsw_cmd_mbox_config_profile_set_kvd_hash_double_size_set( + mbox, 1); + mlxsw_cmd_mbox_config_profile_kvd_hash_double_size_set( + mbox, profile->kvd_hash_double_size); + } for (i = 0; i < MLXSW_CONFIG_PROFILE_SWID_COUNT; i++) mlxsw_pci_config_profile_swid_config(mlxsw_pci, mbox, i, diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 1977e7a5c530..0cc148566677 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -1,9 +1,10 @@ /* * drivers/net/ethernet/mellanox/mlxsw/reg.h * Copyright (c) 2015 Mellanox Technologies. All rights reserved. - * Copyright (c) 2015 Ido Schimmel <idosch@mellanox.com> + * Copyright (c) 2015-2016 Ido Schimmel <idosch@mellanox.com> * Copyright (c) 2015 Elad Raz <eladr@mellanox.com> - * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com> + * Copyright (c) 2015-2016 Jiri Pirko <jiri@mellanox.com> + * Copyright (c) 2016 Yotam Gigi <yotamg@mellanox.com> * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -386,7 +387,9 @@ enum mlxsw_reg_sfd_rec_action { /* forward and trap, trap_id is FDB_TRAP */ MLXSW_REG_SFD_REC_ACTION_MIRROR_TO_CPU = 1, /* trap and do not forward, trap_id is FDB_TRAP */ - MLXSW_REG_SFD_REC_ACTION_TRAP = 3, + MLXSW_REG_SFD_REC_ACTION_TRAP = 2, + /* forward to IP router */ + MLXSW_REG_SFD_REC_ACTION_FORWARD_IP_ROUTER = 3, MLXSW_REG_SFD_REC_ACTION_DISCARD_ERROR = 15, }; @@ -3186,6 +3189,1183 @@ static inline void mlxsw_reg_hpkt_pack(char *payload, u8 action, u16 trap_id) mlxsw_reg_hpkt_ctrl_set(payload, MLXSW_REG_HPKT_CTRL_PACKET_DEFAULT); } +/* RGCR - Router General Configuration Register + * -------------------------------------------- + * The register is used for setting up the router configuration. + */ +#define MLXSW_REG_RGCR_ID 0x8001 +#define MLXSW_REG_RGCR_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_rgcr = { + .id = MLXSW_REG_RGCR_ID, + .len = MLXSW_REG_RGCR_LEN, +}; + +/* reg_rgcr_ipv4_en + * IPv4 router enable. + * Access: RW + */ +MLXSW_ITEM32(reg, rgcr, ipv4_en, 0x00, 31, 1); + +/* reg_rgcr_ipv6_en + * IPv6 router enable. + * Access: RW + */ +MLXSW_ITEM32(reg, rgcr, ipv6_en, 0x00, 30, 1); + +/* reg_rgcr_max_router_interfaces + * Defines the maximum number of active router interfaces for all virtual + * routers. + * Access: RW + */ +MLXSW_ITEM32(reg, rgcr, max_router_interfaces, 0x10, 0, 16); + +/* reg_rgcr_usp + * Update switch priority and packet color. + * 0 - Preserve the value of Switch Priority and packet color. + * 1 - Recalculate the value of Switch Priority and packet color. + * Access: RW + * + * Note: Not supported by SwitchX and SwitchX-2. + */ +MLXSW_ITEM32(reg, rgcr, usp, 0x18, 20, 1); + +/* reg_rgcr_pcp_rw + * Indicates how to handle the pcp_rewrite_en value: + * 0 - Preserve the value of pcp_rewrite_en. + * 2 - Disable PCP rewrite. + * 3 - Enable PCP rewrite. + * Access: RW + * + * Note: Not supported by SwitchX and SwitchX-2. + */ +MLXSW_ITEM32(reg, rgcr, pcp_rw, 0x18, 16, 2); + +/* reg_rgcr_activity_dis + * Activity disable: + * 0 - Activity will be set when an entry is hit (default). + * 1 - Activity will not be set when an entry is hit. + * + * Bit 0 - Disable activity bit in Router Algorithmic LPM Unicast Entry + * (RALUE). + * Bit 1 - Disable activity bit in Router Algorithmic LPM Unicast Host + * Entry (RAUHT). + * Bits 2:7 are reserved. + * Access: RW + * + * Note: Not supported by SwitchX, SwitchX-2 and Switch-IB. + */ +MLXSW_ITEM32(reg, rgcr, activity_dis, 0x20, 0, 8); + +static inline void mlxsw_reg_rgcr_pack(char *payload, bool ipv4_en) +{ + MLXSW_REG_ZERO(rgcr, payload); + mlxsw_reg_rgcr_ipv4_en_set(payload, ipv4_en); +} + +/* RITR - Router Interface Table Register + * -------------------------------------- + * The register is used to configure the router interface table. + */ +#define MLXSW_REG_RITR_ID 0x8002 +#define MLXSW_REG_RITR_LEN 0x40 + +static const struct mlxsw_reg_info mlxsw_reg_ritr = { + .id = MLXSW_REG_RITR_ID, + .len = MLXSW_REG_RITR_LEN, +}; + +/* reg_ritr_enable + * Enables routing on the router interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, enable, 0x00, 31, 1); + +/* reg_ritr_ipv4 + * IPv4 routing enable. Enables routing of IPv4 traffic on the router + * interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv4, 0x00, 29, 1); + +/* reg_ritr_ipv6 + * IPv6 routing enable. Enables routing of IPv6 traffic on the router + * interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv6, 0x00, 28, 1); + +enum mlxsw_reg_ritr_if_type { + MLXSW_REG_RITR_VLAN_IF, + MLXSW_REG_RITR_FID_IF, + MLXSW_REG_RITR_SP_IF, +}; + +/* reg_ritr_type + * Router interface type. + * 0 - VLAN interface. + * 1 - FID interface. + * 2 - Sub-port interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, type, 0x00, 23, 3); + +enum { + MLXSW_REG_RITR_RIF_CREATE, + MLXSW_REG_RITR_RIF_DEL, +}; + +/* reg_ritr_op + * Opcode: + * 0 - Create or edit RIF. + * 1 - Delete RIF. + * Reserved for SwitchX-2. For Spectrum, editing of interface properties + * is not supported. An interface must be deleted and re-created in order + * to update properties. + * Access: WO + */ +MLXSW_ITEM32(reg, ritr, op, 0x00, 20, 2); + +/* reg_ritr_rif + * Router interface index. A pointer to the Router Interface Table. + * Access: Index + */ +MLXSW_ITEM32(reg, ritr, rif, 0x00, 0, 16); + +/* reg_ritr_ipv4_fe + * IPv4 Forwarding Enable. + * Enables routing of IPv4 traffic on the router interface. When disabled, + * forwarding is blocked but local traffic (traps and IP2ME) will be enabled. + * Not supported in SwitchX-2. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv4_fe, 0x04, 29, 1); + +/* reg_ritr_ipv6_fe + * IPv6 Forwarding Enable. + * Enables routing of IPv6 traffic on the router interface. When disabled, + * forwarding is blocked but local traffic (traps and IP2ME) will be enabled. + * Not supported in SwitchX-2. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv6_fe, 0x04, 28, 1); + +/* reg_ritr_virtual_router + * Virtual router ID associated with the router interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, virtual_router, 0x04, 0, 16); + +/* reg_ritr_mtu + * Router interface MTU. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, mtu, 0x34, 0, 16); + +/* reg_ritr_if_swid + * Switch partition ID. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, if_swid, 0x08, 24, 8); + +/* reg_ritr_if_mac + * Router interface MAC address. + * In Spectrum, all MAC addresses must have the same 38 MSBits. + * Access: RW + */ +MLXSW_ITEM_BUF(reg, ritr, if_mac, 0x12, 6); + +/* VLAN Interface */ + +/* reg_ritr_vlan_if_vid + * VLAN ID. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, vlan_if_vid, 0x08, 0, 12); + +/* FID Interface */ + +/* reg_ritr_fid_if_fid + * Filtering ID. Used to connect a bridge to the router. Only FIDs from + * the vFID range are supported. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, fid_if_fid, 0x08, 0, 16); + +static inline void mlxsw_reg_ritr_fid_set(char *payload, + enum mlxsw_reg_ritr_if_type rif_type, + u16 fid) +{ + if (rif_type == MLXSW_REG_RITR_FID_IF) + mlxsw_reg_ritr_fid_if_fid_set(payload, fid); + else + mlxsw_reg_ritr_vlan_if_vid_set(payload, fid); +} + +/* Sub-port Interface */ + +/* reg_ritr_sp_if_lag + * LAG indication. When this bit is set the system_port field holds the + * LAG identifier. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, sp_if_lag, 0x08, 24, 1); + +/* reg_ritr_sp_system_port + * Port unique indentifier. When lag bit is set, this field holds the + * lag_id in bits 0:9. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, sp_if_system_port, 0x08, 0, 16); + +/* reg_ritr_sp_if_vid + * VLAN ID. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, sp_if_vid, 0x18, 0, 12); + +static inline void mlxsw_reg_ritr_rif_pack(char *payload, u16 rif) +{ + MLXSW_REG_ZERO(ritr, payload); + mlxsw_reg_ritr_rif_set(payload, rif); +} + +static inline void mlxsw_reg_ritr_sp_if_pack(char *payload, bool lag, + u16 system_port, u16 vid) +{ + mlxsw_reg_ritr_sp_if_lag_set(payload, lag); + mlxsw_reg_ritr_sp_if_system_port_set(payload, system_port); + mlxsw_reg_ritr_sp_if_vid_set(payload, vid); +} + +static inline void mlxsw_reg_ritr_pack(char *payload, bool enable, + enum mlxsw_reg_ritr_if_type type, + u16 rif, u16 mtu, const char *mac) +{ + bool op = enable ? MLXSW_REG_RITR_RIF_CREATE : MLXSW_REG_RITR_RIF_DEL; + + MLXSW_REG_ZERO(ritr, payload); + mlxsw_reg_ritr_enable_set(payload, enable); + mlxsw_reg_ritr_ipv4_set(payload, 1); + mlxsw_reg_ritr_type_set(payload, type); + mlxsw_reg_ritr_op_set(payload, op); + mlxsw_reg_ritr_rif_set(payload, rif); + mlxsw_reg_ritr_ipv4_fe_set(payload, 1); + mlxsw_reg_ritr_mtu_set(payload, mtu); + mlxsw_reg_ritr_if_mac_memcpy_to(payload, mac); +} + +/* RATR - Router Adjacency Table Register + * -------------------------------------- + * The RATR register is used to configure the Router Adjacency (next-hop) + * Table. + */ +#define MLXSW_REG_RATR_ID 0x8008 +#define MLXSW_REG_RATR_LEN 0x2C + +static const struct mlxsw_reg_info mlxsw_reg_ratr = { + .id = MLXSW_REG_RATR_ID, + .len = MLXSW_REG_RATR_LEN, +}; + +enum mlxsw_reg_ratr_op { + /* Read */ + MLXSW_REG_RATR_OP_QUERY_READ = 0, + /* Read and clear activity */ + MLXSW_REG_RATR_OP_QUERY_READ_CLEAR = 2, + /* Write Adjacency entry */ + MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY = 1, + /* Write Adjacency entry only if the activity is cleared. + * The write may not succeed if the activity is set. There is not + * direct feedback if the write has succeeded or not, however + * the get will reveal the actual entry (SW can compare the get + * response to the set command). + */ + MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY_ON_ACTIVITY = 3, +}; + +/* reg_ratr_op + * Note that Write operation may also be used for updating + * counter_set_type and counter_index. In this case all other + * fields must not be updated. + * Access: OP + */ +MLXSW_ITEM32(reg, ratr, op, 0x00, 28, 4); + +/* reg_ratr_v + * Valid bit. Indicates if the adjacency entry is valid. + * Note: the device may need some time before reusing an invalidated + * entry. During this time the entry can not be reused. It is + * recommended to use another entry before reusing an invalidated + * entry (e.g. software can put it at the end of the list for + * reusing). Trying to access an invalidated entry not yet cleared + * by the device results with failure indicating "Try Again" status. + * When valid is '0' then egress_router_interface,trap_action, + * adjacency_parameters and counters are reserved + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, v, 0x00, 24, 1); + +/* reg_ratr_a + * Activity. Set for new entries. Set if a packet lookup has hit on + * the specific entry. To clear the a bit, use "clear activity". + * Access: RO + */ +MLXSW_ITEM32(reg, ratr, a, 0x00, 16, 1); + +/* reg_ratr_adjacency_index_low + * Bits 15:0 of index into the adjacency table. + * For SwitchX and SwitchX-2, the adjacency table is linear and + * used for adjacency entries only. + * For Spectrum, the index is to the KVD linear. + * Access: Index + */ +MLXSW_ITEM32(reg, ratr, adjacency_index_low, 0x04, 0, 16); + +/* reg_ratr_egress_router_interface + * Range is 0 .. cap_max_router_interfaces - 1 + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, egress_router_interface, 0x08, 0, 16); + +enum mlxsw_reg_ratr_trap_action { + MLXSW_REG_RATR_TRAP_ACTION_NOP, + MLXSW_REG_RATR_TRAP_ACTION_TRAP, + MLXSW_REG_RATR_TRAP_ACTION_MIRROR_TO_CPU, + MLXSW_REG_RATR_TRAP_ACTION_MIRROR, + MLXSW_REG_RATR_TRAP_ACTION_DISCARD_ERRORS, +}; + +/* reg_ratr_trap_action + * see mlxsw_reg_ratr_trap_action + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, trap_action, 0x0C, 28, 4); + +enum mlxsw_reg_ratr_trap_id { + MLXSW_REG_RATR_TRAP_ID_RTR_EGRESS0 = 0, + MLXSW_REG_RATR_TRAP_ID_RTR_EGRESS1 = 1, +}; + +/* reg_ratr_adjacency_index_high + * Bits 23:16 of the adjacency_index. + * Access: Index + */ +MLXSW_ITEM32(reg, ratr, adjacency_index_high, 0x0C, 16, 8); + +/* reg_ratr_trap_id + * Trap ID to be reported to CPU. + * Trap-ID is RTR_EGRESS0 or RTR_EGRESS1. + * For trap_action of NOP, MIRROR and DISCARD_ERROR + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, trap_id, 0x0C, 0, 8); + +/* reg_ratr_eth_destination_mac + * MAC address of the destination next-hop. + * Access: RW + */ +MLXSW_ITEM_BUF(reg, ratr, eth_destination_mac, 0x12, 6); + +static inline void +mlxsw_reg_ratr_pack(char *payload, + enum mlxsw_reg_ratr_op op, bool valid, + u32 adjacency_index, u16 egress_rif) +{ + MLXSW_REG_ZERO(ratr, payload); + mlxsw_reg_ratr_op_set(payload, op); + mlxsw_reg_ratr_v_set(payload, valid); + mlxsw_reg_ratr_adjacency_index_low_set(payload, adjacency_index); + mlxsw_reg_ratr_adjacency_index_high_set(payload, adjacency_index >> 16); + mlxsw_reg_ratr_egress_router_interface_set(payload, egress_rif); +} + +static inline void mlxsw_reg_ratr_eth_entry_pack(char *payload, + const char *dest_mac) +{ + mlxsw_reg_ratr_eth_destination_mac_memcpy_to(payload, dest_mac); +} + +/* RALTA - Router Algorithmic LPM Tree Allocation Register + * ------------------------------------------------------- + * RALTA is used to allocate the LPM trees of the SHSPM method. + */ +#define MLXSW_REG_RALTA_ID 0x8010 +#define MLXSW_REG_RALTA_LEN 0x04 + +static const struct mlxsw_reg_info mlxsw_reg_ralta = { + .id = MLXSW_REG_RALTA_ID, + .len = MLXSW_REG_RALTA_LEN, +}; + +/* reg_ralta_op + * opcode (valid for Write, must be 0 on Read) + * 0 - allocate a tree + * 1 - deallocate a tree + * Access: OP + */ +MLXSW_ITEM32(reg, ralta, op, 0x00, 28, 2); + +enum mlxsw_reg_ralxx_protocol { + MLXSW_REG_RALXX_PROTOCOL_IPV4, + MLXSW_REG_RALXX_PROTOCOL_IPV6, +}; + +/* reg_ralta_protocol + * Protocol. + * Deallocation opcode: Reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, ralta, protocol, 0x00, 24, 4); + +/* reg_ralta_tree_id + * An identifier (numbered from 1..cap_shspm_max_trees-1) representing + * the tree identifier (managed by software). + * Note that tree_id 0 is allocated for a default-route tree. + * Access: Index + */ +MLXSW_ITEM32(reg, ralta, tree_id, 0x00, 0, 8); + +static inline void mlxsw_reg_ralta_pack(char *payload, bool alloc, + enum mlxsw_reg_ralxx_protocol protocol, + u8 tree_id) +{ + MLXSW_REG_ZERO(ralta, payload); + mlxsw_reg_ralta_op_set(payload, !alloc); + mlxsw_reg_ralta_protocol_set(payload, protocol); + mlxsw_reg_ralta_tree_id_set(payload, tree_id); +} + +/* RALST - Router Algorithmic LPM Structure Tree Register + * ------------------------------------------------------ + * RALST is used to set and query the structure of an LPM tree. + * The structure of the tree must be sorted as a sorted binary tree, while + * each node is a bin that is tagged as the length of the prefixes the lookup + * will refer to. Therefore, bin X refers to a set of entries with prefixes + * of X bits to match with the destination address. The bin 0 indicates + * the default action, when there is no match of any prefix. + */ +#define MLXSW_REG_RALST_ID 0x8011 +#define MLXSW_REG_RALST_LEN 0x104 + +static const struct mlxsw_reg_info mlxsw_reg_ralst = { + .id = MLXSW_REG_RALST_ID, + .len = MLXSW_REG_RALST_LEN, +}; + +/* reg_ralst_root_bin + * The bin number of the root bin. + * 0<root_bin=<(length of IP address) + * For a default-route tree configure 0xff + * Access: RW + */ +MLXSW_ITEM32(reg, ralst, root_bin, 0x00, 16, 8); + +/* reg_ralst_tree_id + * Tree identifier numbered from 1..(cap_shspm_max_trees-1). + * Access: Index + */ +MLXSW_ITEM32(reg, ralst, tree_id, 0x00, 0, 8); + +#define MLXSW_REG_RALST_BIN_NO_CHILD 0xff +#define MLXSW_REG_RALST_BIN_OFFSET 0x04 +#define MLXSW_REG_RALST_BIN_COUNT 128 + +/* reg_ralst_left_child_bin + * Holding the children of the bin according to the stored tree's structure. + * For trees composed of less than 4 blocks, the bins in excess are reserved. + * Note that tree_id 0 is allocated for a default-route tree, bins are 0xff + * Access: RW + */ +MLXSW_ITEM16_INDEXED(reg, ralst, left_child_bin, 0x04, 8, 8, 0x02, 0x00, false); + +/* reg_ralst_right_child_bin + * Holding the children of the bin according to the stored tree's structure. + * For trees composed of less than 4 blocks, the bins in excess are reserved. + * Note that tree_id 0 is allocated for a default-route tree, bins are 0xff + * Access: RW + */ +MLXSW_ITEM16_INDEXED(reg, ralst, right_child_bin, 0x04, 0, 8, 0x02, 0x00, + false); + +static inline void mlxsw_reg_ralst_pack(char *payload, u8 root_bin, u8 tree_id) +{ + MLXSW_REG_ZERO(ralst, payload); + + /* Initialize all bins to have no left or right child */ + memset(payload + MLXSW_REG_RALST_BIN_OFFSET, + MLXSW_REG_RALST_BIN_NO_CHILD, MLXSW_REG_RALST_BIN_COUNT * 2); + + mlxsw_reg_ralst_root_bin_set(payload, root_bin); + mlxsw_reg_ralst_tree_id_set(payload, tree_id); +} + +static inline void mlxsw_reg_ralst_bin_pack(char *payload, u8 bin_number, + u8 left_child_bin, + u8 right_child_bin) +{ + int bin_index = bin_number - 1; + + mlxsw_reg_ralst_left_child_bin_set(payload, bin_index, left_child_bin); + mlxsw_reg_ralst_right_child_bin_set(payload, bin_index, + right_child_bin); +} + +/* RALTB - Router Algorithmic LPM Tree Binding Register + * ---------------------------------------------------- + * RALTB is used to bind virtual router and protocol to an allocated LPM tree. + */ +#define MLXSW_REG_RALTB_ID 0x8012 +#define MLXSW_REG_RALTB_LEN 0x04 + +static const struct mlxsw_reg_info mlxsw_reg_raltb = { + .id = MLXSW_REG_RALTB_ID, + .len = MLXSW_REG_RALTB_LEN, +}; + +/* reg_raltb_virtual_router + * Virtual Router ID + * Range is 0..cap_max_virtual_routers-1 + * Access: Index + */ +MLXSW_ITEM32(reg, raltb, virtual_router, 0x00, 16, 16); + +/* reg_raltb_protocol + * Protocol. + * Access: Index + */ +MLXSW_ITEM32(reg, raltb, protocol, 0x00, 12, 4); + +/* reg_raltb_tree_id + * Tree to be used for the {virtual_router, protocol} + * Tree identifier numbered from 1..(cap_shspm_max_trees-1). + * By default, all Unicast IPv4 and IPv6 are bound to tree_id 0. + * Access: RW + */ +MLXSW_ITEM32(reg, raltb, tree_id, 0x00, 0, 8); + +static inline void mlxsw_reg_raltb_pack(char *payload, u16 virtual_router, + enum mlxsw_reg_ralxx_protocol protocol, + u8 tree_id) +{ + MLXSW_REG_ZERO(raltb, payload); + mlxsw_reg_raltb_virtual_router_set(payload, virtual_router); + mlxsw_reg_raltb_protocol_set(payload, protocol); + mlxsw_reg_raltb_tree_id_set(payload, tree_id); +} + +/* RALUE - Router Algorithmic LPM Unicast Entry Register + * ----------------------------------------------------- + * RALUE is used to configure and query LPM entries that serve + * the Unicast protocols. + */ +#define MLXSW_REG_RALUE_ID 0x8013 +#define MLXSW_REG_RALUE_LEN 0x38 + +static const struct mlxsw_reg_info mlxsw_reg_ralue = { + .id = MLXSW_REG_RALUE_ID, + .len = MLXSW_REG_RALUE_LEN, +}; + +/* reg_ralue_protocol + * Protocol. + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, protocol, 0x00, 24, 4); + +enum mlxsw_reg_ralue_op { + /* Read operation. If entry doesn't exist, the operation fails. */ + MLXSW_REG_RALUE_OP_QUERY_READ = 0, + /* Clear on read operation. Used to read entry and + * clear Activity bit. + */ + MLXSW_REG_RALUE_OP_QUERY_CLEAR = 1, + /* Write operation. Used to write a new entry to the table. All RW + * fields are written for new entry. Activity bit is set + * for new entries. + */ + MLXSW_REG_RALUE_OP_WRITE_WRITE = 0, + /* Update operation. Used to update an existing route entry and + * only update the RW fields that are detailed in the field + * op_u_mask. If entry doesn't exist, the operation fails. + */ + MLXSW_REG_RALUE_OP_WRITE_UPDATE = 1, + /* Clear activity. The Activity bit (the field a) is cleared + * for the entry. + */ + MLXSW_REG_RALUE_OP_WRITE_CLEAR = 2, + /* Delete operation. Used to delete an existing entry. If entry + * doesn't exist, the operation fails. + */ + MLXSW_REG_RALUE_OP_WRITE_DELETE = 3, +}; + +/* reg_ralue_op + * Operation. + * Access: OP + */ +MLXSW_ITEM32(reg, ralue, op, 0x00, 20, 3); + +/* reg_ralue_a + * Activity. Set for new entries. Set if a packet lookup has hit on the + * specific entry, only if the entry is a route. To clear the a bit, use + * "clear activity" op. + * Enabled by activity_dis in RGCR + * Access: RO + */ +MLXSW_ITEM32(reg, ralue, a, 0x00, 16, 1); + +/* reg_ralue_virtual_router + * Virtual Router ID + * Range is 0..cap_max_virtual_routers-1 + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, virtual_router, 0x04, 16, 16); + +#define MLXSW_REG_RALUE_OP_U_MASK_ENTRY_TYPE BIT(0) +#define MLXSW_REG_RALUE_OP_U_MASK_BMP_LEN BIT(1) +#define MLXSW_REG_RALUE_OP_U_MASK_ACTION BIT(2) + +/* reg_ralue_op_u_mask + * opcode update mask. + * On read operation, this field is reserved. + * This field is valid for update opcode, otherwise - reserved. + * This field is a bitmask of the fields that should be updated. + * Access: WO + */ +MLXSW_ITEM32(reg, ralue, op_u_mask, 0x04, 8, 3); + +/* reg_ralue_prefix_len + * Number of bits in the prefix of the LPM route. + * Note that for IPv6 prefixes, if prefix_len>64 the entry consumes + * two entries in the physical HW table. + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, prefix_len, 0x08, 0, 8); + +/* reg_ralue_dip* + * The prefix of the route or of the marker that the object of the LPM + * is compared with. The most significant bits of the dip are the prefix. + * The list significant bits must be '0' if the prefix_len is smaller + * than 128 for IPv6 or smaller than 32 for IPv4. + * IPv4 address uses bits dip[31:0] and bits dip[127:32] are reserved. + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, dip4, 0x18, 0, 32); + +enum mlxsw_reg_ralue_entry_type { + MLXSW_REG_RALUE_ENTRY_TYPE_MARKER_ENTRY = 1, + MLXSW_REG_RALUE_ENTRY_TYPE_ROUTE_ENTRY = 2, + MLXSW_REG_RALUE_ENTRY_TYPE_MARKER_AND_ROUTE_ENTRY = 3, +}; + +/* reg_ralue_entry_type + * Entry type. + * Note - for Marker entries, the action_type and action fields are reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, entry_type, 0x1C, 30, 2); + +/* reg_ralue_bmp_len + * The best match prefix length in the case that there is no match for + * longer prefixes. + * If (entry_type != MARKER_ENTRY), bmp_len must be equal to prefix_len + * Note for any update operation with entry_type modification this + * field must be set. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, bmp_len, 0x1C, 16, 8); + +enum mlxsw_reg_ralue_action_type { + MLXSW_REG_RALUE_ACTION_TYPE_REMOTE, + MLXSW_REG_RALUE_ACTION_TYPE_LOCAL, + MLXSW_REG_RALUE_ACTION_TYPE_IP2ME, +}; + +/* reg_ralue_action_type + * Action Type + * Indicates how the IP address is connected. + * It can be connected to a local subnet through local_erif or can be + * on a remote subnet connected through a next-hop router, + * or transmitted to the CPU. + * Reserved when entry_type = MARKER_ENTRY + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, action_type, 0x1C, 0, 2); + +enum mlxsw_reg_ralue_trap_action { + MLXSW_REG_RALUE_TRAP_ACTION_NOP, + MLXSW_REG_RALUE_TRAP_ACTION_TRAP, + MLXSW_REG_RALUE_TRAP_ACTION_MIRROR_TO_CPU, + MLXSW_REG_RALUE_TRAP_ACTION_MIRROR, + MLXSW_REG_RALUE_TRAP_ACTION_DISCARD_ERROR, +}; + +/* reg_ralue_trap_action + * Trap action. + * For IP2ME action, only NOP and MIRROR are possible. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, trap_action, 0x20, 28, 4); + +/* reg_ralue_trap_id + * Trap ID to be reported to CPU. + * Trap ID is RTR_INGRESS0 or RTR_INGRESS1. + * For trap_action of NOP, MIRROR and DISCARD_ERROR, trap_id is reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, trap_id, 0x20, 0, 9); + +/* reg_ralue_adjacency_index + * Points to the first entry of the group-based ECMP. + * Only relevant in case of REMOTE action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, adjacency_index, 0x24, 0, 24); + +/* reg_ralue_ecmp_size + * Amount of sequential entries starting + * from the adjacency_index (the number of ECMPs). + * The valid range is 1-64, 512, 1024, 2048 and 4096. + * Reserved when trap_action is TRAP or DISCARD_ERROR. + * Only relevant in case of REMOTE action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, ecmp_size, 0x28, 0, 13); + +/* reg_ralue_local_erif + * Egress Router Interface. + * Only relevant in case of LOCAL action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, local_erif, 0x24, 0, 16); + +/* reg_ralue_v + * Valid bit for the tunnel_ptr field. + * If valid = 0 then trap to CPU as IP2ME trap ID. + * If valid = 1 and the packet format allows NVE or IPinIP tunnel + * decapsulation then tunnel decapsulation is done. + * If valid = 1 and packet format does not allow NVE or IPinIP tunnel + * decapsulation then trap as IP2ME trap ID. + * Only relevant in case of IP2ME action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, v, 0x24, 31, 1); + +/* reg_ralue_tunnel_ptr + * Tunnel Pointer for NVE or IPinIP tunnel decapsulation. + * For Spectrum, pointer to KVD Linear. + * Only relevant in case of IP2ME action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, tunnel_ptr, 0x24, 0, 24); + +static inline void mlxsw_reg_ralue_pack(char *payload, + enum mlxsw_reg_ralxx_protocol protocol, + enum mlxsw_reg_ralue_op op, + u16 virtual_router, u8 prefix_len) +{ + MLXSW_REG_ZERO(ralue, payload); + mlxsw_reg_ralue_protocol_set(payload, protocol); + mlxsw_reg_ralue_virtual_router_set(payload, virtual_router); + mlxsw_reg_ralue_prefix_len_set(payload, prefix_len); + mlxsw_reg_ralue_entry_type_set(payload, + MLXSW_REG_RALUE_ENTRY_TYPE_ROUTE_ENTRY); + mlxsw_reg_ralue_bmp_len_set(payload, prefix_len); +} + +static inline void mlxsw_reg_ralue_pack4(char *payload, + enum mlxsw_reg_ralxx_protocol protocol, + enum mlxsw_reg_ralue_op op, + u16 virtual_router, u8 prefix_len, + u32 dip) +{ + mlxsw_reg_ralue_pack(payload, protocol, op, virtual_router, prefix_len); + mlxsw_reg_ralue_dip4_set(payload, dip); +} + +static inline void +mlxsw_reg_ralue_act_remote_pack(char *payload, + enum mlxsw_reg_ralue_trap_action trap_action, + u16 trap_id, u32 adjacency_index, u16 ecmp_size) +{ + mlxsw_reg_ralue_action_type_set(payload, + MLXSW_REG_RALUE_ACTION_TYPE_REMOTE); + mlxsw_reg_ralue_trap_action_set(payload, trap_action); + mlxsw_reg_ralue_trap_id_set(payload, trap_id); + mlxsw_reg_ralue_adjacency_index_set(payload, adjacency_index); + mlxsw_reg_ralue_ecmp_size_set(payload, ecmp_size); +} + +static inline void +mlxsw_reg_ralue_act_local_pack(char *payload, + enum mlxsw_reg_ralue_trap_action trap_action, + u16 trap_id, u16 local_erif) +{ + mlxsw_reg_ralue_action_type_set(payload, + MLXSW_REG_RALUE_ACTION_TYPE_LOCAL); + mlxsw_reg_ralue_trap_action_set(payload, trap_action); + mlxsw_reg_ralue_trap_id_set(payload, trap_id); + mlxsw_reg_ralue_local_erif_set(payload, local_erif); +} + +static inline void +mlxsw_reg_ralue_act_ip2me_pack(char *payload) +{ + mlxsw_reg_ralue_action_type_set(payload, + MLXSW_REG_RALUE_ACTION_TYPE_IP2ME); +} + +/* RAUHT - Router Algorithmic LPM Unicast Host Table Register + * ---------------------------------------------------------- + * The RAUHT register is used to configure and query the Unicast Host table in + * devices that implement the Algorithmic LPM. + */ +#define MLXSW_REG_RAUHT_ID 0x8014 +#define MLXSW_REG_RAUHT_LEN 0x74 + +static const struct mlxsw_reg_info mlxsw_reg_rauht = { + .id = MLXSW_REG_RAUHT_ID, + .len = MLXSW_REG_RAUHT_LEN, +}; + +enum mlxsw_reg_rauht_type { + MLXSW_REG_RAUHT_TYPE_IPV4, + MLXSW_REG_RAUHT_TYPE_IPV6, +}; + +/* reg_rauht_type + * Access: Index + */ +MLXSW_ITEM32(reg, rauht, type, 0x00, 24, 2); + +enum mlxsw_reg_rauht_op { + MLXSW_REG_RAUHT_OP_QUERY_READ = 0, + /* Read operation */ + MLXSW_REG_RAUHT_OP_QUERY_CLEAR_ON_READ = 1, + /* Clear on read operation. Used to read entry and clear + * activity bit. + */ + MLXSW_REG_RAUHT_OP_WRITE_ADD = 0, + /* Add. Used to write a new entry to the table. All R/W fields are + * relevant for new entry. Activity bit is set for new entries. + */ + MLXSW_REG_RAUHT_OP_WRITE_UPDATE = 1, + /* Update action. Used to update an existing route entry and + * only update the following fields: + * trap_action, trap_id, mac, counter_set_type, counter_index + */ + MLXSW_REG_RAUHT_OP_WRITE_CLEAR_ACTIVITY = 2, + /* Clear activity. A bit is cleared for the entry. */ + MLXSW_REG_RAUHT_OP_WRITE_DELETE = 3, + /* Delete entry */ + MLXSW_REG_RAUHT_OP_WRITE_DELETE_ALL = 4, + /* Delete all host entries on a RIF. In this command, dip + * field is reserved. + */ +}; + +/* reg_rauht_op + * Access: OP + */ +MLXSW_ITEM32(reg, rauht, op, 0x00, 20, 3); + +/* reg_rauht_a + * Activity. Set for new entries. Set if a packet lookup has hit on + * the specific entry. + * To clear the a bit, use "clear activity" op. + * Enabled by activity_dis in RGCR + * Access: RO + */ +MLXSW_ITEM32(reg, rauht, a, 0x00, 16, 1); + +/* reg_rauht_rif + * Router Interface + * Access: Index + */ +MLXSW_ITEM32(reg, rauht, rif, 0x00, 0, 16); + +/* reg_rauht_dip* + * Destination address. + * Access: Index + */ +MLXSW_ITEM32(reg, rauht, dip4, 0x1C, 0x0, 32); + +enum mlxsw_reg_rauht_trap_action { + MLXSW_REG_RAUHT_TRAP_ACTION_NOP, + MLXSW_REG_RAUHT_TRAP_ACTION_TRAP, + MLXSW_REG_RAUHT_TRAP_ACTION_MIRROR_TO_CPU, + MLXSW_REG_RAUHT_TRAP_ACTION_MIRROR, + MLXSW_REG_RAUHT_TRAP_ACTION_DISCARD_ERRORS, +}; + +/* reg_rauht_trap_action + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, trap_action, 0x60, 28, 4); + +enum mlxsw_reg_rauht_trap_id { + MLXSW_REG_RAUHT_TRAP_ID_RTR_EGRESS0, + MLXSW_REG_RAUHT_TRAP_ID_RTR_EGRESS1, +}; + +/* reg_rauht_trap_id + * Trap ID to be reported to CPU. + * Trap-ID is RTR_EGRESS0 or RTR_EGRESS1. + * For trap_action of NOP, MIRROR and DISCARD_ERROR, + * trap_id is reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, trap_id, 0x60, 0, 9); + +/* reg_rauht_counter_set_type + * Counter set type for flow counters + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, counter_set_type, 0x68, 24, 8); + +/* reg_rauht_counter_index + * Counter index for flow counters + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, counter_index, 0x68, 0, 24); + +/* reg_rauht_mac + * MAC address. + * Access: RW + */ +MLXSW_ITEM_BUF(reg, rauht, mac, 0x6E, 6); + +static inline void mlxsw_reg_rauht_pack(char *payload, + enum mlxsw_reg_rauht_op op, u16 rif, + const char *mac) +{ + MLXSW_REG_ZERO(rauht, payload); + mlxsw_reg_rauht_op_set(payload, op); + mlxsw_reg_rauht_rif_set(payload, rif); + mlxsw_reg_rauht_mac_memcpy_to(payload, mac); +} + +static inline void mlxsw_reg_rauht_pack4(char *payload, + enum mlxsw_reg_rauht_op op, u16 rif, + const char *mac, u32 dip) +{ + mlxsw_reg_rauht_pack(payload, op, rif, mac); + mlxsw_reg_rauht_dip4_set(payload, dip); +} + +/* RALEU - Router Algorithmic LPM ECMP Update Register + * --------------------------------------------------- + * The register enables updating the ECMP section in the action for multiple + * LPM Unicast entries in a single operation. The update is executed to + * all entries of a {virtual router, protocol} tuple using the same ECMP group. + */ +#define MLXSW_REG_RALEU_ID 0x8015 +#define MLXSW_REG_RALEU_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_raleu = { + .id = MLXSW_REG_RALEU_ID, + .len = MLXSW_REG_RALEU_LEN, +}; + +/* reg_raleu_protocol + * Protocol. + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, protocol, 0x00, 24, 4); + +/* reg_raleu_virtual_router + * Virtual Router ID + * Range is 0..cap_max_virtual_routers-1 + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, virtual_router, 0x00, 0, 16); + +/* reg_raleu_adjacency_index + * Adjacency Index used for matching on the existing entries. + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, adjacency_index, 0x10, 0, 24); + +/* reg_raleu_ecmp_size + * ECMP Size used for matching on the existing entries. + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, ecmp_size, 0x14, 0, 13); + +/* reg_raleu_new_adjacency_index + * New Adjacency Index. + * Access: WO + */ +MLXSW_ITEM32(reg, raleu, new_adjacency_index, 0x20, 0, 24); + +/* reg_raleu_new_ecmp_size + * New ECMP Size. + * Access: WO + */ +MLXSW_ITEM32(reg, raleu, new_ecmp_size, 0x24, 0, 13); + +static inline void mlxsw_reg_raleu_pack(char *payload, + enum mlxsw_reg_ralxx_protocol protocol, + u16 virtual_router, + u32 adjacency_index, u16 ecmp_size, + u32 new_adjacency_index, + u16 new_ecmp_size) +{ + MLXSW_REG_ZERO(raleu, payload); + mlxsw_reg_raleu_protocol_set(payload, protocol); + mlxsw_reg_raleu_virtual_router_set(payload, virtual_router); + mlxsw_reg_raleu_adjacency_index_set(payload, adjacency_index); + mlxsw_reg_raleu_ecmp_size_set(payload, ecmp_size); + mlxsw_reg_raleu_new_adjacency_index_set(payload, new_adjacency_index); + mlxsw_reg_raleu_new_ecmp_size_set(payload, new_ecmp_size); +} + +/* RAUHTD - Router Algorithmic LPM Unicast Host Table Dump Register + * ---------------------------------------------------------------- + * The RAUHTD register allows dumping entries from the Router Unicast Host + * Table. For a given session an entry is dumped no more than one time. The + * first RAUHTD access after reset is a new session. A session ends when the + * num_rec response is smaller than num_rec request or for IPv4 when the + * num_entries is smaller than 4. The clear activity affect the current session + * or the last session if a new session has not started. + */ +#define MLXSW_REG_RAUHTD_ID 0x8018 +#define MLXSW_REG_RAUHTD_BASE_LEN 0x20 +#define MLXSW_REG_RAUHTD_REC_LEN 0x20 +#define MLXSW_REG_RAUHTD_REC_MAX_NUM 32 +#define MLXSW_REG_RAUHTD_LEN (MLXSW_REG_RAUHTD_BASE_LEN + \ + MLXSW_REG_RAUHTD_REC_MAX_NUM * MLXSW_REG_RAUHTD_REC_LEN) +#define MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC 4 + +static const struct mlxsw_reg_info mlxsw_reg_rauhtd = { + .id = MLXSW_REG_RAUHTD_ID, + .len = MLXSW_REG_RAUHTD_LEN, +}; + +#define MLXSW_REG_RAUHTD_FILTER_A BIT(0) +#define MLXSW_REG_RAUHTD_FILTER_RIF BIT(3) + +/* reg_rauhtd_filter_fields + * if a bit is '0' then the relevant field is ignored and dump is done + * regardless of the field value + * Bit0 - filter by activity: entry_a + * Bit3 - filter by entry rip: entry_rif + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, filter_fields, 0x00, 0, 8); + +enum mlxsw_reg_rauhtd_op { + MLXSW_REG_RAUHTD_OP_DUMP, + MLXSW_REG_RAUHTD_OP_DUMP_AND_CLEAR, +}; + +/* reg_rauhtd_op + * Access: OP + */ +MLXSW_ITEM32(reg, rauhtd, op, 0x04, 24, 2); + +/* reg_rauhtd_num_rec + * At request: number of records requested + * At response: number of records dumped + * For IPv4, each record has 4 entries at request and up to 4 entries + * at response + * Range is 0..MLXSW_REG_RAUHTD_REC_MAX_NUM + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, num_rec, 0x04, 0, 8); + +/* reg_rauhtd_entry_a + * Dump only if activity has value of entry_a + * Reserved if filter_fields bit0 is '0' + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, entry_a, 0x08, 16, 1); + +enum mlxsw_reg_rauhtd_type { + MLXSW_REG_RAUHTD_TYPE_IPV4, + MLXSW_REG_RAUHTD_TYPE_IPV6, +}; + +/* reg_rauhtd_type + * Dump only if record type is: + * 0 - IPv4 + * 1 - IPv6 + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, type, 0x08, 0, 4); + +/* reg_rauhtd_entry_rif + * Dump only if RIF has value of entry_rif + * Reserved if filter_fields bit3 is '0' + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, entry_rif, 0x0C, 0, 16); + +static inline void mlxsw_reg_rauhtd_pack(char *payload, + enum mlxsw_reg_rauhtd_type type) +{ + MLXSW_REG_ZERO(rauhtd, payload); + mlxsw_reg_rauhtd_filter_fields_set(payload, MLXSW_REG_RAUHTD_FILTER_A); + mlxsw_reg_rauhtd_op_set(payload, MLXSW_REG_RAUHTD_OP_DUMP_AND_CLEAR); + mlxsw_reg_rauhtd_num_rec_set(payload, MLXSW_REG_RAUHTD_REC_MAX_NUM); + mlxsw_reg_rauhtd_entry_a_set(payload, 1); + mlxsw_reg_rauhtd_type_set(payload, type); +} + +/* reg_rauhtd_ipv4_rec_num_entries + * Number of valid entries in this record: + * 0 - 1 valid entry + * 1 - 2 valid entries + * 2 - 3 valid entries + * 3 - 4 valid entries + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_rec_num_entries, + MLXSW_REG_RAUHTD_BASE_LEN, 28, 2, + MLXSW_REG_RAUHTD_REC_LEN, 0x00, false); + +/* reg_rauhtd_rec_type + * Record type. + * 0 - IPv4 + * 1 - IPv6 + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, rec_type, MLXSW_REG_RAUHTD_BASE_LEN, 24, 2, + MLXSW_REG_RAUHTD_REC_LEN, 0x00, false); + +#define MLXSW_REG_RAUHTD_IPV4_ENT_LEN 0x8 + +/* reg_rauhtd_ipv4_ent_a + * Activity. Set for new entries. Set if a packet lookup has hit on the + * specific entry. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_ent_a, MLXSW_REG_RAUHTD_BASE_LEN, 16, 1, + MLXSW_REG_RAUHTD_IPV4_ENT_LEN, 0x00, false); + +/* reg_rauhtd_ipv4_ent_rif + * Router interface. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_ent_rif, MLXSW_REG_RAUHTD_BASE_LEN, 0, + 16, MLXSW_REG_RAUHTD_IPV4_ENT_LEN, 0x00, false); + +/* reg_rauhtd_ipv4_ent_dip + * Destination IPv4 address. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_ent_dip, MLXSW_REG_RAUHTD_BASE_LEN, 0, + 32, MLXSW_REG_RAUHTD_IPV4_ENT_LEN, 0x04, false); + +static inline void mlxsw_reg_rauhtd_ent_ipv4_unpack(char *payload, + int ent_index, u16 *p_rif, + u32 *p_dip) +{ + *p_rif = mlxsw_reg_rauhtd_ipv4_ent_rif_get(payload, ent_index); + *p_dip = mlxsw_reg_rauhtd_ipv4_ent_dip_get(payload, ent_index); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -3924,6 +5104,26 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "HTGT"; case MLXSW_REG_HPKT_ID: return "HPKT"; + case MLXSW_REG_RGCR_ID: + return "RGCR"; + case MLXSW_REG_RITR_ID: + return "RITR"; + case MLXSW_REG_RATR_ID: + return "RATR"; + case MLXSW_REG_RALTA_ID: + return "RALTA"; + case MLXSW_REG_RALST_ID: + return "RALST"; + case MLXSW_REG_RALTB_ID: + return "RALTB"; + case MLXSW_REG_RALUE_ID: + return "RALUE"; + case MLXSW_REG_RAUHT_ID: + return "RAUHT"; + case MLXSW_REG_RALEU_ID: + return "RALEU"; + case MLXSW_REG_RAUHTD_ID: + return "RAUHTD"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 4a7273771028..c812513e079d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -49,7 +49,9 @@ #include <linux/jiffies.h> #include <linux/bitops.h> #include <linux/list.h> +#include <linux/notifier.h> #include <linux/dcbnl.h> +#include <linux/inetdevice.h> #include <net/switchdev.h> #include <generated/utsrelease.h> @@ -209,23 +211,6 @@ static int mlxsw_sp_port_dev_addr_init(struct mlxsw_sp_port *mlxsw_sp_port) return mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr); } -static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, - u16 vid, enum mlxsw_reg_spms_state state) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char *spms_pl; - int err; - - spms_pl = kmalloc(MLXSW_REG_SPMS_LEN, GFP_KERNEL); - if (!spms_pl) - return -ENOMEM; - mlxsw_reg_spms_pack(spms_pl, mlxsw_sp_port->local_port); - mlxsw_reg_spms_vid_pack(spms_pl, vid, state); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spms), spms_pl); - kfree(spms_pl); - return err; -} - static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; @@ -247,15 +232,23 @@ static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl); } -static int mlxsw_sp_port_swid_set(struct mlxsw_sp_port *mlxsw_sp_port, u8 swid) +static int __mlxsw_sp_port_swid_set(struct mlxsw_sp *mlxsw_sp, u8 local_port, + u8 swid) { - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; char pspa_pl[MLXSW_REG_PSPA_LEN]; - mlxsw_reg_pspa_pack(pspa_pl, swid, mlxsw_sp_port->local_port); + mlxsw_reg_pspa_pack(pspa_pl, swid, local_port); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pspa), pspa_pl); } +static int mlxsw_sp_port_swid_set(struct mlxsw_sp_port *mlxsw_sp_port, u8 swid) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + + return __mlxsw_sp_port_swid_set(mlxsw_sp, mlxsw_sp_port->local_port, + swid); +} + static int mlxsw_sp_port_vp_mode_set(struct mlxsw_sp_port *mlxsw_sp_port, bool enable) { @@ -305,9 +298,9 @@ mlxsw_sp_port_system_port_mapping_set(struct mlxsw_sp_port *mlxsw_sp_port) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sspr), sspr_pl); } -static int __mlxsw_sp_port_module_info_get(struct mlxsw_sp *mlxsw_sp, - u8 local_port, u8 *p_module, - u8 *p_width, u8 *p_lane) +static int mlxsw_sp_port_module_info_get(struct mlxsw_sp *mlxsw_sp, + u8 local_port, u8 *p_module, + u8 *p_width, u8 *p_lane) { char pmlp_pl[MLXSW_REG_PMLP_LEN]; int err; @@ -322,16 +315,6 @@ static int __mlxsw_sp_port_module_info_get(struct mlxsw_sp *mlxsw_sp, return 0; } -static int mlxsw_sp_port_module_info_get(struct mlxsw_sp *mlxsw_sp, - u8 local_port, u8 *p_module, - u8 *p_width) -{ - u8 lane; - - return __mlxsw_sp_port_module_info_get(mlxsw_sp, local_port, p_module, - p_width, &lane); -} - static int mlxsw_sp_port_module_map(struct mlxsw_sp *mlxsw_sp, u8 local_port, u8 module, u8 width, u8 lane) { @@ -410,7 +393,11 @@ static netdev_tx_t mlxsw_sp_port_xmit(struct sk_buff *skb, } mlxsw_sp_txhdr_construct(skb, &tx_info); - len = skb->len; + /* TX header is consumed by HW on the way so we shouldn't count its + * bytes as being sent. + */ + len = skb->len - MLXSW_TXHDR_LEN; + /* Due to a race we might fail here because of a full queue. In that * unlikely case we simply drop the packet. */ @@ -634,94 +621,8 @@ static int mlxsw_sp_port_vlan_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port) return 0; } -static struct mlxsw_sp_vfid * -mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, u16 vid) -{ - struct mlxsw_sp_vfid *vfid; - - list_for_each_entry(vfid, &mlxsw_sp->port_vfids.list, list) { - if (vfid->vid == vid) - return vfid; - } - - return NULL; -} - -static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp) -{ - return find_first_zero_bit(mlxsw_sp->port_vfids.mapped, - MLXSW_SP_VFID_PORT_MAX); -} - -static int __mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vfid) -{ - u16 fid = mlxsw_sp_vfid_to_fid(vfid); - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, fid, 0); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); -} - -static void __mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, u16 vfid) -{ - u16 fid = mlxsw_sp_vfid_to_fid(vfid); - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, fid, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); -} - -static struct mlxsw_sp_vfid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, - u16 vid) -{ - struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_vfid *vfid; - u16 n_vfid; - int err; - - n_vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp); - if (n_vfid == MLXSW_SP_VFID_PORT_MAX) { - dev_err(dev, "No available vFIDs\n"); - return ERR_PTR(-ERANGE); - } - - err = __mlxsw_sp_vfid_create(mlxsw_sp, n_vfid); - if (err) { - dev_err(dev, "Failed to create vFID=%d\n", n_vfid); - return ERR_PTR(err); - } - - vfid = kzalloc(sizeof(*vfid), GFP_KERNEL); - if (!vfid) - goto err_allocate_vfid; - - vfid->vfid = n_vfid; - vfid->vid = vid; - - list_add(&vfid->list, &mlxsw_sp->port_vfids.list); - set_bit(n_vfid, mlxsw_sp->port_vfids.mapped); - - return vfid; - -err_allocate_vfid: - __mlxsw_sp_vfid_destroy(mlxsw_sp, n_vfid); - return ERR_PTR(-ENOMEM); -} - -static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_vfid *vfid) -{ - clear_bit(vfid->vfid, mlxsw_sp->port_vfids.mapped); - list_del(&vfid->list); - - __mlxsw_sp_vfid_destroy(mlxsw_sp, vfid->vfid); - - kfree(vfid); -} - static struct mlxsw_sp_port * -mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_vfid *vfid) +mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { struct mlxsw_sp_port *mlxsw_sp_vport; @@ -739,8 +640,7 @@ mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_vport->stp_state = BR_STATE_FORWARDING; mlxsw_sp_vport->lagged = mlxsw_sp_port->lagged; mlxsw_sp_vport->lag_id = mlxsw_sp_port->lag_id; - mlxsw_sp_vport->vport.vfid = vfid; - mlxsw_sp_vport->vport.vid = vfid->vid; + mlxsw_sp_vport->vport.vid = vid; list_add(&mlxsw_sp_vport->vport.list, &mlxsw_sp_port->vports_list); @@ -757,9 +657,8 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_vfid *vfid; + bool untagged = vid == 1; int err; /* VLAN 0 is added to HW filter when device goes up, but it is @@ -773,31 +672,10 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, return 0; } - vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid); - if (!vfid) { - vfid = mlxsw_sp_vfid_create(mlxsw_sp, vid); - if (IS_ERR(vfid)) { - netdev_err(dev, "Failed to create vFID for VID=%d\n", - vid); - return PTR_ERR(vfid); - } - } - - mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, vfid); + mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, vid); if (!mlxsw_sp_vport) { netdev_err(dev, "Failed to create vPort for VID=%d\n", vid); - err = -ENOMEM; - goto err_port_vport_create; - } - - if (!vfid->nr_vports) { - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, - true, false); - if (err) { - netdev_err(dev, "Failed to setup flooding for vFID=%d\n", - vfid->vfid); - goto err_vport_flood_set; - } + return -ENOMEM; } /* When adding the first VLAN interface on a bridged port we need to @@ -812,70 +690,37 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, } } - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - true, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); - if (err) { - netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n", - vid, vfid->vfid); - goto err_port_vid_to_fid_set; - } - err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); if (err) { netdev_err(dev, "Failed to disable learning for VID=%d\n", vid); goto err_port_vid_learning_set; } - err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, true, false); + err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, true, untagged); if (err) { netdev_err(dev, "Failed to set VLAN membership for VID=%d\n", vid); goto err_port_add_vid; } - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_FORWARDING); - if (err) { - netdev_err(dev, "Failed to set STP state for VID=%d\n", vid); - goto err_port_stp_state_set; - } - - vfid->nr_vports++; - return 0; -err_port_stp_state_set: - mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false); err_port_add_vid: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); err_port_vid_learning_set: - mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false, - mlxsw_sp_vfid_to_fid(vfid->vfid), vid); -err_port_vid_to_fid_set: if (list_is_singular(&mlxsw_sp_port->vports_list)) mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: - if (!vfid->nr_vports) - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, - false); -err_vport_flood_set: mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); -err_port_vport_create: - if (!vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp, vfid); return err; } -int mlxsw_sp_port_kill_vid(struct net_device *dev, - __be16 __always_unused proto, u16 vid) +static int mlxsw_sp_port_kill_vid(struct net_device *dev, + __be16 __always_unused proto, u16 vid) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_vfid *vfid; + struct mlxsw_sp_fid *f; int err; /* VLAN 0 is removed from HW filter when device goes down, but @@ -890,15 +735,6 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return 0; } - vfid = mlxsw_sp_vport->vport.vfid; - - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_DISCARDING); - if (err) { - netdev_err(dev, "Failed to set STP state for VID=%d\n", vid); - return err; - } - err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false); if (err) { netdev_err(dev, "Failed to set VLAN membership for VID=%d\n", @@ -912,16 +748,12 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return err; } - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - false, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); - if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID=%d} to vFID=%d mapping\n", - vid, vfid->vfid); - return err; - } + /* Drop FID reference. If this was the last reference the + * resources will be freed. + */ + f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + if (f && !WARN_ON(!f->leave)) + f->leave(mlxsw_sp_vport); /* When removing the last VLAN interface on a bridged port we need to * transition all active 802.1Q bridge VLANs to use VID to FID @@ -935,13 +767,8 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, } } - vfid->nr_vports--; mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); - /* Destroy the vFID if no vPorts are assigned to it anymore. */ - if (!vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp_port->mlxsw_sp, vfid); - return 0; } @@ -949,17 +776,11 @@ static int mlxsw_sp_port_get_phys_port_name(struct net_device *dev, char *name, size_t len) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); - u8 module, width, lane; + u8 module = mlxsw_sp_port->mapping.module; + u8 width = mlxsw_sp_port->mapping.width; + u8 lane = mlxsw_sp_port->mapping.lane; int err; - err = __mlxsw_sp_port_module_info_get(mlxsw_sp_port->mlxsw_sp, - mlxsw_sp_port->local_port, - &module, &width, &lane); - if (err) { - netdev_err(dev, "Failed to retrieve module information\n"); - return err; - } - if (!mlxsw_sp_port->split) err = snprintf(name, len, "p%d", module + 1); else @@ -982,6 +803,8 @@ static const struct net_device_ops mlxsw_sp_port_netdev_ops = { .ndo_get_stats64 = mlxsw_sp_port_get_stats64, .ndo_vlan_rx_add_vid = mlxsw_sp_port_add_vid, .ndo_vlan_rx_kill_vid = mlxsw_sp_port_kill_vid, + .ndo_neigh_construct = mlxsw_sp_router_neigh_construct, + .ndo_neigh_destroy = mlxsw_sp_router_neigh_destroy, .ndo_fdb_add = switchdev_port_fdb_add, .ndo_fdb_del = switchdev_port_fdb_del, .ndo_fdb_dump = switchdev_port_fdb_dump, @@ -1681,8 +1504,8 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port) return 0; } -static int __mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, - bool split, u8 module, u8 width) +static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, + bool split, u8 module, u8 width, u8 lane) { struct mlxsw_sp_port *mlxsw_sp_port; struct net_device *dev; @@ -1697,6 +1520,9 @@ static int __mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, mlxsw_sp_port->mlxsw_sp = mlxsw_sp; mlxsw_sp_port->local_port = local_port; mlxsw_sp_port->split = split; + mlxsw_sp_port->mapping.module = module; + mlxsw_sp_port->mapping.width = width; + mlxsw_sp_port->mapping.lane = lane; bytes = DIV_ROUND_UP(VLAN_N_VID, BITS_PER_BYTE); mlxsw_sp_port->active_vlans = kzalloc(bytes, GFP_KERNEL); if (!mlxsw_sp_port->active_vlans) { @@ -1839,45 +1665,6 @@ err_port_active_vlans_alloc: return err; } -static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, - bool split, u8 module, u8 width, u8 lane) -{ - int err; - - err = mlxsw_sp_port_module_map(mlxsw_sp, local_port, module, width, - lane); - if (err) - return err; - - err = __mlxsw_sp_port_create(mlxsw_sp, local_port, split, module, - width); - if (err) - goto err_port_create; - - return 0; - -err_port_create: - mlxsw_sp_port_module_unmap(mlxsw_sp, local_port); - return err; -} - -static void mlxsw_sp_port_vports_fini(struct mlxsw_sp_port *mlxsw_sp_port) -{ - struct net_device *dev = mlxsw_sp_port->dev; - struct mlxsw_sp_port *mlxsw_sp_vport, *tmp; - - list_for_each_entry_safe(mlxsw_sp_vport, tmp, - &mlxsw_sp_port->vports_list, vport.list) { - u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - - /* vPorts created for VLAN devices should already be gone - * by now, since we unregistered the port netdev. - */ - WARN_ON(is_vlan_dev(mlxsw_sp_vport->dev)); - mlxsw_sp_port_kill_vid(dev, 0, vid); - } -} - static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) { struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; @@ -1888,13 +1675,14 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) mlxsw_core_port_fini(&mlxsw_sp_port->core_port); unregister_netdev(mlxsw_sp_port->dev); /* This calls ndo_stop */ mlxsw_sp_port_dcb_fini(mlxsw_sp_port); - mlxsw_sp_port_vports_fini(mlxsw_sp_port); + mlxsw_sp_port_kill_vid(mlxsw_sp_port->dev, 0, 1); mlxsw_sp_port_switchdev_fini(mlxsw_sp_port); mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT); mlxsw_sp_port_module_unmap(mlxsw_sp, mlxsw_sp_port->local_port); free_percpu(mlxsw_sp_port->pcpu_stats); kfree(mlxsw_sp_port->untagged_vlans); kfree(mlxsw_sp_port->active_vlans); + WARN_ON_ONCE(!list_empty(&mlxsw_sp_port->vports_list)); free_netdev(mlxsw_sp_port->dev); } @@ -1909,8 +1697,8 @@ static void mlxsw_sp_ports_remove(struct mlxsw_sp *mlxsw_sp) static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp) { + u8 module, width, lane; size_t alloc_size; - u8 module, width; int i; int err; @@ -1921,13 +1709,14 @@ static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp) for (i = 1; i < MLXSW_PORT_MAX_PORTS; i++) { err = mlxsw_sp_port_module_info_get(mlxsw_sp, i, &module, - &width); + &width, &lane); if (err) goto err_port_module_info_get; if (!width) continue; mlxsw_sp->port_to_module[i] = module; - err = __mlxsw_sp_port_create(mlxsw_sp, i, false, module, width); + err = mlxsw_sp_port_create(mlxsw_sp, i, false, module, width, + lane); if (err) goto err_port_create; } @@ -1948,12 +1737,85 @@ static u8 mlxsw_sp_cluster_base_port_get(u8 local_port) return local_port - offset; } +static int mlxsw_sp_port_split_create(struct mlxsw_sp *mlxsw_sp, u8 base_port, + u8 module, unsigned int count) +{ + u8 width = MLXSW_PORT_MODULE_MAX_WIDTH / count; + int err, i; + + for (i = 0; i < count; i++) { + err = mlxsw_sp_port_module_map(mlxsw_sp, base_port + i, module, + width, i * width); + if (err) + goto err_port_module_map; + } + + for (i = 0; i < count; i++) { + err = __mlxsw_sp_port_swid_set(mlxsw_sp, base_port + i, 0); + if (err) + goto err_port_swid_set; + } + + for (i = 0; i < count; i++) { + err = mlxsw_sp_port_create(mlxsw_sp, base_port + i, true, + module, width, i * width); + if (err) + goto err_port_create; + } + + return 0; + +err_port_create: + for (i--; i >= 0; i--) + mlxsw_sp_port_remove(mlxsw_sp, base_port + i); + i = count; +err_port_swid_set: + for (i--; i >= 0; i--) + __mlxsw_sp_port_swid_set(mlxsw_sp, base_port + i, + MLXSW_PORT_SWID_DISABLED_PORT); + i = count; +err_port_module_map: + for (i--; i >= 0; i--) + mlxsw_sp_port_module_unmap(mlxsw_sp, base_port + i); + return err; +} + +static void mlxsw_sp_port_unsplit_create(struct mlxsw_sp *mlxsw_sp, + u8 base_port, unsigned int count) +{ + u8 local_port, module, width = MLXSW_PORT_MODULE_MAX_WIDTH; + int i; + + /* Split by four means we need to re-create two ports, otherwise + * only one. + */ + count = count / 2; + + for (i = 0; i < count; i++) { + local_port = base_port + i * 2; + module = mlxsw_sp->port_to_module[local_port]; + + mlxsw_sp_port_module_map(mlxsw_sp, local_port, module, width, + 0); + } + + for (i = 0; i < count; i++) + __mlxsw_sp_port_swid_set(mlxsw_sp, base_port + i * 2, 0); + + for (i = 0; i < count; i++) { + local_port = base_port + i * 2; + module = mlxsw_sp->port_to_module[local_port]; + + mlxsw_sp_port_create(mlxsw_sp, local_port, false, module, + width, 0); + } +} + static int mlxsw_sp_port_split(struct mlxsw_core *mlxsw_core, u8 local_port, unsigned int count) { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); struct mlxsw_sp_port *mlxsw_sp_port; - u8 width = MLXSW_PORT_MODULE_MAX_WIDTH / count; u8 module, cur_width, base_port; int i; int err; @@ -1965,18 +1827,14 @@ static int mlxsw_sp_port_split(struct mlxsw_core *mlxsw_core, u8 local_port, return -EINVAL; } + module = mlxsw_sp_port->mapping.module; + cur_width = mlxsw_sp_port->mapping.width; + if (count != 2 && count != 4) { netdev_err(mlxsw_sp_port->dev, "Port can only be split into 2 or 4 ports\n"); return -EINVAL; } - err = mlxsw_sp_port_module_info_get(mlxsw_sp, local_port, &module, - &cur_width); - if (err) { - netdev_err(mlxsw_sp_port->dev, "Failed to get port's width\n"); - return err; - } - if (cur_width != MLXSW_PORT_MODULE_MAX_WIDTH) { netdev_err(mlxsw_sp_port->dev, "Port cannot be split further\n"); return -EINVAL; @@ -2001,25 +1859,16 @@ static int mlxsw_sp_port_split(struct mlxsw_core *mlxsw_core, u8 local_port, for (i = 0; i < count; i++) mlxsw_sp_port_remove(mlxsw_sp, base_port + i); - for (i = 0; i < count; i++) { - err = mlxsw_sp_port_create(mlxsw_sp, base_port + i, true, - module, width, i * width); - if (err) { - dev_err(mlxsw_sp->bus_info->dev, "Failed to create split port\n"); - goto err_port_create; - } + err = mlxsw_sp_port_split_create(mlxsw_sp, base_port, module, count); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to create split ports\n"); + goto err_port_split_create; } return 0; -err_port_create: - for (i--; i >= 0; i--) - mlxsw_sp_port_remove(mlxsw_sp, base_port + i); - for (i = 0; i < count / 2; i++) { - module = mlxsw_sp->port_to_module[base_port + i * 2]; - mlxsw_sp_port_create(mlxsw_sp, base_port + i * 2, false, - module, MLXSW_PORT_MODULE_MAX_WIDTH, 0); - } +err_port_split_create: + mlxsw_sp_port_unsplit_create(mlxsw_sp, base_port, count); return err; } @@ -2027,10 +1876,9 @@ static int mlxsw_sp_port_unsplit(struct mlxsw_core *mlxsw_core, u8 local_port) { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); struct mlxsw_sp_port *mlxsw_sp_port; - u8 module, cur_width, base_port; + u8 cur_width, base_port; unsigned int count; int i; - int err; mlxsw_sp_port = mlxsw_sp->ports[local_port]; if (!mlxsw_sp_port) { @@ -2044,12 +1892,7 @@ static int mlxsw_sp_port_unsplit(struct mlxsw_core *mlxsw_core, u8 local_port) return -EINVAL; } - err = mlxsw_sp_port_module_info_get(mlxsw_sp, local_port, &module, - &cur_width); - if (err) { - netdev_err(mlxsw_sp_port->dev, "Failed to get port's width\n"); - return err; - } + cur_width = mlxsw_sp_port->mapping.width; count = cur_width == 1 ? 4 : 2; base_port = mlxsw_sp_cluster_base_port_get(local_port); @@ -2061,14 +1904,7 @@ static int mlxsw_sp_port_unsplit(struct mlxsw_core *mlxsw_core, u8 local_port) for (i = 0; i < count; i++) mlxsw_sp_port_remove(mlxsw_sp, base_port + i); - for (i = 0; i < count / 2; i++) { - module = mlxsw_sp->port_to_module[base_port + i * 2]; - err = mlxsw_sp_port_create(mlxsw_sp, base_port + i * 2, false, - module, MLXSW_PORT_MODULE_MAX_WIDTH, - 0); - if (err) - dev_err(mlxsw_sp->bus_info->dev, "Failed to reinstantiate port\n"); - } + mlxsw_sp_port_unsplit_create(mlxsw_sp, base_port, count); return 0; } @@ -2083,11 +1919,8 @@ static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg, local_port = mlxsw_reg_pude_local_port_get(pude_pl); mlxsw_sp_port = mlxsw_sp->ports[local_port]; - if (!mlxsw_sp_port) { - dev_warn(mlxsw_sp->bus_info->dev, "Port %d: Link event received for non-existent port\n", - local_port); + if (!mlxsw_sp_port) return; - } status = mlxsw_reg_pude_oper_status_get(pude_pl); if (status == MLXSW_PORT_OPER_STATUS_UP) { @@ -2242,6 +2075,31 @@ static const struct mlxsw_rx_listener mlxsw_sp_rx_listener[] = { .local_port = MLXSW_PORT_DONT_CARE, .trap_id = MLXSW_TRAP_ID_IGMP_V3_REPORT, }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_ARPBC, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_ARPUC, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IP2ME, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_RTR_INGRESS0, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_HOST_MISS_IPV4, + }, }; static int mlxsw_sp_traps_init(struct mlxsw_sp *mlxsw_sp) @@ -2282,7 +2140,7 @@ err_rx_trap_set: mlxsw_sp); err_rx_listener_register: for (i--; i >= 0; i--) { - mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_DISCARD, mlxsw_sp_rx_listener[i].trap_id); mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); @@ -2299,7 +2157,7 @@ static void mlxsw_sp_traps_fini(struct mlxsw_sp *mlxsw_sp) int i; for (i = 0; i < ARRAY_SIZE(mlxsw_sp_rx_listener); i++) { - mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_DISCARD, mlxsw_sp_rx_listener[i].trap_id); mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); @@ -2378,8 +2236,8 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->core = mlxsw_core; mlxsw_sp->bus_info = mlxsw_bus_info; - INIT_LIST_HEAD(&mlxsw_sp->port_vfids.list); - INIT_LIST_HEAD(&mlxsw_sp->br_vfids.list); + INIT_LIST_HEAD(&mlxsw_sp->fids); + INIT_LIST_HEAD(&mlxsw_sp->vfids.list); INIT_LIST_HEAD(&mlxsw_sp->br_mids.list); err = mlxsw_sp_base_mac_get(mlxsw_sp); @@ -2388,16 +2246,10 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, return err; } - err = mlxsw_sp_ports_create(mlxsw_sp); - if (err) { - dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); - return err; - } - err = mlxsw_sp_event_register(mlxsw_sp, MLXSW_TRAP_ID_PUDE); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Failed to register for PUDE events\n"); - goto err_event_register; + return err; } err = mlxsw_sp_traps_init(mlxsw_sp); @@ -2430,8 +2282,24 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, goto err_switchdev_init; } + err = mlxsw_sp_router_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize router\n"); + goto err_router_init; + } + + err = mlxsw_sp_ports_create(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); + goto err_ports_create; + } + return 0; +err_ports_create: + mlxsw_sp_router_fini(mlxsw_sp); +err_router_init: + mlxsw_sp_switchdev_fini(mlxsw_sp); err_switchdev_init: err_lag_init: mlxsw_sp_buffers_fini(mlxsw_sp); @@ -2440,20 +2308,24 @@ err_flood_init: mlxsw_sp_traps_fini(mlxsw_sp); err_rx_listener_register: mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); -err_event_register: - mlxsw_sp_ports_remove(mlxsw_sp); return err; } static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + int i; + mlxsw_sp_ports_remove(mlxsw_sp); + mlxsw_sp_router_fini(mlxsw_sp); mlxsw_sp_switchdev_fini(mlxsw_sp); mlxsw_sp_buffers_fini(mlxsw_sp); mlxsw_sp_traps_fini(mlxsw_sp); mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); - mlxsw_sp_ports_remove(mlxsw_sp); + WARN_ON(!list_empty(&mlxsw_sp->vfids.list)); + WARN_ON(!list_empty(&mlxsw_sp->fids)); + for (i = 0; i < MLXSW_SP_RIF_MAX; i++) + WARN_ON_ONCE(mlxsw_sp->rifs[i]); } static struct mlxsw_config_profile mlxsw_sp_config_profile = { @@ -2484,6 +2356,10 @@ static struct mlxsw_config_profile mlxsw_sp_config_profile = { .max_ib_mc = 0, .used_max_pkey = 1, .max_pkey = 0, + .used_kvd_sizes = 1, + .kvd_linear_size = MLXSW_SP_KVD_LINEAR_SIZE, + .kvd_hash_single_size = MLXSW_SP_KVD_HASH_SINGLE_SIZE, + .kvd_hash_double_size = MLXSW_SP_KVD_HASH_DOUBLE_SIZE, .swid_config = { { .used_type = 1, @@ -2515,16 +2391,590 @@ static struct mlxsw_driver mlxsw_sp_driver = { .profile = &mlxsw_sp_config_profile, }; -static int -mlxsw_sp_port_fdb_flush_by_port(const struct mlxsw_sp_port *mlxsw_sp_port) +static bool mlxsw_sp_port_dev_check(const struct net_device *dev) +{ + return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; +} + +static struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find(struct net_device *dev) +{ + struct net_device *lower_dev; + struct list_head *iter; + + if (mlxsw_sp_port_dev_check(dev)) + return netdev_priv(dev); + + netdev_for_each_all_lower_dev(dev, lower_dev, iter) { + if (mlxsw_sp_port_dev_check(lower_dev)) + return netdev_priv(lower_dev); + } + return NULL; +} + +static struct mlxsw_sp *mlxsw_sp_lower_get(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find(dev); + return mlxsw_sp_port ? mlxsw_sp_port->mlxsw_sp : NULL; +} + +static struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find_rcu(struct net_device *dev) +{ + struct net_device *lower_dev; + struct list_head *iter; + + if (mlxsw_sp_port_dev_check(dev)) + return netdev_priv(dev); + + netdev_for_each_all_lower_dev_rcu(dev, lower_dev, iter) { + if (mlxsw_sp_port_dev_check(lower_dev)) + return netdev_priv(lower_dev); + } + return NULL; +} + +struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + rcu_read_lock(); + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find_rcu(dev); + if (mlxsw_sp_port) + dev_hold(mlxsw_sp_port->dev); + rcu_read_unlock(); + return mlxsw_sp_port; +} + +void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port) +{ + dev_put(mlxsw_sp_port->dev); +} + +static bool mlxsw_sp_rif_should_config(struct mlxsw_sp_rif *r, + unsigned long event) +{ + switch (event) { + case NETDEV_UP: + if (!r) + return true; + r->ref_count++; + return false; + case NETDEV_DOWN: + if (r && --r->ref_count == 0) + return true; + /* It is possible we already removed the RIF ourselves + * if it was assigned to a netdev that is now a bridge + * or LAG slave. + */ + return false; + } + + return false; +} + +static int mlxsw_sp_avail_rif_get(struct mlxsw_sp *mlxsw_sp) +{ + int i; + + for (i = 0; i < MLXSW_SP_RIF_MAX; i++) + if (!mlxsw_sp->rifs[i]) + return i; + + return MLXSW_SP_RIF_MAX; +} + +static void mlxsw_sp_vport_rif_sp_attr_get(struct mlxsw_sp_port *mlxsw_sp_vport, + bool *p_lagged, u16 *p_system_port) +{ + u8 local_port = mlxsw_sp_vport->local_port; + + *p_lagged = mlxsw_sp_vport->lagged; + *p_system_port = *p_lagged ? mlxsw_sp_vport->lag_id : local_port; +} + +static int mlxsw_sp_vport_rif_sp_op(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *l3_dev, u16 rif, + bool create) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + bool lagged = mlxsw_sp_vport->lagged; + char ritr_pl[MLXSW_REG_RITR_LEN]; + u16 system_port; + + mlxsw_reg_ritr_pack(ritr_pl, create, MLXSW_REG_RITR_SP_IF, rif, + l3_dev->mtu, l3_dev->dev_addr); + + mlxsw_sp_vport_rif_sp_attr_get(mlxsw_sp_vport, &lagged, &system_port); + mlxsw_reg_ritr_sp_if_pack(ritr_pl, lagged, system_port, + mlxsw_sp_vport_vid_get(mlxsw_sp_vport)); + + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); +} + +static void mlxsw_sp_vport_rif_sp_leave(struct mlxsw_sp_port *mlxsw_sp_vport); + +static struct mlxsw_sp_fid * +mlxsw_sp_rfid_alloc(u16 fid, struct net_device *l3_dev) +{ + struct mlxsw_sp_fid *f; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + return NULL; + + f->leave = mlxsw_sp_vport_rif_sp_leave; + f->ref_count = 0; + f->dev = l3_dev; + f->fid = fid; + + return f; +} + +static struct mlxsw_sp_rif * +mlxsw_sp_rif_alloc(u16 rif, struct net_device *l3_dev, struct mlxsw_sp_fid *f) +{ + struct mlxsw_sp_rif *r; + + r = kzalloc(sizeof(*r), GFP_KERNEL); + if (!r) + return NULL; + + ether_addr_copy(r->addr, l3_dev->dev_addr); + r->mtu = l3_dev->mtu; + r->ref_count = 1; + r->dev = l3_dev; + r->rif = rif; + r->f = f; + + return r; +} + +static struct mlxsw_sp_rif * +mlxsw_sp_vport_rif_sp_create(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *l3_dev) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct mlxsw_sp_fid *f; + struct mlxsw_sp_rif *r; + u16 fid, rif; + int err; + + rif = mlxsw_sp_avail_rif_get(mlxsw_sp); + if (rif == MLXSW_SP_RIF_MAX) + return ERR_PTR(-ERANGE); + + err = mlxsw_sp_vport_rif_sp_op(mlxsw_sp_vport, l3_dev, rif, true); + if (err) + return ERR_PTR(err); + + fid = mlxsw_sp_rif_sp_to_fid(rif); + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, fid, true); + if (err) + goto err_rif_fdb_op; + + f = mlxsw_sp_rfid_alloc(fid, l3_dev); + if (!f) { + err = -ENOMEM; + goto err_rfid_alloc; + } + + r = mlxsw_sp_rif_alloc(rif, l3_dev, f); + if (!r) { + err = -ENOMEM; + goto err_rif_alloc; + } + + f->r = r; + mlxsw_sp->rifs[rif] = r; + + return r; + +err_rif_alloc: + kfree(f); +err_rfid_alloc: + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, fid, false); +err_rif_fdb_op: + mlxsw_sp_vport_rif_sp_op(mlxsw_sp_vport, l3_dev, rif, false); + return ERR_PTR(err); +} + +static void mlxsw_sp_vport_rif_sp_destroy(struct mlxsw_sp_port *mlxsw_sp_vport, + struct mlxsw_sp_rif *r) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct net_device *l3_dev = r->dev; + struct mlxsw_sp_fid *f = r->f; + u16 fid = f->fid; + u16 rif = r->rif; + + mlxsw_sp->rifs[rif] = NULL; + f->r = NULL; + + kfree(r); + + kfree(f); + + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, fid, false); + + mlxsw_sp_vport_rif_sp_op(mlxsw_sp_vport, l3_dev, rif, false); +} + +static int mlxsw_sp_vport_rif_sp_join(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *l3_dev) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct mlxsw_sp_rif *r; + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, l3_dev); + if (!r) { + r = mlxsw_sp_vport_rif_sp_create(mlxsw_sp_vport, l3_dev); + if (IS_ERR(r)) + return PTR_ERR(r); + } + + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, r->f); + r->f->ref_count++; + + netdev_dbg(mlxsw_sp_vport->dev, "Joined FID=%d\n", r->f->fid); + + return 0; +} + +static void mlxsw_sp_vport_rif_sp_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + netdev_dbg(mlxsw_sp_vport->dev, "Left