diff options
Diffstat (limited to 'drivers/net/ixgbe/ixgbe_main.c')
| -rw-r--r-- | drivers/net/ixgbe/ixgbe_main.c | 2085 |
1 files changed, 1198 insertions, 887 deletions
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index eee0b298bd36..38ab4f3f8197 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -52,13 +52,14 @@ char ixgbe_driver_name[] = "ixgbe"; static const char ixgbe_driver_string[] = "Intel(R) 10 Gigabit PCI Express Network Driver"; -#define DRV_VERSION "2.0.84-k2" +#define DRV_VERSION "3.0.12-k2" const char ixgbe_driver_version[] = DRV_VERSION; static char ixgbe_copyright[] = "Copyright (c) 1999-2010 Intel Corporation."; static const struct ixgbe_info *ixgbe_info_tbl[] = { [board_82598] = &ixgbe_82598_info, [board_82599] = &ixgbe_82599_info, + [board_X540] = &ixgbe_X540_info, }; /* ixgbe_pci_tbl - PCI Device ID Table @@ -108,10 +109,16 @@ static DEFINE_PCI_DEVICE_TABLE(ixgbe_pci_tbl) = { board_82599 }, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_CX4), board_82599 }, + {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_BACKPLANE_FCOE), + board_82599 }, + {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_SFP_FCOE), + board_82599 }, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_T3_LOM), board_82599 }, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_COMBO_BACKPLANE), board_82599 }, + {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X540T), + board_X540 }, /* required last entry */ {0, } @@ -560,6 +567,7 @@ static void ixgbe_set_ivar(struct ixgbe_adapter *adapter, s8 direction, IXGBE_WRITE_REG(hw, IXGBE_IVAR(index), ivar); break; case ixgbe_mac_82599EB: + case ixgbe_mac_X540: if (direction == -1) { /* other causes */ msix_vector |= IXGBE_IVAR_ALLOC_VAL; @@ -589,29 +597,34 @@ static inline void ixgbe_irq_rearm_queues(struct ixgbe_adapter *adapter, { u32 mask; - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: mask = (IXGBE_EIMS_RTX_QUEUE & qmask); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS, mask); - } else { + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: mask = (qmask & 0xFFFFFFFF); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS_EX(0), mask); mask = (qmask >> 32); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS_EX(1), mask); + break; + default: + break; } } -void ixgbe_unmap_and_free_tx_resource(struct ixgbe_adapter *adapter, - struct ixgbe_tx_buffer - *tx_buffer_info) +void ixgbe_unmap_and_free_tx_resource(struct ixgbe_ring *tx_ring, + struct ixgbe_tx_buffer *tx_buffer_info) { if (tx_buffer_info->dma) { if (tx_buffer_info->mapped_as_page) - dma_unmap_page(&adapter->pdev->dev, + dma_unmap_page(tx_ring->dev, tx_buffer_info->dma, tx_buffer_info->length, DMA_TO_DEVICE); else - dma_unmap_single(&adapter->pdev->dev, + dma_unmap_single(tx_ring->dev, tx_buffer_info->dma, tx_buffer_info->length, DMA_TO_DEVICE); @@ -626,92 +639,166 @@ void ixgbe_unmap_and_free_tx_resource(struct ixgbe_adapter *adapter, } /** - * ixgbe_tx_xon_state - check the tx ring xon state - * @adapter: the ixgbe adapter - * @tx_ring: the corresponding tx_ring + * ixgbe_dcb_txq_to_tc - convert a reg index to a traffic class + * @adapter: driver private struct + * @index: reg idx of queue to query (0-127) * - * If not in DCB mode, checks TFCS.TXOFF, otherwise, find out the - * corresponding TC of this tx_ring when checking TFCS. + * Helper function to determine the traffic index for a paticular + * register index. * - * Returns : true if in xon state (currently not paused) + * Returns : a tc index for use in range 0-7, or 0-3 */ -static inline bool ixgbe_tx_xon_state(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring) +u8 ixgbe_dcb_txq_to_tc(struct ixgbe_adapter *adapter, u8 reg_idx) { - u32 txoff = IXGBE_TFCS_TXOFF; + int tc = -1; + int dcb_i = adapter->ring_feature[RING_F_DCB].indices; -#ifdef CONFIG_IXGBE_DCB - if (adapter->dcb_cfg.pfc_mode_enable) { - int tc; - int reg_idx = tx_ring->reg_idx; - int dcb_i = adapter->ring_feature[RING_F_DCB].indices; + /* if DCB is not enabled the queues have no TC */ + if (!(adapter->flags & IXGBE_FLAG_DCB_ENABLED)) + return tc; + + /* check valid range */ + if (reg_idx >= adapter->hw.mac.max_tx_queues) + return tc; + + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: + tc = reg_idx >> 2; + break; + default: + if (dcb_i != 4 && dcb_i != 8) + break; + + /* if VMDq is enabled the lowest order bits determine TC */ + if (adapter->flags & (IXGBE_FLAG_SRIOV_ENABLED | + IXGBE_FLAG_VMDQ_ENABLED)) { + tc = reg_idx & (dcb_i - 1); + break; + } + + /* + * Convert the reg_idx into the correct TC. This bitmask + * targets the last full 32 ring traffic class and assigns + * it a value of 1. From there the rest of the rings are + * based on shifting the mask further up to include the + * reg_idx / 16 and then reg_idx / 8. It assumes dcB_i + * will only ever be 8 or 4 and that reg_idx will never + * be greater then 128. The code without the power of 2 + * optimizations would be: + * (((reg_idx % 32) + 32) * dcb_i) >> (9 - reg_idx / 32) + */ + tc = ((reg_idx & 0X1F) + 0x20) * dcb_i; + tc >>= 9 - (reg_idx >> 5); + } + + return tc; +} + +static void ixgbe_update_xoff_received(struct ixgbe_adapter *adapter) +{ + struct ixgbe_hw *hw = &adapter->hw; + struct ixgbe_hw_stats *hwstats = &adapter->stats; + u32 data = 0; + u32 xoff[8] = {0}; + int i; - switch (adapter->hw.mac.type) { + if ((hw->fc.current_mode == ixgbe_fc_full) || + (hw->fc.current_mode == ixgbe_fc_rx_pause)) { + switch (hw->mac.type) { case ixgbe_mac_82598EB: - tc = reg_idx >> 2; - txoff = IXGBE_TFCS_TXOFF0; + data = IXGBE_READ_REG(hw, IXGBE_LXOFFRXC); break; - case ixgbe_mac_82599EB: - tc = 0; - txoff = IXGBE_TFCS_TXOFF; - if (dcb_i == 8) { - /* TC0, TC1 */ - tc = reg_idx >> 5; - if (tc == 2) /* TC2, TC3 */ - tc += (reg_idx - 64) >> 4; - else if (tc == 3) /* TC4, TC5, TC6, TC7 */ - tc += 1 + ((reg_idx - 96) >> 3); - } else if (dcb_i == 4) { - /* TC0, TC1 */ - tc = reg_idx >> 6; - if (tc == 1) { - tc += (reg_idx - 64) >> 5; - if (tc == 2) /* TC2, TC3 */ - tc += (reg_idx - 96) >> 4; - } - } + default: + data = IXGBE_READ_REG(hw, IXGBE_LXOFFRXCNT); + } + hwstats->lxoffrxc += data; + + /* refill credits (no tx hang) if we received xoff */ + if (!data) + return; + + for (i = 0; i < adapter->num_tx_queues; i++) + clear_bit(__IXGBE_HANG_CHECK_ARMED, + &adapter->tx_ring[i]->state); + return; + } else if (!(adapter->dcb_cfg.pfc_mode_enable)) + return; + + /* update stats for each tc, only valid with PFC enabled */ + for (i = 0; i < MAX_TX_PACKET_BUFFERS; i++) { + switch (hw->mac.type) { + case ixgbe_mac_82598EB: + xoff[i] = IXGBE_READ_REG(hw, IXGBE_PXOFFRXC(i)); break; default: - tc = 0; + xoff[i] = IXGBE_READ_REG(hw, IXGBE_PXOFFRXCNT(i)); } - txoff <<= tc; + hwstats->pxoffrxc[i] += xoff[i]; + } + + /* disarm tx queues that have received xoff frames */ + for (i = 0; i < adapter->num_tx_queues; i++) { + struct ixgbe_ring *tx_ring = adapter->tx_ring[i]; + u32 tc = ixgbe_dcb_txq_to_tc(adapter, tx_ring->reg_idx); + + if (xoff[tc]) + clear_bit(__IXGBE_HANG_CHECK_ARMED, &tx_ring->state); } -#endif - return IXGBE_READ_REG(&adapter->hw, IXGBE_TFCS) & txoff; } -static inline bool ixgbe_check_tx_hang(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring, - unsigned int eop) +static u64 ixgbe_get_tx_completed(struct ixgbe_ring *ring) { + return ring->tx_stats.completed; +} + +static u64 ixgbe_get_tx_pending(struct ixgbe_ring *ring) +{ + struct ixgbe_adapter *adapter = netdev_priv(ring->netdev); struct ixgbe_hw *hw = &adapter->hw; - /* Detect a transmit hang in hardware, this serializes the - * check with the clearing of time_stamp and movement of eop */ - adapter->detect_tx_hung = false; - if (tx_ring->tx_buffer_info[eop].time_stamp && - time_after(jiffies, tx_ring->tx_buffer_info[eop].time_stamp + HZ) && - ixgbe_tx_xon_state(adapter, tx_ring)) { - /* detected Tx unit hang */ - union ixgbe_adv_tx_desc *tx_desc; - tx_desc = IXGBE_TX_DESC_ADV(tx_ring, eop); - e_err(drv, "Detected Tx Unit Hang\n" - " Tx Queue <%d>\n" - " TDH, TDT <%x>, <%x>\n" - " next_to_use <%x>\n" - " next_to_clean <%x>\n" - "tx_buffer_info[next_to_clean]\n" - " time_stamp <%lx>\n" - " jiffies <%lx>\n", - tx_ring->queue_index, - IXGBE_READ_REG(hw, tx_ring->head), - IXGBE_READ_REG(hw, tx_ring->tail), - tx_ring->next_to_use, eop, - tx_ring->tx_buffer_info[eop].time_stamp, jiffies); - return true; + u32 head = IXGBE_READ_REG(hw, IXGBE_TDH(ring->reg_idx)); + u32 tail = IXGBE_READ_REG(hw, IXGBE_TDT(ring->reg_idx)); + + if (head != tail) + return (head < tail) ? + tail - head : (tail + ring->count - head); + + return 0; +} + +static inline bool ixgbe_check_tx_hang(struct ixgbe_ring *tx_ring) +{ + u32 tx_done = ixgbe_get_tx_completed(tx_ring); + u32 tx_done_old = tx_ring->tx_stats.tx_done_old; + u32 tx_pending = ixgbe_get_tx_pending(tx_ring); + bool ret = false; + + clear_check_for_tx_hang(tx_ring); + + /* + * Check for a hung queue, but be thorough. This verifies + * that a transmit has been completed since the previous + * check AND there is at least one packet pending. The + * ARMED bit is set to indicate a potential hang. The + * bit is cleared if a pause frame is received to remove + * false hang detection due to PFC or 802.3x frames. By + * requiring this to fail twice we avoid races with + * pfc clearing the ARMED bit and conditions where we + * run the check_tx_hang logic with a transmit completion + * pending but without time to complete it yet. + */ + if ((tx_done_old == tx_done) && tx_pending) { + /* make sure it is true for two checks in a row */ + ret = test_and_set_bit(__IXGBE_HANG_CHECK_ARMED, + &tx_ring->state); + } else { + /* update completed stats and continue */ + tx_ring->tx_stats.tx_done_old = tx_done; + /* reset the countdown */ + clear_bit(__IXGBE_HANG_CHECK_ARMED, &tx_ring->state); } - return false; + return ret; } #define IXGBE_MAX_TXD_PWR 