diff options
Diffstat (limited to 'drivers/net/ethernet/intel/e1000e/netdev.c')
-rw-r--r-- | drivers/net/ethernet/intel/e1000e/netdev.c | 699 |
1 files changed, 577 insertions, 122 deletions
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 643c883..a177b8b 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -1,7 +1,7 @@ /******************************************************************************* Intel PRO/1000 Linux driver - Copyright(c) 1999 - 2012 Intel Corporation. + Copyright(c) 1999 - 2013 Intel Corporation. This program is free software; you can redistribute it and/or modify it under the terms and conditions of the GNU General Public License, @@ -42,7 +42,6 @@ #include <linux/slab.h> #include <net/checksum.h> #include <net/ip6_checksum.h> -#include <linux/mii.h> #include <linux/ethtool.h> #include <linux/if_vlan.h> #include <linux/cpu.h> @@ -56,7 +55,7 @@ #define DRV_EXTRAVERSION "-k" -#define DRV_VERSION "2.1.4" DRV_EXTRAVERSION +#define DRV_VERSION "2.2.14" DRV_EXTRAVERSION char e1000e_driver_name[] = "e1000e"; const char e1000e_driver_version[] = DRV_VERSION; @@ -87,20 +86,7 @@ struct e1000_reg_info { char *name; }; -#define E1000_RDFH 0x02410 /* Rx Data FIFO Head - RW */ -#define E1000_RDFT 0x02418 /* Rx Data FIFO Tail - RW */ -#define E1000_RDFHS 0x02420 /* Rx Data FIFO Head Saved - RW */ -#define E1000_RDFTS 0x02428 /* Rx Data FIFO Tail Saved - RW */ -#define E1000_RDFPC 0x02430 /* Rx Data FIFO Packet Count - RW */ - -#define E1000_TDFH 0x03410 /* Tx Data FIFO Head - RW */ -#define E1000_TDFT 0x03418 /* Tx Data FIFO Tail - RW */ -#define E1000_TDFHS 0x03420 /* Tx Data FIFO Head Saved - RW */ -#define E1000_TDFTS 0x03428 /* Tx Data FIFO Tail Saved - RW */ -#define E1000_TDFPC 0x03430 /* Tx Data FIFO Packet Count - RW */ - static const struct e1000_reg_info e1000_reg_info_tbl[] = { - /* General Registers */ {E1000_CTRL, "CTRL"}, {E1000_STATUS, "STATUS"}, @@ -488,20 +474,87 @@ static int e1000_desc_unused(struct e1000_ring *ring) } /** + * e1000e_systim_to_hwtstamp - convert system time value to hw time stamp + * @adapter: board private structure + * @hwtstamps: time stamp structure to update + * @systim: unsigned 64bit system time value. + * + * Convert the system time value stored in the RX/TXSTMP registers into a + * hwtstamp which can be used by the upper level time stamping functions. + * + * The 'systim_lock' spinlock is used to protect the consistency of the + * system time value. This is needed because reading the 64 bit time + * value involves reading two 32 bit registers. The first read latches the + * value. + **/ +static void e1000e_systim_to_hwtstamp(struct e1000_adapter *adapter, + struct skb_shared_hwtstamps *hwtstamps, + u64 systim) +{ + u64 ns; + unsigned long flags; + + spin_lock_irqsave(&adapter->systim_lock, flags); + ns = timecounter_cyc2time(&adapter->tc, systim); + spin_unlock_irqrestore(&adapter->systim_lock, flags); + + memset(hwtstamps, 0, sizeof(*hwtstamps)); + hwtstamps->hwtstamp = ns_to_ktime(ns); +} + +/** + * e1000e_rx_hwtstamp - utility function which checks for Rx time stamp + * @adapter: board private structure + * @status: descriptor extended error and status field + * @skb: particular skb to include time stamp + * + * If the time stamp is valid, convert it into the timecounter ns value + * and store that result into the shhwtstamps structure which is passed + * up the network stack. + **/ +static void e1000e_rx_hwtstamp(struct e1000_adapter *adapter, u32 status, + struct sk_buff *skb) +{ + struct e1000_hw *hw = &adapter->hw; + u64 rxstmp; + + if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP) || + !(status & E1000_RXDEXT_STATERR_TST) || + !(er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID)) + return; + + /* The Rx time stamp registers contain the time stamp. No other + * received packet will be time stamped until the Rx time stamp + * registers are read. Because only one packet can be time stamped + * at a time, the register values must belong to this packet and + * therefore none of the other additional attributes need to be + * compared. + */ + rxstmp = (u64)er32(RXSTMPL); + rxstmp |= (u64)er32(RXSTMPH) << 32; + e1000e_systim_to_hwtstamp(adapter, skb_hwtstamps(skb), rxstmp); + + adapter->flags2 &= ~FLAG2_CHECK_RX_HWTSTAMP; +} + +/** * e1000_receive_skb - helper function to handle Rx indications * @adapter: board private structure - * @status: descriptor status field as written by hardware + * @staterr: descriptor extended error and status field as written by hardware * @vlan: descriptor vlan field as written by hardware (no le/be conversion) * @skb: pointer to sk_buff to be indicated