FID=%d\n", f->fid); + + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); + if (--f->ref_count == 0) + mlxsw_sp_vport_rif_sp_destroy(mlxsw_sp_vport, f->r); +} + +static int mlxsw_sp_inetaddr_vport_event(struct net_device *l3_dev, + struct net_device *port_dev, + unsigned long event, u16 vid) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(port_dev); + struct mlxsw_sp_port *mlxsw_sp_vport; + + mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); + if (WARN_ON(!mlxsw_sp_vport)) + return -EINVAL; + + switch (event) { + case NETDEV_UP: + return mlxsw_sp_vport_rif_sp_join(mlxsw_sp_vport, l3_dev); + case NETDEV_DOWN: + mlxsw_sp_vport_rif_sp_leave(mlxsw_sp_vport); + break; + } + + return 0; +} + +static int mlxsw_sp_inetaddr_port_event(struct net_device *port_dev, + unsigned long event) +{ + if (netif_is_bridge_port(port_dev) || netif_is_lag_port(port_dev)) + return 0; + + return mlxsw_sp_inetaddr_vport_event(port_dev, port_dev, event, 1); +} + +static int __mlxsw_sp_inetaddr_lag_event(struct net_device *l3_dev, + struct net_device *lag_dev, + unsigned long event, u16 vid) +{ + struct net_device *port_dev; + struct list_head *iter; + int err; + + netdev_for_each_lower_dev(lag_dev, port_dev, iter) { + if (mlxsw_sp_port_dev_check(port_dev)) { + err = mlxsw_sp_inetaddr_vport_event(l3_dev, port_dev, + event, vid); + if (err) + return err; + } + } + + return 0; +} + +static int mlxsw_sp_inetaddr_lag_event(struct net_device *lag_dev, + unsigned long event) +{ + if (netif_is_bridge_port(lag_dev)) + return 0; + + return __mlxsw_sp_inetaddr_lag_event(lag_dev, lag_dev, event, 1); +} + +static struct mlxsw_sp_fid *mlxsw_sp_bridge_fid_get(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev) +{ + u16 fid; + + if (is_vlan_dev(l3_dev)) + fid = vlan_dev_vlan_id(l3_dev); + else if (mlxsw_sp->master_bridge.dev == l3_dev) + fid = 1; + else + return mlxsw_sp_vfid_find(mlxsw_sp, l3_dev); + + return mlxsw_sp_fid_find(mlxsw_sp, fid); +} + +static enum mlxsw_reg_ritr_if_type mlxsw_sp_rif_type_get(u16 fid) +{ + if (mlxsw_sp_fid_is_vfid(fid)) + return MLXSW_REG_RITR_FID_IF; + else + return MLXSW_REG_RITR_VLAN_IF; +} + +static int mlxsw_sp_rif_bridge_op(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev, + u16 fid, u16 rif, + bool create) +{ + enum mlxsw_reg_ritr_if_type rif_type; + char ritr_pl[MLXSW_REG_RITR_LEN]; + + rif_type = mlxsw_sp_rif_type_get(fid); + mlxsw_reg_ritr_pack(ritr_pl, create, rif_type, rif, l3_dev->mtu, + l3_dev->dev_addr); + mlxsw_reg_ritr_fid_set(ritr_pl, rif_type, fid); + + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); +} + +static int mlxsw_sp_rif_bridge_create(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev, + struct mlxsw_sp_fid *f) +{ + struct mlxsw_sp_rif *r; + u16 rif; + int err; + + rif = mlxsw_sp_avail_rif_get(mlxsw_sp); + if (rif == MLXSW_SP_RIF_MAX) + return -ERANGE; + + err = mlxsw_sp_rif_bridge_op(mlxsw_sp, l3_dev, f->fid, rif, true); + if (err) + return err; + + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, f->fid, true); + if (err) + goto err_rif_fdb_op; + + r = mlxsw_sp_rif_alloc(rif, l3_dev, f); + if (!r) { + err = -ENOMEM; + goto err_rif_alloc; + } + + f->r = r; + mlxsw_sp->rifs[rif] = r; + + netdev_dbg(l3_dev, "RIF=%d created\n", rif); + + return 0; + +err_rif_alloc: + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, f->fid, false); +err_rif_fdb_op: + mlxsw_sp_rif_bridge_op(mlxsw_sp, l3_dev, f->fid, rif, false); + return err; +} + +void mlxsw_sp_rif_bridge_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *r) +{ + struct net_device *l3_dev = r->dev; + struct mlxsw_sp_fid *f = r->f; + u16 rif = r->rif; + + mlxsw_sp->rifs[rif] = NULL; + f->r = NULL; + + kfree(r); + + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, f->fid, false); + + mlxsw_sp_rif_bridge_op(mlxsw_sp, l3_dev, f->fid, rif, false); + + netdev_dbg(l3_dev, "RIF=%d destroyed\n", rif); +} + +static int mlxsw_sp_inetaddr_bridge_event(struct net_device *l3_dev, + struct net_device *br_dev, + unsigned long event) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(l3_dev); + struct mlxsw_sp_fid *f; + + /* FID can either be an actual FID if the L3 device is the + * VLAN-aware bridge or a VLAN device on top. Otherwise, the + * L3 device is a VLAN-unaware bridge and we get a vFID. + */ + f = mlxsw_sp_bridge_fid_get(mlxsw_sp, l3_dev); + if (WARN_ON(!f)) + return -EINVAL; + + switch (event) { + case NETDEV_UP: + return mlxsw_sp_rif_bridge_create(mlxsw_sp, l3_dev, f); + case NETDEV_DOWN: + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); + break; + } + + return 0; +} + +static int mlxsw_sp_inetaddr_vlan_event(struct net_device *vlan_dev, + unsigned long event) +{ + struct net_device *real_dev = vlan_dev_real_dev(vlan_dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(vlan_dev); + u16 vid = vlan_dev_vlan_id(vlan_dev); + + if (mlxsw_sp_port_dev_check(real_dev)) + return mlxsw_sp_inetaddr_vport_event(vlan_dev, real_dev, event, + vid); + else if (netif_is_lag_master(real_dev)) + return __mlxsw_sp_inetaddr_lag_event(vlan_dev, real_dev, event, + vid); + else if (netif_is_bridge_master(real_dev) && + mlxsw_sp->master_bridge.dev == real_dev) + return mlxsw_sp_inetaddr_bridge_event(vlan_dev, real_dev, + event); + + return 0; +} + +static int mlxsw_sp_inetaddr_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct in_ifaddr *ifa = (struct in_ifaddr *) ptr; + struct net_device *dev = ifa->ifa_dev->dev; + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_rif *r; + int err = 0; + + mlxsw_sp = mlxsw_sp_lower_get(dev); + if (!mlxsw_sp) + goto out; + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (!mlxsw_sp_rif_should_config(r, event)) + goto out; + + if (mlxsw_sp_port_dev_check(dev)) + err = mlxsw_sp_inetaddr_port_event(dev, event); + else if (netif_is_lag_master(dev)) + err = mlxsw_sp_inetaddr_lag_event(dev, event); + else if (netif_is_bridge_master(dev)) + err = mlxsw_sp_inetaddr_bridge_event(dev, dev, event); + else if (is_vlan_dev(dev)) + err = mlxsw_sp_inetaddr_vlan_event(dev, event); + +out: + return notifier_from_errno(err); +} + +static int mlxsw_sp_rif_edit(struct mlxsw_sp *mlxsw_sp, u16 rif, + const char *mac, int mtu) +{ + char ritr_pl[MLXSW_REG_RITR_LEN]; + int err; + + mlxsw_reg_ritr_rif_pack(ritr_pl, rif); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); + if (err) + return err; + + mlxsw_reg_ritr_mtu_set(ritr_pl, mtu); + mlxsw_reg_ritr_if_mac_memcpy_to(ritr_pl, mac); + mlxsw_reg_ritr_op_set(ritr_pl, MLXSW_REG_RITR_RIF_CREATE); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); +} + +static int mlxsw_sp_netdevice_router_port_event(struct net_device *dev) +{ + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_rif *r; + int err; + + mlxsw_sp = mlxsw_sp_lower_get(dev); + if (!mlxsw_sp) + return 0; + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (!r) + return 0; + + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, r->addr, r->f->fid, false); + if (err) + return err; + + err = mlxsw_sp_rif_edit(mlxsw_sp, r->rif, dev->dev_addr, dev->mtu); + if (err) + goto err_rif_edit; + + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, dev->dev_addr, r->f->fid, true); + if (err) + goto err_rif_fdb_op; + + ether_addr_copy(r->addr, dev->dev_addr); + r->mtu = dev->mtu; + + netdev_dbg(dev, "Updated RIF=%d\n", r->rif); + + return 0; + +err_rif_fdb_op: + mlxsw_sp_rif_edit(mlxsw_sp, r->rif, r->addr, r->mtu); +err_rif_edit: + mlxsw_sp_rif_fdb_op(mlxsw_sp, r->addr, r->f->fid, true); + return err; +} + +static bool mlxsw_sp_lag_port_fid_member(struct mlxsw_sp_port *lag_port, + u16 fid) +{ + if (mlxsw_sp_fid_is_vfid(fid)) + return mlxsw_sp_port_vport_find_by_fid(lag_port, fid); + else + return test_bit(fid, lag_port->active_vlans); +} + +static bool mlxsw_sp_port_fdb_should_flush(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char sfdf_pl[MLXSW_REG_SFDF_LEN]; + u8 local_port = mlxsw_sp_port->local_port; + u16 lag_id = mlxsw_sp_port->lag_id; + int i, count = 0; - mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_PORT); - mlxsw_reg_sfdf_system_port_set(sfdf_pl, mlxsw_sp_port->local_port); + if (!mlxsw_sp_port->lagged) + return true; - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); + for (i = 0; i < MLXSW_SP_PORT_PER_LAG_MAX; i++) { + struct mlxsw_sp_port *lag_port; + + lag_port = mlxsw_sp_port_lagged_get(mlxsw_sp, lag_id, i); + if (!lag_port || lag_port->local_port == local_port) + continue; + if (mlxsw_sp_lag_port_fid_member(lag_port, fid)) + count++; + } + + return !count; } static int @@ -2539,17 +2989,8 @@ mlxsw_sp_port_fdb_flush_by_port_fid(const struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_reg_sfdf_port_fid_system_port_set(sfdf_pl, mlxsw_sp_port->local_port); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); -} - -static int -mlxsw_sp_port_fdb_flush_by_lag_id(const struct mlxsw_sp_port *mlxsw_sp_port) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char sfdf_pl[MLXSW_REG_SFDF_LEN]; - - mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_LAG); - mlxsw_reg_sfdf_lag_id_set(sfdf_pl, mlxsw_sp_port->lag_id); + netdev_dbg(mlxsw_sp_port->dev, "FDB flushed using Port=%d, FID=%d\n", + mlxsw_sp_port->local_port, fid); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); } @@ -2565,71 +3006,64 @@ mlxsw_sp_port_fdb_flush_by_lag_id_fid(const struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_reg_sfdf_fid_set(sfdf_pl, fid); mlxsw_reg_sfdf_lag_fid_lag_id_set(sfdf_pl, mlxsw_sp_port->lag_id); + netdev_dbg(mlxsw_sp_port->dev, "FDB flushed using LAG ID=%d, FID=%d\n", + mlxsw_sp_port->lag_id, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); } -static int -__mlxsw_sp_port_fdb_flush(const struct mlxsw_sp_port *mlxsw_sp_port) +int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) { - int err, last_err = 0; - u16 vid; - - for (vid = 1; vid < VLAN_N_VID - 1; vid++) { - err = mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, vid); - if (err) - last_err = err; - } + if (!mlxsw_sp_port_fdb_should_flush(mlxsw_sp_port, fid)) + return 0; - return last_err; + if (mlxsw_sp_port->lagged) + return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_port, + fid); + else + return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, fid); } -static int -__mlxsw_sp_port_fdb_flush_lagged(const struct mlxsw_sp_port *mlxsw_sp_port) +static void mlxsw_sp_master_bridge_gone_sync(struct mlxsw_sp *mlxsw_sp) { - int err, last_err = 0; - u16 vid; + struct mlxsw_sp_fid *f, *tmp; - for (vid = 1; vid < VLAN_N_VID - 1; vid++) { - err = mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_port, vid); - if (err) - last_err = err; - } - - return last_err; + list_for_each_entry_safe(f, tmp, &mlxsw_sp->fids, list) + if (--f->ref_count == 0) + mlxsw_sp_fid_destroy(mlxsw_sp, f); + else + WARN_ON_ONCE(1); } -static int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port) +static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) { - if (!list_empty(&mlxsw_sp_port->vports_list)) - if (mlxsw_sp_port->lagged) - return __mlxsw_sp_port_fdb_flush_lagged(mlxsw_sp_port); - else - return __mlxsw_sp_port_fdb_flush(mlxsw_sp_port); - else - if (mlxsw_sp_port->lagged) - return mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port); - else - return mlxsw_sp_port_fdb_flush_by_port(mlxsw_sp_port); + return !mlxsw_sp->master_bridge.dev || + mlxsw_sp->master_bridge.dev == br_dev; } -static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport) +static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) { - u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_vport); - u16 fid = mlxsw_sp_vfid_to_fid(vfid); - - if (mlxsw_sp_vport->lagged) - return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_vport, - fid); - else - return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_vport, fid); + mlxsw_sp->master_bridge.dev = br_dev; + mlxsw_sp->master_bridge.ref_count++; } -static bool mlxsw_sp_port_dev_check(const struct net_device *dev) +static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp) { - return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; + if (--mlxsw_sp->master_bridge.ref_count == 0) { + mlxsw_sp->master_bridge.dev = NULL; + /* It's possible upper VLAN devices are still holding + * references to underlying FIDs. Drop the reference + * and release the resources if it was the last one. + * If it wasn't, then something bad happened. + */ + mlxsw_sp_master_bridge_gone_sync(mlxsw_sp); + } } -static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) +static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *br_dev) { struct net_device *dev = mlxsw_sp_port->dev; int err; @@ -2643,6 +3077,8 @@ static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) if (err) return err; + mlxsw_sp_master_bridge_inc(mlxsw_sp_port->mlxsw_sp, br_dev); + mlxsw_sp_port->learning = 1; mlxsw_sp_port->learning_sync = 1; mlxsw_sp_port->uc_flood = 1; @@ -2651,16 +3087,14 @@ static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) return 0; } -static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, - bool flush_fdb) +static void mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port) { struct net_device *dev = mlxsw_sp_port->dev; - if (flush_fdb && mlxsw_sp_port_fdb_flush(mlxsw_sp_port)) - netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n"); - mlxsw_sp_port_pvid_set(mlxsw_sp_port, 1); + mlxsw_sp_master_bridge_dec(mlxsw_sp_port->mlxsw_sp); + mlxsw_sp_port->learning = 0; mlxsw_sp_port->learning_sync = 0; mlxsw_sp_port->uc_flood = 0; @@ -2669,28 +3103,7 @@ static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, /* Add implicit VLAN interface in the device, so that untagged * packets will be classified to the default vFID. */ - return mlxsw_sp_port_add_vid(dev, 0, 1); -} - -static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) -{ - return !mlxsw_sp->master_bridge.dev || - mlxsw_sp->master_bridge.dev == br_dev; -} - -static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) -{ - mlxsw_sp->master_bridge.dev = br_dev; - mlxsw_sp->master_bridge.ref_count++; -} - -static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) -{ - if (--mlxsw_sp->master_bridge.ref_count == 0) - mlxsw_sp->master_bridge.dev = NULL; + mlxsw_sp_port_add_vid(dev, 0, 1); } static int mlxsw_sp_lag_create(struct mlxsw_sp *mlxsw_sp, u16 lag_id) @@ -2806,6 +3219,45 @@ static int mlxsw_sp_port_lag_index_get(struct mlxsw_sp *mlxsw_sp, return -EBUSY; } +static void +mlxsw_sp_port_pvid_vport_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, + u16 lag_id) +{ + struct mlxsw_sp_port *mlxsw_sp_vport; + struct mlxsw_sp_fid *f; + + mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, 1); + if (WARN_ON(!mlxsw_sp_vport)) + return; + + /* If vPort is assigned a RIF, then leave it since it's no + * longer valid. + */ + f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + if (f) + f->leave(mlxsw_sp_vport); + + mlxsw_sp_vport->lag_id = lag_id; + mlxsw_sp_vport->lagged = 1; +} + +static void +mlxsw_sp_port_pvid_vport_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_port *mlxsw_sp_vport; + struct mlxsw_sp_fid *f; + + mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, 1); + if (WARN_ON(!mlxsw_sp_vport)) + return; + + f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + if (f) + f->leave(mlxsw_sp_vport); + + mlxsw_sp_vport->lagged = 0; +} + static int mlxsw_sp_port_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *lag_dev) { @@ -2841,6 +3293,9 @@ static int mlxsw_sp_port_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_port->lag_id = lag_id; mlxsw_sp_port->lagged = 1; lag->ref_count++; + + mlxsw_sp_port_pvid_vport_lag_join(mlxsw_sp_port, lag_id); + return 0; err_col_port_enable: @@ -2851,65 +3306,35 @@ err_col_port_add: return err; } -static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, - bool flush_fdb); - -static int mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, - struct net_device *lag_dev) +static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *lag_dev) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_upper *lag; u16 lag_id = mlxsw_sp_port->lag_id; - int err; + struct mlxsw_sp_upper *lag; if (!mlxsw_sp_port->lagged) - return 0; + return; lag = mlxsw_sp_lag_get(mlxsw_sp, lag_id); WARN_ON(lag->ref_count == 0); - err = mlxsw_sp_lag_col_port_disable(mlxsw_sp_port, lag_id); - if (err) - return err; - err = mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id); - if (err) - return err; - - /* In case we leave a LAG device that has bridges built on top, - * then their teardown sequence is never issued and we need to - * invoke the necessary cleanup routines ourselves. - */ - list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, - vport.list) { - struct net_device *br_dev; - - if (!mlxsw_sp_vport->bridged) - continue; - - br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, br_dev, false); - } + mlxsw_sp_lag_col_port_disable(mlxsw_sp_port, lag_id); + mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id); if (mlxsw_sp_port->bridged) { mlxsw_sp_port_active_vlans_del(mlxsw_sp_port); - mlxsw_sp_port_bridge_leave(mlxsw_sp_port, false); - mlxsw_sp_master_bridge_dec(mlxsw_sp, NULL); + mlxsw_sp_port_bridge_leave(mlxsw_sp_port); } - if (lag->ref_count == 1) { - if (mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port)) - netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n"); - err = mlxsw_sp_lag_destroy(mlxsw_sp, lag_id); - if (err) - return err; - } + if (lag->ref_count == 1) + mlxsw_sp_lag_destroy(mlxsw_sp, lag_id); mlxsw_core_lag_mapping_clear(mlxsw_sp->core, lag_id, mlxsw_sp_port->local_port); mlxsw_sp_port->lagged = 0; lag->ref_count--; - return 0; + + mlxsw_sp_port_pvid_vport_lag_leave(mlxsw_sp_port); } static int mlxsw_sp_lag_dist_port_add(struct mlxsw_sp_port *mlxsw_sp_port, @@ -2958,42 +3383,25 @@ static int mlxsw_sp_port_vlan_link(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid = vlan_dev_vlan_id(vlan_dev); mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); - if (!mlxsw_sp_vport) { - WARN_ON(!mlxsw_sp_vport); + if (WARN_ON(!mlxsw_sp_vport)) return -EINVAL; - } mlxsw_sp_vport->dev = vlan_dev; return 0; } -static int mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, - struct net_device *vlan_dev) +static void mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *vlan_dev) { struct mlxsw_sp_port *mlxsw_sp_vport; u16 vid = vlan_dev_vlan_id(vlan_dev); mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); - if (!mlxsw_sp_vport) { - WARN_ON(!mlxsw_sp_vport); - return -EINVAL; - } - - /* When removing a VLAN device while still bridged we should first - * remove it from the bridge, as we receive the bridge's notification - * when the vPort is already gone. - */ - if (mlxsw_sp_vport->bridged) { - struct net_device *br_dev; - - br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, br_dev, true); - } + if (WARN_ON(!mlxsw_sp_vport)) + return; mlxsw_sp_vport->dev = mlxsw_sp_port->dev; - - return 0; } static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, @@ -3003,7 +3411,7 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, struct mlxsw_sp_port *mlxsw_sp_port; struct net_device *upper_dev; struct mlxsw_sp *mlxsw_sp; - int err; + int err = 0; mlxsw_sp_port = netdev_priv(dev); mlxsw_sp = mlxsw_sp_port->mlxsw_sp; @@ -3012,73 +3420,56 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, switch (event) { case NETDEV_PRECHANGEUPPER: upper_dev = info->upper_dev; - if (!info->master || !info->linking) + if (!is_vlan_dev(upper_dev) && + !netif_is_lag_master(upper_dev) && + !netif_is_bridge_master(upper_dev)) + return -EINVAL; + if (!info->linking) break; /* HW limitation forbids to put ports to multiple bridges. */ if (netif_is_bridge_master(upper_dev) && !mlxsw_sp_master_bridge_check(mlxsw_sp, upper_dev)) - return NOTIFY_BAD; + return -EINVAL; if (netif_is_lag_master(upper_dev) && !mlxsw_sp_master_lag_check(mlxsw_sp, upper_dev, info->upper_info)) - return NOTIFY_BAD; + return -EINVAL; + if (netif_is_lag_master(upper_dev) && vlan_uses_dev(dev)) + return -EINVAL; + if (netif_is_lag_port(dev) && is_vlan_dev(upper_dev) && + !netif_is_lag_master(vlan_dev_real_dev(upper_dev))) + return -EINVAL; break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; if (is_vlan_dev(upper_dev)) { - if (info->linking) { + if (info->linking) err = mlxsw_sp_port_vlan_link(mlxsw_sp_port, upper_dev); - if (err) { - netdev_err(dev, "Failed to link VLAN device\n"); - return NOTIFY_BAD; - } - } else { - err = mlxsw_sp_port_vlan_unlink(mlxsw_sp_port, - upper_dev); - if (err) { - netdev_err(dev, "Failed to unlink VLAN device\n"); - return NOTIFY_BAD; - } - } + else + mlxsw_sp_port_vlan_unlink(mlxsw_sp_port, + upper_dev); } else if (netif_is_bridge_master(upper_dev)) { - if (info->linking) { - err = mlxsw_sp_port_bridge_join(mlxsw_sp_port); - if (err) { - netdev_err(dev, "Failed to join bridge\n"); - return NOTIFY_BAD; - } - mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev); - } else { - err = mlxsw_sp_port_bridge_leave(mlxsw_sp_port, - true); - mlxsw_sp_master_bridge_dec(mlxsw_sp, upper_dev); - if (err) { - netdev_err(dev, "Failed to leave bridge\n"); - return NOTIFY_BAD; - } - } + if (info->linking) + err = mlxsw_sp_port_bridge_join(mlxsw_sp_port, + upper_dev); + else + mlxsw_sp_port_bridge_leave(mlxsw_sp_port); } else if (netif_is_lag_master(upper_dev)) { - if (info->linking) { + if (info->linking) err = mlxsw_sp_port_lag_join(mlxsw_sp_port, upper_dev); - if (err) { - netdev_err(dev, "Failed to join link aggregation\n"); - return NOTIFY_BAD; - } - } else { - err = mlxsw_sp_port_lag_leave(mlxsw_sp_port, - upper_dev); - if (err) { - netdev_err(dev, "Failed to leave link aggregation\n"); - return NOTIFY_BAD; - } - } + else + mlxsw_sp_port_lag_leave(mlxsw_sp_port, + upper_dev); + } else { + err = -EINVAL; + WARN_ON(1); } break; } - return NOTIFY_DONE; + return err; } static int mlxsw_sp_netdevice_port_lower_event(struct net_device *dev, @@ -3102,7 +3493,7 @@ static int mlxsw_sp_netdevice_port_lower_event(struct net_device *dev, break; } - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_port_event(struct net_device *dev, @@ -3116,7 +3507,7 @@ static int mlxsw_sp_netdevice_port_event(struct net_device *dev, return mlxsw_sp_netdevice_port_lower_event(dev, event, ptr); } - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, @@ -3129,218 +3520,230 @@ static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, netdev_for_each_lower_dev(lag_dev, dev, iter) { if (mlxsw_sp_port_dev_check(dev)) { ret = mlxsw_sp_netdevice_port_event(dev, event, ptr); - if (ret == NOTIFY_BAD) + if (ret) return ret; } } - return NOTIFY_DONE; + return 0; } -static struct mlxsw_sp_vfid * -mlxsw_sp_br_vfid_find(const struct mlxsw_sp *mlxsw_sp, - const struct net_device *br_dev) +static int mlxsw_sp_master_bridge_vlan_link(struct mlxsw_sp *mlxsw_sp, + struct net_device *vlan_dev) { - struct mlxsw_sp_vfid *vfid; + u16 fid = vlan_dev_vlan_id(vlan_dev); + struct mlxsw_sp_fid *f; - list_for_each_entry(vfid, &mlxsw_sp->br_vfids.list, list) { - if (vfid->br_dev == br_dev) - return vfid; + f = mlxsw_sp_fid_find(mlxsw_sp, fid); + if (!f) { + f = mlxsw_sp_fid_create(mlxsw_sp, fid); + if (IS_ERR(f)) + return PTR_ERR(f); } - return NULL; + f->ref_count++; + + return 0; +} + +static void mlxsw_sp_master_bridge_vlan_unlink(struct mlxsw_sp *mlxsw_sp, + struct net_device *vlan_dev) +{ + u16 fid = vlan_dev_vlan_id(vlan_dev); + struct mlxsw_sp_fid *f; + + f = mlxsw_sp_fid_find(mlxsw_sp, fid); + if (f && f->r) + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); + if (f && --f->ref_count == 0) + mlxsw_sp_fid_destroy(mlxsw_sp, f); } -static u16 mlxsw_sp_vfid_to_br_vfid(u16 vfid) +static int mlxsw_sp_netdevice_bridge_event(struct net_device *br_dev, + unsigned long event, void *ptr) { - return vfid - MLXSW_SP_VFID_PORT_MAX; + struct netdev_notifier_changeupper_info *info; + struct net_device *upper_dev; + struct mlxsw_sp *mlxsw_sp; + int err; + + mlxsw_sp = mlxsw_sp_lower_get(br_dev); + if (!mlxsw_sp) + return 0; + if (br_dev != mlxsw_sp->master_bridge.dev) + return 0; + + info = ptr; + + switch (event) { + case NETDEV_CHANGEUPPER: + upper_dev = info->upper_dev; + if (!is_vlan_dev(upper_dev)) + break; + if (info->linking) { + err = mlxsw_sp_master_bridge_vlan_link(mlxsw_sp, + upper_dev); + if (err) + return err; + } else { + mlxsw_sp_master_bridge_vlan_unlink(mlxsw_sp, upper_dev); + } + break; + } + + return 0; } -static u16 mlxsw_sp_br_vfid_to_vfid(u16 br_vfid) +static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp) { - return MLXSW_SP_VFID_PORT_MAX + br_vfid; + return find_first_zero_bit(mlxsw_sp->vfids.mapped, + MLXSW_SP_VFID_MAX); } -static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp) +static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) { - return find_first_zero_bit(mlxsw_sp->br_vfids.mapped, - MLXSW_SP_VFID_BR_MAX); + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + + mlxsw_reg_sfmr_pack(sfmr_pl, !create, fid, 0); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } -static struct mlxsw_sp_vfid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) +static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); + +static struct mlxsw_sp_fid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) { struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_vfid *vfid; - u16 n_vfid; + struct mlxsw_sp_fid *f; + u16 vfid, fid; int err; - n_vfid = mlxsw_sp_br_vfid_to_vfid(mlxsw_sp_avail_br_vfid_get(mlxsw_sp)); - if (n_vfid == MLXSW_SP_VFID_MAX) { + vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp); + if (vfid == MLXSW_SP_VFID_MAX) { dev_err(dev, "No available vFIDs\n"); return ERR_PTR(-ERANGE); } - err = __mlxsw_sp_vfid_create(mlxsw_sp, n_vfid); + fid = mlxsw_sp_vfid_to_fid(vfid); + err = mlxsw_sp_vfid_op(mlxsw_sp, fid, true); if (err) { - dev_err(dev, "Failed to create vFID=%d\n", n_vfid); + dev_err(dev, "Failed to create FID=%d\n", fid); return ERR_PTR(err); } - vfid = kzalloc(sizeof(*vfid), GFP_KERNEL); - if (!vfid) + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) goto err_allocate_vfid; - vfid->vfid = n_vfid; - vfid->br_dev = br_dev; + f->leave = mlxsw_sp_vport_vfid_leave; + f->fid = fid; + f->dev = br_dev; - list_add(&vfid->list, &mlxsw_sp->br_vfids.list); - set_bit(mlxsw_sp_vfid_to_br_vfid(n_vfid), mlxsw_sp->br_vfids.mapped); + list_add(&f->list, &mlxsw_sp->vfids.list); + set_bit(vfid, mlxsw_sp->vfids.mapped); - return vfid; + return f; err_allocate_vfid: - __mlxsw_sp_vfid_destroy(mlxsw_sp, n_vfid); + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); return ERR_PTR(-ENOMEM); } -static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_vfid *vfid) +static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fid *f) { - u16 br_vfid = mlxsw_sp_vfid_to_br_vfid(vfid->vfid); + u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); + u16 fid = f->fid; + + clear_bit(vfid, mlxsw_sp->vfids.mapped); + list_del(&f->list); - clear_bit(br_vfid, mlxsw_sp->br_vfids.mapped); - list_del(&vfid->list); + if (f->r) + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); - __mlxsw_sp_vfid_destroy(mlxsw_sp, vfid->vfid); + kfree(f); - kfree(vfid); + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); } -static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, - bool flush_fdb) +static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, + bool valid) { - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - struct net_device *dev = mlxsw_sp_vport->dev; - struct mlxsw_sp_vfid *vfid, *new_vfid; - int err; - - vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (!vfid) { - WARN_ON(!vfid); - return -EINVAL; - } - - /* We need a vFID to go back to after leaving the bridge's vFID. */ - new_vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid); - if (!new_vfid) { - new_vfid = mlxsw_sp_vfid_create(mlxsw_sp, vid); - if (IS_ERR(new_vfid)) { - netdev_err(dev, "Failed to create vFID for VID=%d\n", - vid); - return PTR_ERR(new_vfid); - } - } - /* Invalidate existing {Port, VID} to vFID mapping and create a new - * one for the new vFID. - */ - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - false, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); - if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n", - vfid->vfid); - goto err_port_vid_to_fid_invalidate; - } + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, mt, valid, fid, + vid); +} - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - true, - mlxsw_sp_vfid_to_fid(new_vfid->vfid), - vid); - if (err) { - netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", - new_vfid->vfid); - goto err_port_vid_to_fid_validate; - } +static int mlxsw_sp_vport_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *br_dev) +{ + struct mlxsw_sp_fid *f; + int err; - err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); - if (err) { - netdev_err(dev, "Failed to disable learning\n"); - goto err_port_vid_learning_set; + f = mlxsw_sp_vfid_find(mlxsw_sp_vport->mlxsw_sp, br_dev); + if (!f) { + f = mlxsw_sp_vfid_create(mlxsw_sp_vport->mlxsw_sp, br_dev); + if (IS_ERR(f)) + return PTR_ERR(f); } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, - false); - if (err) { - netdev_err(dev, "Failed clear to clear flooding\n"); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); + if (err) goto err_vport_flood_set; - } - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_FORWARDING); - if (err) { - netdev_err(dev, "Failed to set STP state\n"); - goto err_port_stp_state_set; - } - - if (flush_fdb && mlxsw_sp_vport_fdb_flush(mlxsw_sp_vport)) - netdev_err(dev, "Failed to flush FDB\n"); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); + if (err) + goto err_vport_fid_map; - /* Switch between the vFIDs and destroy the old one if needed. */ - new_vfid->nr_vports++; - mlxsw_sp_vport->vport.vfid = new_vfid; - vfid->nr_vports--; - if (!vfid->nr_vports) - mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid); + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, f); + f->ref_count++; - mlxsw_sp_vport->learning = 0; - mlxsw_sp_vport->learning_sync = 0; - mlxsw_sp_vport->uc_flood = 0; - mlxsw_sp_vport->bridged = 0; + netdev_dbg(mlxsw_sp_vport->dev, "Joined FID=%d\n", f->fid); return 0; -err_port_stp_state_set: +err_vport_fid_map: + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); err_vport_flood_set: -err_port_vid_learning_set: -err_port_vid_to_fid_validate: -err_port_vid_to_fid_invalidate: - /* Rollback vFID only if new. */ - if (!new_vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp, new_vfid); + if (!f->ref_count) + mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); return err; } +static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + netdev_dbg(mlxsw_sp_vport->dev, "Left FID=%d\n", f->fid); + + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); + + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); + + mlxsw_sp_port_fdb_flush(mlxsw_sp_vport, f->fid); + + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); + if (--f->ref_count == 0) + mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); +} + static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, struct net_device *br_dev) { - struct mlxsw_sp_vfid *old_vfid = mlxsw_sp_vport->vport.vfid; - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; - struct mlxsw_sp_vfid *vfid; int err; - vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (!vfid) { - vfid = mlxsw_sp_br_vfid_create(mlxsw_sp, br_dev); - if (IS_ERR(vfid)) { - netdev_err(dev, "Failed to create bridge vFID\n"); - return PTR_ERR(vfid); - } - } + if (f && !WARN_ON(!f->leave)) + f->leave(mlxsw_sp_vport); - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, true, false); + err = mlxsw_sp_vport_vfid_join(mlxsw_sp_vport, br_dev); if (err) { - netdev_err(dev, "Failed to setup flooding for vFID=%d\n", - vfid->vfid); - goto err_port_flood_set; + netdev_err(dev, "Failed to join vFID\n"); + return err; } err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); @@ -3349,38 +3752,6 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, goto err_port_vid_learning_set; } - /* We need to invalidate existing {Port, VID} to vFID mapping and - * create a new one for the bridge's vFID. - */ - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - false, - mlxsw_sp_vfid_to_fid(old_vfid->vfid), - vid); - if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n", - old_vfid->vfid); - goto err_port_vid_to_fid_invalidate; - } - - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - true, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); - if (err) { - netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", - vfid->vfid); - goto err_port_vid_to_fid_validate; - } - - /* Switch between the vFIDs and destroy the old one if needed. */ - vfid->nr_vports++; - mlxsw_sp_vport->vport.vfid = vfid; - old_vfid->nr_vports--; - if (!old_vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp, old_vfid); - mlxsw_sp_vport->learning = 1; mlxsw_sp_vport->learning_sync = 1; mlxsw_sp_vport->uc_flood = 1; @@ -3388,20 +3759,25 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, return 0; -err_port_vid_to_fid_validate: - mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false, - mlxsw_sp_vfid_to_fid(old_vfid->vfid), vid); -err_port_vid_to_fid_invalidate: - mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); err_port_vid_learning_set: - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, false); -err_port_flood_set: - if (!vfid->nr_vports) - mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid); + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); return err; } +static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); + + mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); + + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); + + mlxsw_sp_vport->learning = 0; + mlxsw_sp_vport->learning_sync = 0; + mlxsw_sp_vport->uc_flood = 0; + mlxsw_sp_vport->bridged = 0; +} + static bool mlxsw_sp_port_master_bridge_check(const struct mlxsw_sp_port *mlxsw_sp_port, const struct net_device *br_dev) @@ -3410,7 +3786,9 @@ mlxsw_sp_port_master_bridge_check(const struct mlxsw_sp_port *mlxsw_sp_port, list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, vport.list) { - if (mlxsw_sp_vport_br_get(mlxsw_sp_vport) == br_dev) + struct net_device *dev = mlxsw_sp_vport_dev_get(mlxsw_sp_vport); + + if (dev && dev == br_dev) return false; } @@ -3425,56 +3803,39 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, struct netdev_notifier_changeupper_info *info = ptr; struct mlxsw_sp_port *mlxsw_sp_vport; struct net_device *upper_dev; - int err; + int err = 0; mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); switch (event) { case NETDEV_PRECHANGEUPPER: upper_dev = info->upper_dev; - if (!info->master || !info->linking) - break; if (!netif_is_bridge_master(upper_dev)) - return NOTIFY_BAD; + return -EINVAL; + if (!info->linking) + break; /* We can't have multiple VLAN interfaces configured on * the same port and being members in the same bridge. */ if (!mlxsw_sp_port_master_bridge_check(mlxsw_sp_port, upper_dev)) - return NOTIFY_BAD; + return -EINVAL; break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; - if (!info->master) - break; if (info->linking) { - if (!mlxsw_sp_vport) { - WARN_ON(!mlxsw_sp_vport); - return NOTIFY_BAD; - } + if (WARN_ON(!mlxsw_sp_vport)) + return -EINVAL; err = mlxsw_sp_vport_bridge_join(mlxsw_sp_vport, upper_dev); - if (err) { - netdev_err(dev, "Failed to join bridge\n"); - return NOTIFY_BAD; - } } else { - /* We ignore bridge's unlinking notifications if vPort - * is gone, since we already left the bridge when the - * VLAN device was unlinked from the real device. - */ if (!mlxsw_sp_vport) - return NOTIFY_DONE; - err = mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, - upper_dev, true); - if (err) { - netdev_err(dev, "Failed to leave bridge\n"); - return NOTIFY_BAD; - } + return 0; + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); } } - return NOTIFY_DONE; + return err; } static int mlxsw_sp_netdevice_lag_vport_event(struct net_device *lag_dev, @@ -3489,12 +3850,12 @@ static int mlxsw_sp_netdevice_lag_vport_event(struct net_device *lag_dev, if (mlxsw_sp_port_dev_check(dev)) { ret = mlxsw_sp_netdevice_vport_event(dev, event, ptr, vid); - if (ret == NOTIFY_BAD) + if (ret) return ret; } } - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_vlan_event(struct net_device *vlan_dev, @@ -3510,35 +3871,44 @@ static int mlxsw_sp_netdevice_vlan_event(struct net_device *vlan_dev, return mlxsw_sp_netdevice_lag_vport_event(real_dev, event, ptr, vid); - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_event(struct notifier_block *unused, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); + int err = 0; - if (mlxsw_sp_port_dev_check(dev)) - return mlxsw_sp_netdevice_port_event(dev, event, ptr); - - if (netif_is_lag_master(dev)) - return mlxsw_sp_netdevice_lag_event(dev, event, ptr); + if (event == NETDEV_CHANGEADDR || event == NETDEV_CHANGEMTU) + err = mlxsw_sp_netdevice_router_port_event(dev); + else if (mlxsw_sp_port_dev_check(dev)) + err = mlxsw_sp_netdevice_port_event(dev, event, ptr); + else if (netif_is_lag_master(dev)) + err = mlxsw_sp_netdevice_lag_event(dev, event, ptr); + else if (netif_is_bridge_master(dev)) + err = mlxsw_sp_netdevice_bridge_event(dev, event, ptr); + else if (is_vlan_dev(dev)) + err = mlxsw_sp_netdevice_vlan_event(dev, event, ptr); - if (is_vlan_dev(dev)) - return mlxsw_sp_netdevice_vlan_event(dev, event, ptr); - - return NOTIFY_DONE; + return notifier_from_errno(err); } static struct notifier_block mlxsw_sp_netdevice_nb __read_mostly = { .notifier_call = mlxsw_sp_netdevice_event, }; +static struct notifier_block mlxsw_sp_inetaddr_nb __read_mostly = { + .notifier_call = mlxsw_sp_inetaddr_event, + .priority = 10, /* Must be called before FIB notifier block */ +}; + static int __init mlxsw_sp_module_init(void) { int err; register_netdevice_notifier(&mlxsw_sp_netdevice_nb); + register_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); err = mlxsw_core_driver_register(&mlxsw_sp_driver); if (err) goto err_core_driver_register; @@ -3552,6 +3922,7 @@ err_core_driver_register: static void __exit mlxsw_sp_module_exit(void) { mlxsw_core_driver_unregister(&mlxsw_sp_driver); + unregister_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index e2c022d3e2f3..ef4ac8987a2a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -39,19 +39,22 @@ #include <linux/types.h> #include <linux/netdevice.h> +#include <linux/rhashtable.h> #include <linux/bitops.h> #include <linux/if_vlan.h> #include <linux/list.h> #include <linux/dcbnl.h> +#include <linux/in6.h> #include <net/switchdev.h> #include "port.h" #include "core.h" #define MLXSW_SP_VFID_BASE VLAN_N_VID -#define MLXSW_SP_VFID_PORT_MAX 512 /* Non-bridged VLAN interfaces */ -#define MLXSW_SP_VFID_BR_MAX 6144 /* Bridged VLAN interfaces */ -#define MLXSW_SP_VFID_MAX (MLXSW_SP_VFID_PORT_MAX + MLXSW_SP_VFID_BR_MAX) +#define MLXSW_SP_VFID_MAX 6656 /* Bridged VLAN interfaces */ + +#define MLXSW_SP_RFID_BASE 15360 +#define MLXSW_SP_RIF_MAX 800 #define MLXSW_SP_LAG_MAX 64 #define MLXSW_SP_PORT_PER_LAG_MAX 16 @@ -60,6 +63,12 @@ #define MLXSW_SP_PORTS_PER_CLUSTER_MAX 4 +#define MLXSW_SP_LPM_TREE_MIN 2 /* trees 0 and 1 are reserved */ +#define MLXSW_SP_LPM_TREE_MAX 22 +#define MLXSW_SP_LPM_TREE_COUNT (MLXSW_SP_LPM_TREE_MAX - MLXSW_SP_LPM_TREE_MIN) + +#define MLXSW_SP_VIRTUAL_ROUTER_MAX 256 + #define MLXSW_SP_PORT_BASE_SPEED 25000 /* Mb/s */ #define MLXSW_SP_BYTES_PER_CELL 96 @@ -67,6 +76,10 @@ #define MLXSW_SP_BYTES_TO_CELLS(b) DIV_ROUND_UP(b, MLXSW_SP_BYTES_PER_CELL) #define MLXSW_SP_CELLS_TO_BYTES(c) (c * MLXSW_SP_BYTES_PER_CELL) +#define MLXSW_SP_KVD_LINEAR_SIZE 65536 /* entries */ +#define MLXSW_SP_KVD_HASH_SINGLE_SIZE 163840 /* entries */ +#define MLXSW_SP_KVD_HASH_DOUBLE_SIZE 32768 /* entries */ + /* Maximum delay buffer needed in case of PAUSE frames, in cells. * Assumes 100m cable and maximum MTU. */ @@ -87,12 +100,22 @@ struct mlxsw_sp_upper { unsigned int ref_count; }; -struct mlxsw_sp_vfid { +struct mlxsw_sp_fid { + void (*leave)(struct mlxsw_sp_port *mlxsw_sp_vport); struct list_head list; - u16 nr_vports; - u16 vfid; /* Starting at 0 */ - struct net_device *br_dev; - u16 vid; + unsigned int ref_count; + struct net_device *dev; + struct mlxsw_sp_rif *r; + u16 fid; +}; + +struct mlxsw_sp_rif { + struct net_device *dev; + unsigned int ref_count; + struct mlxsw_sp_fid *f; + unsigned char addr[ETH_ALEN]; + int mtu; + u16 rif; }; struct mlxsw_sp_mid { @@ -115,7 +138,17 @@ static inline u16 mlxsw_sp_fid_to_vfid(u16 fid) static inline bool mlxsw_sp_fid_is_vfid(u16 fid) { - return fid >= MLXSW_SP_VFID_BASE; + return fid >= MLXSW_SP_VFID_BASE && fid < MLXSW_SP_RFID_BASE; +} + +static inline bool mlxsw_sp_fid_is_rfid(u16 fid) +{ + return fid >= MLXSW_SP_RFID_BASE; +} + +static inline u16 mlxsw_sp_rif_sp_to_fid(u16 rif) +{ + return MLXSW_SP_RFID_BASE + rif; } struct mlxsw_sp_sb_pr { @@ -152,20 +185,60 @@ struct mlxsw_sp_sb { } ports[MLXSW_PORT_MAX_PORTS]; }; -struct mlxsw_sp { +#define MLXSW_SP_PREFIX_COUNT (sizeof(struct in6_addr) * BITS_PER_BYTE) + +struct mlxsw_sp_prefix_usage { + DECLARE_BITMAP(b, MLXSW_SP_PREFIX_COUNT); +}; + +enum mlxsw_sp_l3proto { + MLXSW_SP_L3_PROTO_IPV4, + MLXSW_SP_L3_PROTO_IPV6, +}; + +struct mlxsw_sp_lpm_tree { + u8 id; /* tree ID */ + unsigned int ref_count; + enum mlxsw_sp_l3proto proto; + struct mlxsw_sp_prefix_usage prefix_usage; +}; + +struct mlxsw_sp_fib; + +struct mlxsw_sp_vr { + u16 id; /* virtual router ID */ + bool used; + enum mlxsw_sp_l3proto proto; + u32 tb_id; /* kernel fib table id */ + struct mlxsw_sp_lpm_tree *lpm_tree; + struct mlxsw_sp_fib *fib; +}; + +struct mlxsw_sp_router { + struct mlxsw_sp_lpm_tree lpm_trees[MLXSW_SP_LPM_TREE_COUNT]; + struct mlxsw_sp_vr vrs[MLXSW_SP_VIRTUAL_ROUTER_MAX]; + struct rhashtable neigh_ht; struct { - struct list_head list; - unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_VFID_PORT_MAX)]; - } port_vfids; + struct delayed_work dw; + unsigned long interval; /* ms */ + } neighs_update; + struct delayed_work nexthop_probe_dw; +#define MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL 5000 /* ms */ + struct list_head nexthop_group_list; + struct list_head nexthop_neighs_list; +}; + +struct mlxsw_sp { struct { struct list_head list; - unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_VFID_BR_MAX)]; - } br_vfids; + DECLARE_BITMAP(mapped, MLXSW_SP_VFID_MAX); + } vfids; struct { struct list_head list; - unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_MID_MAX)]; + DECLARE_BITMAP(mapped, MLXSW_SP_MID_MAX); } br_mids; - unsigned long active_fids[BITS_TO_LONGS(VLAN_N_VID)]; + struct list_head fids; /* VLAN-aware bridge FIDs */ + struct mlxsw_sp_rif *rifs[MLXSW_SP_RIF_MAX]; struct mlxsw_sp_port **ports; struct mlxsw_core *core; const struct mlxsw_bus_info *bus_info; @@ -183,6 +256,10 @@ struct mlxsw_sp { struct mlxsw_sp_upper lags[MLXSW_SP_LAG_MAX]; u8 port_to_module[MLXSW_PORT_MAX_PORTS]; struct mlxsw_sp_sb sb; + struct mlxsw_sp_router router; + struct { + DECLARE_BITMAP(usage, MLXSW_SP_KVD_LINEAR_SIZE); + } kvdl; }; static inline struct mlxsw_sp_upper * @@ -217,7 +294,7 @@ struct mlxsw_sp_port { u16 lag_id; struct { struct list_head list; - struct mlxsw_sp_vfid *vfid; + struct mlxsw_sp_fid *f; u16 vid; } vport; struct { @@ -229,6 +306,11 @@ struct mlxsw_sp_port { struct ieee_maxrate *maxrate; struct ieee_pfc *pfc; } dcb; + struct { + u8 module; + u8 width; + u8 lane; + } mapping; /* 802.1Q bridge VLANs */ unsigned long *active_vlans; unsigned long *untagged_vlans; @@ -236,6 +318,9 @@ struct mlxsw_sp_port { struct list_head vports_list; }; +struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev); +void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port); + static inline bool mlxsw_sp_port_is_pause_en(const struct mlxsw_sp_port *mlxsw_sp_port) { @@ -254,28 +339,38 @@ mlxsw_sp_port_lagged_get(struct mlxsw_sp *mlxsw_sp, u16 lag_id, u8 port_index) return mlxsw_sp_port && mlxsw_sp_port->lagged ? mlxsw_sp_port : NULL; } +static inline u16 +mlxsw_sp_vport_vid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +{ + return mlxsw_sp_vport->vport.vid; +} + static inline bool mlxsw_sp_port_is_vport(const struct mlxsw_sp_port *mlxsw_sp_port) { - return mlxsw_sp_port->vport.vfid; + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_port); + + return vid != 0; } -static inline struct net_device * -mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +static inline void mlxsw_sp_vport_fid_set(struct mlxsw_sp_port *mlxsw_sp_vport, + struct mlxsw_sp_fid *f) { - return mlxsw_sp_vport->vport.vfid->br_dev; + mlxsw_sp_vport->vport.f = f; } -static inline u16 -mlxsw_sp_vport_vid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +static inline struct mlxsw_sp_fid * +mlxsw_sp_vport_fid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { - return mlxsw_sp_vport->vport.vid; + return mlxsw_sp_vport->vport.f; } -static inline u16 -mlxsw_sp_vport_vfid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +static inline struct net_device * +mlxsw_sp_vport_dev_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { - return mlxsw_sp_vport->vport.vfid->vfid; + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + return f ? f->dev : NULL; } static inline struct mlxsw_sp_port * @@ -293,20 +388,60 @@ mlxsw_sp_port_vport_find(const struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) } static inline struct mlxsw_sp_port * -mlxsw_sp_port_vport_find_by_vfid(const struct mlxsw_sp_port *mlxsw_sp_port, - u16 vfid) +mlxsw_sp_port_vport_find_by_fid(const struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) { struct mlxsw_sp_port *mlxsw_sp_vport; list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, vport.list) { - if (mlxsw_sp_vport_vfid_get(mlxsw_sp_vport) == vfid) + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + if (f && f->fid == fid) return mlxsw_sp_vport; } return NULL; } +static inline struct mlxsw_sp_fid *mlxsw_sp_fid_find(struct mlxsw_sp *mlxsw_sp, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + list_for_each_entry(f, &mlxsw_sp->fids, list) + if (f->fid == fid) + return f; + + return NULL; +} + +static inline struct mlxsw_sp_fid * +mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, + const struct net_device *br_dev) +{ + struct mlxsw_sp_fid *f; + + list_for_each_entry(f, &mlxsw_sp->vfids.list, list) + if (f->dev == br_dev) + return f; + + return NULL; +} + +static inline struct mlxsw_sp_rif * +mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, + const struct net_device *dev) +{ + int i; + + for (i = 0; i < MLXSW_SP_RIF_MAX; i++) + if (mlxsw_sp->rifs[i] && mlxsw_sp->rifs[i]->dev == dev) + return mlxsw_sp->rifs[i]; + + return NULL; +} + enum mlxsw_sp_flood_table { MLXSW_SP_FLOOD_TABLE_UC, MLXSW_SP_FLOOD_TABLE_BM, @@ -359,12 +494,17 @@ int mlxsw_sp_port_vlan_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool is_member, bool untagged); int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid); -int mlxsw_sp_port_kill_vid(struct net_device *dev, - __be16 __always_unused proto, u16 vid); -int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid, - bool set, bool only_uc); +int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, + bool set); void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port); int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid); +int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid); +int mlxsw_sp_rif_fdb_op(struct mlxsw_sp *mlxsw_sp, const char *mac, u16 fid, + bool adding); +struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid); +void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f); +void mlxsw_sp_rif_bridge_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *r); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight); @@ -394,4 +534,19 @@ static inline void mlxsw_sp_port_dcb_fini(struct mlxsw_sp_port *mlxsw_sp_port) #endif +int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp); +void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp); +int mlxsw_sp_router_fib4_add(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans); +int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4); +int mlxsw_sp_router_neigh_construct(struct net_device *dev, + struct neighbour *n); +void mlxsw_sp_router_neigh_destroy(struct net_device *dev, + struct neighbour *n); + +int mlxsw_sp_kvdl_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int entry_count); +void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, int entry_index); + #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c new file mode 100644 index 000000000000..ac321e8e5c1a --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c @@ -0,0 +1,91 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c + * Copyright (c) 2016 Mellanox Technologies. All rights reserved. + * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/kernel.h> +#include <linux/bitops.h> + +#include "spectrum.h" + +#define MLXSW_SP_KVDL_SINGLE_BASE 0 +#define MLXSW_SP_KVDL_SINGLE_SIZE 16384 +#define MLXSW_SP_KVDL_CHUNKS_BASE \ + (MLXSW_SP_KVDL_SINGLE_BASE + MLXSW_SP_KVDL_SINGLE_SIZE) +#define MLXSW_SP_KVDL_CHUNKS_SIZE \ + (MLXSW_SP_KVD_LINEAR_SIZE - MLXSW_SP_KVDL_CHUNKS_BASE) +#define MLXSW_SP_CHUNK_MAX 32 + +int mlxsw_sp_kvdl_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int entry_count) +{ + int entry_index; + int size; + int type_base; + int type_size; + int type_entries; + + if (entry_count == 0 || entry_count > MLXSW_SP_CHUNK_MAX) { + return -EINVAL; + } else if (entry_count == 1) { + type_base = MLXSW_SP_KVDL_SINGLE_BASE; + type_size = MLXSW_SP_KVDL_SINGLE_SIZE; + type_entries = 1; + } else { + type_base = MLXSW_SP_KVDL_CHUNKS_BASE; + type_size = MLXSW_SP_KVDL_CHUNKS_SIZE; + type_entries = MLXSW_SP_CHUNK_MAX; + } + + entry_index = type_base; + size = type_base + type_size; + for_each_clear_bit_from(entry_index, mlxsw_sp->kvdl.usage, size) { + int i; + + for (i = 0; i < type_entries; i++) + set_bit(entry_index + i, mlxsw_sp->kvdl.usage); + return entry_index; + } + return -ENOBUFS; +} + +void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, int entry_index) +{ + int type_entries; + int i; + + if (entry_index < MLXSW_SP_KVDL_CHUNKS_BASE) + type_entries = 1; + else + type_entries = MLXSW_SP_CHUNK_MAX; + for (i = 0; i < type_entries; i++) + clear_bit(entry_index + i, mlxsw_sp->kvdl.usage); +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c new file mode 100644 index 000000000000..e084ea5448ac --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -0,0 +1,1814 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c + * Copyright (c) 2016 Mellanox Technologies. All rights reserved. + * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com> + * Copyright (c) 2016 Ido Schimmel <idosch@mellanox.com> + * Copyright (c) 2016 Yotam Gigi <yotamg@mellanox.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/rhashtable.h> +#include <linux/bitops.h> +#include <linux/in6.h> +#include <linux/notifier.h> +#include <net/netevent.h> +#include <net/neighbour.h> +#include <net/arp.h> + +#include "spectrum.h" +#include "core.h" +#include "reg.h" + +#define mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) \ + for_each_set_bit(prefix, (prefix_usage)->b, MLXSW_SP_PREFIX_COUNT) + +static bool +mlxsw_sp_prefix_usage_subset(struct mlxsw_sp_prefix_usage *prefix_usage1, + struct mlxsw_sp_prefix_usage *prefix_usage2) +{ + unsigned char prefix; + + mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage1) { + if (!test_bit(prefix, prefix_usage2->b)) + return false; + } + return true; +} + +static bool +mlxsw_sp_prefix_usage_eq(struct mlxsw_sp_prefix_usage *prefix_usage1, + struct mlxsw_sp_prefix_usage *prefix_usage2) +{ + return !memcmp(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1)); +} + +static bool +mlxsw_sp_prefix_usage_none(struct mlxsw_sp_prefix_usage *prefix_usage) +{ + struct mlxsw_sp_prefix_usage prefix_usage_none = {{ 0 } }; + + return mlxsw_sp_prefix_usage_eq(prefix_usage, &prefix_usage_none); +} + +static void +mlxsw_sp_prefix_usage_cpy(struct mlxsw_sp_prefix_usage *prefix_usage1, + struct mlxsw_sp_prefix_usage *prefix_usage2) +{ + memcpy(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1)); +} + +static void +mlxsw_sp_prefix_usage_zero(struct mlxsw_sp_prefix_usage *prefix_usage) +{ + memset(prefix_usage, 0, sizeof(*prefix_usage)); +} + +static void +mlxsw_sp_prefix_usage_set(struct mlxsw_sp_prefix_usage *prefix_usage, + unsigned char prefix_len) +{ + set_bit(prefix_len, prefix_usage->b); +} + +static void +mlxsw_sp_prefix_usage_clear(struct mlxsw_sp_prefix_usage *prefix_usage, + unsigned char prefix_len) +{ + clear_bit(prefix_len, prefix_usage->b); +} + +struct mlxsw_sp_fib_key { + unsigned char addr[sizeof(struct in6_addr)]; + unsigned char prefix_len; +}; + +enum mlxsw_sp_fib_entry_type { + MLXSW_SP_FIB_ENTRY_TYPE_REMOTE, + MLXSW_SP_FIB_ENTRY_TYPE_LOCAL, + MLXSW_SP_FIB_ENTRY_TYPE_TRAP, +}; + +struct mlxsw_sp_nexthop_group; + +struct mlxsw_sp_fib_entry { + struct rhash_head ht_node; + struct mlxsw_sp_fib_key key; + enum mlxsw_sp_fib_entry_type type; + u8 added:1; + u16 rif; /* used for action local */ + struct mlxsw_sp_vr *vr; + struct list_head nexthop_group_node; + struct mlxsw_sp_nexthop_group *nh_group; +}; + +struct mlxsw_sp_fib { + struct rhashtable ht; + unsigned long prefix_ref_count[MLXSW_SP_PREFIX_COUNT]; + struct mlxsw_sp_prefix_usage prefix_usage; +}; + +static const struct rhashtable_params mlxsw_sp_fib_ht_params = { + .key_offset = offsetof(struct mlxsw_sp_fib_entry, key), + .head_offset = offsetof(struct mlxsw_sp_fib_entry, ht_node), + .key_len = sizeof(struct mlxsw_sp_fib_key), + .automatic_shrinking = true, +}; + +static int mlxsw_sp_fib_entry_insert(struct mlxsw_sp_fib *fib, + struct mlxsw_sp_fib_entry *fib_entry) +{ + unsigned char prefix_len = fib_entry->key.prefix_len; + int err; + + err = rhashtable_insert_fast(&fib->ht, &fib_entry->ht_node, + mlxsw_sp_fib_ht_params); + if (err) + return err; + if (fib->prefix_ref_count[prefix_len]++ == 0) + mlxsw_sp_prefix_usage_set(&fib->prefix_usage, prefix_len); + return 0; +} + +static void mlxsw_sp_fib_entry_remove(struct mlxsw_sp_fib *fib, + struct mlxsw_sp_fib_entry *fib_entry) +{ + unsigned char prefix_len = fib_entry->key.prefix_len; + + if (--fib->prefix_ref_count[prefix_len] == 0) + mlxsw_sp_prefix_usage_clear(&fib->prefix_usage, prefix_len); + rhashtable_remove_fast(&fib->ht, &fib_entry->ht_node, + mlxsw_sp_fib_ht_params); +} + +static struct mlxsw_sp_fib_entry * +mlxsw_sp_fib_entry_create(struct mlxsw_sp_fib *fib, const void *addr, + size_t addr_len, unsigned char prefix_len) +{ + struct mlxsw_sp_fib_entry *fib_entry; + + fib_entry = kzalloc(sizeof(*fib_entry), GFP_KERNEL); + if (!fib_entry) + return NULL; + memcpy(fib_entry->key.addr, addr, addr_len); + fib_entry->key.prefix_len = prefix_len; + return fib_entry; +} + +static void mlxsw_sp_fib_entry_destroy(struct mlxsw_sp_fib_entry *fib_entry) +{ + kfree(fib_entry); +} + +static struct mlxsw_sp_fib_entry * +mlxsw_sp_fib_entry_lookup(struct mlxsw_sp_fib *fib, const void *addr, + size_t addr_len, unsigned char prefix_len) +{ + struct mlxsw_sp_fib_key key = {{ 0 } }; + + memcpy(key.addr, addr, addr_len); + key.prefix_len = prefix_len; + return rhashtable_lookup_fast(&fib->ht, &key, mlxsw_sp_fib_ht_params); +} + +static struct mlxsw_sp_fib *mlxsw_sp_fib_create(void) +{ + struct mlxsw_sp_fib *fib; + int err; + + fib = kzalloc(sizeof(*fib), GFP_KERNEL); + if (!fib) + return ERR_PTR(-ENOMEM); + err = rhashtable_init(&fib->ht, &mlxsw_sp_fib_ht_params); + if (err) + goto err_rhashtable_init; + return fib; + +err_rhashtable_init: + kfree(fib); + return ERR_PTR(err); +} + +static void mlxsw_sp_fib_destroy(struct mlxsw_sp_fib *fib) +{ + rhashtable_destroy(&fib->ht); + kfree(fib); +} + +static struct mlxsw_sp_lpm_tree * +mlxsw_sp_lpm_tree_find_unused(struct mlxsw_sp *mlxsw_sp, bool one_reserved) +{ + static struct mlxsw_sp_lpm_tree *lpm_tree; + int i; + + for (i = 0; i < MLXSW_SP_LPM_TREE_COUNT; i++) { + lpm_tree = &mlxsw_sp->router.lpm_trees[i]; + if (lpm_tree->ref_count == 0) { + if (one_reserved) + one_reserved = false; + else + return lpm_tree; + } + } + return NULL; +} + +static int mlxsw_sp_lpm_tree_alloc(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + char ralta_pl[MLXSW_REG_RALTA_LEN]; + + mlxsw_reg_ralta_pack(ralta_pl, true, lpm_tree->proto, lpm_tree->id); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl); +} + +static int mlxsw_sp_lpm_tree_free(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + char ralta_pl[MLXSW_REG_RALTA_LEN]; + + mlxsw_reg_ralta_pack(ralta_pl, false, lpm_tree->proto, lpm_tree->id); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl); +} + +static int +mlxsw_sp_lpm_tree_left_struct_set(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_prefix_usage *prefix_usage, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + char ralst_pl[MLXSW_REG_RALST_LEN]; + u8 root_bin = 0; + u8 prefix; + u8 last_prefix = MLXSW_REG_RALST_BIN_NO_CHILD; + + mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) + root_bin = prefix; + + mlxsw_reg_ralst_pack(ralst_pl, root_bin, lpm_tree->id); + mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) { + if (prefix == 0) + continue; + mlxsw_reg_ralst_bin_pack(ralst_pl, prefix, last_prefix, + MLXSW_REG_RALST_BIN_NO_CHILD); + last_prefix = prefix; + } + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralst), ralst_pl); +} + +static struct mlxsw_sp_lpm_tree * +mlxsw_sp_lpm_tree_create(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_prefix_usage *prefix_usage, + enum mlxsw_sp_l3proto proto, bool one_reserved) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + int err; + + lpm_tree = mlxsw_sp_lpm_tree_find_unused(mlxsw_sp, one_reserved); + if (!lpm_tree) + return ERR_PTR(-EBUSY); + lpm_tree->proto = proto; + err = mlxsw_sp_lpm_tree_alloc(mlxsw_sp, lpm_tree); + if (err) + return ERR_PTR(err); + + err = mlxsw_sp_lpm_tree_left_struct_set(mlxsw_sp, prefix_usage, + lpm_tree); + if (err) + goto err_left_struct_set; + return lpm_tree; + +err_left_struct_set: + mlxsw_sp_lpm_tree_free(mlxsw_sp, lpm_tree); + return ERR_PTR(err); +} + +static int mlxsw_sp_lpm_tree_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + return mlxsw_sp_lpm_tree_free(mlxsw_sp, lpm_tree); +} + +static struct mlxsw_sp_lpm_tree * +mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_prefix_usage *prefix_usage, + enum mlxsw_sp_l3proto proto, bool one_reserved) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + int i; + + for (i = 0; i < MLXSW_SP_LPM_TREE_COUNT; i++) { + lpm_tree = &mlxsw_sp->router.lpm_trees[i]; + if (lpm_tree->proto == proto && + mlxsw_sp_prefix_usage_eq(&lpm_tree->prefix_usage, + prefix_usage)) + goto inc_ref_count; + } + lpm_tree = mlxsw_sp_lpm_tree_create(mlxsw_sp, prefix_usage, + proto, one_reserved); + if (IS_ERR(lpm_tree)) + return lpm_tree; + +inc_ref_count: + lpm_tree->ref_count++; + return lpm_tree; +} + +static int mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + if (--lpm_tree->ref_count == 0) + return mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree); + return 0; +} + +static void mlxsw_sp_lpm_init(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + int i; + + for (i = 0; i < MLXSW_SP_LPM_TREE_COUNT; i++) { + lpm_tree = &mlxsw_sp->router.lpm_trees[i]; + lpm_tree->id = i + MLXSW_SP_LPM_TREE_MIN; + } +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_find_unused(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_vr *vr; + int i; + + for (i = 0; i < MLXSW_SP_VIRTUAL_ROUTER_MAX; i++) { + vr = &mlxsw_sp->router.vrs[i]; + if (!vr->used) + return vr; + } + return NULL; +} + +static int mlxsw_sp_vr_lpm_tree_bind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr) +{ + char raltb_pl[MLXSW_REG_RALTB_LEN]; + + mlxsw_reg_raltb_pack(raltb_pl, vr->id, vr->proto, vr->lpm_tree->id); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb), raltb_pl); +} + +static int mlxsw_sp_vr_lpm_tree_unbind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr) +{ + char raltb_pl[MLXSW_REG_RALTB_LEN]; + + /* Bind to tree 0 which is default */ + mlxsw_reg_raltb_pack(raltb_pl, vr->id, vr->proto, 0); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb), raltb_pl); +} + +static u32 mlxsw_sp_fix_tb_id(u32 tb_id) +{ + /* For our purpose, squash main and local table into one */ + if (tb_id == RT_TABLE_LOCAL) + tb_id = RT_TABLE_MAIN; + return tb_id; +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_find(struct mlxsw_sp *mlxsw_sp, + u32 tb_id, + enum mlxsw_sp_l3proto proto) +{ + struct mlxsw_sp_vr *vr; + int i; + + tb_id = mlxsw_sp_fix_tb_id(tb_id); + for (i = 0; i < MLXSW_SP_VIRTUAL_ROUTER_MAX; i++) { + vr = &mlxsw_sp->router.vrs[i]; + if (vr->used && vr->proto == proto && vr->tb_id == tb_id) + return vr; + } + return NULL; +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_create(struct mlxsw_sp *mlxsw_sp, + unsigned char prefix_len, + u32 tb_id, + enum mlxsw_sp_l3proto proto) +{ + struct mlxsw_sp_prefix_usage req_prefix_usage; + struct mlxsw_sp_lpm_tree *lpm_tree; + struct mlxsw_sp_vr *vr; + int err; + + vr = mlxsw_sp_vr_find_unused(mlxsw_sp); + if (!vr) + return ERR_PTR(-EBUSY); + vr->fib = mlxsw_sp_fib_create(); + if (IS_ERR(vr->fib)) + return ERR_CAST(vr->fib); + + vr->proto = proto; + vr->tb_id = tb_id; + mlxsw_sp_prefix_usage_zero(&req_prefix_usage); + mlxsw_sp_prefix_usage_set(&req_prefix_usage, prefix_len); + lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, &req_prefix_usage, + proto, true); + if (IS_ERR(lpm_tree)) { + err = PTR_ERR(lpm_tree); + goto err_tree_get; + } + vr->lpm_tree = lpm_tree; + err = mlxsw_sp_vr_lpm_tree_bind(mlxsw_sp, vr); + if (err) + goto err_tree_bind; + + vr->used = true; + return vr; + +err_tree_bind: + mlxsw_sp_lpm_tree_put(mlxsw_sp, vr->lpm_tree); +err_tree_get: + mlxsw_sp_fib_destroy(vr->fib); + + return ERR_PTR(err); +} + +static void mlxsw_sp_vr_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr) +{ + mlxsw_sp_vr_lpm_tree_unbind(mlxsw_sp, vr); + mlxsw_sp_lpm_tree_put(mlxsw_sp, vr->lpm_tree); + mlxsw_sp_fib_destroy(vr->fib); + vr->used = false; +} + +static int +mlxsw_sp_vr_lpm_tree_check(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr, + struct mlxsw_sp_prefix_usage *req_prefix_usage) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + + if (mlxsw_sp_prefix_usage_eq(req_prefix_usage, + &vr->lpm_tree->prefix_usage)) + return 0; + + lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, req_prefix_usage, + vr->proto, false); + if (IS_ERR(lpm_tree)) { + /* We failed to get a tree according to the required + * prefix usage. However, the current tree might be still good + * for us if our requirement is subset of the prefixes used + * in the tree. + */ + if (mlxsw_sp_prefix_usage_subset(req_prefix_usage, + &vr->lpm_tree->prefix_usage)) + return 0; + return PTR_ERR(lpm_tree); + } + + mlxsw_sp_vr_lpm_tree_unbind(mlxsw_sp, vr); + mlxsw_sp_lpm_tree_put(mlxsw_sp, vr->lpm_tree); + vr->lpm_tree = lpm_tree; + return mlxsw_sp_vr_lpm_tree_bind(mlxsw_sp, vr); +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_get(struct mlxsw_sp *mlxsw_sp, + unsigned char prefix_len, + u32 tb_id, + enum mlxsw_sp_l3proto proto) +{ + struct mlxsw_sp_vr *vr; + int err; + + tb_id = mlxsw_sp_fix_tb_id(tb_id); + vr = mlxsw_sp_vr_find(mlxsw_sp, tb_id, proto); + if (!vr) { + vr = mlxsw_sp_vr_create(mlxsw_sp, prefix_len, tb_id, proto); + if (IS_ERR(vr)) + return vr; + } else { + struct mlxsw_sp_prefix_usage req_prefix_usage; + + mlxsw_sp_prefix_usage_cpy(&req_prefix_usage, + &vr->fib->prefix_usage); + mlxsw_sp_prefix_usage_set(&req_prefix_usage, prefix_len); + /* Need to replace LPM tree in case new prefix is required. */ + err = mlxsw_sp_vr_lpm_tree_check(mlxsw_sp, vr, + &req_prefix_usage); + if (err) + return ERR_PTR(err); + } + return vr; +} + +static void mlxsw_sp_vr_put(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr) +{ + /* Destroy virtual router entity in case the associated FIB is empty + * and allow it to be used for other tables in future. Otherwise, + * check if some prefix usage did not disappear and change tree if + * that is the case. Note that in case new, smaller tree cannot be + * allocated, the original one will be kept being used. + */ + if (mlxsw_sp_prefix_usage_none(&vr->fib->prefix_usage)) + mlxsw_sp_vr_destroy(mlxsw_sp, vr); + else + mlxsw_sp_vr_lpm_tree_check(mlxsw_sp, vr, + &vr->fib->prefix_usage); +} + +static void mlxsw_sp_vrs_init(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_vr *vr; + int i; + + for (i = 0; i < MLXSW_SP_VIRTUAL_ROUTER_MAX; i++) { + vr = &mlxsw_sp->router.vrs[i]; + vr->id = i; + } +} + +struct mlxsw_sp_neigh_key { + unsigned char addr[sizeof(struct in6_addr)]; + struct net_device *dev; +}; + +struct mlxsw_sp_neigh_entry { + struct rhash_head ht_node; + struct mlxsw_sp_neigh_key key; + u16 rif; + struct neighbour *n; + bool offloaded; + struct delayed_work dw; + struct mlxsw_sp_port *mlxsw_sp_port; + unsigned char ha[ETH_ALEN]; + struct list_head nexthop_list; /* list of nexthops using + * this neigh entry + */ + struct list_head nexthop_neighs_list_node; +}; + +static const struct rhashtable_params mlxsw_sp_neigh_ht_params = { + .key_offset = offsetof(struct mlxsw_sp_neigh_entry, key), + .head_offset = offsetof(struct mlxsw_sp_neigh_entry, ht_node), + .key_len = sizeof(struct mlxsw_sp_neigh_key), +}; + +static int +mlxsw_sp_neigh_entry_insert(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry) +{ + return rhashtable_insert_fast(&mlxsw_sp->router.neigh_ht, + &neigh_entry->ht_node, + mlxsw_sp_neigh_ht_params); +} + +static void +mlxsw_sp_neigh_entry_remove(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry) +{ + rhashtable_remove_fast(&mlxsw_sp->router.neigh_ht, + &neigh_entry->ht_node, + mlxsw_sp_neigh_ht_params); +} + +static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work); + +static struct mlxsw_sp_neigh_entry * +mlxsw_sp_neigh_entry_create(const void *addr, size_t addr_len, + struct net_device *dev, u16 rif, + struct neighbour *n) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + + neigh_entry = kzalloc(sizeof(*neigh_entry), GFP_ATOMIC); + if (!neigh_entry) + return NULL; + memcpy(neigh_entry->key.addr, addr, addr_len); + neigh_entry->key.dev = dev; + neigh_entry->rif = rif; + neigh_entry->n = n; + INIT_DELAYED_WORK(&neigh_entry->dw, mlxsw_sp_router_neigh_update_hw); + INIT_LIST_HEAD(&neigh_entry->nexthop_list); + return neigh_entry; +} + +static void +mlxsw_sp_neigh_entry_destroy(struct mlxsw_sp_neigh_entry *neigh_entry) +{ + kfree(neigh_entry); +} + +static struct mlxsw_sp_neigh_entry * +mlxsw_sp_neigh_entry_lookup(struct mlxsw_sp *mlxsw_sp, const void *addr, + size_t addr_len, struct net_device *dev) +{ + struct mlxsw_sp_neigh_key key = {{ 0 } }; + + memcpy(key.addr, addr, addr_len); + key.dev = dev; + return rhashtable_lookup_fast(&mlxsw_sp->router.neigh_ht, + &key, mlxsw_sp_neigh_ht_params); +} + +int mlxsw_sp_router_neigh_construct(struct net_device *dev, + struct neighbour *n) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_neigh_entry *neigh_entry; + struct mlxsw_sp_rif *r; + u32 dip; + int err; + + if (n->tbl != &arp_tbl) + return 0; + + dip = ntohl(*((__be32 *) n->primary_key)); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &dip, sizeof(dip), + n->dev); + if (neigh_entry) { + WARN_ON(neigh_entry->n != n); + return 0; + } + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (WARN_ON(!r)) + return -EINVAL; + + neigh_entry = mlxsw_sp_neigh_entry_create(&dip, sizeof(dip), n->dev, + r->rif, n); + if (!neigh_entry) + return -ENOMEM; + err = mlxsw_sp_neigh_entry_insert(mlxsw_sp, neigh_entry); + if (err) + goto err_neigh_entry_insert; + return 0; + +err_neigh_entry_insert: + mlxsw_sp_neigh_entry_destroy(neigh_entry); + return err; +} + +void mlxsw_sp_router_neigh_destroy(struct net_device *dev, + struct neighbour *n) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_neigh_entry *neigh_entry; + u32 dip; + + if (n->tbl != &arp_tbl) + return; + + dip = ntohl(*((__be32 *) n->primary_key)); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &dip, sizeof(dip), + n->dev); + if (!neigh_entry) + return; + mlxsw_sp_neigh_entry_remove(mlxsw_sp, neigh_entry); + mlxsw_sp_neigh_entry_destroy(neigh_entry); +} + +static void +mlxsw_sp_router_neighs_update_interval_init(struct mlxsw_sp *mlxsw_sp) +{ + unsigned long interval = NEIGH_VAR(&arp_tbl.parms, DELAY_PROBE_TIME); + + mlxsw_sp->router.neighs_update.interval = jiffies_to_msecs(interval); +} + +static void mlxsw_sp_router_neigh_ent_ipv4_process(struct mlxsw_sp *mlxsw_sp, + char *rauhtd_pl, + int ent_index) +{ + struct net_device *dev; + struct neighbour *n; + __be32 dipn; + u32 dip; + u16 rif; + + mlxsw_reg_rauhtd_ent_ipv4_unpack(rauhtd_pl, ent_index, &rif, &dip); + + if (!mlxsw_sp->rifs[rif]) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Incorrect RIF in neighbour entry\n"); + return; + } + + dipn = htonl(dip); + dev = mlxsw_sp->rifs[rif]->dev; + n = neigh_lookup(&arp_tbl, &dipn, dev); + if (!n) { + netdev_err(dev, "Failed to find matching neighbour for IP=%pI4h\n", + &dip); + return; + } + + netdev_dbg(dev, "Updating neighbour with IP=%pI4h\n", &dip); + neigh_event_send(n, NULL); + neigh_release(n); +} + +static void mlxsw_sp_router_neigh_rec_ipv4_process(struct mlxsw_sp *mlxsw_sp, + char *rauhtd_pl, + int rec_index) +{ + u8 num_entries; + int i; + + num_entries = mlxsw_reg_rauhtd_ipv4_rec_num_entries_get(rauhtd_pl, + rec_index); + /* Hardware starts counting at 0, so add 1. */ + num_entries++; + + /* Each record consists of several neighbour entries. */ + for (i = 0; i < num_entries; i++) { + int ent_index; + + ent_index = rec_index * MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC + i; + mlxsw_sp_router_neigh_ent_ipv4_process(mlxsw_sp, rauhtd_pl, + ent_index); + } + +} + +static void mlxsw_sp_router_neigh_rec_process(struct mlxsw_sp *mlxsw_sp, + char *rauhtd_pl, int rec_index) +{ + switch (mlxsw_reg_rauhtd_rec_type_get(rauhtd_pl, rec_index)) { + case MLXSW_REG_RAUHTD_TYPE_IPV4: + mlxsw_sp_router_neigh_rec_ipv4_process(mlxsw_sp, rauhtd_pl, + rec_index); + break; + case MLXSW_REG_RAUHTD_TYPE_IPV6: + WARN_ON_ONCE(1); + break; + } +} + +static int mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp) +{ + char *rauhtd_pl; + u8 num_rec; + int i, err; + + rauhtd_pl = kmalloc(MLXSW_REG_RAUHTD_LEN, GFP_KERNEL); + if (!rauhtd_pl) + return -ENOMEM; + + /* Make sure the neighbour's netdev isn't removed in the + * process. + */ + rtnl_lock(); + do { + mlxsw_reg_rauhtd_pack(rauhtd_pl, MLXSW_REG_RAUHTD_TYPE_IPV4); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(rauhtd), + rauhtd_pl); + if (err) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to dump neighbour talbe\n"); + break; + } + num_rec = mlxsw_reg_rauhtd_num_rec_get(rauhtd_pl); + for (i = 0; i < num_rec; i++) + mlxsw_sp_router_neigh_rec_process(mlxsw_sp, rauhtd_pl, + i); + } while (num_rec); + rtnl_unlock(); + + kfree(rauhtd_pl); + return err; +} + +static void mlxsw_sp_router_neighs_update_nh(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + + /* Take RTNL mutex here to prevent lists from changes */ + rtnl_lock(); + list_for_each_entry(neigh_entry, &mlxsw_sp->router.nexthop_neighs_list, + nexthop_neighs_list_node) { + /* If this neigh have nexthops, make the kernel think this neigh + * is active regardless of the traffic. + */ + if (!list_empty(&neigh_entry->nexthop_list)) + neigh_event_send(neigh_entry->n, NULL); + } + rtnl_unlock(); +} + +static void +mlxsw_sp_router_neighs_update_work_schedule(struct mlxsw_sp *mlxsw_sp) +{ + unsigned long interval = mlxsw_sp->router.neighs_update.interval; + + mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, + msecs_to_jiffies(interval)); +} + +static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) +{ + struct mlxsw_sp *mlxsw_sp = container_of(work, struct mlxsw_sp, + router.neighs_update.dw.work); + int err; + + err = mlxsw_sp_router_neighs_update_rauhtd(mlxsw_sp); + if (err) + dev_err(mlxsw_sp->bus_info->dev, "Could not update kernel for neigh activity"); + + mlxsw_sp_router_neighs_update_nh(mlxsw_sp); + + mlxsw_sp_router_neighs_update_work_schedule(mlxsw_sp); +} + +static void mlxsw_sp_router_probe_unresolved_nexthops(struct work_struct *work) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + struct mlxsw_sp *mlxsw_sp = container_of(work, struct mlxsw_sp, + router.nexthop_probe_dw.work); + + /* Iterate over nexthop neighbours, find those who are unresolved and + * send arp on them. This solves the chicken-egg problem when + * the nexthop wouldn't get offloaded until the neighbor is resolved + * but it wouldn't get resolved ever in case traffic is flowing in HW + * using different nexthop. + * + * Take RTNL mutex here to prevent lists from changes. + */ + rtnl_lock(); + list_for_each_entry(neigh_entry, &mlxsw_sp->router.nexthop_neighs_list, + nexthop_neighs_list_node) { + if (!