14 @@ -734,11 +821,10 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector, struct ixgbe_ring *tx_ring) { struct ixgbe_adapter *adapter = q_vector->adapter; - struct net_device *netdev = adapter->netdev; union ixgbe_adv_tx_desc *tx_desc, *eop_desc; struct ixgbe_tx_buffer *tx_buffer_info; - unsigned int i, eop, count = 0; unsigned int total_bytes = 0, total_packets = 0; + u16 i, eop, count = 0; i = tx_ring->next_to_clean; eop = tx_ring->tx_buffer_info[i].next_to_watch; @@ -749,148 +835,182 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector, bool cleaned = false; rmb(); /* read buffer_info after eop_desc */ for ( ; !cleaned; count++) { - struct sk_buff *skb; tx_desc = IXGBE_TX_DESC_ADV(tx_ring, i); tx_buffer_info = &tx_ring->tx_buffer_info[i]; - cleaned = (i == eop); - skb = tx_buffer_info->skb; - - if (cleaned && skb) { - unsigned int segs, bytecount; - unsigned int hlen = skb_headlen(skb); - - /* gso_segs is currently only valid for tcp */ - segs = skb_shinfo(skb)->gso_segs ?: 1; -#ifdef IXGBE_FCOE - /* adjust for FCoE Sequence Offload */ - if ((adapter->flags & IXGBE_FLAG_FCOE_ENABLED) - && skb_is_gso(skb) - && vlan_get_protocol(skb) == - htons(ETH_P_FCOE)) { - hlen = skb_transport_offset(skb) + - sizeof(struct fc_frame_header) + - sizeof(struct fcoe_crc_eof); - segs = DIV_ROUND_UP(skb->len - hlen, - skb_shinfo(skb)->gso_size); - } -#endif /* IXGBE_FCOE */ - /* multiply data chunks by size of headers */ - bytecount = ((segs - 1) * hlen) + skb->len; - total_packets += segs; - total_bytes += bytecount; - } - - ixgbe_unmap_and_free_tx_resource(adapter, - tx_buffer_info); tx_desc->wb.status = 0; + cleaned = (i == eop); i++; if (i == tx_ring->count) i = 0; + + if (cleaned && tx_buffer_info->skb) { + total_bytes += tx_buffer_info->bytecount; + total_packets += tx_buffer_info->gso_segs; + } + + ixgbe_unmap_and_free_tx_resource(tx_ring, + tx_buffer_info); } + tx_ring->tx_stats.completed++; eop = tx_ring->tx_buffer_info[i].next_to_watch; eop_desc = IXGBE_TX_DESC_ADV(tx_ring, eop); } tx_ring->next_to_clean = i; + tx_ring->total_bytes += total_bytes; + tx_ring->total_packets += total_packets; + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->stats.packets += total_packets; + tx_ring->stats.bytes += total_bytes; + u64_stats_update_end(&tx_ring->syncp); + + if (check_for_tx_hang(tx_ring) && ixgbe_check_tx_hang(tx_ring)) { + /* schedule immediate reset if we believe we hung */ + struct ixgbe_hw *hw = &adapter->hw; + tx_desc = IXGBE_TX_DESC_ADV(tx_ring, eop); + e_err(drv, "Detected Tx Unit Hang\n" + " Tx Queue <%d>\n" + " TDH, TDT <%x>, <%x>\n" + " next_to_use <%x>\n" + " next_to_clean <%x>\n" + "tx_buffer_info[next_to_clean]\n" + " time_stamp <%lx>\n" + " jiffies <%lx>\n", + tx_ring->queue_index, + IXGBE_READ_REG(hw, IXGBE_TDH(tx_ring->reg_idx)), + IXGBE_READ_REG(hw, IXGBE_TDT(tx_ring->reg_idx)), + tx_ring->next_to_use, eop, + tx_ring->tx_buffer_info[eop].time_stamp, jiffies); + + netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); + + e_info(probe, + "tx hang %d detected on queue %d, resetting adapter\n", + adapter->tx_timeout_count + 1, tx_ring->queue_index); + + /* schedule immediate reset if we believe we hung */ + ixgbe_tx_timeout(adapter->netdev); + + /* the adapter is about to reset, no point in enabling stuff */ + return true; + } #define TX_WAKE_THRESHOLD (DESC_NEEDED * 2) - if (unlikely(count && netif_carrier_ok(netdev) && + if (unlikely(count && netif_carrier_ok(tx_ring->netdev) && (IXGBE_DESC_UNUSED(tx_ring) >= TX_WAKE_THRESHOLD))) { /* Make sure that anybody stopping the queue after this * sees the new next_to_clean. */ smp_mb(); - if (__netif_subqueue_stopped(netdev, tx_ring->queue_index) && + if (__netif_subqueue_stopped(tx_ring->netdev, tx_ring->queue_index) && !test_bit(__IXGBE_DOWN, &adapter->state)) { - netif_wake_subqueue(netdev, tx_ring->queue_index); - ++tx_ring->restart_queue; - } - } - - if (adapter->detect_tx_hung) { - if (ixgbe_check_tx_hang(adapter, tx_ring, i)) { - /* schedule immediate reset if we believe we hung */ - e_info(probe, "tx hang %d detected, resetting " - "adapter\n", adapter->tx_timeout_count + 1); - ixgbe_tx_timeout(adapter->netdev); + netif_wake_subqueue(tx_ring->netdev, tx_ring->queue_index); + ++tx_ring->tx_stats.restart_queue; } } - /* re-arm the interrupt */ - if (count >= tx_ring->work_limit) - ixgbe_irq_rearm_queues(adapter, ((u64)1 << q_vector->v_idx)); - - tx_ring->total_bytes += total_bytes; - tx_ring->total_packets += total_packets; - u64_stats_update_begin(&tx_ring->syncp); - tx_ring->stats.packets += total_packets; - tx_ring->stats.bytes += total_bytes; - u64_stats_update_end(&tx_ring->syncp); return count < tx_ring->work_limit; } #ifdef CONFIG_IXGBE_DCA static void ixgbe_update_rx_dca(struct ixgbe_adapter *adapter, - struct ixgbe_ring *rx_ring) + struct ixgbe_ring *rx_ring, + int cpu) { + struct ixgbe_hw *hw = &adapter->hw; u32 rxctrl; - int cpu = get_cpu(); - int q = rx_ring->reg_idx; - - if (rx_ring->cpu != cpu) { - rxctrl = IXGBE_READ_REG(&adapter->hw, IXGBE_DCA_RXCTRL(q)); - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { - rxctrl &= ~IXGBE_DCA_RXCTRL_CPUID_MASK; - rxctrl |= dca3_get_tag(&adapter->pdev->dev, cpu); - } else if (adapter->hw.mac.type == ixgbe_mac_82599EB) { - rxctrl &= ~IXGBE_DCA_RXCTRL_CPUID_MASK_82599; - rxctrl |= (dca3_get_tag(&adapter->pdev->dev, cpu) << - IXGBE_DCA_RXCTRL_CPUID_SHIFT_82599); - } - rxctrl |= IXGBE_DCA_RXCTRL_DESC_DCA_EN; - rxctrl |= IXGBE_DCA_RXCTRL_HEAD_DCA_EN; - rxctrl &= ~(IXGBE_DCA_RXCTRL_DESC_RRO_EN); - rxctrl &= ~(IXGBE_DCA_RXCTRL_DESC_WRO_EN | - IXGBE_DCA_RXCTRL_DESC_HSRO_EN); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_DCA_RXCTRL(q), rxctrl); - rx_ring->cpu = cpu; + u8 reg_idx = rx_ring->reg_idx; + + rxctrl = IXGBE_READ_REG(hw, IXGBE_DCA_RXCTRL(reg_idx)); + switch (hw->mac.type) { + case ixgbe_mac_82598EB: + rxctrl &= ~IXGBE_DCA_RXCTRL_CPUID_MASK; + rxctrl |= dca3_get_tag(&adapter->pdev->dev, cpu); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + rxctrl &= ~IXGBE_DCA_RXCTRL_CPUID_MASK_82599; + rxctrl |= (dca3_get_tag(&adapter->pdev->dev, cpu) << + IXGBE_DCA_RXCTRL_CPUID_SHIFT_82599); + break; + default: + break; } - put_cpu(); + rxctrl |= IXGBE_DCA_RXCTRL_DESC_DCA_EN; + rxctrl |= IXGBE_DCA_RXCTRL_HEAD_DCA_EN; + rxctrl &= ~(IXGBE_DCA_RXCTRL_DESC_RRO_EN); + rxctrl &= ~(IXGBE_DCA_RXCTRL_DESC_WRO_EN | + IXGBE_DCA_RXCTRL_DESC_HSRO_EN); + IXGBE_WRITE_REG(hw, IXGBE_DCA_RXCTRL(reg_idx), rxctrl); } static void ixgbe_update_tx_dca(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring) + struct ixgbe_ring *tx_ring, + int cpu) { + struct ixgbe_hw *hw = &adapter->hw; u32 txctrl; + u8 reg_idx = tx_ring->reg_idx; + + switch (hw->mac.type) { + case ixgbe_mac_82598EB: + txctrl = IXGBE_READ_REG(hw, IXGBE_DCA_TXCTRL(reg_idx)); + txctrl &= ~IXGBE_DCA_TXCTRL_CPUID_MASK; + txctrl |= dca3_get_tag(&adapter->pdev->dev, cpu); + txctrl |= IXGBE_DCA_TXCTRL_DESC_DCA_EN; + txctrl &= ~IXGBE_DCA_TXCTRL_TX_WB_RO_EN; + IXGBE_WRITE_REG(hw, IXGBE_DCA_TXCTRL(reg_idx), txctrl); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + txctrl = IXGBE_READ_REG(hw, IXGBE_DCA_TXCTRL_82599(reg_idx)); + txctrl &= ~IXGBE_DCA_TXCTRL_CPUID_MASK_82599; + txctrl |= (dca3_get_tag(&adapter->pdev->dev, cpu) << + IXGBE_DCA_TXCTRL_CPUID_SHIFT_82599); + txctrl |= IXGBE_DCA_TXCTRL_DESC_DCA_EN; + txctrl &= ~IXGBE_DCA_TXCTRL_TX_WB_RO_EN; + IXGBE_WRITE_REG(hw, IXGBE_DCA_TXCTRL_82599(reg_idx), txctrl); + break; + default: + break; + } +} + +static void ixgbe_update_dca(struct ixgbe_q_vector *q_vector) +{ + struct ixgbe_adapter *adapter = q_vector->adapter; int cpu = get_cpu(); - int q = tx_ring->reg_idx; - struct ixgbe_hw *hw = &adapter->hw; + long r_idx; + int i; - if (tx_ring->cpu != cpu) { - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { - txctrl = IXGBE_READ_REG(hw, IXGBE_DCA_TXCTRL(q)); - txctrl &= ~IXGBE_DCA_TXCTRL_CPUID_MASK; - txctrl |= dca3_get_tag(&adapter->pdev->dev, cpu); - txctrl |= IXGBE_DCA_TXCTRL_DESC_DCA_EN; - IXGBE_WRITE_REG(hw, IXGBE_DCA_TXCTRL(q), txctrl); - } else if (adapter->hw.mac.type == ixgbe_mac_82599EB) { - txctrl = IXGBE_READ_REG(hw, IXGBE_DCA_TXCTRL_82599(q)); - txctrl &= ~IXGBE_DCA_TXCTRL_CPUID_MASK_82599; - txctrl |= (dca3_get_tag(&adapter->pdev->dev, cpu) << - IXGBE_DCA_TXCTRL_CPUID_SHIFT_82599); - txctrl |= IXGBE_DCA_TXCTRL_DESC_DCA_EN; - IXGBE_WRITE_REG(hw, IXGBE_DCA_TXCTRL_82599(q), txctrl); - } - tx_ring->cpu = cpu; + if (q_vector->cpu == cpu) + goto out_no_update; + + r_idx = find_first_bit(q_vector->txr_idx, adapter->num_tx_queues); + for (i = 0; i < q_vector->txr_count; i++) { + ixgbe_update_tx_dca(adapter, adapter->tx_ring[r_idx], cpu); + r_idx = find_next_bit(q_vector->txr_idx, adapter->num_tx_queues, + r_idx + 1); } + + r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); + for (i = 0; i < q_vector->rxr_count; i++) { + ixgbe_update_rx_dca(adapter, adapter->rx_ring[r_idx], cpu); + r_idx = find_next_bit(q_vector->rxr_idx, adapter->num_rx_queues, + r_idx + 1); + } + + q_vector->cpu = cpu; +out_no_update: put_cpu(); } static void ixgbe_setup_dca(struct ixgbe_adapter *adapter) { + int num_q_vectors; int i; if (!(adapter->flags & IXGBE_FLAG_DCA_ENABLED)) @@ -899,22 +1019,25 @@ static void ixgbe_setup_dca(struct ixgbe_adapter *adapter) /* always use CB2 mode, difference is masked in the CB driver */ IXGBE_WRITE_REG(&adapter->hw, IXGBE_DCA_CTRL, 2); - for (i = 0; i < adapter->num_tx_queues; i++) { - adapter->tx_ring[i]->cpu = -1; - ixgbe_update_tx_dca(adapter, adapter->tx_ring[i]); - } - for (i = 0; i < adapter->num_rx_queues; i++) { - adapter->rx_ring[i]->cpu = -1; - ixgbe_update_rx_dca(adapter, adapter->rx_ring[i]); + if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED) + num_q_vectors = adapter->num_msix_vectors - NON_Q_VECTORS; + else + num_q_vectors = 1; + + for (i = 0; i < num_q_vectors; i++) { + adapter->q_vector[i]->cpu = -1; + ixgbe_update_dca(adapter->q_vector[i]); } } static int __ixgbe_notify_dca(struct device *dev, void *data) { - struct net_device *netdev = dev_get_drvdata(dev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = dev_get_drvdata(dev); unsigned long event = *(unsigned long *)data; + if (!