to stack **/ static void e1000_receive_skb(struct e1000_adapter *adapter, struct net_device *netdev, struct sk_buff *skb, - u8 status, __le16 vlan) + u32 staterr, __le16 vlan) { u16 tag = le16_to_cpu(vlan); + + e1000e_rx_hwtstamp(adapter, staterr, skb); + skb->protocol = eth_type_trans(skb, netdev); - if (status & E1000_RXD_STAT_VP) + if (staterr & E1000_RXD_STAT_VP) __vlan_hwaccel_put_tag(skb, tag); napi_gro_receive(&adapter->napi, skb); @@ -765,7 +818,7 @@ static void e1000_alloc_jumbo_rx_buffers(struct e1000_ring *rx_ring, struct e1000_buffer *buffer_info; struct sk_buff *skb; unsigned int i; - unsigned int bufsz = 256 - 16 /* for skb_reserve */; + unsigned int bufsz = 256 - 16; /* for skb_reserve */ i = rx_ring->next_to_use; buffer_info = &rx_ring->buffer_info[i]; @@ -1050,9 +1103,9 @@ static void e1000_print_hw_hang(struct work_struct *work) adapter->tx_hang_recheck = false; netif_stop_queue(netdev); - e1e_rphy(hw, PHY_STATUS, &phy_status); - e1e_rphy(hw, PHY_1000T_STATUS, &phy_1000t_status); - e1e_rphy(hw, PHY_EXT_STATUS, &phy_ext_status); + e1e_rphy(hw, MII_BMSR, &phy_status); + e1e_rphy(hw, MII_STAT1000, &phy_1000t_status); + e1e_rphy(hw, MII_ESTATUS, &phy_ext_status); pci_read_config_word(adapter->pdev, PCI_STATUS, &pci_status); @@ -1092,6 +1145,41 @@ static void e1000_print_hw_hang(struct work_struct *work) } /** + * e1000e_tx_hwtstamp_work - check for Tx time stamp + * @work: pointer to work struct + * + * This work function polls the TSYNCTXCTL valid bit to determine when a + * timestamp has been taken for the current stored skb. The timestamp must + * be for this skb because only one such packet is allowed in the queue. + */ +static void e1000e_tx_hwtstamp_work(struct work_struct *work) +{ + struct e1000_adapter *adapter = container_of(work, struct e1000_adapter, + tx_hwtstamp_work); + struct e1000_hw *hw = &adapter->hw; + + if (!adapter->tx_hwtstamp_skb) + return; + + if (er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_VALID) { + struct skb_shared_hwtstamps shhwtstamps; + u64 txstmp; + + txstmp = er32(TXSTMPL); + txstmp |= (u64)er32(TXSTMPH) << 32; + + e1000e_systim_to_hwtstamp(adapter, &shhwtstamps, txstmp); + + skb_tstamp_tx(adapter->tx_hwtstamp_skb, &shhwtstamps); + dev_kfree_skb_any(adapter->tx_hwtstamp_skb); + adapter->tx_hwtstamp_skb = NULL; + } else { + /* reschedule to check later */ + schedule_work(&adapter->tx_hwtstamp_work); + } +} + +/** * e1000_clean_tx_irq - Reclaim resources after transmit completes * @tx_ring: Tx descriptor ring * @@ -1345,8 +1433,8 @@ copydone: cpu_to_le16(E1000_RXDPS_HDRSTAT_HDRSP)) adapter->rx_hdr_split++; - e1000_receive_skb(adapter, netdev, skb, - staterr, rx_desc->wb.middle.vlan); + e1000_receive_skb(adapter, netdev, skb, staterr, + rx_desc->wb.middle.vlan); next_desc: rx_desc->wb.middle.status_error &= cpu_to_le32(~0xFF); @@ -1645,7 +1733,7 @@ static void e1000e_downshift_workaround(struct work_struct *work) * @irq: interrupt number * @data: pointer to a network interface device structure **/ -static irqreturn_t e1000_intr_msi(int irq, void *data) +static irqreturn_t e1000_intr_msi(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -1671,7 +1759,7 @@ static irqreturn_t e1000_intr_msi(int irq, void *data) /* disable receives */ u32 rctl = er32(RCTL); ew32(RCTL, rctl & ~E1000_RCTL_EN); - adapter->flags |= FLAG_RX_RESTART_NOW; + adapter->flags |= FLAG_RESTART_NOW; } /* guard against interrupt when we're going down */ if (!test_bit(__E1000_DOWN, &adapter->state)) @@ -1711,7 +1799,7 @@ static irqreturn_t e1000_intr_msi(int irq, void *data) * @irq: interrupt number * @data: pointer to a network interface device structure **/ -static irqreturn_t e1000_intr(int irq, void *data) +static irqreturn_t e1000_intr(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -1751,7 +1839,7 @@ static irqreturn_t e1000_intr(int irq, void *data) /* disable receives */ rctl = er32(RCTL); ew32(RCTL, rctl & ~E1000_RCTL_EN); - adapter->flags |= FLAG_RX_RESTART_NOW; + adapter->flags |= FLAG_RESTART_NOW; } /* guard against interrupt when we're going down */ if (!test_bit(__E1000_DOWN, &adapter->state)) @@ -1786,7 +1874,7 @@ static irqreturn_t e1000_intr(int irq, void *data) return IRQ_HANDLED; } -static irqreturn_t e1000_msix_other(int irq, void *data) +static irqreturn_t e1000_msix_other(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -1818,8 +1906,7 @@ no_link_interrupt: return IRQ_HANDLED; } - -static irqreturn_t e1000_intr_msix_tx(int irq, void *data) +static irqreturn_t e1000_intr_msix_tx(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -1837,7 +1924,7 @@ static irqreturn_t e1000_intr_msix_tx(int irq, void *data) return IRQ_HANDLED; } -static irqreturn_t e1000_intr_msix_rx(int irq, void *data) +static irqreturn_t e1000_intr_msix_rx(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -1924,7 +2011,6 @@ static void e1000_configure_msix(struct e1000_adapter *adapter) ctrl_ext |= E1000_CTRL_EXT_PBA_CLR; /* Auto-Mask Other interrupts upon ICR read */ -#define E1000_EIAC_MASK_82574 0x01F00000 ew32(IAM, ~E1000_EIAC_MASK_82574 | E1000_IMS_OTHER); ctrl_ext |= E1000_CTRL_EXT_EIAME; ew32(CTRL_EXT, ctrl_ext); @@ -2394,9 +2480,7 @@ void e1000e_free_rx_resources(struct e1000_ring *rx_ring) * while increasing bulk throughput. This functionality is controlled * by the InterruptThrottleRate module parameter. **/ -static unsigned int e1000_update_itr(struct e1000_adapter *adapter, - u16 itr_setting, int packets, - int bytes) +static unsigned int e1000_update_itr(u16 itr_setting, int packets, int bytes) { unsigned int retval = itr_setting; @@ -2441,7 +2525,6 @@ static unsigned int e1000_update_itr(struct e1000_adapter *adapter, static void e1000_set_itr(struct e1000_adapter *adapter) { - struct e1000_hw *hw = &adapter->hw; u16 current_itr; u32 new_itr = adapter->itr; @@ -2457,18 +2540,16 @@ static void e1000_set_itr(struct e1000_adapter *adapter) goto set_itr_now; } - adapter->tx_itr = e1000_update_itr(adapter, - adapter->tx_itr, - adapter->total_tx_packets, - adapter->total_tx_bytes); + adapter->tx_itr = e1000_update_itr(adapter->tx_itr, + adapter->total_tx_packets, + adapter->total_tx_bytes); /* conservative mode (itr 3) eliminates the lowest_latency setting */ if (adapter->itr_setting == 3 && adapter->tx_itr == lowest_latency) adapter->tx_itr = low_latency; - adapter->rx_itr = e1000_update_itr(adapter, - adapter->rx_itr, - adapter->total_rx_packets, - adapter->total_rx_bytes); + adapter->rx_itr = e1000_update_itr(adapter->rx_itr, + adapter->total_rx_packets, + adapter->total_rx_bytes); /* conservative mode (itr 3) eliminates the lowest_latency setting */ if (adapter->itr_setting == 3 && adapter->rx_itr == lowest_latency) adapter->rx_itr = low_latency; @@ -2504,10 +2585,7 @@ set_itr_now: if (adapter->msix_entries) adapter->rx_ring->set_itr = 1; else - if (new_itr) - ew32(ITR, 1000000000 / (new_itr * 256)); - else - ew32(ITR, 0); + e1000e_write_itr(adapter, new_itr); } } @@ -3049,7 +3127,7 @@ static void e1000_setup_rctl(struct e1000_adapter *adapter) ew32(RCTL, rctl); /* just started the receive unit, no need to restart */ - adapter->flags &= ~FLAG_RX_RESTART_NOW; + adapter->flags &= ~FLAG_RESTART_NOW; } /** @@ -3144,18 +3222,23 @@ static void e1000_configure_rx(struct e1000_adapter *adapter) rxcsum &= ~E1000_RXCSUM_TUOFL; ew32(RXCSUM, rxcsum); - if (adapter->hw.mac.type == e1000_pch2lan) { - /* With jumbo frames, excessive C-state transition - * latencies result in dropped transactions. - */ - if (adapter->netdev->mtu > ETH_DATA_LEN) { + /* With jumbo frames, excessive C-state transition latencies result + * in dropped transactions. + */ + if (adapter->netdev->mtu > ETH_DATA_LEN) { + u32 lat = + ((er32(PBA) & E1000_PBA_RXA_MASK) * 1024 - + adapter->max_frame_size) * 8 / 1000; + + if (adapter->flags & FLAG_IS_ICH) { u32 rxdctl = er32(RXDCTL(0)); ew32(RXDCTL(0), rxdctl | 0x3); - pm_qos_update_request(&adapter->netdev->pm_qos_req, 55); - } else { - pm_qos_update_request(&adapter->netdev->pm_qos_req, - PM_QOS_DEFAULT_VALUE); } + + pm_qos_update_request(&adapter->netdev->pm_qos_req, lat); + } else { + pm_qos_update_request(&adapter->netdev->pm_qos_req, + PM_QOS_DEFAULT_VALUE); } /* Enable Receives */ @@ -3344,6 +3427,241 @@ static void e1000e_setup_rss_hash(struct e1000_adapter *adapter) } /** + * e1000e_get_base_timinca - get default SYSTIM time increment attributes + * @adapter: board private structure + * @timinca: pointer to returned time increment attributes + * + * Get attributes for incrementing the System Time Register SYSTIML/H at + * the default base frequency, and set the cyclecounter shift value. + **/ +s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca) +{ + struct e1000_hw *hw = &adapter->hw; + u32 incvalue, incperiod, shift; + + /* Make sure clock is enabled on I217 before checking the frequency */ + if ((hw->mac.type == e1000_pch_lpt) && + !(er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) && + !(er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_ENABLED)) { + u32 fextnvm7 = er32(FEXTNVM7); + + if (!