(neigh_entry->n->nud_state & NUD_VALID) && + !list_empty(&neigh_entry->nexthop_list)) + neigh_event_send(neigh_entry->n, NULL); + } + rtnl_unlock(); + + mlxsw_core_schedule_dw(&mlxsw_sp->router.nexthop_probe_dw, + MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL); +} + +static void +mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry, + bool removing); + +static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work) +{ + struct mlxsw_sp_neigh_entry *neigh_entry = + container_of(work, struct mlxsw_sp_neigh_entry, dw.work); + struct neighbour *n = neigh_entry->n; + struct mlxsw_sp_port *mlxsw_sp_port = neigh_entry->mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char rauht_pl[MLXSW_REG_RAUHT_LEN]; + struct net_device *dev; + bool entry_connected; + u8 nud_state; + bool updating; + bool removing; + bool adding; + u32 dip; + int err; + + read_lock_bh(&n->lock); + dip = ntohl(*((__be32 *) n->primary_key)); + memcpy(neigh_entry->ha, n->ha, sizeof(neigh_entry->ha)); + nud_state = n->nud_state; + dev = n->dev; + read_unlock_bh(&n->lock); + + entry_connected = nud_state & NUD_VALID; + adding = (!neigh_entry->offloaded) && entry_connected; + updating = neigh_entry->offloaded && entry_connected; + removing = neigh_entry->offloaded && !entry_connected; + + if (adding || updating) { + mlxsw_reg_rauht_pack4(rauht_pl, MLXSW_REG_RAUHT_OP_WRITE_ADD, + neigh_entry->rif, + neigh_entry->ha, dip); + err = mlxsw_reg_write(mlxsw_sp->core, + MLXSW_REG(rauht), rauht_pl); + if (err) { + netdev_err(dev, "Could not add neigh %pI4h\n", &dip); + neigh_entry->offloaded = false; + } else { + neigh_entry->offloaded = true; + } + mlxsw_sp_nexthop_neigh_update(mlxsw_sp, neigh_entry, false); + } else if (removing) { + mlxsw_reg_rauht_pack4(rauht_pl, MLXSW_REG_RAUHT_OP_WRITE_DELETE, + neigh_entry->rif, + neigh_entry->ha, dip); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rauht), + rauht_pl); + if (err) { + netdev_err(dev, "Could not delete neigh %pI4h\n", &dip); + neigh_entry->offloaded = true; + } else { + neigh_entry->offloaded = false; + } + mlxsw_sp_nexthop_neigh_update(mlxsw_sp, neigh_entry, true); + } + + neigh_release(n); + mlxsw_sp_port_dev_put(mlxsw_sp_port); +} + +static int mlxsw_sp_router_netevent_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + struct mlxsw_sp_port *mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp; + unsigned long interval; + struct net_device *dev; + struct neigh_parms *p; + struct neighbour *n; + u32 dip; + + switch (event) { + case NETEVENT_DELAY_PROBE_TIME_UPDATE: + p = ptr; + + /* We don't care about changes in the default table. */ + if (!p->dev || p->tbl != &arp_tbl) + return NOTIFY_DONE; + + /* We are in atomic context and can't take RTNL mutex, + * so use RCU variant to walk the device chain. + */ + mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(p->dev); + if (!mlxsw_sp_port) + return NOTIFY_DONE; + + mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + interval = jiffies_to_msecs(NEIGH_VAR(p, DELAY_PROBE_TIME)); + mlxsw_sp->router.neighs_update.interval = interval; + + mlxsw_sp_port_dev_put(mlxsw_sp_port); + break; + case NETEVENT_NEIGH_UPDATE: + n = ptr; + dev = n->dev; + + if (n->tbl != &arp_tbl) + return NOTIFY_DONE; + + mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(dev); + if (!mlxsw_sp_port) + return NOTIFY_DONE; + + mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + dip = ntohl(*((__be32 *) n->primary_key)); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, + &dip, + sizeof(__be32), + dev); + if (WARN_ON(!neigh_entry) || WARN_ON(neigh_entry->n != n)) { + mlxsw_sp_port_dev_put(mlxsw_sp_port); + return NOTIFY_DONE; + } + neigh_entry->mlxsw_sp_port = mlxsw_sp_port; + + /* Take a reference to ensure the neighbour won't be + * destructed until we drop the reference in delayed + * work. + */ + neigh_clone(n); + if (!mlxsw_core_schedule_dw(&neigh_entry->dw, 0)) { + neigh_release(n); + mlxsw_sp_port_dev_put(mlxsw_sp_port); + } + break; + } + + return NOTIFY_DONE; +} + +static struct notifier_block mlxsw_sp_router_netevent_nb __read_mostly = { + .notifier_call = mlxsw_sp_router_netevent_event, +}; + +static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) +{ + int err; + + err = rhashtable_init(&mlxsw_sp->router.neigh_ht, + &mlxsw_sp_neigh_ht_params); + if (err) + return err; + + /* Initialize the polling interval according to the default + * table. + */ + mlxsw_sp_router_neighs_update_interval_init(mlxsw_sp); + + err = register_netevent_notifier(&mlxsw_sp_router_netevent_nb); + if (err) + goto err_register_netevent_notifier; + + /* Create the delayed works for the activity_update */ + INIT_DELAYED_WORK(&mlxsw_sp->router.neighs_update.dw, + mlxsw_sp_router_neighs_update_work); + INIT_DELAYED_WORK(&mlxsw_sp->router.nexthop_probe_dw, + mlxsw_sp_router_probe_unresolved_nexthops); + mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, 0); + mlxsw_core_schedule_dw(&mlxsw_sp->router.nexthop_probe_dw, 0); + return 0; + +err_register_netevent_notifier: + rhashtable_destroy(&mlxsw_sp->router.neigh_ht); + return err; +} + +static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp) +{ + cancel_delayed_work_sync(&mlxsw_sp->router.neighs_update.dw); + cancel_delayed_work_sync(&mlxsw_sp->router.nexthop_probe_dw); + unregister_netevent_notifier(&mlxsw_sp_router_netevent_nb); + rhashtable_destroy(&mlxsw_sp->router.neigh_ht); +} + +struct mlxsw_sp_nexthop { + struct list_head neigh_list_node; /* member of neigh entry list */ + struct mlxsw_sp_nexthop_group *nh_grp; /* pointer back to the group + * this belongs to + */ + u8 should_offload:1, /* set indicates this neigh is connected and + * should be put to KVD linear area of this group. + */ + offloaded:1, /* set in case the neigh is actually put into + * KVD linear area of this group. + */ + update:1; /* set indicates that MAC of this neigh should be + * updated in HW + */ + struct mlxsw_sp_neigh_entry *neigh_entry; +}; + +struct mlxsw_sp_nexthop_group { + struct list_head list; /* node in mlxsw->router.nexthop_group_list */ + struct list_head fib_list; /* list of fib entries that use this group */ + u8 adj_index_valid:1; + u32 adj_index; + u16 ecmp_size; + u16 count; + struct mlxsw_sp_nexthop nexthops[0]; +}; + +static int mlxsw_sp_adj_index_mass_update_vr(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr, + u32 adj_index, u16 ecmp_size, + u32 new_adj_index, + u16 new_ecmp_size) +{ + char raleu_pl[MLXSW_REG_RALEU_LEN]; + + mlxsw_reg_raleu_pack(raleu_pl, vr->proto, vr->id, + adj_index, ecmp_size, + new_adj_index, new_ecmp_size); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raleu), raleu_pl); +} + +static int mlxsw_sp_adj_index_mass_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + u32 old_adj_index, u16 old_ecmp_size) +{ + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr = NULL; + int err; + + list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) { + if (vr == fib_entry->vr) + continue; + vr = fib_entry->vr; + err = mlxsw_sp_adj_index_mass_update_vr(mlxsw_sp, vr, + old_adj_index, + old_ecmp_size, + nh_grp->adj_index, + nh_grp->ecmp_size); + if (err) + return err; + } + return 0; +} + +static int mlxsw_sp_nexthop_mac_update(struct mlxsw_sp *mlxsw_sp, u32 adj_index, + struct mlxsw_sp_nexthop *nh) +{ + struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry; + char ratr_pl[MLXSW_REG_RATR_LEN]; + + mlxsw_reg_ratr_pack(ratr_pl, MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY, + true, adj_index, neigh_entry->rif); + mlxsw_reg_ratr_eth_entry_pack(ratr_pl, neigh_entry->ha); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ratr), ratr_pl); +} + +static int +mlxsw_sp_nexthop_group_mac_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + u32 adj_index = nh_grp->adj_index; /* base */ + struct mlxsw_sp_nexthop *nh; + int i; + int err; + + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + + if (!nh->should_offload) { + nh->offloaded = 0; + continue; + } + + if (nh->update) { + err = mlxsw_sp_nexthop_mac_update(mlxsw_sp, + adj_index, nh); + if (err) + return err; + nh->update = 0; + nh->offloaded = 1; + } + adj_index++; + } + return 0; +} + +static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry); + +static int +mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_fib_entry *fib_entry; + int err; + + list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) { + err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); + if (err) + return err; + } + return 0; +} + +static void +mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_nexthop *nh; + bool offload_change = false; + u32 adj_index; + u16 ecmp_size = 0; + bool old_adj_index_valid; + u32 old_adj_index; + u16 old_ecmp_size; + int ret; + int i; + int err; + + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + + if (nh->should_offload ^ nh->offloaded) { + offload_change = true; + if (nh->should_offload) + nh->update = 1; + } + if (nh->should_offload) + ecmp_size++; + } + if (!offload_change) { + /* Nothing was added or removed, so no need to reallocate. Just + * update MAC on existing adjacency indexes. + */ + err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n"); + goto set_trap; + } + return; + } + if (!ecmp_size) + /* No neigh of this group is connected so we just set + * the trap and let everthing flow through kernel. + */ + goto set_trap; + + ret = mlxsw_sp_kvdl_alloc(mlxsw_sp, ecmp_size); + if (ret < 0) { + /* We ran out of KVD linear space, just set the + * trap and let everything flow through kernel. + */ + dev_warn(mlxsw_sp->bus_info->dev, "Failed to allocate KVD linear area for nexthop group.\n"); + goto set_trap; + } + adj_index = ret; + old_adj_index_valid = nh_grp->adj_index_valid; + old_adj_index = nh_grp->adj_index; + old_ecmp_size = nh_grp->ecmp_size; + nh_grp->adj_index_valid = 1; + nh_grp->adj_index = adj_index; + nh_grp->ecmp_size = ecmp_size; + err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n"); + goto set_trap; + } + + if (!old_adj_index_valid) { + /* The trap was set for fib entries, so we have to call + * fib entry update to unset it and use adjacency index. + */ + err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to add adjacency index to fib entries.\n"); + goto set_trap; + } + return; + } + + err = mlxsw_sp_adj_index_mass_update(mlxsw_sp, nh_grp, + old_adj_index, old_ecmp_size); + mlxsw_sp_kvdl_free(mlxsw_sp, old_adj_index); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to mass-update adjacency index for nexthop group.\n"); + goto set_trap; + } + return; + +set_trap: + old_adj_index_valid = nh_grp->adj_index_valid; + nh_grp->adj_index_valid = 0; + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + nh->offloaded = 0; + } + err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp); + if (err) + dev_warn(mlxsw_sp->bus_info->dev, "Failed to set traps for fib entries.\n"); + if (old_adj_index_valid) + mlxsw_sp_kvdl_free(mlxsw_sp, nh_grp->adj_index); +} + +static void __mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp_nexthop *nh, + bool removing) +{ + if (!removing && !nh->should_offload) + nh->should_offload = 1; + else if (removing && nh->offloaded) + nh->should_offload = 0; + nh->update = 1; +} + +static void +mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry, + bool removing) +{ + struct mlxsw_sp_nexthop *nh; + + /* Take RTNL mutex here to prevent lists from changes */ + rtnl_lock(); + list_for_each_entry(nh, &neigh_entry->nexthop_list, + neigh_list_node) { + __mlxsw_sp_nexthop_neigh_update(nh, removing); + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nh_grp); + } + rtnl_unlock(); +} + +static int mlxsw_sp_nexthop_init(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + struct mlxsw_sp_nexthop *nh, + struct fib_nh *fib_nh) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + u32 gwip = ntohl(fib_nh->nh_gw); + struct net_device *dev = fib_nh->nh_dev; + struct neighbour *n; + u8 nud_state; + + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &gwip, + sizeof(gwip), dev); + if (!neigh_entry) { + __be32 gwipn = htonl(gwip); + + n = neigh_create(&arp_tbl, &gwipn, dev); + if (IS_ERR(n)) + return PTR_ERR(n); + neigh_event_send(n, NULL); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &gwip, + sizeof(gwip), dev); + if (!neigh_entry) { + neigh_release(n); + return -EINVAL; + } + } else { + /* Take a reference of neigh here ensuring that neigh would + * not be detructed before the nexthop entry is finished. + * The second branch takes the reference in neith_create() + */ + n = neigh_entry->n; + neigh_clone(n); + } + + /* If that is the first nexthop connected to that neigh, add to + * nexthop_neighs_list + */ + if (list_empty(&neigh_entry->nexthop_list)) + list_add_tail(&neigh_entry->nexthop_neighs_list_node, + &mlxsw_sp->router.nexthop_neighs_list); + + nh->nh_grp = nh_grp; + nh->neigh_entry = neigh_entry; + list_add_tail(&nh->neigh_list_node, &neigh_entry->nexthop_list); + read_lock_bh(&n->lock); + nud_state = n->nud_state; + read_unlock_bh(&n->lock); + __mlxsw_sp_nexthop_neigh_update(nh, !(nud_state & NUD_VALID)); + + return 0; +} + +static void mlxsw_sp_nexthop_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry; + + list_del(&nh->neigh_list_node); + + /* If that is the last nexthop connected to that neigh, remove from + * nexthop_neighs_list + */ + if (list_empty(&nh->neigh_entry->nexthop_list)) + list_del(&nh->neigh_entry->nexthop_neighs_list_node); + + neigh_release(neigh_entry->n); +} + +static struct mlxsw_sp_nexthop_group * +mlxsw_sp_nexthop_group_create(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + struct mlxsw_sp_nexthop *nh; + struct fib_nh *fib_nh; + size_t alloc_size; + int i; + int err; + + alloc_size = sizeof(*nh_grp) + + fi->fib_nhs * sizeof(struct mlxsw_sp_nexthop); + nh_grp = kzalloc(alloc_size, GFP_KERNEL); + if (!nh_grp) + return ERR_PTR(-ENOMEM); + INIT_LIST_HEAD(&nh_grp->fib_list); + nh_grp->count = fi->fib_nhs; + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + fib_nh = &fi->fib_nh[i]; + err = mlxsw_sp_nexthop_init(mlxsw_sp, nh_grp, nh, fib_nh); + if (err) + goto err_nexthop_init; + } + list_add_tail(&nh_grp->list, &mlxsw_sp->router.nexthop_group_list); + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp); + return nh_grp; + +err_nexthop_init: + for (i--; i >= 0; i--) + mlxsw_sp_nexthop_fini(mlxsw_sp, nh); + kfree(nh_grp); + return ERR_PTR(err); +} + +static void +mlxsw_sp_nexthop_group_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_nexthop *nh; + int i; + + list_del(&nh_grp->list); + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + mlxsw_sp_nexthop_fini(mlxsw_sp, nh); + } + kfree(nh_grp); +} + +static bool mlxsw_sp_nexthop_match(struct mlxsw_sp_nexthop *nh, + struct fib_info *fi) +{ + int i; + + for (i = 0; i < fi->fib_nhs; i++) { + struct fib_nh *fib_nh = &fi->fib_nh[i]; + u32 gwip = ntohl(fib_nh->nh_gw); + + if (memcmp(nh->neigh_entry->key.addr, + &gwip, sizeof(u32)) == 0 && + nh->neigh_entry->key.dev == fib_nh->nh_dev) + return true; + } + return false; +} + +static bool mlxsw_sp_nexthop_group_match(struct mlxsw_sp_nexthop_group *nh_grp, + struct fib_info *fi) +{ + int i; + + if (nh_grp->count != fi->fib_nhs) + return false; + for (i = 0; i < nh_grp->count; i++) { + struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i]; + + if (!mlxsw_sp_nexthop_match(nh, fi)) + return false; + } + return true; +} + +static struct mlxsw_sp_nexthop_group * +mlxsw_sp_nexthop_group_find(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + + list_for_each_entry(nh_grp, &mlxsw_sp->router.nexthop_group_list, + list) { + if (mlxsw_sp_nexthop_group_match(nh_grp, fi)) + return nh_grp; + } + return NULL; +} + +static int mlxsw_sp_nexthop_group_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + struct fib_info *fi) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + + nh_grp = mlxsw_sp_nexthop_group_find(mlxsw_sp, fi); + if (!nh_grp) { + nh_grp = mlxsw_sp_nexthop_group_create(mlxsw_sp, fi); + if (IS_ERR(nh_grp)) + return PTR_ERR(nh_grp); + } + list_add_tail(&fib_entry->nexthop_group_node, &nh_grp->fib_list); + fib_entry->nh_group = nh_grp; + return 0; +} + +static void mlxsw_sp_nexthop_group_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group; + + list_del(&fib_entry->nexthop_group_node); + if (!list_empty(&nh_grp->fib_list)) + return; + mlxsw_sp_nexthop_group_destroy(mlxsw_sp, nh_grp); +} + +static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) +{ + char rgcr_pl[MLXSW_REG_RGCR_LEN]; + + mlxsw_reg_rgcr_pack(rgcr_pl, true); + mlxsw_reg_rgcr_max_router_interfaces_set(rgcr_pl, MLXSW_SP_RIF_MAX); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); +} + +static void __mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) +{ + char rgcr_pl[MLXSW_REG_RGCR_LEN]; + + mlxsw_reg_rgcr_pack(rgcr_pl, false); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); +} + +int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) +{ + int err; + + INIT_LIST_HEAD(&mlxsw_sp->router.nexthop_neighs_list); + INIT_LIST_HEAD(&mlxsw_sp->router.nexthop_group_list); + err = __mlxsw_sp_router_init(mlxsw_sp); + if (err) + return err; + mlxsw_sp_lpm_init(mlxsw_sp); + mlxsw_sp_vrs_init(mlxsw_sp); + return mlxsw_sp_neigh_init(mlxsw_sp); +} + +void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) +{ + mlxsw_sp_neigh_fini(mlxsw_sp); + __mlxsw_sp_router_fini(mlxsw_sp); +} + +static int mlxsw_sp_fib_entry_op4_remote(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + char ralue_pl[MLXSW_REG_RALUE_LEN]; + u32 *p_dip = (u32 *) fib_entry->key.addr; + struct mlxsw_sp_vr *vr = fib_entry->vr; + enum mlxsw_reg_ralue_trap_action trap_action; + u16 trap_id = 0; + u32 adjacency_index = 0; + u16 ecmp_size = 0; + + /* In case the nexthop group adjacency index is valid, use it + * with provided ECMP size. Otherwise, setup trap and pass + * traffic to kernel. + */ + if (fib_entry->nh_group->adj_index_valid) { + trap_action = MLXSW_REG_RALUE_TRAP_ACTION_NOP; + adjacency_index = fib_entry->nh_group->adj_index; + ecmp_size = fib_entry->nh_group->ecmp_size; + } else { + trap_action = MLXSW_REG_RALUE_TRAP_ACTION_TRAP; + trap_id = MLXSW_TRAP_ID_RTR_INGRESS0; + } + + mlxsw_reg_ralue_pack4(ralue_pl, vr->proto, op, vr->id, + fib_entry->key.prefix_len, *p_dip); + mlxsw_reg_ralue_act_remote_pack(ralue_pl, trap_action, trap_id, + adjacency_index, ecmp_size); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl); +} + +static int mlxsw_sp_fib_entry_op4_local(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + char ralue_pl[MLXSW_REG_RALUE_LEN]; + u32 *p_dip = (u32 *) fib_entry->key.addr; + struct mlxsw_sp_vr *vr = fib_entry->vr; + + mlxsw_reg_ralue_pack4(ralue_pl, vr->proto, op, vr->id, + fib_entry->key.prefix_len, *p_dip); + mlxsw_reg_ralue_act_local_pack(ralue_pl, + MLXSW_REG_RALUE_TRAP_ACTION_NOP, 0, + fib_entry->rif); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl); +} + +static int mlxsw_sp_fib_entry_op4_trap(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + char ralue_pl[MLXSW_REG_RALUE_LEN]; + u32 *p_dip = (u32 *) fib_entry->key.addr; + struct mlxsw_sp_vr *vr = fib_entry->vr; + + mlxsw_reg_ralue_pack4(ralue_pl, vr->proto, op, vr->id, + fib_entry->key.prefix_len, *p_dip); + mlxsw_reg_ralue_act_ip2me_pack(ralue_pl); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl); +} + +static int mlxsw_sp_fib_entry_op4(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + switch (fib_entry->type) { + case MLXSW_SP_FIB_ENTRY_TYPE_REMOTE: + return mlxsw_sp_fib_entry_op4_remote(mlxsw_sp, fib_entry, op); + case MLXSW_SP_FIB_ENTRY_TYPE_LOCAL: + return mlxsw_sp_fib_entry_op4_local(mlxsw_sp, fib_entry, op); + case MLXSW_SP_FIB_ENTRY_TYPE_TRAP: + return mlxsw_sp_fib_entry_op4_trap(mlxsw_sp, fib_entry, op); + } + return -EINVAL; +} + +static int mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + switch (fib_entry->vr->proto) { + case MLXSW_SP_L3_PROTO_IPV4: + return mlxsw_sp_fib_entry_op4(mlxsw_sp, fib_entry, op); + case MLXSW_SP_L3_PROTO_IPV6: + return -EINVAL; + } + return -EINVAL; +} + +static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + enum mlxsw_reg_ralue_op op; + + op = !fib_entry->added ? MLXSW_REG_RALUE_OP_WRITE_WRITE : + MLXSW_REG_RALUE_OP_WRITE_UPDATE; + return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, op); +} + +static int mlxsw_sp_fib_entry_del(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, + MLXSW_REG_RALUE_OP_WRITE_DELETE); +} + +struct mlxsw_sp_router_fib4_add_info { + struct switchdev_trans_item tritem; + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_fib_entry *fib_entry; +}; + +static void mlxsw_sp_router_fib4_add_info_destroy(void const *data) +{ + const struct mlxsw_sp_router_fib4_add_info *info = data; + struct mlxsw_sp_fib_entry *fib_entry = info->fib_entry; + struct mlxsw_sp *mlxsw_sp = info->mlxsw_sp; + + mlxsw_sp_fib_entry_destroy(fib_entry); + mlxsw_sp_vr_put(mlxsw_sp, fib_entry->vr); + kfree(info); +} + +static int +mlxsw_sp_router_fib4_entry_init(struct mlxsw_sp *mlxsw_sp, + const struct switchdev_obj_ipv4_fib *fib4, + struct mlxsw_sp_fib_entry *fib_entry) +{ + struct fib_info *fi = fib4->fi; + + if (fib4->type == RTN_LOCAL || fib4->type == RTN_BROADCAST) { + fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP; + return 0; + } + if (fib4->type != RTN_UNICAST) + return -EINVAL; + + if (fi->fib_scope != RT_SCOPE_UNIVERSE) { + struct mlxsw_sp_rif *r; + + fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_LOCAL; + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, fi->fib_dev); + if (!r) + return -EINVAL; + fib_entry->rif = r->rif; + return 0; + } + fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_REMOTE; + return mlxsw_sp_nexthop_group_get(mlxsw_sp, fib_entry, fi); +} + +static void +mlxsw_sp_router_fib4_entry_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + if (fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_REMOTE) + return; + mlxsw_sp_nexthop_group_put(mlxsw_sp, fib_entry); +} + +static int +mlxsw_sp_router_fib4_add_prepare(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_router_fib4_add_info *info; + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr; + int err; + + vr = mlxsw_sp_vr_get(mlxsw_sp, fib4->dst_len, fib4->tb_id, + MLXSW_SP_L3_PROTO_IPV4); + if (IS_ERR(vr)) + return PTR_ERR(vr); + + fib_entry = mlxsw_sp_fib_entry_create(vr->fib, &fib4->dst, + sizeof(fib4->dst), fib4->dst_len); + if (!fib_entry) { + err = -ENOMEM; + goto err_fib_entry_create; + } + fib_entry->vr = vr; + + err = mlxsw_sp_router_fib4_entry_init(mlxsw_sp, fib4, fib_entry); + if (err) + goto err_fib4_entry_init; + + info = kmalloc(sizeof(*info), GFP_KERNEL); + if (!info) { + err = -ENOMEM; + goto err_alloc_info; + } + info->mlxsw_sp = mlxsw_sp; + info->fib_entry = fib_entry; + switchdev_trans_item_enqueue(trans, info, + mlxsw_sp_router_fib4_add_info_destroy, + &info->tritem); + return 0; + +err_alloc_info: + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); +err_fib4_entry_init: + mlxsw_sp_fib_entry_destroy(fib_entry); +err_fib_entry_create: + mlxsw_sp_vr_put(mlxsw_sp, vr); + return err; +} + +static int +mlxsw_sp_router_fib4_add_commit(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_router_fib4_add_info *info; + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr; + int err; + + info = switchdev_trans_item_dequeue(trans); + fib_entry = info->fib_entry; + kfree(info); + + vr = fib_entry->vr; + err = mlxsw_sp_fib_entry_insert(fib_entry->vr->fib, fib_entry); + if (err) + goto err_fib_entry_insert; + err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); + if (err) + goto err_fib_entry_add; + return 0; + +err_fib_entry_add: + mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); +err_fib_entry_insert: + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); + mlxsw_sp_fib_entry_destroy(fib_entry); + mlxsw_sp_vr_put(mlxsw_sp, vr); + return err; +} + +int mlxsw_sp_router_fib4_add(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans) +{ + if (switchdev_trans_ph_prepare(trans)) + return mlxsw_sp_router_fib4_add_prepare(mlxsw_sp_port, + fib4, trans); + return mlxsw_sp_router_fib4_add_commit(mlxsw_sp_port, + fib4, trans); +} + +int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr; + + vr = mlxsw_sp_vr_find(mlxsw_sp, fib4->tb_id, MLXSW_SP_L3_PROTO_IPV4); + if (!vr) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to find virtual router for FIB4 entry being removed.\n"); + return -ENOENT; + } + fib_entry = mlxsw_sp_fib_entry_lookup(vr->fib, &fib4->dst, + sizeof(fib4->dst), fib4->dst_len); + if (!fib_entry) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to find FIB4 entry being removed.\n"); + return PTR_ERR(vr); + } + mlxsw_sp_fib_entry_del(mlxsw_sp_port->mlxsw_sp, fib_entry); + mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); + mlxsw_sp_fib_entry_destroy(fib_entry); + mlxsw_sp_vr_put(mlxsw_sp, vr); + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 3710f19ed6bb..a1ad5e6bdfa8 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -55,13 +55,10 @@ static u16 mlxsw_sp_port_vid_to_fid_get(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_port); u16 fid = vid; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_port); - - fid = mlxsw_sp_vfid_to_fid(vfid); - } + fid = f ? f->fid : fid; if (!fid) fid = mlxsw_sp_port->pvid; @@ -169,11 +166,6 @@ static int mlxsw_sp_port_attr_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, return mlxsw_sp_port_stp_state_set(mlxsw_sp_port, state); } -static bool mlxsw_sp_vfid_is_vport_br(u16 vfid) -{ - return vfid >= MLXSW_SP_VFID_PORT_MAX; -} - static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 idx_begin, u16 idx_end, bool set, bool only_uc) @@ -185,15 +177,10 @@ static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, char *sftr_pl; int err; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { + if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID; - if (mlxsw_sp_vfid_is_vport_br(idx_begin)) - local_port = mlxsw_sp_port->local_port; - else - local_port = MLXSW_PORT_CPU_PORT; - } else { + else table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST; - } sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL); if (!sftr_pl) @@ -236,7 +223,8 @@ static int mlxsw_sp_port_uc_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, int err; if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_port); + u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port)->fid; + u16 vfid = mlxsw_sp_fid_to_vfid(fid); return __mlxsw_sp_port_flood_set(mlxsw_sp_port, vfid, vfid, set, true); @@ -260,14 +248,17 @@ err_port_flood_set: return err; } -int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid, - bool set, bool only_uc) +int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, + bool set) { + u16 vfid; + /* In case of vFIDs, index into the flooding table is relative to * the start of the vFIDs range. */ + vfid = mlxsw_sp_fid_to_vfid(fid); return __mlxsw_sp_port_flood_set(mlxsw_sp_vport, vfid, vfid, set, - only_uc); + false); } static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port, @@ -383,6 +374,187 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev, return err; } +static int mlxsw_sp_fid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) +{ + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + + mlxsw_reg_sfmr_pack(sfmr_pl, !create, fid, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); +} + +static int mlxsw_sp_fid_map(struct mlxsw_sp *mlxsw_sp, u16 fid, bool valid) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_VID_TO_FID; + char svfa_pl[MLXSW_REG_SVFA_LEN]; + + mlxsw_reg_svfa_pack(svfa_pl, 0, mt, valid, fid, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl); +} + +static struct mlxsw_sp_fid *mlxsw_sp_fid_alloc(u16 fid) +{ + struct mlxsw_sp_fid *f; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + return NULL; + + f->fid = fid; + + return f; +} + +struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid) +{ + struct mlxsw_sp_fid *f; + int err; + + err = mlxsw_sp_fid_op(mlxsw_sp, fid, true); + if (err) + return ERR_PTR(err); + + /* Although all the ports member in the FID might be using a + * {Port, VID} to FID mapping, we create a global VID-to-FID + * mapping. This allows a port to transition to VLAN mode, + * knowing the global mapping exists. + */ + err = mlxsw_sp_fid_map(mlxsw_sp, fid, true); + if (err) + goto err_fid_map; + + f = mlxsw_sp_fid_alloc(fid); + if (!f) { + err = -ENOMEM; + goto err_allocate_fid; + } + + list_add(&f->list, &mlxsw_sp->fids); + + return f; + +err_allocate_fid: + mlxsw_sp_fid_map(mlxsw_sp, fid, false); +err_fid_map: + mlxsw_sp_fid_op(mlxsw_sp, fid, false); + return ERR_PTR(err); +} + +void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f) +{ + u16 fid = f->fid; + + list_del(&f->list); + + if (f->r) + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); + + kfree(f); + + mlxsw_sp_fid_op(mlxsw_sp, fid, false); +} + +static int __mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + f = mlxsw_sp_fid_find(mlxsw_sp_port->mlxsw_sp, fid); + if (!f) { + f = mlxsw_sp_fid_create(mlxsw_sp_port->mlxsw_sp, fid); + if (IS_ERR(f)) + return PTR_ERR(f); + } + + f->ref_count++; + + netdev_dbg(mlxsw_sp_port->dev, "Joined FID=%d\n", fid); + + return 0; +} + +static void __mlxsw_sp_port_fid_leave(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + f = mlxsw_sp_fid_find(mlxsw_sp_port->mlxsw_sp, fid); + if (WARN_ON(!f)) + return; + + netdev_dbg(mlxsw_sp_port->dev, "Left FID=%d\n", fid); + + mlxsw_sp_port_fdb_flush(mlxsw_sp_port, fid); + + if (--f->ref_count == 0) + mlxsw_sp_fid_destroy(mlxsw_sp_port->mlxsw_sp, f); +} + +static int mlxsw_sp_port_fid_map(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid, + bool valid) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + + /* If port doesn't have vPorts, then it can use the global + * VID-to-FID mapping. + */ + if (list_empty(&mlxsw_sp_port->vports_list)) + return 0; + + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, valid, fid, fid); +} + +static int mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid_begin, u16 fid_end) +{ + int fid, err; + + for (fid = fid_begin; fid <= fid_end; fid++) { + err = __mlxsw_sp_port_fid_join(mlxsw_sp_port, fid); + if (err) + goto err_port_fid_join; + } + + err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, + true, false); + if (err) + goto err_port_flood_set; + + for (fid = fid_begin; fid <= fid_end; fid++) { + err = mlxsw_sp_port_fid_map(mlxsw_sp_port, fid, true); + if (err) + goto err_port_fid_map; + } + + return 0; + +err_port_fid_map: + for (fid--; fid >= fid_begin; fid--) + mlxsw_sp_port_fid_map(mlxsw_sp_port, fid, false); + __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, false, + false); +err_port_flood_set: + fid = fid_end; +err_port_fid_join: + for (fid--; fid >= fid_begin; fid--) + __mlxsw_sp_port_fid_leave(mlxsw_sp_port, fid); + return err; +} + +static void mlxsw_sp_port_fid_leave(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid_begin, u16 fid_end) +{ + int fid; + + for (fid = fid_begin; fid <= fid_end; fid++) + mlxsw_sp_port_fid_map(mlxsw_sp_port, fid, false); + + __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, false, + false); + + for (fid = fid_begin; fid <= fid_end; fid++) + __mlxsw_sp_port_fid_leave(mlxsw_sp_port, fid); +} + static int __mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { @@ -440,74 +612,6 @@ err_port_allow_untagged_set: return err; } -static int mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid) -{ - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - int err; - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, fid, fid); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); - - if (err) - return err; - - set_bit(fid, mlxsw_sp->active_fids); - return 0; -} - -static void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, u16 fid) -{ - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - - clear_bit(fid, mlxsw_sp->active_fids); - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, - fid, fid); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); -} - -static int mlxsw_sp_port_fid_map(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) -{ - enum mlxsw_reg_svfa_mt mt; - - if (!list_empty(&mlxsw_sp_port->vports_list)) - mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; - else - mt = MLXSW_REG_SVFA_MT_VID_TO_FID; - - return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, true, fid, fid); -} - -static int mlxsw_sp_port_fid_unmap(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) -{ - enum mlxsw_reg_svfa_mt mt; - - if (list_empty(&mlxsw_sp_port->vports_list)) - return 0; - - mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; - return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false, fid, fid); -} - -static int mlxsw_sp_port_add_vids(struct net_device *dev, u16 vid_begin, - u16 vid_end) -{ - u16 vid; - int err; - - for (vid = vid_begin; vid <= vid_end; vid++) { - err = mlxsw_sp_port_add_vid(dev, 0, vid); - if (err) - goto err_port_add_vid; - } - return 0; - -err_port_add_vid: - for (vid--; vid >= vid_begin; vid--) - mlxsw_sp_port_kill_vid(dev, 0, vid); - return err; -} - static int __mlxsw_sp_port_vlans_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool is_member, bool untagged) @@ -533,57 +637,17 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool flag_untagged, bool flag_pvid) { - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct net_device *dev = mlxsw_sp_port->dev; - u16 vid, last_visited_vid, old_pvid; - enum mlxsw_reg_svfa_mt mt; + u16 vid, old_pvid; int err; - /* In case this is invoked with BRIDGE_FLAGS_SELF and port is - * not bridged, then packets ingressing through the port with - * the specified VIDs will be directed to CPU. - */ if (!mlxsw_sp_port->bridged) - return mlxsw_sp_port_add_vids(dev, vid_begin, vid_end); - - for (vid = vid_begin; vid <= vid_end; vid++) { - if (!test_bit(vid, mlxsw_sp->active_fids)) { - err = mlxsw_sp_fid_create(mlxsw_sp, vid); - if (err) { - netdev_err(dev, "Failed to create FID=%d\n", - vid); - return err; - } - - /* When creating a FID, we set a VID to FID mapping - * regardless of the port's mode. - */ - mt = MLXSW_REG_SVFA_MT_VID_TO_FID; - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, - true, vid, vid); - if (err) { - netdev_err(dev, "Failed to create FID=VID=%d mapping\n", - vid); - goto err_port_vid_to_fid_set; - } - } - } - - /* Set FID mapping according to port's mode */ - for (vid = vid_begin; vid <= vid_end; vid++) { - err = mlxsw_sp_port_fid_map(mlxsw_sp_port, vid); - if (err) { - netdev_err(dev, "Failed to map FID=%d", vid); - last_visited_vid = --vid; - goto err_port_fid_map; - } - } + return -EINVAL; - err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, - true, false); + err = mlxsw_sp_port_fid_join(mlxsw_sp_port, vid_begin, vid_end); if (err) { - netdev_err(dev, "Failed to configure flooding\n"); - goto err_port_flood_set; + netdev_err(dev, "Failed to join FIDs\n"); + return err; } err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, @@ -628,10 +692,6 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, return 0; -err_port_vid_to_fid_set: - mlxsw_sp_fid_destroy(mlxsw_sp, vid); - return err; - err_port_stp_state_set: for (vid = vid_begin; vid <= vid_end; vid++) clear_bit(vid, mlxsw_sp_port->active_vlans); @@ -641,13 +701,7 @@ err_port_pvid_set: __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, false, false); err_port_vlans_set: - __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, false, - false); -err_port_flood_set: - last_visited_vid = vid_end; -err_port_fid_map: - for (vid = last_visited_vid; vid >= vid_begin; vid--) - mlxsw_sp_port_fid_unmap(mlxsw_sp_port, vid); + mlxsw_sp_port_fid_leave(mlxsw_sp_port, vid_begin, vid_end); return err; } @@ -678,9 +732,10 @@ static enum mlxsw_reg_sfd_op mlxsw_sp_sfd_op(bool adding) MLXSW_REG_SFD_OP_WRITE_REMOVE; } -static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, - const char *mac, u16 fid, bool adding, - bool dynamic) +static int __mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, + const char *mac, u16 fid, bool adding, + enum mlxsw_reg_sfd_rec_action action, + bool dynamic) { char *sfd_pl; int err; @@ -691,14 +746,29 @@ static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, mlxsw_reg_sfd_pack(sfd_pl, mlxsw_sp_sfd_op(adding), 0); mlxsw_reg_sfd_uc_pack(sfd_pl, 0, mlxsw_sp_sfd_rec_policy(dynamic), - mac, fid, MLXSW_REG_SFD_REC_ACTION_NOP, - local_port); + mac, fid, action, local_port); err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfd), sfd_pl); kfree(sfd_pl); return err; } +static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, + const char *mac, u16 fid, bool adding, + bool dynamic) +{ + return __mlxsw_sp_port_fdb_uc_op(mlxsw_sp, local_port, mac, fid, adding, + MLXSW_REG_SFD_REC_ACTION_NOP, dynamic); +} + +int mlxsw_sp_rif_fdb_op(struct mlxsw_sp *mlxsw_sp, const char *mac, u16 fid, + bool adding) +{ + return __mlxsw_sp_port_fdb_uc_op(mlxsw_sp, 0, mac, fid, adding, + MLXSW_REG_SFD_REC_ACTION_FORWARD_IP_ROUTER, + false); +} + static int mlxsw_sp_port_fdb_uc_lag_op(struct mlxsw_sp *mlxsw_sp, u16 lag_id, const char *mac, u16 fid, u16 lag_vid, bool adding, bool dynamic) @@ -903,6 +973,11 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev, SWITCHDEV_OBJ_PORT_VLAN(obj), trans); break; + case SWITCHDEV_OBJ_ID_IPV4_FIB: + err = mlxsw_sp_router_fib4_add(mlxsw_sp_port, + SWITCHDEV_OBJ_IPV4_FIB(obj), + trans); + break; case SWITCHDEV_OBJ_ID_PORT_FDB: err = mlxsw_sp_port_fdb_static_add(mlxsw_sp_port, SWITCHDEV_OBJ_PORT_FDB(obj), @@ -921,21 +996,6 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev, return err; } -static int mlxsw_sp_port_kill_vids(struct net_device *dev, u16 vid_begin, - u16 vid_end) -{ - u16 vid; - int err; - - for (vid = vid_begin; vid <= vid_end; vid++) { - err = mlxsw_sp_port_kill_vid(dev, 0, vid); - if (err) - return err; - } - - return 0; -} - static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool init) { @@ -943,12 +1003,8 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid, pvid; int err; - /* In case this is invoked with BRIDGE_FLAGS_SELF and port is - * not bridged, then prevent packets ingressing through the - * port with the specified VIDs from being trapped to CPU. - */ if (!init && !mlxsw_sp_port->bridged) - return mlxsw_sp_port_kill_vids(dev, vid_begin, vid_end); + return -EINVAL; err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, false, false); @@ -970,21 +1026,7 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, } } - err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, - false, false); - if (err) { - netdev_err(dev, "Failed to clear flooding\n"); - return err; - } - - for (vid = vid_begin; vid <= vid_end; vid++) { - /* Remove FID mapping in case of Virtual mode */ - err = mlxsw_sp_port_fid_unmap(mlxsw_sp_port, vid); - if (err) { - netdev_err(dev, "Failed to unmap FID=%d", vid); - return err; - } - } + mlxsw_sp_port_fid_leave(mlxsw_sp_port, vid_begin, vid_end); out: /* Changing activity bits only if HW operation succeded */ @@ -1081,6 +1123,10 @@ static int mlxsw_sp_port_obj_del(struct net_device *dev, err = mlxsw_sp_port_vlans_del(mlxsw_sp_port, SWITCHDEV_OBJ_PORT_VLAN(obj)); break; + case SWITCHDEV_OBJ_ID_IPV4_FIB: + err = mlxsw_sp_router_fib4_del(mlxsw_sp_port, + SWITCHDEV_OBJ_IPV4_FIB(obj)); + break; case SWITCHDEV_OBJ_ID_PORT_FDB: err = mlxsw_sp_port_fdb_static_del(mlxsw_sp_port, SWITCHDEV_OBJ_PORT_FDB(obj)); @@ -1118,7 +1164,8 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *tmp; - u16 vport_fid = 0; + struct mlxsw_sp_fid *f; + u16 vport_fid; char *sfd_pl; char mac[ETH_ALEN]; u16 fid; @@ -1133,12 +1180,8 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, if (!sfd_pl) return -ENOMEM; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 tmp; - - tmp = mlxsw_sp_vport_vfid_get(mlxsw_sp_port); - vport_fid = mlxsw_sp_vfid_to_fid(tmp); - } + f = mlxsw_sp_vport_fid_get(mlxsw_sp_port); + vport_fid = f ? f->fid : 0; mlxsw_reg_sfd_pack(sfd_pl, MLXSW_REG_SFD_OP_QUERY_DUMP, 0); do { @@ -1310,11 +1353,10 @@ static void mlxsw_sp_fdb_notify_mac_process(struct mlxsw_sp *mlxsw_sp, } if (mlxsw_sp_fid_is_vfid(fid)) { - u16 vfid = mlxsw_sp_fid_to_vfid(fid); struct mlxsw_sp_port *mlxsw_sp_vport; - mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_vfid(mlxsw_sp_port, - vfid); + mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_fid(mlxsw_sp_port, + fid); if (!mlxsw_sp_vport) { netdev_err(mlxsw_sp_port->dev, "Failed to find a matching vPort following FDB notification\n"); goto just_remove; @@ -1370,11 +1412,10 @@ static void mlxsw_sp_fdb_notify_mac_lag_process(struct mlxsw_sp *mlxsw_sp, } if (mlxsw_sp_fid_is_vfid(fid)) { - u16 vfid = mlxsw_sp_fid_to_vfid(fid); struct mlxsw_sp_port *mlxsw_sp_vport; - mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_vfid(mlxsw_sp_port, - vfid); + mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_fid(mlxsw_sp_port, + fid); if (!mlxsw_sp_vport) { netdev_err(mlxsw_sp_port->dev, "Failed to find a matching vPort following FDB notification\n"); goto just_remove; @@ -1495,14 +1536,6 @@ static void mlxsw_sp_fdb_fini(struct mlxsw_sp *mlxsw_sp) cancel_delayed_work_sync(&mlxsw_sp->fdb_notify.dw); } -static void mlxsw_sp_fids_fini(struct mlxsw_sp *mlxsw_sp) -{ - u16 fid; - - for_each_set_bit(fid, mlxsw_sp->active_fids, VLAN_N_VID) - mlxsw_sp_fid_destroy(mlxsw_sp, fid); -} - int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp) { return mlxsw_sp_fdb_init(mlxsw_sp); @@ -1511,7 +1544,6 @@ int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp) void mlxsw_sp_switchdev_fini(struct mlxsw_sp *mlxsw_sp) { mlxsw_sp_fdb_fini(mlxsw_sp); - mlxsw_sp_fids_fini(mlxsw_sp); } int mlxsw_sp_port_vlan_init(struct mlxsw_sp_port *mlxsw_sp_port) diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c index 3842eab9449a..25f658b3849a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c +++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c @@ -316,7 +316,10 @@ static netdev_tx_t mlxsw_sx_port_xmit(struct sk_buff *skb, } } mlxsw_sx_txhdr_construct(skb, &tx_info); - len = skb->len; + /* TX header is consumed by HW on the way so we shouldn't count its + * bytes as being sent. + */ + len = skb->len - MLXSW_TXHDR_LEN; /* Due to a race we might fail here because of a full queue. In that * unlikely case we simply drop the packet. */ diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h index 53a9550be75e..470d7696e9fe 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h @@ -54,6 +54,11 @@ enum { MLXSW_TRAP_ID_IGMP_V2_REPORT = 0x32, MLXSW_TRAP_ID_IGMP_V2_LEAVE = 0x33, MLXSW_TRAP_ID_IGMP_V3_REPORT = 0x34, + MLXSW_TRAP_ID_ARPBC = 0x50, + MLXSW_TRAP_ID_ARPUC = 0x51, + MLXSW_TRAP_ID_IP2ME = 0x5F, + MLXSW_TRAP_ID_RTR_INGRESS0 = 0x70, + MLXSW_TRAP_ID_HOST_MISS_IPV4 = 0x90, MLXSW_TRAP_ID_MAX = 0x1FF }; diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h index e744acc18ef4..690635660195 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net.h +++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h @@ -63,7 +63,7 @@ #define NFP_NET_POLL_TIMEOUT 5 /* Bar allocation */ -#define NFP_NET_CRTL_BAR 0 +#define NFP_NET_CTRL_BAR 0 #define NFP_NET_Q0_BAR 2 #define NFP_NET_Q1_BAR 4 /* OBSOLETE */ diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index fa47c14c743a..1e74b911accb 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1845,13 +1845,14 @@ void nfp_net_coalesce_write_cfg(struct nfp_net *nn) } /** - * nfp_net_write_mac_addr() - Write mac address to device registers + * nfp_net_write_mac_addr() - Write mac address to the device control BAR * @nn: NFP Net device to reconfigure - * @mac: Six-byte MAC address to be written * - * We do a bit of byte swapping dance because firmware is LE. + * Writes the MAC address from the netdev to the device control BAR. Does not + * perform the required reconfig. We do a bit of byte swapping dance because + * firmware is LE. */ -static void nfp_net_write_mac_addr(struct nfp_net *nn, const u8 *mac) +static void nfp_net_write_mac_addr(struct nfp_net *nn) { nn_writel(nn, NFP_NET_CFG_MACADDR + 0, get_unaligned_be32(nn->netdev->dev_addr)); @@ -1952,7 +1953,7 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn) nn_writeq(nn, NFP_NET_CFG_RXRS_ENABLE, nn->num_rx_rings == 64 ? 0xffffffffffffffffULL : ((u64)1 << nn->num_rx_rings) - 1); - nfp_net_write_mac_addr(nn, nn->netdev->dev_addr); + nfp_net_write_mac_addr(nn); nn_writel(nn, NFP_NET_CFG_MTU, nn->netdev->mtu); nn_writel(nn, NFP_NET_CFG_FLBUFSZ, nn->fl_bufsz); @@ -1979,7 +1980,7 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn) if (nn->ctrl & NFP_NET_CFG_CTRL_VXLAN) { memset(&nn->vxlan_ports, 0, sizeof(nn->vxlan_ports)); memset(&nn->vxlan_usecnt, 0, sizeof(nn->vxlan_usecnt)); - vxlan_get_rx_port(nn->netdev); + udp_tunnel_get_rx_info(nn->netdev); } return err; @@ -2015,7 +2016,7 @@ static void nfp_net_open_stack(struct nfp_net *nn) netif_tx_wake_all_queues(nn->netdev); - enable_irq(nn->irq_entries[NFP_NET_CFG_LSC].vector); + enable_irq(nn->irq_entries[NFP_NET_IRQ_LSC_IDX].vector); nfp_net_read_link_status(nn); } @@ -2044,7 +2045,7 @@ static int nfp_net_netdev_open(struct net_device *netdev) NFP_NET_IRQ_LSC_IDX, nn->lsc_handler); if (err) goto err_free_exn; - disable_irq(nn->irq_entries[NFP_NET_CFG_LSC].vector); + disable_irq(nn->irq_entries[NFP_NET_IRQ_LSC_IDX].vector); nn->rx_rings = kcalloc(nn->num_rx_rings, sizeof(*nn->rx_rings), GFP_KERNEL); @@ -2133,7 +2134,7 @@ static void nfp_net_close_stack(struct nfp_net *nn) { unsigned int r; - disable_irq(nn->irq_entries[NFP_NET_CFG_LSC].vector); + disable_irq(nn->irq_entries[NFP_NET_IRQ_LSC_IDX].vector); netif_carrier_off(nn->netdev); nn->link_up = false; @@ -2551,26 +2552,32 @@ static int nfp_net_find_vxlan_idx(struct nfp_net *nn, __be16 port) } static void nfp_net_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct nfp_net *nn = netdev_priv(netdev); int idx; - idx = nfp_net_find_vxlan_idx(nn, port); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + idx = nfp_net_find_vxlan_idx(nn, ti->port); if (idx == -ENOSPC) return; if (!nn->vxlan_usecnt[idx]++) - nfp_net_set_vxlan_port(nn, idx, port); + nfp_net_set_vxlan_port(nn, idx, ti->port); } static void nfp_net_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct nfp_net *nn = netdev_priv(netdev); int idx; - idx = nfp_net_find_vxlan_idx(nn, port); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + idx = nfp_net_find_vxlan_idx(nn, ti->port); if (!nn->vxlan_usecnt[idx] || idx == -ENOSPC) return; @@ -2589,8 +2596,8 @@ static const struct net_device_ops nfp_net_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_set_features = nfp_net_set_features, .ndo_features_check = nfp_net_features_check, - .ndo_add_vxlan_port = nfp_net_add_vxlan_port, - .ndo_del_vxlan_port = nfp_net_del_vxlan_port, + .ndo_udp_tunnel_add = nfp_net_add_vxlan_port, + .ndo_udp_tunnel_del = nfp_net_del_vxlan_port, }; /** @@ -2733,7 +2740,7 @@ int nfp_net_netdev_init(struct net_device *netdev) nn->cap = nn_readl(nn, NFP_NET_CFG_CAP); nn->max_mtu = nn_readl(nn, NFP_NET_CFG_MAX_MTU); - nfp_net_write_mac_addr(nn, nn->netdev->dev_addr); + nfp_net_write_mac_addr(nn); /* Set default MTU and Freelist buffer size */ if (nn->max_mtu < NFP_NET_DEFAULT_MTU) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c index ccfef1f17627..7d7933d00b8f 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c @@ -605,6 +605,7 @@ static int nfp_net_set_coalesce(struct net_device *netdev, static const struct ethtool_ops nfp_net_ethtool_ops = { .get_drvinfo = nfp_net_get_drvinfo, + .get_link = ethtool_op_get_link, .get_ringparam = nfp_net_get_ringparam, .set_ringparam = nfp_net_set_ringparam, .get_strings = nfp_net_get_strings, diff --git a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c index e2b22b8a20f1..37abef016a0a 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c @@ -124,11 +124,11 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev, * first NFP_NET_CFG_BAR_SZ of the BAR. This keeps the code * the identical for PF and VF drivers. */ - ctrl_bar = ioremap_nocache(pci_resource_start(pdev, NFP_NET_CRTL_BAR), + ctrl_bar = ioremap_nocache(pci_resource_start(pdev, NFP_NET_CTRL_BAR), NFP_NET_CFG_BAR_SZ); if (!ctrl_bar) { dev_err(&pdev->dev, - "Failed to map resource %d\n", NFP_NET_CRTL_BAR); + "Failed to map resource %d\n", NFP_NET_CTRL_BAR); err = -EIO; goto err_pci_regions; } diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index b1ce7aaa8f8b..4d4ecba0aad9 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -425,7 +425,6 @@ struct netdata_local { unsigned int last_tx_idx; unsigned int num_used_tx_buffs; struct mii_bus *mii_bus; - struct phy_device *phy_dev; struct clk *clk; dma_addr_t dma_buff_base_p; void *dma_buff_base_v; @@ -750,7 +749,7 @@ static int lpc_mdio_reset(struct mii_bus *bus) static void lpc_handle_link_change(struct net_device *ndev) { struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; + struct phy_device *phydev = ndev->phydev; unsigned long flags; bool status_change = false; @@ -814,7 +813,6 @@ static int lpc_mii_probe(struct net_device *ndev) pldat->link = 0; pldat->speed = 0; pldat->duplex = -1; - pldat->phy_dev = phydev; phy_attached_info(phydev); @@ -1048,8 +1046,8 @@ static int lpc_eth_close(struct net_device *ndev) napi_disable(&pldat->napi); netif_stop_queue(ndev); - if (pldat->phy_dev) - phy_stop(pldat->phy_dev); + if (ndev->phydev) + phy_stop(ndev->phydev); spin_lock_irqsave(&pldat->lock, flags); __lpc_eth_reset(pldat); @@ -1185,8 +1183,7 @@ static void lpc_eth_set_multicast_list(struct net_device *ndev) static int lpc_eth_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) { - struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!netif_running(ndev)) return -EINVAL; @@ -1207,14 +1204,14 @@ static int lpc_eth_open(struct net_device *ndev) __lpc_eth_clock_enable(pldat, true); /* Suspended PHY makes LPC ethernet core block, so resume now */ - phy_resume(pldat->phy_dev); + phy_resume(ndev->phydev); /* Reset and initialize */ __lpc_eth_reset(pldat); __lpc_eth_init(pldat); /* schedule a link state check */ - phy_start(pldat->phy_dev); + phy_start(ndev->phydev); netif_start_queue(ndev); napi_enable(&pldat->napi); @@ -1247,37 +1244,13 @@ static void lpc_eth_ethtool_setmsglevel(struct net_device *ndev, u32 level) pldat->msg_enable = level; } -static int lpc_eth_ethtool_getsettings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_gset(phydev, cmd); -} - -static int lpc_eth_ethtool_setsettings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_sset(phydev, cmd); -} - static const struct ethtool_ops lpc_eth_ethtool_ops = { .get_drvinfo = lpc_eth_ethtool_getdrvinfo, - .get_settings = lpc_eth_ethtool_getsettings, - .set_settings = lpc_eth_ethtool_setsettings, .get_msglevel = lpc_eth_ethtool_getmsglevel, .set_msglevel = lpc_eth_ethtool_setmsglevel, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops lpc_netdev_ops = { @@ -1460,7 +1433,7 @@ static int lpc_eth_drv_probe(struct platform_device *pdev) netdev_info(ndev, "LPC mac at 0x%08x irq %d\n", res->start, ndev->irq); - phydev = pldat->phy_dev; + phydev = ndev->phydev; device_init_wakeup(&pdev->dev, 1); device_set_wakeup_enable(&pdev->dev, 0); diff --git a/drivers/net/ethernet/qlogic/Kconfig b/drivers/net/ethernet/qlogic/Kconfig index 680d8c736d2b..6ba48406899e 100644 --- a/drivers/net/ethernet/qlogic/Kconfig +++ b/drivers/net/ethernet/qlogic/Kconfig @@ -54,16 +54,6 @@ config QLCNIC_DCB mode of DCB is supported. PG and PFC values are related only to Tx. -config QLCNIC_VXLAN - bool "Virtual eXtensible Local Area Network (VXLAN) offload support" - default n - depends on QLCNIC && VXLAN && !(QLCNIC=y && VXLAN=m) - ---help--- - This enables hardware offload support for VXLAN protocol over QLogic's - 84XX series adapters. - Say Y here if you want to enable hardware offload support for - Virtual eXtensible Local Area Network (VXLAN) in the driver. - config QLCNIC_HWMON bool "QLOGIC QLCNIC 82XX and 83XX family HWMON support" depends on QLCNIC && HWMON && !(QLCNIC=y && HWMON=m) @@ -114,24 +104,4 @@ config QEDE ---help--- This enables the support for ... -config QEDE_VXLAN - bool "Virtual eXtensible Local Area Network support" - default n - depends on QEDE && VXLAN && !(QEDE=y && VXLAN=m) - ---help--- - This enables hardware offload support for VXLAN protocol over - qede module. Say Y here if you want to enable hardware offload - support for Virtual eXtensible Local Area Network (VXLAN) - in the driver. - -config QEDE_GENEVE - bool "Generic Network Virtualization Encapsulation (GENEVE) support" - depends on QEDE && GENEVE && !(QEDE=y && GENEVE=m) - ---help--- - This allows one to create GENEVE virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. GENEVE is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to enable hardware offload support for - Generic Network Virtualization Encapsulation (GENEVE) in the driver. - endif # NET_VENDOR_QLOGIC diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index 9a63df1184f1..35e53771533f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -489,8 +489,8 @@ struct qed_dev { u32 int_mode; enum qed_coalescing_mode int_coalescing_mode; - u8 rx_coalesce_usecs; - u8 tx_coalesce_usecs; + u16 rx_coalesce_usecs; + u16 tx_coalesce_usecs; /* Start Bar offset of first hwfn */ void __iomem *regview; diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index d85b7ba8cb9a..1c35f376143e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -850,7 +850,7 @@ static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn) val = 0; entries[j].next = cpu_to_be64(val); - conn_num -= ent_per_page; + conn_num -= ent_num; } return 0; diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 21ec1c2df2c7..d0dc28f93c0e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -9,6 +9,7 @@ #include <linux/types.h> #include <asm/byteorder.h> #include <linux/bitops.h> +#include <linux/dcbnl.h> #include <linux/errno.h> #include <linux/kernel.h> #include <linux/slab.h> @@ -18,6 +19,9 @@ #include "qed_dcbx.h" #include "qed_hsi.h" #include "qed_sp.h" +#ifdef CONFIG_DCB +#include <linux/qed/qed_eth_if.h> +#endif #define QED_DCBX_MAX_MIB_READ_TRY (100) #define QED_ETH_TYPE_DEFAULT (0) @@ -252,7 +256,7 @@ qed_dcbx_process_tlv(struct qed_hwfn *p_hwfn, if (p_data->arr[type].update) continue; - enable = (type == DCBX_PROTOCOL_ETH) ? false : dcbx_enabled; + enable = !(type == DCBX_PROTOCOL_ETH); qed_dcbx_update_app_info(p_data, p_hwfn, enable, true, priority, tc, type); } @@ -351,6 +355,293 @@ qed_dcbx_copy_mib(struct qed_hwfn *p_hwfn, return rc; } +#ifdef CONFIG_DCB +static void +qed_dcbx_get_priority_info(struct qed_hwfn *p_hwfn, + struct qed_dcbx_app_prio *p_prio, + struct qed_dcbx_results *p_results) +{ + u8 val; + + p_prio->roce = QED_DCBX_INVALID_PRIORITY; + p_prio->roce_v2 = QED_DCBX_INVALID_PRIORITY; + p_prio->iscsi = QED_DCBX_INVALID_PRIORITY; + p_prio->fcoe = QED_DCBX_INVALID_PRIORITY; + + if (p_results->arr[DCBX_PROTOCOL_ROCE].update && + p_results->arr[DCBX_PROTOCOL_ROCE].enable) + p_prio->roce = p_results->arr[DCBX_PROTOCOL_ROCE].priority; + + if (p_results->arr[DCBX_PROTOCOL_ROCE_V2].update && + p_results->arr[DCBX_PROTOCOL_ROCE_V2].enable) { + val = p_results->arr[DCBX_PROTOCOL_ROCE_V2].priority; + p_prio->roce_v2 = val; + } + + if (p_results->arr[DCBX_PROTOCOL_ISCSI].update && + p_results->arr[DCBX_PROTOCOL_ISCSI].enable) + p_prio->iscsi = p_results->arr[DCBX_PROTOCOL_ISCSI].priority; + + if (p_results->arr[DCBX_PROTOCOL_FCOE].update && + p_results->arr[DCBX_PROTOCOL_FCOE].enable) + p_prio->fcoe = p_results->arr[DCBX_PROTOCOL_FCOE].priority; + + if (p_results->arr[DCBX_PROTOCOL_ETH].update && + p_results->arr[DCBX_PROTOCOL_ETH].enable) + p_prio->eth = p_results->arr[DCBX_PROTOCOL_ETH].priority; + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "Priorities: iscsi %d, roce %d, roce v2 %d, fcoe %d, eth %d\n", + p_prio->iscsi, p_prio->roce, p_prio->roce_v2, p_prio->fcoe, + p_prio->eth); +} + +static void +qed_dcbx_get_app_data(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct dcbx_app_priority_entry *p_tbl, + struct qed_dcbx_params *p_params) +{ + struct qed_app_entry *entry; + u8 pri_map; + int i; + + p_params->app_willing = QED_MFW_GET_FIELD(p_app->flags, + DCBX_APP_WILLING); + p_params->app_valid = QED_MFW_GET_FIELD(p_app->flags, DCBX_APP_ENABLED); + p_params->app_error = QED_MFW_GET_FIELD(p_app->flags, DCBX_APP_ERROR); + p_params->num_app_entries = QED_MFW_GET_FIELD(p_app->flags, + DCBX_APP_NUM_ENTRIES); + for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { + entry = &p_params->app_entry[i]; + entry->ethtype = !