(adapter->flags & IXGBE_FLAG_DCA_ENABLED)) + return 0; + switch (event) { case DCA_PROVIDER_ADD: /* if we're already enabled, don't do it again */ @@ -1013,8 +1136,7 @@ static inline void ixgbe_rx_checksum(struct ixgbe_adapter *adapter, skb->ip_summed = CHECKSUM_UNNECESSARY; } -static inline void ixgbe_release_rx_desc(struct ixgbe_hw *hw, - struct ixgbe_ring *rx_ring, u32 val) +static inline void ixgbe_release_rx_desc(struct ixgbe_ring *rx_ring, u32 val) { /* * Force memory writes to complete before letting h/w @@ -1023,72 +1145,81 @@ static inline void ixgbe_release_rx_desc(struct ixgbe_hw *hw, * such as IA-64). */ wmb(); - IXGBE_WRITE_REG(hw, IXGBE_RDT(rx_ring->reg_idx), val); + writel(val, rx_ring->tail); } /** * ixgbe_alloc_rx_buffers - Replace used receive buffers; packet split - * @adapter: address of board private structure + * @rx_ring: ring to place buffers on + * @cleaned_count: number of buffers to replace **/ -void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter, - struct ixgbe_ring *rx_ring, - int cleaned_count) +void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count) { - struct net_device *netdev = adapter->netdev; - struct pci_dev *pdev = adapter->pdev; union ixgbe_adv_rx_desc *rx_desc; struct ixgbe_rx_buffer *bi; - unsigned int i; - unsigned int bufsz = rx_ring->rx_buf_len; + struct sk_buff *skb; + u16 i = rx_ring->next_to_use; - i = rx_ring->next_to_use; - bi = &rx_ring->rx_buffer_info[i]; + /* do nothing if no valid netdev defined */ + if (!rx_ring->netdev) + return; while (cleaned_count--) { rx_desc = IXGBE_RX_DESC_ADV(rx_ring, i); + bi = &rx_ring->rx_buffer_info[i]; + skb = bi->skb; - if (!bi->page_dma && - (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED)) { - if (!bi->page) { - bi->page = netdev_alloc_page(netdev); - if (!bi->page) { - adapter->alloc_rx_page_failed++; - goto no_buffers; - } - bi->page_offset = 0; - } else { - /* use a half page if we're re-using */ - bi->page_offset ^= (PAGE_SIZE / 2); - } - - bi->page_dma = dma_map_page(&pdev->dev, bi->page, - bi->page_offset, - (PAGE_SIZE / 2), - DMA_FROM_DEVICE); - } - - if (!bi->skb) { - struct sk_buff *skb = netdev_alloc_skb_ip_align(netdev, - bufsz); - bi->skb = skb; - + if (!skb) { + skb = netdev_alloc_skb_ip_align(rx_ring->netdev, + rx_ring->rx_buf_len); if (!skb) { - adapter->alloc_rx_buff_failed++; + rx_ring->rx_stats.alloc_rx_buff_failed++; goto no_buffers; } /* initialize queue mapping */ skb_record_rx_queue(skb, rx_ring->queue_index); + bi->skb = skb; } if (!bi->dma) { - bi->dma = dma_map_single(&pdev->dev, - bi->skb->data, + bi->dma = dma_map_single(rx_ring->dev, + skb->data, rx_ring->rx_buf_len, DMA_FROM_DEVICE); + if (dma_mapping_error(rx_ring->dev, bi->dma)) { + rx_ring->rx_stats.alloc_rx_buff_failed++; + bi->dma = 0; + goto no_buffers; + } } - /* Refresh the desc even if buffer_addrs didn't change because - * each write-back erases this info. */ - if (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED) { + + if (ring_is_ps_enabled(rx_ring)) { + if (!bi->page) { + bi->page = netdev_alloc_page(rx_ring->netdev); + if (!bi->page) { + rx_ring->rx_stats.alloc_rx_page_failed++; + goto no_buffers; + } + } + + if (!bi->page_dma) { + /* use a half page if we're re-using */ + bi->page_offset ^= PAGE_SIZE / 2; + bi->page_dma = dma_map_page(rx_ring->dev, + bi->page, + bi->page_offset, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + if (dma_mapping_error(rx_ring->dev, + bi->page_dma)) { + rx_ring->rx_stats.alloc_rx_page_failed++; + bi->page_dma = 0; + goto no_buffers; + } + } + + /* Refresh the desc even if buffer_addrs didn't change + * because each write-back erases this info. */ rx_desc->read.pkt_addr = cpu_to_le64(bi->page_dma); rx_desc->read.hdr_addr = cpu_to_le64(bi->dma); } else { @@ -1099,56 +1230,48 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter, i++; if (i == rx_ring->count) i = 0; - bi = &rx_ring->rx_buffer_info[i]; } no_buffers: if (rx_ring->next_to_use != i) { rx_ring->next_to_use = i; - if (i-- == 0) - i = (rx_ring->count - 1); - - ixgbe_release_rx_desc(&adapter->hw, rx_ring, i); + ixgbe_release_rx_desc(rx_ring, i); } } -static inline u16 ixgbe_get_hdr_info(union ixgbe_adv_rx_desc *rx_desc) -{ - return rx_desc->wb.lower.lo_dword.hs_rss.hdr_info; -} - -static inline u16 ixgbe_get_pkt_info(union ixgbe_adv_rx_desc *rx_desc) -{ - return rx_desc->wb.lower.lo_dword.hs_rss.pkt_info; -} - -static inline u32 ixgbe_get_rsc_count(union ixgbe_adv_rx_desc *rx_desc) +static inline u16 ixgbe_get_hlen(union ixgbe_adv_rx_desc *rx_desc) { - return (le32_to_cpu(rx_desc->wb.lower.lo_dword.data) & - IXGBE_RXDADV_RSCCNT_MASK) >> - IXGBE_RXDADV_RSCCNT_SHIFT; + /* HW will not DMA in data larger than the given buffer, even if it + * parses the (NFS, of course) header to be larger. In that case, it + * fills the header buffer and spills the rest into the page. + */ + u16 hdr_info = le16_to_cpu(rx_desc->wb.lower.lo_dword.hs_rss.hdr_info); + u16 hlen = (hdr_info & IXGBE_RXDADV_HDRBUFLEN_MASK) >> + IXGBE_RXDADV_HDRBUFLEN_SHIFT; + if (hlen > IXGBE_RX_HDR_SIZE) + hlen = IXGBE_RX_HDR_SIZE; + return hlen; } /** * ixgbe_transform_rsc_queue - change rsc queue into a full packet * @skb: pointer to the last skb in the rsc queue - * @count: pointer to number of packets coalesced in this context * * This function changes a queue full of hw rsc buffers into a completed * packet. It uses the ->prev pointers to find the first packet and then * turns it into the frag list owner. **/ -static inline struct sk_buff *ixgbe_transform_rsc_queue(struct sk_buff *skb, - u64 *count) +static inline struct sk_buff *ixgbe_transform_rsc_queue(struct sk_buff *skb) { unsigned int frag_list_size = 0; + unsigned int skb_cnt = 1; while (skb->prev) { struct sk_buff *prev = skb->prev; frag_list_size += skb->len; skb->prev = NULL; skb = prev; - *count += 1; + skb_cnt++; } skb_shinfo(skb)->frag_list = skb->next; @@ -1156,68 +1279,59 @@ static inline struct sk_buff *ixgbe_transform_rsc_queue(struct sk_buff *skb, skb->len += frag_list_size; skb->data_len += frag_list_size; skb->truesize += frag_list_size; + IXGBE_RSC_CB(skb)->skb_cnt = skb_cnt; + return skb; } -struct ixgbe_rsc_cb { - dma_addr_t dma; - bool delay_unmap; -}; - -#define IXGBE_RSC_CB(skb) ((struct ixgbe_rsc_cb *)(skb)->cb) +static inline bool ixgbe_get_rsc_state(union ixgbe_adv_rx_desc *rx_desc) +{ + return !!(le32_to_cpu(rx_desc->wb.lower.lo_dword.data) & + IXGBE_RXDADV_RSCCNT_MASK); +} -static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, +static void ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, struct ixgbe_ring *rx_ring, int *work_done, int work_to_do) { struct ixgbe_adapter *adapter = q_vector->adapter; - struct pci_dev *pdev = adapter->pdev; union ixgbe_adv_rx_desc *rx_desc, *next_rxd; struct ixgbe_rx_buffer *rx_buffer_info, *next_buffer; struct sk_buff *skb; - unsigned int i, rsc_count = 0; - u32 len, staterr; - u16 hdr_info; - bool cleaned = false; - int cleaned_count = 0; unsigned int total_rx_bytes = 0, total_rx_packets = 0; + const int current_node = numa_node_id(); #ifdef IXGBE_FCOE int ddp_bytes = 0; #endif /* IXGBE_FCOE */ + u32 staterr; + u16 i; + u16 cleaned_count = 0; + bool pkt_is_rsc = false; i = rx_ring->next_to_clean; rx_desc = IXGBE_RX_DESC_ADV(rx_ring, i); staterr = le32_to_cpu(rx_desc->wb.upper.status_error); - rx_buffer_info = &rx_ring->rx_buffer_info[i]; while (staterr & IXGBE_RXD_STAT_DD) { u32 upper_len = 0; - if (*work_done >= work_to_do) - break; - (*work_done)++; rmb(); /* read descriptor and rx_buffer_info after status DD */ - if (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED) { - hdr_info = le16_to_cpu(ixgbe_get_hdr_info(rx_desc)); - len = (hdr_info & IXGBE_RXDADV_HDRBUFLEN_MASK) >> - IXGBE_RXDADV_HDRBUFLEN_SHIFT; - upper_len = le16_to_cpu(rx_desc->wb.upper.length); - if ((len > IXGBE_RX_HDR_SIZE) || - (upper_len && !(hdr_info & IXGBE_RXDADV_SPH))) - len = IXGBE_RX_HDR_SIZE; - } else { - len = le16_to_cpu(rx_desc->wb.upper.length); - } - cleaned = true; + rx_buffer_info = &rx_ring->rx_buffer_info[i]; + skb = rx_buffer_info->skb; - prefetch(skb->data); rx_buffer_info->skb = NULL; + prefetch(skb->data); + + if (ring_is_rsc_enabled(rx_ring)) + pkt_is_rsc = ixgbe_get_rsc_state(rx_desc); + /* if this is a skb from previous receive DMA will be 0 */ if (rx_buffer_info->dma) { - if ((adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) && - (!(staterr & IXGBE_RXD_STAT_EOP)) && - (!(skb->prev))) { + u16 hlen; + if (pkt_is_rsc && + !