(fextnvm7 & (1 << 0))) { + ew32(FEXTNVM7, fextnvm7 | (1 << 0)); + e1e_flush(); + } + } + + switch (hw->mac.type) { + case e1000_pch2lan: + case e1000_pch_lpt: + /* On I217, the clock frequency is 25MHz or 96MHz as + * indicated by the System Clock Frequency Indication + */ + if ((hw->mac.type != e1000_pch_lpt) || + (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)) { + /* Stable 96MHz frequency */ + incperiod = INCPERIOD_96MHz; + incvalue = INCVALUE_96MHz; + shift = INCVALUE_SHIFT_96MHz; + adapter->cc.shift = shift + INCPERIOD_SHIFT_96MHz; + break; + } + /* fall-through */ + case e1000_82574: + case e1000_82583: + /* Stable 25MHz frequency */ + incperiod = INCPERIOD_25MHz; + incvalue = INCVALUE_25MHz; + shift = INCVALUE_SHIFT_25MHz; + adapter->cc.shift = shift; + break; + default: + return -EINVAL; + } + + *timinca = ((incperiod << E1000_TIMINCA_INCPERIOD_SHIFT) | + ((incvalue << shift) & E1000_TIMINCA_INCVALUE_MASK)); + + return 0; +} + +/** + * e1000e_config_hwtstamp - configure the hwtstamp registers and enable/disable + * @adapter: board private structure + * + * Outgoing time stamping can be enabled and disabled. Play nice and + * disable it when requested, although it shouldn't cause any overhead + * when no packet needs it. At most one packet in the queue may be + * marked for time stamping, otherwise it would be impossible to tell + * for sure to which packet the hardware time stamp belongs. + * + * Incoming time stamping has to be configured via the hardware filters. + * Not all combinations are supported, in particular event type has to be + * specified. Matching the kind of event packet is not supported, with the + * exception of "all V2 events regardless of level 2 or 4". + **/ +static int e1000e_config_hwtstamp(struct e1000_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct hwtstamp_config *config = &adapter->hwtstamp_config; + u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED; + u32 tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED; + u32 rxmtrl = 0; + u16 rxudp = 0; + bool is_l4 = false; + bool is_l2 = false; + u32 regval; + s32 ret_val; + + if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP)) + return -EINVAL; + + /* flags reserved for future extensions - must be zero */ + if (config->flags) + return -EINVAL; + + switch (config->tx_type) { + case HWTSTAMP_TX_OFF: + tsync_tx_ctl = 0; + break; + case HWTSTAMP_TX_ON: + break; + default: + return -ERANGE; + } + + switch (config->rx_filter) { + case HWTSTAMP_FILTER_NONE: + tsync_rx_ctl = 0; + break; + case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L4_V1; + rxmtrl = E1000_RXMTRL_PTP_V1_SYNC_MESSAGE; + is_l4 = true; + break; + case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L4_V1; + rxmtrl = E1000_RXMTRL_PTP_V1_DELAY_REQ_MESSAGE; + is_l4 = true; + break; + case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: + /* Also time stamps V2 L2 Path Delay Request/Response */ + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_V2; + rxmtrl = E1000_RXMTRL_PTP_V2_SYNC_MESSAGE; + is_l2 = true; + break; + case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: + /* Also time stamps V2 L2 Path Delay Request/Response. */ + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_V2; + rxmtrl = E1000_RXMTRL_PTP_V2_DELAY_REQ_MESSAGE; + is_l2 = true; + break; + case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: + /* Hardware cannot filter just V2 L4 Sync messages; + * fall-through to V2 (both L2 and L4) Sync. + */ + case HWTSTAMP_FILTER_PTP_V2_SYNC: + /* Also time stamps V2 Path Delay Request/Response. */ + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_L4_V2; + rxmtrl = E1000_RXMTRL_PTP_V2_SYNC_MESSAGE; + is_l2 = true; + is_l4 = true; + break; + case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: + /* Hardware cannot filter just V2 L4 Delay Request messages; + * fall-through to V2 (both L2 and L4) Delay Request. + */ + case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: + /* Also time stamps V2 Path Delay Request/Response. */ + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_L4_V2; + rxmtrl = E1000_RXMTRL_PTP_V2_DELAY_REQ_MESSAGE; + is_l2 = true; + is_l4 = true; + break; + case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: + /* Hardware cannot filter just V2 L4 or L2 Event messages; + * fall-through to all V2 (both L2 and L4) Events. + */ + case HWTSTAMP_FILTER_PTP_V2_EVENT: + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_EVENT_V2; + config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; + is_l2 = true; + is_l4 = true; + break; + case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: + /* For V1, the hardware can only filter Sync messages or + * Delay Request messages but not both so fall-through to + * time stamp all packets. + */ + case HWTSTAMP_FILTER_ALL: + is_l2 = true; + is_l4 = true; + tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_ALL; + config->rx_filter = HWTSTAMP_FILTER_ALL; + break; + default: + return -ERANGE; + } + + /* enable/disable Tx h/w time stamping */ + regval = er32(TSYNCTXCTL); + regval &= ~E1000_TSYNCTXCTL_ENABLED; + regval |= tsync_tx_ctl; + ew32(TSYNCTXCTL, regval); + if ((er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) != + (regval & E1000_TSYNCTXCTL_ENABLED)) { + e_err("Timesync Tx Control register not set as expected\n"); + return -EAGAIN; + } + + /* enable/disable Rx h/w time stamping */ + regval = er32(TSYNCRXCTL); + regval &= ~(E1000_TSYNCRXCTL_ENABLED | E1000_TSYNCRXCTL_TYPE_MASK); + regval |= tsync_rx_ctl; + ew32(TSYNCRXCTL, regval); + if ((er32(TSYNCRXCTL) & (E1000_TSYNCRXCTL_ENABLED | + E1000_TSYNCRXCTL_TYPE_MASK)) != + (regval & (E1000_TSYNCRXCTL_ENABLED | + E1000_TSYNCRXCTL_TYPE_MASK))) { + e_err("Timesync Rx Control register not set as expected\n"); + return -EAGAIN; + } + + /* L2: define ethertype filter for time stamped packets */ + if (is_l2) + rxmtrl |= ETH_P_1588; + + /* define which PTP packets get time stamped */ + ew32(RXMTRL, rxmtrl); + + /* Filter by destination port */ + if (is_l4) { + rxudp = PTP_EV_PORT; + cpu_to_be16s(&rxudp); + } + ew32(RXUDP, rxudp); + + e1e_flush(); + + /* Clear TSYNCRXCTL_VALID & TSYNCTXCTL_VALID bit */ + er32(RXSTMPH); + er32(TXSTMPH); + + /* Get and set the System Time Register SYSTIM base frequency */ + ret_val = e1000e_get_base_timinca(adapter, ®val); + if (ret_val) + return ret_val; + ew32(TIMINCA, regval); + + /* reset the ns time counter */ + timecounter_init(&adapter->tc, &adapter->cc, + ktime_to_ns(ktime_get_real())); + + return 0; +} + +/** * e1000_configure - configure the hardware for Rx and Tx * @adapter: private board structure **/ @@ -3509,14 +3827,17 @@ void e1000e_reset(struct e1000_adapter *adapter) break; case e1000_pch2lan: case e1000_pch_lpt: - fc->high_water = 0x05C20; - fc->low_water = 0x05048; - fc->pause_time = 0x0650; fc->refresh_time = 0x0400; - if (adapter->netdev->mtu > ETH_DATA_LEN) { - pba = 14; - ew32(PBA, pba); + + if (adapter->netdev->mtu <= ETH_DATA_LEN) { + fc->high_water = 0x05C20; + fc->low_water = 0x05048; + fc->pause_time = 0x0650; + break; } + + fc->high_water = ((pba << 10) * 9 / 10) & E1000_FCRTH_RTH; + fc->low_water = ((pba << 10) * 8 / 10) & E1000_FCRTL_RTL; break; } @@ -3569,6 +3890,9 @@ void e1000e_reset(struct e1000_adapter *adapter) e1000e_reset_adaptive(hw); + /* initialize systim and reset the ns time counter */ + e1000e_config_hwtstamp(adapter); + if (!netif_running(adapter->netdev) && !test_bit(__E1000_TESTING, &adapter->state)) { e1000_power_down_phy(adapter); @@ -3705,6 +4029,24 @@ void e1000e_reinit_locked(struct e1000_adapter *adapter) } /** + * e1000e_cyclecounter_read - read raw cycle counter (used by time counter) + * @cc: cyclecounter structure + **/ +static cycle_t e1000e_cyclecounter_read(const struct cyclecounter *cc) +{ + struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter, + cc); + struct e1000_hw *hw = &adapter->hw; + cycle_t systim; + + /* latch SYSTIMH on read of SYSTIML */ + systim = (cycle_t)er32(SYSTIML); + systim |= (cycle_t)er32(SYSTIMH) << 32; + + return systim; +} + +/** * e1000_sw_init - Initialize general software structures (struct e1000_adapter) * @adapter: board private structure to initialize * @@ -3730,6 +4072,17 @@ static int e1000_sw_init(struct e1000_adapter *adapter) if (e1000_alloc_queues(adapter)) return -ENOMEM; + /* Setup hardware time stamping cyclecounter */ + if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) { + adapter->cc.read = e1000e_cyclecounter_read; + adapter->cc.mask = CLOCKSOURCE_MASK(64); + adapter->cc.mult = 1; + /* cc.shift set in e1000e_get_base_tininca() */ + + spin_lock_init(&adapter->systim_lock); + INIT_WORK(&adapter->tx_hwtstamp_work, e1000e_tx_hwtstamp_work); + } + /* Explicitly disable IRQ since the NIC can be in any state. */ e1000_irq_disable(adapter); @@ -3742,7 +4095,7 @@ static int e1000_sw_init(struct e1000_adapter *adapter) * @irq: interrupt number * @data: pointer to a network interface device structure **/ -static irqreturn_t e1000_intr_msi_test(int irq, void *data) +static irqreturn_t e1000_intr_msi_test(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -3913,10 +4266,8 @@ static int e1000_open(struct net_device *netdev) e1000_update_mng_vlan(adapter); /* DMA latency requirement to workaround jumbo issue */ - if (adapter->hw.mac.type == e1000_pch2lan) - pm_qos_add_request(&adapter->netdev->pm_qos_req, - PM_QOS_CPU_DMA_LATENCY, - PM_QOS_DEFAULT_VALUE); + pm_qos_add_request(&adapter->netdev->pm_qos_req, PM_QOS_CPU_DMA_LATENCY, + PM_QOS_DEFAULT_VALUE); /* before we allocate