(QED_MFW_GET_FIELD(p_tbl[i].entry, + DCBX_APP_SF)); + pri_map = QED_MFW_GET_FIELD(p_tbl[i].entry, DCBX_APP_PRI_MAP); + entry->prio = ffs(pri_map) - 1; + entry->proto_id = QED_MFW_GET_FIELD(p_tbl[i].entry, + DCBX_APP_PROTOCOL_ID); + qed_dcbx_get_app_protocol_type(p_hwfn, p_tbl[i].entry, + entry->proto_id, + &entry->proto_type); + } + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "APP params: willing %d, valid %d error = %d\n", + p_params->app_willing, p_params->app_valid, + p_params->app_error); +} + +static void +qed_dcbx_get_pfc_data(struct qed_hwfn *p_hwfn, + u32 pfc, struct qed_dcbx_params *p_params) +{ + u8 pfc_map; + + p_params->pfc.willing = QED_MFW_GET_FIELD(pfc, DCBX_PFC_WILLING); + p_params->pfc.max_tc = QED_MFW_GET_FIELD(pfc, DCBX_PFC_CAPS); + p_params->pfc.enabled = QED_MFW_GET_FIELD(pfc, DCBX_PFC_ENABLED); + pfc_map = QED_MFW_GET_FIELD(pfc, DCBX_PFC_PRI_EN_BITMAP); + p_params->pfc.prio[0] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_0); + p_params->pfc.prio[1] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_1); + p_params->pfc.prio[2] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_2); + p_params->pfc.prio[3] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_3); + p_params->pfc.prio[4] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_4); + p_params->pfc.prio[5] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_5); + p_params->pfc.prio[6] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_6); + p_params->pfc.prio[7] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_7); + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "PFC params: willing %d, pfc_bitmap %d\n", + p_params->pfc.willing, pfc_map); +} + +static void +qed_dcbx_get_ets_data(struct qed_hwfn *p_hwfn, + struct dcbx_ets_feature *p_ets, + struct qed_dcbx_params *p_params) +{ + u32 bw_map[2], tsa_map[2], pri_map; + int i; + + p_params->ets_willing = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_WILLING); + p_params->ets_enabled = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_ENABLED); + p_params->ets_cbs = QED_MFW_GET_FIELD(p_ets->flags, DCBX_ETS_CBS); + p_params->max_ets_tc = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_MAX_TCS); + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "ETS params: willing %d, ets_cbs %d pri_tc_tbl_0 %x max_ets_tc %d\n", + p_params->ets_willing, + p_params->ets_cbs, + p_ets->pri_tc_tbl[0], p_params->max_ets_tc); + + /* 8 bit tsa and bw data corresponding to each of the 8 TC's are + * encoded in a type u32 array of size 2. + */ + bw_map[0] = be32_to_cpu(p_ets->tc_bw_tbl[0]); + bw_map[1] = be32_to_cpu(p_ets->tc_bw_tbl[1]); + tsa_map[0] = be32_to_cpu(p_ets->tc_tsa_tbl[0]); + tsa_map[1] = be32_to_cpu(p_ets->tc_tsa_tbl[1]); + pri_map = be32_to_cpu(p_ets->pri_tc_tbl[0]); + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + p_params->ets_tc_bw_tbl[i] = ((u8 *)bw_map)[i]; + p_params->ets_tc_tsa_tbl[i] = ((u8 *)tsa_map)[i]; + p_params->ets_pri_tc_tbl[i] = QED_DCBX_PRIO2TC(pri_map, i); + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "elem %d bw_tbl %x tsa_tbl %x\n", + i, p_params->ets_tc_bw_tbl[i], + p_params->ets_tc_tsa_tbl[i]); + } +} + +static void +qed_dcbx_get_common_params(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct dcbx_app_priority_entry *p_tbl, + struct dcbx_ets_feature *p_ets, + u32 pfc, struct qed_dcbx_params *p_params) +{ + qed_dcbx_get_app_data(p_hwfn, p_app, p_tbl, p_params); + qed_dcbx_get_ets_data(p_hwfn, p_ets, p_params); + qed_dcbx_get_pfc_data(p_hwfn, pfc, p_params); +} + +static void +qed_dcbx_get_local_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, struct qed_dcbx_get *params) +{ + struct dcbx_features *p_feat; + + p_feat = &p_hwfn->p_dcbx_info->local_admin.features; + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->local.params); + params->local.valid = true; +} + +static void +qed_dcbx_get_remote_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, struct qed_dcbx_get *params) +{ + struct dcbx_features *p_feat; + + p_feat = &p_hwfn->p_dcbx_info->remote.features; + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->remote.params); + params->remote.valid = true; +} + +static void +qed_dcbx_get_operational_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct qed_dcbx_operational_params *p_operational; + struct qed_dcbx_results *p_results; + struct dcbx_features *p_feat; + bool enabled, err; + u32 flags; + bool val; + + flags = p_hwfn->p_dcbx_info->operational.flags; + + /* If DCBx version is non zero, then negotiation + * was successfuly performed + */ + p_operational = ¶ms->operational; + enabled = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) != + DCBX_CONFIG_VERSION_DISABLED); + if (!enabled) { + p_operational->enabled = enabled; + p_operational->valid = false; + return; + } + + p_feat = &p_hwfn->p_dcbx_info->operational.features; + p_results = &p_hwfn->p_dcbx_info->results; + + val = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) == + DCBX_CONFIG_VERSION_IEEE); + p_operational->ieee = val; + val = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) == + DCBX_CONFIG_VERSION_CEE); + p_operational->cee = val; + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, "Version support: ieee %d, cee %d\n", + p_operational->ieee, p_operational->cee); + + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->operational.params); + qed_dcbx_get_priority_info(p_hwfn, &p_operational->app_prio, p_results); + err = QED_MFW_GET_FIELD(p_feat->app.flags, DCBX_APP_ERROR); + p_operational->err = err; + p_operational->enabled = enabled; + p_operational->valid = true; +} + +static void +qed_dcbx_get_local_lldp_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct lldp_config_params_s *p_local; + + p_local = &p_hwfn->p_dcbx_info->lldp_local[LLDP_NEAREST_BRIDGE]; + + memcpy(params->lldp_local.local_chassis_id, p_local->local_chassis_id, + ARRAY_SIZE(p_local->local_chassis_id)); + memcpy(params->lldp_local.local_port_id, p_local->local_port_id, + ARRAY_SIZE(p_local->local_port_id)); +} + +static void +qed_dcbx_get_remote_lldp_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct lldp_status_params_s *p_remote; + + p_remote = &p_hwfn->p_dcbx_info->lldp_remote[LLDP_NEAREST_BRIDGE]; + + memcpy(params->lldp_remote.peer_chassis_id, p_remote->peer_chassis_id, + ARRAY_SIZE(p_remote->peer_chassis_id)); + memcpy(params->lldp_remote.peer_port_id, p_remote->peer_port_id, + ARRAY_SIZE(p_remote->peer_port_id)); +} + +static int +qed_dcbx_get_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + struct qed_dcbx_get *p_params, + enum qed_mib_read_type type) +{ + switch (type) { + case QED_DCBX_REMOTE_MIB: + qed_dcbx_get_remote_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_LOCAL_MIB: + qed_dcbx_get_local_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_OPERATIONAL_MIB: + qed_dcbx_get_operational_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_REMOTE_LLDP_MIB: + qed_dcbx_get_remote_lldp_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_LOCAL_LLDP_MIB: + qed_dcbx_get_local_lldp_params(p_hwfn, p_ptt, p_params); + break; + default: + DP_ERR(p_hwfn, "MIB read err, unknown mib type %d\n", type); + return -EINVAL; + } + + return 0; +} +#endif + static int qed_dcbx_read_local_lldp_mib(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { @@ -561,3 +852,1333 @@ void qed_dcbx_set_pf_update_params(struct qed_dcbx_results *p_src, p_dcb_data = &p_dest->eth_dcb_data; qed_dcbx_update_protocol_data(p_dcb_data, p_src, DCBX_PROTOCOL_ETH); } + +#ifdef CONFIG_DCB +static int qed_dcbx_query_params(struct qed_hwfn *p_hwfn, + struct qed_dcbx_get *p_get, + enum qed_mib_read_type type) +{ + struct qed_ptt *p_ptt; + int rc; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) + return -EBUSY; + + rc = qed_dcbx_read_mib(p_hwfn, p_ptt, type); + if (rc) + goto out; + + rc = qed_dcbx_get_params(p_hwfn, p_ptt, p_get, type); + +out: + qed_ptt_release(p_hwfn, p_ptt); + return rc; +} + +static void +qed_dcbx_set_pfc_data(struct qed_hwfn *p_hwfn, + u32 *pfc, struct qed_dcbx_params *p_params) +{ + u8 pfc_map = 0; + int i; + + if (p_params->pfc.willing) + *pfc |= DCBX_PFC_WILLING_MASK; + else + *pfc &= ~DCBX_PFC_WILLING_MASK; + + if (p_params->pfc.enabled) + *pfc |= DCBX_PFC_ENABLED_MASK; + else + *pfc &= ~DCBX_PFC_ENABLED_MASK; + + *pfc &= ~DCBX_PFC_CAPS_MASK; + *pfc |= (u32)p_params->pfc.max_tc << DCBX_PFC_CAPS_SHIFT; + + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (p_params->pfc.prio[i]) + pfc_map |= BIT(i); + + *pfc |= (pfc_map << DCBX_PFC_PRI_EN_BITMAP_SHIFT); + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, "pfc = 0x%x\n", *pfc); +} + +static void +qed_dcbx_set_ets_data(struct qed_hwfn *p_hwfn, + struct dcbx_ets_feature *p_ets, + struct qed_dcbx_params *p_params) +{ + u8 *bw_map, *tsa_map; + u32 val; + int i; + + if (p_params->ets_willing) + p_ets->flags |= DCBX_ETS_WILLING_MASK; + else + p_ets->flags &= ~DCBX_ETS_WILLING_MASK; + + if (p_params->ets_cbs) + p_ets->flags |= DCBX_ETS_CBS_MASK; + else + p_ets->flags &= ~DCBX_ETS_CBS_MASK; + + if (p_params->ets_enabled) + p_ets->flags |= DCBX_ETS_ENABLED_MASK; + else + p_ets->flags &= ~DCBX_ETS_ENABLED_MASK; + + p_ets->flags &= ~DCBX_ETS_MAX_TCS_MASK; + p_ets->flags |= (u32)p_params->max_ets_tc << DCBX_ETS_MAX_TCS_SHIFT; + + bw_map = (u8 *)&p_ets->tc_bw_tbl[0]; + tsa_map = (u8 *)&p_ets->tc_tsa_tbl[0]; + p_ets->pri_tc_tbl[0] = 0; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + bw_map[i] = p_params->ets_tc_bw_tbl[i]; + tsa_map[i] = p_params->ets_tc_tsa_tbl[i]; + /* Copy the priority value to the corresponding 4 bits in the + * traffic class table. + */ + val = (((u32)p_params->ets_pri_tc_tbl[i]) << ((7 - i) * 4)); + p_ets->pri_tc_tbl[0] |= val; + } + p_ets->pri_tc_tbl[0] = cpu_to_be32(p_ets->pri_tc_tbl[0]); + for (i = 0; i < 2; i++) { + p_ets->tc_bw_tbl[i] = cpu_to_be32(p_ets->tc_bw_tbl[i]); + p_ets->tc_tsa_tbl[i] = cpu_to_be32(p_ets->tc_tsa_tbl[i]); + } +} + +static void +qed_dcbx_set_app_data(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct qed_dcbx_params *p_params) +{ + u32 *entry; + int i; + + if (p_params->app_willing) + p_app->flags |= DCBX_APP_WILLING_MASK; + else + p_app->flags &= ~DCBX_APP_WILLING_MASK; + + if (p_params->app_valid) + p_app->flags |= DCBX_APP_ENABLED_MASK; + else + p_app->flags &= ~DCBX_APP_ENABLED_MASK; + + p_app->flags &= ~DCBX_APP_NUM_ENTRIES_MASK; + p_app->flags |= (u32)p_params->num_app_entries << + DCBX_APP_NUM_ENTRIES_SHIFT; + + for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { + entry = &p_app->app_pri_tbl[i].entry; + *entry &= ~DCBX_APP_SF_MASK; + if (p_params->app_entry[i].ethtype) + *entry |= ((u32)DCBX_APP_SF_ETHTYPE << + DCBX_APP_SF_SHIFT); + else + *entry |= ((u32)DCBX_APP_SF_PORT << DCBX_APP_SF_SHIFT); + *entry &= ~DCBX_APP_PROTOCOL_ID_MASK; + *entry |= ((u32)p_params->app_entry[i].proto_id << + DCBX_APP_PROTOCOL_ID_SHIFT); + *entry &= ~DCBX_APP_PRI_MAP_MASK; + *entry |= ((u32)(p_params->app_entry[i].prio) << + DCBX_APP_PRI_MAP_SHIFT); + } +} + +static void +qed_dcbx_set_local_params(struct qed_hwfn *p_hwfn, + struct dcbx_local_params *local_admin, + struct qed_dcbx_set *params) +{ + local_admin->flags = 0; + memcpy(&local_admin->features, + &p_hwfn->p_dcbx_info->operational.features, + sizeof(local_admin->features)); + + if (params->enabled) + local_admin->config = params->ver_num; + else + local_admin->config = DCBX_CONFIG_VERSION_DISABLED; + + if (params->override_flags & QED_DCBX_OVERRIDE_PFC_CFG) + qed_dcbx_set_pfc_data(p_hwfn, &local_admin->features.pfc, + ¶ms->config.params); + + if (params->override_flags & QED_DCBX_OVERRIDE_ETS_CFG) + qed_dcbx_set_ets_data(p_hwfn, &local_admin->features.ets, + ¶ms->config.params); + + if (params->override_flags & QED_DCBX_OVERRIDE_APP_CFG) + qed_dcbx_set_app_data(p_hwfn, &local_admin->features.app, + ¶ms->config.params); +} + +int qed_dcbx_config_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + struct qed_dcbx_set *params, bool hw_commit) +{ + struct dcbx_local_params local_admin; + struct qed_dcbx_mib_meta_data data; + u32 resp = 0, param = 0; + int rc = 0; + + if (!hw_commit) { + memcpy(&p_hwfn->p_dcbx_info->set, params, + sizeof(struct qed_dcbx_set)); + return 0; + } + + /* clear set-parmas cache */ + memset(&p_hwfn->p_dcbx_info->set, 0, sizeof(p_hwfn->p_dcbx_info->set)); + + memset(&local_admin, 0, sizeof(local_admin)); + qed_dcbx_set_local_params(p_hwfn, &local_admin, params); + + data.addr = p_hwfn->mcp_info->port_addr + + offsetof(struct public_port, local_admin_dcbx_mib); + data.local_admin = &local_admin; + data.size = sizeof(struct dcbx_local_params); + qed_memcpy_to(p_hwfn, p_ptt, data.addr, data.local_admin, data.size); + + rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_SET_DCBX, + 1 << DRV_MB_PARAM_LLDP_SEND_SHIFT, &resp, ¶m); + if (rc) + DP_NOTICE(p_hwfn, "Failed to send DCBX update request\n"); + + return rc; +} + +int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, + struct qed_dcbx_set *params) +{ + struct qed_dcbx_get *dcbx_info; + int rc; + + if (p_hwfn->p_dcbx_info->set.config.valid) { + memcpy(params, &p_hwfn->p_dcbx_info->set, + sizeof(struct qed_dcbx_set)); + return 0; + } + + dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + if (!dcbx_info) { + DP_ERR(p_hwfn, "Failed to allocate struct qed_dcbx_info\n"); + return -ENOMEM; + } + + rc = qed_dcbx_query_params(p_hwfn, dcbx_info, QED_DCBX_OPERATIONAL_MIB); + if (rc) { + kfree(dcbx_info); + return rc; + } + + p_hwfn->p_dcbx_info->set.override_flags = 0; + p_hwfn->p_dcbx_info->set.ver_num = DCBX_CONFIG_VERSION_DISABLED; + if (dcbx_info->operational.cee) + p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_CEE; + if (dcbx_info->operational.ieee) + p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_IEEE; + + p_hwfn->p_dcbx_info->set.enabled = dcbx_info->operational.enabled; + memcpy(&p_hwfn->p_dcbx_info->set.config.params, + &dcbx_info->operational.params, + sizeof(struct qed_dcbx_admin_params)); + p_hwfn->p_dcbx_info->set.config.valid = true; + + memcpy(params, &p_hwfn->p_dcbx_info->set, sizeof(struct qed_dcbx_set)); + + kfree(dcbx_info); + + return 0; +} + +static struct qed_dcbx_get *qed_dcbnl_get_dcbx(struct qed_hwfn *hwfn, + enum qed_mib_read_type type) +{ + struct qed_dcbx_get *dcbx_info; + + dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + if (!dcbx_info) { + DP_ERR(hwfn->cdev, "Failed to allocate memory for dcbx_info\n"); + return NULL; + } + + if (qed_dcbx_query_params(hwfn, dcbx_info, type)) { + kfree(dcbx_info); + return NULL; + } + + if ((type == QED_DCBX_OPERATIONAL_MIB) && + !dcbx_info->operational.enabled) { + DP_INFO(hwfn, "DCBX is not enabled/operational\n"); + kfree(dcbx_info); + return NULL; + } + + return dcbx_info; +} + +static u8 qed_dcbnl_getstate(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + bool enabled; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + enabled = dcbx_info->operational.enabled; + DP_VERBOSE(hwfn, QED_MSG_DCB, "DCB state = %d\n", enabled); + kfree(dcbx_info); + + return enabled; +} + +static u8 qed_dcbnl_setstate(struct qed_dev *cdev, u8 state) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "DCB state = %d\n", state); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + dcbx_set.enabled = !!state; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc ? 1 : 0; +} + +static void qed_dcbnl_getpgtccfgtx(struct qed_dev *cdev, int tc, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tc = %d\n", tc); + *prio_type = *pgid = *bw_pct = *up_map = 0; + if (tc < 0 || tc >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid tc %d\n", tc); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *pgid = dcbx_info->operational.params.ets_pri_tc_tbl[tc]; + kfree(dcbx_info); +} + +static void qed_dcbnl_getpgbwgcfgtx(struct qed_dev *cdev, int pgid, u8 *bw_pct) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + *bw_pct = 0; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pgid = %d\n", pgid); + if (pgid < 0 || pgid >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid pgid %d\n", pgid); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *bw_pct = dcbx_info->operational.params.ets_tc_bw_tbl[pgid]; + DP_VERBOSE(hwfn, QED_MSG_DCB, "bw_pct = %d\n", *bw_pct); + kfree(dcbx_info); +} + +static void qed_dcbnl_getpgtccfgrx(struct qed_dev *cdev, int tc, u8 *prio, + u8 *bwg_id, u8 *bw_pct, u8 *up_map) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); + *prio = *bwg_id = *bw_pct = *up_map = 0; +} + +static void qed_dcbnl_getpgbwgcfgrx(struct qed_dev *cdev, + int bwg_id, u8 *bw_pct) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); + *bw_pct = 0; +} + +static void qed_dcbnl_getpfccfg(struct qed_dev *cdev, + int priority, u8 *setting) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "priority = %d\n", priority); + if (priority < 0 || priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", priority); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *setting = dcbx_info->operational.params.pfc.prio[priority]; + DP_VERBOSE(hwfn, QED_MSG_DCB, "setting = %d\n", *setting); + kfree(dcbx_info); +} + +static void qed_dcbnl_setpfccfg(struct qed_dev *cdev, int priority, u8 setting) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "priority = %d setting = %d\n", + priority, setting); + if (priority < 0 || priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", priority); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.prio[priority] = !!setting; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static u8 qed_dcbnl_getcap(struct qed_dev *cdev, int capid, u8 *cap) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int rc = 0; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "capid = %d\n", capid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 1; + + switch (capid) { + case DCB_CAP_ATTR_PG: + case DCB_CAP_ATTR_PFC: + case DCB_CAP_ATTR_UP2TC: + case DCB_CAP_ATTR_GSP: + *cap = true; + break; + case DCB_CAP_ATTR_PG_TCS: + case DCB_CAP_ATTR_PFC_TCS: + *cap = 0x80; + break; + case DCB_CAP_ATTR_DCBX: + *cap = (DCB_CAP_DCBX_LLD_MANAGED | DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_VER_IEEE); + break; + default: + *cap = false; + rc = 1; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "id = %d caps = %d\n", capid, *cap); + kfree(dcbx_info); + + return rc; +} + +static int qed_dcbnl_getnumtcs(struct qed_dev *cdev, int tcid, u8 *num) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int rc = 0; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tcid = %d\n", tcid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + switch (tcid) { + case DCB_NUMTCS_ATTR_PG: + *num = dcbx_info->operational.params.max_ets_tc; + break; + case DCB_NUMTCS_ATTR_PFC: + *num = dcbx_info->operational.params.pfc.max_tc; + break; + default: + rc = -EINVAL; + } + + kfree(dcbx_info); + DP_VERBOSE(hwfn, QED_MSG_DCB, "numtcs = %d\n", *num); + + return rc; +} + +static u8 qed_dcbnl_getpfcstate(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + bool enabled; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + enabled = dcbx_info->operational.params.pfc.enabled; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pfc state = %d\n", enabled); + kfree(dcbx_info); + + return enabled; +} + +static u8 qed_dcbnl_getdcbx(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + u8 mode = 0; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + if (dcbx_info->operational.enabled) + mode |= DCB_CAP_DCBX_LLD_MANAGED; + if (dcbx_info->operational.ieee) + mode |= DCB_CAP_DCBX_VER_IEEE; + if (dcbx_info->operational.cee) + mode |= DCB_CAP_DCBX_VER_CEE; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "dcb mode = %d\n", mode); + kfree(dcbx_info); + + return mode; +} + +static void qed_dcbnl_setpgtccfgtx(struct qed_dev *cdev, + int tc, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, + "tc = %d pri_type = %d pgid = %d bw_pct = %d up_map = %d\n", + tc, pri_type, pgid, bw_pct, up_map); + + if (tc < 0 || tc >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid tc %d\n", tc); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_pri_tc_tbl[tc] = pgid; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static void qed_dcbnl_setpgtccfgrx(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); +} + +static void qed_dcbnl_setpgbwgcfgtx(struct qed_dev *cdev, int pgid, u8 bw_pct) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "pgid = %d bw_pct = %d\n", pgid, bw_pct); + if (pgid < 0 || pgid >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid pgid %d\n", pgid); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_tc_bw_tbl[pgid] = bw_pct; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static void qed_dcbnl_setpgbwgcfgrx(struct qed_dev *cdev, int pgid, u8 bw_pct) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); +} + +static u8 qed_dcbnl_setall(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 1); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static int qed_dcbnl_setnumtcs(struct qed_dev *cdev, int tcid, u8 num) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tcid = %d num = %d\n", tcid, num); + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + switch (tcid) { + case DCB_NUMTCS_ATTR_PG: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.max_ets_tc = num; + break; + case DCB_NUMTCS_ATTR_PFC: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.max_tc = num; + break; + default: + DP_INFO(hwfn, "Invalid tcid %d\n", tcid); + return -EINVAL; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static void qed_dcbnl_setpfcstate(struct qed_dev *cdev, u8 state) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "new state = %d\n", state); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.enabled = !!state; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static int qed_dcbnl_getapp(struct qed_dev *cdev, u8 idtype, u16 idval) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_app_entry *entry; + bool ethtype; + u8 prio = 0; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + ethtype = !!(idtype == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_info->operational.params.app_entry[i]; + if ((entry->ethtype == ethtype) && (entry->proto_id == idval)) { + prio = entry->prio; + break; + } + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App entry (%d, %d) not found\n", idtype, idval); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + return prio; +} + +static int qed_dcbnl_setapp(struct qed_dev *cdev, + u8 idtype, u16 idval, u8 pri_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_app_entry *entry; + struct qed_ptt *ptt; + bool ethtype; + int rc, i; + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + ethtype = !!(idtype == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_set.config.params.app_entry[i]; + if ((entry->ethtype == ethtype) && (entry->proto_id == idval)) + break; + /* First empty slot */ + if (!entry->proto_id) + break; + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App table is full\n"); + return -EBUSY; + } + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_entry[i].ethtype = ethtype; + dcbx_set.config.params.app_entry[i].proto_id = idval; + dcbx_set.config.params.app_entry[i].prio = pri_map; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EBUSY; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static u8 qed_dcbnl_setdcbx(struct qed_dev *cdev, u8 mode) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "new mode = %x\n", mode); + + if (!(mode & DCB_CAP_DCBX_VER_IEEE) && !(mode & DCB_CAP_DCBX_VER_CEE)) { + DP_INFO(hwfn, "Allowed mode is cee, ieee or both\n"); + return 1; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + dcbx_set.ver_num = 0; + if (mode & DCB_CAP_DCBX_VER_CEE) { + dcbx_set.ver_num |= DCBX_CONFIG_VERSION_CEE; + dcbx_set.enabled = true; + } + + if (mode & DCB_CAP_DCBX_VER_IEEE) { + dcbx_set.ver_num |= DCBX_CONFIG_VERSION_IEEE; + dcbx_set.enabled = true; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static u8 qed_dcbnl_getfeatcfg(struct qed_dev *cdev, int featid, u8 *flags) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "Feature id = %d\n", featid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 1; + + *flags = 0; + switch (featid) { + case DCB_FEATCFG_ATTR_PG: + if (dcbx_info->operational.params.ets_enabled) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + case DCB_FEATCFG_ATTR_PFC: + if (dcbx_info->operational.params.pfc.enabled) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + case DCB_FEATCFG_ATTR_APP: + if (dcbx_info->operational.params.app_valid) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + default: + DP_INFO(hwfn, "Invalid feature-ID %d\n", featid); + kfree(dcbx_info); + return 1; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "flags = %d\n", *flags); + kfree(dcbx_info); + + return 0; +} + +static u8 qed_dcbnl_setfeatcfg(struct qed_dev *cdev, int featid, u8 flags) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + bool enabled, willing; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "featid = %d flags = %d\n", + featid, flags); + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + enabled = !!(flags & DCB_FEATCFG_ENABLE); + willing = !!(flags & DCB_FEATCFG_WILLING); + switch (featid) { + case DCB_FEATCFG_ATTR_PG: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_enabled = enabled; + dcbx_set.config.params.ets_willing = willing; + break; + case DCB_FEATCFG_ATTR_PFC: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.enabled = enabled; + dcbx_set.config.params.pfc.willing = willing; + break; + case DCB_FEATCFG_ATTR_APP: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_willing = willing; + break; + default: + DP_INFO(hwfn, "Invalid feature-ID %d\n", featid); + return 1; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static int qed_dcbnl_peer_getappinfo(struct qed_dev *cdev, + struct dcb_peer_app_info *info, + u16 *app_count) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + info->willing = dcbx_info->remote.params.app_willing; + info->error = dcbx_info->remote.params.app_error; + *app_count = dcbx_info->remote.params.num_app_entries; + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_peer_getapptable(struct qed_dev *cdev, + struct dcb_app *table) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + for (i = 0; i < dcbx_info->remote.params.num_app_entries; i++) { + if (dcbx_info->remote.params.app_entry[i].ethtype) + table[i].selector = DCB_APP_IDTYPE_ETHTYPE; + else + table[i].selector = DCB_APP_IDTYPE_PORTNUM; + table[i].priority = dcbx_info->remote.params.app_entry[i].prio; + table[i].protocol = + dcbx_info->remote.params.app_entry[i].proto_id; + } + + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_cee_peer_getpfc(struct qed_dev *cdev, struct cee_pfc *pfc) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (dcbx_info->remote.params.pfc.prio[i]) + pfc->pfc_en |= BIT(i); + + pfc->tcs_supported = dcbx_info->remote.params.pfc.max_tc; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pfc state = %d tcs_supported = %d\n", + pfc->pfc_en, pfc->tcs_supported); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_cee_peer_getpg(struct qed_dev *cdev, struct cee_pg *pg) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + pg->willing = dcbx_info->remote.params.ets_willing; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + pg->pg_bw[i] = dcbx_info->remote.params.ets_tc_bw_tbl[i]; + pg->prio_pg[i] = dcbx_info->remote.params.ets_pri_tc_tbl[i]; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "willing = %d", pg->willing); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_get_ieee_pfc(struct qed_dev *cdev, + struct ieee_pfc *pfc, bool remote) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_params *params; + struct qed_dcbx_get *dcbx_info; + int rc, i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + return -EINVAL; + } + + if (remote) { + memset(dcbx_info, 0, sizeof(*dcbx_info)); + rc = qed_dcbx_query_params(hwfn, dcbx_info, + QED_DCBX_REMOTE_MIB); + if (rc) { + kfree(dcbx_info); + return -EINVAL; + } + + params = &dcbx_info->remote.params; + } else { + params = &dcbx_info->operational.params; + } + + pfc->pfc_cap = params->pfc.max_tc; + pfc->pfc_en = 0; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (params->pfc.prio[i]) + pfc->pfc_en |= BIT(i); + + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_ieee_getpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + return qed_dcbnl_get_ieee_pfc(cdev, pfc, false); +} + +static int qed_dcbnl_ieee_setpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc, i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + dcbx_set.config.params.pfc.prio[i] = !!(pfc->pfc_en & BIT(i)); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static int qed_dcbnl_get_ieee_ets(struct qed_dev *cdev, + struct ieee_ets *ets, bool remote) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_params *params; + int rc; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + if (remote) { + memset(dcbx_info, 0, sizeof(*dcbx_info)); + rc = qed_dcbx_query_params(hwfn, dcbx_info, + QED_DCBX_REMOTE_MIB); + if (rc) { + kfree(dcbx_info); + return -EINVAL; + } + + params = &dcbx_info->remote.params; + } else { + params = &dcbx_info->operational.params; + } + + ets->ets_cap = params->max_ets_tc; + ets->willing = params->ets_willing; + ets->cbs = params->ets_cbs; + memcpy(ets->tc_tx_bw, params->ets_tc_bw_tbl, sizeof(ets->tc_tx_bw)); + memcpy(ets->tc_tsa, params->ets_tc_tsa_tbl, sizeof(ets->tc_tsa)); + memcpy(ets->prio_tc, params->ets_pri_tc_tbl, sizeof(ets->prio_tc)); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_ieee_getets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + return qed_dcbnl_get_ieee_ets(cdev, ets, false); +} + +static int qed_dcbnl_ieee_setets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.max_ets_tc = ets->ets_cap; + dcbx_set.config.params.ets_willing = ets->willing; + dcbx_set.config.params.ets_cbs = ets->cbs; + memcpy(dcbx_set.config.params.ets_tc_bw_tbl, ets->tc_tx_bw, + sizeof(ets->tc_tx_bw)); + memcpy(dcbx_set.config.params.ets_tc_tsa_tbl, ets->tc_tsa, + sizeof(ets->tc_tsa)); + memcpy(dcbx_set.config.params.ets_pri_tc_tbl, ets->prio_tc, + sizeof(ets->prio_tc)); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +int qed_dcbnl_ieee_peer_getets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + return qed_dcbnl_get_ieee_ets(cdev, ets, true); +} + +int qed_dcbnl_ieee_peer_getpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + return qed_dcbnl_get_ieee_pfc(cdev, pfc, true); +} + +int qed_dcbnl_ieee_getapp(struct qed_dev *cdev, struct dcb_app *app) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_app_entry *entry; + bool ethtype; + u8 prio = 0; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + /* ieee defines the selector field value for ethertype to be 1 */ + ethtype = !!((app->selector - 1) == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_info->operational.params.app_entry[i]; + if ((entry->ethtype == ethtype) && + (entry->proto_id == app->protocol)) { + prio = entry->prio; + break; + } + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App entry (%d, %d) not found\n", app->selector, + app->protocol); + kfree(dcbx_info); + return -EINVAL; + } + + app->priority = ffs(prio) - 1; + + kfree(dcbx_info); + + return 0; +} + +int qed_dcbnl_ieee_setapp(struct qed_dev *cdev, struct dcb_app *app) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_app_entry *entry; + struct qed_ptt *ptt; + bool ethtype; + int rc, i; + + if (app->priority < 0 || app->priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", app->priority); + return -EINVAL; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + /* ieee defines the selector field value for ethertype to be 1 */ + ethtype = !!((app->selector - 1) == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_set.config.params.app_entry[i]; + if ((entry->ethtype == ethtype) && + (entry->proto_id == app->protocol)) + break; + /* First empty slot */ + if (!entry->proto_id) + break; + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App table is full\n"); + return -EBUSY; + } + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_entry[i].ethtype = ethtype; + dcbx_set.config.params.app_entry[i].proto_id = app->protocol; + dcbx_set.config.params.app_entry[i].prio = BIT(app->priority); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EBUSY; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +const struct qed_eth_dcbnl_ops qed_dcbnl_ops_pass = { + .getstate = qed_dcbnl_getstate, + .setstate = qed_dcbnl_setstate, + .getpgtccfgtx = qed_dcbnl_getpgtccfgtx, + .getpgbwgcfgtx = qed_dcbnl_getpgbwgcfgtx, + .getpgtccfgrx = qed_dcbnl_getpgtccfgrx, + .getpgbwgcfgrx = qed_dcbnl_getpgbwgcfgrx, + .getpfccfg = qed_dcbnl_getpfccfg, + .setpfccfg = qed_dcbnl_setpfccfg, + .getcap = qed_dcbnl_getcap, + .getnumtcs = qed_dcbnl_getnumtcs, + .getpfcstate = qed_dcbnl_getpfcstate, + .getdcbx = qed_dcbnl_getdcbx, + .setpgtccfgtx = qed_dcbnl_setpgtccfgtx, + .setpgtccfgrx = qed_dcbnl_setpgtccfgrx, + .setpgbwgcfgtx = qed_dcbnl_setpgbwgcfgtx, + .setpgbwgcfgrx = qed_dcbnl_setpgbwgcfgrx, + .setall = qed_dcbnl_setall, + .setnumtcs = qed_dcbnl_setnumtcs, + .setpfcstate = qed_dcbnl_setpfcstate, + .setapp = qed_dcbnl_setapp, + .setdcbx = qed_dcbnl_setdcbx, + .setfeatcfg = qed_dcbnl_setfeatcfg, + .getfeatcfg = qed_dcbnl_getfeatcfg, + .getapp = qed_dcbnl_getapp, + .peer_getappinfo = qed_dcbnl_peer_getappinfo, + .peer_getapptable = qed_dcbnl_peer_getapptable, + .cee_peer_getpfc = qed_dcbnl_cee_peer_getpfc, + .cee_peer_getpg = qed_dcbnl_cee_peer_getpg, + .ieee_getpfc = qed_dcbnl_ieee_getpfc, + .ieee_setpfc = qed_dcbnl_ieee_setpfc, + .ieee_getets = qed_dcbnl_ieee_getets, + .ieee_setets = qed_dcbnl_ieee_setets, + .ieee_peer_getpfc = qed_dcbnl_ieee_peer_getpfc, + .ieee_peer_getets = qed_dcbnl_ieee_peer_getets, + .ieee_getapp = qed_dcbnl_ieee_getapp, + .ieee_setapp = qed_dcbnl_ieee_setapp, +}; + +#endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.h b/drivers/net/ethernet/qlogic/qed/qed_dcbx.h index e7f834dbda2d..9ba681643d05 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.h @@ -33,6 +33,24 @@ struct qed_dcbx_app_data { u8 tc; /* Traffic Class */ }; +#ifdef CONFIG_DCB +#define QED_DCBX_VERSION_DISABLED 0 +#define QED_DCBX_VERSION_IEEE 1 +#define QED_DCBX_VERSION_CEE 2 + +struct qed_dcbx_set { +#define QED_DCBX_OVERRIDE_STATE BIT(0) +#define QED_DCBX_OVERRIDE_PFC_CFG BIT(1) +#define QED_DCBX_OVERRIDE_ETS_CFG BIT(2) +#define QED_DCBX_OVERRIDE_APP_CFG BIT(3) +#define QED_DCBX_OVERRIDE_DSCP_CFG BIT(4) + u32 override_flags; + bool enabled; + struct qed_dcbx_admin_params config; + u32 ver_num; +}; +#endif + struct qed_dcbx_results { bool dcbx_enabled; u8 pf_id; @@ -55,6 +73,9 @@ struct qed_dcbx_info { struct qed_dcbx_results results; struct dcbx_mib operational; struct dcbx_mib remote; +#ifdef CONFIG_DCB + struct qed_dcbx_set set; +#endif u8 dcbx_cap; }; @@ -67,6 +88,13 @@ struct qed_dcbx_mib_meta_data { u32 addr; }; +#ifdef CONFIG_DCB +int qed_dcbx_get_config_params(struct qed_hwfn *, struct qed_dcbx_set *); + +int qed_dcbx_config_params(struct qed_hwfn *, + struct qed_ptt *, struct qed_dcbx_set *, bool); +#endif + /* QED local interface routines */ int qed_dcbx_mib_update_event(struct qed_hwfn *, diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index e45cff4df280..b26fe267a150 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -2222,6 +2222,110 @@ int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, return 0; } +static int qed_set_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u32 hw_addr, void *p_eth_qzone, + size_t eth_qzone_size, u8 timeset) +{ + struct coalescing_timeset *p_coal_timeset; + + if (p_hwfn->cdev->int_coalescing_mode != QED_COAL_MODE_ENABLE) { + DP_NOTICE(p_hwfn, "Coalescing configuration not enabled\n"); + return -EINVAL; + } + + p_coal_timeset = p_eth_qzone; + memset(p_coal_timeset, 0, eth_qzone_size); + SET_FIELD(p_coal_timeset->value, COALESCING_TIMESET_TIMESET, timeset); + SET_FIELD(p_coal_timeset->value, COALESCING_TIMESET_VALID, 1); + qed_memcpy_to(p_hwfn, p_ptt, hw_addr, p_eth_qzone, eth_qzone_size); + + return 0; +} + +int qed_set_rxq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id) +{ + struct ustorm_eth_queue_zone eth_qzone; + u8 timeset, timer_res; + u16 fw_qid = 0; + u32 address; + int rc; + + /* Coalesce = (timeset << timer-resolution), timeset is 7bit wide */ + if (coalesce <= 0x7F) { + timer_res = 0; + } else if (coalesce <= 0xFF) { + timer_res = 1; + } else if (coalesce <= 0x1FF) { + timer_res = 2; + } else { + DP_ERR(p_hwfn, "Invalid coalesce value - %d\n", coalesce); + return -EINVAL; + } + timeset = (u8)(coalesce >> timer_res); + + rc = qed_fw_l2_queue(p_hwfn, (u16)qid, &fw_qid); + if (rc) + return rc; + + rc = qed_int_set_timer_res(p_hwfn, p_ptt, timer_res, sb_id, false); + if (rc) + goto out; + + address = BAR0_MAP_REG_USDM_RAM + USTORM_ETH_QUEUE_ZONE_OFFSET(fw_qid); + + rc = qed_set_coalesce(p_hwfn, p_ptt, address, ð_qzone, + sizeof(struct ustorm_eth_queue_zone), timeset); + if (rc) + goto out; + + p_hwfn->cdev->rx_coalesce_usecs = coalesce; +out: + return rc; +} + +int qed_set_txq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id) +{ + struct xstorm_eth_queue_zone eth_qzone; + u8 timeset, timer_res; + u16 fw_qid = 0; + u32 address; + int rc; + + /* Coalesce = (timeset << timer-resolution), timeset is 7bit wide */ + if (coalesce <= 0x7F) { + timer_res = 0; + } else if (coalesce <= 0xFF) { + timer_res = 1; + } else if (coalesce <= 0x1FF) { + timer_res = 2; + } else { + DP_ERR(p_hwfn, "Invalid coalesce value - %d\n", coalesce); + return -EINVAL; + } + timeset = (u8)(coalesce >> timer_res); + + rc = qed_fw_l2_queue(p_hwfn, (u16)qid, &fw_qid); + if (rc) + return rc; + + rc = qed_int_set_timer_res(p_hwfn, p_ptt, timer_res, sb_id, true); + if (rc) + goto out; + + address = BAR0_MAP_REG_XSDM_RAM + XSTORM_ETH_QUEUE_ZONE_OFFSET(fw_qid); + + rc = qed_set_coalesce(p_hwfn, p_ptt, address, ð_qzone, + sizeof(struct xstorm_eth_queue_zone), timeset); + if (rc) + goto out; + + p_hwfn->cdev->tx_coalesce_usecs = coalesce; +out: + return rc; +} + /* Calculate final WFQ values for all vports and configure them. * After this configuration each vport will have * approx min rate = min_pf_rate * (vport_wfq / QED_WFQ_UNIT) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h index f810ce45d463..343bb0344f62 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h @@ -212,6 +212,20 @@ qed_dmae_host2grc(struct qed_hwfn *p_hwfn, u32 size_in_dwords, u32 flags); + /** + * @brief qed_dmae_grc2host - Read data from dmae data offset + * to source address using the given ptt + * + * @param p_ptt + * @param grc_addr (dmae_data_offset) + * @param dest_addr + * @param size_in_dwords + * @param flags - one of the flags defined above + */ +int qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u32 grc_addr, dma_addr_t dest_addr, u32 size_in_dwords, + u32 flags); + /** * @brief qed_dmae_host2host - copy data from to source address * to a destination adress (for SRIOV) using the given ptt @@ -308,4 +322,37 @@ int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, int qed_final_cleanup(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 id, bool is_vf); +/** + * @brief qed_set_rxq_coalesce - Configure coalesce parameters for an Rx queue + * The fact that we can configure coalescing to up to 511, but on varying + * accuracy [the bigger the value the less accurate] up to a mistake of 3usec + * for the highest values. + * + * @param p_hwfn + * @param p_ptt + * @param coalesce - Coalesce value in micro seconds. + * @param qid - Queue index. + * @param qid - SB Id + * + * @return int + */ +int qed_set_rxq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id); + +/** + * @brief qed_set_txq_coalesce - Configure coalesce parameters for a Tx queue + * While the API allows setting coalescing per-qid, all tx queues sharing a + * SB should be in same range [i.e., either 0-0x7f, 0x80-0xff or 0x100-0x1ff] + * otherwise configuration would break. + * + * @param p_hwfn + * @param p_ptt + * @param coalesce - Coalesce value in micro seconds. + * @param qid - Queue index. + * @param qid - SB Id + * + * @return int + */ +int qed_set_txq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id); #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 72e0ad6d16c6..592784019994 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -7244,6 +7244,9 @@ struct public_drv_mb { #define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_MASK 0x000000FF #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_SHIFT 8 #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_MASK 0x0000FF00 +#define DRV_MB_PARAM_LLDP_SEND_MASK 0x00000001 +#define DRV_MB_PARAM_LLDP_SEND_SHIFT 0 + #define DRV_MB_PARAM_SET_LED_MODE_OPER 0x0 #define DRV_MB_PARAM_SET_LED_MODE_ON 0x1 diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 2693c30981eb..e17885321faf 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -769,6 +769,29 @@ int qed_dmae_host2grc(struct qed_hwfn *p_hwfn, } int +qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 grc_addr, + dma_addr_t dest_addr, u32 size_in_dwords, u32 flags) +{ + u32 grc_addr_in_dw = grc_addr / sizeof(u32); + struct qed_dmae_params params; + int rc; + + memset(¶ms, 0, sizeof(struct qed_dmae_params)); + params.flags = flags; + + mutex_lock(&p_hwfn->dmae_info.mutex); + + rc = qed_dmae_execute_command(p_hwfn, p_ptt, grc_addr_in_dw, + dest_addr, QED_DMAE_ADDRESS_GRC, + QED_DMAE_ADDRESS_HOST_VIRT, + size_in_dwords, ¶ms); + + mutex_unlock(&p_hwfn->dmae_info.mutex); + + return rc; +} + +int qed_dmae_host2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, dma_addr_t source_addr, diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index 09a6ad3d22dd..8fa50fa23c8d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -2418,6 +2418,7 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, { struct qed_dev *cdev = p_hwfn->cdev; u32 cau_state; + u8 timer_res; memset(p_sb_entry, 0, sizeof(*p_sb_entry)); @@ -2443,6 +2444,23 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, cdev->tx_coalesce_usecs = QED_CAU_DEF_TX_USECS; } + /* Coalesce = (timeset << timer-res), timeset is 7bit wide */ + if (cdev->rx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (cdev->rx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES0, timer_res); + + if (cdev->tx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (cdev->tx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES1, timer_res); + SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE0, cau_state); SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE1, cau_state); } @@ -2484,17 +2502,28 @@ void qed_int_cau_conf_sb(struct qed_hwfn *p_hwfn, /* Configure pi coalescing if set */ if (p_hwfn->cdev->int_coalescing_mode == QED_COAL_MODE_ENABLE) { - u8 timeset = p_hwfn->cdev->rx_coalesce_usecs >> - (QED_CAU_DEF_RX_TIMER_RES + 1); + u8 timeset, timer_res; u8 num_tc = 1, i; + /* timeset = (coalesce >> timer-res), timeset is 7bit wide */ + if (p_hwfn->cdev->rx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (p_hwfn->cdev->rx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + timeset = (u8)(p_hwfn->cdev->rx_coalesce_usecs >> timer_res); qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, RX_PI, QED_COAL_RX_STATE_MACHINE, timeset); - timeset = p_hwfn->cdev->tx_coalesce_usecs >> - (QED_CAU_DEF_TX_TIMER_RES + 1); - + if (p_hwfn->cdev->tx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (p_hwfn->cdev->tx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + timeset = (u8)(p_hwfn->cdev->tx_coalesce_usecs >> timer_res); for (i = 0; i < num_tc; i++) { qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, TX_PI(i), @@ -3199,3 +3228,39 @@ void qed_int_disable_post_isr_release(struct qed_dev *cdev) for_each_hwfn(cdev, i) cdev->hwfns[i].b_int_requested = false; } + +int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u8 timer_res, u16 sb_id, bool tx) +{ + struct cau_sb_entry sb_entry; + int rc; + + if (!p_hwfn->hw_init_done) { + DP_ERR(p_hwfn, "hardware not initialized yet\n"); + return -EINVAL; + } + + rc = qed_dmae_grc2host(p_hwfn, p_ptt, CAU_REG_SB_VAR_MEMORY + + sb_id * sizeof(u64), + (u64)(uintptr_t)&sb_entry, 2, 0); + if (rc) { + DP_ERR(p_hwfn, "dmae_grc2host failed %d\n", rc); + return rc; + } + + if (tx) + SET_FIELD(sb_entry.params, CAU_SB_ENTRY_TIMER_RES1, timer_res); + else + SET_FIELD(sb_entry.params, CAU_SB_ENTRY_TIMER_RES0, timer_res); + + rc = qed_dmae_host2grc(p_hwfn, p_ptt, + (u64)(uintptr_t)&sb_entry, + CAU_REG_SB_VAR_MEMORY + + sb_id * sizeof(u64), 2, 0); + if (rc) { + DP_ERR(p_hwfn, "dmae_host2grc failed %d\n", rc); + return rc; + } + + return rc; +} diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h index 20b468637504..0948be64dc78 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.h +++ b/drivers/net/ethernet/qlogic/qed/qed_int.h @@ -389,6 +389,9 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, u16 vf_number, u8 vf_valid); +int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u8 timer_res, u16 sb_id, bool tx); + #define QED_MAPPING_MEMORY_SIZE(dev) (NUM_OF_SBS(dev)) #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 2ee496e26df6..a12c6caa6c66 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -72,6 +72,7 @@ int qed_sp_eth_vport_start(struct qed_hwfn *p_hwfn, p_ramrod->mtu = cpu_to_le16(p_params->mtu); p_ramrod->inner_vlan_removal_en = p_params->remove_inner_vlan; p_ramrod->drop_ttl0_en = p_params->drop_ttl0; + p_ramrod->untagged = p_params->only_untagged; SET_FIELD(rx_mode, ETH_VPORT_RX_MODE_UCAST_DROP_ALL, 1); SET_FIELD(rx_mode, ETH_VPORT_RX_MODE_MCAST_DROP_ALL, 1); @@ -247,10 +248,6 @@ qed_sp_update_accept_mode(struct qed_hwfn *p_hwfn, SET_FIELD(state, ETH_VPORT_TX_MODE_UCAST_DROP_ALL, !!(accept_filter & QED_ACCEPT_NONE)); - SET_FIELD(state, ETH_VPORT_TX_MODE_UCAST_ACCEPT_ALL, - (!!(accept_filter & QED_ACCEPT_UCAST_MATCHED) && - !!(accept_filter & QED_ACCEPT_UCAST_UNMATCHED))); - SET_FIELD(state, ETH_VPORT_TX_MODE_MCAST_DROP_ALL, !!(accept_filter & QED_ACCEPT_NONE)); @@ -1756,7 +1753,8 @@ static int qed_start_vport(struct qed_dev *cdev, start.vport_id, start.mtu); } - qed_reset_vport_stats(cdev); + if (params->clear_stats) + qed_reset_vport_stats(cdev); return 0; } @@ -2166,11 +2164,18 @@ static int qed_fp_cqe_completion(struct qed_dev *dev, extern const struct qed_iov_hv_ops qed_iov_ops_pass; #endif +#ifdef CONFIG_DCB +extern const struct qed_eth_dcbnl_ops qed_dcbnl_ops_pass; +#endif + static const struct qed_eth_ops qed_eth_ops_pass = { .common = &qed_common_ops_pass, #ifdef CONFIG_QED_SRIOV .iov = &qed_iov_ops_pass, #endif +#ifdef CONFIG_DCB + .dcb = &qed_dcbnl_ops_pass, +#endif .fill_dev_info = &qed_fill_eth_dev_info, .register_ops = &qed_register_eth_ops, .check_mac = &qed_check_mac, diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index f41f78d6a3f9..1f13abb5c316 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -1088,6 +1088,7 @@ static int qed_get_port_type(u32 media_type) case MEDIA_SFPP_10G_FIBER: case MEDIA_SFP_1G_FIBER: case MEDIA_XFP_FIBER: + case MEDIA_MODULE_FIBER: case MEDIA_KR: port_type = PORT_FIBRE; break; @@ -1108,6 +1109,39 @@ static int qed_get_port_type(u32 media_type) return port_type; } +static int qed_get_link_data(struct qed_hwfn *hwfn, + struct qed_mcp_link_params *params, + struct qed_mcp_link_state *link, + struct qed_mcp_link_capabilities *link_caps) +{ + void *p; + + if (!IS_PF(hwfn->cdev)) { + qed_vf_get_link_params(hwfn, params); + qed_vf_get_link_state(hwfn, link); + qed_vf_get_link_caps(hwfn, link_caps); + + return 0; + } + + p = qed_mcp_get_link_params(hwfn); + if (!p) + return -ENXIO; + memcpy(params, p, sizeof(*params)); + + p = qed_mcp_get_link_state(hwfn); + if (!p) + return -ENXIO; + memcpy(link, p, sizeof(*link)); + + p = qed_mcp_get_link_capabilities(hwfn); + if (!p) + return -ENXIO; + memcpy(link_caps, p, sizeof(*link_caps)); + + return 0; +} + static void qed_fill_link(struct qed_hwfn *hwfn, struct qed_link_output *if_link) { @@ -1119,15 +1153,9 @@ static void qed_fill_link(struct qed_hwfn *hwfn, memset(if_link, 0, sizeof(*if_link)); /* Prepare source inputs */ - if (IS_PF(hwfn->cdev)) { - memcpy(¶ms, qed_mcp_get_link_params(hwfn), sizeof(params)); - memcpy(&link, qed_mcp_get_link_state(hwfn), sizeof(link)); - memcpy(&link_caps, qed_mcp_get_link_capabilities(hwfn), - sizeof(link_caps)); - } else { - qed_vf_get_link_params(hwfn, ¶ms); - qed_vf_get_link_state(hwfn, &link); - qed_vf_get_link_caps(hwfn, &link_caps); + if (qed_get_link_data(hwfn, ¶ms, &link, &link_caps)) { + dev_warn(&hwfn->cdev->pdev->dev, "no link data available\n"); + return; } /* Set the link parameters to pass to protocol driver */ @@ -1276,6 +1304,38 @@ static int qed_drain(struct qed_dev *cdev) return 0; } +static void qed_get_coalesce(struct qed_dev *cdev, u16 *rx_coal, u16 *tx_coal) +{ + *rx_coal = cdev->rx_coalesce_usecs; + *tx_coal = cdev->tx_coalesce_usecs; +} + +static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, + u8 qid, u16 sb_id) +{ + struct qed_hwfn *hwfn; + struct qed_ptt *ptt; + int hwfn_index; + int status = 0; + + hwfn_index = qid % cdev->num_hwfns; + hwfn = &cdev->hwfns[hwfn_index]; + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EAGAIN; + + status = qed_set_rxq_coalesce(hwfn, ptt, rx_coal, + qid / cdev->num_hwfns, sb_id); + if (status) + goto out; + status = qed_set_txq_coalesce(hwfn, ptt, tx_coal, + qid / cdev->num_hwfns, sb_id); +out: + qed_ptt_release(hwfn, ptt); + + return status; +} + static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); @@ -1322,5 +1382,7 @@ const struct qed_common_ops qed_common_ops_pass = { .update_msglvl = &qed_init_dp, .chain_alloc = &qed_chain_alloc, .chain_free = &qed_chain_free, + .get_coalesce = &qed_get_coalesce, + .set_coalesce = &qed_set_coalesce, .set_led = &qed_set_led, }; diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index aa08ddbd95df..f6b86ca1ff79 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -80,6 +80,8 @@ 0x1f00000UL #define BAR0_MAP_REG_TSDM_RAM \ 0x1c80000UL +#define BAR0_MAP_REG_XSDM_RAM \ + 0x1e00000UL #define NIG_REG_RX_LLH_BRB_GATE_DNTFWD_PERPF \ 0x5011f4UL #define PRS_REG_SEARCH_TCP \ diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index ad9bf5c85c3f..97ffeae262bb 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -213,19 +213,15 @@ static int qed_spq_hw_post(struct qed_hwfn *p_hwfn, SET_FIELD(db.params, CORE_DB_DATA_AGG_VAL_SEL, DQ_XCM_CORE_SPQ_PROD_CMD); db.agg_flags = DQ_XCM_CORE_DQ_CF_CMD; - - /* validate producer is up to-date */ - rmb(); - db.spq_prod = cpu_to_le16(qed_chain_get_prod_idx(p_chain)); - /* do not reorder */ - barrier(); + /* make sure the SPQE is updated before the doorbell */ + wmb(); DOORBELL(p_hwfn, qed_db_addr(p_spq->cid, DQ_DEMS_LEGACY), *(u32 *)&db); /* make sure doorbell is rang */ - mmiowb(); + wmb(); DP_VERBOSE(p_hwfn, QED_MSG_SPQ, "Doorbelled [0x%08x, CID 0x%08x] with Flags: %02x agg_params: %02x, prod: %04x\n", @@ -620,7 +616,9 @@ qed_spq_add_entry(struct qed_hwfn *p_hwfn, *p_en2 = *p_ent; - kfree(p_ent); + /* EBLOCK responsible to free the allocated p_ent */ + if (p_ent->comp_mode != QED_SPQ_MODE_EBLOCK) + kfree(p_ent); p_ent = p_en2; } @@ -755,6 +753,15 @@ int qed_spq_post(struct qed_hwfn *p_hwfn, * Thus, after gaining the answer perform the cleanup here. */ rc = qed_spq_block(p_hwfn, p_ent, fw_return_code); + + if (p_ent->queue == &p_spq->unlimited_pending) { + /* This is an allocated p_ent which does not need to + * return to pool. + */ + kfree(p_ent); + return rc; + } + if (rc) goto spq_post_fail2; @@ -850,8 +857,12 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn, found->comp_cb.function(p_hwfn, found->comp_cb.cookie, p_data, fw_return_code); - if (found->comp_mode != QED_SPQ_MODE_EBLOCK) - /* EBLOCK is responsible for freeing its own entry */ + if ((found->comp_mode != QED_SPQ_MODE_EBLOCK) || + (found->queue == &p_spq->unlimited_pending)) + /* EBLOCK is responsible for returning its own entry into the + * free list, unless it originally added the entry into the + * unlimited pending list. + */ qed_spq_return_entry(p_hwfn, found); /* Attempt to post pending requests */ diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index eb75b8234a7c..4d161c751c12 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -21,18 +21,18 @@ #include "qed_vf.h" /* IOV ramrods */ -static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, - u32 concrete_vfid, u16 opaque_vfid) +static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, struct qed_vf_info *p_vf) { struct vf_start_ramrod_data *p_ramrod = NULL; struct qed_spq_entry *p_ent = NULL; struct qed_sp_init_data init_data; int rc = -EINVAL; + u8 fp_minor; /* Get SPQ entry */ memset(&init_data, 0, sizeof(init_data)); init_data.cid = qed_spq_get_cid(p_hwfn); - init_data.opaque_fid = opaque_vfid; + init_data.opaque_fid = p_vf->opaque_fid; init_data.comp_mode = QED_SPQ_MODE_EBLOCK; rc = qed_sp_init_request(p_hwfn, &p_ent, @@ -43,12 +43,39 @@ static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, p_ramrod = &p_ent->ramrod.vf_start; - p_ramrod->vf_id = GET_FIELD(concrete_vfid, PXP_CONCRETE_FID_VFID); - p_ramrod->opaque_fid = cpu_to_le16(opaque_vfid); + p_ramrod->vf_id = GET_FIELD(p_vf->concrete_fid, PXP_CONCRETE_FID_VFID); + p_ramrod->opaque_fid = cpu_to_le16(p_vf->opaque_fid); + + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ETH: + p_ramrod->personality = PERSONALITY_ETH; + break; + case QED_PCI_ETH_ROCE: + p_ramrod->personality = PERSONALITY_RDMA_AND_ETH; + break; + default: + DP_NOTICE(p_hwfn, "Unknown VF personality %d\n", + p_hwfn->hw_info.personality); + return -EINVAL; + } + + fp_minor = p_vf->acquire.vfdev_info.eth_fp_hsi_minor; + if (fp_minor > ETH_HSI_VER_MINOR) { + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF [%d] - Requested fp hsi %02x.%02x which is slightly newer than PF's %02x.%02x; Configuring PFs version\n", + p_vf->abs_vf_id, + ETH_HSI_VER_MAJOR, + fp_minor, ETH_HSI_VER_MAJOR, ETH_HSI_VER_MINOR); + fp_minor = ETH_HSI_VER_MINOR; + } - p_ramrod->personality = PERSONALITY_ETH; p_ramrod->hsi_fp_ver.major_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MAJOR; - p_ramrod->hsi_fp_ver.minor_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MINOR; + p_ramrod->hsi_fp_ver.minor_ver_arr[ETH_VER_KEY] = fp_minor; + + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "VF[%d] - Starting using HSI %02x.%02x\n", + p_vf->abs_vf_id, ETH_HSI_VER_MAJOR, fp_minor); return qed_spq_post(p_hwfn, p_ent, NULL); } @@ -119,6 +146,45 @@ static struct qed_vf_info *qed_iov_get_vf_info(struct qed_hwfn *p_hwfn, return vf; } +static bool qed_iov_validate_rxq(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, u16 rx_qid) +{ + if (rx_qid >= p_vf->num_rxqs) + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[0x%02x] - can't touch Rx queue[%04x]; Only 0x%04x are allocated\n", + p_vf->abs_vf_id, rx_qid, p_vf->num_rxqs); + return rx_qid < p_vf->num_rxqs; +} + +static bool qed_iov_validate_txq(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, u16 tx_qid) +{ + if (tx_qid >= p_vf->num_txqs) + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[0x%02x] - can't touch Tx queue[%04x]; Only 0x%04x are allocated\n", + p_vf->abs_vf_id, tx_qid, p_vf->num_txqs); + return tx_qid < p_vf->num_txqs; +} + +static bool qed_iov_validate_sb(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, u16 sb_idx) +{ + int i; + + for (i = 0; i < p_vf->num_sbs; i++) + if (p_vf->igu_sbs[i] == sb_idx) + return true; + + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[0%02x] - tried using sb_idx %04x which doesn't exist as one of its 0x%02x SBs\n", + p_vf->abs_vf_id, sb_idx, p_vf->num_sbs); + + return false; +} + int qed_iov_post_vf_bulletin(struct qed_hwfn *p_hwfn, int vfid, struct qed_ptt *p_ptt) { @@ -295,6 +361,9 @@ static void qed_iov_setup_vfdb(struct qed_hwfn *p_hwfn) vf->opaque_fid = (p_hwfn->hw_info.opaque_fid & 0xff) | (vf->abs_vf_id << 8); vf->vport_id = idx + 1; + + vf->num_mac_filters = QED_ETH_VF_NUM_MAC_FILTERS; + vf->num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; } } @@ -600,17 +669,6 @@ static int qed_iov_enable_vf_access(struct qed_hwfn *p_hwfn, /* unpretend */ qed_fid_pretend(p_hwfn, p_ptt, (u16) p_hwfn->hw_info.concrete_fid); - if (vf->state != VF_STOPPED) { - DP_NOTICE(p_hwfn, "VF[%02x] is already started\n", - vf->abs_vf_id); - return -EINVAL; - } - - /* Start VF */ - rc = qed_sp_vf_start(p_hwfn, vf->concrete_fid, vf->opaque_fid); - if (rc) - DP_NOTICE(p_hwfn, "Failed to start VF[%02x]\n", vf->abs_vf_id); - vf->state = VF_FREE; return rc; @@ -854,7 +912,6 @@ static int qed_iov_release_hw_for_vf(struct qed_hwfn *p_hwfn, struct qed_mcp_link_params params; struct qed_mcp_link_state link; struct qed_vf_info *vf = NULL; - int rc = 0; vf = qed_iov_get_vf_info(p_hwfn, rel_vf_id, true); if (!vf) { @@ -876,18 +933,8 @@ static int qed_iov_release_hw_for_vf(struct qed_hwfn *p_hwfn, memcpy(&caps, qed_mcp_get_link_capabilities(p_hwfn), sizeof(caps)); qed_iov_set_link(p_hwfn, rel_vf_id, ¶ms, &link, &caps); - if (vf->state != VF_STOPPED) { - /* Stopping the VF */ - rc = qed_sp_vf_stop(p_hwfn, vf->concrete_fid, vf->opaque_fid); - - if (rc != 0) { - DP_ERR(p_hwfn, "qed_sp_vf_stop returned error %d\n", - rc); - return rc; - } - - vf->state = VF_STOPPED; - } + /* Forget the VF's acquisition message */ + memset(&vf->acquire, 0, sizeof(vf->acquire)); /* disablng interrupts and resetting permission table was done during * vf-close, however, we could get here without going through vf_close @@ -1118,8 +1165,6 @@ static void qed_iov_vf_cleanup(struct qed_hwfn *p_hwfn, p_vf->vf_bulletin = 0; p_vf->vport_instance = 0; - p_vf->num_mac_filters = 0; - p_vf->num_vlan_filters = 0; p_vf->configured_features = 0; /* If VF previously requested less resources, go back to default */ @@ -1132,9 +1177,95 @@ static void qed_iov_vf_cleanup(struct qed_hwfn *p_hwfn, p_vf->vf_queues[i].rxq_active = 0; memset(&p_vf->shadow_config, 0, sizeof(p_vf->shadow_config)); + memset(&p_vf->acquire, 0, sizeof(p_vf->acquire)); qed_iov_clean_vf(p_hwfn, p_vf->relative_vf_id); } +static u8 qed_iov_vf_mbx_acquire_resc(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_vf_info *p_vf, + struct vf_pf_resc_request *p_req, + struct pf_vf_resc *p_resp) +{ + int i; + + /* Queue related information */ + p_resp->num_rxqs = p_vf->num_rxqs; + p_resp->num_txqs = p_vf->num_txqs; + p_resp->num_sbs = p_vf->num_sbs; + + for (i = 0; i < p_resp->num_sbs; i++) { + p_resp->hw_sbs[i].hw_sb_id = p_vf->igu_sbs[i]; + p_resp->hw_sbs[i].sb_qid = 0; + } + + /* These fields are filled for backward compatibility. + * Unused by modern vfs. + */ + for (i = 0; i < p_resp->num_rxqs; i++) { + qed_fw_l2_queue(p_hwfn, p_vf->vf_queues[i].fw_rx_qid, + (u16 *)&p_resp->hw_qid[i]); + p_resp->cid[i] = p_vf->vf_queues[i].fw_cid; + } + + /* Filter related information */ + p_resp->num_mac_filters = min_t(u8, p_vf->num_mac_filters, + p_req->num_mac_filters); + p_resp->num_vlan_filters = min_t(u8, p_vf->num_vlan_filters, + p_req->num_vlan_filters); + + /* This isn't really needed/enforced, but some legacy VFs might depend + * on the correct filling of this field. + */ + p_resp->num_mc_filters = QED_MAX_MC_ADDRS; + + /* Validate sufficient resources for VF */ + if (p_resp->num_rxqs < p_req->num_rxqs || + p_resp->num_txqs < p_req->num_txqs || + p_resp->num_sbs < p_req->num_sbs || + p_resp->num_mac_filters < p_req->num_mac_filters || + p_resp->num_vlan_filters < p_req->num_vlan_filters || + p_resp->num_mc_filters < p_req->num_mc_filters) { + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[%d] - Insufficient resources: rxq [%02x/%02x] txq [%02x/%02x] sbs [%02x/%02x] mac [%02x/%02x] vlan [%02x/%02x] mc [%02x/%02x]\n", + p_vf->abs_vf_id, + p_req->num_rxqs, + p_resp->num_rxqs, + p_req->num_rxqs, + p_resp->num_txqs, + p_req->num_sbs, + p_resp->num_sbs, + p_req->num_mac_filters, + p_resp->num_mac_filters, + p_req->num_vlan_filters, + p_resp->num_vlan_filters, + p_req->num_mc_filters, p_resp->num_mc_filters); + return PFVF_STATUS_NO_RESOURCE; + } + + return PFVF_STATUS_SUCCESS; +} + +static void qed_iov_vf_mbx_acquire_stats(struct qed_hwfn *p_hwfn, + struct pfvf_stats_info *p_stats) +{ + p_stats->mstats.address = PXP_VF_BAR0_START_MSDM_ZONE_B + + offsetof(struct mstorm_vf_zone, + non_trigger.eth_queue_stat); + p_stats->mstats.len = sizeof(struct eth_mstorm_per_queue_stat); + p_stats->ustats.address = PXP_VF_BAR0_START_USDM_ZONE_B + + offsetof(struct ustorm_vf_zone, + non_trigger.eth_queue_stat); + p_stats->ustats.len = sizeof(struct eth_ustorm_per_queue_stat); + p_stats->pstats.address = PXP_VF_BAR0_START_PSDM_ZONE_B + + offsetof(struct pstorm_vf_zone, + non_trigger.eth_queue_stat); + p_stats->pstats.len = sizeof(struct eth_pstorm_per_queue_stat); + p_stats->tstats.address = 0; + p_stats->tstats.len = 0; +} + static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_vf_info *vf) @@ -1143,25 +1274,27 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, struct pfvf_acquire_resp_tlv *resp = &mbx->reply_virt->acquire_resp; struct pf_vf_pfdev_info *pfdev_info = &resp->pfdev_info; struct vfpf_acquire_tlv *req = &mbx->req_virt->acquire; - u8 i, vfpf_status = PFVF_STATUS_SUCCESS; + u8 vfpf_status = PFVF_STATUS_NOT_SUPPORTED; struct pf_vf_resc *resc = &resp->resc; + int rc; + + memset(resp, 0, sizeof(*resp)); /* Validate FW compatibility */ - if (req->vfdev_info.fw_major != FW_MAJOR_VERSION || - req->vfdev_info.fw_minor != FW_MINOR_VERSION || - req->vfdev_info.fw_revision != FW_REVISION_VERSION || - req->vfdev_info.fw_engineering != FW_ENGINEERING_VERSION) { + if (req->vfdev_info.eth_fp_hsi_major != ETH_HSI_VER_MAJOR) { DP_INFO(p_hwfn, - "VF[%d] is running an incompatible driver [VF needs FW %02x:%02x:%02x:%02x but Hypervisor is using %02x:%02x:%02x:%02x]\n", + "VF[%d] needs fastpath HSI %02x.%02x, which is incompatible with loaded FW's faspath HSI %02x.%02x\n", vf->abs_vf_id, - req->vfdev_info.fw_major, - req->vfdev_info.fw_minor, - req->vfdev_info.fw_revision, - req->vfdev_info.fw_engineering, - FW_MAJOR_VERSION, - FW_MINOR_VERSION, - FW_REVISION_VERSION, FW_ENGINEERING_VERSION); - vfpf_status = PFVF_STATUS_NOT_SUPPORTED; + req->vfdev_info.eth_fp_hsi_major, + req->vfdev_info.eth_fp_hsi_minor, + ETH_HSI_VER_MAJOR, ETH_HSI_VER_MINOR); + + /* Write the PF version so that VF would know which version + * is supported. + */ + pfdev_info->major_fp_hsi = ETH_HSI_VER_MAJOR; + pfdev_info->minor_fp_hsi = ETH_HSI_VER_MINOR; + goto out; } @@ -1171,16 +1304,13 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, DP_INFO(p_hwfn, "VF[%d] is running an old driver that doesn't support 100g\n", vf->abs_vf_id); - vfpf_status = PFVF_STATUS_NOT_SUPPORTED; goto out; } - memset(resp, 0, sizeof(*resp)); + /* Store the acquire message */ + memcpy(&vf->acquire, req, sizeof(vf->acquire)); - /* Fill in vf info stuff */ vf->opaque_fid = req->vfdev_info.opaque_fid; - vf->num_mac_filters = 1; - vf->num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; vf->vf_bulletin = req->bulletin_addr; vf->bulletin.size = (vf->bulletin.size < req->bulletin_size) ? @@ -1196,26 +1326,7 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, if (p_hwfn->cdev->num_hwfns > 1) pfdev_info->capabilities |= PFVF_ACQUIRE_CAP_100G; - pfdev_info->stats_info.mstats.address = - PXP_VF_BAR0_START_MSDM_ZONE_B + - offsetof(struct mstorm_vf_zone, non_trigger.eth_queue_stat); - pfdev_info->stats_info.mstats.len = - sizeof(struct eth_mstorm_per_queue_stat); - - pfdev_info->stats_info.ustats.address = - PXP_VF_BAR0_START_USDM_ZONE_B + - offsetof(struct ustorm_vf_zone, non_trigger.eth_queue_stat); - pfdev_info->stats_info.ustats.len = - sizeof(struct eth_ustorm_per_queue_stat); - - pfdev_info->stats_info.pstats.address = - PXP_VF_BAR0_START_PSDM_ZONE_B + - offsetof(struct pstorm_vf_zone, non_trigger.eth_queue_stat); - pfdev_info->stats_info.pstats.len = - sizeof(struct eth_pstorm_per_queue_stat); - - pfdev_info->stats_info.tstats.address = 0; - pfdev_info->stats_info.tstats.len = 0; + qed_iov_vf_mbx_acquire_stats(p_hwfn, &pfdev_info->stats_info); memcpy(pfdev_info->port_mac, p_hwfn->hw_info.hw_mac_addr, ETH_ALEN); @@ -1223,36 +1334,31 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, pfdev_info->fw_minor = FW_MINOR_VERSION; pfdev_info->fw_rev = FW_REVISION_VERSION; pfdev_info->fw_eng = FW_ENGINEERING_VERSION; + pfdev_info->minor_fp_hsi = min_t(u8, + ETH_HSI_VER_MINOR, + req->vfdev_info.eth_fp_hsi_minor); pfdev_info->os_type = VFPF_ACQUIRE_OS_LINUX; qed_mcp_get_mfw_ver(p_hwfn, p_ptt, &pfdev_info->mfw_ver, NULL); pfdev_info->dev_type = p_hwfn->cdev->type; pfdev_info->chip_rev = p_hwfn->cdev->chip_rev; - resc->num_rxqs = vf->num_rxqs; - resc->num_txqs = vf->num_txqs; - resc->num_sbs = vf->num_sbs; - for (i = 0; i < resc->num_sbs; i++) { - resc->hw_sbs[i].hw_sb_id = vf->igu_sbs[i]; - resc->hw_sbs[i].sb_qid = 0; - } + /* Fill resources available to VF; Make sure there are enough to + * satisfy the VF's request. + */ + vfpf_status = qed_iov_vf_mbx_acquire_resc(p_hwfn, p_ptt, vf, + &req->resc_request, resc); + if (vfpf_status != PFVF_STATUS_SUCCESS) + goto out; - for (i = 0; i < resc->num_rxqs; i++) { - qed_fw_l2_queue(p_hwfn, vf->vf_queues[i].fw_rx_qid, - (u16 *)&resc->hw_qid[i]); - resc->cid[i] = vf->vf_queues[i].fw_cid; + /* Start the VF in FW */ + rc = qed_sp_vf_start(p_hwfn, vf); + if (rc) { + DP_NOTICE(p_hwfn, "Failed to start VF[%02x]\n", vf->abs_vf_id); + vfpf_status = PFVF_STATUS_FAILURE; + goto out; } - resc->num_mac_filters = min_t(u8, vf->num_mac_filters, - req->resc_request.num_mac_filters); - resc->num_vlan_filters = min_t(u8, vf->num_vlan_filters, - req->resc_request.num_vlan_filters); - - /* This isn't really required as VF isn't limited, but some VFs might - * actually test this value, so need to provide it. - */ - resc->num_mc_filters = req->resc_request.num_mc_filters; - /* Fill agreed size of bulletin board in response */ resp->bulletin_size = vf->bulletin.size; qed_iov_post_vf_bulletin(p_hwfn, vf->relative_vf_id, p_ptt); @@ -1620,12 +1726,17 @@ static void qed_iov_vf_mbx_start_rxq(struct qed_hwfn *p_hwfn, { struct qed_queue_start_common_params params; struct qed_iov_vf_mbx *mbx = &vf->vf_mbx; - u8 status = PFVF_STATUS_SUCCESS; + u8 status = PFVF_STATUS_NO_RESOURCE; struct vfpf_start_rxq_tlv *req; int rc; memset(¶ms, 0, sizeof(params)); req = &mbx->req_virt->start_rxq; + + if (!qed_iov_validate_rxq(p_hwfn, vf, req->rx_qid) || + !qed_iov_validate_sb(p_hwfn, vf, req->hw_sb)) + goto out; + params.queue_id = vf->vf_queues[req->rx_qid].fw_rx_qid; params.vf_qid = req->rx_qid; params.vport_id = vf->vport_id; @@ -1643,22 +1754,48 @@ static void qed_iov_vf_mbx_start_rxq(struct qed_hwfn *p_hwfn, if (rc) { status = PFVF_STATUS_FAILURE; } else { + status = PFVF_STATUS_SUCCESS; vf->vf_queues[req->rx_qid].rxq_active = true; vf->num_active_rxqs++; } +out: qed_iov_vf_mbx_start_rxq_resp(p_hwfn, p_ptt, vf, status); } +static void qed_iov_vf_mbx_start_txq_resp(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_vf_info *p_vf, u8 status) +{ + struct qed_iov_vf_mbx *mbx = &p_vf->vf_mbx; + struct pfvf_start_queue_resp_tlv *p_tlv; + + mbx->offset = (u8 *)mbx->reply_virt; + + p_tlv = qed_add_tlv(p_hwfn, &mbx->offset, CHANNEL_TLV_START_TXQ, + sizeof(*p_tlv)); + qed_add_tlv(p_hwfn, &mbx->offset, CHANNEL_TLV_LIST_END, + sizeof(struct channel_list_end_tlv)); + + /* Update the TLV with the response */ + if (status == PFVF_STATUS_SUCCESS) { + u16 qid = mbx->req_virt->start_txq.tx_qid; + + p_tlv->offset = qed_db_addr(p_vf->vf_queues[qid].fw_cid, + DQ_DEMS_LEGACY); + } + + qed_iov_send_response(p_hwfn, p_ptt, p_vf, sizeof(*p_tlv), status); +} + static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_vf_info *vf) { - u16 length = sizeof(struct pfvf_def_resp_tlv); struct qed_queue_start_common_params params; struct qed_iov_vf_mbx *mbx = &vf->vf_mbx; + u8 status = PFVF_STATUS_NO_RESOURCE; union qed_qm_pq_params pq_params; - u8 status = PFVF_STATUS_SUCCESS; struct vfpf_start_txq_tlv *req; int rc; @@ -1669,6 +1806,11 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, memset(¶ms, 0, sizeof(params)); req = &mbx->req_virt->start_txq; + + if (!qed_iov_validate_txq(p_hwfn, vf, req->tx_qid) || + !qed_iov_validate_sb(p_hwfn, vf, req->hw_sb)) + goto out; + params.queue_id = vf->vf_queues[req->tx_qid].fw_tx_qid; params.vport_id = vf->vport_id; params.sb = req->hw_sb; @@ -1682,13 +1824,15 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, req->pbl_addr, req->pbl_size, &pq_params); - if (rc) + if (rc) { status = PFVF_STATUS_FAILURE; - else + } else { + status = PFVF_STATUS_SUCCESS; vf->vf_queues[req->tx_qid].txq_active = true; + } - qed_iov_prepare_resp(p_hwfn, p_ptt, vf, CHANNEL_TLV_START_TXQ, - length, status); +out: + qed_iov_vf_mbx_start_txq_resp(p_hwfn, p_ptt, vf, status); } static int qed_iov_vf_stop_rxqs(struct qed_hwfn *p_hwfn, @@ -2113,6 +2257,16 @@ static void qed_iov_vf_mbx_vport_update(struct qed_hwfn *p_hwfn, u16 length; int rc; + /* Valiate PF can send such a request */ + if (!vf->vport_instance) { + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "No VPORT instance available for VF[%d], failing vport update\n", + vf->abs_vf_id); + status = PFVF_STATUS_FAILURE; + goto out; + } + memset(¶ms, 0, sizeof(params)); params.opaque_fid = vf->opaque_fid; params.vport_id = vf->vport_id; @@ -2155,15 +2309,12 @@ out: qed_iov_send_response(p_hwfn, p_ptt, vf, length, status); } -static int qed_iov_vf_update_unicast_shadow(struct qed_hwfn *p_hwfn, - struct qed_vf_info *p_vf, - struct qed_filter_ucast *p_params) +static int qed_iov_vf_update_vlan_shadow(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, + struct qed_filter_ucast *p_params) { int i; - if (p_params->type == QED_FILTER_MAC) - return 0; - /* First remove entries and then add new ones */ if (p_params->opcode == QED_FILTER_REMOVE) { for (i = 0; i < QED_ETH_VF_NUM_VLAN_FILTERS + 1; i++) @@ -2216,6 +2367,80 @@ static int qed_iov_vf_update_unicast_shadow(struct qed_hwfn *p_hwfn, return 0; } +static int qed_iov_vf_update_mac_shadow(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, + struct qed_filter_ucast *p_params) +{ + int i; + + /* If we're in forced-mode, we don't allow any change */ + if (p_vf->bulletin.p_virt->valid_bitmap & (1 << MAC_ADDR_FORCED)) + return 0; + + /* First remove entries and then add new ones */ + if (p_params->opcode == QED_FILTER_REMOVE) { + for (i = 0; i < QED_ETH_VF_NUM_MAC_FILTERS; i++) { + if (ether_addr_equal(p_vf->shadow_config.macs[i], + p_params->mac)) { + memset(p_vf->shadow_config.macs[i], 0, + ETH_ALEN); + break; + } + } + + if (i == QED_ETH_VF_NUM_MAC_FILTERS) { + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "MAC isn't configured\n"); + return -EINVAL; + } + } else if (p_params->opcode == QED_FILTER_REPLACE || + p_params->opcode == QED_FILTER_FLUSH) { + for (i = 0; i < QED_ETH_VF_NUM_MAC_FILTERS; i++) + memset(p_vf->shadow_config.macs[i], 0, ETH_ALEN); + } + + /* List the new MAC address */ + if (p_params->opcode != QED_FILTER_ADD && + p_params->opcode != QED_FILTER_REPLACE) + return 0; + + for (i = 0; i < QED_ETH_VF_NUM_MAC_FILTERS; i++) { + if (is_zero_ether_addr(p_vf->shadow_config.macs[i])) { + ether_addr_copy(p_vf->shadow_config.macs[i], + p_params->mac); + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "Added MAC at %d entry in shadow\n", i); + break; + } + } + + if (i == QED_ETH_VF_NUM_MAC_FILTERS) { + DP_VERBOSE(p_hwfn, QED_MSG_IOV, "No available place for MAC\n"); + return -EINVAL; + } + + return 0; +} + +static int +qed_iov_vf_update_unicast_shadow(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, + struct qed_filter_ucast *p_params) +{ + int rc = 0; + + if (p_params->type == QED_FILTER_MAC) { + rc = qed_iov_vf_update_mac_shadow(p_hwfn, p_vf, p_params); + if (rc) + return rc; + } + + if (p_params->type == QED_FILTER_VLAN) + rc = qed_iov_vf_update_vlan_shadow(p_hwfn, p_vf, p_params); + + return rc; +} + int qed_iov_chk_ucast(struct qed_hwfn *hwfn, int vfid, struct qed_filter_ucast *params) { @@ -2360,11 +2585,27 @@ static void qed_iov_vf_mbx_release(struct qed_hwfn *p_hwfn, struct qed_vf_info *p_vf) { u16 length = sizeof(struct pfvf_def_resp_tlv); + u8 status = PFVF_STATUS_SUCCESS; + int rc = 0; qed_iov_vf_cleanup(p_hwfn, p_vf); + if (p_vf->state != VF_STOPPED && p_vf->state != VF_FREE) { + /* Stopping the VF */ + rc = qed_sp_vf_stop(p_hwfn, p_vf->concrete_fid, + p_vf->opaque_fid); + + if (rc) { + DP_ERR(p_hwfn, "qed_sp_vf_stop returned error %d\n", + rc); + status = PFVF_STATUS_FAILURE; + } + + p_vf->state = VF_STOPPED; + } + qed_iov_prepare_resp(p_hwfn, p_ptt, p_vf, CHANNEL_TLV_RELEASE, - length, PFVF_STATUS_SUCCESS); + length, status); } static int @@ -2616,7 +2857,6 @@ static void qed_iov_process_mbx_req(struct qed_hwfn *p_hwfn, { struct qed_iov_vf_mbx *mbx; struct qed_vf_info *p_vf; - int i; p_vf = qed_iov_get_vf_info(p_hwfn, (u16) vfid, true); if (!p_vf) @@ -2625,9 +2865,8 @@ static void qed_iov_process_mbx_req(struct qed_hwfn *p_hwfn, mbx = &p_vf->vf_mbx; /* qed_iov_process_mbx_request */ - DP_VERBOSE(p_hwfn, - QED_MSG_IOV, - "qed_iov_process_mbx_req vfid %d\n", p_vf->abs_vf_id); + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "VF[%02x]: Processing mailbox message\n", p_vf->abs_vf_id); mbx->first_tlv = mbx->req_virt->first_tlv; @@ -2681,15 +2920,28 @@ static void qed_iov_process_mbx_req(struct qed_hwfn *p_hwfn, * support them. Or this may be because someone wrote a crappy * VF driver and is sending garbage over the channel. */ - DP_ERR(p_hwfn, - "unknown TLV. type %d length %d. first 20 bytes of mailbox buffer:\n", - mbx->first_tlv.tl.type, mbx->first_tlv.tl.length); - - for (i = 0; i < 20; i++) { + DP_NOTICE(p_hwfn, + "VF[%02x]: unknown TLV. type %04x length %04x padding %08x reply address %llu\n", + p_vf->abs_vf_id, + mbx->first_tlv.tl.type, + mbx->first_tlv.tl.length, + mbx->first_tlv.padding, mbx->first_tlv.reply_address); + + /* Try replying in case reply address matches the acquisition's + * posted address. + */ + if (p_vf->acquire.first_tlv.reply_address && + (mbx->first_tlv.reply_address == + p_vf->acquire.first_tlv.reply_address)) { + qed_iov_prepare_resp(p_hwfn, p_ptt, p_vf, + mbx->first_tlv.tl.type, + sizeof(struct pfvf_def_resp_tlv), + PFVF_STATUS_NOT_SUPPORTED); + } else { DP_VERBOSE(p_hwfn, QED_MSG_IOV, - "%x ", - mbx->req_virt->tlv_buf_size.tlv_buffer[i]); + "VF[%02x]: Can't respond to TLV - no valid reply address\n", + p_vf->abs_vf_id); } } } diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.h b/drivers/net/ethernet/qlogic/qed/qed_sriov.h index c8667c65e685..0dd23e409b3f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.h @@ -10,19 +10,23 @@ #define _QED_SRIOV_H #include <linux/types.h> #include "qed_vf.h" + +#define QED_ETH_VF_NUM_MAC_FILTERS 1 +#define QED_ETH_VF_NUM_VLAN_FILTERS 2 #define QED_VF_ARRAY_LENGTH (3) +#ifdef CONFIG_QED_SRIOV #define IS_VF(cdev) ((cdev)->b_is_vf) #define IS_PF(cdev) (!