(staterr & IXGBE_RXD_STAT_EOP) && + !skb->prev) { /* * When HWRSC is enabled, delay unmapping * of the first packet. It carries the @@ -1228,29 +1342,42 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, IXGBE_RSC_CB(skb)->delay_unmap = true; IXGBE_RSC_CB(skb)->dma = rx_buffer_info->dma; } else { - dma_unmap_single(&pdev->dev, + dma_unmap_single(rx_ring->dev, rx_buffer_info->dma, rx_ring->rx_buf_len, DMA_FROM_DEVICE); } rx_buffer_info->dma = 0; - skb_put(skb, len); + + if (ring_is_ps_enabled(rx_ring)) { + hlen = ixgbe_get_hlen(rx_desc); + upper_len = le16_to_cpu(rx_desc->wb.upper.length); + } else { + hlen = le16_to_cpu(rx_desc->wb.upper.length); + } + + skb_put(skb, hlen); + } else { + /* assume packet split since header is unmapped */ + upper_len = le16_to_cpu(rx_desc->wb.upper.length); } if (upper_len) { - dma_unmap_page(&pdev->dev, rx_buffer_info->page_dma, - PAGE_SIZE / 2, DMA_FROM_DEVICE); + dma_unmap_page(rx_ring->dev, + rx_buffer_info->page_dma, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); rx_buffer_info->page_dma = 0; skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, rx_buffer_info->page, rx_buffer_info->page_offset, upper_len); - if ((rx_ring->rx_buf_len > (PAGE_SIZE / 2)) || - (page_count(rx_buffer_info->page) != 1)) - rx_buffer_info->page = NULL; - else + if ((page_count(rx_buffer_info->page) == 1) && + (page_to_nid(rx_buffer_info->page) == current_node)) get_page(rx_buffer_info->page); + else + rx_buffer_info->page = NULL; skb->len += upper_len; skb->data_len += upper_len; @@ -1265,10 +1392,7 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, prefetch(next_rxd); cleaned_count++; - if (adapter->flags2 & IXGBE_FLAG2_RSC_CAPABLE) - rsc_count = ixgbe_get_rsc_count(rx_desc); - - if (rsc_count) { + if (pkt_is_rsc) { u32 nextp = (staterr & IXGBE_RXDADV_NEXTP_MASK) >> IXGBE_RXDADV_NEXTP_SHIFT; next_buffer = &rx_ring->rx_buffer_info[nextp]; @@ -1276,32 +1400,8 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, next_buffer = &rx_ring->rx_buffer_info[i]; } - if (staterr & IXGBE_RXD_STAT_EOP) { - if (skb->prev) - skb = ixgbe_transform_rsc_queue(skb, - &(rx_ring->rsc_count)); - if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) { - if (IXGBE_RSC_CB(skb)->delay_unmap) { - dma_unmap_single(&pdev->dev, - IXGBE_RSC_CB(skb)->dma, - rx_ring->rx_buf_len, - DMA_FROM_DEVICE); - IXGBE_RSC_CB(skb)->dma = 0; - IXGBE_RSC_CB(skb)->delay_unmap = false; - } - if (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED) - rx_ring->rsc_count += - skb_shinfo(skb)->nr_frags; - else - rx_ring->rsc_count++; - rx_ring->rsc_flush++; - } - u64_stats_update_begin(&rx_ring->syncp); - rx_ring->stats.packets++; - rx_ring->stats.bytes += skb->len; - u64_stats_update_end(&rx_ring->syncp); - } else { - if (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED) { + if (!(staterr & IXGBE_RXD_STAT_EOP)) { + if (ring_is_ps_enabled(rx_ring)) { rx_buffer_info->skb = next_buffer->skb; rx_buffer_info->dma = next_buffer->dma; next_buffer->skb = skb; @@ -1310,12 +1410,45 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, skb->next = next_buffer->skb; skb->next->prev = skb; } - rx_ring->non_eop_descs++; + rx_ring->rx_stats.non_eop_descs++; goto next_desc; } + if (skb->prev) { + skb = ixgbe_transform_rsc_queue(skb); + /* if we got here without RSC the packet is invalid */ + if (!pkt_is_rsc) { + __pskb_trim(skb, 0); + rx_buffer_info->skb = skb; + goto next_desc; + } + } + + if (ring_is_rsc_enabled(rx_ring)) { + if (IXGBE_RSC_CB(skb)->delay_unmap) { + dma_unmap_single(rx_ring->dev, + IXGBE_RSC_CB(skb)->dma, + rx_ring->rx_buf_len, + DMA_FROM_DEVICE); + IXGBE_RSC_CB(skb)->dma = 0; + IXGBE_RSC_CB(skb)->delay_unmap = false; + } + } + if (pkt_is_rsc) { + if (ring_is_ps_enabled(rx_ring)) + rx_ring->rx_stats.rsc_count += + skb_shinfo(skb)->nr_frags; + else + rx_ring->rx_stats.rsc_count += + IXGBE_RSC_CB(skb)->skb_cnt; + rx_ring->rx_stats.rsc_flush++; + } + + /* ERR_MASK will only have valid bits if EOP set */ if (staterr & IXGBE_RXDADV_ERR_FRAME_ERR_MASK) { - dev_kfree_skb_irq(skb); + /* trim packet back to size 0 and recycle it */ + __pskb_trim(skb, 0); + rx_buffer_info->skb = skb; goto next_desc; } @@ -1325,7 +1458,7 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, total_rx_bytes += skb->len; total_rx_packets++; - skb->protocol = eth_type_trans(skb, adapter->netdev); + skb->protocol = eth_type_trans(skb, rx_ring->netdev); #ifdef IXGBE_FCOE /* if ddp, not passing to ULD unless for FCP_RSP or error */ if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) { @@ -1339,16 +1472,18 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, next_desc: rx_desc->wb.upper.status_error = 0; + (*work_done)++; + if (*work_done >= work_to_do) + break; + /* return some buffers to hardware, one at a time is too slow */ if (cleaned_count >= IXGBE_RX_BUFFER_WRITE) { - ixgbe_alloc_rx_buffers(adapter, rx_ring, cleaned_count); + ixgbe_alloc_rx_buffers(rx_ring, cleaned_count); cleaned_count = 0; } /* use prefetched values */ rx_desc = next_rxd; - rx_buffer_info = &rx_ring->rx_buffer_info[i]; - staterr = le32_to_cpu(rx_desc->wb.upper.status_error); } @@ -1356,14 +1491,14 @@ next_desc: cleaned_count = IXGBE_DESC_UNUSED(rx_ring); if (cleaned_count) - ixgbe_alloc_rx_buffers(adapter, rx_ring, cleaned_count); + ixgbe_alloc_rx_buffers(rx_ring, cleaned_count); #ifdef IXGBE_FCOE /* include DDPed FCoE data */ if (ddp_bytes > 0) { unsigned int mss; - mss = adapter->netdev->mtu - sizeof(struct fcoe_hdr) - + mss = rx_ring->netdev->mtu - sizeof(struct fcoe_hdr) - sizeof(struct fc_frame_header) - sizeof(struct fcoe_crc_eof); if (mss > 512) @@ -1375,8 +1510,10 @@ next_desc: rx_ring->total_packets += total_rx_packets; rx_ring->total_bytes += total_rx_bytes; - - return cleaned; + u64_stats_update_begin(&rx_ring->syncp); + rx_ring->stats.packets += total_rx_packets; + rx_ring->stats.bytes += total_rx_bytes; + u64_stats_update_end(&rx_ring->syncp); } static int ixgbe_clean_rxonly(struct napi_struct *, int); @@ -1390,7 +1527,7 @@ static int ixgbe_clean_rxonly(struct napi_struct *, int); static void ixgbe_configure_msix(struct ixgbe_adapter *adapter) { struct ixgbe_q_vector *q_vector; - int i, j, q_vectors, v_idx, r_idx; + int i, q_vectors, v_idx, r_idx; u32 mask; q_vectors = adapter->num_msix_vectors - NON_Q_VECTORS; @@ -1406,8 +1543,8 @@ static void ixgbe_configure_msix(struct ixgbe_adapter *adapter) adapter->num_rx_queues); for (i = 0; i < q_vector->rxr_count; i++) { - j = adapter->rx_ring[r_idx]->reg_idx; - ixgbe_set_ivar(adapter, 0, j, v_idx); + u8 reg_idx = adapter->rx_ring[r_idx]->reg_idx; + ixgbe_set_ivar(adapter, 0, reg_idx, v_idx); r_idx = find_next_bit(q_vector->rxr_idx, adapter->num_rx_queues, r_idx + 1); @@ -1416,8 +1553,8 @@ static void ixgbe_configure_msix(struct ixgbe_adapter *adapter) adapter->num_tx_queues); for (i = 0; i < q_vector->txr_count; i++) { - j = adapter->tx_ring[r_idx]->reg_idx; - ixgbe_set_ivar(adapter, 1, j, v_idx); + u8 reg_idx = adapter->tx_ring[r_idx]->reg_idx; + ixgbe_set_ivar(adapter, 1, reg_idx, v_idx); r_idx = find_next_bit(q_vector->txr_idx, adapter->num_tx_queues, r_idx + 1); @@ -1448,11 +1585,19 @@ static void ixgbe_configure_msix(struct ixgbe_adapter *adapter) } } - if (adapter->hw.mac.type == ixgbe_mac_82598EB) + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: ixgbe_set_ivar(adapter, -1, IXGBE_IVAR_OTHER_CAUSES_INDEX, v_idx); - else if (adapter->hw.mac.type == ixgbe_mac_82599EB) + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: ixgbe_set_ivar(adapter, -1, 1, v_idx); + break; + + default: + break; + } IXGBE_WRITE_REG(&adapter->hw, IXGBE_EITR(v_idx), 1950); /* set up to autoclear timer, and the vectors */ @@ -1548,12 +1693,15 @@ void ixgbe_write_eitr(struct ixgbe_q_vector *q_vector) int v_idx = q_vector->v_idx; u32 itr_reg = EITR_INTS_PER_SEC_TO_REG(q_vector->eitr); - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: /* must write high and low 16 bits to reset counter */ itr_reg |= (itr_reg << 16); - } else if (adapter->hw.mac.type == ixgbe_mac_82599EB) { + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: /* - * 82599 can support a value of zero, so allow it for + * 82599 and X540 can support a value of zero, so allow it for * max interrupt rate, but there is an errata where it can * not be zero with RSC */ @@ -1566,6 +1714,9 @@ void ixgbe_write_eitr(struct ixgbe_q_vector *q_vector) * immediate assertion of the interrupt */ itr_reg |= IXGBE_EITR_CNT_WDIS; + break; + default: + break; } IXGBE_WRITE_REG(hw, IXGBE_EITR(v_idx), itr_reg); } @@ -1573,14 +1724,13 @@ void ixgbe_write_eitr(struct ixgbe_q_vector *q_vector) static void ixgbe_set_itr_msix(struct ixgbe_q_vector *q_vector) { struct ixgbe_adapter *adapter = q_vector->adapter; + int i, r_idx; u32 new_itr; u8 current_itr, ret_itr; - int i, r_idx; - struct ixgbe_ring *rx_ring, *tx_ring; r_idx = find_first_bit(q_vector->txr_idx, adapter->num_tx_queues); for (i = 0; i < q_vector->txr_count; i++) { - tx_ring = adapter->tx_ring[r_idx]; + struct ixgbe_ring *tx_ring = adapter->tx_ring[r_idx]; ret_itr = ixgbe_update_itr(adapter, q_vector->eitr, q_vector->tx_itr, tx_ring->total_packets, @@ -1595,7 +1745,7 @@ static void ixgbe_set_itr_msix(struct ixgbe_q_vector *q_vector) r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); for (i = 0; i < q_vector->rxr_count; i++) { - rx_ring = adapter->rx_ring[r_idx]; + struct ixgbe_ring *rx_ring = adapter->rx_ring[r_idx]; ret_itr = ixgbe_update_itr(adapter, q_vector->eitr, q_vector->rx_itr, rx_ring->total_packets, @@ -1626,7 +1776,7 @@ static void ixgbe_set_itr_msix(struct ixgbe_q_vector *q_vector) if (new_itr != q_vector->eitr) { /* do an exponential smoothing */ - new_itr = ((q_vector->eitr * 90)/100) + ((new_itr * 10)/100); + new_itr = ((q_vector->eitr * 9) + new_itr)/10; /* save the algorithm value here, not the smoothed one */ q_vector->eitr = new_itr; @@ -1694,17 +1844,18 @@ static void ixgbe_check_sfp_event(struct ixgbe_adapter *adapter, u32 eicr) { struct ixgbe_hw *hw = &adapter->hw; + if (eicr & IXGBE_EICR_GPI_SDP2) { + /* Clear the interrupt */ + IXGBE_WRITE_REG(hw, IXGBE_EICR, IXGBE_EICR_GPI_SDP2); + if (!test_bit(__IXGBE_DOWN, &adapter->state)) + schedule_work(&adapter->sfp_config_module_task); + } + if (eicr & IXGBE_EICR_GPI_SDP1) { /* Clear the interrupt */ IXGBE_WRITE_REG(hw, IXGBE_EICR, IXGBE_EICR_GPI_SDP1); - schedule_work(&adapter->multispeed_fiber_task); - } else if (eicr & IXGBE_EICR_GPI_SDP2) { - /* Clear the interrupt */ - IXGBE_WRITE_REG(hw, IXGBE_EICR, IXGBE_EICR_GPI_SDP2); - schedule_work(&adapter->sfp_config_module_task); - } else { - /* Interrupt isn't for us... */ - return; + if (!test_bit(__IXGBE_DOWN, &adapter->state)) + schedule_work(&adapter->multispeed_fiber_task); } } @@ -1744,16 +1895,16 @@ static irqreturn_t ixgbe_msix_lsc(int irq, void *data) if (eicr & IXGBE_EICR_MAILBOX) ixgbe_msg_task(adapter); - if (hw->mac.type == ixgbe_mac_82598EB) - ixgbe_check_fan_failure(adapter, eicr); - - if (hw->mac.type == ixgbe_mac_82599EB) { + switch (hw->mac.type) { + case ixgbe_mac_82599EB: ixgbe_check_sfp_event(adapter, eicr); - adapter->interrupt_event = eicr; if ((adapter->flags2 & IXGBE_FLAG2_TEMP_SENSOR_CAPABLE) && - ((eicr & IXGBE_EICR_GPI_SDP0) || (eicr & IXGBE_EICR_LSC))) + ((eicr & IXGBE_EICR_GPI_SDP0) || (eicr & IXGBE_EICR_LSC))) { + adapter->interrupt_event = eicr; schedule_work(&adapter->check_overtemp_task); - + } + /* now fallthrough to handle Flow Director */ + case ixgbe_mac_X540: /* Handle Flow Director Full threshold interrupt */ if (eicr & IXGBE_EICR_FLOW_DIR) { int i; @@ -1763,12 +1914,18 @@ static irqreturn_t ixgbe_msix_lsc(int irq, void *data) for (i = 0; i < adapter->num_tx_queues; i++) { struct ixgbe_ring *tx_ring = adapter->tx_ring[i]; - if (test_and_clear_bit(__IXGBE_FDIR_INIT_DONE, - &tx_ring->reinit_state)) + if (test_and_clear_bit(__IXGBE_TX_FDIR_INIT_DONE, + &tx_ring->state)) schedule_work(&adapter->fdir_reinit_task); } } + break; + default: + break; } + + ixgbe_check_fan_failure(adapter, eicr); + if (!test_bit(__IXGBE_DOWN, &adapter->state)) IXGBE_WRITE_REG(hw, IXGBE_EIMS, IXGBE_EIMS_OTHER); @@ -1779,15 +1936,24 @@ static inline void ixgbe_irq_enable_queues(struct ixgbe_adapter *adapter, u64 qmask) { u32 mask; + struct ixgbe_hw *hw = &adapter->hw; - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { + switch (hw->mac.type) { + case ixgbe_mac_82598EB: mask = (IXGBE_EIMS_RTX_QUEUE & qmask); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMS, mask); - } else { + IXGBE_WRITE_REG(hw, IXGBE_EIMS, mask); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: mask = (qmask & 0xFFFFFFFF); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMS_EX(0), mask); + if (mask) + IXGBE_WRITE_REG(hw, IXGBE_EIMS_EX(0), mask); mask = (qmask >> 32); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMS_EX(1), mask); + if (mask) + IXGBE_WRITE_REG(hw, IXGBE_EIMS_EX(1), mask); + break; + default: + break; } /* skip the flush */ } @@ -1796,15 +1962,24 @@ static inline void ixgbe_irq_disable_queues(struct ixgbe_adapter *adapter, u64 qmask) { u32 mask; + struct ixgbe_hw *hw = &adapter->hw; - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { + switch (hw->mac.type) { + case ixgbe_mac_82598EB: mask = (IXGBE_EIMS_RTX_QUEUE & qmask); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, mask); - } else { + IXGBE_WRITE_REG(hw, IXGBE_EIMC, mask); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: mask = (qmask & 0xFFFFFFFF); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC_EX(0), mask); + if (mask) + IXGBE_WRITE_REG(hw, IXGBE_EIMC_EX(0), mask); mask = (qmask >> 32); - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC_EX(1), mask); + if (mask) + IXGBE_WRITE_REG(hw, IXGBE_EIMC_EX(1), mask); + break; + default: + break; } /* skip the flush */ } @@ -1847,8 +2022,13 @@ static irqreturn_t ixgbe_msix_clean_rx(int irq, void *data) int r_idx; int i; +#ifdef CONFIG_IXGBE_DCA + if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) + ixgbe_update_dca(q_vector); +#endif + r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); - for (i = 0; i < q_vector->rxr_count; i++) { + for (i = 0; i < q_vector->rxr_count; i++) { rx_ring = adapter->rx_ring[r_idx]; rx_ring->total_bytes = 0; rx_ring->total_packets = 0; @@ -1859,7 +2039,6 @@ static irqreturn_t ixgbe_msix_clean_rx(int irq, void *data) if (!q_vector->rxr_count) return IRQ_HANDLED; - /* disable interrupts on this vector only */ /* EIAM disabled interrupts (on this vector) for us */ napi_schedule(&q_vector->napi); @@ -1918,13 +2097,14 @@ static int ixgbe_clean_rxonly(struct napi_struct *napi, int budget) int work_done = 0; long r_idx; - r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); - rx_ring = adapter->rx_ring[r_idx]; #ifdef CONFIG_IXGBE_DCA if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) - ixgbe_update_rx_dca(adapter, rx_ring); + ixgbe_update_dca(q_vector); #endif + r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); + rx_ring = adapter->rx_ring[r_idx]; + ixgbe_clean_rx_irq(q_vector, rx_ring, &work_done, budget); /* If all Rx work done, exit the polling mode */ @@ -1958,13 +2138,14 @@ static int ixgbe_clean_rxtx_many(struct napi_struct *napi, int budget) long r_idx; bool tx_clean_complete = true; +#ifdef CONFIG_IXGBE_DCA + if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) + ixgbe_update_dca(q_vector); +#endif + r_idx = find_first_bit(q_vector->txr_idx, adapter->num_tx_queues); for (i = 0; i < q_vector->txr_count; i++) { ring = adapter->tx_ring[r_idx]; -#ifdef CONFIG_IXGBE_DCA - if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) - ixgbe_update_tx_dca(adapter, ring); -#endif tx_clean_complete &= ixgbe_clean_tx_irq(q_vector, ring); r_idx = find_next_bit(q_vector->txr_idx, adapter->num_tx_queues, r_idx + 1); @@ -1977,10 +2158,6 @@ static int ixgbe_clean_rxtx_many(struct napi_struct *napi, int budget) r_idx = find_first_bit(q_vector->rxr_idx, adapter->num_rx_queues); for (i = 0; i < q_vector->rxr_count; i++) { ring = adapter->rx_ring[r_idx]; -#ifdef CONFIG_IXGBE_DCA - if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) - ixgbe_update_rx_dca(adapter, ring); -#endif ixgbe_clean_rx_irq(q_vector, ring, &work_done, budget); r_idx = find_next_bit(q_vector->rxr_idx, adapter->num_rx_queues, r_idx + 1); @@ -2019,13 +2196,14 @@ static int ixgbe_clean_txonly(struct napi_struct *napi, int budget) int work_done = 0; long r_idx; - r_idx = find_first_bit(q_vector->txr_idx, adapter->num_tx_queues); - tx_ring = adapter->tx_ring[r_idx]; #ifdef CONFIG_IXGBE_DCA if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) - ixgbe_update_tx_dca(adapter, tx_ring); + ixgbe_update_dca(q_vector); #endif + r_idx = find_first_bit(q_vector->txr_idx, adapter->num_tx_queues); + tx_ring = adapter->tx_ring[r_idx]; + if (!ixgbe_clean_tx_irq(q_vector, tx_ring)) work_done = budget; @@ -2046,24 +2224,27 @@ static inline void map_vector_to_rxq(struct ixgbe_adapter *a, int v_idx, int r_idx) { struct ixgbe_q_vector *q_vector = a->q_vector[v_idx]; + struct ixgbe_ring *rx_ring = a->rx_ring[r_idx]; set_bit(r_idx, q_vector->rxr_idx); q_vector->rxr_count++; + rx_ring->q_vector = q_vector; } static inline void map_vector_to_txq(struct ixgbe_adapter *a, int v_idx, int t_idx) { struct ixgbe_q_vector *q_vector = a->q_vector[v_idx]; + struct ixgbe_ring *tx_ring = a->tx_ring[t_idx]; set_bit(t_idx, q_vector->txr_idx); q_vector->txr_count++; + tx_ring->q_vector = q_vector; } /** * ixgbe_map_rings_to_vectors - Maps descriptor rings to vectors * @adapter: board private structure to initialize - * @vectors: allotted vector count for descriptor rings * * This function maps descriptor rings to the queue-specific vectors * we were allotted through the MSI-X enabling code. Ideally, we'd have @@ -2071,9 +2252,9 @@ static inline void map_vector_to_txq(struct ixgbe_adapter *a, int v_idx, * group the rings as "efficiently" as possible. You would add new * mapping configurations in here. **/ -static int ixgbe_map_rings_to_vectors(struct ixgbe_adapter *adapter, - int vectors) +static int ixgbe_map_rings_to_vectors(struct ixgbe_adapter *adapter) { + int q_vectors; int v_start = 0; int rxr_idx = 0, txr_idx = 0; int rxr_remaining = adapter->num_rx_queues; @@ -2086,11 +2267,13 @@ static int ixgbe_map_rings_to_vectors(struct ixgbe_adapter *adapter, if (!(adapter->flags & IXGBE_FLAG_MSIX_ENABLED)) goto out; + q_vectors = adapter->num_msix_vectors - NON_Q_VECTORS; + /* * The ideal configuration... * We have enough vectors to map one per queue. */ - if (vectors == adapter->num_rx_queues + adapter->num_tx_queues) { + if (q_vectors == adapter->num_rx_queues + adapter->num_tx_queues) { for (; rxr_idx < rxr_remaining; v_start++, rxr_idx++) map_vector_to_rxq(adapter, v_start, rxr_idx); @@ -2106,23 +2289,20 @@ static int ixgbe_map_rings_to_vectors(struct ixgbe_adapter *adapter, * multiple queues per vector. */ /* Re-adjusting *qpv takes care of the remainder. */ - for (i = v_start; i < vectors; i++) { - rqpv = DIV_ROUND_UP(rxr_remaining, vectors - i); + for (i = v_start; i < q_vectors; i++) { + rqpv = DIV_ROUND_UP(rxr_remaining, q_vectors - i); for (j = 0; j < rqpv; j++) { map_vector_to_rxq(adapter, i, rxr_idx); rxr_idx++; rxr_remaining--; } - } - for (i = v_start; i < vectors; i++) { - tqpv = DIV_ROUND_UP(txr_remaining, vectors - i); + tqpv = DIV_ROUND_UP(txr_remaining, q_vectors - i); for (j = 0; j < tqpv; j++) { map_vector_to_txq(adapter, i, txr_idx); txr_idx++; txr_remaining--; } } - out: return err; } @@ -2144,30 +2324,36 @@ static int ixgbe_request_msix_irqs(struct ixgbe_adapter *adapter) /* Decrement for Other and TCP Timer vectors */ q_vectors = adapter->num_msix_vectors - NON_Q_VECTORS; - /* Map the Tx/Rx rings to the vectors we were allotted. */ - err = ixgbe_map_rings_to_vectors(adapter, q_vectors); + err = ixgbe_map_rings_to_vectors(adapter); if (err) - goto out; + return err; -#define SET_HANDLER(_v) ((!(_v)->rxr_count) ? &ixgbe_msix_clean_tx : \ - (!