an interrupt, we must be ready to handle it. * Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt @@ -4024,8 +4375,7 @@ static int e1000_close(struct net_device *netdev) !test_bit(__E1000_TESTING, &adapter->state)) e1000e_release_hw_control(adapter); - if (adapter->hw.mac.type == e1000_pch2lan) - pm_qos_remove_request(&adapter->netdev->pm_qos_req); + pm_qos_remove_request(&adapter->netdev->pm_qos_req); pm_runtime_put_sync(&pdev->dev); @@ -4312,14 +4662,14 @@ static void e1000_phy_read_status(struct e1000_adapter *adapter) (adapter->hw.phy.media_type == e1000_media_type_copper)) { int ret_val; - ret_val = e1e_rphy(hw, PHY_CONTROL, &phy->bmcr); - ret_val |= e1e_rphy(hw, PHY_STATUS, &phy->bmsr); - ret_val |= e1e_rphy(hw, PHY_AUTONEG_ADV, &phy->advertise); - ret_val |= e1e_rphy(hw, PHY_LP_ABILITY, &phy->lpa); - ret_val |= e1e_rphy(hw, PHY_AUTONEG_EXP, &phy->expansion); - ret_val |= e1e_rphy(hw, PHY_1000T_CTRL, &phy->ctrl1000); - ret_val |= e1e_rphy(hw, PHY_1000T_STATUS, &phy->stat1000); - ret_val |= e1e_rphy(hw, PHY_EXT_STATUS, &phy->estatus); + ret_val = e1e_rphy(hw, MII_BMCR, &phy->bmcr); + ret_val |= e1e_rphy(hw, MII_BMSR, &phy->bmsr); + ret_val |= e1e_rphy(hw, MII_ADVERTISE, &phy->advertise); + ret_val |= e1e_rphy(hw, MII_LPA, &phy->lpa); + ret_val |= e1e_rphy(hw, MII_EXPANSION, &phy->expansion); + ret_val |= e1e_rphy(hw, MII_CTRL1000, &phy->ctrl1000); + ret_val |= e1e_rphy(hw, MII_STAT1000, &phy->stat1000); + ret_val |= e1e_rphy(hw, MII_ESTATUS, &phy->estatus); if (ret_val) e_warn("Error reading PHY register\n"); } else { @@ -4346,9 +4696,8 @@ static void e1000_print_link_info(struct e1000_adapter *adapter) u32 ctrl = er32(CTRL); /* Link status message must follow this format for user tools */ - printk(KERN_INFO "e1000e: %s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n", - adapter->netdev->name, - adapter->link_speed, + pr_info("%s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n", + adapter->netdev->name, adapter->link_speed, adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half", (ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE) ? "Rx/Tx" : (ctrl & E1000_CTRL_RFCE) ? "Rx" : @@ -4401,11 +4750,11 @@ static void e1000e_enable_receives(struct e1000_adapter *adapter) { /* make sure the receive unit is started */ if ((adapter->flags & FLAG_RX_NEEDS_RESTART) && - (adapter->flags & FLAG_RX_RESTART_NOW)) { + (adapter->flags & FLAG_RESTART_NOW)) { struct e1000_hw *hw = &adapter->hw; u32 rctl = er32(RCTL); ew32(RCTL, rctl | E1000_RCTL_EN); - adapter->flags &= ~FLAG_RX_RESTART_NOW; + adapter->flags &= ~FLAG_RESTART_NOW; } } @@ -4481,6 +4830,13 @@ static void e1000_watchdog_task(struct work_struct *work) &adapter->link_speed, &adapter->link_duplex); e1000_print_link_info(adapter); + + /* check if SmartSpeed worked */ + e1000e_check_downshift(hw); + if (phy->speed_downgraded) + netdev_warn(netdev, + "Link Speed was downgraded by SmartSpeed\n"); + /* On supported PHYs, check for duplex mismatch only * if link has autonegotiated at 10/100 half */ @@ -4492,9 +4848,9 @@ static void e1000_watchdog_task(struct work_struct *work) (adapter->link_duplex == HALF_DUPLEX)) { u16 autoneg_exp; - e1e_rphy(hw, PHY_AUTONEG_EXP, &autoneg_exp); + e1e_rphy(hw, MII_EXPANSION, &autoneg_exp); - if (!(autoneg_exp & NWAY_ER_LP_NWAY_CAPS)) + if (!(autoneg_exp & EXPANSION_NWAY)) e_info("Autonegotiated half duplex but link partner cannot autoneg. Try forcing full duplex if link gets many collisions.\n"); } @@ -4567,15 +4923,22 @@ static void e1000_watchdog_task(struct work_struct *work) adapter->link_speed = 0; adapter->link_duplex = 0; /* Link status message must follow this format */ - printk(KERN_INFO "e1000e: %s NIC Link is Down\n", - adapter->netdev->name); + pr_info("%s NIC Link is Down\n", adapter->netdev->name); netif_carrier_off(netdev); if (!test_bit(__E1000_DOWN, &adapter->state)) mod_timer(&adapter->phy_info_timer, round_jiffies(jiffies + 2 * HZ)); - if (adapter->flags & FLAG_RX_NEEDS_RESTART) - schedule_work(&adapter->reset_task); + /* The link is lost so the controller stops DMA. + * If there is queued Tx work that cannot be done + * or if on an 8000ES2LAN which requires a Rx packet + * buffer work-around on link down event, reset the + * controller to flush the Tx/Rx packet buffers. + * (Do the reset outside of interrupt context). + */ + if ((adapter->flags & FLAG_RX_NEEDS_RESTART) || + (e1000_desc_unused(tx_ring) + 1 < tx_ring->count)) + adapter->flags |= FLAG_RESTART_NOW; else pm_schedule_suspend(netdev->dev.parent, LINK_TIMEOUT); @@ -4597,20 +4960,14 @@ link_up: adapter->gotc_old = adapter->stats.gotc; spin_unlock(&adapter->stats64_lock); - e1000e_update_adaptive(&adapter->hw); - - if (!netif_carrier_ok(netdev) && - (e1000_desc_unused(tx_ring) + 1 < tx_ring->count)) { - /* We've lost link, so the controller stops DMA, - * but we've got queued Tx work that's never going - * to get done, so reset controller to flush Tx. - * (Do the reset outside of interrupt context). - */ + if (adapter->flags & FLAG_RESTART_NOW) { schedule_work(&adapter->reset_task); /* return immediately since reset is imminent */ return; } + e1000e_update_adaptive(&adapter->hw); + /* Simple mode for Interrupt Throttle Rate (ITR) */ if (adapter->itr_setting == 4) { /* Symmetric Tx/Rx gets a reduced ITR=2000; @@ -4647,6 +5004,17 @@ link_up: if (adapter->flags2 & FLAG2_CHECK_PHY_HANG) e1000e_check_82574_phy_workaround(adapter); + /* Clear valid timestamp stuck in RXSTMPL/H due to a Rx error */ + if (adapter->hwtstamp_config.rx_filter != HWTSTAMP_FILTER_NONE) { + if ((adapter->flags2 & FLAG2_CHECK_RX_HWTSTAMP) && + (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID)) { + er32(RXSTMPH); + adapter->rx_hwtstamp_cleared++; + } else { + adapter->flags2 |= FLAG2_CHECK_RX_HWTSTAMP; + } + } + /* Reset the timer */ if (!test_bit(__E1000_DOWN, &adapter->state)) mod_timer(&adapter->watchdog_timer, @@ -4658,6 +5026,7 @@ link_up: #define E1000_TX_FLAGS_TSO 0x00000004 #define E1000_TX_FLAGS_IPV4 0x00000008 #define E1000_TX_FLAGS_NO_FCS 0x00000010 +#define E1000_TX_FLAGS_HWTSTAMP 0x00000020 #define E1000_TX_FLAGS_VLAN_MASK 0xffff0000 #define E1000_TX_FLAGS_VLAN_SHIFT 16 @@ -4916,6 +5285,11 @@ static void e1000_tx_queue(struct e1000_ring *tx_ring, int tx_flags, int count) if (unlikely(tx_flags & E1000_TX_FLAGS_NO_FCS)) txd_lower &= ~(E1000_TXD_CMD_IFCS); + if (unlikely(tx_flags & E1000_TX_FLAGS_HWTSTAMP)) { + txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D; + txd_upper |= E1000_TXD_EXTCMD_TSTAMP; + } + i = tx_ring->next_to_use; do { @@ -4964,12 +5338,11 @@ static int e1000_transfer_dhcp_info(struct e1000_adapter *adapter, struct e1000_hw *hw = &adapter->hw; u16 length, offset; - if (vlan_tx_tag_present(skb)) { - if (!((vlan_tx_tag_get(skb) == adapter->hw.mng_cookie.vlan_id) && - (adapter->hw.mng_cookie.status & - E1000_MNG_DHCP_COOKIE_STATUS_VLAN))) - return 0; - } + if (vlan_tx_tag_present(skb) && + !((vlan_tx_tag_get(skb) == adapter->hw.mng_cookie.vlan_id) && + (adapter->hw.mng_cookie.status & + E1000_MNG_DHCP_COOKIE_STATUS_VLAN))) + return 0; if (skb->len <= MINIMUM_DHCP_PACKET_SIZE) return 0; @@ -5140,7 +5513,15 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb, count = e1000_tx_map(tx_ring, skb, first, adapter->tx_fifo_limit, nr_frags); if (count) { - skb_tx_timestamp(skb); + if (unlikely((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && + !adapter->tx_hwtstamp_skb)) { + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + tx_flags |= E1000_TX_FLAGS_HWTSTAMP; + adapter->tx_hwtstamp_skb = skb_get(skb); + schedule_work(&adapter->tx_hwtstamp_work); + } else { + skb_tx_timestamp(skb); + } netdev_sent_queue(netdev, skb->len); e1000_tx_queue(tx_ring, tx_flags, count); @@ -5180,10 +5561,9 @@ static void e1000_reset_task(struct work_struct *work) if (test_bit(__E1000_DOWN, &adapter->state)) return; - if (!((adapter->flags & FLAG_RX_NEEDS_RESTART) && - (adapter->flags & FLAG_RX_RESTART_NOW))) { + if (!(adapter->flags & FLAG_RESTART_NOW)) { e1000e_dump(adapter); - e_err("Reset adapter\n"); + e_err("Reset adapter unexpectedly\n"); } e1000e_reinit_locked(adapter); } @@ -5369,6 +5749,61 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, return 0; } +/** + * e1000e_hwtstamp_ioctl - control hardware time stamping + * @netdev: network interface device structure + * @ifreq: interface request + * + * Outgoing time stamping can be enabled and disabled. Play nice and + * disable it when requested, although it shouldn't cause any overhead + * when no packet needs it. At most one packet in the queue may be + * marked for time stamping, otherwise it would be impossible to tell + * for sure to which packet the hardware time stamp belongs. + * + * Incoming time stamping has to be configured via the hardware filters. + * Not all combinations are supported, in particular event type has to be + * specified. Matching the kind of event packet is not supported, with the + * exception of "all V2 events regardless of level 2 or 4". + **/ +static int e1000e_hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) +{ + struct e1000_adapter *adapter = netdev_priv(netdev); + struct hwtstamp_config config; + int ret_val; + + if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) + return -EFAULT; + + adapter->hwtstamp_config = config; + + ret_val = e1000e_config_hwtstamp(adapter); + if (ret_val) + return ret_val; + + config = adapter->hwtstamp_config; + + switch (config.rx_filter) { + case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: + /* With V2 type filters which specify a Sync or Delay Request, + * Path Delay Request/Response messages are also time stamped + * by hardware so notify the caller the requested packets plus + * some others are time stamped. + */ + config.rx_filter = HWTSTAMP_FILTER_SOME; + break; + default: + break; + } + + return copy_to_user(ifr->ifr_data, &config, + sizeof(config)) ? -EFAULT : 0; +} + static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { switch (cmd) { @@ -5376,6 +5811,8 @@ static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) case SIOCGMIIREG: case SIOCSMIIREG: return e1000_mii_ioctl(netdev, ifr, cmd); + case SIOCSHWTSTAMP: + return e1000e_hwtstamp_ioctl(netdev, ifr); default: return -EOPNOTSUPP; } @@ -5386,7 +5823,7 @@ static int e1000_init_phy_wakeup(struct e1000_adapter *adapter, u32 wufc) struct e1000_hw *hw = &adapter->hw; u32 i, mac_reg; u16 phy_reg, wuc_enable; - int retval = 0; + int retval; /* copy MAC RARs to PHY RARs */ e1000_copy_rx_addrs_to_phy_ich8lan(hw); @@ -5600,14 +6037,21 @@ static void __e1000e_disable_aspm(struct pci_dev *pdev, u16 state) #else static void __e1000e_disable_aspm(struct pci_dev *pdev, u16 state) { + u16 aspm_ctl = 0; + + if (state & PCIE_LINK_STATE_L0S) + aspm_ctl |= PCI_EXP_LNKCTL_ASPM_L0S; + if (state & PCIE_LINK_STATE_L1) + aspm_ctl |= PCI_EXP_LNKCTL_ASPM_L1; + /* Both device and parent should have the same ASPM setting. * Disable ASPM in downstream component first and then upstream. */ - pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, state); + pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, aspm_ctl); if (pdev->bus->self) pcie_capability_clear_word(pdev->bus->self, PCI_EXP_LNKCTL, - state); + aspm_ctl); } #endif static void e1000e_disable_aspm(struct pci_dev *pdev, u16 state) @@ -5792,7 +6236,7 @@ static void e1000_shutdown(struct pci_dev *pdev) #ifdef CONFIG_NET_POLL_CONTROLLER -static irqreturn_t e1000_intr_msix(int irq, void *data) +static irqreturn_t e1000_intr_msix(int __always_unused irq, void *data) { struct net_device *netdev = data; struct e1000_adapter *adapter = netdev_priv(netdev); @@ -5956,7 +6400,6 @@ static void e1000_io_resume(struct pci_dev *pdev) */ if (!(adapter->flags & FLAG_HAS_AMT)) e1000e_get_hw_control(adapter); - } static void e1000_print_device_info(struct e1000_adapter *adapter) @@ -6114,8 +6557,8 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } err = pci_request_selected_regions_exclusive(pdev, - pci_select_bars(pdev, IORESOURCE_MEM), - e1000e_driver_name); + pci_select_bars(pdev, IORESOURCE_MEM), + e1000e_driver_name); if (err) goto err_pci_reg; @@ -6274,11 +6717,10 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) "NVM Read Error while reading MAC address\n"); memcpy(netdev->dev_addr, adapter->hw.mac.addr, netdev->addr_len); - memcpy(netdev->perm_addr, adapter->hw.mac.addr, netdev->addr_len); - if (!is_valid_ether_addr(netdev->perm_addr)) { + if (!is_valid_ether_addr(netdev->dev_addr)) { dev_err(&pdev->dev, "Invalid MAC Address: %pM\n", - netdev->perm_addr); + netdev->dev_addr); err = -EIO; goto err_eeprom; } @@ -6364,6 +6806,9 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* carrier off reporting is important to ethtool even BEFORE open */ netif_carrier_off(netdev); + /* init PTP hardware clock */ + e1000e_ptp_init(adapter); + e1000_print_device_info(adapter); if (pci_dev_run_wake(pdev)) @@ -6412,6 +6857,8 @@ static void e1000_remove(struct pci_dev *pdev) struct e1000_adapter *adapter = netdev_priv(netdev); bool down = test_bit(__E1000_DOWN, &adapter->state); + e1000e_ptp_remove(adapter); + /* The timers may be rescheduled, so explicitly disable them * from being rescheduled. */ @@ -6426,6 +6873,14 @@ static void e1000_remove(struct pci_dev *pdev) cancel_work_sync(&adapter->update_phy_task); cancel_work_sync(&adapter->print_hang_task); + if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) { + cancel_work_sync(&adapter->tx_hwtstamp_work); + if (adapter->tx_hwtstamp_skb) { + dev_kfree_skb_any(adapter->tx_hwtstamp_skb); + adapter->tx_hwtstamp_skb = NULL; + } + } + if (!(netdev->flags & IFF_UP)) e1000_power_down_phy(adapter); @@ -6578,7 +7033,7 @@ static int __init e1000_init_module(void) int ret; pr_info("Intel(R) PRO/1000 Network Driver - %s\n", e1000e_driver_version); - pr_info("Copyright(c) 1999 - 2012 Intel Corporation.\n"); + pr_info("Copyright(c) 1999 - 2013 Intel Corporation.\n"); ret = pci_register_driver(&e1000_driver); return ret; |