((cdev)->b_is_vf)) -#ifdef CONFIG_QED_SRIOV #define IS_PF_SRIOV(p_hwfn) (!!((p_hwfn)->cdev->p_iov_info)) #else +#define IS_VF(cdev) (0) +#define IS_PF(cdev) (1) #define IS_PF_SRIOV(p_hwfn) (0) #endif #define IS_PF_SRIOV_ALLOC(p_hwfn) (!!((p_hwfn)->pf_iov_info)) #define QED_MAX_VF_CHAINS_PER_PF 16 -#define QED_ETH_VF_NUM_VLAN_FILTERS 2 #define QED_ETH_MAX_VF_NUM_VLAN_FILTERS \ (MAX_NUM_VFS * QED_ETH_VF_NUM_VLAN_FILTERS) @@ -118,6 +122,8 @@ struct qed_vf_shadow_config { /* Shadow copy of all guest vlans */ struct qed_vf_vlan_shadow vlans[QED_ETH_VF_NUM_VLAN_FILTERS + 1]; + /* Shadow copy of all configured MACs; Empty if forcing MACs */ + u8 macs[QED_ETH_VF_NUM_MAC_FILTERS][ETH_ALEN]; u8 inner_vlan_removal; }; @@ -131,6 +137,9 @@ struct qed_vf_info { struct qed_bulletin bulletin; dma_addr_t vf_bulletin; + /* PF saves a copy of the last VF acquire message */ + struct vfpf_acquire_tlv acquire; + u32 concrete_fid; u16 opaque_fid; u16 mtu; diff --git a/drivers/net/ethernet/qlogic/qed/qed_vf.c b/drivers/net/ethernet/qlogic/qed/qed_vf.c index 72e69c0ec10d..9819230947bf 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_vf.c +++ b/drivers/net/ethernet/qlogic/qed/qed_vf.c @@ -117,36 +117,64 @@ exit: } #define VF_ACQUIRE_THRESH 3 -#define VF_ACQUIRE_MAC_FILTERS 1 +static void qed_vf_pf_acquire_reduce_resc(struct qed_hwfn *p_hwfn, + struct vf_pf_resc_request *p_req, + struct pf_vf_resc *p_resp) +{ + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "PF unwilling to fullill resource request: rxq [%02x/%02x] txq [%02x/%02x] sbs [%02x/%02x] mac [%02x/%02x] vlan [%02x/%02x] mc [%02x/%02x]. Try PF recommended amount\n", + p_req->num_rxqs, + p_resp->num_rxqs, + p_req->num_rxqs, + p_resp->num_txqs, + p_req->num_sbs, + p_resp->num_sbs, + p_req->num_mac_filters, + p_resp->num_mac_filters, + p_req->num_vlan_filters, + p_resp->num_vlan_filters, + p_req->num_mc_filters, p_resp->num_mc_filters); + + /* humble our request */ + p_req->num_txqs = p_resp->num_txqs; + p_req->num_rxqs = p_resp->num_rxqs; + p_req->num_sbs = p_resp->num_sbs; + p_req->num_mac_filters = p_resp->num_mac_filters; + p_req->num_vlan_filters = p_resp->num_vlan_filters; + p_req->num_mc_filters = p_resp->num_mc_filters; +} static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) { struct qed_vf_iov *p_iov = p_hwfn->vf_iov_info; struct pfvf_acquire_resp_tlv *resp = &p_iov->pf2vf_reply->acquire_resp; struct pf_vf_pfdev_info *pfdev_info = &resp->pfdev_info; - u8 rx_count = 1, tx_count = 1, num_sbs = 1; - u8 num_mac = VF_ACQUIRE_MAC_FILTERS; + struct vf_pf_resc_request *p_resc; bool resources_acquired = false; struct vfpf_acquire_tlv *req; int rc = 0, attempts = 0; /* clear mailbox and prep first tlv */ req = qed_vf_pf_prep(p_hwfn, CHANNEL_TLV_ACQUIRE, sizeof(*req)); + p_resc = &req->resc_request; /* starting filling the request */ req->vfdev_info.opaque_fid = p_hwfn->hw_info.opaque_fid; - req->resc_request.num_rxqs = rx_count; - req->resc_request.num_txqs = tx_count; - req->resc_request.num_sbs = num_sbs; - req->resc_request.num_mac_filters = num_mac; - req->resc_request.num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; + p_resc->num_rxqs = QED_MAX_VF_CHAINS_PER_PF; + p_resc->num_txqs = QED_MAX_VF_CHAINS_PER_PF; + p_resc->num_sbs = QED_MAX_VF_CHAINS_PER_PF; + p_resc->num_mac_filters = QED_ETH_VF_NUM_MAC_FILTERS; + p_resc->num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; req->vfdev_info.os_type = VFPF_ACQUIRE_OS_LINUX; req->vfdev_info.fw_major = FW_MAJOR_VERSION; req->vfdev_info.fw_minor = FW_MINOR_VERSION; req->vfdev_info.fw_revision = FW_REVISION_VERSION; req->vfdev_info.fw_engineering = FW_ENGINEERING_VERSION; + req->vfdev_info.eth_fp_hsi_major = ETH_HSI_VER_MAJOR; + req->vfdev_info.eth_fp_hsi_minor = ETH_HSI_VER_MINOR; /* Fill capability field with any non-deprecated config we support */ req->vfdev_info.capabilities |= VFPF_ACQUIRE_CAP_100G; @@ -185,21 +213,21 @@ static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) resources_acquired = true; } else if (resp->hdr.status == PFVF_STATUS_NO_RESOURCE && attempts < VF_ACQUIRE_THRESH) { - DP_VERBOSE(p_hwfn, - QED_MSG_IOV, - "PF unwilling to fullfill resource request. Try PF recommended amount\n"); - - /* humble our request */ - req->resc_request.num_txqs = resp->resc.num_txqs; - req->resc_request.num_rxqs = resp->resc.num_rxqs; - req->resc_request.num_sbs = resp->resc.num_sbs; - req->resc_request.num_mac_filters = - resp->resc.num_mac_filters; - req->resc_request.num_vlan_filters = - resp->resc.num_vlan_filters; + qed_vf_pf_acquire_reduce_resc(p_hwfn, p_resc, + &resp->resc); /* Clear response buffer */ memset(p_iov->pf2vf_reply, 0, sizeof(union pfvf_tlvs)); + } else if ((resp->hdr.status == PFVF_STATUS_NOT_SUPPORTED) && + pfdev_info->major_fp_hsi && + (pfdev_info->major_fp_hsi != ETH_HSI_VER_MAJOR)) { + DP_NOTICE(p_hwfn, + "PF uses an incompatible fastpath HSI %02x.%02x [VF requires %02x.%02x]. Please change to a VF driver using %02x.xx.\n", + pfdev_info->major_fp_hsi, + pfdev_info->minor_fp_hsi, + ETH_HSI_VER_MAJOR, + ETH_HSI_VER_MINOR, pfdev_info->major_fp_hsi); + return -EINVAL; } else { DP_ERR(p_hwfn, "PF returned error %d to VF acquisition request\n", @@ -225,6 +253,13 @@ static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) } } + if (ETH_HSI_VER_MINOR && + (resp->pfdev_info.minor_fp_hsi < ETH_HSI_VER_MINOR)) { + DP_INFO(p_hwfn, + "PF is using older fastpath HSI; %02x.%02x is configured\n", + ETH_HSI_VER_MAJOR, resp->pfdev_info.minor_fp_hsi); + } + return 0; } @@ -405,8 +440,8 @@ int qed_vf_pf_txq_start(struct qed_hwfn *p_hwfn, u16 pbl_size, void __iomem **pp_doorbell) { struct qed_vf_iov *p_iov = p_hwfn->vf_iov_info; + struct pfvf_start_queue_resp_tlv *resp; struct vfpf_start_txq_tlv *req; - struct pfvf_def_resp_tlv *resp; int rc; /* clear mailbox and prep first tlv */ @@ -424,20 +459,24 @@ int qed_vf_pf_txq_start(struct qed_hwfn *p_hwfn, qed_add_tlv(p_hwfn, &p_iov->offset, CHANNEL_TLV_LIST_END, sizeof(struct channel_list_end_tlv)); - resp = &p_iov->pf2vf_reply->default_resp; + resp = &p_iov->pf2vf_reply->queue_start; rc = qed_send_msg2pf(p_hwfn, &resp->hdr.status, sizeof(*resp)); if (rc) - return rc; + goto exit; - if (resp->hdr.status != PFVF_STATUS_SUCCESS) - return -EINVAL; + if (resp->hdr.status != PFVF_STATUS_SUCCESS) { + rc = -EINVAL; + goto exit; + } if (pp_doorbell) { - u8 cid = p_iov->acquire_resp.resc.cid[tx_queue_id]; + *pp_doorbell = (u8 __iomem *)p_hwfn->doorbells + resp->offset; - *pp_doorbell = (u8 __iomem *)p_hwfn->doorbells + - qed_db_addr(cid, DQ_DEMS_LEGACY); + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "Txq[0x%02x]: doorbell at %p [offset 0x%08x]\n", + tx_queue_id, *pp_doorbell, resp->offset); } +exit: return rc; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_vf.h b/drivers/net/ethernet/qlogic/qed/qed_vf.h index b82fda964bbd..b23ce58e932f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_vf.h +++ b/drivers/net/ethernet/qlogic/qed/qed_vf.h @@ -96,7 +96,9 @@ struct vfpf_acquire_tlv { u32 driver_version; u16 opaque_fid; /* ME register value */ u8 os_type; /* VFPF_ACQUIRE_OS_* value */ - u8 padding[5]; + u8 eth_fp_hsi_major; + u8 eth_fp_hsi_minor; + u8 padding[3]; } vfdev_info; struct vf_pf_resc_request resc_request; @@ -171,7 +173,14 @@ struct pfvf_acquire_resp_tlv { struct pfvf_stats_info stats_info; u8 port_mac[ETH_ALEN]; - u8 padding2[2]; + + /* It's possible PF had to configure an older fastpath HSI + * [in case VF is newer than PF]. This is communicated back + * to the VF. It can also be used in case of error due to + * non-matching versions to shed light in VF about failure. + */ + u8 major_fp_hsi; + u8 minor_fp_hsi; } pfdev_info; struct pf_vf_resc { diff --git a/drivers/net/ethernet/qlogic/qede/Makefile b/drivers/net/ethernet/qlogic/qede/Makefile index 06ff90d87572..74a49850d74d 100644 --- a/drivers/net/ethernet/qlogic/qede/Makefile +++ b/drivers/net/ethernet/qlogic/qede/Makefile @@ -1,3 +1,4 @@ obj-$(CONFIG_QEDE) := qede.o qede-y := qede_main.o qede_ethtool.o +qede-$(CONFIG_DCB) += qede_dcbnl.o diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 47d6b22252f6..02b06d4e40ae 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -24,7 +24,7 @@ #include <linux/qed/qed_eth_if.h> #define QEDE_MAJOR_VERSION 8 -#define QEDE_MINOR_VERSION 7 +#define QEDE_MINOR_VERSION 10 #define QEDE_REVISION_VERSION 1 #define QEDE_ENGINEERING_VERSION 20 #define DRV_MODULE_VERSION __stringify(QEDE_MAJOR_VERSION) "." \ @@ -143,6 +143,8 @@ struct qede_dev { struct mutex qede_lock; u32 state; /* Protected by qede_lock */ u16 rx_buf_size; + u32 rx_copybreak; + /* L2 header size + 2*VLANs (8 bytes) + LLC SNAP (8 bytes) */ #define ETH_OVERHEAD (ETH_HLEN + 8 + 8) /* Max supported alignment is 256 (8 shift) @@ -235,6 +237,7 @@ struct qede_rx_queue { u64 rx_hw_errors; u64 rx_alloc_errors; + u64 rx_ip_frags; }; union db_prod { @@ -304,6 +307,9 @@ union qede_reload_args { u16 mtu; }; +#ifdef CONFIG_DCB +void qede_set_dcbnl_ops(struct net_device *ndev); +#endif void qede_config_debug(uint debug, u32 *p_dp_module, u8 *p_dp_level); void qede_set_ethtool_ops(struct net_device *netdev); void qede_reload(struct qede_dev *edev, @@ -329,6 +335,7 @@ void qede_recycle_rx_bd_ring(struct qede_rx_queue *rxq, struct qede_dev *edev, #define NUM_TX_BDS_MIN 128 #define NUM_TX_BDS_DEF NUM_TX_BDS_MAX +#define QEDE_MIN_PKT_LEN 64 #define QEDE_RX_HDR_SIZE 256 #define for_each_rss(i) for (i = 0; i < edev->num_rss; i++) diff --git a/drivers/net/ethernet/qlogic/qede/qede_dcbnl.c b/drivers/net/ethernet/qlogic/qede/qede_dcbnl.c new file mode 100644 index 000000000000..03e8c0212433 --- /dev/null +++ b/drivers/net/ethernet/qlogic/qede/qede_dcbnl.c @@ -0,0 +1,348 @@ +/* QLogic qede NIC Driver +* Copyright (c) 2015 QLogic Corporation +* +* This software is available under the terms of the GNU General Public License +* (GPL) Version 2, available from the file COPYING in the main directory of +* this source tree. +*/ + +#include <linux/types.h> +#include <linux/netdevice.h> +#include <linux/rtnetlink.h> +#include <net/dcbnl.h> +#include "qede.h" + +static u8 qede_dcbnl_getstate(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getstate(edev->cdev); +} + +static u8 qede_dcbnl_setstate(struct net_device *netdev, u8 state) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setstate(edev->cdev, state); +} + +static void qede_dcbnl_getpermhwaddr(struct net_device *netdev, + u8 *perm_addr) +{ + memcpy(perm_addr, netdev->dev_addr, netdev->addr_len); +} + +static void qede_dcbnl_getpgtccfgtx(struct net_device *netdev, int prio, + u8 *prio_type, u8 *pgid, u8 *bw_pct, + u8 *up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgtccfgtx(edev->cdev, prio, prio_type, + pgid, bw_pct, up_map); +} + +static void qede_dcbnl_getpgbwgcfgtx(struct net_device *netdev, + int pgid, u8 *bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgbwgcfgtx(edev->cdev, pgid, bw_pct); +} + +static void qede_dcbnl_getpgtccfgrx(struct net_device *netdev, int prio, + u8 *prio_type, u8 *pgid, u8 *bw_pct, + u8 *up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgtccfgrx(edev->cdev, prio, prio_type, pgid, bw_pct, + up_map); +} + +static void qede_dcbnl_getpgbwgcfgrx(struct net_device *netdev, + int pgid, u8 *bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgbwgcfgrx(edev->cdev, pgid, bw_pct); +} + +static void qede_dcbnl_getpfccfg(struct net_device *netdev, int prio, + u8 *setting) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpfccfg(edev->cdev, prio, setting); +} + +static void qede_dcbnl_setpfccfg(struct net_device *netdev, int prio, + u8 setting) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->setpfccfg(edev->cdev, prio, setting); +} + +static u8 qede_dcbnl_getcap(struct net_device *netdev, int capid, u8 *cap) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getcap(edev->cdev, capid, cap); +} + +static int qede_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getnumtcs(edev->cdev, tcid, num); +} + +static u8 qede_dcbnl_getpfcstate(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getpfcstate(edev->cdev); +} + +static int qede_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getapp(edev->cdev, idtype, id); +} + +static u8 qede_dcbnl_getdcbx(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getdcbx(edev->cdev); +} + +static void qede_dcbnl_setpgtccfgtx(struct net_device *netdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgtccfgtx(edev->cdev, prio, pri_type, pgid, + bw_pct, up_map); +} + +static void qede_dcbnl_setpgtccfgrx(struct net_device *netdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgtccfgrx(edev->cdev, prio, pri_type, pgid, + bw_pct, up_map); +} + +static void qede_dcbnl_setpgbwgcfgtx(struct net_device *netdev, int pgid, + u8 bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgbwgcfgtx(edev->cdev, pgid, bw_pct); +} + +static void qede_dcbnl_setpgbwgcfgrx(struct net_device *netdev, int pgid, + u8 bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgbwgcfgrx(edev->cdev, pgid, bw_pct); +} + +static u8 qede_dcbnl_setall(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setall(edev->cdev); +} + +static int qede_dcbnl_setnumtcs(struct net_device *netdev, int tcid, u8 num) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setnumtcs(edev->cdev, tcid, num); +} + +static void qede_dcbnl_setpfcstate(struct net_device *netdev, u8 state) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpfcstate(edev->cdev, state); +} + +static int qede_dcbnl_setapp(struct net_device *netdev, u8 idtype, u16 idval, + u8 up) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setapp(edev->cdev, idtype, idval, up); +} + +static u8 qede_dcbnl_setdcbx(struct net_device *netdev, u8 state) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setdcbx(edev->cdev, state); +} + +static u8 qede_dcbnl_getfeatcfg(struct net_device *netdev, int featid, + u8 *flags) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getfeatcfg(edev->cdev, featid, flags); +} + +static u8 qede_dcbnl_setfeatcfg(struct net_device *netdev, int featid, u8 flags) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setfeatcfg(edev->cdev, featid, flags); +} + +static int qede_dcbnl_peer_getappinfo(struct net_device *netdev, + struct dcb_peer_app_info *info, + u16 *count) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->peer_getappinfo(edev->cdev, info, count); +} + +static int qede_dcbnl_peer_getapptable(struct net_device *netdev, + struct dcb_app *app) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->peer_getapptable(edev->cdev, app); +} + +static int qede_dcbnl_cee_peer_getpfc(struct net_device *netdev, + struct cee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->cee_peer_getpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_cee_peer_getpg(struct net_device *netdev, + struct cee_pg *pg) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->cee_peer_getpg(edev->cdev, pg); +} + +static int qede_dcbnl_ieee_getpfc(struct net_device *netdev, + struct ieee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_getpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_ieee_setpfc(struct net_device *netdev, + struct ieee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_setpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_ieee_getets(struct net_device *netdev, + struct ieee_ets *ets) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_getets(edev->cdev, ets); +} + +static int qede_dcbnl_ieee_setets(struct net_device *netdev, + struct ieee_ets *ets) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_setets(edev->cdev, ets); +} + +static int qede_dcbnl_ieee_getapp(struct net_device *netdev, + struct dcb_app *app) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_getapp(edev->cdev, app); +} + +static int qede_dcbnl_ieee_setapp(struct net_device *netdev, + struct dcb_app *app) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_setapp(edev->cdev, app); +} + +static int qede_dcbnl_ieee_peer_getpfc(struct net_device *netdev, + struct ieee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_peer_getpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_ieee_peer_getets(struct net_device *netdev, + struct ieee_ets *ets) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_peer_getets(edev->cdev, ets); +} + +static const struct dcbnl_rtnl_ops qede_dcbnl_ops = { + .ieee_getpfc = qede_dcbnl_ieee_getpfc, + .ieee_setpfc = qede_dcbnl_ieee_setpfc, + .ieee_getets = qede_dcbnl_ieee_getets, + .ieee_setets = qede_dcbnl_ieee_setets, + .ieee_getapp = qede_dcbnl_ieee_getapp, + .ieee_setapp = qede_dcbnl_ieee_setapp, + .getdcbx = qede_dcbnl_getdcbx, + .ieee_peer_getpfc = qede_dcbnl_ieee_peer_getpfc, + .ieee_peer_getets = qede_dcbnl_ieee_peer_getets, + .getstate = qede_dcbnl_getstate, + .setstate = qede_dcbnl_setstate, + .getpermhwaddr = qede_dcbnl_getpermhwaddr, + .getpgtccfgtx = qede_dcbnl_getpgtccfgtx, + .getpgbwgcfgtx = qede_dcbnl_getpgbwgcfgtx, + .getpgtccfgrx = qede_dcbnl_getpgtccfgrx, + .getpgbwgcfgrx = qede_dcbnl_getpgbwgcfgrx, + .getpfccfg = qede_dcbnl_getpfccfg, + .setpfccfg = qede_dcbnl_setpfccfg, + .getcap = qede_dcbnl_getcap, + .getnumtcs = qede_dcbnl_getnumtcs, + .getpfcstate = qede_dcbnl_getpfcstate, + .getapp = qede_dcbnl_getapp, + .getdcbx = qede_dcbnl_getdcbx, + .setpgtccfgtx = qede_dcbnl_setpgtccfgtx, + .setpgtccfgrx = qede_dcbnl_setpgtccfgrx, + .setpgbwgcfgtx = qede_dcbnl_setpgbwgcfgtx, + .setpgbwgcfgrx = qede_dcbnl_setpgbwgcfgrx, + .setall = qede_dcbnl_setall, + .setnumtcs = qede_dcbnl_setnumtcs, + .setpfcstate = qede_dcbnl_setpfcstate, + .setapp = qede_dcbnl_setapp, + .setdcbx = qede_dcbnl_setdcbx, + .setfeatcfg = qede_dcbnl_setfeatcfg, + .getfeatcfg = qede_dcbnl_getfeatcfg, + .peer_getappinfo = qede_dcbnl_peer_getappinfo, + .peer_getapptable = qede_dcbnl_peer_getapptable, + .cee_peer_getpfc = qede_dcbnl_cee_peer_getpfc, + .cee_peer_getpg = qede_dcbnl_cee_peer_getpg, +}; + +void qede_set_dcbnl_ops(struct net_device *dev) +{ + dev->dcbnl_ops = &qede_dcbnl_ops; +} diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 6836d44b6b89..f8492cac9290 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -37,6 +37,7 @@ static const struct { } qede_rqstats_arr[] = { QEDE_RQSTAT(rx_hw_errors), QEDE_RQSTAT(rx_alloc_errors), + QEDE_RQSTAT(rx_ip_frags), }; #define QEDE_NUM_RQSTATS ARRAY_SIZE(qede_rqstats_arr) @@ -426,6 +427,59 @@ static u32 qede_get_link(struct net_device *dev) return current_link.link_up; } +static int qede_get_coalesce(struct net_device *dev, + struct ethtool_coalesce *coal) +{ + struct qede_dev *edev = netdev_priv(dev); + u16 rxc, txc; + + memset(coal, 0, sizeof(struct ethtool_coalesce)); + edev->ops->common->get_coalesce(edev->cdev, &rxc, &txc); + + coal->rx_coalesce_usecs = rxc; + coal->tx_coalesce_usecs = txc; + + return 0; +} + +static int qede_set_coalesce(struct net_device *dev, + struct ethtool_coalesce *coal) +{ + struct qede_dev *edev = netdev_priv(dev); + int i, rc = 0; + u16 rxc, txc; + u8 sb_id; + + if (!netif_running(dev)) { + DP_INFO(edev, "Interface is down\n"); + return -EINVAL; + } + + if (coal->rx_coalesce_usecs > QED_COALESCE_MAX || + coal->tx_coalesce_usecs > QED_COALESCE_MAX) { + DP_INFO(edev, + "Can't support requested %s coalesce value [max supported value %d]\n", + coal->rx_coalesce_usecs > QED_COALESCE_MAX ? "rx" + : "tx", + QED_COALESCE_MAX); + return -EINVAL; + } + + rxc = (u16)coal->rx_coalesce_usecs; + txc = (u16)coal->tx_coalesce_usecs; + for_each_rss(i) { + sb_id = edev->fp_array[i].sb_info->igu_sb_id; + rc = edev->ops->common->set_coalesce(edev->cdev, rxc, txc, + (u8)i, sb_id); + if (rc) { + DP_INFO(edev, "Set coalesce error, rc = %d\n", rc); + return rc; + } + } + + return rc; +} + static void qede_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) { @@ -1131,6 +1185,48 @@ static void qede_self_test(struct net_device *dev, } } +static int qede_set_tunable(struct net_device *dev, + const struct ethtool_tunable *tuna, + const void *data) +{ + struct qede_dev *edev = netdev_priv(dev); + u32 val; + + switch (tuna->id) { + case ETHTOOL_RX_COPYBREAK: + val = *(u32 *)data; + if (val < QEDE_MIN_PKT_LEN || val > QEDE_RX_HDR_SIZE) { + DP_VERBOSE(edev, QED_MSG_DEBUG, + "Invalid rx copy break value, range is [%u, %u]", + QEDE_MIN_PKT_LEN, QEDE_RX_HDR_SIZE); + return -EINVAL; + } + + edev->rx_copybreak = *(u32 *)data; + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int qede_get_tunable(struct net_device *dev, + const struct ethtool_tunable *tuna, void *data) +{ + struct qede_dev *edev = netdev_priv(dev); + + switch (tuna->id) { + case ETHTOOL_RX_COPYBREAK: + *(u32 *)data = edev->rx_copybreak; + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + static const struct ethtool_ops qede_ethtool_ops = { .get_settings = qede_get_settings, .set_settings = qede_set_settings, @@ -1139,6 +1235,8 @@ static const struct ethtool_ops qede_ethtool_ops = { .set_msglevel = qede_set_msglevel, .nway_reset = qede_nway_reset, .get_link = qede_get_link, + .get_coalesce = qede_get_coalesce, + .set_coalesce = qede_set_coalesce, .get_ringparam = qede_get_ringparam, .set_ringparam = qede_set_ringparam, .get_pauseparam = qede_get_pauseparam, @@ -1157,6 +1255,8 @@ static const struct ethtool_ops qede_ethtool_ops = { .get_channels = qede_get_channels, .set_channels = qede_set_channels, .self_test = qede_self_test, + .get_tunable = qede_get_tunable, + .set_tunable = qede_set_tunable, }; static const struct ethtool_ops qede_vf_ethtool_ops = { @@ -1179,6 +1279,8 @@ static const struct ethtool_ops qede_vf_ethtool_ops = { .set_rxfh = qede_set_rxfh, .get_channels = qede_get_channels, .set_channels = qede_set_channels, + .get_tunable = qede_get_tunable, + .set_tunable = qede_set_tunable, }; void qede_set_ethtool_ops(struct net_device *dev) diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index edd2d908c52e..91e7bb0b85c8 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -24,12 +24,7 @@ #include <linux/netdev_features.h> #include <linux/udp.h> #include <linux/tcp.h> -#ifdef CONFIG_QEDE_VXLAN -#include <net/vxlan.h> -#endif -#ifdef CONFIG_QEDE_GENEVE -#include <net/geneve.h> -#endif +#include <net/udp_tunnel.h> #include <linux/ip.h> #include <net/ipv6.h> #include <net/tcp.h> @@ -87,7 +82,9 @@ static const struct pci_device_id qede_pci_tbl[] = { {PCI_VDEVICE(QLOGIC, PCI_DEVICE_ID_57980S_100), QEDE_PRIVATE_PF}, {PCI_VDEVICE(QLOGIC, PCI_DEVICE_ID_57980S_50), QEDE_PRIVATE_PF}, {PCI_VDEVICE(QLOGIC, PCI_DEVICE_ID_57980S_25), QEDE_PRIVATE_PF}, +#ifdef CONFIG_QED_SRIOV {PCI_VDEVICE(QLOGIC, PCI_DEVICE_ID_57980S_IOV), QEDE_PRIVATE_VF}, +#endif { 0 } }; @@ -488,6 +485,24 @@ static bool qede_pkt_req_lin(struct qede_dev *edev, struct sk_buff *skb, } #endif +static inline void qede_update_tx_producer(struct qede_tx_queue *txq) +{ + /* wmb makes sure that the BDs data is updated before updating the + * producer, otherwise FW may read old data from the BDs. + */ + wmb(); + barrier(); + writel(txq->tx_db.raw, txq->doorbell_addr); + + /* mmiowb is needed to synchronize doorbell writes from more than one + * processor. It guarantees that the write arrives to the device before + * the queue lock is released and another start_xmit is called (possibly + * on another CPU). Without this barrier, the next doorbell can bypass + * this doorbell. This is applicable to IA64/Altix systems. + */ + mmiowb(); +} + /* Main transmit function */ static netdev_tx_t qede_start_xmit(struct sk_buff *skb, @@ -546,6 +561,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) { DP_NOTICE(edev, "SKB mapping failed\n"); qede_free_failed_tx_pkt(edev, txq, first_bd, 0, false); + qede_update_tx_producer(txq); return NETDEV_TX_OK; } nbd++; @@ -660,6 +676,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (rc) { qede_free_failed_tx_pkt(edev, txq, first_bd, nbd, data_split); + qede_update_tx_producer(txq); return NETDEV_TX_OK; } @@ -684,6 +701,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (rc) { qede_free_failed_tx_pkt(edev, txq, first_bd, nbd, data_split); + qede_update_tx_producer(txq); return NETDEV_TX_OK; } } @@ -704,20 +722,8 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, txq->tx_db.data.bd_prod = cpu_to_le16(qed_chain_get_prod_idx(&txq->tx_pbl)); - /* wmb makes sure that the BDs data is updated before updating the - * producer, otherwise FW may read old data from the BDs. - */ - wmb(); - barrier(); - writel(txq->tx_db.raw, txq->doorbell_addr); - - /* mmiowb is needed to synchronize doorbell writes from more than one - * processor. It guarantees that the write arrives to the device before - * the queue lock is released and another start_xmit is called (possibly - * on another CPU). Without this barrier, the next doorbell can bypass - * this doorbell. This is applicable to IA64/Altix systems. - */ - mmiowb(); + if (!skb->xmit_more || netif_tx_queue_stopped(netdev_txq)) + qede_update_tx_producer(txq); if (unlikely(qed_chain_get_elem_left(&txq->tx_pbl) < (MAX_SKB_FRAGS + 1))) { @@ -1351,6 +1357,20 @@ static u8 qede_check_csum(u16 flag) return qede_check_tunn_csum(flag); } +static bool qede_pkt_is_ip_fragmented(struct eth_fast_path_rx_reg_cqe *cqe, + u16 flag) +{ + u8 tun_pars_flg = cqe->tunnel_pars_flags.flags; + + if ((tun_pars_flg & (ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK << + ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT)) || + (flag & (PARSING_AND_ERR_FLAGS_IPV4FRAG_MASK << + PARSING_AND_ERR_FLAGS_IPV4FRAG_SHIFT))) + return true; + + return false; +} + static int qede_rx_int(struct qede_fastpath *fp, int budget) { struct qede_dev *edev = fp->edev; @@ -1429,6 +1449,12 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget) csum_flag = qede_check_csum(parse_flag); if (unlikely(csum_flag == QEDE_CSUM_ERROR)) { + if (qede_pkt_is_ip_fragmented(&cqe->fast_path_regular, + parse_flag)) { + rxq->rx_ip_frags++; + goto alloc_skb; + } + DP_NOTICE(edev, "CQE in CONS = %u has error, flags = %x, dropping incoming packet\n", sw_comp_cons, parse_flag); @@ -1437,6 +1463,7 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget) goto next_cqe; } +alloc_skb: skb = netdev_alloc_skb(edev->ndev, QEDE_RX_HDR_SIZE); if (unlikely(!skb)) { DP_NOTICE(edev, @@ -1447,7 +1474,7 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget) } /* Copy data into SKB */ - if (len + pad <= QEDE_RX_HDR_SIZE) { + if (len + pad <= edev->rx_copybreak) { memcpy(skb_put(skb, len), page_address(data) + pad + sw_rx_data->page_offset, len); @@ -1579,56 +1606,49 @@ next_cqe: /* don't consume bd rx buffer */ static int qede_poll(struct napi_struct *napi, int budget) { - int work_done = 0; struct qede_fastpath *fp = container_of(napi, struct qede_fastpath, - napi); + napi); struct qede_dev *edev = fp->edev; + int rx_work_done = 0; + u8 tc; - while (1) { - u8 tc; - - for (tc = 0; tc < edev->num_tc; tc++) - if (qede_txq_has_work(&fp->txqs[tc])) - qede_tx_int(edev, &fp->txqs[tc]); - - if (qede_has_rx_work(fp->rxq)) { - work_done += qede_rx_int(fp, budget - work_done); - - /* must not complete if we consumed full budget */ - if (work_done >= budget) - break; - } + for (tc = 0; tc < edev->num_tc; tc++) + if (qede_txq_has_work(&fp->txqs[tc])) + qede_tx_int(edev, &fp->txqs[tc]); + + rx_work_done = qede_has_rx_work(fp->rxq) ? + qede_rx_int(fp, budget) : 0; + if (rx_work_done < budget) { + qed_sb_update_sb_idx(fp->sb_info); + /* *_has_*_work() reads the status block, + * thus we need to ensure that status block indices + * have been actually read (qed_sb_update_sb_idx) + * prior to this check (*_has_*_work) so that + * we won't write the "newer" value of the status block + * to HW (if there was a DMA right after + * qede_has_rx_work and if there is no rmb, the memory + * reading (qed_sb_update_sb_idx) may be postponed + * to right before *_ack_sb). In this case there + * will never be another interrupt until there is + * another update of the status block, while there + * is still unhandled work. + */ + rmb(); /* Fall out from the NAPI loop if needed */ - if (!(qede_has_rx_work(fp->rxq) || qede_has_tx_work(fp))) { - qed_sb_update_sb_idx(fp->sb_info); - /* *_has_*_work() reads the status block, - * thus we need to ensure that status block indices - * have been actually read (qed_sb_update_sb_idx) - * prior to this check (*_has_*_work) so that - * we won't write the "newer" value of the status block - * to HW (if there was a DMA right after - * qede_has_rx_work and if there is no rmb, the memory - * reading (qed_sb_update_sb_idx) may be postponed - * to right before *_ack_sb). In this case there - * will never be another interrupt until there is - * another update of the status block, while there - * is still unhandled work. - */ - rmb(); - - if (!(qede_has_rx_work(fp->rxq) || - qede_has_tx_work(fp))) { - napi_complete(napi); - /* Update and reenable interrupts */ - qed_sb_ack(fp->sb_info, IGU_INT_ENABLE, - 1 /*update*/); - break; - } + if (!(qede_has_rx_work(fp->rxq) || + qede_has_tx_work(fp))) { + napi_complete(napi); + + /* Update and reenable interrupts */ + qed_sb_ack(fp->sb_info, IGU_INT_ENABLE, + 1 /*update*/); + } else { + rx_work_done = budget; } } - return work_done; + return rx_work_done; } static irqreturn_t qede_msix_fp_int(int irq, void *fp_cookie) @@ -2110,75 +2130,75 @@ int qede_set_features(struct net_device *dev, netdev_features_t features) return 0; } -#ifdef CONFIG_QEDE_VXLAN -static void qede_add_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) +static void qede_udp_tunnel_add(struct net_device *dev, + struct udp_tunnel_info *ti) { struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - if (edev->vxlan_dst_port) - return; + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (edev->vxlan_dst_port) + return; - edev->vxlan_dst_port = t_port; + edev->vxlan_dst_port = t_port; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Added vxlan port=%d", t_port); + DP_VERBOSE(edev, QED_MSG_DEBUG, "Added vxlan port=%d", + t_port); - set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); - schedule_delayed_work(&edev->sp_task, 0); -} + set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (edev->geneve_dst_port) + return; -static void qede_del_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) -{ - struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + edev->geneve_dst_port = t_port; - if (t_port != edev->vxlan_dst_port) + DP_VERBOSE(edev, QED_MSG_DEBUG, "Added geneve port=%d", + t_port); + set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); + break; + default: return; + } - edev->vxlan_dst_port = 0; - - DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted vxlan port=%d", t_port); - - set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); schedule_delayed_work(&edev->sp_task, 0); } -#endif -#ifdef CONFIG_QEDE_GENEVE -static void qede_add_geneve_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) +static void qede_udp_tunnel_del(struct net_device *dev, + struct udp_tunnel_info *ti) { struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - if (edev->geneve_dst_port) - return; + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (t_port != edev->vxlan_dst_port) + return; - edev->geneve_dst_port = t_port; + edev->vxlan_dst_port = 0; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Added geneve port=%d", t_port); - set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); - schedule_delayed_work(&edev->sp_task, 0); -} + DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted vxlan port=%d", + t_port); -static void qede_del_geneve_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) -{ - struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (t_port != edev->geneve_dst_port) + return; - if (t_port != edev->geneve_dst_port) - return; + edev->geneve_dst_port = 0; - edev->geneve_dst_port = 0; + DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted geneve port=%d", + t_port); + set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); + break; + default: + return; + } - DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted geneve port=%d", t_port); - set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); schedule_delayed_work(&edev->sp_task, 0); } -#endif static const struct net_device_ops qede_netdev_ops = { .ndo_open = qede_open, @@ -2202,14 +2222,8 @@ static const struct net_device_ops qede_netdev_ops = { .ndo_get_vf_config = qede_get_vf_config, .ndo_set_vf_rate = qede_set_vf_rate, #endif -#ifdef CONFIG_QEDE_VXLAN - .ndo_add_vxlan_port = qede_add_vxlan_port, - .ndo_del_vxlan_port = qede_del_vxlan_port, -#endif -#ifdef CONFIG_QEDE_GENEVE - .ndo_add_geneve_port = qede_add_geneve_port, - .ndo_del_geneve_port = qede_del_geneve_port, -#endif + .ndo_udp_tunnel_add = qede_udp_tunnel_add, + .ndo_udp_tunnel_del = qede_udp_tunnel_del, }; /* ------------------------------------------------------------------------- @@ -2499,8 +2513,13 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, edev->ops->register_ops(cdev, &qede_ll_ops, edev); +#ifdef CONFIG_DCB + qede_set_dcbnl_ops(edev->ndev); +#endif + INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); mutex_init(&edev->qede_lock); + edev->rx_copybreak = QEDE_RX_HDR_SIZE; DP_INFO(edev, "Ending successfully qede probe\n"); @@ -3227,7 +3246,7 @@ static int qede_stop_queues(struct qede_dev *edev) return rc; } -static int qede_start_queues(struct qede_dev *edev) +static int qede_start_queues(struct qede_dev *edev, bool clear_stats) { int rc, tc, i; int vlan_removal_en = 1; @@ -3458,6 +3477,7 @@ out: enum qede_load_mode { QEDE_LOAD_NORMAL, + QEDE_LOAD_RELOAD, }; static int qede_load(struct qede_dev *edev, enum qede_load_mode mode) @@ -3496,7 +3516,7 @@ static int qede_load(struct qede_dev *edev, enum qede_load_mode mode) goto err3; DP_INFO(edev, "Setup IRQs succeeded\n"); - rc = qede_start_queues(edev); + rc = qede_start_queues(edev, mode != QEDE_LOAD_RELOAD); if (rc) goto err4; DP_INFO(edev, "Start VPORT, RXQ and TXQ succeeded\n"); @@ -3551,7 +3571,7 @@ void qede_reload(struct qede_dev *edev, if (func) func(edev, args); - qede_load(edev, QEDE_LOAD_NORMAL); + qede_load(edev, QEDE_LOAD_RELOAD); mutex_lock(&edev->qede_lock); qede_config_rx_mode(edev->ndev); @@ -3573,12 +3593,8 @@ static int qede_open(struct net_device *ndev) if (rc) return rc; -#ifdef CONFIG_QEDE_VXLAN - vxlan_get_rx_port(ndev); -#endif -#ifdef CONFIG_QEDE_GENEVE - geneve_get_rx_port(ndev); -#endif + udp_tunnel_get_rx_info(ndev); + return 0; } diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index caf6ddb7ea76..fd973f4f16c7 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -1026,10 +1026,8 @@ struct qlcnic_ipaddr { #define QLCNIC_HAS_PHYS_PORT_ID 0x40000 #define QLCNIC_TSS_RSS 0x80000 -#ifdef CONFIG_QLCNIC_VXLAN #define QLCNIC_ADD_VXLAN_PORT 0x100000 #define QLCNIC_DEL_VXLAN_PORT 0x200000 -#endif #define QLCNIC_VLAN_FILTERING 0x800000 diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c index f9640d5ce6ba..bdbcd2b088a0 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c @@ -2159,7 +2159,6 @@ int qlcnic_83xx_get_mac_address(struct qlcnic_adapter *adapter, u8 *mac, struct qlcnic_cmd_args cmd; u32 mac_low, mac_high; - function = 0; err = qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_MAC_ADDRESS); if (err) return err; diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c index bf892160dd5f..a496390b8632 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c @@ -1020,7 +1020,6 @@ static int qlcnic_83xx_idc_check_state_validity(struct qlcnic_adapter *adapter, return 0; } -#ifdef CONFIG_QLCNIC_VXLAN #define QLC_83XX_ENCAP_TYPE_VXLAN BIT_1 #define QLC_83XX_MATCH_ENCAP_ID BIT_2 #define QLC_83XX_SET_VXLAN_UDP_DPORT BIT_3 @@ -1089,14 +1088,12 @@ static int qlcnic_set_vxlan_parsing(struct qlcnic_adapter *adapter, return ret; } -#endif static void qlcnic_83xx_periodic_tasks(struct qlcnic_adapter *adapter) { if (adapter->fhash.fnum) qlcnic_prune_lb_filters(adapter); -#ifdef CONFIG_QLCNIC_VXLAN if (adapter->flags & QLCNIC_ADD_VXLAN_PORT) { if (qlcnic_set_vxlan_port(adapter)) return; @@ -1112,7 +1109,6 @@ static void qlcnic_83xx_periodic_tasks(struct qlcnic_adapter *adapter) adapter->ahw->vxlan_port = 0; adapter->flags &= ~QLCNIC_DEL_VXLAN_PORT; } -#endif } /** diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c index 7bd6f25b4625..607bb7d4514d 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c @@ -2220,7 +2220,7 @@ void qlcnic_83xx_process_rcv_ring_diag(struct qlcnic_host_sds_ring *sds_ring) if (!opcode) return; - ring = QLCNIC_FETCH_RING_ID(qlcnic_83xx_hndl(sts_data[0])); + ring = QLCNIC_FETCH_RING_ID(sts_data[0]); qlcnic_83xx_process_rcv_diag(adapter, ring, sts_data); desc = &sds_ring->desc_head[consumer]; desc->status_desc_data[0] = cpu_to_le64(STATUS_OWNER_PHANTOM); diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index 1c29105b6c36..3ebef27e0964 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -16,9 +16,7 @@ #include <linux/aer.h> #include <linux/log2.h> #include <linux/pci.h> -#ifdef CONFIG_QLCNIC_VXLAN #include <net/vxlan.h> -#endif #include "qlcnic.h" #include "qlcnic_sriov.h" @@ -474,13 +472,15 @@ static int qlcnic_get_phys_port_id(struct net_device *netdev, return 0; } -#ifdef CONFIG_QLCNIC_VXLAN static void qlcnic_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct qlcnic_adapter *adapter = netdev_priv(netdev); struct qlcnic_hardware_context *ahw = adapter->ahw; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + /* Adapter supports only one VXLAN port. Use very first port * for enabling offload */ @@ -488,23 +488,26 @@ static void qlcnic_add_vxlan_port(struct net_device *netdev, return; if (!ahw->vxlan_port_count) { ahw->vxlan_port_count = 1; - ahw->vxlan_port = ntohs(port); + ahw->vxlan_port = ntohs(ti->port); adapter->flags |= QLCNIC_ADD_VXLAN_PORT; return; } - if (ahw->vxlan_port == ntohs(port)) + if (ahw->vxlan_port == ntohs(ti->port)) ahw->vxlan_port_count++; } static void qlcnic_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct qlcnic_adapter *adapter = netdev_priv(netdev); struct qlcnic_hardware_context *ahw = adapter->ahw; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (!qlcnic_encap_rx_offload(adapter) || !ahw->vxlan_port_count || - (ahw->vxlan_port != ntohs(port))) + (ahw->vxlan_port != ntohs(ti->port))) return; ahw->vxlan_port_count--; @@ -519,7 +522,6 @@ static netdev_features_t qlcnic_features_check(struct sk_buff *skb, features = vlan_features_check(skb, features); return vxlan_features_check(skb, features); } -#endif static const struct net_device_ops qlcnic_netdev_ops = { .ndo_open = qlcnic_open, @@ -539,11 +541,9 @@ static const struct net_device_ops qlcnic_netdev_ops = { .ndo_fdb_del = qlcnic_fdb_del, .ndo_fdb_dump = qlcnic_fdb_dump, .ndo_get_phys_port_id = qlcnic_get_phys_port_id, -#ifdef CONFIG_QLCNIC_VXLAN - .ndo_add_vxlan_port = qlcnic_add_vxlan_port, - .ndo_del_vxlan_port = qlcnic_del_vxlan_port, + .ndo_udp_tunnel_add = qlcnic_add_vxlan_port, + .ndo_udp_tunnel_del = qlcnic_del_vxlan_port, .ndo_features_check = qlcnic_features_check, -#endif #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = qlcnic_poll_controller, #endif @@ -2015,10 +2015,8 @@ qlcnic_attach(struct qlcnic_adapter *adapter) qlcnic_create_sysfs_entries(adapter); -#ifdef CONFIG_QLCNIC_VXLAN if (qlcnic_encap_rx_offload(adapter)) - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); adapter->is_up = QLCNIC_ADAPTER_UP_MAGIC; return 0; diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 6b541e57c96a..cb29ee24cf1b 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -4,7 +4,7 @@ * Copyright (C) 2004 Sten Wang <sten.wang@rdc.com.tw> * Copyright (C) 2007 * Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us> - * Copyright (C) 2007-2012 Florian Fainelli <florian@openwrt.org> + * Copyright (C) 2007-2012 Florian Fainelli <f.fainelli@gmail.com> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -48,8 +48,8 @@ #include <asm/processor.h> #define DRV_NAME "r6040" -#define DRV_VERSION "0.28" -#define DRV_RELDATE "07Oct2011" +#define DRV_VERSION "0.29" +#define DRV_RELDATE "04Jul2016" /* Time in jiffies before concluding the transmitter is hung. */ #define TX_TIMEOUT (6000 * HZ / 1000) @@ -162,7 +162,7 @@ MODULE_AUTHOR("Sten Wang <sten.wang@rdc.com.tw>," "Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>," - "Florian Fainelli <florian@openwrt.org>"); + "Florian Fainelli <f.fainelli@gmail.com>"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("RDC R6040 NAPI PCI FastEthernet driver"); MODULE_VERSION(DRV_VERSION " " DRV_RELDATE); @@ -200,7 +200,6 @@ struct r6040_private { struct mii_bus *mii_bus; struct napi_struct napi; void __iomem *base; - struct phy_device *phydev; int old_link; int old_duplex; }; @@ -474,7 +473,7 @@ static void r6040_down(struct net_device *dev) iowrite16(adrp[1], ioaddr + MID_0M); iowrite16(adrp[2], ioaddr + MID_0H); - phy_stop(lp->phydev); + phy_stop(dev->phydev); } static int r6040_close(struct net_device *dev) @@ -515,12 +514,10 @@ static int r6040_close(struct net_device *dev) static int r6040_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct r6040_private *lp = netdev_priv(dev); - - if (!lp->phydev) + if (!dev->phydev) return -EINVAL; - return phy_mii_ioctl(lp->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static int r6040_rx(struct net_device *dev, int limit) @@ -617,10 +614,15 @@ static void r6040_tx(struct net_device *dev) if (descptr->status & DSC_OWNER_MAC) break; /* Not complete */ skb_ptr = descptr->skb_ptr; + + /* Statistic Counter */ + dev->stats.tx_packets++; + dev->stats.tx_bytes += skb_ptr->len; + pci_unmap_single(priv->pdev, le32_to_cpu(descptr->buf), skb_ptr->len, PCI_DMA_TODEVICE); /* Free buffer */ - dev_kfree_skb_irq(skb_ptr); + dev_kfree_skb(skb_ptr); descptr->skb_ptr = NULL; /* To next descriptor */ descptr = descptr->vndescp; @@ -641,12 +643,15 @@ static int r6040_poll(struct napi_struct *napi, int budget) void __iomem *ioaddr = priv->base; int work_done; + r6040_tx(dev); + work_done = r6040_rx(dev, budget); if (work_done < budget) { - napi_complete(napi); - /* Enable RX interrupt */ - iowrite16(ioread16(ioaddr + MIER) | RX_INTS, ioaddr + MIER); + napi_complete_done(napi, work_done); + /* Enable RX/TX interrupt */ + iowrite16(ioread16(ioaddr + MIER) | RX_INTS | TX_INTS, + ioaddr + MIER); } return work_done; } @@ -673,7 +678,7 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id) } /* RX interrupt request */ - if (status & RX_INTS) { + if (status & (RX_INTS | TX_INTS)) { if (status & RX_NO_DESC) { /* RX descriptor unavailable */ dev->stats.rx_dropped++; @@ -684,15 +689,11 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id) if (likely(napi_schedule_prep(&lp->napi))) { /* Mask off RX interrupt */ - misr &= ~RX_INTS; - __napi_schedule(&lp->napi); + misr &= ~(RX_INTS | TX_INTS); + __napi_schedule_irqoff(&lp->napi); } } - /* TX interrupt request */ - if (status & TX_INTS) - r6040_tx(dev); - /* Restore RDC MAC interrupt */ iowrite16(misr, ioaddr + MIER); @@ -732,7 +733,7 @@ static int r6040_up(struct net_device *dev) /* Initialize all MAC registers */ r6040_init_mac_regs(dev); - phy_start(lp->phydev); + phy_start(dev->phydev); return 0; } @@ -813,6 +814,9 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, void __iomem *ioaddr = lp->base; unsigned long flags; + if (skb_put_padto(skb, ETH_ZLEN) < 0) + return NETDEV_TX_OK; + /* Critical Section */ spin_lock_irqsave(&lp->lock, flags); @@ -824,17 +828,10 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, return NETDEV_TX_BUSY; } - /* Statistic Counter */ - dev->stats.tx_packets++; - dev->stats.tx_bytes += skb->len; /* Set TX descriptor & Transmit it */ lp->tx_free_desc--; descptr = lp->tx_insert_ptr; - if (skb->len < ETH_ZLEN) - descptr->len = ETH_ZLEN; - else - descptr->len = skb->len; - + descptr->len = skb->len; descptr->skb_ptr = skb; descptr->buf = cpu_to_le32(pci_map_single(lp->pdev, skb->data, skb->len, PCI_DMA_TODEVICE)); @@ -843,7 +840,8 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, skb_tx_timestamp(skb); /* Trigger the MAC to check the TX descriptor */ - iowrite16(TM2TX, ioaddr + MTPR); + if (!skb->xmit_more || netif_queue_stopped(dev)) + iowrite16(TM2TX, ioaddr + MTPR); lp->tx_insert_ptr = descptr->vndescp; /* If no tx resource, stop */ @@ -957,26 +955,12 @@ static void netdev_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, pci_name(rp->pdev), sizeof(info->bus_info)); } -static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct r6040_private *rp = netdev_priv(dev); - - return phy_ethtool_gset(rp->phydev, cmd); -} - -static int netdev_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct r6040_private *rp = netdev_priv(dev); - - return phy_ethtool_sset(rp->phydev, cmd); -} - static const struct ethtool_ops netdev_ethtool_ops = { .get_drvinfo = netdev_get_drvinfo, - .get_settings = netdev_get_settings, - .set_settings = netdev_set_settings, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops r6040_netdev_ops = { @@ -998,7 +982,7 @@ static const struct net_device_ops r6040_netdev_ops = { static void r6040_adjust_link(struct net_device *dev) { struct r6040_private *lp = netdev_priv(dev); - struct phy_device *phydev = lp->phydev; + struct phy_device *phydev = dev->phydev; int status_changed = 0; void __iomem *ioaddr = lp->base; @@ -1018,14 +1002,8 @@ static void r6040_adjust_link(struct net_device *dev) lp->old_duplex = phydev->duplex; } - if (status_changed) { - pr_info("%s: link %s", dev->name, phydev->link ? - "UP" : "DOWN"); - if (phydev->link) - pr_cont(" - %d/%s", phydev->speed, - DUPLEX_FULL == phydev->duplex ? "full" : "half"); - pr_cont("\n"); - } + if (status_changed) + phy_print_status(phydev); } static int r6040_mii_probe(struct net_device *dev) @@ -1057,7 +1035,6 @@ static int r6040_mii_probe(struct net_device *dev) | SUPPORTED_TP); phydev->advertising = phydev->supported; - lp->phydev = phydev; lp->old_link = 0; lp->old_duplex = -1; diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c index 28b775e5a9ad..f0b09b05ed3f 100644 --- a/drivers/net/ethernet/rocker/rocker_main.c +++ b/drivers/net/ethernet/rocker/rocker_main.c @@ -1996,7 +1996,8 @@ static int rocker_port_change_proto_down(struct net_device *dev, return 0; } -static void rocker_port_neigh_destroy(struct neighbour *n) +static void rocker_port_neigh_destroy(struct net_device *dev, + struct neighbour *n) { struct rocker_port *rocker_port = netdev_priv(n->dev); int err; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h b/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h index 45019649bbbd..5cb51b609f02 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h @@ -475,7 +475,6 @@ struct sxgbe_priv_data { int rxcsum_insertion; spinlock_t stats_lock; /* lock for tx/rx statatics */ - struct phy_device *phydev; int oldlink; int speed; int oldduplex; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c index c0981ae45874..542b67d436df 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c @@ -147,7 +147,7 @@ static int sxgbe_get_eee(struct net_device *dev, edata->eee_active = priv->eee_active; edata->tx_lpi_timer = priv->tx_lpi_timer; - return phy_ethtool_get_eee(priv->phydev, edata); + return phy_ethtool_get_eee(dev->phydev, edata); } static int sxgbe_set_eee(struct net_device *dev, @@ -172,7 +172,7 @@ static int sxgbe_set_eee(struct net_device *dev, priv->tx_lpi_timer = edata->tx_lpi_timer; } - return phy_ethtool_set_eee(priv->phydev, edata); + return phy_ethtool_set_eee(dev->phydev, edata); } static void sxgbe_getdrvinfo(struct net_device *dev, @@ -182,27 +182,6 @@ static void sxgbe_getdrvinfo(struct net_device *dev, strlcpy(info->version, DRV_VERSION, sizeof(info->version)); } -static int sxgbe_getsettings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - struct sxgbe_priv_data *priv = netdev_priv(dev); - - if (priv->phydev) - return phy_ethtool_gset(priv->phydev, cmd); - - return -EOPNOTSUPP; -} - -static int sxgbe_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct sxgbe_priv_data *priv = netdev_priv(dev); - - if (priv->phydev) - return phy_ethtool_sset(priv->phydev, cmd); - - return -EOPNOTSUPP; -} - static u32 sxgbe_getmsglevel(struct net_device *dev) { struct sxgbe_priv_data *priv = netdev_priv(dev); @@ -255,7 +234,7 @@ static void sxgbe_get_ethtool_stats(struct net_device *dev, char *p; if (priv->eee_enabled) { - int val = phy_get_eee_err(priv->phydev); + int val = phy_get_eee_err(dev->phydev); if (val) priv->xstats.eee_wakeup_error_n = val; @@ -499,8 +478,6 @@ static int sxgbe_get_regs_len(struct net_device *dev) static const struct ethtool_ops sxgbe_ethtool_ops = { .get_drvinfo = sxgbe_getdrvinfo, - .get_settings = sxgbe_getsettings, - .set_settings = sxgbe_setsettings, .get_msglevel = sxgbe_getmsglevel, .set_msglevel = sxgbe_setmsglevel, .get_link = ethtool_op_get_link, @@ -516,6 +493,8 @@ static const struct ethtool_ops sxgbe_ethtool_ops = { .get_regs_len = sxgbe_get_regs_len, .get_eee = sxgbe_get_eee, .set_eee = sxgbe_set_eee, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; void sxgbe_set_ethtool_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c index 413ea14ab91f..ea44a2456ce1 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c @@ -124,12 +124,13 @@ static void sxgbe_eee_ctrl_timer(unsigned long arg) */ bool sxgbe_eee_init(struct sxgbe_priv_data * const priv) { + struct net_device *ndev = priv->dev; bool ret = false; /* MAC core supports the EEE feature. */ if (priv->hw_cap.eee) { /* Check if the PHY supports EEE */ - if (phy_init_eee(priv->phydev, 1)) + if (phy_init_eee(ndev->phydev, 1)) return false; priv->eee_active = 1; @@ -152,12 +153,14 @@ bool sxgbe_eee_init(struct sxgbe_priv_data * const priv) static void sxgbe_eee_adjust(const struct sxgbe_priv_data *priv) { + struct net_device *ndev = priv->dev; + /* When the EEE has been already initialised we have to * modify the PLS bit in the LPI ctrl & status reg according * to the PHY link status. For this reason. */ if (priv->eee_enabled) - priv->hw->mac->set_eee_pls(priv->ioaddr, priv->phydev->link); + priv->hw->mac->set_eee_pls(priv->ioaddr, ndev->phydev->link); } /** @@ -203,7 +206,7 @@ static inline u32 sxgbe_tx_avail(struct sxgbe_tx_queue *queue, int tx_qsize) static void sxgbe_adjust_link(struct net_device *dev) { struct sxgbe_priv_data *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; u8 new_state = 0; u8 speed = 0xff; @@ -306,9 +309,6 @@ static int sxgbe_init_phy(struct net_device *ndev) netdev_dbg(ndev, "%s: attached to PHY (UID 0x%x) Link = %d\n", __func__, phydev->phy_id, phydev->link); - /* save phy device in private structure */ - priv->phydev = phydev; - return 0; } @@ -1173,8 +1173,8 @@ static int sxgbe_open(struct net_device *dev) priv->hw->dma->start_tx(priv->ioaddr, SXGBE_TX_QUEUES); priv->hw->dma->start_rx(priv->ioaddr, SXGBE_RX_QUEUES); - if (priv->phydev) - phy_start(priv->phydev); + if (dev->phydev) + phy_start(dev->phydev); /* initialise TX coalesce parameters */ sxgbe_tx_init_coalesce(priv); @@ -1194,8 +1194,8 @@ static int sxgbe_open(struct net_device *dev) init_error: free_dma_desc_resources(priv); - if (priv->phydev) - phy_disconnect(priv->phydev); + if (dev->phydev) + phy_disconnect(dev->phydev); phy_error: clk_disable_unprepare(priv->sxgbe_clk); @@ -1216,10 +1216,9 @@ static int sxgbe_release(struct net_device *dev) del_timer_sync(&priv->eee_ctrl_timer); /* Stop and disconnect the PHY */ - if (priv->phydev) { - phy_stop(priv->phydev); - phy_disconnect(priv->phydev); - priv->phydev = NULL; + if (dev->phydev) { + phy_stop(dev->phydev); + phy_disconnect(dev->phydev); } netif_tx_stop_all_queues(dev); @@ -1969,7 +1968,6 @@ static void sxgbe_poll_controller(struct net_device *dev) */ static int sxgbe_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct sxgbe_priv_data *priv = netdev_priv(dev); int ret = -EOPNOTSUPP; if (!netif_running(dev)) @@ -1979,9 +1977,9 @@ static int sxgbe_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: - if (!priv->phydev) + if (!dev->phydev) return -EINVAL; - ret = phy_mii_ioctl(priv->phydev, rq, cmd); + ret = phy_mii_ioctl(dev->phydev, rq, cmd); break; default: break; diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 1f309127457d..f658fee74f18 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -50,14 +50,34 @@ enum { #define HUNT_FILTER_TBL_ROWS 8192 #define EFX_EF10_FILTER_ID_INVALID 0xffff + +#define EFX_EF10_FILTER_DEV_UC_MAX 32 +#define EFX_EF10_FILTER_DEV_MC_MAX 256 + +/* VLAN list entry */ +struct efx_ef10_vlan { + struct list_head list; + u16 vid; +}; + +/* Per-VLAN filters information */ +struct efx_ef10_filter_vlan { + struct list_head list; + u16 vid; + u16 uc[EFX_EF10_FILTER_DEV_UC_MAX]; + u16 mc[EFX_EF10_FILTER_DEV_MC_MAX]; + u16 ucdef; + u16 bcast; + u16 mcdef; +}; + struct efx_ef10_dev_addr { u8 addr[ETH_ALEN]; - u16 id; }; struct efx_ef10_filter_table { -/* The RX match field masks supported by this fw & hw, in order of priority */ - enum efx_filter_match_flags rx_match_flags[ +/* The MCDI match masks supported by this fw & hw, in order of priority */ + u32 rx_match_mcdi_flags[ MC_CMD_GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES_MAXNUM]; unsigned int rx_match_count; @@ -73,16 +93,16 @@ struct efx_ef10_filter_table { } *entry; wait_queue_head_t waitq; /* Shadow of net_device address lists, guarded by mac_lock */ -#define EFX_EF10_FILTER_DEV_UC_MAX 32 -#define EFX_EF10_FILTER_DEV_MC_MAX 256 struct efx_ef10_dev_addr dev_uc_list[EFX_EF10_FILTER_DEV_UC_MAX]; struct efx_ef10_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX]; int dev_uc_count; int dev_mc_count; -/* Indices (like efx_ef10_dev_addr.id) for promisc/allmulti filters */ - u16 ucdef_id; - u16 bcast_id; - u16 mcdef_id; + bool uc_promisc; + bool mc_promisc; +/* Whether in multicast promiscuous mode when last changed */ + bool mc_promisc_last; + bool vlan_filter; + struct list_head vlan_list; }; /* An arbitrary search limit for the software hash table */ @@ -90,6 +110,10 @@ struct efx_ef10_filter_table { static void efx_ef10_rx_free_indir_table(struct efx_nic *efx); static void efx_ef10_filter_table_remove(struct efx_nic *efx); +static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid); +static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan); +static void efx_ef10_filter_del_vlan(struct efx_nic *efx, u16 vid); static int efx_ef10_get_warm_boot_count(struct efx_nic *efx) { @@ -275,6 +299,131 @@ static ssize_t efx_ef10_show_primary_flag(struct device *dev, ? 1 : 0); } +static struct efx_ef10_vlan *efx_ef10_find_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan; + + WARN_ON(!mutex_is_locked(&nic_data->vlan_lock)); + + list_for_each_entry(vlan, &nic_data->vlan_list, list) { + if (vlan->vid == vid) + return vlan; + } + + return NULL; +} + +static int efx_ef10_add_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan; + int rc; + + mutex_lock(&nic_data->vlan_lock); + + vlan = efx_ef10_find_vlan(efx, vid); + if (vlan) { + /* We add VID 0 on init. 8021q adds it on module init + * for all interfaces with VLAN filtring feature. + */ + if (vid == 0) + goto done_unlock; + netif_warn(efx, drv, efx->net_dev, + "VLAN %u already added\n", vid); + rc = -EALREADY; + goto fail_exist; + } + + rc = -ENOMEM; + vlan = kzalloc(sizeof(*vlan), GFP_KERNEL); + if (!vlan) + goto fail_alloc; + + vlan->vid = vid; + + list_add_tail(&vlan->list, &nic_data->vlan_list); + + if (efx->filter_state) { + mutex_lock(&efx->mac_lock); + down_write(&efx->filter_sem); + rc = efx_ef10_filter_add_vlan(efx, vlan->vid); + up_write(&efx->filter_sem); + mutex_unlock(&efx->mac_lock); + if (rc) + goto fail_filter_add_vlan; + } + +done_unlock: + mutex_unlock(&nic_data->vlan_lock); + return 0; + +fail_filter_add_vlan: + list_del(&vlan->list); + kfree(vlan); +fail_alloc: +fail_exist: + mutex_unlock(&nic_data->vlan_lock); + return rc; +} + +static void efx_ef10_del_vlan_internal(struct efx_nic *efx, + struct efx_ef10_vlan *vlan) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + WARN_ON(!mutex_is_locked(&nic_data->vlan_lock)); + + if (efx->filter_state) { + down_write(&efx->filter_sem); + efx_ef10_filter_del_vlan(efx, vlan->vid); + up_write(&efx->filter_sem); + } + + list_del(&vlan->list); + kfree(vlan); +} + +static int efx_ef10_del_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan; + int rc = 0; + + /* 8021q removes VID 0 on module unload for all interfaces + * with VLAN filtering feature. We need to keep it to receive + * untagged traffic. + */ + if (vid == 0) + return 0; + + mutex_lock(&nic_data->vlan_lock); + + vlan = efx_ef10_find_vlan(efx, vid); + if (!vlan) { + netif_err(efx, drv, efx->net_dev, + "VLAN %u to be deleted not found\n", vid); + rc = -ENOENT; + } else { + efx_ef10_del_vlan_internal(efx, vlan); + } + + mutex_unlock(&nic_data->vlan_lock); + + return rc; +} + +static void efx_ef10_cleanup_vlans(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan, *next_vlan; + + mutex_lock(&nic_data->vlan_lock); + list_for_each_entry_safe(vlan, next_vlan, &nic_data->vlan_list, list) + efx_ef10_del_vlan_internal(efx, vlan); + mutex_unlock(&nic_data->vlan_lock); +} + static DEVICE_ATTR(link_control_flag, 0444, efx_ef10_show_link_control_flag, NULL); static DEVICE_ATTR(primary_flag, 0444, efx_ef10_show_primary_flag, NULL); @@ -421,8 +570,30 @@ static int efx_ef10_probe(struct efx_nic *efx) #endif ether_addr_copy(nic_data->port_id, efx->net_dev->perm_addr); + INIT_LIST_HEAD(&nic_data->vlan_list); + mutex_init(&nic_data->vlan_lock); + + /* Add unspecified VID to support VLAN filtering being disabled */ + rc = efx_ef10_add_vlan(efx, EFX_FILTER_VID_UNSPEC); + if (rc) + goto fail_add_vid_unspec; + + /* If VLAN filtering is enabled, we need VID 0 to get untagged + * traffic. It is added automatically if 8021q module is loaded, + * but we can't rely on it since module may be not loaded. + */ + rc = efx_ef10_add_vlan(efx, 0); + if (rc) + goto fail_add_vid_0; + return 0; +fail_add_vid_0: + efx_ef10_cleanup_vlans(efx); +fail_add_vid_unspec: + mutex_destroy(&nic_data->vlan_lock); + efx_ptp_remove(efx); + efx_mcdi_mon_remove(efx); fail5: device_remove_file(&efx->pci_dev->dev, &dev_attr_primary_flag); fail4: @@ -676,6 +847,9 @@ static void efx_ef10_remove(struct efx_nic *efx) } #endif + efx_ef10_cleanup_vlans(efx); + mutex_destroy(&nic_data->vlan_lock); + efx_ptp_remove(efx); efx_mcdi_mon_remove(efx); @@ -704,6 +878,45 @@ static int efx_ef10_probe_pf(struct efx_nic *efx) return efx_ef10_probe(efx); } +int efx_ef10_vadaptor_query(struct efx_nic *efx, unsigned int port_id, + u32 *port_flags, u32 *vadaptor_flags, + unsigned int *vlan_tags) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_QUERY_IN_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_VADAPTOR_QUERY_OUT_LEN); + size_t outlen; + int rc; + + if (nic_data->datapath_caps & + (1 << MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_QUERY_LBN)) { + MCDI_SET_DWORD(inbuf, VADAPTOR_QUERY_IN_UPSTREAM_PORT_ID, + port_id); + + rc = efx_mcdi_rpc(efx, MC_CMD_VADAPTOR_QUERY, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), &outlen); + if (rc) + return rc; + + if (outlen < sizeof(outbuf)) { + rc = -EIO; + return rc; + } + } + + if (port_flags) + *port_flags = MCDI_DWORD(outbuf, VADAPTOR_QUERY_OUT_PORT_FLAGS); + if (vadaptor_flags) + *vadaptor_flags = + MCDI_DWORD(outbuf, VADAPTOR_QUERY_OUT_VADAPTOR_FLAGS); + if (vlan_tags) + *vlan_tags = + MCDI_DWORD(outbuf, + VADAPTOR_QUERY_OUT_NUM_AVAILABLE_VLAN_TAGS); + + return 0; +} + int efx_ef10_vadaptor_alloc(struct efx_nic *efx, unsigned int port_id) { MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_ALLOC_IN_LEN); @@ -3040,15 +3253,55 @@ static int efx_ef10_filter_push(struct efx_nic *efx, return rc; } -static int efx_ef10_filter_rx_match_pri(struct efx_ef10_filter_table *table, - enum efx_filter_match_flags match_flags) +static u32 efx_ef10_filter_mcdi_flags_from_spec(const struct efx_filter_spec *spec) { + unsigned int match_flags = spec->match_flags; + u32 mcdi_flags = 0; + + if (match_flags & EFX_FILTER_MATCH_LOC_MAC_IG) { + match_flags &= ~EFX_FILTER_MATCH_LOC_MAC_IG; + mcdi_flags |= + is_multicast_ether_addr(spec->loc_mac) ? + (1 << MC_CMD_FILTER_OP_IN_MATCH_UNKNOWN_MCAST_DST_LBN) : + (1 << MC_CMD_FILTER_OP_IN_MATCH_UNKNOWN_UCAST_DST_LBN); + } + +#define MAP_FILTER_TO_MCDI_FLAG(gen_flag, mcdi_field) { \ + unsigned int old_match_flags = match_flags; \ + match_flags &= ~EFX_FILTER_MATCH_ ## gen_flag; \ + if (match_flags != old_match_flags) \ + mcdi_flags |= \ + (1 << MC_CMD_FILTER_OP_IN_MATCH_ ## \ + mcdi_field ## _LBN); \ + } + MAP_FILTER_TO_MCDI_FLAG(REM_HOST, SRC_IP); + MAP_FILTER_TO_MCDI_FLAG(LOC_HOST, DST_IP); + MAP_FILTER_TO_MCDI_FLAG(REM_MAC, SRC_MAC); + MAP_FILTER_TO_MCDI_FLAG(REM_PORT, SRC_PORT); + MAP_FILTER_TO_MCDI_FLAG(LOC_MAC, DST_MAC); + MAP_FILTER_TO_MCDI_FLAG(LOC_PORT, DST_PORT); + MAP_FILTER_TO_MCDI_FLAG(ETHER_TYPE, ETHER_TYPE); + MAP_FILTER_TO_MCDI_FLAG(INNER_VID, INNER_VLAN); + MAP_FILTER_TO_MCDI_FLAG(OUTER_VID, OUTER_VLAN); + MAP_FILTER_TO_MCDI_FLAG(IP_PROTO, IP_PROTO); +#undef MAP_FILTER_TO_MCDI_FLAG + + /* Did we map them all? */ + WARN_ON_ONCE(match_flags); + + return mcdi_flags; +} + +static int efx_ef10_filter_pri(struct efx_ef10_filter_table *table, + const struct efx_filter_spec *spec) +{ + u32 mcdi_flags = efx_ef10_filter_mcdi_flags_from_spec(spec); unsigned int match_pri; for (match_pri = 0; match_pri < table->rx_match_count; match_pri++) - if (table->rx_match_flags[match_pri] == match_flags) + if (table->rx_match_mcdi_flags[match_pri] == mcdi_flags) return match_pri; return -EPROTONOSUPPORT; @@ -3074,7 +3327,7 @@ static s32 efx_ef10_filter_insert(struct efx_nic *efx, EFX_FILTER_FLAG_RX) return -EINVAL; - rc = efx_ef10_filter_rx_match_pri(table, spec->match_flags); + rc = efx_ef10_filter_pri(table, spec); if (rc < 0) return rc; match_pri = rc; @@ -3313,7 +3566,7 @@ static int efx_ef10_filter_remove_internal(struct efx_nic *efx, spec = efx_ef10_filter_entry_spec(table, filter_idx); if (!spec || (!by_index && - efx_ef10_filter_rx_match_pri(table, spec->match_flags) != + efx_ef10_filter_pri(table, spec) != filter_id / HUNT_FILTER_TBL_ROWS)) { rc = -ENOENT; goto out_unlock; @@ -3394,12 +3647,13 @@ static u32 efx_ef10_filter_get_unsafe_id(struct efx_nic *efx, u32 filter_id) return filter_id % HUNT_FILTER_TBL_ROWS; } -static int efx_ef10_filter_remove_unsafe(struct efx_nic *efx, - enum efx_filter_priority priority, - u32 filter_id) +static void efx_ef10_filter_remove_unsafe(struct efx_nic *efx, + enum efx_filter_priority priority, + u32 filter_id) { - return efx_ef10_filter_remove_internal(efx, 1U << priority, - filter_id, true); + if (filter_id == EFX_EF10_FILTER_ID_INVALID) + return; + efx_ef10_filter_remove_internal(efx, 1U << priority, filter_id, true); } static int efx_ef10_filter_get_safe(struct efx_nic *efx, @@ -3414,7 +3668,7 @@ static int efx_ef10_filter_get_safe(struct efx_nic *efx, spin_lock_bh(&efx->filter_lock); saved_spec = efx_ef10_filter_entry_spec(table, filter_idx); if (saved_spec && saved_spec->priority == priority && - efx_ef10_filter_rx_match_pri(table, saved_spec->match_flags) == + efx_ef10_filter_pri(table, saved_spec) == filter_id / HUNT_FILTER_TBL_ROWS) { *spec = *saved_spec; rc = 0; @@ -3487,8 +3741,7 @@ static s32 efx_ef10_filter_get_rx_ids(struct efx_nic *efx, count = -EMSGSIZE; break; } - buf[count++] = (efx_ef10_filter_rx_match_pri( - table, spec->match_flags) * + buf[count++] = (efx_ef10_filter_pri(table, spec) * HUNT_FILTER_TBL_ROWS + filter_idx); } @@ -3724,15 +3977,58 @@ static int efx_ef10_filter_match_flags_from_mcdi(u32 mcdi_flags) return match_flags; } +static void efx_ef10_filter_cleanup_vlans(struct efx_nic *efx) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan, *next_vlan; + + /* See comment in efx_ef10_filter_table_remove() */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + + if (!table) + return; + + list_for_each_entry_safe(vlan, next_vlan, &table->vlan_list, list) + efx_ef10_filter_del_vlan_internal(efx, vlan); +} + +static bool efx_ef10_filter_match_supported(struct efx_ef10_filter_table *table, + enum efx_filter_match_flags match_flags) +{ + unsigned int match_pri; + int mf; + + for (match_pri = 0; + match_pri < table->rx_match_count; + match_pri++) { + mf = efx_ef10_filter_match_flags_from_mcdi( + table->rx_match_mcdi_flags[match_pri]); + if (mf == match_flags) + return true; + } + + return false; +} + static int efx_ef10_filter_table_probe(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN); MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX); + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct net_device *net_dev = efx->net_dev; unsigned int pd_match_pri, pd_match_count; struct efx_ef10_filter_table *table; + struct efx_ef10_vlan *vlan; size_t outlen; int rc; + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return -EINVAL; + + if (efx->filter_state) /* already probed */ + return 0; + table = kzalloc(sizeof(*table), GFP_KERNEL); if (!table) return -ENOMEM; @@ -3765,24 +4061,48 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) "%s: fw flags %#x pri %u supported as driver flags %#x pri %u\n", __func__, mcdi_flags, pd_match_pri, rc, table->rx_match_count); - table->rx_match_flags[table->rx_match_count++] = rc; + table->rx_match_mcdi_flags[table->rx_match_count] = mcdi_flags; + table->rx_match_count++; } } + if ((efx_supported_features(efx) & NETIF_F_HW_VLAN_CTAG_FILTER) && + !(efx_ef10_filter_match_supported(table, + (EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC)) && + efx_ef10_filter_match_supported(table, + (EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC_IG)))) { + netif_info(efx, probe, net_dev, + "VLAN filters are not supported in this firmware variant\n"); + net_dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + efx->fixed_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + net_dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + } + table->entry = vzalloc(HUNT_FILTER_TBL_ROWS * sizeof(*table->entry)); if (!table->entry) { rc = -ENOMEM; goto fail; } - table->ucdef_id = EFX_EF10_FILTER_ID_INVALID; - table->bcast_id = EFX_EF10_FILTER_ID_INVALID; - table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; + table->mc_promisc_last = false; + table->vlan_filter = + !!(efx->net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER); + INIT_LIST_HEAD(&table->vlan_list); efx->filter_state = table; init_waitqueue_head(&table->waitq); + + list_for_each_entry(vlan, &nic_data->vlan_list, list) { + rc = efx_ef10_filter_add_vlan(efx, vlan->vid); + if (rc) + goto fail_add_vlan; + } + return 0; +fail_add_vlan: + efx_ef10_filter_cleanup_vlans(efx); + efx->filter_state = NULL; fail: kfree(table); return rc; @@ -3843,7 +4163,6 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx) nic_data->must_restore_filters = false; } -/* Caller must hold efx->filter_sem for write */ static void efx_ef10_filter_table_remove(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; @@ -3852,7 +4171,17 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) unsigned int filter_idx; int rc; + efx_ef10_filter_cleanup_vlans(efx); efx->filter_state = NULL; + /* If we were called without locking, then it's not safe to free + * the table as others might be using it. So we just WARN, leak + * the memory, and potentially get an inconsistent filter table + * state. + * This should never actually happen. + */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + if (!table) return; @@ -3880,37 +4209,54 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) kfree(table); } -#define EFX_EF10_FILTER_DO_MARK_OLD(id) \ - if (id != EFX_EF10_FILTER_ID_INVALID) { \ - filter_idx = efx_ef10_filter_get_unsafe_id(efx, id); \ - if (!table->entry[filter_idx].spec) \ - netif_dbg(efx, drv, efx->net_dev, \ - "%s: marked null spec old %04x:%04x\n", \ - __func__, id, filter_idx); \ - table->entry[filter_idx].spec |= EFX_EF10_FILTER_FLAG_AUTO_OLD;\ - } -static void efx_ef10_filter_mark_old(struct efx_nic *efx) +static void efx_ef10_filter_mark_one_old(struct efx_nic *efx, uint16_t *id) { struct efx_ef10_filter_table *table = efx->filter_state; - unsigned int filter_idx, i; + unsigned int filter_idx; - if (!table) - return; + if (*id != EFX_EF10_FILTER_ID_INVALID) { + filter_idx = efx_ef10_filter_get_unsafe_id(efx, *id); + if (!table->entry[filter_idx].spec) + netif_dbg(efx, drv, efx->net_dev, + "marked null spec old %04x:%04x\n", *id, + filter_idx); + table->entry[filter_idx].spec |= EFX_EF10_FILTER_FLAG_AUTO_OLD; + *id = EFX_EF10_FILTER_ID_INVALID; + } +} + +/* Mark old per-VLAN filters that may need to be removed */ +static void _efx_ef10_filter_vlan_mark_old(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + unsigned int i; - /* Mark old filters that may need to be removed */ - spin_lock_bh(&efx->filter_lock); for (i = 0; i < table->dev_uc_count; i++) - EFX_EF10_FILTER_DO_MARK_OLD(table->dev_uc_list[i].id); + efx_ef10_filter_mark_one_old(efx, &vlan->uc[i]); for (i = 0; i < table->dev_mc_count; i++) - EFX_EF10_FILTER_DO_MARK_OLD(table->dev_mc_list[i].id); - EFX_EF10_FILTER_DO_MARK_OLD(table->ucdef_id); - EFX_EF10_FILTER_DO_MARK_OLD(table->bcast_id); - EFX_EF10_FILTER_DO_MARK_OLD(table->mcdef_id); + efx_ef10_filter_mark_one_old(efx, &vlan->mc[i]); + efx_ef10_filter_mark_one_old(efx, &vlan->ucdef); + efx_ef10_filter_mark_one_old(efx, &vlan->bcast); + efx_ef10_filter_mark_one_old(efx, &vlan->mcdef); +} + +/* Mark old filters that may need to be removed. + * Caller must hold efx->filter_sem for read if race against + * efx_ef10_filter_table_remove() is possible + */ +static void efx_ef10_filter_mark_old(struct efx_nic *efx) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan; + + spin_lock_bh(&efx->filter_lock); + list_for_each_entry(vlan, &table->vlan_list, list) + _efx_ef10_filter_vlan_mark_old(efx, vlan); spin_unlock_bh(&efx->filter_lock); } -#undef EFX_EF10_FILTER_DO_MARK_OLD -static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) +static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; struct net_device *net_dev = efx->net_dev; @@ -3918,45 +4264,38 @@ static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) int addr_count; unsigned int i; - table->ucdef_id = EFX_EF10_FILTER_ID_INVALID; addr_count = netdev_uc_count(net_dev); - if (net_dev->flags & IFF_PROMISC) - *promisc = true; + table->uc_promisc = !!