(_v)->txr_count) ? &ixgbe_msix_clean_rx : \ - &ixgbe_msix_clean_many) +#define SET_HANDLER(_v) (((_v)->rxr_count && (_v)->txr_count) \ + ? &ixgbe_msix_clean_many : \ + (_v)->rxr_count ? &ixgbe_msix_clean_rx : \ + (_v)->txr_count ? &ixgbe_msix_clean_tx : \ + NULL) for (vector = 0; vector < q_vectors; vector++) { - handler = SET_HANDLER(adapter->q_vector[vector]); + struct ixgbe_q_vector *q_vector = adapter->q_vector[vector]; + handler = SET_HANDLER(q_vector); if (handler == &ixgbe_msix_clean_rx) { - sprintf(adapter->name[vector], "%s-%s-%d", - netdev->name, "rx", ri++); + snprintf(q_vector->name, sizeof(q_vector->name) - 1, + "%s-%s-%d", netdev->name, "rx", ri++); } else if (handler == &ixgbe_msix_clean_tx) { - sprintf(adapter->name[vector], "%s-%s-%d", - netdev->name, "tx", ti++); - } else - sprintf(adapter->name[vector], "%s-%s-%d", - netdev->name, "TxRx", vector); - + snprintf(q_vector->name, sizeof(q_vector->name) - 1, + "%s-%s-%d", netdev->name, "tx", ti++); + } else if (handler == &ixgbe_msix_clean_many) { + snprintf(q_vector->name, sizeof(q_vector->name) - 1, + "%s-%s-%d", netdev->name, "TxRx", ri++); + ti++; + } else { + /* skip this unused q_vector */ + continue; + } err = request_irq(adapter->msix_entries[vector].vector, - handler, 0, adapter->name[vector], - adapter->q_vector[vector]); + handler, 0, q_vector->name, + q_vector); if (err) { e_err(probe, "request_irq failed for MSIX interrupt " "Error: %d\n", err); @@ -2175,9 +2361,9 @@ static int ixgbe_request_msix_irqs(struct ixgbe_adapter *adapter) } } - sprintf(adapter->name[vector], "%s:lsc", netdev->name); + sprintf(adapter->lsc_int_name, "%s:lsc", netdev->name); err = request_irq(adapter->msix_entries[vector].vector, - ixgbe_msix_lsc, 0, adapter->name[vector], netdev); + ixgbe_msix_lsc, 0, adapter->lsc_int_name, netdev); if (err) { e_err(probe, "request_irq for msix_lsc failed: %d\n", err); goto free_queue_irqs; @@ -2193,17 +2379,16 @@ free_queue_irqs: pci_disable_msix(adapter->pdev); kfree(adapter->msix_entries); adapter->msix_entries = NULL; -out: return err; } static void ixgbe_set_itr(struct ixgbe_adapter *adapter) { struct ixgbe_q_vector *q_vector = adapter->q_vector[0]; - u8 current_itr; - u32 new_itr = q_vector->eitr; struct ixgbe_ring *rx_ring = adapter->rx_ring[0]; struct ixgbe_ring *tx_ring = adapter->tx_ring[0]; + u32 new_itr = q_vector->eitr; + u8 current_itr; q_vector->tx_itr = ixgbe_update_itr(adapter, new_itr, q_vector->tx_itr, @@ -2233,9 +2418,9 @@ static void ixgbe_set_itr(struct ixgbe_adapter *adapter) if (new_itr != q_vector->eitr) { /* do an exponential smoothing */ - new_itr = ((q_vector->eitr * 90)/100) + ((new_itr * 10)/100); + new_itr = ((q_vector->eitr * 9) + new_itr)/10; - /* save the algorithm value here, not the smoothed one */ + /* save the algorithm value here */ q_vector->eitr = new_itr; ixgbe_write_eitr(q_vector); @@ -2256,12 +2441,17 @@ static inline void ixgbe_irq_enable(struct ixgbe_adapter *adapter, bool queues, mask |= IXGBE_EIMS_GPI_SDP0; if (adapter->flags & IXGBE_FLAG_FAN_FAIL_CAPABLE) mask |= IXGBE_EIMS_GPI_SDP1; - if (adapter->hw.mac.type == ixgbe_mac_82599EB) { + switch (adapter->hw.mac.type) { + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: mask |= IXGBE_EIMS_ECC; mask |= IXGBE_EIMS_GPI_SDP1; mask |= IXGBE_EIMS_GPI_SDP2; if (adapter->num_vfs) mask |= IXGBE_EIMS_MAILBOX; + break; + default: + break; } if (adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE || adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE) @@ -2317,13 +2507,20 @@ static irqreturn_t ixgbe_intr(int irq, void *data) if (eicr & IXGBE_EICR_LSC) ixgbe_check_lsc(adapter); - if (hw->mac.type == ixgbe_mac_82599EB) + switch (hw->mac.type) { + case ixgbe_mac_82599EB: ixgbe_check_sfp_event(adapter, eicr); + if ((adapter->flags2 & IXGBE_FLAG2_TEMP_SENSOR_CAPABLE) && + ((eicr & IXGBE_EICR_GPI_SDP0) || (eicr & IXGBE_EICR_LSC))) { + adapter->interrupt_event = eicr; + schedule_work(&adapter->check_overtemp_task); + } + break; + default: + break; + } ixgbe_check_fan_failure(adapter, eicr); - if ((adapter->flags2 & IXGBE_FLAG2_TEMP_SENSOR_CAPABLE) && - ((eicr & IXGBE_EICR_GPI_SDP0) || (eicr & IXGBE_EICR_LSC))) - schedule_work(&adapter->check_overtemp_task); if (napi_schedule_prep(&(q_vector->napi))) { adapter->tx_ring[0]->total_packets = 0; @@ -2416,14 +2613,20 @@ static void ixgbe_free_irq(struct ixgbe_adapter *adapter) **/ static inline void ixgbe_irq_disable(struct ixgbe_adapter *adapter) { - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, ~0); - } else { + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, 0xFFFF0000); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC_EX(0), ~0); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC_EX(1), ~0); if (adapter->num_vfs > 32) IXGBE_WRITE_REG(&adapter->hw, IXGBE_EITRSEL, 0); + break; + default: + break; } IXGBE_WRITE_FLUSH(&adapter->hw); if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED) { @@ -2469,7 +2672,7 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter, u64 tdba = ring->dma; int wait_loop = 10; u32 txdctl; - u16 reg_idx = ring->reg_idx; + u8 reg_idx = ring->reg_idx; /* disable queue to avoid issues while updating state */ txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(reg_idx)); @@ -2484,8 +2687,7 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter, ring->count * sizeof(union ixgbe_adv_tx_desc)); IXGBE_WRITE_REG(hw, IXGBE_TDH(reg_idx), 0); IXGBE_WRITE_REG(hw, IXGBE_TDT(reg_idx), 0); - ring->head = IXGBE_TDH(reg_idx); - ring->tail = IXGBE_TDT(reg_idx); + ring->tail = hw->hw_addr + IXGBE_TDT(reg_idx); /* configure fetching thresholds */ if (adapter->rx_itr_setting == 0) { @@ -2501,7 +2703,16 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter, } /* reinitialize flowdirector state */ - set_bit(__IXGBE_FDIR_INIT_DONE, &ring->reinit_state); + if ((adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) && + adapter->atr_sample_rate) { + ring->atr_sample_rate = adapter->atr_sample_rate; + ring->atr_count = 0; + set_bit(__IXGBE_TX_FDIR_INIT_DONE, &ring->state); + } else { + ring->atr_sample_rate = 0; + } + + clear_bit(__IXGBE_HANG_CHECK_ARMED, &ring->state); /* enable queue */ txdctl |= IXGBE_TXDCTL_ENABLE; @@ -2592,16 +2803,22 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter, struct ixgbe_ring *rx_ring) { u32 srrctl; - int index; - struct ixgbe_ring_feature *feature = adapter->ring_feature; + u8 reg_idx = rx_ring->reg_idx; - index = rx_ring->reg_idx; - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { - unsigned long mask; - mask = (unsigned long) feature[RING_F_RSS].mask; - index = index & mask; + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: { + struct ixgbe_ring_feature *feature = adapter->ring_feature; + const int mask = feature[RING_F_RSS].mask; + reg_idx = reg_idx & mask; + } + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + default: + break; } - srrctl = IXGBE_READ_REG(&adapter->hw, IXGBE_SRRCTL(index)); + + srrctl = IXGBE_READ_REG(&adapter->hw, IXGBE_SRRCTL(reg_idx)); srrctl &= ~IXGBE_SRRCTL_BSIZEHDR_MASK; srrctl &= ~IXGBE_SRRCTL_BSIZEPKT_MASK; @@ -2611,7 +2828,7 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter, srrctl |= (IXGBE_RX_HDR_SIZE << IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT) & IXGBE_SRRCTL_BSIZEHDR_MASK; - if (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED) { + if (ring_is_ps_enabled(rx_ring)) { #if (PAGE_SIZE / 2) > IXGBE_MAX_RXBUFFER srrctl |= IXGBE_MAX_RXBUFFER >> IXGBE_SRRCTL_BSIZEPKT_SHIFT; #else @@ -2624,7 +2841,7 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter, srrctl |= IXGBE_SRRCTL_DESCTYPE_ADV_ONEBUF; } - IXGBE_WRITE_REG(&adapter->hw, IXGBE_SRRCTL(index), srrctl); + IXGBE_WRITE_REG(&adapter->hw, IXGBE_SRRCTL(reg_idx), srrctl); } static void ixgbe_setup_mrqc(struct ixgbe_adapter *adapter) @@ -2694,19 +2911,36 @@ static void ixgbe_setup_mrqc(struct ixgbe_adapter *adapter) } /** + * ixgbe_clear_rscctl - disable RSC for the indicated ring + * @adapter: address of board private structure + * @ring: structure containing ring specific data + **/ +void ixgbe_clear_rscctl(struct ixgbe_adapter *adapter, + struct ixgbe_ring *ring) +{ + struct ixgbe_hw *hw = &adapter->hw; + u32 rscctrl; + u8 reg_idx = ring->reg_idx; + + rscctrl = IXGBE_READ_REG(hw, IXGBE_RSCCTL(reg_idx)); + rscctrl &= ~IXGBE_RSCCTL_RSCEN; + IXGBE_WRITE_REG(hw, IXGBE_RSCCTL(reg_idx), rscctrl); +} + +/** * ixgbe_configure_rscctl - enable RSC for the indicated ring * @adapter: address of board private structure * @index: index of ring to set **/ -static void ixgbe_configure_rscctl(struct ixgbe_adapter *adapter, +void ixgbe_configure_rscctl(struct ixgbe_adapter *adapter, struct ixgbe_ring *ring) { struct ixgbe_hw *hw = &adapter->hw; u32 rscctrl; int rx_buf_len; - u16 reg_idx = ring->reg_idx; + u8 reg_idx = ring->reg_idx; - if (!(adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)) + if (!ring_is_rsc_enabled(ring)) return; rx_buf_len = ring->rx_buf_len; @@ -2717,7 +2951,7 @@ static void ixgbe_configure_rscctl(struct ixgbe_adapter *adapter, * total size of max desc * buf_len is not greater * than 65535 */ - if (ring->flags & IXGBE_RING_RX_PS_ENABLED) { + if (ring_is_ps_enabled(ring)) { #if (MAX_SKB_FRAGS > 16) rscctrl |= IXGBE_RSCCTL_MAXDESC_16; #elif (MAX_SKB_FRAGS > 8) @@ -2770,9 +3004,9 @@ static void ixgbe_rx_desc_queue_enable(struct ixgbe_adapter *adapter, struct ixgbe_ring *ring) { struct ixgbe_hw *hw = &adapter->hw; - int reg_idx = ring->reg_idx; int wait_loop = IXGBE_MAX_RX_DESC_POLL; u32 rxdctl; + u8 reg_idx = ring->reg_idx; /* RXDCTL.EN will return 0 on 82598 if link is down, so skip it */ if (hw->mac.type == ixgbe_mac_82598EB && @@ -2796,7 +3030,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, struct ixgbe_hw *hw = &adapter->hw; u64 rdba = ring->dma; u32 rxdctl; - u16 reg_idx = ring->reg_idx; + u8 reg_idx = ring->reg_idx; /* disable queue to avoid issues while updating state */ rxdctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(reg_idx)); @@ -2810,8 +3044,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, ring->count * sizeof(union ixgbe_adv_rx_desc)); IXGBE_WRITE_REG(hw, IXGBE_RDH(reg_idx), 0); IXGBE_WRITE_REG(hw, IXGBE_RDT(reg_idx), 0); - ring->head = IXGBE_RDH(reg_idx); - ring->tail = IXGBE_RDT(reg_idx); + ring->tail = hw->hw_addr + IXGBE_RDT(reg_idx); ixgbe_configure_srrctl(adapter, ring); ixgbe_configure_rscctl(adapter, ring); @@ -2833,7 +3066,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(reg_idx), rxdctl); ixgbe_rx_desc_queue_enable(adapter, ring); - ixgbe_alloc_rx_buffers(adapter, ring, IXGBE_DESC_UNUSED(ring)); + ixgbe_alloc_rx_buffers(ring, IXGBE_DESC_UNUSED(ring)); } static void ixgbe_setup_psrtype(struct ixgbe_adapter *adapter) @@ -2899,6 +3132,9 @@ static void ixgbe_configure_virtualization(struct ixgbe_adapter *adapter) /* enable Tx loopback for VF/PF communication */ IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, IXGBE_PFDTXGSWC_VT_LBEN); + /* Enable MAC Anti-Spoofing */ + hw->mac.ops.set_mac_anti_spoofing(hw, (adapter->num_vfs != 0), + adapter->num_vfs); } static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter) @@ -2956,24 +3192,32 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter) rx_ring->rx_buf_len = rx_buf_len; if (adapter->flags & IXGBE_FLAG_RX_PS_ENABLED) - rx_ring->flags |= IXGBE_RING_RX_PS_ENABLED; + set_ring_ps_enabled(rx_ring); else - rx_ring->flags &= ~IXGBE_RING_RX_PS_ENABLED; + clear_ring_ps_enabled(rx_ring); + + if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) + set_ring_rsc_enabled(rx_ring); + else + clear_ring_rsc_enabled(rx_ring); #ifdef IXGBE_FCOE if (netdev->features & NETIF_F_FCOE_MTU) { struct ixgbe_ring_feature *f; f = &adapter->ring_feature[RING_F_FCOE]; if ((i >= f->mask) && (i < f->mask + f->indices)) { - rx_ring->flags &= ~IXGBE_RING_RX_PS_ENABLED; + clear_ring_ps_enabled(rx_ring); if (rx_buf_len < IXGBE_FCOE_JUMBO_FRAME_SIZE) rx_ring->rx_buf_len = IXGBE_FCOE_JUMBO_FRAME_SIZE; + } else if (!ring_is_rsc_enabled(rx_ring) && + !ring_is_ps_enabled(rx_ring)) { + rx_ring->rx_buf_len = + IXGBE_FCOE_JUMBO_FRAME_SIZE; } } #endif /* IXGBE_FCOE */ } - } static void ixgbe_setup_rdrxctl(struct ixgbe_adapter *adapter) @@ -2996,6 +3240,7 @@ static void ixgbe_setup_rdrxctl(struct ixgbe_adapter *adapter) rdrxctl |= IXGBE_RDRXCTL_MVMEN; break; case ixgbe_mac_82599EB: + case ixgbe_mac_X540: /* Disable RSC for ACK packets */ IXGBE_WRITE_REG(hw, IXGBE_RSCDBU, (IXGBE_RSCDBU_RSCACKDIS | IXGBE_READ_REG(hw, IXGBE_RSCDBU))); @@ -3123,6 +3368,7 @@ static void ixgbe_vlan_strip_disable(struct ixgbe_adapter *adapter) IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl); break; case ixgbe_mac_82599EB: + case ixgbe_mac_X540: for (i = 0; i < adapter->num_rx_queues; i++) { j = adapter->rx_ring[i]->reg_idx; vlnctrl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(j)); @@ -3152,6 +3398,7 @@ static void ixgbe_vlan_strip_enable(struct ixgbe_adapter *adapter) IXGBE_WRITE_REG(hw, IXGBE_VLNCTRL, vlnctrl); break; case ixgbe_mac_82599EB: + case ixgbe_mac_X540: for (i = 0; i < adapter->num_rx_queues; i++) { j = adapter->rx_ring[i]->reg_idx; vlnctrl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(j)); @@ -3349,8 +3596,6 @@ static void ixgbe_configure_dcb(struct ixgbe_adapter *adapter) { struct ixgbe_hw *hw = &adapter->hw; int max_frame = adapter->netdev->mtu + ETH_HLEN + ETH_FCS_LEN; - u32 txdctl; - int i, j; if (!(adapter->flags & IXGBE_FLAG_DCB_ENABLED)) { if (hw->mac.type == ixgbe_mac_82598EB) @@ -3366,25 +3611,18 @@ static void ixgbe_configure_dcb(struct ixgbe_adapter *adapter) max_frame = max(max_frame, IXGBE_FCOE_JUMBO_FRAME_SIZE); #endif - ixgbe_dcb_calculate_tc_credits(&adapter->dcb_cfg, max_frame, + ixgbe_dcb_calculate_tc_credits(hw, &adapter->dcb_cfg, max_frame, DCB_TX_CONFIG); - ixgbe_dcb_calculate_tc_credits(&adapter->dcb_cfg, max_frame, + ixgbe_dcb_calculate_tc_credits(hw, &adapter->dcb_cfg, max_frame, DCB_RX_CONFIG); - /* reconfigure the hardware */ - ixgbe_dcb_hw_config(&adapter->hw, &adapter->dcb_cfg); - - for (i = 0; i < adapter->num_tx_queues; i++) { - j = adapter->tx_ring[i]->reg_idx; - txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(j)); - /* PThresh workaround for Tx hang with DFP enabled. */ - txdctl |= 32; - IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(j), txdctl); - } /* Enable VLAN tag insert/strip */ adapter->netdev->features |= NETIF_F_HW_VLAN_RX; hw->mac.ops.set_vfta(&adapter->hw, 0, 0, true); + + /* reconfigure the hardware */ + ixgbe_dcb_hw_config(hw, &adapter->dcb_cfg); } #endif @@ -3516,8 +3754,9 @@ static void ixgbe_setup_gpie(struct ixgbe_adapter *adapter) case ixgbe_mac_82598EB: IXGBE_WRITE_REG(hw, IXGBE_EIAM, IXGBE_EICS_RTX_QUEUE); break; - default: case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + default: IXGBE_WRITE_REG(hw, IXGBE_EIAM_EX(0), 0xFFFFFFFF); IXGBE_WRITE_REG(hw, IXGBE_EIAM_EX(1), 0xFFFFFFFF); break; @@ -3561,13 +3800,24 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter) else ixgbe_configure_msi_and_legacy(adapter); - /* enable the optics */ - if (hw->phy.multispeed_fiber) + /* enable the optics for both mult-speed fiber and 82599 SFP+ fiber */ + if (hw->mac.ops.enable_tx_laser && + ((hw->phy.multispeed_fiber) || + ((hw->mac.ops.get_media_type(hw) == ixgbe_media_type_fiber) && + (hw->mac.type == ixgbe_mac_82599EB)))) hw->mac.ops.enable_tx_laser(hw); clear_bit(__IXGBE_DOWN, &adapter->state); ixgbe_napi_enable_all(adapter); + if (ixgbe_is_sfp(hw)) { + ixgbe_sfp_link_config(adapter); + } else { + err = ixgbe_non_sfp_link_config(hw); + if (err) + e_err(probe, "link_config FAILED %d\n", err); + } + /* clear any pending interrupts, may auto mask */ IXGBE_READ_REG(hw, IXGBE_EICR); ixgbe_irq_enable(adapter, true, true); @@ -3590,26 +3840,8 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter) * If we're not hot-pluggable SFP+, we just need to configure link * and bring it up. */ - if (hw->phy.type == ixgbe_phy_unknown) { - err = hw->phy.ops.identify(hw); - if (err == IXGBE_ERR_SFP_NOT_SUPPORTED) { - /* - * Take the device down and schedule the sfp tasklet - * which will unregister_netdev and log it. - */ - ixgbe_down(adapter); - schedule_work(&adapter->sfp_config_module_task); - return err; - } - } - - if (ixgbe_is_sfp(hw)) { - ixgbe_sfp_link_config(adapter); - } else { - err = ixgbe_non_sfp_link_config(hw); - if (err) - e_err(probe, "link_config FAILED %d\n", err); - } + if (hw->phy.type == ixgbe_phy_unknown) + schedule_work(&adapter->sfp_config_module_task); /* enable transmits */ netif_tx_start_all_queues(adapter->netdev); @@ -3687,15 +3919,13 @@ void ixgbe_reset(struct ixgbe_adapter *adapter) /** * ixgbe_clean_rx_ring - Free Rx Buffers per Queue - * @adapter: board private structure * @rx_ring: ring to free buffers from **/ -static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter, - struct ixgbe_ring *rx_ring) +static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring) { - struct pci_dev *pdev = adapter->pdev; + struct device *dev = rx_ring->dev; unsigned long size; - unsigned int i; + u16 i; /* ring already cleared, nothing to do */ if (!rx_ring->rx_buffer_info) @@ -3707,7 +3937,7 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter, rx_buffer_info = &rx_ring->rx_buffer_info[i]; if (rx_buffer_info->dma) { - dma_unmap_single(&pdev->dev, rx_buffer_info->dma, + dma_unmap_single(rx_ring->dev, rx_buffer_info->dma, rx_ring->rx_buf_len, DMA_FROM_DEVICE); rx_buffer_info->dma = 0; @@ -3718,7 +3948,7 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter, do { struct sk_buff *this = skb; if (IXGBE_RSC_CB(this)->delay_unmap) { - dma_unmap_single(&pdev->dev, + dma_unmap_single(dev, IXGBE_RSC_CB(this)->dma, rx_ring->rx_buf_len, DMA_FROM_DEVICE); @@ -3732,7 +3962,7 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter, if (!rx_buffer_info->page) continue; if (rx_buffer_info->page_dma) { - dma_unmap_page(&pdev->dev, rx_buffer_info->page_dma, + dma_unmap_page(dev, rx_buffer_info->page_dma, PAGE_SIZE / 2, DMA_FROM_DEVICE); rx_buffer_info->page_dma = 0; } @@ -3749,24 +3979,17 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter, rx_ring->next_to_clean = 0; rx_ring->next_to_use = 0; - - if (rx_ring->head) - writel(0, adapter->hw.hw_addr + rx_ring->head); - if (rx_ring->tail) - writel(0, adapter->hw.hw_addr + rx_ring->tail); } /** * ixgbe_clean_tx_ring - Free Tx Buffers - * @adapter: board private structure * @tx_ring: ring to be cleaned **/ -static void ixgbe_clean_tx_ring(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring) +static void ixgbe_clean_tx_ring(struct ixgbe_ring *tx_ring) { struct ixgbe_tx_buffer *tx_buffer_info; unsigned long size; - unsigned int i; + u16 i; /* ring already cleared, nothing to do */ if (!tx_ring->tx_buffer_info) @@ -3775,7 +3998,7 @@ static void ixgbe_clean_tx_ring(struct ixgbe_adapter *adapter, /* Free all the Tx ring sk_buffs */ for (i = 0; i < tx_ring->count; i++) { tx_buffer_info = &tx_ring->tx_buffer_info[i]; - ixgbe_unmap_and_free_tx_resource(adapter, tx_buffer_info); + ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer_info); } size = sizeof(struct ixgbe_tx_buffer) * tx_ring->count; @@ -3786,11 +4009,6 @@ static void ixgbe_clean_tx_ring(struct ixgbe_adapter *adapter, tx_ring->next_to_use = 0; tx_ring->next_to_clean = 0; - - if (tx_ring->head) - writel(0, adapter->hw.hw_addr + tx_ring->head); - if (tx_ring->tail) - writel(0, adapter->hw.hw_addr + tx_ring->tail); } /** @@ -3802,7 +4020,7 @@ static void ixgbe_clean_all_rx_rings(struct ixgbe_adapter *adapter) int i; for (i = 0; i < adapter->num_rx_queues; i++) - ixgbe_clean_rx_ring(adapter, adapter->rx_ring[i]); + ixgbe_clean_rx_ring(adapter->rx_ring[i]); } /** @@ -3814,7 +4032,7 @@ static void ixgbe_clean_all_tx_rings(struct ixgbe_adapter *adapter) int i; for (i = 0; i < adapter->num_tx_queues; i++) - ixgbe_clean_tx_ring(adapter, adapter->tx_ring[i]); + ixgbe_clean_tx_ring(adapter->tx_ring[i]); } void ixgbe_down(struct ixgbe_adapter *adapter) @@ -3823,7 +4041,7 @@ void ixgbe_down(struct ixgbe_adapter *adapter) struct ixgbe_hw *hw = &adapter->hw; u32 rxctrl; u32 txdctl; - int i, j; + int i; int num_q_vectors = adapter->num_msix_vectors - NON_Q_VECTORS; /* signal that we are down to the interrupt handler */ @@ -3881,26 +4099,36 @@ void ixgbe_down(struct ixgbe_adapter *adapter) /* disable transmits in the hardware now that interrupts are off */ for (i = 0; i < adapter->num_tx_queues; i++) { - j = adapter->tx_ring[i]->reg_idx; - txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(j)); - IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(j), + u8 