(net_dev->flags & IFF_PROMISC); table->dev_uc_count = 1 + addr_count; ether_addr_copy(table->dev_uc_list[0].addr, net_dev->dev_addr); i = 1; netdev_for_each_uc_addr(uc, net_dev) { if (i >= EFX_EF10_FILTER_DEV_UC_MAX) { - *promisc = true; + table->uc_promisc = true; break; } ether_addr_copy(table->dev_uc_list[i].addr, uc->addr); - table->dev_uc_list[i].id = EFX_EF10_FILTER_ID_INVALID; i++; } } -static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx, bool *promisc) +static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; struct net_device *net_dev = efx->net_dev; struct netdev_hw_addr *mc; unsigned int i, addr_count; - table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; - table->bcast_id = EFX_EF10_FILTER_ID_INVALID; - if (net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) - *promisc = true; + table->mc_promisc = !!(net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI)); addr_count = netdev_mc_count(net_dev); i = 0; netdev_for_each_mc_addr(mc, net_dev) { if (i >= EFX_EF10_FILTER_DEV_MC_MAX) { - *promisc = true; + table->mc_promisc = true; break; } ether_addr_copy(table->dev_mc_list[i].addr, mc->addr); - table->dev_mc_list[i].id = EFX_EF10_FILTER_ID_INVALID; i++; } @@ -3964,7 +4303,8 @@ static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx, bool *promisc) } static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, - bool multicast, bool rollback) + struct efx_ef10_filter_vlan *vlan, + bool multicast, bool rollback) { struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_dev_addr *addr_list; @@ -3973,14 +4313,17 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, u8 baddr[ETH_ALEN]; unsigned int i, j; int addr_count; + u16 *ids; int rc; if (multicast) { addr_list = table->dev_mc_list; addr_count = table->dev_mc_count; + ids = vlan->mc; } else { addr_list = table->dev_uc_list; addr_count = table->dev_uc_count; + ids = vlan->uc; } filter_flags = efx_rss_enabled(efx) ? EFX_FILTER_FLAG_RX_RSS : 0; @@ -3988,8 +4331,7 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, /* Insert/renew filters */ for (i = 0; i < addr_count; i++) { efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); - efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, - addr_list[i].addr); + efx_filter_set_eth_local(&spec, vlan->vid, addr_list[i].addr); rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { if (rollback) { @@ -3998,12 +4340,10 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, rc); /* Fall back to promiscuous */ for (j = 0; j < i; j++) { - if (addr_list[j].id == EFX_EF10_FILTER_ID_INVALID) - continue; efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, - addr_list[j].id); - addr_list[j].id = EFX_EF10_FILTER_ID_INVALID; + ids[j]); + ids[j] = EFX_EF10_FILTER_ID_INVALID; } return rc; } else { @@ -4011,40 +4351,40 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, rc = EFX_EF10_FILTER_ID_INVALID; } } - addr_list[i].id = efx_ef10_filter_get_unsafe_id(efx, rc); + ids[i] = efx_ef10_filter_get_unsafe_id(efx, rc); } if (multicast && rollback) { /* Also need an Ethernet broadcast filter */ efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); eth_broadcast_addr(baddr); - efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, baddr); + efx_filter_set_eth_local(&spec, vlan->vid, baddr); rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { netif_warn(efx, drv, efx->net_dev, "Broadcast filter insert failed rc=%d\n", rc); /* Fall back to promiscuous */ for (j = 0; j < i; j++) { - if (addr_list[j].id == EFX_EF10_FILTER_ID_INVALID) - continue; efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, - addr_list[j].id); - addr_list[j].id = EFX_EF10_FILTER_ID_INVALID; + ids[j]); + ids[j] = EFX_EF10_FILTER_ID_INVALID; } return rc; } else { - table->bcast_id = efx_ef10_filter_get_unsafe_id(efx, rc); + EFX_WARN_ON_PARANOID(vlan->bcast != + EFX_EF10_FILTER_ID_INVALID); + vlan->bcast = efx_ef10_filter_get_unsafe_id(efx, rc); } } return 0; } -static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, - bool rollback) +static int efx_ef10_filter_insert_def(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan, + bool multicast, bool rollback) { - struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; enum efx_filter_flags filter_flags; struct efx_filter_spec spec; @@ -4060,6 +4400,9 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, else efx_filter_set_uc_def(&spec); + if (vlan->vid != EFX_FILTER_VID_UNSPEC) + efx_filter_set_eth_local(&spec, vlan->vid, NULL); + rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { netif_printk(efx, drv, rc == -EPERM ? KERN_DEBUG : KERN_WARNING, @@ -4067,14 +4410,14 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, "%scast mismatch filter insert failed rc=%d\n", multicast ? "Multi" : "Uni", rc); } else if (multicast) { - table->mcdef_id = efx_ef10_filter_get_unsafe_id(efx, rc); + EFX_WARN_ON_PARANOID(vlan->mcdef != EFX_EF10_FILTER_ID_INVALID); + vlan->mcdef = efx_ef10_filter_get_unsafe_id(efx, rc); if (!nic_data->workaround_26807) { /* Also need an Ethernet broadcast filter */ efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); eth_broadcast_addr(baddr); - efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, - baddr); + efx_filter_set_eth_local(&spec, vlan->vid, baddr); rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { netif_warn(efx, drv, efx->net_dev, @@ -4084,17 +4427,20 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, /* Roll back the mc_def filter */ efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, - table->mcdef_id); - table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; + vlan->mcdef); + vlan->mcdef = EFX_EF10_FILTER_ID_INVALID; return rc; } } else { - table->bcast_id = efx_ef10_filter_get_unsafe_id(efx, rc); + EFX_WARN_ON_PARANOID(vlan->bcast != + EFX_EF10_FILTER_ID_INVALID); + vlan->bcast = efx_ef10_filter_get_unsafe_id(efx, rc); } } rc = 0; } else { - table->ucdef_id = rc; + EFX_WARN_ON_PARANOID(vlan->ucdef != EFX_EF10_FILTER_ID_INVALID); + vlan->ucdef = rc; rc = 0; } return rc; @@ -4203,64 +4549,55 @@ reset_nic: /* Caller must hold efx->filter_sem for read if race against * efx_ef10_filter_table_remove() is possible */ -static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) +static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan) { struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; - struct net_device *net_dev = efx->net_dev; - bool uc_promisc = false, mc_promisc = false; - if (!efx_dev_registered(efx)) - return; - - if (!table) - return; - - efx_ef10_filter_mark_old(efx); - - /* Copy/convert the address lists; add the primary station - * address and broadcast address + /* Do not install unspecified VID if VLAN filtering is enabled. + * Do not install all specified VIDs if VLAN filtering is disabled. */ - netif_addr_lock_bh(net_dev); - efx_ef10_filter_uc_addr_list(efx, &uc_promisc); - efx_ef10_filter_mc_addr_list(efx, &mc_promisc); - netif_addr_unlock_bh(net_dev); + if ((vlan->vid == EFX_FILTER_VID_UNSPEC) == table->vlan_filter) + return; /* Insert/renew unicast filters */ - if (uc_promisc) { - efx_ef10_filter_insert_def(efx, false, false); - efx_ef10_filter_insert_addr_list(efx, false, false); + if (table->uc_promisc) { + efx_ef10_filter_insert_def(efx, vlan, false, false); + efx_ef10_filter_insert_addr_list(efx, vlan, false, false); } else { /* If any of the filters failed to insert, fall back to * promiscuous mode - add in the uc_def filter. But keep * our individual unicast filters. */ - if (efx_ef10_filter_insert_addr_list(efx, false, false)) - efx_ef10_filter_insert_def(efx, false, false); + if (efx_ef10_filter_insert_addr_list(efx, vlan, false, false)) + efx_ef10_filter_insert_def(efx, vlan, false, false); } /* Insert/renew multicast filters */ /* If changing promiscuous state with cascaded multicast filters, remove * old filters first, so that packets are dropped rather than duplicated */ - if (nic_data->workaround_26807 && efx->mc_promisc != mc_promisc) + if (nic_data->workaround_26807 && + table->mc_promisc_last != table->mc_promisc) efx_ef10_filter_remove_old(efx); - if (mc_promisc) { + if (table->mc_promisc) { if (nic_data->workaround_26807) { /* If we failed to insert promiscuous filters, rollback * and fall back to individual multicast filters */ - if (efx_ef10_filter_insert_def(efx, true, true)) { + if (efx_ef10_filter_insert_def(efx, vlan, true, true)) { /* Changing promisc state, so remove old filters */ efx_ef10_filter_remove_old(efx); - efx_ef10_filter_insert_addr_list(efx, true, false); + efx_ef10_filter_insert_addr_list(efx, vlan, + true, false); } } else { /* If we failed to insert promiscuous filters, don't * rollback. Regardless, also insert the mc_list */ - efx_ef10_filter_insert_def(efx, true, false); - efx_ef10_filter_insert_addr_list(efx, true, false); + efx_ef10_filter_insert_def(efx, vlan, true, false); + efx_ef10_filter_insert_addr_list(efx, vlan, true, false); } } else { /* If any filters failed to insert, rollback and fall back to @@ -4268,17 +4605,153 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) * that fails, roll back again and insert as many of our * individual multicast filters as we can. */ - if (efx_ef10_filter_insert_addr_list(efx, true, true)) { + if (efx_ef10_filter_insert_addr_list(efx, vlan, true, true)) { /* Changing promisc state, so remove old filters */ if (nic_data->workaround_26807) efx_ef10_filter_remove_old(efx); - if (efx_ef10_filter_insert_def(efx, true, true)) - efx_ef10_filter_insert_addr_list(efx, true, false); + if (efx_ef10_filter_insert_def(efx, vlan, true, true)) + efx_ef10_filter_insert_addr_list(efx, vlan, + true, false); } } +} + +/* Caller must hold efx->filter_sem for read if race against + * efx_ef10_filter_table_remove() is possible + */ +static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct net_device *net_dev = efx->net_dev; + struct efx_ef10_filter_vlan *vlan; + bool vlan_filter; + + if (!efx_dev_registered(efx)) + return; + + if (!table) + return; + + efx_ef10_filter_mark_old(efx); + + /* Copy/convert the address lists; add the primary station + * address and broadcast address + */ + netif_addr_lock_bh(net_dev); + efx_ef10_filter_uc_addr_list(efx); + efx_ef10_filter_mc_addr_list(efx); + netif_addr_unlock_bh(net_dev); + + /* If VLAN filtering changes, all old filters are finally removed. + * Do it in advance to avoid conflicts for unicast untagged and + * VLAN 0 tagged filters. + */ + vlan_filter = !!(net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER); + if (table->vlan_filter != vlan_filter) { + table->vlan_filter = vlan_filter; + efx_ef10_filter_remove_old(efx); + } + + list_for_each_entry(vlan, &table->vlan_list, list) + efx_ef10_filter_vlan_sync_rx_mode(efx, vlan); efx_ef10_filter_remove_old(efx); - efx->mc_promisc = mc_promisc; + table->mc_promisc_last = table->mc_promisc; +} + +static struct efx_ef10_filter_vlan *efx_ef10_filter_find_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan; + + WARN_ON(!rwsem_is_locked(&efx->filter_sem)); + + list_for_each_entry(vlan, &table->vlan_list, list) { + if (vlan->vid == vid) + return vlan; + } + + return NULL; +} + +static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan; + unsigned int i; + + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return -EINVAL; + + vlan = efx_ef10_filter_find_vlan(efx, vid); + if (WARN_ON(vlan)) { + netif_err(efx, drv, efx->net_dev, + "VLAN %u already added\n", vid); + return -EALREADY; + } + + vlan = kzalloc(sizeof(*vlan), GFP_KERNEL); + if (!vlan) + return -ENOMEM; + + vlan->vid = vid; + + for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) + vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID; + for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) + vlan->mc[i] = EFX_EF10_FILTER_ID_INVALID; + vlan->ucdef = EFX_EF10_FILTER_ID_INVALID; + vlan->bcast = EFX_EF10_FILTER_ID_INVALID; + vlan->mcdef = EFX_EF10_FILTER_ID_INVALID; + + list_add_tail(&vlan->list, &table->vlan_list); + + if (efx_dev_registered(efx)) + efx_ef10_filter_vlan_sync_rx_mode(efx, vlan); + + return 0; +} + +static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan) +{ + unsigned int i; + + /* See comment in efx_ef10_filter_table_remove() */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + + list_del(&vlan->list); + + for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->uc[i]); + for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->mc[i]); + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, vlan->ucdef); + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, vlan->bcast); + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, vlan->mcdef); + + kfree(vlan); +} + +static void efx_ef10_filter_del_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_filter_vlan *vlan; + + /* See comment in efx_ef10_filter_table_remove() */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + + vlan = efx_ef10_filter_find_vlan(efx, vid); + if (!vlan) { + netif_err(efx, drv, efx->net_dev, + "VLAN %u not found in filter state\n", vid); + return; + } + + efx_ef10_filter_del_vlan_internal(efx, vlan); } static int efx_ef10_set_mac_address(struct efx_nic *efx) @@ -4290,6 +4763,8 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx) efx_device_detach_sync(efx); efx_net_stop(efx->net_dev); + + mutex_lock(&efx->mac_lock); down_write(&efx->filter_sem); efx_ef10_filter_table_remove(efx); @@ -4302,6 +4777,8 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx) efx_ef10_filter_table_probe(efx); up_write(&efx->filter_sem); + mutex_unlock(&efx->mac_lock); + if (was_enabled) efx_net_open(efx->net_dev); netif_device_attach(efx->net_dev); @@ -4703,6 +5180,29 @@ static int efx_ef10_ptp_set_ts_config(struct efx_nic *efx, } } +static int efx_ef10_vlan_rx_add_vid(struct efx_nic *efx, __be16 proto, u16 vid) +{ + if (proto != htons(ETH_P_8021Q)) + return -EINVAL; + + return efx_ef10_add_vlan(efx, vid); +} + +static int efx_ef10_vlan_rx_kill_vid(struct efx_nic *efx, __be16 proto, u16 vid) +{ + if (proto != htons(ETH_P_8021Q)) + return -EINVAL; + + return efx_ef10_del_vlan(efx, vid); +} + +#define EF10_OFFLOAD_FEATURES \ + (NETIF_F_IP_CSUM | \ + NETIF_F_HW_VLAN_CTAG_FILTER | \ + NETIF_F_IPV6_CSUM | \ + NETIF_F_RXHASH | \ + NETIF_F_NTUPLE) + const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .is_vf = true, .mem_bar = EFX_MEM_VF_BAR, @@ -4780,6 +5280,8 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { #endif .ptp_write_host_time = efx_ef10_ptp_write_host_time_vf, .ptp_set_ts_config = efx_ef10_ptp_set_ts_config_vf, + .vlan_rx_add_vid = efx_ef10_vlan_rx_add_vid, + .vlan_rx_kill_vid = efx_ef10_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV .vswitching_probe = efx_ef10_vswitching_probe_vf, .vswitching_restore = efx_ef10_vswitching_restore_vf, @@ -4798,8 +5300,7 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .always_rx_scatter = true, .max_interrupt_mode = EFX_INT_MODE_MSIX, .timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH, - .offload_features = (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | - NETIF_F_RXHASH | NETIF_F_NTUPLE), + .offload_features = EF10_OFFLOAD_FEATURES, .mcdi_max_ver = 2, .max_rx_ip_filters = HUNT_FILTER_TBL_ROWS, .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | @@ -4891,6 +5392,8 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .ptp_write_host_time = efx_ef10_ptp_write_host_time, .ptp_set_ts_sync_events = efx_ef10_ptp_set_ts_sync_events, .ptp_set_ts_config = efx_ef10_ptp_set_ts_config, + .vlan_rx_add_vid = efx_ef10_vlan_rx_add_vid, + .vlan_rx_kill_vid = efx_ef10_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV .sriov_configure = efx_ef10_sriov_configure, .sriov_init = efx_ef10_sriov_init, @@ -4919,8 +5422,7 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .always_rx_scatter = true, .max_interrupt_mode = EFX_INT_MODE_MSIX, .timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH, - .offload_features = (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | - NETIF_F_RXHASH | NETIF_F_NTUPLE), + .offload_features = EF10_OFFLOAD_FEATURES, .mcdi_max_ver = 2, .max_rx_ip_filters = HUNT_FILTER_TBL_ROWS, .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c index 3c17f274e802..a949b9d27329 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.c +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -232,6 +232,35 @@ fail: return rc; } +static int efx_ef10_vadaptor_alloc_set_features(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + u32 port_flags; + int rc; + + rc = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + if (rc) + goto fail_vadaptor_alloc; + + rc = efx_ef10_vadaptor_query(efx, nic_data->vport_id, + &port_flags, NULL, NULL); + if (rc) + goto fail_vadaptor_query; + + if (port_flags & + (1 << MC_CMD_VPORT_ALLOC_IN_FLAG_VLAN_RESTRICT_LBN)) + efx->fixed_features |= NETIF_F_HW_VLAN_CTAG_FILTER; + else + efx->fixed_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + + return 0; + +fail_vadaptor_query: + efx_ef10_vadaptor_free(efx, EVB_PORT_ID_ASSIGNED); +fail_vadaptor_alloc: + return rc; +} + /* On top of the default firmware vswitch setup, create a VEB vswitch and * expansion vport for use by this function. */ @@ -243,7 +272,7 @@ int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) if (pci_sriov_get_totalvfs(efx->pci_dev) <= 0) { /* vswitch not needed as we have no VFs */ - efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + efx_ef10_vadaptor_alloc_set_features(efx); return 0; } @@ -263,7 +292,7 @@ int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) goto fail3; ether_addr_copy(nic_data->vport_mac, net_dev->dev_addr); - rc = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + rc = efx_ef10_vadaptor_alloc_set_features(efx); if (rc) goto fail4; @@ -282,9 +311,7 @@ fail1: int efx_ef10_vswitching_probe_vf(struct efx_nic *efx) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; - - return efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + return efx_ef10_vadaptor_alloc_set_features(efx); } int efx_ef10_vswitching_restore_pf(struct efx_nic *efx) @@ -554,6 +581,7 @@ int efx_ef10_sriov_set_vf_vlan(struct efx_nic *efx, int vf_i, u16 vlan, efx_device_detach_sync(vf->efx); efx_net_stop(vf->efx->net_dev); + mutex_lock(&vf->efx->mac_lock); down_write(&vf->efx->filter_sem); vf->efx->type->filter_table_remove(vf->efx); @@ -630,6 +658,7 @@ restore_filters: goto reset_nic_up_write; up_write(&vf->efx->filter_sem); + mutex_unlock(&vf->efx->mac_lock); up_write(&vf->efx->filter_sem); @@ -642,9 +671,10 @@ restore_filters: return rc; reset_nic_up_write: - if (vf->efx) + if (vf->efx) { up_write(&vf->efx->filter_sem); - + mutex_unlock(&vf->efx->mac_lock); + } reset_nic: if (vf->efx) { netif_err(efx, drv, efx->net_dev, diff --git a/drivers/net/ethernet/sfc/ef10_sriov.h b/drivers/net/ethernet/sfc/ef10_sriov.h index 6d25b92cb45e..9ceb7ef0a210 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.h +++ b/drivers/net/ethernet/sfc/ef10_sriov.h @@ -70,6 +70,9 @@ int efx_ef10_vport_add_mac(struct efx_nic *efx, int efx_ef10_vport_del_mac(struct efx_nic *efx, unsigned int port_id, u8 *mac); int efx_ef10_vadaptor_alloc(struct efx_nic *efx, unsigned int port_id); +int efx_ef10_vadaptor_query(struct efx_nic *efx, unsigned int port_id, + u32 *port_flags, u32 *vadaptor_flags, + unsigned int *vlan_tags); int efx_ef10_vadaptor_free(struct efx_nic *efx, unsigned int port_id); #endif /* EF10_SRIOV_H */ diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 097f363f1630..14b821b1c880 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -600,6 +600,7 @@ fail: */ static void efx_start_datapath(struct efx_nic *efx) { + netdev_features_t old_features = efx->net_dev->features; bool old_rx_scatter = efx->rx_scatter; struct efx_tx_queue *tx_queue; struct efx_rx_queue *rx_queue; @@ -644,6 +645,15 @@ static void efx_start_datapath(struct efx_nic *efx) efx->rx_dma_len, efx->rx_page_buf_step, efx->rx_bufs_per_page, efx->rx_pages_per_batch); + /* Restore previously fixed features in hw_features and remove + * features which are fixed now + */ + efx->net_dev->hw_features |= efx->net_dev->features; + efx->net_dev->hw_features &= ~efx->fixed_features; + efx->net_dev->features |= efx->fixed_features; + if (efx->net_dev->features != old_features) + netdev_features_change(efx->net_dev); + /* RX filters may also have scatter-enabled flags */ if (efx->rx_scatter != old_rx_scatter) efx->type->filter_update_rx_scatter(efx); @@ -1719,6 +1729,7 @@ static int efx_probe_filters(struct efx_nic *efx) spin_lock_init(&efx->filter_lock); init_rwsem(&efx->filter_sem); + mutex_lock(&efx->mac_lock); down_write(&efx->filter_sem); rc = efx->type->filter_table_probe(efx); if (rc) @@ -1757,6 +1768,7 @@ static int efx_probe_filters(struct efx_nic *efx) #endif out_unlock: up_write(&efx->filter_sem); + mutex_unlock(&efx->mac_lock); return rc; } @@ -2312,14 +2324,46 @@ static void efx_set_rx_mode(struct net_device *net_dev) static int efx_set_features(struct net_device *net_dev, netdev_features_t data) { struct efx_nic *efx = netdev_priv(net_dev); + int rc; /* If disabling RX n-tuple filtering, clear existing filters */ - if (net_dev->features & ~data & NETIF_F_NTUPLE) - return efx->type->filter_clear_rx(efx, EFX_FILTER_PRI_MANUAL); + if (net_dev->features & ~data & NETIF_F_NTUPLE) { + rc = efx->type->filter_clear_rx(efx, EFX_FILTER_PRI_MANUAL); + if (rc) + return rc; + } + + /* If Rx VLAN filter is changed, update filters via mac_reconfigure */ + if ((net_dev->features ^ data) & NETIF_F_HW_VLAN_CTAG_FILTER) { + /* efx_set_rx_mode() will schedule MAC work to update filters + * when a new features are finally set in net_dev. + */ + efx_set_rx_mode(net_dev); + } return 0; } +static int efx_vlan_rx_add_vid(struct net_device *net_dev, __be16 proto, u16 vid) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->vlan_rx_add_vid) + return efx->type->vlan_rx_add_vid(efx, proto, vid); + else + return -EOPNOTSUPP; +} + +static int efx_vlan_rx_kill_vid(struct net_device *net_dev, __be16 proto, u16 vid) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->vlan_rx_kill_vid) + return efx->type->vlan_rx_kill_vid(efx, proto, vid); + else + return -EOPNOTSUPP; +} + static const struct net_device_ops efx_netdev_ops = { .ndo_open = efx_net_open, .ndo_stop = efx_net_stop, @@ -2332,6 +2376,8 @@ static const struct net_device_ops efx_netdev_ops = { .ndo_set_mac_address = efx_set_mac_address, .ndo_set_rx_mode = efx_set_rx_mode, .ndo_set_features = efx_set_features, + .ndo_vlan_rx_add_vid = efx_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = efx_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV .ndo_set_vf_mac = efx_sriov_set_vf_mac, .ndo_set_vf_vlan = efx_sriov_set_vf_vlan, @@ -3147,17 +3193,25 @@ static int efx_pci_probe(struct pci_dev *pci_dev, return -ENOMEM; efx = netdev_priv(net_dev); efx->type = (const struct efx_nic_type *) entry->driver_data; + efx->fixed_features |= NETIF_F_HIGHDMA; net_dev->features |= (efx->type->offload_features | NETIF_F_SG | - NETIF_F_HIGHDMA | NETIF_F_TSO | - NETIF_F_RXCSUM); + NETIF_F_TSO | NETIF_F_RXCSUM); if (efx->type->offload_features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM)) net_dev->features |= NETIF_F_TSO6; /* Mask for features that also apply to VLAN devices */ net_dev->vlan_features |= (NETIF_F_HW_CSUM | NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_ALL_TSO | NETIF_F_RXCSUM); - /* All offloads can be toggled */ - net_dev->hw_features = net_dev->features & ~NETIF_F_HIGHDMA; + + net_dev->hw_features = net_dev->features & ~efx->fixed_features; + + /* Disable VLAN filtering by default. It may be enforced if + * the feature is fixed (i.e. VLAN filters are required to + * receive VLAN tagged packets due to vPort restrictions). + */ + net_dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + net_dev->features |= efx->fixed_features; + pci_set_drvdata(pci_dev, efx); SET_NETDEV_DEV(net_dev, &pci_dev->dev); rc = efx_init_struct(efx, pci_dev, net_dev); diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index 5e3f93f04e62..c3ae739e9c7a 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -274,4 +274,13 @@ static inline void efx_device_detach_sync(struct efx_nic *efx) netif_tx_unlock_bh(dev); } +static inline bool efx_rwsem_assert_write_locked(struct rw_semaphore *sem) +{ + if (WARN_ON(down_read_trylock(sem))) { + up_read(sem); + return false; + } + return true; +} + #endif /* EFX_EFX_H */ diff --git a/drivers/net/ethernet/sfc/farch.c b/drivers/net/ethernet/sfc/farch.c index 133e9e35be9e..4c83739d158f 100644 --- a/drivers/net/ethernet/sfc/farch.c +++ b/drivers/net/ethernet/sfc/farch.c @@ -104,7 +104,8 @@ int efx_farch_test_registers(struct efx_nic *efx, const struct efx_farch_register_test *regs, size_t n_regs) { - unsigned address = 0, i, j; + unsigned address = 0; + int i, j; efx_oword_t mask, imask, original, reg, buf; for (i = 0; i < n_regs; ++i) { diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h index 4cc772164a79..c9a5b003caaf 100644 --- a/drivers/net/ethernet/sfc/mcdi_pcol.h +++ b/drivers/net/ethernet/sfc/mcdi_pcol.h @@ -273,6 +273,9 @@ * have already installed filters. See the comment at * MC_CMD_WORKAROUND_BUG26807. */ #define MC_CMD_ERR_FILTERS_PRESENT 0x1014 +/* The clock whose frequency you've attempted to set set + * doesn't exist on this NIC */ +#define MC_CMD_ERR_NO_CLOCK 0x1015 #define MC_CMD_ERR_CODE_OFST 0 @@ -292,9 +295,11 @@ /* Point to the copycode entry point. */ #define SIENA_MC_BOOTROM_COPYCODE_VEC (0x800 - 3 * 0x4) #define HUNT_MC_BOOTROM_COPYCODE_VEC (0x8000 - 3 * 0x4) +#define MEDFORD_MC_BOOTROM_COPYCODE_VEC (0x10000 - 3 * 0x4) /* Points to the recovery mode entry point. */ #define SIENA_MC_BOOTROM_NOFLASH_VEC (0x800 - 2 * 0x4) #define HUNT_MC_BOOTROM_NOFLASH_VEC (0x8000 - 2 * 0x4) +#define MEDFORD_MC_BOOTROM_NOFLASH_VEC (0x10000 - 2 * 0x4) /* The command set exported by the boot ROM (MCDI v0) */ #define MC_CMD_GET_VERSION_V0_SUPPORTED_FUNCS { \ @@ -686,6 +691,12 @@ #define FCDI_EVENT_CODE_PTP_STATUS 0x9 /* enum: Port id config to map MC-FC port idx */ #define FCDI_EVENT_CODE_PORT_CONFIG 0xa +/* enum: Boot result or error code */ +#define FCDI_EVENT_CODE_BOOT_RESULT 0xb +#define FCDI_EVENT_REBOOT_SRC_LBN 36 +#define FCDI_EVENT_REBOOT_SRC_WIDTH 8 +#define FCDI_EVENT_REBOOT_FC_FW 0x0 /* enum */ +#define FCDI_EVENT_REBOOT_FC_BOOTLOADER 0x1 /* enum */ #define FCDI_EVENT_ASSERT_INSTR_ADDRESS_OFST 0 #define FCDI_EVENT_ASSERT_INSTR_ADDRESS_LBN 0 #define FCDI_EVENT_ASSERT_INSTR_ADDRESS_WIDTH 32 @@ -717,6 +728,11 @@ #define FCDI_EVENT_PORT_CONFIG_DATA_OFST 0 #define FCDI_EVENT_PORT_CONFIG_DATA_LBN 0 #define FCDI_EVENT_PORT_CONFIG_DATA_WIDTH 32 +#define FCDI_EVENT_BOOT_RESULT_OFST 0 +/* Enum values, see field(s): */ +/* MC_CMD_AOE/MC_CMD_AOE_OUT_INFO/FC_BOOT_RESULT */ +#define FCDI_EVENT_BOOT_RESULT_LBN 0 +#define FCDI_EVENT_BOOT_RESULT_WIDTH 32 /* FCDI_EXTENDED_EVENT_PPS structuredef: Extended FCDI event to send PPS events * to the MC. Note that this structure | is overlayed over a normal FCDI event @@ -1649,15 +1665,30 @@ /* MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS msgresponse */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_LEN 16 -/* Uncorrected error on transmit timestamps in NIC clock format */ +/* Uncorrected error on PTP transmit timestamps in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_TRANSMIT_OFST 0 -/* Uncorrected error on receive timestamps in NIC clock format */ +/* Uncorrected error on PTP receive timestamps in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_RECEIVE_OFST 4 /* Uncorrected error on PPS output in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_PPS_OUT_OFST 8 /* Uncorrected error on PPS input in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_PPS_IN_OFST 12 +/* MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2 msgresponse */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_LEN 24 +/* Uncorrected error on PTP transmit timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PTP_TX_OFST 0 +/* Uncorrected error on PTP receive timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PTP_RX_OFST 4 +/* Uncorrected error on PPS output in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PPS_OUT_OFST 8 +/* Uncorrected error on PPS input in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PPS_IN_OFST 12 +/* Uncorrected error on non-PTP transmit timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_GENERAL_TX_OFST 16 +/* Uncorrected error on non-PTP receive timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_GENERAL_RX_OFST 20 + /* MC_CMD_PTP_OUT_MANFTEST_PPS msgresponse */ #define MC_CMD_PTP_OUT_MANFTEST_PPS_LEN 4 /* Results of testing */ @@ -2158,8 +2189,12 @@ /* MC_CMD_DRV_ATTACH_IN msgrequest */ #define MC_CMD_DRV_ATTACH_IN_LEN 12 -/* new state (0=detached, 1=attached) to set if UPDATE=1 */ +/* new state to set if UPDATE=1 */ #define MC_CMD_DRV_ATTACH_IN_NEW_STATE_OFST 0 +#define MC_CMD_DRV_ATTACH_LBN 0 +#define MC_CMD_DRV_ATTACH_WIDTH 1 +#define MC_CMD_DRV_PREBOOT_LBN 1 +#define MC_CMD_DRV_PREBOOT_WIDTH 1 /* 1 to set new state, or 0 to just report the existing state */ #define MC_CMD_DRV_ATTACH_IN_UPDATE_OFST 4 /* preferred datapath firmware (for Huntington; ignored for Siena) */ @@ -2181,12 +2216,12 @@ /* MC_CMD_DRV_ATTACH_OUT msgresponse */ #define MC_CMD_DRV_ATTACH_OUT_LEN 4 -/* previous or existing state (0=detached, 1=attached) */ +/* previous or existing state, see the bitmask at NEW_STATE */ #define MC_CMD_DRV_ATTACH_OUT_OLD_STATE_OFST 0 /* MC_CMD_DRV_ATTACH_EXT_OUT msgresponse */ #define MC_CMD_DRV_ATTACH_EXT_OUT_LEN 8 -/* previous or existing state (0=detached, 1=attached) */ +/* previous or existing state, see the bitmask at NEW_STATE */ #define MC_CMD_DRV_ATTACH_EXT_OUT_OLD_STATE_OFST 0 /* Flags associated with this function */ #define MC_CMD_DRV_ATTACH_EXT_OUT_FUNC_FLAGS_OFST 4 @@ -2198,6 +2233,10 @@ #define MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_LINKCTRL 0x1 /* enum: The function can perform privileged operations */ #define MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_TRUSTED 0x2 +/* enum: The function does not have an active port associated with it. The port + * refers to the Sorrento external FPGA port. + */ +#define MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_NO_ACTIVE_PORT 0x3 /***********************************/ @@ -2892,7 +2931,7 @@ */ #define MC_CMD_SET_MAC 0x2c -#define MC_CMD_0x2c_PRIVILEGE_CTG SRIOV_CTG_LINK +#define MC_CMD_0x2c_PRIVILEGE_CTG SRIOV_CTG_GENERAL /* MC_CMD_SET_MAC_IN msgrequest */ #define MC_CMD_SET_MAC_IN_LEN 28 @@ -2927,9 +2966,66 @@ #define MC_CMD_SET_MAC_IN_FLAG_INCLUDE_FCS_LBN 0 #define MC_CMD_SET_MAC_IN_FLAG_INCLUDE_FCS_WIDTH 1 +/* MC_CMD_SET_MAC_EXT_IN msgrequest */ +#define MC_CMD_SET_MAC_EXT_IN_LEN 32 +/* The MTU is the MTU programmed directly into the XMAC/GMAC (inclusive of + * EtherII, VLAN, bug16011 padding). + */ +#define MC_CMD_SET_MAC_EXT_IN_MTU_OFST 0 +#define MC_CMD_SET_MAC_EXT_IN_DRAIN_OFST 4 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_OFST 8 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_LEN 8 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_LO_OFST 8 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_HI_OFST 12 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_OFST 16 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_UNCST_LBN 0 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_UNCST_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_BRDCST_LBN 1 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_BRDCST_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_FCNTL_OFST 20 +/* enum: Flow control is off. */ +/* MC_CMD_FCNTL_OFF 0x0 */ +/* enum: Respond to flow control. */ +/* MC_CMD_FCNTL_RESPOND 0x1 */ +/* enum: Respond to and Issue flow control. */ +/* MC_CMD_FCNTL_BIDIR 0x2 */ +/* enum: Auto neg flow control. */ +/* MC_CMD_FCNTL_AUTO 0x3 */ +/* enum: Priority flow control (eftest builds only). */ +/* MC_CMD_FCNTL_QBB 0x4 */ +/* enum: Issue flow control. */ +/* MC_CMD_FCNTL_GENERATE 0x5 */ +#define MC_CMD_SET_MAC_EXT_IN_FLAGS_OFST 24 +#define MC_CMD_SET_MAC_EXT_IN_FLAG_INCLUDE_FCS_LBN 0 +#define MC_CMD_SET_MAC_EXT_IN_FLAG_INCLUDE_FCS_WIDTH 1 +/* Select which parameters to configure. A parameter will only be modified if + * the corresponding control flag is set. If SET_MAC_ENHANCED is not set in + * capabilities then this field is ignored (and all flags are assumed to be + * set). + */ +#define MC_CMD_SET_MAC_EXT_IN_CONTROL_OFST 28 +#define MC_CMD_SET_MAC_EXT_IN_CFG_MTU_LBN 0 +#define MC_CMD_SET_MAC_EXT_IN_CFG_MTU_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_DRAIN_LBN 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_DRAIN_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_REJECT_LBN 2 +#define MC_CMD_SET_MAC_EXT_IN_CFG_REJECT_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCNTL_LBN 3 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCNTL_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCS_LBN 4 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCS_WIDTH 1 + /* MC_CMD_SET_MAC_OUT msgresponse */ #define MC_CMD_SET_MAC_OUT_LEN 0 +/* MC_CMD_SET_MAC_V2_OUT msgresponse */ +#define MC_CMD_SET_MAC_V2_OUT_LEN 4 +/* MTU as configured after processing the request. See comment at + * MC_CMD_SET_MAC_IN/MTU. To query MTU without doing any changes, set CONTROL + * to 0. + */ +#define MC_CMD_SET_MAC_V2_OUT_MTU_OFST 0 + /***********************************/ /* MC_CMD_PHY_STATS @@ -3521,6 +3617,26 @@ #define MC_CMD_NVRAM_INFO_OUT_PHYSDEV_OFST 16 #define MC_CMD_NVRAM_INFO_OUT_PHYSADDR_OFST 20 +/* MC_CMD_NVRAM_INFO_V2_OUT msgresponse */ +#define MC_CMD_NVRAM_INFO_V2_OUT_LEN 28 +#define MC_CMD_NVRAM_INFO_V2_OUT_TYPE_OFST 0 +/* Enum values, see field(s): */ +/* MC_CMD_NVRAM_TYPES/MC_CMD_NVRAM_TYPES_OUT/TYPES */ +#define MC_CMD_NVRAM_INFO_V2_OUT_SIZE_OFST 4 +#define MC_CMD_NVRAM_INFO_V2_OUT_ERASESIZE_OFST 8 +#define MC_CMD_NVRAM_INFO_V2_OUT_FLAGS_OFST 12 +#define MC_CMD_NVRAM_INFO_V2_OUT_PROTECTED_LBN 0 +#define MC_CMD_NVRAM_INFO_V2_OUT_PROTECTED_WIDTH 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_TLV_LBN 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_TLV_WIDTH 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_A_B_LBN 7 +#define MC_CMD_NVRAM_INFO_V2_OUT_A_B_WIDTH 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_PHYSDEV_OFST 16 +#define MC_CMD_NVRAM_INFO_V2_OUT_PHYSADDR_OFST 20 +/* Writes must be multiples of this size. Added to support the MUM on Sorrento. + */ +#define MC_CMD_NVRAM_INFO_V2_OUT_WRITESIZE_OFST 24 + /***********************************/ /* MC_CMD_NVRAM_UPDATE_START @@ -3561,6 +3677,37 @@ /* amount to read in bytes */ #define MC_CMD_NVRAM_READ_IN_LENGTH_OFST 8 +/* MC_CMD_NVRAM_READ_IN_V2 msgrequest */ +#define MC_CMD_NVRAM_READ_IN_V2_LEN 16 +#define MC_CMD_NVRAM_READ_IN_V2_TYPE_OFST 0 +/* Enum values, see field(s): */ +/* MC_CMD_NVRAM_TYPES/MC_CMD_NVRAM_TYPES_OUT/TYPES */ +#define MC_CMD_NVRAM_READ_IN_V2_OFFSET_OFST 4 +/* amount to read in bytes */ +#define MC_CMD_NVRAM_READ_IN_V2_LENGTH_OFST 8 +/* Optional control info. If a partition is stored with an A/B versioning + * scheme (i.e. in more than one physical partition in NVRAM) the host can set + * this to control which underlying physical partition is used to read data + * from. This allows it to perform a read-modify-write-verify with the write + * lock continuously held by calling NVRAM_UPDATE_START, reading the old + * contents using MODE=TARGET_CURRENT, overwriting the old partition and then + * verifying by reading with MODE=TARGET_BACKUP. + */ +#define MC_CMD_NVRAM_READ_IN_V2_MODE_OFST 12 +/* enum: Same as omitting MODE: caller sees data in current partition unless it + * holds the write lock in which case it sees data in the partition it is + * updating. + */ +#define MC_CMD_NVRAM_READ_IN_V2_DEFAULT 0x0 +/* enum: Read from the current partition of an A/B pair, even if holding the + * write lock. + */ +#define MC_CMD_NVRAM_READ_IN_V2_TARGET_CURRENT 0x1 +/* enum: Read from the non-current (i.e. to be updated) partition of an A/B + * pair + */ +#define MC_CMD_NVRAM_READ_IN_V2_TARGET_BACKUP 0x2 + /* MC_CMD_NVRAM_READ_OUT msgresponse */ #define MC_CMD_NVRAM_READ_OUT_LENMIN 1 #define MC_CMD_NVRAM_READ_OUT_LENMAX 252 @@ -3895,6 +4042,8 @@ #define MC_CMD_SENSOR_CCOM_AVREG_1V8_SUPPLY 0x39 /* enum: CCOM AVREG 1v8 supply (external ADC): mV */ #define MC_CMD_SENSOR_CCOM_AVREG_1V8_SUPPLY_EXTADC 0x3a +/* enum: CCOM RTS temperature: degC */ +#define MC_CMD_SENSOR_CONTROLLER_RTS 0x3b /* enum: Not a sensor: reserved for the next page flag */ #define MC_CMD_SENSOR_PAGE1_NEXT 0x3f /* enum: controller internal temperature sensor voltage on master core @@ -3931,6 +4080,12 @@ #define MC_CMD_SENSOR_PHY0_VCC 0x4c /* enum: Voltage supplied to the QSFP #1 from their power supply: mV */ #define MC_CMD_SENSOR_PHY1_VCC 0x4d +/* enum: Controller die temperature (TDIODE): degC */ +#define MC_CMD_SENSOR_CONTROLLER_TDIODE_TEMP 0x4e +/* enum: Board temperature (front): degC */ +#define MC_CMD_SENSOR_BOARD_FRONT_TEMP 0x4f +/* enum: Board temperature (back): degC */ +#define MC_CMD_SENSOR_BOARD_BACK_TEMP 0x50 /* MC_CMD_SENSOR_INFO_ENTRY_TYPEDEF */ #define MC_CMD_SENSOR_ENTRY_OFST 4 #define MC_CMD_SENSOR_ENTRY_LEN 8 @@ -4007,7 +4162,7 @@ /* MC_CMD_READ_SENSORS_EXT_IN msgrequest */ #define MC_CMD_READ_SENSORS_EXT_IN_LEN 12 -/* DMA address of host buffer for sensor readings */ +/* DMA address of host buffer for sensor readings (must be 4Kbyte aligned). */ #define MC_CMD_READ_SENSORS_EXT_IN_DMA_ADDR_OFST 0 #define MC_CMD_READ_SENSORS_EXT_IN_DMA_ADDR_LEN 8 #define MC_CMD_READ_SENSORS_EXT_IN_DMA_ADDR_LO_OFST 0 @@ -4608,6 +4763,10 @@ * operations */ #define MC_CMD_MUM_OP_QSFP 0xc +/* enum: Request discrete and SODIMM DDR info (type, size, speed grade, voltage + * level) from MUM + */ +#define MC_CMD_MUM_OP_READ_DDR_INFO 0xd /* MC_CMD_MUM_IN_NULL msgrequest */ #define MC_CMD_MUM_IN_NULL_LEN 4 @@ -4793,6 +4952,10 @@ #define MC_CMD_MUM_IN_PROGRAM_CLOCKS_FLAGS_OFST 8 #define MC_CMD_MUM_IN_PROGRAM_CLOCKS_OVERCLOCK_110_LBN 0 #define MC_CMD_MUM_IN_PROGRAM_CLOCKS_OVERCLOCK_110_WIDTH 1 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_NIC_FROM_FPGA_LBN 1 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_NIC_FROM_FPGA_WIDTH 1 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_REF_FROM_XO_LBN 2 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_REF_FROM_XO_WIDTH 1 /* MC_CMD_MUM_IN_FPGA_LOAD msgrequest */ #define MC_CMD_MUM_IN_FPGA_LOAD_LEN 8 @@ -4862,6 +5025,11 @@ #define MC_CMD_MUM_IN_QSFP_POLL_BIST_HDR_OFST 4 #define MC_CMD_MUM_IN_QSFP_POLL_BIST_IDX_OFST 8 +/* MC_CMD_MUM_IN_READ_DDR_INFO msgrequest */ +#define MC_CMD_MUM_IN_READ_DDR_INFO_LEN 4 +/* MUM cmd header */ +/* MC_CMD_MUM_IN_CMD_OFST 0 */ + /* MC_CMD_MUM_OUT msgresponse */ #define MC_CMD_MUM_OUT_LEN 0 @@ -5004,6 +5172,69 @@ #define MC_CMD_MUM_OUT_QSFP_POLL_BIST_LEN 4 #define MC_CMD_MUM_OUT_QSFP_POLL_BIST_TEST_OFST 0 +/* MC_CMD_MUM_OUT_READ_DDR_INFO msgresponse */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_LENMIN 24 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_LENMAX 248 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_LEN(num) (8+8*(num)) +/* Discrete (soldered) DDR resistor strap info */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_DISCRETE_DDR_INFO_OFST 0 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VRATIO_LBN 0 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VRATIO_WIDTH 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED1_LBN 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED1_WIDTH 16 +/* Number of SODIMM info records */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NUM_RECORDS_OFST 4 +/* Array of SODIMM info records */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_OFST 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_LEN 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_LO_OFST 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_HI_OFST 12 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_MINNUM 2 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_MAXNUM 30 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK_ID_LBN 0 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK_ID_WIDTH 8 +/* enum: SODIMM bank 1 (Top SODIMM for Sorrento) */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK1 0x0 +/* enum: SODIMM bank 2 (Bottom SODDIMM for Sorrento) */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK2 0x1 +/* enum: Total number of SODIMM banks */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NUM_BANKS 0x2 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_TYPE_LBN 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_TYPE_WIDTH 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RANK_LBN 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RANK_WIDTH 4 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VOLTAGE_LBN 20 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VOLTAGE_WIDTH 4 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NOT_POWERED 0x0 /* enum */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V25 0x1 /* enum */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V35 0x2 /* enum */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V5 0x3 /* enum */ +/* enum: Values 5-15 are reserved for future usage */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V8 0x4 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SIZE_LBN 24 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SIZE_WIDTH 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SPEED_LBN 32 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SPEED_WIDTH 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_STATE_LBN 48 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_STATE_WIDTH 4 +/* enum: No module present */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_ABSENT 0x0 +/* enum: Module present supported and powered on */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_POWERED 0x1 +/* enum: Module present but bad type */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_TYPE 0x2 +/* enum: Module present but incompatible voltage */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_VOLTAGE 0x3 +/* enum: Module present but unknown SPD */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_SPD 0x4 +/* enum: Module present but slot cannot support it */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_SLOT 0x5 +/* enum: Modules may or may not be present, but cannot establish contact by I2C + */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NOT_REACHABLE 0x6 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED2_LBN 52 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED2_WIDTH 12 + /* MC_CMD_RESOURCE_SPECIFIER enum */ /* enum: Any */ #define MC_CMD_RESOURCE_INSTANCE_ANY 0xffffffff @@ -5076,6 +5307,8 @@ #define NVRAM_PARTITION_TYPE_DYNAMIC_CONFIG 0x500 /* enum: Expansion ROM configuration data for port 0 */ #define NVRAM_PARTITION_TYPE_EXPROM_CONFIG_PORT0 0x600 +/* enum: Synonym for EXPROM_CONFIG_PORT0 as used in pmap files */ +#define NVRAM_PARTITION_TYPE_EXPROM_CONFIG 0x600 /* enum: Expansion ROM configuration data for port 1 */ #define NVRAM_PARTITION_TYPE_EXPROM_CONFIG_PORT1 0x601 /* enum: Expansion ROM configuration data for port 2 */ @@ -5084,6 +5317,8 @@ #define NVRAM_PARTITION_TYPE_EXPROM_CONFIG_PORT3 0x603 /* enum: Non-volatile log output partition */ #define NVRAM_PARTITION_TYPE_LOG 0x700 +/* enum: Non-volatile log output of second core on dual-core device */ +#define NVRAM_PARTITION_TYPE_LOG_SLAVE 0x701 /* enum: Device state dump output partition */ #define NVRAM_PARTITION_TYPE_DUMP 0x800 /* enum: Application license key storage partition */ @@ -5116,6 +5351,20 @@ #define NVRAM_PARTITION_TYPE_MUM_USER_ROM 0xc05 /* enum: MUM fuses and lockbits partition. */ #define NVRAM_PARTITION_TYPE_MUM_FUSELOCK 0xc06 +/* enum: UEFI expansion ROM if separate from PXE */ +#define NVRAM_PARTITION_TYPE_EXPANSION_UEFI 0xd00 +/* enum: Spare partition 0 */ +#define NVRAM_PARTITION_TYPE_SPARE_0 0x1000 +/* enum: Spare partition 1 */ +#define NVRAM_PARTITION_TYPE_SPARE_1 0x1100 +/* enum: Spare partition 2 */ +#define NVRAM_PARTITION_TYPE_SPARE_2 0x1200 +/* enum: Spare partition 3 */ +#define NVRAM_PARTITION_TYPE_SPARE_3 0x1300 +/* enum: Spare partition 4 */ +#define NVRAM_PARTITION_TYPE_SPARE_4 0x1400 +/* enum: Spare partition 5 */ +#define NVRAM_PARTITION_TYPE_SPARE_5 0x1500 /* enum: Start of reserved value range (firmware may use for any purpose) */ #define NVRAM_PARTITION_TYPE_RESERVED_VALUES_MIN 0xff00 /* enum: End of reserved value range (firmware may use for any purpose) */ @@ -5149,6 +5398,90 @@ #define LICENSED_APP_ID_ID_LBN 0 #define LICENSED_APP_ID_ID_WIDTH 32 +/* LICENSED_FEATURES structuredef */ +#define LICENSED_FEATURES_LEN 8 +/* Bitmask of licensed firmware features */ +#define LICENSED_FEATURES_MASK_OFST 0 +#define LICENSED_FEATURES_MASK_LEN 8 +#define LICENSED_FEATURES_MASK_LO_OFST 0 +#define LICENSED_FEATURES_MASK_HI_OFST 4 +#define LICENSED_FEATURES_RX_CUT_THROUGH_LBN 0 +#define LICENSED_FEATURES_RX_CUT_THROUGH_WIDTH 1 +#define LICENSED_FEATURES_PIO_LBN 1 +#define LICENSED_FEATURES_PIO_WIDTH 1 +#define LICENSED_FEATURES_EVQ_TIMER_LBN 2 +#define LICENSED_FEATURES_EVQ_TIMER_WIDTH 1 +#define LICENSED_FEATURES_CLOCK_LBN 3 +#define LICENSED_FEATURES_CLOCK_WIDTH 1 +#define LICENSED_FEATURES_RX_TIMESTAMPS_LBN 4 +#define LICENSED_FEATURES_RX_TIMESTAMPS_WIDTH 1 +#define LICENSED_FEATURES_TX_TIMESTAMPS_LBN 5 +#define LICENSED_FEATURES_TX_TIMESTAMPS_WIDTH 1 +#define LICENSED_FEATURES_RX_SNIFF_LBN 6 +#define LICENSED_FEATURES_RX_SNIFF_WIDTH 1 +#define LICENSED_FEATURES_TX_SNIFF_LBN 7 +#define LICENSED_FEATURES_TX_SNIFF_WIDTH 1 +#define LICENSED_FEATURES_PROXY_FILTER_OPS_LBN 8 +#define LICENSED_FEATURES_PROXY_FILTER_OPS_WIDTH 1 +#define LICENSED_FEATURES_EVENT_CUT_THROUGH_LBN 9 +#define LICENSED_FEATURES_EVENT_CUT_THROUGH_WIDTH 1 +#define LICENSED_FEATURES_MASK_LBN 0 +#define LICENSED_FEATURES_MASK_WIDTH 64 + +/* LICENSED_V3_APPS structuredef */ +#define LICENSED_V3_APPS_LEN 8 +/* Bitmask of licensed applications */ +#define LICENSED_V3_APPS_MASK_OFST 0 +#define LICENSED_V3_APPS_MASK_LEN 8 +#define LICENSED_V3_APPS_MASK_LO_OFST 0 +#define LICENSED_V3_APPS_MASK_HI_OFST 4 +#define LICENSED_V3_APPS_ONLOAD_LBN 0 +#define LICENSED_V3_APPS_ONLOAD_WIDTH 1 +#define LICENSED_V3_APPS_PTP_LBN 1 +#define LICENSED_V3_APPS_PTP_WIDTH 1 +#define LICENSED_V3_APPS_SOLARCAPTURE_PRO_LBN 2 +#define LICENSED_V3_APPS_SOLARCAPTURE_PRO_WIDTH 1 +#define LICENSED_V3_APPS_SOLARSECURE_LBN 3 +#define LICENSED_V3_APPS_SOLARSECURE_WIDTH 1 +#define LICENSED_V3_APPS_PERF_MONITOR_LBN 4 +#define LICENSED_V3_APPS_PERF_MONITOR_WIDTH 1 +#define LICENSED_V3_APPS_SOLARCAPTURE_LIVE_LBN 5 +#define LICENSED_V3_APPS_SOLARCAPTURE_LIVE_WIDTH 1 +#define LICENSED_V3_APPS_CAPTURE_SOLARSYSTEM_LBN 6 +#define LICENSED_V3_APPS_CAPTURE_SOLARSYSTEM_WIDTH 1 +#define LICENSED_V3_APPS_NETWORK_ACCESS_CONTROL_LBN 7 +#define LICENSED_V3_APPS_NETWORK_ACCESS_CONTROL_WIDTH 1 +#define LICENSED_V3_APPS_MASK_LBN 0 +#define LICENSED_V3_APPS_MASK_WIDTH 64 + +/* LICENSED_V3_FEATURES structuredef */ +#define LICENSED_V3_FEATURES_LEN 8 +/* Bitmask of licensed firmware features */ +#define LICENSED_V3_FEATURES_MASK_OFST 0 +#define LICENSED_V3_FEATURES_MASK_LEN 8 +#define LICENSED_V3_FEATURES_MASK_LO_OFST 0 +#define LICENSED_V3_FEATURES_MASK_HI_OFST 4 +#define LICENSED_V3_FEATURES_RX_CUT_THROUGH_LBN 0 +#define LICENSED_V3_FEATURES_RX_CUT_THROUGH_WIDTH 1 +#define LICENSED_V3_FEATURES_PIO_LBN 1 +#define LICENSED_V3_FEATURES_PIO_WIDTH 1 +#define LICENSED_V3_FEATURES_EVQ_TIMER_LBN 2 +#define LICENSED_V3_FEATURES_EVQ_TIMER_WIDTH 1 +#define LICENSED_V3_FEATURES_CLOCK_LBN 3 +#define LICENSED_V3_FEATURES_CLOCK_WIDTH 1 +#define LICENSED_V3_FEATURES_RX_TIMESTAMPS_LBN 4 +#define LICENSED_V3_FEATURES_RX_TIMESTAMPS_WIDTH 1 +#define LICENSED_V3_FEATURES_TX_TIMESTAMPS_LBN 5 +#define LICENSED_V3_FEATURES_TX_TIMESTAMPS_WIDTH 1 +#define LICENSED_V3_FEATURES_RX_SNIFF_LBN 6 +#define LICENSED_V3_FEATURES_RX_SNIFF_WIDTH 1 +#define LICENSED_V3_FEATURES_TX_SNIFF_LBN 7 +#define LICENSED_V3_FEATURES_TX_SNIFF_WIDTH 1 +#define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_LBN 8 +#define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_WIDTH 1 +#define LICENSED_V3_FEATURES_MASK_LBN 0 +#define LICENSED_V3_FEATURES_MASK_WIDTH 64 + /* TX_TIMESTAMP_EVENT structuredef */ #define TX_TIMESTAMP_EVENT_LEN 6 /* lower 16 bits of timestamp data */ @@ -5258,6 +5591,8 @@ #define MC_CMD_INIT_EVQ_IN_FLAG_RX_MERGE_WIDTH 1 #define MC_CMD_INIT_EVQ_IN_FLAG_TX_MERGE_LBN 5 #define MC_CMD_INIT_EVQ_IN_FLAG_TX_MERGE_WIDTH 1 +#define MC_CMD_INIT_EVQ_IN_FLAG_USE_TIMER_LBN 6 +#define MC_CMD_INIT_EVQ_IN_FLAG_USE_TIMER_WIDTH 1 #define MC_CMD_INIT_EVQ_IN_TMR_MODE_OFST 20 /* enum: Disabled */ #define MC_CMD_INIT_EVQ_IN_TMR_MODE_DIS 0x0 @@ -5362,6 +5697,8 @@ #define MC_CMD_INIT_RXQ_IN_FLAG_PREFIX_WIDTH 1 #define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_LBN 9 #define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_WIDTH 1 +#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_LBN 10 +#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_RXQ_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -5422,6 +5759,8 @@ #define MC_CMD_INIT_RXQ_EXT_IN_PS_BUFF_64K 0x4 /* enum */ #define MC_CMD_INIT_RXQ_EXT_IN_FLAG_WANT_OUTER_CLASSES_LBN 18 #define MC_CMD_INIT_RXQ_EXT_IN_FLAG_WANT_OUTER_CLASSES_WIDTH 1 +#define MC_CMD_INIT_RXQ_EXT_IN_FLAG_FORCE_EV_MERGING_LBN 19 +#define MC_CMD_INIT_RXQ_EXT_IN_FLAG_FORCE_EV_MERGING_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_RXQ_EXT_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -5535,6 +5874,8 @@ #define MC_CMD_INIT_TXQ_EXT_IN_FLAG_INNER_IP_CSUM_EN_WIDTH 1 #define MC_CMD_INIT_TXQ_EXT_IN_FLAG_INNER_TCP_CSUM_EN_LBN 11 #define MC_CMD_INIT_TXQ_EXT_IN_FLAG_INNER_TCP_CSUM_EN_WIDTH 1 +#define MC_CMD_INIT_TXQ_EXT_IN_FLAG_TSOV2_EN_LBN 12 +#define MC_CMD_INIT_TXQ_EXT_IN_FLAG_TSOV2_EN_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_TXQ_EXT_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -5747,6 +6088,46 @@ #define MC_CMD_PROXY_CONFIGURE_IN_ALLOWED_MCDI_MASK_OFST 44 #define MC_CMD_PROXY_CONFIGURE_IN_ALLOWED_MCDI_MASK_LEN 64 +/* MC_CMD_PROXY_CONFIGURE_EXT_IN msgrequest */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_LEN 112 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_FLAGS_OFST 0 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ENABLE_LBN 0 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ENABLE_WIDTH 1 +/* Host provides a contiguous memory buffer that contains at least NUM_BLOCKS + * of blocks, each of the size REQUEST_BLOCK_SIZE. + */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_OFST 4 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_LEN 8 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_LO_OFST 4 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_HI_OFST 8 +/* Must be a power of 2 */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BLOCK_SIZE_OFST 12 +/* Host provides a contiguous memory buffer that contains at least NUM_BLOCKS + * of blocks, each of the size REPLY_BLOCK_SIZE. + */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_OFST 16 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_LEN 8 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_LO_OFST 16 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_HI_OFST 20 +/* Must be a power of 2 */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BLOCK_SIZE_OFST 24 +/* Host provides a contiguous memory buffer that contains at least NUM_BLOCKS + * of blocks, each of the size STATUS_BLOCK_SIZE. This buffer is only needed if + * host intends to complete proxied operations by using MC_CMD_PROXY_CMD. + */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_OFST 28 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_LEN 8 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_LO_OFST 28 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_HI_OFST 32 +/* Must be a power of 2, or zero if this buffer is not provided */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BLOCK_SIZE_OFST 36 +/* Applies to all three buffers */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_NUM_BLOCKS_OFST 40 +/* A bit mask defining which MCDI operations may be proxied */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ALLOWED_MCDI_MASK_OFST 44 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ALLOWED_MCDI_MASK_LEN 64 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_RESERVED_OFST 108 + /* MC_CMD_PROXY_CONFIGURE_OUT msgresponse */ #define MC_CMD_PROXY_CONFIGURE_OUT_LEN 0 @@ -6323,6 +6704,15 @@ * client */ #define MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_RESTRICTIONS 0x2 +/* enum: read properties relating to security rules (Medford-only; for use by + * SolarSecure apps, not directly by drivers. See SF-114946-SW.) + */ +#define MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SECURITY_RULE_INFO 0x3 +/* enum: read the list of supported RX filter matches for VXLAN/NVGRE + * encapsulated frames, which follow a different match sequence to normal + * frames (Medford only) + */ +#define MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SUPPORTED_ENCAP_RX_MATCHES 0x4 /* MC_CMD_GET_PARSER_DISP_INFO_OUT msgresponse */ #define MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMIN 8 @@ -6356,7 +6746,10 @@ /***********************************/ /* MC_CMD_PARSER_DISP_RW - * Direct read/write of parser-dispatcher state (DICPUs and LUE) for debugging + * Direct read/write of parser-dispatcher state (DICPUs and LUE) for debugging. + * Please note that this interface is only of use to debug tools which have + * knowledge of firmware and hardware data structures; nothing here is intended + * for use by normal driver code. */ #define MC_CMD_PARSER_DISP_RW 0xe5 @@ -6374,6 +6767,12 @@ #define MC_CMD_PARSER_DISP_RW_IN_LUE 0x2 /* enum: Lookup engine (with requested metadata format) */ #define MC_CMD_PARSER_DISP_RW_IN_LUE_VERSIONED_METADATA 0x3 +/* enum: RX0 dispatcher CPU (alias for RX_DICPU; Medford has 2 RX DICPUs) */ +#define MC_CMD_PARSER_DISP_RW_IN_RX0_DICPU 0x0 +/* enum: RX1 dispatcher CPU (only valid for Medford) */ +#define MC_CMD_PARSER_DISP_RW_IN_RX1_DICPU 0x4 +/* enum: Miscellaneous other state (only valid for Medford) */ +#define MC_CMD_PARSER_DISP_RW_IN_MISC_STATE 0x5 /* identifies the type of operation requested */ #define MC_CMD_PARSER_DISP_RW_IN_OP_OFST 4 /* enum: read a word of DICPU DMEM or a LUE entry */ @@ -6382,8 +6781,12 @@ #define MC_CMD_PARSER_DISP_RW_IN_WRITE 0x1 /* enum: read-modify-write a word of DICPU DMEM (not valid for LUE) */ #define MC_CMD_PARSER_DISP_RW_IN_RMW 0x2 -/* data memory address or LUE index */ +/* data memory address (DICPU targets) or LUE index (LUE targets) */ #define MC_CMD_PARSER_DISP_RW_IN_ADDRESS_OFST 8 +/* selector (for MISC_STATE target) */ +#define MC_CMD_PARSER_DISP_RW_IN_SELECTOR_OFST 8 +/* enum: Port to datapath mapping */ +#define MC_CMD_PARSER_DISP_RW_IN_PORT_DP_MAPPING 0x1 /* value to write (for DMEM writes) */ #define MC_CMD_PARSER_DISP_RW_IN_DMEM_WRITE_VALUE_OFST 12 /* XOR value (for DMEM read-modify-writes: new = (old & mask) ^ value) */ @@ -6408,6 +6811,12 @@ */ #define MC_CMD_PARSER_DISP_RW_OUT_LUE_MGR_STATE_OFST 20 #define MC_CMD_PARSER_DISP_RW_OUT_LUE_MGR_STATE_LEN 32 +/* datapath(s) used for each port (for MISC_STATE PORT_DP_MAPPING selector) */ +#define MC_CMD_PARSER_DISP_RW_OUT_PORT_DP_MAPPING_OFST 0 +#define MC_CMD_PARSER_DISP_RW_OUT_PORT_DP_MAPPING_LEN 4 +#define MC_CMD_PARSER_DISP_RW_OUT_PORT_DP_MAPPING_NUM 4 +#define MC_CMD_PARSER_DISP_RW_OUT_DP0 0x1 /* enum */ +#define MC_CMD_PARSER_DISP_RW_OUT_DP1 0x2 /* enum */ /***********************************/ @@ -7071,6 +7480,24 @@ #define MC_CMD_GET_CAPABILITIES_OUT_LEN 20 /* First word of flags. */ #define MC_CMD_GET_CAPABILITIES_OUT_FLAGS1_OFST 0 +#define MC_CMD_GET_CAPABILITIES_OUT_VPORT_RECONFIGURE_LBN 3 +#define MC_CMD_GET_CAPABILITIES_OUT_VPORT_RECONFIGURE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_TX_STRIPING_LBN 4 +#define MC_CMD_GET_CAPABILITIES_OUT_TX_STRIPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_QUERY_LBN 5 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_QUERY_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_EVB_PORT_VLAN_RESTRICT_LBN 6 +#define MC_CMD_GET_CAPABILITIES_OUT_EVB_PORT_VLAN_RESTRICT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_DRV_ATTACH_PREBOOT_LBN 7 +#define MC_CMD_GET_CAPABILITIES_OUT_DRV_ATTACH_PREBOOT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_RX_FORCE_EVENT_MERGING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_OUT_RX_FORCE_EVENT_MERGING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_SET_MAC_ENHANCED_LBN 9 +#define MC_CMD_GET_CAPABILITIES_OUT_SET_MAC_ENHANCED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_LBN 10 +#define MC_CMD_GET_CAPABILITIES_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 11 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 #define MC_CMD_GET_CAPABILITIES_OUT_TX_MAC_SECURITY_FILTERING_LBN 12 #define MC_CMD_GET_CAPABILITIES_OUT_TX_MAC_SECURITY_FILTERING_WIDTH 1 #define MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN 13 @@ -7138,6 +7565,8 @@ #define MC_CMD_GET_CAPABILITIES_OUT_RXDP_TEST_FW_RX_HDR_SPLIT 0x107 /* enum: RXDP Test firmware image 8 */ #define MC_CMD_GET_CAPABILITIES_OUT_RXDP_TEST_FW_DISABLE_DL 0x108 +/* enum: RXDP Test firmware image 9 */ +#define MC_CMD_GET_CAPABILITIES_OUT_RXDP_TEST_FW_DOORBELL_DELAY 0x10b /* TxDPCPU firmware id. */ #define MC_CMD_GET_CAPABILITIES_OUT_TX_DPCPU_FW_ID_OFST 6 #define MC_CMD_GET_CAPABILITIES_OUT_TX_DPCPU_FW_ID_LEN 2 @@ -7153,6 +7582,8 @@ #define MC_CMD_GET_CAPABILITIES_OUT_TXDP_TEST_FW_TSO_EDIT 0x101 /* enum: TXDP Test firmware image 2 */ #define MC_CMD_GET_CAPABILITIES_OUT_TXDP_TEST_FW_PACKET_EDITS 0x102 +/* enum: TXDP CSR bus test firmware */ +#define MC_CMD_GET_CAPABILITIES_OUT_TXDP_TEST_FW_CSR 0x103 #define MC_CMD_GET_CAPABILITIES_OUT_RXPD_FW_VERSION_OFST 8 #define MC_CMD_GET_CAPABILITIES_OUT_RXPD_FW_VERSION_LEN 2 #define MC_CMD_GET_CAPABILITIES_OUT_RXPD_FW_VERSION_REV_LBN 0 @@ -7227,6 +7658,258 @@ /* Licensed capabilities */ #define MC_CMD_GET_CAPABILITIES_OUT_LICENSE_CAPABILITIES_OFST 16 +/* MC_CMD_GET_CAPABILITIES_V2_IN msgrequest */ +#define MC_CMD_GET_CAPABILITIES_V2_IN_LEN 0 + +/* MC_CMD_GET_CAPABILITIES_V2_OUT msgresponse */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_LEN 72 +/* First word of flags. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_FLAGS1_OFST 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VPORT_RECONFIGURE_LBN 3 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VPORT_RECONFIGURE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_STRIPING_LBN 4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_STRIPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_QUERY_LBN 5 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_QUERY_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_PORT_VLAN_RESTRICT_LBN 6 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_PORT_VLAN_RESTRICT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_DRV_ATTACH_PREBOOT_LBN 7 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_DRV_ATTACH_PREBOOT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_FORCE_EVENT_MERGING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_FORCE_EVENT_MERGING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SET_MAC_ENHANCED_LBN 9 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SET_MAC_ENHANCED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_LBN 10 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 11 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_SECURITY_FILTERING_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_SECURITY_FILTERING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_ADDITIONAL_RSS_MODES_LBN 13 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_ADDITIONAL_RSS_MODES_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_QBB_LBN 14 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_QBB_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_VAR_BUFFERS_LBN 15 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_VAR_BUFFERS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_RSS_LIMITED_LBN 16 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_RSS_LIMITED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_LBN 17 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_INCLUDE_FCS_LBN 18 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_INCLUDE_FCS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VLAN_INSERTION_LBN 19 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VLAN_INSERTION_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_VLAN_STRIPPING_LBN 20 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_VLAN_STRIPPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_LBN 21 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_0_LBN 22 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_0_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_14_LBN 23 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_14_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_TIMESTAMP_LBN 24 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_TIMESTAMP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_BATCHING_LBN 25 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_BATCHING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_MCAST_FILTER_CHAINING_LBN 26 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_MCAST_FILTER_CHAINING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PM_AND_RXDP_COUNTERS_LBN 27 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PM_AND_RXDP_COUNTERS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DISABLE_SCATTER_LBN 28 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DISABLE_SCATTER_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MCAST_UDP_LOOPBACK_LBN 29 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MCAST_UDP_LOOPBACK_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_LBN 30 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VXLAN_NVGRE_LBN 31 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VXLAN_NVGRE_WIDTH 1 +/* RxDPCPU firmware id. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DPCPU_FW_ID_OFST 4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DPCPU_FW_ID_LEN 2 +/* enum: Standard RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP 0x0 +/* enum: Low latency RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_LOW_LATENCY 0x1 +/* enum: Packed stream RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_PACKED_STREAM 0x2 +/* enum: BIST RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_BIST 0x10a +/* enum: RXDP Test firmware image 1 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_TO_MC_CUT_THROUGH 0x101 +/* enum: RXDP Test firmware image 2 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD 0x102 +/* enum: RXDP Test firmware image 3 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD_FIRST 0x103 +/* enum: RXDP Test firmware image 4 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_EVERY_EVENT_BATCHABLE 0x104 +/* enum: RXDP Test firmware image 5 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_BACKPRESSURE 0x105 +/* enum: RXDP Test firmware image 6 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_PACKET_EDITS 0x106 +/* enum: RXDP Test firmware image 7 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_RX_HDR_SPLIT 0x107 +/* enum: RXDP Test firmware image 8 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_DISABLE_DL 0x108 +/* enum: RXDP Test firmware image 9 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_DOORBELL_DELAY 0x10b +/* TxDPCPU firmware id. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DPCPU_FW_ID_OFST 6 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DPCPU_FW_ID_LEN 2 +/* enum: Standard TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP 0x0 +/* enum: Low latency TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_LOW_LATENCY 0x1 +/* enum: High packet rate TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_HIGH_PACKET_RATE 0x3 +/* enum: BIST TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_BIST 0x12d +/* enum: TXDP Test firmware image 1 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_TEST_FW_TSO_EDIT 0x101 +/* enum: TXDP Test firmware image 2 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_TEST_FW_PACKET_EDITS 0x102 +/* enum: TXDP CSR bus test firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_TEST_FW_CSR 0x103 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_OFST 8 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_REV_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_REV_WIDTH 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_TYPE_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_TYPE_WIDTH 4 +/* enum: reserved value - do not use (may indicate alternative interpretation + * of REV field in future) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_RESERVED 0x0 +/* enum: Trivial RX PD firmware for early Huntington development (Huntington + * development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_FIRST_PKT 0x1 +/* enum: RX PD firmware with approximately Siena-compatible behaviour + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_SIENA_COMPAT 0x2 +/* enum: Virtual switching (full feature) RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_VSWITCH 0x3 +/* enum: siena_compat variant RX PD firmware using PM rather than MAC + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_SIENA_COMPAT_PM 0x4 +/* enum: Low latency RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_LOW_LATENCY 0x5 +/* enum: Packed stream RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_PACKED_STREAM 0x6 +/* enum: RX PD firmware handling layer 2 only for high packet rate performance + * tests (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_LAYER2_PERF 0x7 +/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe +/* enum: RX PD firmware parsing but not filtering network overlay tunnel + * encapsulations (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_TESTFW_ENCAP_PARSING_ONLY 0xf +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_OFST 10 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_REV_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_REV_WIDTH 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_TYPE_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_TYPE_WIDTH 4 +/* enum: reserved value - do not use (may indicate alternative interpretation + * of REV field in future) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_RESERVED 0x0 +/* enum: Trivial TX PD firmware for early Huntington development (Huntington + * development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_FIRST_PKT 0x1 +/* enum: TX PD firmware with approximately Siena-compatible behaviour + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_SIENA_COMPAT 0x2 +/* enum: Virtual switching (full feature) TX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_VSWITCH 0x3 +/* enum: siena_compat variant TX PD firmware using PM rather than MAC + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_SIENA_COMPAT_PM 0x4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_LOW_LATENCY 0x5 /* enum */ +/* enum: TX PD firmware handling layer 2 only for high packet rate performance + * tests (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_LAYER2_PERF 0x7 +/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe +/* Hardware capabilities of NIC */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_HW_CAPABILITIES_OFST 12 +/* Licensed capabilities */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_LICENSE_CAPABILITIES_OFST 16 +/* Second word of flags. Not present on older firmware (check the length). */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_FLAGS2_OFST 20 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_ENCAP_LBN 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_ENCAP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVQ_TIMER_CTRL_LBN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVQ_TIMER_CTRL_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVENT_CUT_THROUGH_LBN 3 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVENT_CUT_THROUGH_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_LBN 4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_WIDTH 1 +/* Number of FATSOv2 contexts per datapath supported by this NIC. Not present + * on older firmware (check the length). + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_N_CONTEXTS_OFST 24 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_N_CONTEXTS_LEN 2 +/* One byte per PF containing the number of the external port assigned to this + * PF, indexed by PF number. Special values indicate that a PF is either not + * present or not assigned. + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PFS_TO_PORTS_ASSIGNMENT_OFST 26 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PFS_TO_PORTS_ASSIGNMENT_LEN 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PFS_TO_PORTS_ASSIGNMENT_NUM 16 +/* enum: The caller is not permitted to access information on this PF. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_ACCESS_NOT_PERMITTED 0xff +/* enum: PF does not exist. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PF_NOT_PRESENT 0xfe +/* enum: PF does exist but is not assigned to any external port. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PF_NOT_ASSIGNED 0xfd +/* enum: This value indicates that PF is assigned, but it cannot be expressed + * in this field. It is intended for a possible future situation where a more + * complex scheme of PFs to ports mapping is being used. The future driver + * should look for a new field supporting the new scheme. The current/old + * driver should treat this value as PF_NOT_ASSIGNED. + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_INCOMPATIBLE_ASSIGNMENT 0xfc +/* One byte per PF containing the number of its VFs, indexed by PF number. A + * special value indicates that a PF is not present. + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VFS_PER_PF_OFST 42 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VFS_PER_PF_LEN 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VFS_PER_PF_NUM 16 +/* enum: The caller is not permitted to access information on this PF. */ +/* MC_CMD_GET_CAPABILITIES_V2_OUT_ACCESS_NOT_PERMITTED 0xff */ +/* enum: PF does not exist. */ +/* MC_CMD_GET_CAPABILITIES_V2_OUT_PF_NOT_PRESENT 0xfe */ +/* Number of VIs available for each external port */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VIS_PER_PORT_OFST 58 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VIS_PER_PORT_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VIS_PER_PORT_NUM 4 +/* Size of RX descriptor cache expressed as binary logarithm The actual size + * equals (2 ^ RX_DESC_CACHE_SIZE) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DESC_CACHE_SIZE_OFST 66 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DESC_CACHE_SIZE_LEN 1 +/* Size of TX descriptor cache expressed as binary logarithm The actual size + * equals (2 ^ TX_DESC_CACHE_SIZE) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DESC_CACHE_SIZE_OFST 67 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DESC_CACHE_SIZE_LEN 1 +/* Total number of available PIO buffers */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_PIO_BUFFS_OFST 68 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_PIO_BUFFS_LEN 2 +/* Size of a single PIO buffer */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_OFST 70 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_LEN 2 + /***********************************/ /* MC_CMD_V2_EXTN @@ -7475,6 +8158,25 @@ /***********************************/ +/* MC_CMD_VSWITCH_QUERY + * read some config of v-switch. For now this command is an empty placeholder. + * It may be used to check if a v-switch is connected to a given EVB port (if + * not, then the command returns ENOENT). + */ +#define MC_CMD_VSWITCH_QUERY 0x63 + +#define MC_CMD_0x63_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VSWITCH_QUERY_IN msgrequest */ +#define MC_CMD_VSWITCH_QUERY_IN_LEN 4 +/* The port to which the v-switch is connected. */ +#define MC_CMD_VSWITCH_QUERY_IN_UPSTREAM_PORT_ID_OFST 0 + +/* MC_CMD_VSWITCH_QUERY_OUT msgresponse */ +#define MC_CMD_VSWITCH_QUERY_OUT_LEN 0 + + +/***********************************/ /* MC_CMD_VPORT_ALLOC * allocate a v-port. */ @@ -7510,6 +8212,8 @@ #define MC_CMD_VPORT_ALLOC_IN_FLAGS_OFST 8 #define MC_CMD_VPORT_ALLOC_IN_FLAG_AUTO_PORT_LBN 0 #define MC_CMD_VPORT_ALLOC_IN_FLAG_AUTO_PORT_WIDTH 1 +#define MC_CMD_VPORT_ALLOC_IN_FLAG_VLAN_RESTRICT_LBN 1 +#define MC_CMD_VPORT_ALLOC_IN_FLAG_VLAN_RESTRICT_WIDTH 1 /* The number of VLAN tags to insert/remove. An error will be returned if * incompatible with the number of VLAN tags specified for the upstream * v-switch. @@ -7561,6 +8265,8 @@ #define MC_CMD_VADAPTOR_ALLOC_IN_FLAGS_OFST 8 #define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_AUTO_VADAPTOR_LBN 0 #define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_AUTO_VADAPTOR_WIDTH 1 +#define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 1 +#define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 /* The number of VLAN tags to strip on receive */ #define MC_CMD_VADAPTOR_ALLOC_IN_NUM_VLANS_OFST 12 /* The number of VLAN tags to transparently insert/remove. */ @@ -7639,6 +8345,29 @@ /***********************************/ +/* MC_CMD_VADAPTOR_QUERY + * read some config of v-adaptor. + */ +#define MC_CMD_VADAPTOR_QUERY 0x61 + +#define MC_CMD_0x61_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VADAPTOR_QUERY_IN msgrequest */ +#define MC_CMD_VADAPTOR_QUERY_IN_LEN 4 +/* The port to which the v-adaptor is connected. */ +#define MC_CMD_VADAPTOR_QUERY_IN_UPSTREAM_PORT_ID_OFST 0 + +/* MC_CMD_VADAPTOR_QUERY_OUT msgresponse */ +#define MC_CMD_VADAPTOR_QUERY_OUT_LEN 12 +/* The EVB port flags as defined at MC_CMD_VPORT_ALLOC. */ +#define MC_CMD_VADAPTOR_QUERY_OUT_PORT_FLAGS_OFST 0 +/* The v-adaptor flags as defined at MC_CMD_VADAPTOR_ALLOC. */ +#define MC_CMD_VADAPTOR_QUERY_OUT_VADAPTOR_FLAGS_OFST 4 +/* The number of VLAN tags that may still be added */ +#define MC_CMD_VADAPTOR_QUERY_OUT_NUM_AVAILABLE_VLAN_TAGS_OFST 8 + + +/***********************************/ /* MC_CMD_EVB_PORT_ASSIGN * assign a port to a PCI function. */ @@ -7875,10 +8604,17 @@ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN 8 /* The handle of the RSS context */ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID_OFST 0 -/* Hash control flags. The _EN bits are always supported. The _MODE bits only - * work when the firmware reports ADDITIONAL_RSS_MODES in - * MC_CMD_GET_CAPABILITIES and override the _EN bits if any of them are not 0. - * See the RSS_MODE structure for the meaning of the mode bits. +/* Hash control flags. The _EN bits are always supported, but new modes are + * available when ADDITIONAL_RSS_MODES is reported by MC_CMD_GET_CAPABILITIES: + * in this case, the MODE fields may be set to non-zero values, and will take + * effect regardless of the settings of the _EN flags. See the RSS_MODE + * structure for the meaning of the mode bits. Drivers must check the + * capability before trying to set any _MODE fields, as older firmware will + * reject any attempt to set the FLAGS field to a value > 0xff with EINVAL. In + * the case where all the _MODE flags are zero, the _EN flags take effect, + * providing backward compatibility for existing drivers. (Setting all _MODE + * *and* all _EN flags to zero is valid, to disable RSS spreading for that + * particular packet type.) */ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_FLAGS_OFST 4 #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_IPV4_EN_LBN 0 @@ -7923,11 +8659,18 @@ /* MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT msgresponse */ #define MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN 8 -/* Hash control flags. If any _MODE bits are non-zero (which will only be true - * when the firmware reports ADDITIONAL_RSS_MODES) then the _EN bits should be - * disregarded (but are guaranteed to be consistent with the _MODE bits if - * RSS_CONTEXT_SET_FLAGS has never been called for this context since it was - * allocated). +/* Hash control flags. If all _MODE bits are zero (which will always be true + * for older firmware which does not report the ADDITIONAL_RSS_MODES + * capability), the _EN bits report the state. If any _MODE bits are non-zero + * (which will only be true when the firmware reports ADDITIONAL_RSS_MODES) + * then the _EN bits should be disregarded, although the _MODE flags are + * guaranteed to be consistent with the _EN flags for a freshly-allocated RSS + * context and in the case where the _EN flags were used in the SET. This + * provides backward compatibility: old drivers will not be attempting to + * derive any meaning from the _MODE bits (and can never set them to any value + * not representable by the _EN bits); new drivers can always determine the + * mode by looking only at the _MODE bits; the value returned by a GET can + * always be used for a SET regardless of old/new driver vs. old/new firmware. */ #define MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_FLAGS_OFST 4 #define MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV4_EN_LBN 0 @@ -8155,6 +8898,74 @@ /***********************************/ +/* MC_CMD_VPORT_RECONFIGURE + * Replace VLAN tags and/or MAC addresses of an existing v-port. If the v-port + * has already been passed to another function (v-port's user), then that + * function will be reset before applying the changes. + */ +#define MC_CMD_VPORT_RECONFIGURE 0xeb + +#define MC_CMD_0xeb_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VPORT_RECONFIGURE_IN msgrequest */ +#define MC_CMD_VPORT_RECONFIGURE_IN_LEN 44 +/* The handle of the v-port */ +#define MC_CMD_VPORT_RECONFIGURE_IN_VPORT_ID_OFST 0 +/* Flags requesting what should be changed. */ +#define MC_CMD_VPORT_RECONFIGURE_IN_FLAGS_OFST 4 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_VLAN_TAGS_LBN 0 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_VLAN_TAGS_WIDTH 1 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_MACADDRS_LBN 1 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_MACADDRS_WIDTH 1 +/* The number of VLAN tags to insert/remove. An error will be returned if + * incompatible with the number of VLAN tags specified for the upstream + * v-switch. + */ +#define MC_CMD_VPORT_RECONFIGURE_IN_NUM_VLAN_TAGS_OFST 8 +/* The actual VLAN tags to insert/remove */ +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAGS_OFST 12 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_0_LBN 0 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_0_WIDTH 16 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_1_LBN 16 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_1_WIDTH 16 +/* The number of MAC addresses to add */ +#define MC_CMD_VPORT_RECONFIGURE_IN_NUM_MACADDRS_OFST 16 +/* MAC addresses to add */ +#define MC_CMD_VPORT_RECONFIGURE_IN_MACADDRS_OFST 20 +#define MC_CMD_VPORT_RECONFIGURE_IN_MACADDRS_LEN 6 +#define MC_CMD_VPORT_RECONFIGURE_IN_MACADDRS_NUM 4 + +/* MC_CMD_VPORT_RECONFIGURE_OUT msgresponse */ +#define MC_CMD_VPORT_RECONFIGURE_OUT_LEN 4 +#define MC_CMD_VPORT_RECONFIGURE_OUT_FLAGS_OFST 0 +#define MC_CMD_VPORT_RECONFIGURE_OUT_RESET_DONE_LBN 0 +#define MC_CMD_VPORT_RECONFIGURE_OUT_RESET_DONE_WIDTH 1 + + +/***********************************/ +/* MC_CMD_EVB_PORT_QUERY + * read some config of v-port. + */ +#define MC_CMD_EVB_PORT_QUERY 0x62 + +#define MC_CMD_0x62_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_EVB_PORT_QUERY_IN msgrequest */ +#define MC_CMD_EVB_PORT_QUERY_IN_LEN 4 +/* The handle of the v-port */ +#define MC_CMD_EVB_PORT_QUERY_IN_PORT_ID_OFST 0 + +/* MC_CMD_EVB_PORT_QUERY_OUT msgresponse */ +#define MC_CMD_EVB_PORT_QUERY_OUT_LEN 8 +/* The EVB port flags as defined at MC_CMD_VPORT_ALLOC. */ +#define MC_CMD_EVB_PORT_QUERY_OUT_PORT_FLAGS_OFST 0 +/* The number of VLAN tags that may be used on a v-adaptor connected to this + * EVB port. + */ +#define MC_CMD_EVB_PORT_QUERY_OUT_NUM_AVAILABLE_VLAN_TAGS_OFST 4 + + +/***********************************/ /* MC_CMD_DUMP_BUFTBL_ENTRIES * Dump buffer table entries, mainly for command client debug use. Dumps * absolute entries, and does not use chunk handles. All entries must be in @@ -8196,6 +9007,14 @@ #define MC_CMD_SET_RXDP_CONFIG_IN_DATA_OFST 0 #define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_DMA_LBN 0 #define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_DMA_WIDTH 1 +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_LEN_LBN 1 +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_LEN_WIDTH 2 +/* enum: pad to 64 bytes */ +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_64 0x0 +/* enum: pad to 128 bytes (Medford only) */ +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_128 0x1 +/* enum: pad to 256 bytes (Medford only) */ +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_256 0x2 /* MC_CMD_SET_RXDP_CONFIG_OUT msgresponse */ #define MC_CMD_SET_RXDP_CONFIG_OUT_LEN 0 @@ -8217,6 +9036,10 @@ #define MC_CMD_GET_RXDP_CONFIG_OUT_DATA_OFST 0 #define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_DMA_LBN 0 #define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_DMA_WIDTH 1 +#define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_LEN_LBN 1 +#define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_LEN_WIDTH 2 +/* Enum values, see field(s): */ +/* MC_CMD_SET_RXDP_CONFIG/MC_CMD_SET_RXDP_CONFIG_IN/PAD_HOST_LEN */ /***********************************/ @@ -8788,32 +9611,38 @@ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_MAXNUM 63 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_ID_LBN 0 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_ID_WIDTH 8 -/* enum: Attenuation (0-15, TBD for Medford) */ +/* enum: Attenuation (0-15, Huntington) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_ATT 0x0 -/* enum: CTLE Boost (0-15, TBD for Medford) */ +/* enum: CTLE Boost (0-15, Huntington) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_BOOST 0x1 -/* enum: Edge DFE Tap1 (0 - max negative, 64 - zero, 127 - max positive, TBD - * for Medford) +/* enum: Edge DFE Tap1 (Huntington - 0 - max negative, 64 - zero, 127 - max + * positive, Medford - 0-31) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP1 0x2 -/* enum: Edge DFE Tap2 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap2 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-31) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP2 0x3 -/* enum: Edge DFE Tap3 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap3 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-16) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP3 0x4 -/* enum: Edge DFE Tap4 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap4 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-16) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP4 0x5 -/* enum: Edge DFE Tap5 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap5 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-16) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP5 0x6 -/* enum: Edge DFE DLEV (TBD for Medford) */ +/* enum: Edge DFE DLEV (0-128 for Medford) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_DLEV 0x7 +/* enum: Variable Gain Amplifier (0-15, Medford) */ +#define MC_CMD_KR_TUNE_RXEQ_GET_OUT_VGA 0x8 +/* enum: CTLE EQ Capacitor (0-15, Medford) */ +#define MC_CMD_KR_TUNE_RXEQ_GET_OUT_CTLE_EQC 0x9 +/* enum: CTLE EQ Resistor (0-7, Medford) */ +#define MC_CMD_KR_TUNE_RXEQ_GET_OUT_CTLE_EQRES 0xa #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_LANE_LBN 8 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_LANE_WIDTH 3 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_LANE_0 0x0 /* enum */ @@ -8885,26 +9714,32 @@ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_MAXNUM 63 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_ID_LBN 0 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_ID_WIDTH 8 -/* enum: TX Amplitude */ +/* enum: TX Amplitude (Huntington, Medford) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_LEV 0x0 -/* enum: De-Emphasis Tap1 Magnitude (0-7) */ +/* enum: De-Emphasis Tap1 Magnitude (0-7) (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_MODE 0x1 /* enum: De-Emphasis Tap1 Fine */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_DTLEV 0x2 -/* enum: De-Emphasis Tap2 Magnitude (0-6) */ +/* enum: De-Emphasis Tap2 Magnitude (0-6) (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_D2 0x3 -/* enum: De-Emphasis Tap2 Fine */ +/* enum: De-Emphasis Tap2 Fine (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_D2TLEV 0x4 -/* enum: Pre-Emphasis Magnitude */ +/* enum: Pre-Emphasis Magnitude (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_E 0x5 -/* enum: Pre-Emphasis Fine */ +/* enum: Pre-Emphasis Fine (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_ETLEV 0x6 -/* enum: TX Slew Rate Coarse control */ +/* enum: TX Slew Rate Coarse control (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_PREDRV_DLY 0x7 -/* enum: TX Slew Rate Fine control */ +/* enum: TX Slew Rate Fine control (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_SR_SET 0x8 -/* enum: TX Termination Impedance control */ +/* enum: TX Termination Impedance control (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_RT_SET 0x9 +/* enum: TX Amplitude Fine control (Medford) */ +#define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_LEV_FINE 0xa +/* enum: Pre-shoot Tap (Medford) */ +#define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TAP_ADV 0xb +/* enum: De-emphasis Tap (Medford) */ +#define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TAP_DLY 0xc #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_LANE_LBN 8 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_LANE_WIDTH 3 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_LANE_0 0x0 /* enum */ @@ -9086,8 +9921,16 @@ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_DFE_TAP4 0x5 /* enum: DFE Tap5 (0 - max negative, 32 - zero, 63 - max positive) */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_DFE_TAP5 0x6 +/* enum: DFE DLev */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_DFE_DLEV 0x7 +/* enum: Figure of Merit */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_FOM 0x8 +/* enum: CTLE EQ Capacitor (HF Gain) */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_CTLE_EQC 0x9 +/* enum: CTLE EQ Resistor (DC Gain) */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_CTLE_EQRES 0xa #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_LANE_LBN 8 -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_LANE_WIDTH 4 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_LANE_WIDTH 5 #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_0 0x0 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_1 0x1 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_2 0x2 /* enum */ @@ -9096,12 +9939,57 @@ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_5 0x5 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_6 0x6 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_7 0x7 /* enum */ -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_ALL 0x8 /* enum */ -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_LBN 12 -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_WIDTH 12 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_8 0x8 /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_9 0x9 /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_10 0xa /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_11 0xb /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_12 0xc /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_13 0xd /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_14 0xe /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_15 0xf /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_ALL 0x10 /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_AUTOCAL_LBN 13 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_AUTOCAL_WIDTH 1 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_LBN 14 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_WIDTH 10 #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_CURRENT_LBN 24 #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_CURRENT_WIDTH 8 +/* MC_CMD_PCIE_TUNE_RXEQ_SET_IN msgrequest */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_LENMIN 8 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_LENMAX 252 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_LEN(num) (4+4*(num)) +/* Requested operation */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_OP_OFST 0 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_OP_LEN 1 +/* Align the arguments to 32 bits */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_RSVD_OFST 1 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_RSVD_LEN 3 +/* RXEQ Parameter */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_OFST 4 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_LEN 4 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_MINNUM 1 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_MAXNUM 62 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_ID_LBN 0 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_ID_WIDTH 8 +/* Enum values, see field(s): */ +/* MC_CMD_PCIE_TUNE_RXEQ_GET_OUT/PARAM_ID */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_LANE_LBN 8 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_LANE_WIDTH 5 +/* Enum values, see field(s): */ +/* MC_CMD_PCIE_TUNE_RXEQ_GET_OUT/PARAM_LANE */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_AUTOCAL_LBN 13 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_AUTOCAL_WIDTH 1 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED_LBN 14 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED_WIDTH 2 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_INITIAL_LBN 16 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_INITIAL_WIDTH 8 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED2_LBN 24 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED2_WIDTH 8 + +/* MC_CMD_PCIE_TUNE_RXEQ_SET_OUT msgresponse */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_OUT_LEN 0 + /* MC_CMD_PCIE_TUNE_TXEQ_GET_IN msgrequest */ #define MC_CMD_PCIE_TUNE_TXEQ_GET_IN_LEN 4 /* Requested operation */ @@ -9176,6 +10064,7 @@ /***********************************/ /* MC_CMD_LICENSING * Operations on the NVRAM_PARTITION_TYPE_LICENSE application license partition + * - not used for V3 licensing */ #define MC_CMD_LICENSING 0xf3 @@ -9220,6 +10109,93 @@ /***********************************/ +/* MC_CMD_LICENSING_V3 + * Operations on the NVRAM_PARTITION_TYPE_LICENSE application license partition + * - V3 licensing (Medford) + */ +#define MC_CMD_LICENSING_V3 0xd0 + +#define MC_CMD_0xd0_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSING_V3_IN msgrequest */ +#define MC_CMD_LICENSING_V3_IN_LEN 4 +/* identifies the type of operation requested */ +#define MC_CMD_LICENSING_V3_IN_OP_OFST 0 +/* enum: re-read and apply licenses after a license key partition update; note + * that this operation returns a zero-length response + */ +#define MC_CMD_LICENSING_V3_IN_OP_UPDATE_LICENSE 0x0 +/* enum: report counts of installed licenses */ +#define MC_CMD_LICENSING_V3_IN_OP_REPORT_LICENSE 0x1 + +/* MC_CMD_LICENSING_V3_OUT msgresponse */ +#define MC_CMD_LICENSING_V3_OUT_LEN 88 +/* count of keys which are valid */ +#define MC_CMD_LICENSING_V3_OUT_VALID_KEYS_OFST 0 +/* sum of UNVERIFIABLE_KEYS + WRONG_NODE_KEYS (for compatibility with + * MC_CMD_FC_OP_LICENSE) + */ +#define MC_CMD_LICENSING_V3_OUT_INVALID_KEYS_OFST 4 +/* count of keys which are invalid due to being unverifiable */ +#define MC_CMD_LICENSING_V3_OUT_UNVERIFIABLE_KEYS_OFST 8 +/* count of keys which are invalid due to being for the wrong node */ +#define MC_CMD_LICENSING_V3_OUT_WRONG_NODE_KEYS_OFST 12 +/* licensing state (for diagnostics; the exact meaning of the bits in this + * field are private to the firmware) + */ +#define MC_CMD_LICENSING_V3_OUT_LICENSING_STATE_OFST 16 +/* licensing subsystem self-test report (for manftest) */ +#define MC_CMD_LICENSING_V3_OUT_LICENSING_SELF_TEST_OFST 20 +/* enum: licensing subsystem self-test failed */ +#define MC_CMD_LICENSING_V3_OUT_SELF_TEST_FAIL 0x0 +/* enum: licensing subsystem self-test passed */ +#define MC_CMD_LICENSING_V3_OUT_SELF_TEST_PASS 0x1 +/* bitmask of licensed applications */ +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_OFST 24 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_LEN 8 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_LO_OFST 24 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_HI_OFST 28 +/* reserved for future use */ +#define MC_CMD_LICENSING_V3_OUT_RESERVED_0_OFST 32 +#define MC_CMD_LICENSING_V3_OUT_RESERVED_0_LEN 24 +/* bitmask of licensed features */ +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_OFST 56 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_LEN 8 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_LO_OFST 56 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_HI_OFST 60 +/* reserved for future use */ +#define MC_CMD_LICENSING_V3_OUT_RESERVED_1_OFST 64 +#define MC_CMD_LICENSING_V3_OUT_RESERVED_1_LEN 24 + + +/***********************************/ +/* MC_CMD_LICENSING_GET_ID_V3 + * Get ID and type from the NVRAM_PARTITION_TYPE_LICENSE application license + * partition - V3 licensing (Medford) + */ +#define MC_CMD_LICENSING_GET_ID_V3 0xd1 + +#define MC_CMD_0xd1_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSING_GET_ID_V3_IN msgrequest */ +#define MC_CMD_LICENSING_GET_ID_V3_IN_LEN 0 + +/* MC_CMD_LICENSING_GET_ID_V3_OUT msgresponse */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LENMIN 8 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LENMAX 252 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LEN(num) (8+1*(num)) +/* type of license (eg 3) */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_TYPE_OFST 0 +/* length of the license ID (in bytes) */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_LENGTH_OFST 4 +/* the unique license ID of the adapter */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_OFST 8 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_LEN 1 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_MINNUM 0 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_MAXNUM 244 + + +/***********************************/ /* MC_CMD_MC2MC_PROXY * Execute an arbitrary MCDI command on the slave MC of a dual-core device. * This will fail on a single-core system. @@ -9239,7 +10215,7 @@ /* MC_CMD_GET_LICENSED_APP_STATE * Query the state of an individual licensed application. (Note that the actual * state may be invalidated by the MC_CMD_LICENSING OP_UPDATE_LICENSE operation - * or a reboot of the MC.) + * or a reboot of the MC.) Not used for V3 licensing */ #define MC_CMD_GET_LICENSED_APP_STATE 0xf5 @@ -9261,8 +10237,68 @@ /***********************************/ +/* MC_CMD_GET_LICENSED_V3_APP_STATE + * Query the state of an individual licensed application. (Note that the actual + * state may be invalidated by the MC_CMD_LICENSING_V3 OP_UPDATE_LICENSE + * operation or a reboot of the MC.) Used for V3 licensing (Medford) + */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE 0xd2 + +#define MC_CMD_0xd2_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_GET_LICENSED_V3_APP_STATE_IN msgrequest */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_LEN 8 +/* application ID to query (LICENSED_V3_APPS_xxx) expressed as a single bit + * mask + */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_OFST 0 +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_LEN 8 +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_LO_OFST 0 +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_HI_OFST 4 + +/* MC_CMD_GET_LICENSED_V3_APP_STATE_OUT msgresponse */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_LEN 4 +/* state of this application */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_STATE_OFST 0 +/* enum: no (or invalid) license is present for the application */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_NOT_LICENSED 0x0 +/* enum: a valid license is present for the application */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_LICENSED 0x1 + + +/***********************************/ +/* MC_CMD_GET_LICENSED_V3_FEATURE_STATES + * Query the state of an one or more licensed features. (Note that the actual + * state may be invalidated by the MC_CMD_LICENSING_V3 OP_UPDATE_LICENSE + * operation or a reboot of the MC.) Used for V3 licensing (Medford) + */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES 0xd3 + +#define MC_CMD_0xd3_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN msgrequest */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_LEN 8 +/* features to query (LICENSED_V3_FEATURES_xxx) expressed as a mask with one or + * more bits set + */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_LEN 8 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_LO_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_HI_OFST 4 + +/* MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT msgresponse */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_LEN 8 +/* states of these features - bit set for licensed, clear for not licensed */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_LEN 8 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_LO_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_HI_OFST 4 + + +/***********************************/ /* MC_CMD_LICENSED_APP_OP - * Perform an action for an individual licensed application. + * Perform an action for an individual licensed application - not used for V3 + * licensing. */ #define MC_CMD_LICENSED_APP_OP 0xf6 @@ -9328,6 +10364,67 @@ /***********************************/ +/* MC_CMD_LICENSED_V3_VALIDATE_APP + * Perform validation for an individual licensed application - V3 licensing + * (Medford) + */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP 0xd4 + +#define MC_CMD_0xd4_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSED_V3_VALIDATE_APP_IN msgrequest */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_LEN 72 +/* application ID expressed as a single bit mask */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_OFST 0 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_LEN 8 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_LO_OFST 0 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_HI_OFST 4 +/* challenge for validation */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_CHALLENGE_OFST 8 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_CHALLENGE_LEN 64 + +/* MC_CMD_LICENSED_V3_VALIDATE_APP_OUT msgresponse */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_LEN 72 +/* application expiry time */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_TIME_OFST 0 +/* application expiry units */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_UNITS_OFST 4 +/* enum: expiry units are accounting units */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_UNIT_ACC 0x0 +/* enum: expiry units are calendar days */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_UNIT_DAYS 0x1 +/* validation response to challenge */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_RESPONSE_OFST 8 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_RESPONSE_LEN 64 + + +/***********************************/ +/* MC_CMD_LICENSED_V3_MASK_FEATURES + * Mask features - V3 licensing (Medford) + */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES 0xd5 + +#define MC_CMD_0xd5_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSED_V3_MASK_FEATURES_IN msgrequest */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_LEN 12 +/* mask to be applied to features to be changed */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_OFST 0 +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_LEN 8 +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_LO_OFST 0 +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_HI_OFST 4 +/* whether to turn on or turn off the masked features */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_FLAG_OFST 8 +/* enum: turn the features off */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_OFF 0x0 +/* enum: turn the features back on */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_ON 0x1 + +/* MC_CMD_LICENSED_V3_MASK_FEATURES_OUT msgresponse */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_OUT_LEN 0 + + +/***********************************/ /* MC_CMD_SET_PORT_SNIFF_CONFIG * Configure RX port sniffing for the physical port associated with the calling * function. Only a privileged function may change the port sniffing @@ -9696,12 +10793,27 @@ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_ONLOAD 0x4 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_PTP 0x8 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_INSECURE_FILTERS 0x10 /* enum */ -#define MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING 0x20 /* enum */ +/* enum: Deprecated. Equivalent to MAC_SPOOFING_TX combined with CHANGE_MAC. */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING 0x20 #define MC_CMD_PRIVILEGE_MASK_IN_GRP_UNICAST 0x40 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_MULTICAST 0x80 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_BROADCAST 0x100 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_ALL_MULTICAST 0x200 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_PROMISCUOUS 0x400 /* enum */ +/* enum: Allows to set the TX packets' source MAC address to any arbitrary MAC + * adress. + */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING_TX 0x800 +/* enum: Privilege that allows a Function to change the MAC address configured + * in its associated vAdapter/vPort. + */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_CHANGE_MAC 0x1000 +/* enum: Privilege that allows a Function to install filters that specify VLANs + * that are not in the permit list for the associated vPort. This privilege is + * primarily to support ESX where vPorts are created that restrict traffic to + * only a set of permitted VLANs. See the vPort flag FLAG_VLAN_RESTRICT. + */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_UNRESTRICTED_VLAN 0x2000 /* enum: Set this bit to indicate that a new privilege mask is to be set, * otherwise the command will only read the existing mask. */ @@ -9951,7 +11063,7 @@ /* Sector type */ #define MC_CMD_XPM_WRITE_SECTOR_IN_TYPE_OFST 4 /* Enum values, see field(s): */ -/* MC_CMD_XPM_READ_SECTOR_OUT/TYPE */ +/* MC_CMD_XPM_READ_SECTOR/MC_CMD_XPM_READ_SECTOR_OUT/TYPE */ /* Sector size */ #define MC_CMD_XPM_WRITE_SECTOR_IN_SIZE_OFST 8 /* Sector data */ @@ -10067,4 +11179,123 @@ #define MC_CMD_XPM_WRITE_TEST_OUT_LEN 0 +/***********************************/ +/* MC_CMD_EXEC_SIGNED + * Check the CMAC of the contents of IMEM and DMEM against the value supplied + * and if correct begin execution from the start of IMEM. The caller supplies a + * key ID, the length of IMEM and DMEM to validate and the expected CMAC. CMAC + * computation runs from the start of IMEM, and from the start of DMEM + 16k, + * to match flash booting. The command will respond with EINVAL if the CMAC + * does match, otherwise it will respond with success before it jumps to IMEM. + */ +#define MC_CMD_EXEC_SIGNED 0x10c + +#define MC_CMD_0x10c_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_EXEC_SIGNED_IN msgrequest */ +#define MC_CMD_EXEC_SIGNED_IN_LEN 28 +/* the length of code to include in the CMAC */ +#define MC_CMD_EXEC_SIGNED_IN_CODELEN_OFST 0 +/* the length of date to include in the CMAC */ +#define MC_CMD_EXEC_SIGNED_IN_DATALEN_OFST 4 +/* the XPM sector containing the key to use */ +#define MC_CMD_EXEC_SIGNED_IN_KEYSECTOR_OFST 8 +/* the expected CMAC value */ +#define MC_CMD_EXEC_SIGNED_IN_CMAC_OFST 12 +#define MC_CMD_EXEC_SIGNED_IN_CMAC_LEN 16 + +/* MC_CMD_EXEC_SIGNED_OUT msgresponse */ +#define MC_CMD_EXEC_SIGNED_OUT_LEN 0 + + +/***********************************/ +/* MC_CMD_PREPARE_SIGNED + * Prepare to upload a signed image. This will scrub the specified length of + * the data region, which must be at least as large as the DATALEN supplied to + * MC_CMD_EXEC_SIGNED. + */ +#define MC_CMD_PREPARE_SIGNED 0x10d + +#define MC_CMD_0x10d_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_PREPARE_SIGNED_IN msgrequest */ +#define MC_CMD_PREPARE_SIGNED_IN_LEN 4 +/* the length of data area to clear */ +#define MC_CMD_PREPARE_SIGNED_IN_DATALEN_OFST 0 + +/* MC_CMD_PREPARE_SIGNED_OUT msgresponse */ +#define MC_CMD_PREPARE_SIGNED_OUT_LEN 0 + + +/***********************************/ +/* MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS + * Configure UDP ports for tunnel encapsulation hardware acceleration. The + * parser-dispatcher will attempt to parse traffic on these ports as tunnel + * encapsulation PDUs and filter them using the tunnel encapsulation filter + * chain rather than the standard filter chain. Note that this command can + * cause all functions to see a reset. (Available on Medford only.) + */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS 0x117 + +#define MC_CMD_0x117_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN msgrequest */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_LENMIN 4 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_LENMAX 68 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_LEN(num) (4+4*(num)) +/* Flags */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_FLAGS_OFST 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_FLAGS_LEN 2 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_UNLOADING_LBN 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_UNLOADING_WIDTH 1 +/* The number of entries in the ENTRIES array */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_NUM_ENTRIES_OFST 2 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_NUM_ENTRIES_LEN 2 +/* Entries defining the UDP port to protocol mapping, each laid out as a + * TUNNEL_ENCAP_UDP_PORT_ENTRY + */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_OFST 4 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_LEN 4 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_MINNUM 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_MAXNUM 16 + +/* MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT msgresponse */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_LEN 2 +/* Flags */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_FLAGS_OFST 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_FLAGS_LEN 2 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_RESETTING_LBN 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_RESETTING_WIDTH 1 + + +/***********************************/ +/* MC_CMD_RX_BALANCING + * Configure a port upconverter to distribute the packets on both RX engines. + * Packets are distributed based on a table with the destination vFIFO. The + * index of the table is a hash of source and destination of IPV4 and VLAN + * priority. + */ +#define MC_CMD_RX_BALANCING 0x118 + +#define MC_CMD_0x118_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_RX_BALANCING_IN msgrequest */ +#define MC_CMD_RX_BALANCING_IN_LEN 4 +/* The RX port whose upconverter table will be modified */ +#define MC_CMD_RX_BALANCING_IN_PORT_OFST 0 +#define MC_CMD_RX_BALANCING_IN_PORT_LEN 1 +/* The VLAN priority associated to the table index and vFIFO */ +#define MC_CMD_RX_BALANCING_IN_PRIORITY_OFST 1 +#define MC_CMD_RX_BALANCING_IN_PRIORITY_LEN 1 +/* The resulting bit of SRC^DST for indexing the table */ +#define MC_CMD_RX_BALANCING_IN_SRC_DST_OFST 2 +#define MC_CMD_RX_BALANCING_IN_SRC_DST_LEN 1 +/* The RX engine to which the vFIFO in the table entry will point to */ +#define MC_CMD_RX_BALANCING_IN_ENG_OFST 3 +#define MC_CMD_RX_BALANCING_IN_ENG_LEN 1 + +/* MC_CMD_RX_BALANCING_OUT msgresponse */ +#define MC_CMD_RX_BALANCING_OUT_LEN 0 + + #endif /* MCDI_PCOL_H */ diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c index 7f295c4d7b80..2a9228a6e4a0 100644 --- a/drivers/net/ethernet/sfc/mcdi_port.c +++ b/drivers/net/ethernet/sfc/mcdi_port.c @@ -189,11 +189,12 @@ static u32 mcdi_to_ethtool_cap(u32 media, u32 cap) case MC_CMD_MEDIA_XFP: case MC_CMD_MEDIA_SFP_PLUS: - result |= SUPPORTED_FIBRE; - break; - case MC_CMD_MEDIA_QSFP_PLUS: result |= SUPPORTED_FIBRE; + if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN)) + result |= SUPPORTED_1000baseT_Full; + if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN)) + result |= SUPPORTED_10000baseT_Full; if (cap & (1 << MC_CMD_PHY_CAP_40000FDX_LBN)) result |= SUPPORTED_40000baseCR4_Full; break; diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index d13ddf9703ff..9ff062a36ea8 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -868,6 +868,7 @@ struct vfdi_status; * be held to modify it. * @port_initialized: Port initialized? * @net_dev: Operating system network device. Consider holding the rtnl lock + * @fixed_features: Features which cannot be turned off * @stats_buffer: DMA buffer for statistics * @phy_type: PHY type * @phy_op: PHY interface @@ -916,7 +917,6 @@ struct vfdi_status; * @stats_lock: Statistics update lock. Must be held when calling * efx_nic_type::{update,start,stop}_stats. * @n_rx_noskb_drops: Count of RX packets dropped due to failure to allocate an skb - * @mc_promisc: Whether in multicast promiscuous mode when last changed * * This is stored in the private area of the &struct net_device. */ @@ -1008,6 +1008,8 @@ struct efx_nic { bool port_initialized; struct net_device *net_dev; + netdev_features_t fixed_features; + struct efx_buffer stats_buffer; u64 rx_nodesc_drops_total; u64 rx_nodesc_drops_while_down; @@ -1065,7 +1067,6 @@ struct efx_nic { int last_irq_cpu; spinlock_t stats_lock; atomic_t n_rx_noskb_drops; - bool mc_promisc; }; static inline int efx_dev_registered(struct efx_nic *efx) @@ -1333,6 +1334,8 @@ struct efx_nic_type { int (*ptp_set_ts_config)(struct efx_nic *efx, struct hwtstamp_config *init); int (*sriov_configure)(struct efx_nic *efx, int num_vfs); + int (*vlan_rx_add_vid)(struct efx_nic *efx, __be16 proto, u16 vid); + int (*vlan_rx_kill_vid)(struct efx_nic *efx, __be16 proto, u16 vid); int (*sriov_init)(struct efx_nic *efx); void (*sriov_fini)(struct efx_nic *efx); bool (*sriov_wanted)(struct efx_nic *efx); @@ -1521,4 +1524,16 @@ static inline void efx_xmit_hwtstamp_pending(struct sk_buff *skb) skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; } +/* Get all supported features. + * If a feature is not fixed, it is present in hw_features. + * If a feature is fixed, it does not present in hw_features, but + * always in features. + */ +static inline netdev_features_t efx_supported_features(const struct efx_nic *efx) +{ + const struct net_device *net_dev = efx->net_dev; + + return net_dev->features | net_dev->hw_features; +} + #endif /* EFX_NET_DRIVER_H */ diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h index 0b536e27d3b2..96944c3c9d14 100644 --- a/drivers/net/ethernet/sfc/nic.h +++ b/drivers/net/ethernet/sfc/nic.h @@ -519,6 +519,9 @@ enum { #ifdef CONFIG_SFC_SRIOV * @vf: Pointer to VF data structure #endif + * @vport_mac: The MAC address on the vport, only for PFs; VFs will be zero + * @vlan_list: List of VLANs added over the interface. Serialised by vlan_lock. + * @vlan_lock: Lock to serialize access to vlan_list. */ struct efx_ef10_nic_data { struct efx_buffer mcdi_buf; @@ -550,6 +553,8 @@ struct efx_ef10_nic_data { struct ef10_vf *vf; #endif u8 vport_mac[ETH_ALEN]; + struct list_head vlan_list; + struct mutex vlan_lock; }; int efx_init_sriov(void); diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 8af25563f627..ca3134540d2d 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -114,9 +114,7 @@ struct smsc911x_data { /* spinlock to ensure register accesses are serialised */ spinlock_t dev_lock; - struct phy_device *phy_dev; struct mii_bus *mii_bus; - int phy_irq[PHY_MAX_ADDR]; unsigned int using_extphy; int last_duplex; int last_carrier; @@ -834,7 +832,7 @@ static int smsc911x_phy_reset(struct smsc911x_data *pdata) static int smsc911x_phy_loopbacktest(struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); - struct phy_device *phy_dev = pdata->phy_dev; + struct phy_device *phy_dev = dev->phydev; int result = -EIO; unsigned int i, val; unsigned long flags; @@ -904,7 +902,8 @@ static int smsc911x_phy_loopbacktest(struct net_device *dev) static void smsc911x_phy_update_flowcontrol(struct smsc911x_data *pdata) { - struct phy_device *phy_dev = pdata->phy_dev; + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; u32 afc = smsc911x_reg_read(pdata, AFC_CFG); u32 flow; unsigned long flags; @@ -945,7 +944,7 @@ static void smsc911x_phy_update_flowcontrol(struct smsc911x_data *pdata) static void smsc911x_phy_adjust_link(struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); - struct phy_device *phy_dev = pdata->phy_dev; + struct phy_device *phy_dev = dev->phydev; unsigned long flags; int carrier; @@ -1038,7 +1037,6 @@ static int smsc911x_mii_probe(struct net_device *dev) SUPPORTED_Asym_Pause); phydev->advertising = phydev->supported; - pdata->phy_dev = phydev; pdata->last_duplex = -1; pdata->last_carrier = -1; @@ -1073,7 +1071,6 @@ static int smsc911x_mii_init(struct platform_device *pdev, pdata->mii_bus->priv = pdata; pdata->mii_bus->read = smsc911x_mii_read; pdata->mii_bus->write = smsc911x_mii_write; - memcpy(pdata->mii_bus->irq, pdata->phy_irq, sizeof(pdata->mii_bus)); pdata->mii_bus->parent = &pdev->dev; @@ -1340,9 +1337,11 @@ static void smsc911x_rx_multicast_update_workaround(struct smsc911x_data *pdata) static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) { + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; int rc = 0; - if (!pdata->phy_dev) + if (!phy_dev) return rc; /* If the internal PHY is in General Power-Down mode, all, except the @@ -1352,7 +1351,7 @@ static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) * In that case, clear the bit 0.11, so the PHY powers up and we can * access to the phy registers. */ - rc = phy_read(pdata->phy_dev, MII_BMCR); + rc = phy_read(phy_dev, MII_BMCR); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed reading PHY control reg"); return rc; @@ -1362,7 +1361,7 @@ static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) * disable the general power down-mode. */ if (rc & BMCR_PDOWN) { - rc = phy_write(pdata->phy_dev, MII_BMCR, rc & ~BMCR_PDOWN); + rc = phy_write(phy_dev, MII_BMCR, rc & ~BMCR_PDOWN); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed writing PHY control reg"); return rc; @@ -1376,12 +1375,14 @@ static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) static int smsc911x_phy_disable_energy_detect(struct smsc911x_data *pdata) { + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; int rc = 0; - if (!pdata->phy_dev) + if (!phy_dev) return rc; - rc = phy_read(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS); + rc = phy_read(phy_dev, MII_LAN83C185_CTRL_STATUS); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed reading PHY control reg"); @@ -1391,7 +1392,7 @@ static int smsc911x_phy_disable_energy_detect(struct smsc911x_data *pdata) /* Only disable if energy detect mode is already enabled */ if (rc & MII_LAN83C185_EDPWRDOWN) { /* Disable energy detect mode for this SMSC Transceivers */ - rc = phy_write(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS, + rc = phy_write(phy_dev, MII_LAN83C185_CTRL_STATUS, rc & (~MII_LAN83C185_EDPWRDOWN)); if (rc < 0) { @@ -1407,12 +1408,14 @@ static int smsc911x_phy_disable_energy_detect(struct smsc911x_data *pdata) static int smsc911x_phy_enable_energy_detect(struct smsc911x_data *pdata) { + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; int rc = 0; - if (!pdata->phy_dev) + if (!phy_dev) return rc; - rc = phy_read(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS); + rc = phy_read(phy_dev, MII_LAN83C185_CTRL_STATUS); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed reading PHY control reg"); @@ -1422,7 +1425,7 @@ static int smsc911x_phy_enable_energy_detect(struct smsc911x_data *pdata) /* Only enable if energy detect mode is already disabled */ if (!(rc & MII_LAN83C185_EDPWRDOWN)) { /* Enable energy detect mode for this SMSC Transceivers */ - rc = phy_write(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS, + rc = phy_write(phy_dev, MII_LAN83C185_CTRL_STATUS, rc | MII_LAN83C185_EDPWRDOWN); if (rc < 0) { @@ -1519,7 +1522,7 @@ static int smsc911x_open(struct net_device *dev) unsigned int intcfg; /* if the phy is not yet registered, retry later*/ - if (!pdata->phy_dev) { + if (!dev->phydev) { SMSC_WARN(pdata, hw, "phy_dev is NULL"); return -EAGAIN; } @@ -1610,7 +1613,7 @@ static int smsc911x_open(struct net_device *dev) pdata->last_carrier = -1; /* Bring the PHY up */ - phy_start(pdata->phy_dev); + phy_start(dev->phydev); temp = smsc911x_reg_read(pdata, HW_CFG); /* Preserve TX FIFO size and external PHY configuration */ @@ -1665,8 +1668,8 @@ static int smsc911x_stop(struct net_device *dev) smsc911x_tx_update_txcounters(dev); /* Bring the PHY down */ - if (pdata->phy_dev) - phy_stop(pdata->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); SMSC_TRACE(pdata, ifdown, "Interface stopped"); return 0; @@ -1906,30 +1909,10 @@ static int smsc911x_set_mac_address(struct net_device *dev, void *p) /* Standard ioctls for mii-tool */ static int smsc911x_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct smsc911x_data *pdata = netdev_priv(dev); - - if (!netif_running(dev) || !pdata->phy_dev) + if (!netif_running(dev) || !dev->phydev) return -EINVAL; - return phy_mii_ioctl(pdata->phy_dev, ifr, cmd); -} - -static int -smsc911x_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct smsc911x_data *pdata = netdev_priv(dev); - - cmd->maxtxpkt = 1; - cmd->maxrxpkt = 1; - return phy_ethtool_gset(pdata->phy_dev, cmd); -} - -static int -smsc911x_ethtool_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct smsc911x_data *pdata = netdev_priv(dev); - - return phy_ethtool_sset(pdata->phy_dev, cmd); + return phy_mii_ioctl(dev->phydev, ifr, cmd); } static void smsc911x_ethtool_getdrvinfo(struct net_device *dev, @@ -1943,9 +1926,7 @@ static void smsc911x_ethtool_getdrvinfo(struct net_device *dev, static int smsc911x_ethtool_nwayreset(struct net_device *dev) { - struct smsc911x_data *pdata = netdev_priv(dev); - - return phy_start_aneg(pdata->phy_dev); + return phy_start_aneg(dev->phydev); } static u32 smsc911x_ethtool_getmsglevel(struct net_device *dev) @@ -1971,7 +1952,7 @@ smsc911x_ethtool_getregs(struct net_device *dev, struct ethtool_regs *regs, void *buf) { struct smsc911x_data *pdata = netdev_priv(dev); - struct phy_device *phy_dev = pdata->phy_dev; + struct phy_device *phy_dev = dev->phydev; unsigned long flags; unsigned int i; unsigned int j = 0; @@ -2117,8 +2098,6 @@ static int smsc911x_ethtool_set_eeprom(struct net_device *dev, } static const struct ethtool_ops smsc911x_ethtool_ops = { - .get_settings = smsc911x_ethtool_getsettings, - .set_settings = smsc911x_ethtool_setsettings, .get_link = ethtool_op_get_link, .get_drvinfo = smsc911x_ethtool_getdrvinfo, .nway_reset = smsc911x_ethtool_nwayreset, @@ -2130,6 +2109,8 @@ static const struct ethtool_ops smsc911x_ethtool_ops = { .get_eeprom = smsc911x_ethtool_get_eeprom, .set_eeprom = smsc911x_ethtool_set_eeprom, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops smsc911x_netdev_ops = { @@ -2310,12 +2291,11 @@ static int smsc911x_drv_remove(struct platform_device *pdev) pdata = netdev_priv(dev); BUG_ON(!pdata); BUG_ON(!pdata->ioaddr); - BUG_ON(!pdata->phy_dev); + BUG_ON(!dev->phydev); SMSC_TRACE(pdata, ifdown, "Stopping driver"); - phy_disconnect(pdata->phy_dev); - pdata->phy_dev = NULL; + phy_disconnect(dev->phydev); mdiobus_unregister(pdata->mii_bus); mdiobus_free(pdata->mii_bus); diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index fc60368df2e7..2533b91f1421 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -232,6 +232,11 @@ struct stmmac_extra_stats { #define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY iface */ #define DEFAULT_DMA_PBL 8 +/* PCS status and mask defines */ +#define PCS_ANE_IRQ BIT(2) /* PCS Auto-Negotiation */ +#define PCS_LINK_IRQ BIT(1) /* PCS Link */ +#define PCS_RGSMIIIS_IRQ BIT(0) /* RGMII or SMII Interrupt */ + /* Max/Min RI Watchdog Timer count value */ #define MAX_DMA_RIWT 0xff #define MIN_DMA_RIWT 0x20 @@ -272,9 +277,6 @@ enum dma_irq_status { #define CORE_IRQ_RX_PATH_IN_LPI_MODE (1 << 2) #define CORE_IRQ_RX_PATH_EXIT_LPI_MODE (1 << 3) -#define CORE_PCS_ANE_COMPLETE (1 << 5) -#define CORE_PCS_LINK_STATUS (1 << 6) -#define CORE_RGMII_IRQ (1 << 7) #define CORE_IRQ_MTL_RX_OVERFLOW BIT(8) /* Physical Coding Sublayer */ @@ -469,9 +471,12 @@ struct stmmac_ops { void (*reset_eee_mode)(struct mac_device_info *hw); void (*set_eee_timer)(struct mac_device_info *hw, int ls, int tw); void (*set_eee_pls)(struct mac_device_info *hw, int link); - void (*ctrl_ane)(struct mac_device_info *hw, bool restart); - void (*get_adv)(struct mac_device_info *hw, struct rgmii_adv *adv); void (*debug)(void __iomem *ioaddr, struct stmmac_extra_stats *x); + /* PCS calls */ + void (*pcs_ctrl_ane)(void __iomem *ioaddr, bool ane, bool srgmi_ral, + bool loopback); + void (*pcs_rane)(void __iomem *ioaddr, bool restart); + void (*pcs_get_adv_lp)(void __iomem *ioaddr, struct rgmii_adv *adv); }; /* PTP and HW Timer helpers */ @@ -524,6 +529,9 @@ struct mac_device_info { int unicast_filter_entries; int mcast_bits_log2; unsigned int rx_csum; + unsigned int pcs; + unsigned int pmt; + unsigned int ps; }; struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr, int mcbins, @@ -546,6 +554,7 @@ void stmmac_dwmac4_get_mac_addr(void __iomem *ioaddr, unsigned char *addr, void stmmac_dwmac4_set_mac(void __iomem *ioaddr, bool enable); void dwmac_dma_flush_tx_fifo(void __iomem *ioaddr); + extern const struct stmmac_mode_ops ring_mode_ops; extern const struct stmmac_mode_ops chain_mode_ops; extern const struct stmmac_desc_ops dwmac4_desc_ops; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 0cd3ecff768b..92105916ef40 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -46,6 +46,7 @@ struct rk_priv_data { struct platform_device *pdev; int phy_iface; struct regulator *regulator; + bool suspended; const struct rk_gmac_ops *ops; bool clk_enabled; @@ -72,6 +73,122 @@ struct rk_priv_data { #define GRF_BIT(nr) (BIT(nr) | BIT(nr+16)) #define GRF_CLR_BIT(nr) (BIT(nr+16)) +#define RK3228_GRF_MAC_CON0 0x0900 +#define RK3228_GRF_MAC_CON1 0x0904 + +/* RK3228_GRF_MAC_CON0 */ +#define RK3228_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 7) +#define RK3228_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) + +/* RK3228_GRF_MAC_CON1 */ +#define RK3228_GMAC_PHY_INTF_SEL_RGMII \ + (GRF_BIT(4) | GRF_CLR_BIT(5) | GRF_CLR_BIT(6)) +#define RK3228_GMAC_PHY_INTF_SEL_RMII \ + (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | GRF_BIT(6)) +#define RK3228_GMAC_FLOW_CTRL GRF_BIT(3) +#define RK3228_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) +#define RK3228_GMAC_SPEED_10M GRF_CLR_BIT(2) +#define RK3228_GMAC_SPEED_100M GRF_BIT(2) +#define RK3228_GMAC_RMII_CLK_25M GRF_BIT(7) +#define RK3228_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(7) +#define RK3228_GMAC_CLK_125M (GRF_CLR_BIT(8) | GRF_CLR_BIT(9)) +#define RK3228_GMAC_CLK_25M (GRF_BIT(8) | GRF_BIT(9)) +#define RK3228_GMAC_CLK_2_5M (GRF_CLR_BIT(8) | GRF_BIT(9)) +#define RK3228_GMAC_RMII_MODE GRF_BIT(10) +#define RK3228_GMAC_RMII_MODE_CLR GRF_CLR_BIT(10) +#define RK3228_GMAC_TXCLK_DLY_ENABLE GRF_BIT(0) +#define RK3228_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(0) +#define RK3228_GMAC_RXCLK_DLY_ENABLE GRF_BIT(1) +#define RK3228_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(1) + +static void rk3228_set_to_rgmii(struct rk_priv_data *bsp_priv, + int tx_delay, int rx_delay) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_PHY_INTF_SEL_RGMII | + RK3228_GMAC_RMII_MODE_CLR | + RK3228_GMAC_RXCLK_DLY_ENABLE | + RK3228_GMAC_TXCLK_DLY_ENABLE); + + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON0, + RK3228_GMAC_CLK_RX_DL_CFG(rx_delay) | + RK3228_GMAC_CLK_TX_DL_CFG(tx_delay)); +} + +static void rk3228_set_to_rmii(struct rk_priv_data *bsp_priv) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_PHY_INTF_SEL_RMII | + RK3228_GMAC_RMII_MODE); + + /* set MAC to RMII mode */ + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, GRF_BIT(11)); +} + +static void rk3228_set_rgmii_speed(struct rk_priv_data *bsp_priv, int speed) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + if (speed == 10) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_CLK_2_5M); + else if (speed == 100) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_CLK_25M); + else if (speed == 1000) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_CLK_125M); + else + dev_err(dev, "unknown speed value for RGMII! speed=%d", speed); +} + +static void rk3228_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + if (speed == 10) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_RMII_CLK_2_5M | + RK3228_GMAC_SPEED_10M); + else if (speed == 100) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_RMII_CLK_25M | + RK3228_GMAC_SPEED_100M); + else + dev_err(dev, "unknown speed value for RMII! speed=%d", speed); +} + +static const struct rk_gmac_ops rk3228_ops = { + .set_to_rgmii = rk3228_set_to_rgmii, + .set_to_rmii = rk3228_set_to_rmii, + .set_rgmii_speed = rk3228_set_rgmii_speed, + .set_rmii_speed = rk3228_set_rmii_speed, +}; + #define RK3288_GRF_SOC_CON1 0x0248 #define RK3288_GRF_SOC_CON3 0x0250 @@ -529,9 +646,8 @@ static struct rk_priv_data *rk_gmac_setup(struct platform_device *pdev, return bsp_priv; } -static int rk_gmac_init(struct platform_device *pdev, void *priv) +static int rk_gmac_powerup(struct rk_priv_data *bsp_priv) { - struct rk_priv_data *bsp_priv = priv; int ret; ret = phy_power_on(bsp_priv, true); @@ -545,14 +661,50 @@ static int rk_gmac_init(struct platform_device *pdev, void *priv) return 0; } -static void rk_gmac_exit(struct platform_device *pdev, void *priv) +static void rk_gmac_powerdown(struct rk_priv_data *gmac) { - struct rk_priv_data *gmac = priv; - phy_power_on(gmac, false); gmac_clk_enable(gmac, false); } +static int rk_gmac_init(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + return rk_gmac_powerup(bsp_priv); +} + +static void rk_gmac_exit(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + rk_gmac_powerdown(bsp_priv); +} + +static void rk_gmac_suspend(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + /* Keep the PHY up if we use Wake-on-Lan. */ + if (device_may_wakeup(&pdev->dev)) + return; + + rk_gmac_powerdown(bsp_priv); + bsp_priv->suspended = true; +} + +static void rk_gmac_resume(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + /* The PHY was up for Wake-on-Lan. */ + if (!bsp_priv->suspended) + return; + + rk_gmac_powerup(bsp_priv); + bsp_priv->suspended = false; +} + static void rk_fix_speed(void *priv, unsigned int speed) { struct rk_priv_data *bsp_priv = priv; @@ -591,6 +743,8 @@ static int rk_gmac_probe(struct platform_device *pdev) plat_dat->init = rk_gmac_init; plat_dat->exit = rk_gmac_exit; plat_dat->fix_mac_speed = rk_fix_speed; + plat_dat->suspend = rk_gmac_suspend; + plat_dat->resume = rk_gmac_resume; plat_dat->bsp_priv = rk_gmac_setup(pdev, data); if (IS_ERR(plat_dat->bsp_priv)) @@ -604,6 +758,7 @@ static int rk_gmac_probe(struct platform_device *pdev) } static const struct of_device_id rk_gmac_dwmac_match[] = { + { .compatible = "rockchip,rk3228-gmac", .data = &rk3228_ops }, { .compatible = "rockchip,rk3288-gmac", .data = &rk3288_ops }, { .compatible = "rockchip,rk3368-gmac", .data = &rk3368_ops }, { } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h index b0593a4268ee..ff3e5ab39bd0 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h @@ -38,19 +38,26 @@ #define GMAC_WAKEUP_FILTER 0x00000028 /* Wake-up Frame Filter */ #define GMAC_INT_STATUS 0x00000038 /* interrupt status register */ -enum dwmac1000_irq_status { - lpiis_irq = 0x400, - time_stamp_irq = 0x0200, - mmc_rx_csum_offload_irq = 0x0080, - mmc_tx_irq = 0x0040, - mmc_rx_irq = 0x0020, - mmc_irq = 0x0010, - pmt_irq = 0x0008, - pcs_ane_irq = 0x0004, - pcs_link_irq = 0x0002, - rgmii_irq = 0x0001, -}; -#define GMAC_INT_MASK 0x0000003c /* interrupt mask register */ +#define GMAC_INT_STATUS_PMT BIT(3) +#define GMAC_INT_STATUS_MMCIS BIT(4) +#define GMAC_INT_STATUS_MMCRIS BIT(5) +#define GMAC_INT_STATUS_MMCTIS BIT(6) +#define GMAC_INT_STATUS_MMCCSUM BIT(7) +#define GMAC_INT_STATUS_TSTAMP BIT(9) +#define GMAC_INT_STATUS_LPIIS BIT(10) + +/* interrupt mask register */ +#define GMAC_INT_MASK 0x0000003c +#define GMAC_INT_DISABLE_RGMII BIT(0) +#define GMAC_INT_DISABLE_PCSLINK BIT(1) +#define GMAC_INT_DISABLE_PCSAN BIT(2) +#define GMAC_INT_DISABLE_PMT BIT(3) +#define GMAC_INT_DISABLE_TIMESTAMP BIT(9) +#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_RGMII | \ + GMAC_INT_DISABLE_PCSLINK | \ + GMAC_INT_DISABLE_PCSAN) +#define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_TIMESTAMP | \ + GMAC_INT_DISABLE_PCS) /* PMT Control and Status */ #define GMAC_PMT 0x0000002c @@ -90,42 +97,23 @@ enum power_event { (reg * 8)) #define GMAC_MAX_PERFECT_ADDRESSES 1 -/* PCS registers (AN/TBI/SGMII/RGMII) offset */ -#define GMAC_AN_CTRL 0x000000c0 /* AN control */ -#define GMAC_AN_STATUS 0x000000c4 /* AN status */ -#define GMAC_ANE_ADV 0x000000c8 /* Auto-Neg. Advertisement */ -#define GMAC_ANE_LPA 0x000000cc /* Auto-Neg. link partener ability */ -#define GMAC_ANE_EXP 0x000000d0 /* ANE expansion */ -#define GMAC_TBI 0x000000d4 /* TBI extend status */ -#define GMAC_S_R_GMII 0x000000d8 /* SGMII RGMII status */ - -/* AN Configuration defines */ -#define GMAC_AN_CTRL_RAN 0x00000200 /* Restart Auto-Negotiation */ -#define GMAC_AN_CTRL_ANE 0x00001000 /* Auto-Negotiation Enable */ -#define GMAC_AN_CTRL_ELE 0x00004000 /* External Loopback Enable */ -#define GMAC_AN_CTRL_ECD 0x00010000 /* Enable Comma Detect */ -#define GMAC_AN_CTRL_LR 0x00020000 /* Lock to Reference */ -#define GMAC_AN_CTRL_SGMRAL 0x00040000 /* SGMII RAL Control */ - -/* AN Status defines */ -#define GMAC_AN_STATUS_LS 0x00000004 /* Link Status 0:down 1:up */ -#define GMAC_AN_STATUS_ANA 0x00000008 /* Auto-Negotiation Ability */ -#define GMAC_AN_STATUS_ANC 0x00000020 /* Auto-Negotiation Complete */ -#define GMAC_AN_STATUS_ES 0x00000100 /* Extended Status */ - -/* Register 54 (SGMII/RGMII status register) */ -#define GMAC_S_R_GMII_LINK 0x8 -#define GMAC_S_R_GMII_SPEED 0x5 -#define GMAC_S_R_GMII_SPEED_SHIFT 0x1 -#define GMAC_S_R_GMII_MODE 0x1 -#define GMAC_S_R_GMII_SPEED_125 2 -#define GMAC_S_R_GMII_SPEED_25 1 - -/* Common ADV and LPA defines */ -#define GMAC_ANE_FD (1 << 5) -#define GMAC_ANE_HD (1 << 6) -#define GMAC_ANE_PSE (3 << 7) -#define GMAC_ANE_PSE_SHIFT 7 +#define GMAC_PCS_BASE 0x000000c0 /* PCS register base */ +#define GMAC_RGSMIIIS 0x000000d8 /* RGMII/SMII status */ + +/* SGMII/RGMII status register */ +#define GMAC_RGSMIIIS_LNKMODE BIT(0) +#define GMAC_RGSMIIIS_SPEED GENMASK(2, 1) +#define GMAC_RGSMIIIS_SPEED_SHIFT 1 +#define GMAC_RGSMIIIS_LNKSTS BIT(3) +#define GMAC_RGSMIIIS_JABTO BIT(4) +#define GMAC_RGSMIIIS_FALSECARDET BIT(5) +#define GMAC_RGSMIIIS_SMIDRXS BIT(16) +/* LNKMOD */ +#define GMAC_RGSMIIIS_LNKMOD_MASK 0x1 +/* LNKSPEED */ +#define GMAC_RGSMIIIS_SPEED_125 0x2 +#define GMAC_RGSMIIIS_SPEED_25 0x1 +#define GMAC_RGSMIIIS_SPEED_2_5 0x0 /* GMAC Configuration defines */ #define GMAC_CONTROL_2K 0x08000000 /* IEEE 802.3as 2K packets */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index fb1eb578e34e..cbefe9e2207c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -30,22 +30,48 @@ #include <linux/slab.h> #include <linux/ethtool.h> #include <asm/io.h> +#include "stmmac_pcs.h" #include "dwmac1000.h" static void dwmac1000_core_init(struct mac_device_info *hw, int mtu) { void __iomem *ioaddr = hw->pcsr; u32 value = readl(ioaddr + GMAC_CONTROL); + + /* Configure GMAC core */ value |= GMAC_CORE_INIT; + if (mtu > 1500) value |= GMAC_CONTROL_2K; if (mtu > 2000) value |= GMAC_CONTROL_JE; + if (hw->ps) { + value |= GMAC_CONTROL_TE; + + if (hw->ps == SPEED_1000) { + value &= ~GMAC_CONTROL_PS; + } else { + value |= GMAC_CONTROL_PS; + + if (hw->ps == SPEED_10) + value &= ~GMAC_CONTROL_FES; + else + value |= GMAC_CONTROL_FES; + } + } + writel(value, ioaddr + GMAC_CONTROL); /* Mask GMAC interrupts */ - writel(0x207, ioaddr + GMAC_INT_MASK); + value = GMAC_INT_DEFAULT_MASK; + + if (hw->pmt) + value &= ~GMAC_INT_DISABLE_PMT; + if (hw->pcs) + value &= ~GMAC_INT_DISABLE_PCS; + + writel(value, ioaddr + GMAC_INT_MASK); #ifdef STMMAC_VLAN_TAG_USED /* Tag detection without filtering */ @@ -241,6 +267,39 @@ static void dwmac1000_pmt(struct mac_device_info *hw, unsigned long mode) writel(pmt, ioaddr + GMAC_PMT); } +/* RGMII or SMII interface */ +static void dwmac1000_rgsmii(void __iomem *ioaddr, struct stmmac_extra_stats *x) +{ + u32 status; + + status = readl(ioaddr + GMAC_RGSMIIIS); + x->irq_rgmii_n++; + + /* Check the link status */ + if (status & GMAC_RGSMIIIS_LNKSTS) { + int speed_value; + + x->pcs_link = 1; + + speed_value = ((status & GMAC_RGSMIIIS_SPEED) >> + GMAC_RGSMIIIS_SPEED_SHIFT); + if (speed_value == GMAC_RGSMIIIS_SPEED_125) + x->pcs_speed = SPEED_1000; + else if (speed_value == GMAC_RGSMIIIS_SPEED_25) + x->pcs_speed = SPEED_100; + else + x->pcs_speed = SPEED_10; + + x->pcs_duplex = (status & GMAC_RGSMIIIS_LNKMOD_MASK); + + pr_info("Link is Up - %d/%s\n", (int)x->pcs_speed, + x->pcs_duplex ? "Full" : "Half"); + } else { + x->pcs_link = 0; + pr_info("Link is Down\n"); + } +} + static int dwmac1000_irq_status(struct mac_device_info *hw, struct stmmac_extra_stats *x) { @@ -249,19 +308,20 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, int ret = 0; /* Not used events (e.g. MMC interrupts) are not handled. */ - if ((intr_status & mmc_tx_irq)) + if ((intr_status & GMAC_INT_STATUS_MMCTIS)) x->mmc_tx_irq_n++; - if (unlikely(intr_status & mmc_rx_irq)) + if (unlikely(intr_status & GMAC_INT_STATUS_MMCRIS)) x->mmc_rx_irq_n++; - if (unlikely(intr_status & mmc_rx_csum_offload_irq)) + if (unlikely(intr_status & GMAC_INT_STATUS_MMCCSUM)) x->mmc_rx_csum_offload_irq_n++; - if (unlikely(intr_status & pmt_irq)) { + if (unlikely(intr_status & GMAC_INT_DISABLE_PMT)) { /* clear the PMT bits 5 and 6 by reading the PMT status reg */ readl(ioaddr + GMAC_PMT); x->irq_receive_pmt_irq_n++; } - /* MAC trx/rx EEE LPI entry/exit interrupts */ - if (intr_status & lpiis_irq) { + + /* MAC tx/rx EEE LPI entry/exit interrupts */ + if (intr_status & GMAC_INT_STATUS_LPIIS) { /* Clean LPI interrupt by reading the Reg 12 */ ret = readl(ioaddr + LPI_CTRL_STATUS); @@ -275,36 +335,10 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, x->irq_rx_path_exit_lpi_mode_n++; } - if ((intr_status & pcs_ane_irq) || (intr_status & pcs_link_irq)) { - readl(ioaddr + GMAC_AN_STATUS); - x->irq_pcs_ane_n++; - } - if (intr_status & rgmii_irq) { - u32 status = readl(ioaddr + GMAC_S_R_GMII); - x->irq_rgmii_n++; - - /* Save and dump the link status. */ - if (status & GMAC_S_R_GMII_LINK) { - int speed_value = (status & GMAC_S_R_GMII_SPEED) >> - GMAC_S_R_GMII_SPEED_SHIFT; - x->pcs_duplex = (status & GMAC_S_R_GMII_MODE); - - if (speed_value == GMAC_S_R_GMII_SPEED_125) - x->pcs_speed = SPEED_1000; - else if (speed_value == GMAC_S_R_GMII_SPEED_25) - x->pcs_speed = SPEED_100; - else - x->pcs_speed = SPEED_10; + dwmac_pcs_isr(ioaddr, GMAC_PCS_BASE, intr_status, x); - x->pcs_link = 1; - pr_debug("%s: Link is Up - %d/%s\n", __func__, - (int)x->pcs_speed, - x->pcs_duplex ? "Full" : "Half"); - } else { - x->pcs_link = 0; - pr_debug("%s: Link is Down\n", __func__); - } - } + if (intr_status & PCS_RGSMIIIS_IRQ) + dwmac1000_rgsmii(ioaddr, x); return ret; } @@ -363,38 +397,20 @@ static void dwmac1000_set_eee_timer(struct mac_device_info *hw, int ls, int tw) writel(value, ioaddr + LPI_TIMER_CTRL); } -static void dwmac1000_ctrl_ane(struct mac_device_info *hw, bool restart) +static void dwmac1000_ctrl_ane(void __iomem *ioaddr, bool ane, bool srgmi_ral, + bool loopback) { - void __iomem *ioaddr = hw->pcsr; - /* auto negotiation enable and External Loopback enable */ - u32 value = GMAC_AN_CTRL_ANE | GMAC_AN_CTRL_ELE; - - if (restart) - value |= GMAC_AN_CTRL_RAN; - - writel(value, ioaddr + GMAC_AN_CTRL); + dwmac_ctrl_ane(ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); } -static void dwmac1000_get_adv(struct mac_device_info *hw, struct rgmii_adv *adv) +static void dwmac1000_rane(void __iomem *ioaddr, bool restart) { - void __iomem *ioaddr = hw->pcsr; - u32 value = readl(ioaddr + GMAC_ANE_ADV); - - if (value & GMAC_ANE_FD) - adv->duplex = DUPLEX_FULL; - if (value & GMAC_ANE_HD) - adv->duplex |= DUPLEX_HALF; - - adv->pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; - - value = readl(ioaddr + GMAC_ANE_LPA); - - if (value & GMAC_ANE_FD) - adv->lp_duplex = DUPLEX_FULL; - if (value & GMAC_ANE_HD) - adv->lp_duplex = DUPLEX_HALF; + dwmac_rane(ioaddr, GMAC_PCS_BASE, restart); +} - adv->lp_pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; +static void dwmac1000_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv) +{ + dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv); } static void dwmac1000_debug(void __iomem *ioaddr, struct stmmac_extra_stats *x) @@ -485,9 +501,10 @@ static const struct stmmac_ops dwmac1000_ops = { .reset_eee_mode = dwmac1000_reset_eee_mode, .set_eee_timer = dwmac1000_set_eee_timer, .set_eee_pls = dwmac1000_set_eee_pls, - .ctrl_ane = dwmac1000_ctrl_ane, - .get_adv = dwmac1000_get_adv, .debug = dwmac1000_debug, + .pcs_ctrl_ane = dwmac1000_ctrl_ane, + .pcs_rane = dwmac1000_rane, + .pcs_get_adv_lp = dwmac1000_get_adv_lp, }; struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr, int mcbins, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index bc50952a18e7..6f4f5ce25114 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -24,10 +24,8 @@ #define GMAC_QX_TX_FLOW_CTRL(x) (0x70 + x * 4) #define GMAC_INT_STATUS 0x000000b0 #define GMAC_INT_EN 0x000000b4 -#define GMAC_AN_CTRL 0x000000e0 -#define GMAC_AN_STATUS 0x000000e4 -#define GMAC_AN_ADV 0x000000e8 -#define GMAC_AN_LPA 0x000000ec +#define GMAC_PCS_BASE 0x000000e0 +#define GMAC_PHYIF_CONTROL_STATUS 0x000000f8 #define GMAC_PMT 0x000000c0 #define GMAC_VERSION 0x00000110 #define GMAC_DEBUG 0x00000114 @@ -54,9 +52,18 @@ #define GMAC_TX_FLOW_CTRL_PT_SHIFT 16 /* MAC Interrupt bitmap*/ +#define GMAC_INT_RGSMIIS BIT(0) +#define GMAC_INT_PCS_LINK BIT(1) +#define GMAC_INT_PCS_ANE BIT(2) +#define GMAC_INT_PCS_PHYIS BIT(3) #define GMAC_INT_PMT_EN BIT(4) #define GMAC_INT_LPI_EN BIT(5) +#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_RGSMIIS | GMAC_INT_PCS_LINK | \ + GMAC_INT_PCS_ANE) + +#define GMAC_INT_DEFAULT_MASK GMAC_INT_PMT_EN + enum dwmac4_irq_status { time_stamp_irq = 0x00001000, mmc_rx_csum_offload_irq = 0x00000800, @@ -64,19 +71,8 @@ enum dwmac4_irq_status { mmc_rx_irq = 0x00000200, mmc_irq = 0x00000100, pmt_irq = 0x00000010, - pcs_ane_irq = 0x00000004, - pcs_link_irq = 0x00000002, }; -/* MAC Auto-Neg bitmap*/ -#define GMAC_AN_CTRL_RAN BIT(9) -#define GMAC_AN_CTRL_ANE BIT(12) -#define GMAC_AN_CTRL_ELE BIT(14) -#define GMAC_AN_FD BIT(5) -#define GMAC_AN_HD BIT(6) -#define GMAC_AN_PSE_MASK GENMASK(8, 7) -#define GMAC_AN_PSE_SHIFT 7 - /* MAC PMT bitmap */ enum power_event { pointer_reset = 0x80000000, @@ -250,6 +246,23 @@ enum power_event { #define MTL_DEBUG_RRCSTS_FLUSH 3 #define MTL_DEBUG_RWCSTS BIT(0) +/* SGMII/RGMII status register */ +#define GMAC_PHYIF_CTRLSTATUS_TC BIT(0) +#define GMAC_PHYIF_CTRLSTATUS_LUD BIT(1) +#define GMAC_PHYIF_CTRLSTATUS_SMIDRXS BIT(4) +#define GMAC_PHYIF_CTRLSTATUS_LNKMOD BIT(16) +#define GMAC_PHYIF_CTRLSTATUS_SPEED GENMASK(18, 17) +#define GMAC_PHYIF_CTRLSTATUS_SPEED_SHIFT 17 +#define GMAC_PHYIF_CTRLSTATUS_LNKSTS BIT(19) +#define GMAC_PHYIF_CTRLSTATUS_JABTO BIT(20) +#define GMAC_PHYIF_CTRLSTATUS_FALSECARDET BIT(21) +/* LNKMOD */ +#define GMAC_PHYIF_CTRLSTATUS_LNKMOD_MASK 0x1 +/* LNKSPEED */ +#define GMAC_PHYIF_CTRLSTATUS_SPEED_125 0x2 +#define GMAC_PHYIF_CTRLSTATUS_SPEED_25 0x1 +#define GMAC_PHYIF_CTRLSTATUS_SPEED_2_5 0x0 + extern const struct stmmac_dma_ops dwmac4_dma_ops; extern const struct stmmac_dma_ops dwmac410_dma_ops; #endif /* __DWMAC4_H__ */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 4f7283d05588..df5580dcdfed 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -17,6 +17,7 @@ #include <linux/slab.h> #include <linux/ethtool.h> #include <linux/io.h> +#include "stmmac_pcs.h" #include "dwmac4.h" static void dwmac4_core_init(struct mac_device_info *hw, int mtu) @@ -31,10 +32,31 @@ static void dwmac4_core_init(struct mac_device_info *hw, int mtu) if (mtu > 2000) value |= GMAC_CONFIG_JE; + if (hw->ps) { + value |= GMAC_CONFIG_TE; + + if (hw->ps == SPEED_1000) { + value &= ~GMAC_CONFIG_PS; + } else { + value |= GMAC_CONFIG_PS; + + if (hw->ps == SPEED_10) + value &= ~GMAC_CONFIG_FES; + else + value |= GMAC_CONFIG_FES; + } + } + writel(value, ioaddr + GMAC_CONFIG); /* Mask GMAC interrupts */ - writel(GMAC_INT_PMT_EN, ioaddr + GMAC_INT_EN); + value = GMAC_INT_DEFAULT_MASK; + if (hw->pmt) + value |= GMAC_INT_PMT_EN; + if (hw->pcs) + value |= GMAC_PCS_IRQ_DEFAULT; + + writel(value, ioaddr + GMAC_INT_EN); } static void dwmac4_dump_regs(struct mac_device_info *hw) @@ -156,7 +178,7 @@ static void dwmac4_set_filter(struct mac_device_info *hw, struct netdev_hw_addr *ha; netdev_for_each_uc_addr(ha, dev) { - dwmac4_set_umac_addr(ioaddr, ha->addr, reg); + dwmac4_set_umac_addr(hw, ha->addr, reg); reg++; } } @@ -190,39 +212,53 @@ static void dwmac4_flow_ctrl(struct mac_device_info *hw, unsigned int duplex, } } -static void dwmac4_ctrl_ane(struct mac_device_info *hw, bool restart) +static void dwmac4_ctrl_ane(void __iomem *ioaddr, bool ane, bool srgmi_ral, + bool loopback) { - void __iomem *ioaddr = hw->pcsr; - - /* auto negotiation enable and External Loopback enable */ - u32 value = GMAC_AN_CTRL_ANE | GMAC_AN_CTRL_ELE; + dwmac_ctrl_ane(ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); +} - if (restart) - value |= GMAC_AN_CTRL_RAN; +static void dwmac4_rane(void __iomem *ioaddr, bool restart) +{ + dwmac_rane(ioaddr, GMAC_PCS_BASE, restart); +} - writel(value, ioaddr + GMAC_AN_CTRL); +static void dwmac4_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv) +{ + dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv); } -static void dwmac4_get_adv(struct mac_device_info *hw, struct rgmii_adv *adv) +/* RGMII or SMII interface */ +static void dwmac4_phystatus(void __iomem *ioaddr, struct stmmac_extra_stats *x) { - void __iomem *ioaddr = hw->pcsr; - u32 value = readl(ioaddr + GMAC_AN_ADV); + u32 status; - if (value & GMAC_AN_FD) - adv->duplex = DUPLEX_FULL; - if (value & GMAC_AN_HD) - adv->duplex |= DUPLEX_HALF; + status = readl(ioaddr + GMAC_PHYIF_CONTROL_STATUS); + x->irq_rgmii_n++; - adv->pause = (value & GMAC_AN_PSE_MASK) >> GMAC_AN_PSE_SHIFT; + /* Check the link status */ + if (status & GMAC_PHYIF_CTRLSTATUS_LNKSTS) { + int speed_value; - value = readl(ioaddr + GMAC_AN_LPA); + x->pcs_link = 1; + + speed_value = ((status & GMAC_PHYIF_CTRLSTATUS_SPEED) >> + GMAC_PHYIF_CTRLSTATUS_SPEED_SHIFT); + if (speed_value == GMAC_PHYIF_CTRLSTATUS_SPEED_125) + x->pcs_speed = SPEED_1000; + else if (speed_value == GMAC_PHYIF_CTRLSTATUS_SPEED_25) + x->pcs_speed = SPEED_100; + else + x->pcs_speed = SPEED_10; - if (value & GMAC_AN_FD) - adv->lp_duplex = DUPLEX_FULL; - if (value & GMAC_AN_HD) - adv->lp_duplex = DUPLEX_HALF; + x->pcs_duplex = (status & GMAC_PHYIF_CTRLSTATUS_LNKMOD_MASK); - adv->lp_pause = (value & GMAC_AN_PSE_MASK) >> GMAC_AN_PSE_SHIFT; + pr_info("Link is Up - %d/%s\n", (int)x->pcs_speed, + x->pcs_duplex ? "Full" : "Half"); + } else { + x->pcs_link = 0; + pr_info("Link is Down\n"); + } } static int dwmac4_irq_status(struct mac_device_info *hw, @@ -248,11 +284,6 @@ static int dwmac4_irq_status(struct mac_device_info *hw, x->irq_receive_pmt_irq_n++; } - if ((intr_status & pcs_ane_irq) || (intr_status & pcs_link_irq)) { - readl(ioaddr + GMAC_AN_STATUS); - x->irq_pcs_ane_n++; - } - mtl_int_qx_status = readl(ioaddr + MTL_INT_STATUS); /* Check MTL Interrupt: Currently only one queue is used: Q0. */ if (mtl_int_qx_status & MTL_INT_Q0) { @@ -267,6 +298,10 @@ static int dwmac4_irq_status(struct mac_device_info *hw, } } + dwmac_pcs_isr(ioaddr, GMAC_PCS_BASE, intr_status, x); + if (intr_status & PCS_RGSMIIIS_IRQ) + dwmac4_phystatus(ioaddr, x); + return ret; } @@ -363,8 +398,9 @@ static const struct stmmac_ops dwmac4_ops = { .pmt = dwmac4_pmt, .set_umac_addr = dwmac4_set_umac_addr, .get_umac_addr = dwmac4_get_umac_addr, - .ctrl_ane = dwmac4_ctrl_ane, - .get_adv = dwmac4_get_adv, + .pcs_ctrl_ane = dwmac4_ctrl_ane, + .pcs_rane = dwmac4_rane, + .pcs_get_adv_lp = dwmac4_get_adv_lp, .debug = dwmac4_debug, .set_filter = dwmac4_set_filter, }; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 59ae6088cd22..8dc9056c1001 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -117,7 +117,6 @@ struct stmmac_priv { int eee_enabled; int eee_active; int tx_lpi_timer; - int pcs; unsigned int mode; int extend_desc; struct ptp_clock *ptp_clock; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index e2b98b01647e..1e06173fc9d7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -276,7 +276,8 @@ static int stmmac_ethtool_getsettings(struct net_device *dev, struct phy_device *phy = priv->phydev; int rc; - if ((priv->pcs & STMMAC_PCS_RGMII) || (priv->pcs & STMMAC_PCS_SGMII)) { + if (priv->hw->pcs & STMMAC_PCS_RGMII || + priv->hw->pcs & STMMAC_PCS_SGMII) { struct rgmii_adv adv; if (!priv->xstats.pcs_link) { @@ -289,10 +290,10 @@ static int stmmac_ethtool_getsettings(struct net_device *dev, ethtool_cmd_speed_set(cmd, priv->xstats.pcs_speed); /* Get and convert ADV/LP_ADV from the HW AN registers */ - if (!priv->hw->mac->get_adv) + if (!priv->hw->mac->pcs_get_adv_lp) return -EOPNOTSUPP; /* should never happen indeed */ - priv->hw->mac->get_adv(priv->hw, &adv); + priv->hw->mac->pcs_get_adv_lp(priv->ioaddr, &adv); /* Encoding of PSE bits is defined in 802.3z, 37.2.1.4 */ @@ -361,7 +362,8 @@ static int stmmac_ethtool_setsettings(struct net_device *dev, struct phy_device *phy = priv->phydev; int rc; - if ((priv->pcs & STMMAC_PCS_RGMII) || (priv->pcs & STMMAC_PCS_SGMII)) { + if (priv->hw->pcs & STMMAC_PCS_RGMII || + priv->hw->pcs & STMMAC_PCS_SGMII) { u32 mask = ADVERTISED_Autoneg | ADVERTISED_Pause; /* Only support ANE */ @@ -376,8 +378,11 @@ static int stmmac_ethtool_setsettings(struct net_device *dev, ADVERTISED_10baseT_Full); spin_lock(&priv->lock); - if (priv->hw->mac->ctrl_ane) - priv->hw->mac->ctrl_ane(priv->hw, 1); + + if (priv->hw->mac->pcs_ctrl_ane) + priv->hw->mac->pcs_ctrl_ane(priv->ioaddr, 1, + priv->hw->ps, 0); + spin_unlock(&priv->lock); return 0; @@ -452,11 +457,22 @@ stmmac_get_pauseparam(struct net_device *netdev, { struct stmmac_priv *priv = netdev_priv(netdev); - if (priv->pcs) /* FIXME */ - return; - pause->rx_pause = 0; pause->tx_pause = 0; + + if (priv->hw->pcs && priv->hw->mac->pcs_get_adv_lp) { + struct rgmii_adv adv_lp; + + pause->autoneg = 1; + priv->hw->mac->pcs_get_adv_lp(priv->ioaddr, &adv_lp); + if (!adv_lp.pause) + return; + } else { + if (!(priv->phydev->supported & SUPPORTED_Pause) || + !(priv->phydev->supported & SUPPORTED_Asym_Pause)) + return; + } + pause->autoneg = priv->phydev->autoneg; if (priv->flow_ctrl & FLOW_RX) @@ -473,10 +489,19 @@ stmmac_set_pauseparam(struct net_device *netdev, struct stmmac_priv *priv = netdev_priv(netdev); struct phy_device *phy = priv->phydev; int new_pause = FLOW_OFF; - int ret = 0; - if (priv->pcs) /* FIXME */ - return -EOPNOTSUPP; + if (priv->hw->pcs && priv->hw->mac->pcs_get_adv_lp) { + struct rgmii_adv adv_lp; + + pause->autoneg = 1; + priv->hw->mac->pcs_get_adv_lp(priv->ioaddr, &adv_lp); + if (!adv_lp.pause) + return -EOPNOTSUPP; + } else { + if (!(phy->supported & SUPPORTED_Pause) || + !(phy->supported & SUPPORTED_Asym_Pause)) + return -EOPNOTSUPP; + } if (pause->rx_pause) new_pause |= FLOW_RX; @@ -488,11 +513,12 @@ stmmac_set_pauseparam(struct net_device *netdev, if (phy->autoneg) { if (netif_running(netdev)) - ret = phy_start_aneg(phy); - } else - priv->hw->mac->flow_ctrl(priv->hw, phy->duplex, - priv->flow_ctrl, priv->pause); - return ret; + return phy_start_aneg(phy); + } + + priv->hw->mac->flow_ctrl(priv->hw, phy->duplex, priv->flow_ctrl, + priv->pause); + return 0; } static void stmmac_get_ethtool_stats(struct net_device *dev, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index eac45d0c75e2..aab777c1ba33 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -285,8 +285,9 @@ bool stmmac_eee_init(struct stmmac_priv *priv) /* Using PCS we cannot dial with the phy registers at this stage * so we do not support extra feature like EEE. */ - if ((priv->pcs == STMMAC_PCS_RGMII) || (priv->pcs == STMMAC_PCS_TBI) || - (priv->pcs == STMMAC_PCS_RTBI)) + if ((priv->hw->pcs == STMMAC_PCS_RGMII) || + (priv->hw->pcs == STMMAC_PCS_TBI) || + (priv->hw->pcs == STMMAC_PCS_RTBI)) goto out; /* MAC core supports the EEE feature. */ @@ -799,10 +800,10 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) (interface == PHY_INTERFACE_MODE_RGMII_RXID) || (interface == PHY_INTERFACE_MODE_RGMII_TXID)) { pr_debug("STMMAC: PCS RGMII support enable\n"); - priv->pcs = STMMAC_PCS_RGMII; + priv->hw->pcs = STMMAC_PCS_RGMII; } else if (interface == PHY_INTERFACE_MODE_SGMII) { pr_debug("STMMAC: PCS SGMII support enable\n"); - priv->pcs = STMMAC_PCS_SGMII; + priv->hw->pcs = STMMAC_PCS_SGMII; } } } @@ -1665,6 +1666,19 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) if (priv->plat->bus_setup) priv->plat->bus_setup(priv->ioaddr); + /* PS and related bits will be programmed according to the speed */ + if (priv->hw->pcs) { + int speed = priv->plat->mac_port_sel_speed; + + if ((speed == SPEED_10) || (speed == SPEED_100) || + (speed == SPEED_1000)) { + priv->hw->ps = speed; + } else { + dev_warn(priv->device, "invalid port speed\n"); + priv->hw->ps = 0; + } + } + /* Initialize the MAC Core */ priv->hw->mac->core_init(priv->hw, dev->mtu); @@ -1714,8 +1728,8 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) priv->hw->dma->rx_watchdog(priv->ioaddr, MAX_DMA_RIWT); } - if (priv->pcs && priv->hw->mac->ctrl_ane) - priv->hw->mac->ctrl_ane(priv->hw, 0); + if (priv->hw->pcs && priv->hw->mac->pcs_ctrl_ane) + priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, priv->hw->ps, 0); /* set TX ring length */ if (priv->hw->dma->set_tx_ring_len) @@ -1748,8 +1762,9 @@ static int stmmac_open(struct net_device *dev) stmmac_check_ether_addr(priv); - if (priv->pcs != STMMAC_PCS_RGMII && priv->pcs != STMMAC_PCS_TBI && - priv->pcs != STMMAC_PCS_RTBI) { + if (priv->hw->pcs != STMMAC_PCS_RGMII && + priv->hw->pcs != STMMAC_PCS_TBI && + priv->hw->pcs != STMMAC_PCS_RTBI) { ret = stmmac_init_phy(dev); if (ret) { pr_err("%s: Cannot attach to PHY (error: %d)\n", @@ -2809,6 +2824,14 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id) priv->rx_tail_addr, STMMAC_CHAN0); } + + /* PCS link status */ + if (priv->hw->pcs) { + if (priv->xstats.pcs_link) + netif_carrier_on(dev); + else + netif_carrier_off(dev); + } } /* To handle DMA interrupts */ @@ -3130,6 +3153,7 @@ static int stmmac_hw_init(struct stmmac_priv *priv) */ priv->plat->enh_desc = priv->dma_cap.enh_desc; priv->plat->pmt = priv->dma_cap.pmt_remote_wake_up; + priv->hw->pmt = priv->plat->pmt; /* TXCOE doesn't work in thresh DMA mode */ if (priv->plat->force_thresh_dma_mode) @@ -3325,8 +3349,9 @@ int stmmac_dvr_probe(struct device *device, stmmac_check_pcs_mode(priv); - if (priv->pcs != STMMAC_PCS_RGMII && priv->pcs != STMMAC_PCS_TBI && - priv->pcs != STMMAC_PCS_RTBI) { + if (priv->hw->pcs != STMMAC_PCS_RGMII && + priv->hw->pcs != STMMAC_PCS_TBI && + priv->hw->pcs != STMMAC_PCS_RTBI) { /* MDIO bus Registration */ ret = stmmac_mdio_register(ndev); if (ret < 0) { @@ -3376,8 +3401,9 @@ int stmmac_dvr_remove(struct device *dev) reset_control_assert(priv->stmmac_rst); clk_disable_unprepare(priv->pclk); clk_disable_unprepare(priv->stmmac_clk); - if (priv->pcs != STMMAC_PCS_RGMII && priv->pcs != STMMAC_PCS_TBI && - priv->pcs != STMMAC_PCS_RTBI) + if (priv->hw->pcs != STMMAC_PCS_RGMII && + priv->hw->pcs != STMMAC_PCS_TBI && + priv->hw->pcs != STMMAC_PCS_RTBI) stmmac_mdio_unregister(ndev); free_netdev(ndev); @@ -3450,8 +3476,6 @@ int stmmac_resume(struct device *dev) if (!netif_running(ndev)) return 0; - spin_lock_irqsave(&priv->lock, flags); - /* Power Down bit, into the PM register, is cleared * automatically as soon as a magic packet or a Wake-up frame * is received. Anyway, it's better to manually clear @@ -3459,7 +3483,9 @@ int stmmac_resume(struct device *dev) * from another devices (e.g. serial console). */ if (device_may_wakeup(priv->device)) { + spin_lock_irqsave(&priv->lock, flags); priv->hw->mac->pmt(priv->hw, 0); + spin_unlock_irqrestore(&priv->lock, flags); priv->irq_wake = 0; } else { pinctrl_pm_select_default_state(priv->device); @@ -3473,6 +3499,8 @@ int stmmac_resume(struct device *dev) netif_device_attach(ndev); + spin_lock_irqsave(&priv->lock, flags); + priv->cur_rx = 0; priv->dirty_rx = 0; priv->dirty_tx = 0; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h new file mode 100644 index 000000000000..eba41c24b7a7 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h @@ -0,0 +1,159 @@ +/* + * stmmac_pcs.h: Physical Coding Sublayer Header File + * + * Copyright (C) 2016 STMicroelectronics (R&D) Limited + * Author: Giuseppe Cavallaro <peppe.cavallaro@st.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __STMMAC_PCS_H__ +#define __STMMAC_PCS_H__ + +#include <linux/slab.h> +#include <linux/io.h> +#include "common.h" + +/* PCS registers (AN/TBI/SGMII/RGMII) offsets */ +#define GMAC_AN_CTRL(x) (x) /* AN control */ +#define GMAC_AN_STATUS(x) (x + 0x4) /* AN status */ +#define GMAC_ANE_ADV(x) (x + 0x8) /* ANE Advertisement */ +#define GMAC_ANE_LPA(x) (x + 0xc) /* ANE link partener ability */ +#define GMAC_ANE_EXP(x) (x + 0x10) /* ANE expansion */ +#define GMAC_TBI(x) (x + 0x14) /* TBI extend status */ + +/* AN Configuration defines */ +#define GMAC_AN_CTRL_RAN BIT(9) /* Restart Auto-Negotiation */ +#define GMAC_AN_CTRL_ANE BIT(12) /* Auto-Negotiation Enable */ +#define GMAC_AN_CTRL_ELE BIT(14) /* External Loopback Enable */ +#define GMAC_AN_CTRL_ECD BIT(16) /* Enable Comma Detect */ +#define GMAC_AN_CTRL_LR BIT(17) /* Lock to Reference */ +#define GMAC_AN_CTRL_SGMRAL BIT(18) /* SGMII RAL Control */ + +/* AN Status defines */ +#define GMAC_AN_STATUS_LS BIT(2) /* Link Status 0:down 1:up */ +#define GMAC_AN_STATUS_ANA BIT(3) /* Auto-Negotiation Ability */ +#define GMAC_AN_STATUS_ANC BIT(5) /* Auto-Negotiation Complete */ +#define GMAC_AN_STATUS_ES BIT(8) /* Extended Status */ + +/* ADV and LPA defines */ +#define GMAC_ANE_FD BIT(5) +#define GMAC_ANE_HD BIT(6) +#define GMAC_ANE_PSE GENMASK(8, 7) +#define GMAC_ANE_PSE_SHIFT 7 +#define GMAC_ANE_RFE GENMASK(13, 12) +#define GMAC_ANE_RFE_SHIFT 12 +#define GMAC_ANE_ACK BIT(14) + +/** + * dwmac_pcs_isr - TBI, RTBI, or SGMII PHY ISR + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @intr_status: GMAC core interrupt status + * @x: pointer to log these events as stats + * Description: it is the ISR for PCS events: Auto-Negotiation Completed and + * Link status. + */ +static inline void dwmac_pcs_isr(void __iomem *ioaddr, u32 reg, + unsigned int intr_status, + struct stmmac_extra_stats *x) +{ + u32 val = readl(ioaddr + GMAC_AN_STATUS(reg)); + + if (intr_status & PCS_ANE_IRQ) { + x->irq_pcs_ane_n++; + if (val & GMAC_AN_STATUS_ANC) + pr_info("stmmac_pcs: ANE process completed\n"); + } + + if (intr_status & PCS_LINK_IRQ) { + x->irq_pcs_link_n++; + if (val & GMAC_AN_STATUS_LS) + pr_info("stmmac_pcs: Link Up\n"); + else + pr_info("stmmac_pcs: Link Down\n"); + } +} + +/** + * dwmac_rane - To restart ANE + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @restart: to restart ANE + * Description: this is to just restart the Auto-Negotiation. + */ +static inline void dwmac_rane(void __iomem *ioaddr, u32 reg, bool restart) +{ + u32 value = readl(ioaddr + GMAC_AN_CTRL(reg)); + + if (restart) + value |= GMAC_AN_CTRL_RAN; + + writel(value, ioaddr + GMAC_AN_CTRL(reg)); +} + +/** + * dwmac_ctrl_ane - To program the AN Control Register. + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @ane: to enable the auto-negotiation + * @srgmi_ral: to manage MAC-2-MAC SGMII connections. + * @loopback: to cause the PHY to loopback tx data into rx path. + * Description: this is the main function to configure the AN control register + * and init the ANE, select loopback (usually for debugging purpose) and + * configure SGMII RAL. + */ +static inline void dwmac_ctrl_ane(void __iomem *ioaddr, u32 reg, bool ane, + bool srgmi_ral, bool loopback) +{ + u32 value = readl(ioaddr + GMAC_AN_CTRL(reg)); + + /* Enable and restart the Auto-Negotiation */ + if (ane) + value |= GMAC_AN_CTRL_ANE | GMAC_AN_CTRL_RAN; + + /* In case of MAC-2-MAC connection, block is configured to operate + * according to MAC conf register. + */ + if (srgmi_ral) + value |= GMAC_AN_CTRL_SGMRAL; + + if (loopback) + value |= GMAC_AN_CTRL_ELE; + + writel(value, ioaddr + GMAC_AN_CTRL(reg)); +} + +/** + * dwmac_get_adv_lp - Get ADV and LP cap + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @adv_lp: structure to store the adv,lp status + * Description: this is to expose the ANE advertisement and Link partner ability + * status to ethtool support. + */ +static inline void dwmac_get_adv_lp(void __iomem *ioaddr, u32 reg, + struct rgmii_adv *adv_lp) +{ + u32 value = readl(ioaddr + GMAC_ANE_ADV(reg)); + + if (value & GMAC_ANE_FD) + adv_lp->duplex = DUPLEX_FULL; + if (value & GMAC_ANE_HD) + adv_lp->duplex |= DUPLEX_HALF; + + adv_lp->pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; + + value = readl(ioaddr + GMAC_ANE_LPA(reg)); + + if (value & GMAC_ANE_FD) + adv_lp->lp_duplex = DUPLEX_FULL; + if (value & GMAC_ANE_HD) + adv_lp->lp_duplex = DUPLEX_HALF; + + adv_lp->lp_pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; +} +#endif /* __STMMAC_PCS_H__ */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 409db913b117..f7dfc0ae8e9c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -319,6 +319,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac) pr_warn("force_sf_dma_mode is ignored if force_thresh_dma_mode is set."); } + of_property_read_u32(np, "snps,ps-speed", &plat->mac_port_sel_speed); + plat->axi = stmmac_axi_setup(pdev); return plat; @@ -411,7 +413,9 @@ static int stmmac_pltfr_suspend(struct device *dev) struct platform_device *pdev = to_platform_device(dev); ret = stmmac_suspend(dev); - if (priv->plat->exit) + if (priv->plat->suspend) + priv->plat->suspend(pdev, priv->plat->bsp_priv); + else if (priv->plat->exit) priv->plat->exit(pdev, priv->plat->bsp_priv); return ret; @@ -430,7 +434,9 @@ static int stmmac_pltfr_resume(struct device *dev) struct stmmac_priv *priv = netdev_priv(ndev); struct platform_device *pdev = to_platform_device(dev); - if (priv->plat->init) + if (priv->plat->resume) + priv->plat->resume(pdev, priv->plat->bsp_priv); + else if (priv->plat->init) priv->plat->init(pdev, priv->plat->bsp_priv); return stmmac_resume(dev); diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index 158213cd6cdd..c34111b390c7 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -46,7 +46,6 @@ #include <linux/delay.h> #include <linux/dma-mapping.h> #include <linux/vmalloc.h> -#include <linux/version.h> #include <linux/device.h> #include <linux/bitrev.h> @@ -598,7 +597,6 @@ struct net_local { struct work_struct txtimeout_reinit; phy_interface_t phy_interface; - struct phy_device *phy_dev; struct mii_bus *mii_bus; unsigned int link; @@ -816,7 +814,7 @@ static int dwceqos_mdio_write(struct mii_bus *bus, int mii_id, int phyreg, static int dwceqos_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) { struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!netif_running(ndev)) return -EINVAL; @@ -850,6 +848,7 @@ static void dwceqos_link_down(struct net_local *lp) static void dwceqos_link_up(struct net_local *lp) { + struct net_device *ndev = lp->ndev; u32 regval; unsigned long flags; @@ -860,7 +859,7 @@ static void dwceqos_link_up(struct net_local *lp) dwceqos_write(lp, REG_DWCEQOS_MAC_LPI_CTRL_STATUS, regval); spin_unlock_irqrestore(&lp->hw_lock, flags); - lp->eee_active = !phy_init_eee(lp->phy_dev, 0); + lp->eee_active = !phy_init_eee(ndev->phydev, 0); /* Check for changed EEE capability */ if (!lp->eee_active && lp->eee_enabled) { @@ -876,7 +875,8 @@ static void dwceqos_link_up(struct net_local *lp) static void dwceqos_set_speed(struct net_local *lp) { - struct phy_device *phydev = lp->phy_dev; + struct net_device *ndev = lp->ndev; + struct phy_device *phydev = ndev->phydev; u32 regval; regval = dwceqos_read(lp, REG_DWCEQOS_MAC_CFG); @@ -903,7 +903,7 @@ static void dwceqos_set_speed(struct net_local *lp) static void dwceqos_adjust_link(struct net_device *ndev) { struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; int status_change = 0; if (lp->phy_defer) @@ -987,7 +987,6 @@ static int dwceqos_mii_probe(struct net_device *ndev) lp->link = 0; lp->speed = 0; lp->duplex = DUPLEX_UNKNOWN; - lp->phy_dev = phydev; return 0; } @@ -1531,6 +1530,7 @@ static void dwceqos_configure_bus(struct net_local *lp) static void dwceqos_init_hw(struct net_local *lp) { + struct net_device *ndev = lp->ndev; u32 regval; u32 buswidth; u32 dma_skip; @@ -1645,10 +1645,10 @@ static void dwceqos_init_hw(struct net_local *lp) regval | DWCEQOS_MAC_CFG_TE | DWCEQOS_MAC_CFG_RE); lp->phy_defer = false; - mutex_lock(&lp->phy_dev->lock); - phy_read_status(lp->phy_dev); + mutex_lock(&ndev->phydev->lock); + phy_read_status(ndev->phydev); dwceqos_adjust_link(lp->ndev); - mutex_unlock(&lp->phy_dev->lock); + mutex_unlock(&ndev->phydev->lock); } static void dwceqos_tx_reclaim(unsigned long data) @@ -1898,7 +1898,7 @@ static int dwceqos_open(struct net_device *ndev) * hence the unusual init order with phy_start first. */ lp->phy_defer = true; - phy_start(lp->phy_dev); + phy_start(ndev->phydev); dwceqos_init_hw(lp); napi_enable(&lp->napi); @@ -1943,7 +1943,7 @@ static int dwceqos_stop(struct net_device *ndev) dwceqos_drain_dma(lp); dwceqos_reset_hw(lp); - phy_stop(lp->phy_dev); + phy_stop(ndev->phydev); dwceqos_descriptor_free(lp); @@ -2523,30 +2523,6 @@ dwceqos_get_stats64(struct net_device *ndev, struct rtnl_link_stats64 *s) return s; } -static int -dwceqos_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) -{ - struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, ecmd); -} - -static int -dwceqos_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) -{ - struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, ecmd); -} - static void dwceqos_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *ed) { @@ -2574,17 +2550,17 @@ static int dwceqos_set_pauseparam(struct net_device *ndev, lp->flowcontrol.autoneg = pp->autoneg; if (pp->autoneg) { - lp->phy_dev->advertising |= ADVERTISED_Pause; - lp->phy_dev->advertising |= ADVERTISED_Asym_Pause; + ndev->phydev->advertising |= ADVERTISED_Pause; + ndev->phydev->advertising |= ADVERTISED_Asym_Pause; } else { - lp->phy_dev->advertising &= ~ADVERTISED_Pause; - lp->phy_dev->advertising &= ~ADVERTISED_Asym_Pause; + ndev->phydev->advertising &= ~ADVERTISED_Pause; + ndev->phydev->advertising &= ~ADVERTISED_Asym_Pause; lp->flowcontrol.rx = pp->rx_pause; lp->flowcontrol.tx = pp->tx_pause; } if (netif_running(ndev)) - ret = phy_start_aneg(lp->phy_dev); + ret = phy_start_aneg(ndev->phydev); return ret; } @@ -2705,7 +2681,7 @@ static int dwceqos_get_eee(struct net_device *ndev, struct ethtool_eee *edata) dwceqos_get_tx_lpi_state(regval)); } - return phy_ethtool_get_eee(lp->phy_dev, edata); + return phy_ethtool_get_eee(ndev->phydev, edata); } static int dwceqos_set_eee(struct net_device *ndev, struct ethtool_eee *edata) @@ -2747,7 +2723,7 @@ static int dwceqos_set_eee(struct net_device *ndev, struct ethtool_eee *edata) spin_unlock_irqrestore(&lp->hw_lock, flags); } - return phy_ethtool_set_eee(lp->phy_dev, edata); + return phy_ethtool_set_eee(ndev->phydev, edata); } static u32 dwceqos_get_msglevel(struct net_device *ndev) @@ -2765,8 +2741,6 @@ static void dwceqos_set_msglevel(struct net_device *ndev, u32 msglevel) } static struct ethtool_ops dwceqos_ethtool_ops = { - .get_settings = dwceqos_get_settings, - .set_settings = dwceqos_set_settings, .get_drvinfo = dwceqos_get_drvinfo, .get_link = ethtool_op_get_link, .get_pauseparam = dwceqos_get_pauseparam, @@ -2780,6 +2754,8 @@ static struct ethtool_ops dwceqos_ethtool_ops = { .set_eee = dwceqos_set_eee, .get_msglevel = dwceqos_get_msglevel, .set_msglevel = dwceqos_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static struct net_device_ops netdev_ops = { @@ -2981,8 +2957,8 @@ static int dwceqos_remove(struct platform_device *pdev) if (ndev) { lp = netdev_priv(ndev); - if (lp->phy_dev) - phy_disconnect(lp->phy_dev); + if (ndev->phydev) + phy_disconnect(ndev->phydev); mdiobus_unregister(lp->mii_bus); mdiobus_free(lp->mii_bus); diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 635979bddd34..1a93a1f28433 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1243,6 +1243,7 @@ static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_priv *priv) slave->phy = NULL; cpsw_ale_control_set(priv->ale, slave_port, ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); + soft_reset_slave(slave); } static int cpsw_ndo_open(struct net_device *ndev) @@ -1251,7 +1252,11 @@ static int cpsw_ndo_open(struct net_device *ndev) int i, ret; u32 reg; - pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } if (!cpsw_common_res_usage_state(priv)) cpsw_intr_disable(priv); @@ -1277,6 +1282,7 @@ static int cpsw_ndo_open(struct net_device *ndev) if (!cpsw_common_res_usage_state(priv)) { struct cpsw_priv *priv_sl0 = cpsw_get_slave_priv(priv, 0); + int buf_num; /* setup tx dma to fixed prio and zero offset */ cpdma_control_set(priv->dma, CPDMA_TX_PRIO_FIXED, 1); @@ -1304,10 +1310,8 @@ static int cpsw_ndo_open(struct net_device *ndev) enable_irq(priv->irqs_table[0]); } - if (WARN_ON(!priv->data.rx_descs)) - priv->data.rx_descs = 128; - - for (i = 0; i < priv->data.rx_descs; i++) { + buf_num = cpdma_chan_get_rx_buf_num(priv->dma); + for (i = 0; i < buf_num; i++) { struct sk_buff *skb; ret = -ENOMEM; @@ -1338,7 +1342,7 @@ static int cpsw_ndo_open(struct net_device *ndev) if (priv->coal_intvl != 0) { struct ethtool_coalesce coal; - coal.rx_coalesce_usecs = (priv->coal_intvl << 4); + coal.rx_coalesce_usecs = priv->coal_intvl; cpsw_set_coalesce(ndev, &coal); } @@ -1610,10 +1614,17 @@ static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) struct sockaddr *addr = (struct sockaddr *)p; int flags = 0; u16 vid = 0; + int ret; if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } + if (priv->data.dual_emac) { vid = priv->slaves[priv->emac_port].port_vlan; flags = ALE_VLAN; @@ -1628,6 +1639,8 @@ static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) memcpy(ndev->dev_addr, priv->mac_addr, ETH_ALEN); for_each_slave(priv, cpsw_set_slave_mac, priv); + pm_runtime_put(&priv->pdev->dev); + return 0; } @@ -1692,10 +1705,17 @@ static int cpsw_ndo_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) { struct cpsw_priv *priv = netdev_priv(ndev); + int ret; if (vid == priv->data.default_vlan) return 0; + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } + if (priv->data.dual_emac) { /* In dual EMAC, reserved VLAN id should not be used for * creating VLAN interfaces as this can break the dual @@ -1710,7 +1730,10 @@ static int cpsw_ndo_vlan_rx_add_vid(struct net_device *ndev, } dev_info(priv->dev, "Adding vlanid %d to vlan filter\n", vid); - return cpsw_add_vlan_ale_entry(priv, vid); + ret = cpsw_add_vlan_ale_entry(priv, vid); + + pm_runtime_put(&priv->pdev->dev); + return ret; } static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, @@ -1722,6 +1745,12 @@ static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, if (vid == priv->data.default_vlan) return 0; + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } + if (priv->data.dual_emac) { int i; @@ -1741,8 +1770,10 @@ static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, if (ret != 0) return ret; - return cpsw_ale_del_mcast(priv->ale, priv->ndev->broadcast, - 0, ALE_VLAN, vid); + ret = cpsw_ale_del_mcast(priv->ale, priv->ndev->broadcast, + 0, ALE_VLAN, vid); + pm_runtime_put(&priv->pdev->dev); + return ret; } static const struct net_device_ops cpsw_netdev_ops = { @@ -1901,10 +1932,33 @@ static int cpsw_set_pauseparam(struct net_device *ndev, priv->tx_pause = pause->tx_pause ? true : false; for_each_slave(priv, _cpsw_adjust_link, priv, &link); - return 0; } +static int cpsw_ethtool_op_begin(struct net_device *ndev) +{ + struct cpsw_priv *priv = netdev_priv(ndev); + int ret; + + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + cpsw_err(priv, drv, "ethtool begin failed %d\n", ret); + pm_runtime_put_noidle(&priv->pdev->dev); + } + + return ret; +} + +static void cpsw_ethtool_op_complete(struct net_device *ndev) +{ + struct cpsw_priv *priv = netdev_priv(ndev); + int ret; + + ret = pm_runtime_put(&priv->pdev->dev); + if (ret < 0) + cpsw_err(priv, drv, "ethtool complete failed %d\n", ret); +} + static const struct ethtool_ops cpsw_ethtool_ops = { .get_drvinfo = cpsw_get_drvinfo, .get_msglevel = cpsw_get_msglevel, @@ -1924,6 +1978,8 @@ static const struct ethtool_ops cpsw_ethtool_ops = { .set_wol = cpsw_set_wol, .get_regs_len = cpsw_get_regs_len, .get_regs = cpsw_get_regs, + .begin = cpsw_ethtool_op_begin, + .complete = cpsw_ethtool_op_complete, }; static void cpsw_slave_init(struct cpsw_slave *slave, struct cpsw_priv *priv, @@ -1998,12 +2054,6 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, } data->bd_ram_size = prop; - if (of_property_read_u32(node, "rx_descs", &prop)) { - dev_err(&pdev->dev, "Missing rx_descs property in the DT.\n"); - return -EINVAL; - } - data->rx_descs = prop; - if (of_property_read_u32(node, "mac_control", &prop)) { dev_err(&pdev->dev, "Missing mac_control property in the DT.\n"); return -EINVAL; @@ -2021,7 +2071,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, if (ret) dev_warn(&pdev->dev, "Doesn't have any child node\n"); - for_each_child_of_node(node, slave_node) { + for_each_available_child_of_node(node, slave_node) { struct cpsw_slave_data *slave_data = data->slave_data + i; const void *mac_addr = NULL; int lenp; @@ -2318,7 +2368,11 @@ static int cpsw_probe(struct platform_device *pdev) /* Need to enable clocks with runtime PM api to access module * registers */ - pm_runtime_get_sync(&pdev->dev); + ret = pm_runtime_get_sync(&pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&pdev->dev); + goto clean_runtime_disable_ret; + } priv->version = readl(&priv->regs->id_ver); pm_runtime_put_sync(&pdev->dev); @@ -2502,8 +2556,6 @@ static int cpsw_probe(struct platform_device *pdev) clean_ale_ret: cpsw_ale_destroy(priv->ale); clean_dma_ret: - cpdma_chan_destroy(priv->txch); - cpdma_chan_destroy(priv->rxch); cpdma_ctlr_destroy(priv->dma); clean_runtime_disable_ret: pm_runtime_disable(&pdev->dev); @@ -2531,8 +2583,6 @@ static int cpsw_remove(struct platform_device *pdev) unregister_netdev(ndev); cpsw_ale_destroy(priv->ale); - cpdma_chan_destroy(priv->txch); - cpdma_chan_destroy(priv->rxch); cpdma_ctlr_destroy(priv->dma); pm_runtime_disable(&pdev->dev); device_for_each_child(&pdev->dev, NULL, cpsw_remove_child_device); @@ -2555,16 +2605,12 @@ static int cpsw_suspend(struct device *dev) for (i = 0; i < priv->data.slaves; i++) { if (netif_running(priv->slaves[i].ndev)) cpsw_ndo_stop(priv->slaves[i].ndev); - soft_reset_slave(priv->slaves + i); } } else { if (netif_running(ndev)) cpsw_ndo_stop(ndev); - for_each_slave(priv, soft_reset_slave); } - pm_runtime_put_sync(&pdev->dev); - /* Select sleep pin state */ pinctrl_pm_select_sleep_state(&pdev->dev); @@ -2577,8 +2623,6 @@ static int cpsw_resume(struct device *dev) struct net_device *ndev = platform_get_drvdata(pdev); struct cpsw_priv *priv = netdev_priv(ndev); - pm_runtime_get_sync(&pdev->dev); - /* Select default pin state */ pinctrl_pm_select_default_state(&pdev->dev); diff --git a/drivers/net/ethernet/ti/cpsw.h b/drivers/net/ethernet/ti/cpsw.h index e50afd1b2eda..16b54c6f32c2 100644 --- a/drivers/net/ethernet/ti/cpsw.h +++ b/drivers/net/ethernet/ti/cpsw.h @@ -35,7 +35,6 @@ struct cpsw_platform_data { u32 cpts_clock_shift; /* convert input clock ticks to nanoseconds */ u32 ale_entries; /* ale table size */ u32 bd_ram_size; /*buffer descriptor ram size */ - u32 rx_descs; /* Number of Rx Descriptios */ u32 mac_control; /* Mac control register */ u16 default_vlan; /* Def VLAN for ALE lookup in VLAN aware mode*/ bool dual_emac; /* Enable Dual EMAC mode */ diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index 18bf3a8fdc50..1c653ca7c316 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -21,7 +21,7 @@ #include <linux/dma-mapping.h> #include <linux/io.h> #include <linux/delay.h> - +#include <linux/genalloc.h> #include "davinci_cpdma.h" /* DMA Registers */ @@ -87,9 +87,8 @@ struct cpdma_desc_pool { void *cpumap; /* dma_alloc map */ int desc_size, mem_size; int num_desc, used_desc; - unsigned long *bitmap; struct device *dev; - spinlock_t lock; + struct gen_pool *gen_pool; }; enum cpdma_state { @@ -117,6 +116,7 @@ struct cpdma_chan { int chan_num; spinlock_t lock; int count; + u32 desc_num; u32 mask; cpdma_handler_fn handler; enum dma_data_direction dir; @@ -145,6 +145,19 @@ struct cpdma_chan { (directed << CPDMA_TO_PORT_SHIFT)); \ } while (0) +static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool) +{ + if (!pool) + return; + + WARN_ON(pool->used_desc); + if (pool->cpumap) + dma_free_coherent(pool->dev, pool->mem_size, pool->cpumap, + pool->phys); + else + iounmap(pool->iomap); +} + /* * Utility constructs for a cpdma descriptor pool. Some devices (e.g. davinci * emac) have dedicated on-chip memory for these descriptors. Some other @@ -155,24 +168,25 @@ static struct cpdma_desc_pool * cpdma_desc_pool_create(struct device *dev, u32 phys, dma_addr_t hw_addr, int size, int align) { - int bitmap_size; struct cpdma_desc_pool *pool; + int ret; pool = devm_kzalloc(dev, sizeof(*pool), GFP_KERNEL); if (!pool) - goto fail; - - spin_lock_init(&pool->lock); + goto gen_pool_create_fail; pool->dev = dev; pool->mem_size = size; pool->desc_size = ALIGN(sizeof(struct cpdma_desc), align); pool->num_desc = size / pool->desc_size; - bitmap_size = (pool->num_desc / BITS_PER_LONG) * sizeof(long); - pool->bitmap = devm_kzalloc(dev, bitmap_size, GFP_KERNEL); - if (!pool->bitmap) - goto fail; + pool->gen_pool = devm_gen_pool_create(dev, ilog2(pool->desc_size), -1, + "cpdma"); + if (IS_ERR(pool->gen_pool)) { + dev_err(dev, "pool create failed %ld\n", + PTR_ERR(pool->gen_pool)); + goto gen_pool_create_fail; + } if (phys) { pool->phys = phys; @@ -185,24 +199,22 @@ cpdma_desc_pool_create(struct device *dev, u32 phys, dma_addr_t hw_addr, pool->phys = pool->hw_addr; /* assumes no IOMMU, don't use this value */ } - if (pool->iomap) - return pool; -fail: - return NULL; -} - -static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool) -{ - if (!pool) - return; + if (!pool->iomap) + goto gen_pool_create_fail; - WARN_ON(pool->used_desc); - if (pool->cpumap) { - dma_free_coherent(pool->dev, pool->mem_size, pool->cpumap, - pool->phys); - } else { - iounmap(pool->iomap); + ret = gen_pool_add_virt(pool->gen_pool, (unsigned long)pool->iomap, + pool->phys, pool->mem_size, -1); + if (ret < 0) { + dev_err(dev, "pool add failed %d\n", ret); + goto gen_pool_add_virt_fail; } + + return pool; + +gen_pool_add_virt_fail: + cpdma_desc_pool_destroy(pool); +gen_pool_create_fail: + return NULL; } static inline dma_addr_t desc_phys(struct cpdma_desc_pool *pool, @@ -220,47 +232,23 @@ desc_from_phys(struct cpdma_desc_pool *pool, dma_addr_t dma) } static struct cpdma_desc __iomem * -cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc, bool is_rx) +cpdma_desc_alloc(struct cpdma_desc_pool *pool) { - unsigned long flags; - int index; - int desc_start; - int desc_end; struct cpdma_desc __iomem *desc = NULL; - spin_lock_irqsave(&pool->lock, flags); - - if (is_rx) { - desc_start = 0; - desc_end = pool->num_desc/2; - } else { - desc_start = pool->num_desc/2; - desc_end = pool->num_desc; - } - - index = bitmap_find_next_zero_area(pool->bitmap, - desc_end, desc_start, num_desc, 0); - if (index < desc_end) { - bitmap_set(pool->bitmap, index, num_desc); - desc = pool->iomap + pool->desc_size * index; + desc = (struct cpdma_desc __iomem *)gen_pool_alloc(pool->gen_pool, + pool->desc_size); + if (desc) pool->used_desc++; - } - spin_unlock_irqrestore(&pool->lock, flags); return desc; } static void cpdma_desc_free(struct cpdma_desc_pool *pool, struct cpdma_desc __iomem *desc, int num_desc) { - unsigned long flags, index; - - index = ((unsigned long)desc - (unsigned long)pool->iomap) / - pool->desc_size; - spin_lock_irqsave(&pool->lock, flags); - bitmap_clear(pool->bitmap, index, num_desc); + gen_pool_free(pool->gen_pool, (unsigned long)desc, pool->desc_size); pool->used_desc--; - spin_unlock_irqrestore(&pool->lock, flags); } struct cpdma_ctlr *cpdma_ctlr_create(struct cpdma_params *params) @@ -516,6 +504,7 @@ struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, chan->state = CPDMA_STATE_IDLE; chan->chan_num = chan_num; chan->handler = handler; + chan->desc_num = ctlr->pool->num_desc / 2; if (is_rx_chan(chan)) { chan->hdp = ctlr->params.rxhdp + offset; @@ -543,6 +532,12 @@ struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, } EXPORT_SYMBOL_GPL(cpdma_chan_create); +int cpdma_chan_get_rx_buf_num(struct cpdma_ctlr *ctlr) +{ + return ctlr->pool->num_desc / 2; +} +EXPORT_SYMBOL_GPL(cpdma_chan_get_rx_buf_num); + int cpdma_chan_destroy(struct cpdma_chan *chan) { struct cpdma_ctlr *ctlr; @@ -675,7 +670,13 @@ int cpdma_chan_submit(struct cpdma_chan *chan, void *token, void *data, goto unlock_ret; } - desc = cpdma_desc_alloc(ctlr->pool, 1, is_rx_chan(chan)); + if (chan->count >= chan->desc_num) { + chan->stats.desc_alloc_fail++; + ret = -ENOMEM; + goto unlock_ret; + } + + desc = cpdma_desc_alloc(ctlr->pool); if (!desc) { chan->stats.desc_alloc_fail++; ret = -ENOMEM; @@ -721,24 +722,16 @@ EXPORT_SYMBOL_GPL(cpdma_chan_submit); bool cpdma_check_free_tx_desc(struct cpdma_chan *chan) { - unsigned long flags; - int index; - bool ret; struct cpdma_ctlr *ctlr = chan->ctlr; struct cpdma_desc_pool *pool = ctlr->pool; + bool free_tx_desc; + unsigned long flags; - spin_lock_irqsave(&pool->lock, flags); - - index = bitmap_find_next_zero_area(pool->bitmap, - pool->num_desc, pool->num_desc/2, 1, 0); - - if (index < pool->num_desc) - ret = true; - else - ret = false; - - spin_unlock_irqrestore(&pool->lock, flags); - return ret; + spin_lock_irqsave(&chan->lock, flags); + free_tx_desc = (chan->count < chan->desc_num) && + gen_pool_avail(pool->gen_pool); + spin_unlock_irqrestore(&chan->lock, flags); + return free_tx_desc; } EXPORT_SYMBOL_GPL(cpdma_check_free_tx_desc); diff --git a/drivers/net/ethernet/ti/davinci_cpdma.h b/drivers/net/ethernet/ti/davinci_cpdma.h index 86dee487f2f0..80c015cbbce5 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.h +++ b/drivers/net/ethernet/ti/davinci_cpdma.h @@ -81,6 +81,7 @@ int cpdma_ctlr_dump(struct cpdma_ctlr *ctlr); struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, cpdma_handler_fn handler); +int cpdma_chan_get_rx_buf_num(struct cpdma_ctlr *ctlr); int cpdma_chan_destroy(struct cpdma_chan *chan); int cpdma_chan_start(struct cpdma_chan *chan); int cpdma_chan_stop(struct cpdma_chan *chan); diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index f56d66e6ec15..c6c54659f8d4 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -348,7 +348,6 @@ struct emac_priv { u32 rx_addr_type; const char *phy_id; struct device_node *phy_node; - struct phy_device *phydev; spinlock_t lock; /*platform specific members*/ void (*int_enable) (void); @@ -486,43 +485,6 @@ static void emac_get_drvinfo(struct net_device *ndev, } /** - * emac_get_settings - Get EMAC settings - * @ndev: The DaVinci EMAC network adapter - * @ecmd: ethtool command - * - * Executes ethool get command - * - */ -static int emac_get_settings(struct net_device *ndev, - struct ethtool_cmd *ecmd) -{ - struct emac_priv *priv = netdev_priv(ndev); - if (priv->phydev) - return phy_ethtool_gset(priv->phydev, ecmd); - else - return -EOPNOTSUPP; - -} - -/** - * emac_set_settings - Set EMAC settings - * @ndev: The DaVinci EMAC network adapter - * @ecmd: ethtool command - * - * Executes ethool set command - * - */ -static int emac_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) -{ - struct emac_priv *priv = netdev_priv(ndev); - if (priv->phydev) - return phy_ethtool_sset(priv->phydev, ecmd); - else - return -EOPNOTSUPP; - -} - -/** * emac_get_coalesce - Get interrupt coalesce settings for this device * @ndev : The DaVinci EMAC network adapter * @coal : ethtool coalesce settings structure @@ -625,12 +587,12 @@ static int emac_set_coalesce(struct net_device *ndev, */ static const struct ethtool_ops ethtool_ops = { .get_drvinfo = emac_get_drvinfo, - .get_settings = emac_get_settings, - .set_settings = emac_set_settings, .get_link = ethtool_op_get_link, .get_coalesce = emac_get_coalesce, .set_coalesce = emac_set_coalesce, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /** @@ -651,8 +613,8 @@ static void emac_update_phystatus(struct emac_priv *priv) mac_control = emac_read(EMAC_MACCONTROL); cur_duplex = (mac_control & EMAC_MACCONTROL_FULLDUPLEXEN) ? DUPLEX_FULL : DUPLEX_HALF; - if (priv->phydev) - new_duplex = priv->phydev->duplex; + if (ndev->phydev) + new_duplex = ndev->phydev->duplex; else new_duplex = DUPLEX_FULL; @@ -1454,7 +1416,7 @@ static void emac_poll_controller(struct net_device *ndev) static void emac_adjust_link(struct net_device *ndev) { struct emac_priv *priv = netdev_priv(ndev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = ndev->phydev; unsigned long flags; int new_state = 0; @@ -1483,7 +1445,7 @@ static void emac_adjust_link(struct net_device *ndev) } if (new_state) { emac_update_phystatus(priv); - phy_print_status(priv->phydev); + phy_print_status(ndev->phydev); } spin_unlock_irqrestore(&priv->lock, flags); @@ -1505,15 +1467,13 @@ static void emac_adjust_link(struct net_device *ndev) */ static int emac_devioctl(struct net_device *ndev, struct ifreq *ifrq, int cmd) { - struct emac_priv *priv = netdev_priv(ndev); - if (!(netif_running(ndev))) return -EINVAL; /* TODO: Add phy read and write and private statistics get feature */ - if (priv->phydev) - return phy_mii_ioctl(priv->phydev, ifrq, cmd); + if (ndev->phydev) + return phy_mii_ioctl(ndev->phydev, ifrq, cmd); else return -EOPNOTSUPP; } @@ -1542,6 +1502,7 @@ static int emac_dev_open(struct net_device *ndev) int res_num = 0, irq_num = 0; int i = 0; struct emac_priv *priv = netdev_priv(ndev); + struct phy_device *phydev = NULL; ret = pm_runtime_get_sync(&priv->pdev->dev); if (ret < 0) { @@ -1607,12 +1568,10 @@ static int emac_dev_open(struct net_device *ndev) cpdma_ctlr_start(priv->dma); - priv->phydev = NULL; - if (priv->phy_node) { - priv->phydev = of_phy_connect(ndev, priv->phy_node, - &emac_adjust_link, 0, 0); - if (!priv->phydev) { + phydev = of_phy_connect(ndev, priv->phy_node, + &emac_adjust_link, 0, 0); + if (!phydev) { dev_err(emac_dev, "could not connect to phy %s\n", priv->phy_node->full_name); ret = -ENODEV; @@ -1621,7 +1580,7 @@ static int emac_dev_open(struct net_device *ndev) } /* use the first phy on the bus if pdata did not give us a phy id */ - if (!priv->phydev && !priv->phy_id) { + if (!phydev && !priv->phy_id) { struct device *phy; phy = bus_find_device(&mdio_bus_type, NULL, NULL, @@ -1630,16 +1589,15 @@ static int emac_dev_open(struct net_device *ndev) priv->phy_id = dev_name(phy); } - if (!priv->phydev && priv->phy_id && *priv->phy_id) { - priv->phydev = phy_connect(ndev, priv->phy_id, - &emac_adjust_link, - PHY_INTERFACE_MODE_MII); + if (!phydev && priv->phy_id && *priv->phy_id) { + phydev = phy_connect(ndev, priv->phy_id, + &emac_adjust_link, + PHY_INTERFACE_MODE_MII); - if (IS_ERR(priv->phydev)) { + if (IS_ERR(phydev)) { dev_err(emac_dev, "could not connect to phy %s\n", priv->phy_id); - ret = PTR_ERR(priv->phydev); - priv->phydev = NULL; + ret = PTR_ERR(phydev); goto err; } @@ -1647,10 +1605,10 @@ static int emac_dev_open(struct net_device *ndev) priv->speed = 0; priv->duplex = ~0; - phy_attached_info(priv->phydev); + phy_attached_info(phydev); } - if (!priv->phydev) { + if (!phydev) { /* No PHY , fix the link, speed and duplex settings */ dev_notice(emac_dev, "no phy, defaulting to 100/full\n"); priv->link = 1; @@ -1665,8 +1623,8 @@ static int emac_dev_open(struct net_device *ndev) if (netif_msg_drv(priv)) dev_notice(emac_dev, "DaVinci EMAC: Opened %s\n", ndev->name); - if (priv->phydev) - phy_start(priv->phydev); + if (phydev) + phy_start(phydev); return 0; @@ -1717,8 +1675,8 @@ static int emac_dev_stop(struct net_device *ndev) cpdma_ctlr_stop(priv->dma); emac_write(EMAC_SOFTRESET, 1); - if (priv->phydev) - phy_disconnect(priv->phydev); + if (ndev->phydev) + phy_disconnect(ndev->phydev); /* Free IRQ */ while ((res = platform_get_resource(priv->pdev, IORESOURCE_IRQ, i))) { diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 4e7c9b9b042a..33df340db1f1 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -53,6 +53,10 @@ #define DEF_OUT_FREQ 2200000 /* 2.2 MHz */ +struct davinci_mdio_of_param { + int autosuspend_delay_ms; +}; + struct davinci_mdio_regs { u32 version; u32 control; @@ -90,19 +94,19 @@ static const struct mdio_platform_data default_pdata = { struct davinci_mdio_data { struct mdio_platform_data pdata; struct davinci_mdio_regs __iomem *regs; - spinlock_t lock; struct clk *clk; struct device *dev; struct mii_bus *bus; - bool suspended; + bool active_in_suspend; unsigned long access_time; /* jiffies */ /* Indicates that driver shouldn't modify phy_mask in case * if MDIO bus is registered from DT. */ bool skip_scan; + u32 clk_div; }; -static void __davinci_mdio_reset(struct davinci_mdio_data *data) +static void davinci_mdio_init_clk(struct davinci_mdio_data *data) { u32 mdio_in, div, mdio_out_khz, access_time; @@ -111,9 +115,7 @@ static void __davinci_mdio_reset(struct davinci_mdio_data *data) if (div > CONTROL_MAX_DIV) div = CONTROL_MAX_DIV; - /* set enable and clock divider */ - __raw_writel(div | CONTROL_ENABLE, &data->regs->control); - + data->clk_div = div; /* * One mdio transaction consists of: * 32 bits of preamble @@ -134,12 +136,23 @@ static void __davinci_mdio_reset(struct davinci_mdio_data *data) data->access_time = 1; } +static void davinci_mdio_enable(struct davinci_mdio_data *data) +{ + /* set enable and clock divider */ + __raw_writel(data->clk_div | CONTROL_ENABLE, &data->regs->control); +} + static int davinci_mdio_reset(struct mii_bus *bus) { struct davinci_mdio_data *data = bus->priv; u32 phy_mask, ver; + int ret; - __davinci_mdio_reset(data); + ret = pm_runtime_get_sync(data->dev); + if (ret < 0) { + pm_runtime_put_noidle(data->dev); + return ret; + } /* wait for scan logic to settle */ msleep(PHY_MAX_ADDR * data->access_time); @@ -150,7 +163,7 @@ static int davinci_mdio_reset(struct mii_bus *bus) (ver >> 8) & 0xff, ver & 0xff); if (data->skip_scan) - return 0; + goto done; /* get phy mask from the alive register */ phy_mask = __raw_readl(&data->regs->alive); @@ -165,6 +178,10 @@ static int davinci_mdio_reset(struct mii_bus *bus) } data->bus->phy_mask = phy_mask; +done: + pm_runtime_mark_last_busy(data->dev); + pm_runtime_put_autosuspend(data->dev); + return 0; } @@ -190,7 +207,7 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data) * operation */ dev_warn(data->dev, "resetting idled controller\n"); - __davinci_mdio_reset(data); + davinci_mdio_enable(data); return -EAGAIN; } @@ -225,11 +242,10 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) if (phy_reg & ~PHY_REG_MASK || phy_id & ~PHY_ID_MASK) return -EINVAL; - spin_lock(&data->lock); - - if (data->suspended) { - spin_unlock(&data->lock); - return -ENODEV; + ret = pm_runtime_get_sync(data->dev); + if (ret < 0) { + pm_runtime_put_noidle(data->dev); + return ret; } reg = (USERACCESS_GO | USERACCESS_READ | (phy_reg << 21) | @@ -255,8 +271,8 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) break; } - spin_unlock(&data->lock); - + pm_runtime_mark_last_busy(data->dev); + pm_runtime_put_autosuspend(data->dev); return ret; } @@ -270,11 +286,10 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, if (phy_reg & ~PHY_REG_MASK || phy_id & ~PHY_ID_MASK) return -EINVAL; - spin_lock(&data->lock); - - if (data->suspended) { - spin_unlock(&data->lock); - return -ENODEV; + ret = pm_runtime_get_sync(data->dev); + if (ret < 0) { + pm_runtime_put_noidle(data->dev); + return ret; } reg = (USERACCESS_GO | USERACCESS_WRITE | (phy_reg << 21) | @@ -295,9 +310,10 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, break; } - spin_unlock(&data->lock); + pm_runtime_mark_last_busy(data->dev); + pm_runtime_put_autosuspend(data->dev); - return 0; + return ret; } #if IS_ENABLED(CONFIG_OF) @@ -320,6 +336,19 @@ static int davinci_mdio_probe_dt(struct mdio_platform_data *data, } #endif +#if IS_ENABLED(CONFIG_OF) +static const struct davinci_mdio_of_param of_cpsw_mdio_data = { + .autosuspend_delay_ms = 100, +}; + +static const struct of_device_id davinci_mdio_of_mtable[] = { + { .compatible = "ti,davinci_mdio", }, + { .compatible = "ti,cpsw-mdio", .data = &of_cpsw_mdio_data}, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, davinci_mdio_of_mtable); +#endif + static int davinci_mdio_probe(struct platform_device *pdev) { struct mdio_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -328,6 +357,7 @@ static int davinci_mdio_probe(struct platform_device *pdev) struct resource *res; struct phy_device *phy; int ret, addr; + int autosuspend_delay_ms = -1; data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); if (!data) @@ -340,9 +370,22 @@ static int davinci_mdio_probe(struct platform_device *pdev) } if (dev->of_node) { - if (davinci_mdio_probe_dt(&data->pdata, pdev)) - data->pdata = default_pdata; + const struct of_device_id *of_id; + + ret = davinci_mdio_probe_dt(&data->pdata, pdev); + if (ret) + return ret; snprintf(data->bus->id, MII_BUS_ID_SIZE, "%s", pdev->name); + + of_id = of_match_device(davinci_mdio_of_mtable, &pdev->dev); + if (of_id) { + const struct davinci_mdio_of_param *of_mdio_data; + + of_mdio_data = of_id->data; + if (of_mdio_data) + autosuspend_delay_ms = + of_mdio_data->autosuspend_delay_ms; + } } else { data->pdata = pdata ? (*pdata) : default_pdata; snprintf(data->bus->id, MII_BUS_ID_SIZE, "%s-%x", @@ -356,26 +399,25 @@ static int davinci_mdio_probe(struct platform_device *pdev) data->bus->parent = dev; data->bus->priv = data; - pm_runtime_enable(&pdev->dev); - pm_runtime_get_sync(&pdev->dev); data->clk = devm_clk_get(dev, "fck"); if (IS_ERR(data->clk)) { dev_err(dev, "failed to get device clock\n"); - ret = PTR_ERR(data->clk); - data->clk = NULL; - goto bail_out; + return PTR_ERR(data->clk); } dev_set_drvdata(dev, data); data->dev = dev; - spin_lock_init(&data->lock); res = platform_get_resource(pdev, IORESOURCE_MEM, 0); data->regs = devm_ioremap_resource(dev, res); - if (IS_ERR(data->regs)) { - ret = PTR_ERR(data->regs); - goto bail_out; - } + if (IS_ERR(data->regs)) + return PTR_ERR(data->regs); + + davinci_mdio_init_clk(data); + + pm_runtime_set_autosuspend_delay(&pdev->dev, autosuspend_delay_ms); + pm_runtime_use_autosuspend(&pdev->dev); + pm_runtime_enable(&pdev->dev); /* register the mii bus * Create PHYs from DT only in case if PHY child nodes are explicitly @@ -404,9 +446,8 @@ static int davinci_mdio_probe(struct platform_device *pdev) return 0; bail_out: - pm_runtime_put_sync(&pdev->dev); + pm_runtime_dont_use_autosuspend(&pdev->dev); pm_runtime_disable(&pdev->dev); - return ret; } @@ -417,29 +458,47 @@ static int davinci_mdio_remove(struct platform_device *pdev) if (data->bus) mdiobus_unregister(data->bus); - pm_runtime_put_sync(&pdev->dev); + pm_runtime_dont_use_autosuspend(&pdev->dev); pm_runtime_disable(&pdev->dev); return 0; } -#ifdef CONFIG_PM_SLEEP -static int davinci_mdio_suspend(struct device *dev) +#ifdef CONFIG_PM +static int davinci_mdio_runtime_suspend(struct device *dev) { struct davinci_mdio_data *data = dev_get_drvdata(dev); u32 ctrl; - spin_lock(&data->lock); - /* shutdown the scan state machine */ ctrl = __raw_readl(&data->regs->control); ctrl &= ~CONTROL_ENABLE; __raw_writel(ctrl, &data->regs->control); wait_for_idle(data); - data->suspended = true; - spin_unlock(&data->lock); - pm_runtime_put_sync(data->dev); + return 0; +} + +static int davinci_mdio_runtime_resume(struct device *dev) +{ + struct davinci_mdio_data *data = dev_get_drvdata(dev); + + davinci_mdio_enable(data); + return 0; +} +#endif + +#ifdef CONFIG_PM_SLEEP +static int davinci_mdio_suspend(struct device *dev) +{ + struct davinci_mdio_data *data = dev_get_drvdata(dev); + int ret = 0; + + data->active_in_suspend = !pm_runtime_status_suspended(dev); + if (data->active_in_suspend) + ret = pm_runtime_force_suspend(dev); + if (ret < 0) + return ret; /* Select sleep pin state */ pinctrl_pm_select_sleep_state(dev); @@ -454,31 +513,19 @@ static int davinci_mdio_resume(struct device *dev) /* Select default pin state */ pinctrl_pm_select_default_state(dev); - pm_runtime_get_sync(data->dev); - - spin_lock(&data->lock); - /* restart the scan state machine */ - __davinci_mdio_reset(data); - - data->suspended = false; - spin_unlock(&data->lock); + if (data->active_in_suspend) + pm_runtime_force_resume(dev); return 0; } #endif static const struct dev_pm_ops davinci_mdio_pm_ops = { + SET_RUNTIME_PM_OPS(davinci_mdio_runtime_suspend, + davinci_mdio_runtime_resume, NULL) SET_LATE_SYSTEM_SLEEP_PM_OPS(davinci_mdio_suspend, davinci_mdio_resume) }; -#if IS_ENABLED(CONFIG_OF) -static const struct of_device_id davinci_mdio_of_mtable[] = { - { .compatible = "ti,davinci_mdio", }, - { /* sentinel */ }, -}; -MODULE_DEVICE_TABLE(of, davinci_mdio_of_mtable); -#endif - static struct platform_driver davinci_mdio_driver = { .driver = { .name = "davinci_mdio", diff --git a/drivers/net/ethernet/ti/tlan.c b/drivers/net/ethernet/ti/tlan.c index 561703317312..ece0ea0f6b38 100644 --- a/drivers/net/ethernet/ti/tlan.c +++ b/drivers/net/ethernet/ti/tlan.c @@ -1651,7 +1651,6 @@ static u32 tlan_handle_tx_eoc(struct net_device *dev, u16 host_int) dma_addr_t head_list_phys; u32 ack = 1; - host_int = 0; if (priv->tlan_rev < 0x30) { TLAN_DBG(TLAN_DEBUG_TX, "TRANSMIT: handling TX EOC (Head=%d Tail=%d) -- IRQ\n", diff --git a/drivers/net/ethernet/tile/tilegx.c b/drivers/net/ethernet/tile/tilegx.c index 0a15acc075b3..11213a38c795 100644 --- a/drivers/net/ethernet/tile/tilegx.c +++ b/drivers/net/ethernet/tile/tilegx.c @@ -462,7 +462,7 @@ static void tile_tx_timestamp(struct sk_buff *skb, int instance) if (unlikely((shtx->tx_flags & SKBTX_HW_TSTAMP) != 0)) { struct mpipe_data *md = &mpipe_data[instance]; struct skb_shared_hwtstamps shhwtstamps; - struct timespec ts; + struct timespec64 ts; shtx->tx_flags |= SKBTX_IN_PROGRESS; gxio_mpipe_get_timestamp(&md->context, &ts); @@ -886,9 +886,9 @@ static struct ptp_clock_info ptp_mpipe_caps = { /* Sync mPIPE's timestamp up with Linux system time and register PTP clock. */ static void register_ptp_clock(struct net_device *dev, struct mpipe_data *md) { - struct timespec ts; + struct timespec64 ts; - getnstimeofday(&ts); + ktime_get_ts64(&ts); gxio_mpipe_set_timestamp(&md->context, &ts); mutex_init(&md->ptp_lock); diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 54874783476a..74e671906ddb 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -280,7 +280,7 @@ struct tc35815_regs { * Descriptors */ -/* Frame descripter */ +/* Frame descriptor */ struct FDesc { volatile __u32 FDNext; volatile __u32 FDSystem; @@ -288,7 +288,7 @@ struct FDesc { volatile __u32 FDCtl; }; -/* Buffer descripter */ +/* Buffer descriptor */ struct BDesc { volatile __u32 BuffData; volatile __u32 BDCtl; @@ -296,7 +296,7 @@ struct BDesc { #define FD_ALIGN 16 -/* Frame Descripter bit assign ---------------------------------------------- */ +/* Frame Descriptor bit assign ---------------------------------------------- */ #define FD_FDLength_MASK 0x0000FFFF /* Length MASK */ #define FD_BDCnt_MASK 0x001F0000 /* BD count MASK in FD */ #define FD_FrmOpt_MASK 0x7C000000 /* Frame option MASK */ @@ -309,7 +309,7 @@ struct BDesc { #define FD_Next_EOL 0x00000001 /* FD EOL indicator */ #define FD_BDCnt_SHIFT 16 -/* Buffer Descripter bit assign --------------------------------------------- */ +/* Buffer Descriptor bit assign --------------------------------------------- */ #define BD_BuffLength_MASK 0x0000FFFF /* Receive Data Size */ #define BD_RxBDID_MASK 0x00FF0000 /* BD ID Number MASK */ #define BD_RxBDSeqN_MASK 0x7F000000 /* Rx BD Sequence Number */ @@ -1387,7 +1387,7 @@ static int tc35815_do_interrupt(struct net_device *dev, u32 status, int limit) if (status & Int_IntExBD) { if (netif_msg_rx_err(lp)) dev_warn(&dev->dev, - "Excessive Buffer Descriptiors (%#x).\n", + "Excessive Buffer Descriptors (%#x).\n", status); dev->stats.rx_length_errors++; ret = 0; diff --git a/drivers/net/ethernet/xircom/xirc2ps_cs.c b/drivers/net/ethernet/xircom/xirc2ps_cs.c index 7b44968e02e6..ddced28e8247 100644 --- a/drivers/net/ethernet/xircom/xirc2ps_cs.c +++ b/drivers/net/ethernet/xircom/xirc2ps_cs.c @@ -1144,8 +1144,8 @@ xirc2ps_interrupt(int irq, void *dev_id) dev->stats.tx_packets += lp->last_ptr_value - n; netif_wake_queue(dev); } - if (tx_status & 0x0002) { /* Execessive collissions */ - pr_debug("tx restarted due to execssive collissions\n"); + if (tx_status & 0x0002) { /* Excessive collisions */ + pr_debug("tx restarted due to excessive collisions\n"); PutByte(XIRCREG_CR, RestartTx); /* restart transmitter process */ } if (tx_status & 0x0040) diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c index 5138407941cf..7f127dc1b7ba 100644 --- a/drivers/net/ethernet/xscale/ixp4xx_eth.c +++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c @@ -171,7 +171,6 @@ struct port { struct npe *npe; struct net_device *netdev; struct napi_struct napi; - struct phy_device *phydev; struct eth_plat_info *plat; buffer_t *rx_buff_tab[RX_DESCS], *tx_buff_tab[TX_DESCS]; struct desc *desc_tab; /* coherent */ @@ -562,7 +561,7 @@ static void ixp4xx_mdio_remove(void) static void ixp4xx_adjust_link(struct net_device *dev) { struct port *port = netdev_priv(dev); - struct phy_device *phydev = port->phydev; + struct phy_device *phydev = dev->phydev; if (!phydev->link) { if (port->speed) { @@ -976,8 +975,6 @@ static void eth_set_mcast_list(struct net_device *dev) static int eth_ioctl(struct net_device *dev, struct ifreq *req, int cmd) { - struct port *port = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; @@ -988,7 +985,7 @@ static int eth_ioctl(struct net_device *dev, struct ifreq *req, int cmd) return hwtstamp_get(dev, req); } - return phy_mii_ioctl(port->phydev, req, cmd); + return phy_mii_ioctl(dev->phydev, req, cmd); } /* ethtool support */ @@ -1005,22 +1002,9 @@ static void ixp4xx_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, "internal", sizeof(info->bus_info)); } -static int ixp4xx_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct port *port = netdev_priv(dev); - return phy_ethtool_gset(port->phydev, cmd); -} - -static int ixp4xx_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct port *port = netdev_priv(dev); - return phy_ethtool_sset(port->phydev, cmd); -} - static int ixp4xx_nway_reset(struct net_device *dev) { - struct port *port = netdev_priv(dev); - return phy_start_aneg(port->phydev); + return phy_start_aneg(dev->phydev); } int ixp46x_phc_index = -1; @@ -1054,11 +1038,11 @@ static int ixp4xx_get_ts_info(struct net_device *dev, static const struct ethtool_ops ixp4xx_ethtool_ops = { .get_drvinfo = ixp4xx_get_drvinfo, - .get_settings = ixp4xx_get_settings, - .set_settings = ixp4xx_set_settings, .nway_reset = ixp4xx_nway_reset, .get_link = ethtool_op_get_link, .get_ts_info = ixp4xx_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; @@ -1259,7 +1243,7 @@ static int eth_open(struct net_device *dev) } port->speed = 0; /* force "link up" message */ - phy_start(port->phydev); + phy_start(dev->phydev); for (i = 0; i < ETH_ALEN; i++) __raw_writel(dev->dev_addr[i], &port->regs->hw_addr[i]); @@ -1380,7 +1364,7 @@ static int eth_close(struct net_device *dev) printk(KERN_CRIT "%s: unable to disable loopback\n", dev->name); - phy_stop(port->phydev); + phy_stop(dev->phydev); if (!ports_open) qmgr_disable_irq(TXDONE_QUEUE); @@ -1405,6 +1389,7 @@ static int eth_init_one(struct platform_device *pdev) struct port *port; struct net_device *dev; struct eth_plat_info *plat = dev_get_platdata(&pdev->dev); + struct phy_device *phydev = NULL; u32 regs_phys; char phy_id[MII_BUS_ID_SIZE + 3]; int err; @@ -1466,14 +1451,14 @@ static int eth_init_one(struct platform_device *pdev) snprintf(phy_id, MII_BUS_ID_SIZE + 3, PHY_ID_FMT, mdio_bus->id, plat->phy); - port->phydev = phy_connect(dev, phy_id, &ixp4xx_adjust_link, - PHY_INTERFACE_MODE_MII); - if (IS_ERR(port->phydev)) { - err = PTR_ERR(port->phydev); + phydev = phy_connect(dev, phy_id, &ixp4xx_adjust_link, + PHY_INTERFACE_MODE_MII); + if (IS_ERR(phydev)) { + err = PTR_ERR(phydev); goto err_free_mem; } - port->phydev->irq = PHY_POLL; + phydev->irq = PHY_POLL; if ((err = register_netdev(dev))) goto err_phy_dis; @@ -1484,7 +1469,7 @@ static int eth_init_one(struct platform_device *pdev) return 0; err_phy_dis: - phy_disconnect(port->phydev); + phy_disconnect(phydev); err_free_mem: npe_port_tab[NPE_ID(port->id)] = NULL; release_resource(port->mem_res); @@ -1498,10 +1483,11 @@ err_free: static int eth_remove_one(struct platform_device *pdev) { struct net_device *dev = platform_get_drvdata(pdev); + struct phy_device *phydev = dev->phydev; struct port *port = netdev_priv(dev); unregister_netdev(dev); - phy_disconnect(port->phydev); + phy_disconnect(phydev); npe_port_tab[NPE_ID(port->id)] = NULL; npe_release(port->npe); release_resource(port->mem_res); |