reg_idx = adapter->tx_ring[i]->reg_idx; + txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(reg_idx)); + IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(reg_idx), (txdctl & ~IXGBE_TXDCTL_ENABLE)); } /* Disable the Tx DMA engine on 82599 */ - if (hw->mac.type == ixgbe_mac_82599EB) + switch (hw->mac.type) { + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: IXGBE_WRITE_REG(hw, IXGBE_DMATXCTL, (IXGBE_READ_REG(hw, IXGBE_DMATXCTL) & ~IXGBE_DMATXCTL_TE)); - - /* power down the optics */ - if (hw->phy.multispeed_fiber) - hw->mac.ops.disable_tx_laser(hw); + break; + default: + break; + } /* clear n-tuple filters that are cached */ ethtool_ntuple_flush(netdev); if (!pci_channel_offline(adapter->pdev)) ixgbe_reset(adapter); + + /* power down the optics for multispeed fiber and 82599 SFP+ fiber */ + if (hw->mac.ops.disable_tx_laser && + ((hw->phy.multispeed_fiber) || + ((hw->mac.ops.get_media_type(hw) == ixgbe_media_type_fiber) && + (hw->mac.type == ixgbe_mac_82599EB)))) + hw->mac.ops.disable_tx_laser(hw); + ixgbe_clean_all_tx_rings(adapter); ixgbe_clean_all_rx_rings(adapter); @@ -3925,10 +4153,8 @@ static int ixgbe_poll(struct napi_struct *napi, int budget) int tx_clean_complete, work_done = 0; #ifdef CONFIG_IXGBE_DCA - if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) { - ixgbe_update_tx_dca(adapter, adapter->tx_ring[0]); - ixgbe_update_rx_dca(adapter, adapter->rx_ring[0]); - } + if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) + ixgbe_update_dca(q_vector); #endif tx_clean_complete = ixgbe_clean_tx_irq(q_vector, adapter->tx_ring[0]); @@ -3956,6 +4182,8 @@ static void ixgbe_tx_timeout(struct net_device *netdev) { struct ixgbe_adapter *adapter = netdev_priv(netdev); + adapter->tx_timeout_count++; + /* Do the reset outside of interrupt context */ schedule_work(&adapter->reset_task); } @@ -3970,8 +4198,6 @@ static void ixgbe_reset_task(struct work_struct *work) test_bit(__IXGBE_RESETTING, &adapter->state)) return; - adapter->tx_timeout_count++; - ixgbe_dump(adapter); netdev_err(adapter->netdev, "Reset adapter\n"); ixgbe_reinit_locked(adapter); @@ -4221,19 +4447,16 @@ static void ixgbe_acquire_msix_vectors(struct ixgbe_adapter *adapter, static inline bool ixgbe_cache_ring_rss(struct ixgbe_adapter *adapter) { int i; - bool ret = false; - if (adapter->flags & IXGBE_FLAG_RSS_ENABLED) { - for (i = 0; i < adapter->num_rx_queues; i++) - adapter->rx_ring[i]->reg_idx = i; - for (i = 0; i < adapter->num_tx_queues; i++) - adapter->tx_ring[i]->reg_idx = i; - ret = true; - } else { - ret = false; - } + if (!(adapter->flags & IXGBE_FLAG_RSS_ENABLED)) + return false; - return ret; + for (i = 0; i < adapter->num_rx_queues; i++) + adapter->rx_ring[i]->reg_idx = i; + for (i = 0; i < adapter->num_tx_queues; i++) + adapter->tx_ring[i]->reg_idx = i; + + return true; } #ifdef CONFIG_IXGBE_DCB @@ -4250,71 +4473,67 @@ static inline bool ixgbe_cache_ring_dcb(struct ixgbe_adapter *adapter) bool ret = false; int dcb_i = adapter->ring_feature[RING_F_DCB].indices; - if (adapter->flags & IXGBE_FLAG_DCB_ENABLED) { - if (adapter->hw.mac.type == ixgbe_mac_82598EB) { - /* the number of queues is assumed to be symmetric */ - for (i = 0; i < dcb_i; i++) { - adapter->rx_ring[i]->reg_idx = i << 3; - adapter->tx_ring[i]->reg_idx = i << 2; - } - ret = true; - } else if (adapter->hw.mac.type == ixgbe_mac_82599EB) { - if (dcb_i == 8) { - /* - * Tx TC0 starts at: descriptor queue 0 - * Tx TC1 starts at: descriptor queue 32 - * Tx TC2 starts at: descriptor queue 64 - * Tx TC3 starts at: descriptor queue 80 - * Tx TC4 starts at: descriptor queue 96 - * Tx TC5 starts at: descriptor queue 104 - * Tx TC6 starts at: descriptor queue 112 - * Tx TC7 starts at: descriptor queue 120 - * - * Rx TC0-TC7 are offset by 16 queues each - */ - for (i = 0; i < 3; i++) { - adapter->tx_ring[i]->reg_idx = i << 5; - adapter->rx_ring[i]->reg_idx = i << 4; - } - for ( ; i < 5; i++) { - adapter->tx_ring[i]->reg_idx = - ((i + 2) << 4); - adapter->rx_ring[i]->reg_idx = i << 4; - } - for ( ; i < dcb_i; i++) { - adapter->tx_ring[i]->reg_idx = - ((i + 8) << 3); - adapter->rx_ring[i]->reg_idx = i << 4; - } + if (!(adapter->flags & IXGBE_FLAG_DCB_ENABLED)) + return false; - ret = true; - } else if (dcb_i == 4) { - /* - * Tx TC0 starts at: descriptor queue 0 - * Tx TC1 starts at: descriptor queue 64 - * Tx TC2 starts at: descriptor queue 96 - * Tx TC3 starts at: descriptor queue 112 - * - * Rx TC0-TC3 are offset by 32 queues each - */ - adapter->tx_ring[0]->reg_idx = 0; - adapter->tx_ring[1]->reg_idx = 64; - adapter->tx_ring[2]->reg_idx = 96; - adapter->tx_ring[3]->reg_idx = 112; - for (i = 0 ; i < dcb_i; i++) - adapter->rx_ring[i]->reg_idx = i << 5; - - ret = true; - } else { - ret = false; + /* the number of queues is assumed to be symmetric */ + switch (adapter->hw.mac.type) { + case ixgbe_mac_82598EB: + for (i = 0; i < dcb_i; i++) { + adapter->rx_ring[i]->reg_idx = i << 3; + adapter->tx_ring[i]->reg_idx = i << 2; + } + ret = true; + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + if (dcb_i == 8) { + /* + * Tx TC0 starts at: descriptor queue 0 + * Tx TC1 starts at: descriptor queue 32 + * Tx TC2 starts at: descriptor queue 64 + * Tx TC3 starts at: descriptor queue 80 + * Tx TC4 starts at: descriptor queue 96 + * Tx TC5 starts at: descriptor queue 104 + * Tx TC6 starts at: descriptor queue 112 + * Tx TC7 starts at: descriptor queue 120 + * + * Rx TC0-TC7 are offset by 16 queues each + */ + for (i = 0; i < 3; i++) { + adapter->tx_ring[i]->reg_idx = i << 5; + adapter->rx_ring[i]->reg_idx = i << 4; } - } else { - ret = false; + for ( ; i < 5; i++) { + adapter->tx_ring[i]->reg_idx = ((i + 2) << 4); + adapter->rx_ring[i]->reg_idx = i << 4; + } + for ( ; i < dcb_i; i++) { + adapter->tx_ring[i]->reg_idx = ((i + 8) << 3); + adapter->rx_ring[i]->reg_idx = i << 4; + } + ret = true; + } else if (dcb_i == 4) { + /* + * Tx TC0 starts at: descriptor queue 0 + * Tx TC1 starts at: descriptor queue 64 + * Tx TC2 starts at: descriptor queue 96 + * Tx TC3 starts at: descriptor queue 112 + * + * Rx TC0-TC3 are offset by 32 queues each + */ + adapter->tx_ring[0]->reg_idx = 0; + adapter->tx_ring[1]->reg_idx = 64; + adapter->tx_ring[2]->reg_idx = 96; + adapter->tx_ring[3]->reg_idx = 112; + for (i = 0 ; i < dcb_i; i++) + adapter->rx_ring[i]->reg_idx = i << 5; + ret = true; } - } else { - ret = false; + break; + default: + break; } - return ret; } #endif @@ -4354,55 +4573,55 @@ static inline bool ixgbe_cache_ring_fdir(struct ixgbe_adapter *adapter) */ static inline bool ixgbe_cache_ring_fcoe(struct ixgbe_adapter *adapter) { - int i, fcoe_rx_i = 0, fcoe_tx_i = 0; - bool ret = false; struct ixgbe_ring_feature *f = &adapter->ring_feature[RING_F_FCOE]; + int i; + u8 fcoe_rx_i = 0, fcoe_tx_i = 0; + + if (!(adapter->flags & IXGBE_FLAG_FCOE_ENABLED)) + return false; - if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) { #ifdef CONFIG_IXGBE_DCB - if (adapter->flags & IXGBE_FLAG_DCB_ENABLED) { - struct ixgbe_fcoe *fcoe = &adapter->fcoe; + if (adapter->flags & IXGBE_FLAG_DCB_ENABLED) { + struct ixgbe_fcoe *fcoe = &adapter->fcoe; - ixgbe_cache_ring_dcb(adapter); - /* find out queues in TC for FCoE */ - fcoe_rx_i = adapter->rx_ring[fcoe->tc]->reg_idx + 1; - fcoe_tx_i = adapter->tx_ring[fcoe->tc]->reg_idx + 1; - /* - * In 82599, the number of Tx queues for each traffic - * class for both 8-TC and 4-TC modes are: - * TCs : TC0 TC1 TC2 TC3 TC4 TC5 TC6 TC7 - * 8 TCs: 32 32 16 16 8 8 8 8 - * 4 TCs: 64 64 32 32 - * We have max 8 queues for FCoE, where 8 the is - * FCoE redirection table size. If TC for FCoE is - * less than or equal to TC3, we have enough queues - * to add max of 8 queues for FCoE, so we start FCoE - * tx descriptor from the next one, i.e., reg_idx + 1. - * If TC for FCoE is above TC3, implying 8 TC mode, - * and we need 8 for FCoE, we have to take all queues - * in that traffic class for FCoE. - */ - if ((f->indices == IXGBE_FCRETA_SIZE) && (fcoe->tc > 3)) - fcoe_tx_i--; - } + ixgbe_cache_ring_dcb(adapter); + /* find out queues in TC for FCoE */ + fcoe_rx_i = adapter->rx_ring[fcoe->tc]->reg_idx + 1; + fcoe_tx_i = adapter->tx_ring[fcoe->tc]->reg_idx + 1; + /* + * In 82599, the number of Tx queues for each traffic + * class for both 8-TC and 4-TC modes are: + * TCs : TC0 TC1 TC2 TC3 TC4 TC5 TC6 TC7 + * 8 TCs: 32 32 16 16 8 8 8 8 + * 4 TCs: 64 64 32 32 + * We have max 8 queues for FCoE, where 8 the is + * FCoE redirection table size. If TC for FCoE is + * less than or equal to TC3, we have enough queues + * to add max of 8 queues for FCoE, so we start FCoE + * Tx queue from the next one, i.e., reg_idx + 1. + * If TC for FCoE is above TC3, implying 8 TC mode, + * and we need 8 for FCoE, we have to take all queues + * in that traffic class for FCoE. + */ + if ((f->indices == IXGBE_FCRETA_SIZE) && (fcoe->tc > 3)) + fcoe_tx_i--; + } #endif /* CONFIG_IXGBE_DCB */ - if (adapter->flags & IXGBE_FLAG_RSS_ENABLED) { - if ((adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) || - (adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE)) - ixgbe_cache_ring_fdir(adapter); - else - ixgbe_cache_ring_rss(adapter); + if (adapter->flags & IXGBE_FLAG_RSS_ENABLED) { + if ((adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) || + (adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE)) + ixgbe_cache_ring_fdir(adapter); + else + ixgbe_cache_ring_rss(adapter); - fcoe_rx_i = f->mask; - fcoe_tx_i = f->mask; - } - for (i = 0; i < f->indices; i++, fcoe_rx_i++, fcoe_tx_i++) { - adapter->rx_ring[f->mask + i]->reg_idx = fcoe_rx_i; - adapter->tx_ring[f->mask + i]->reg_idx = fcoe_tx_i; - } - ret = true; + fcoe_rx_i = f->mask; + fcoe_tx_i = f->mask; } - return ret; + for (i = 0; i < f->indices; i++, fcoe_rx_i++, fcoe_tx_i++) { + adapter->rx_ring[f->mask + i]->reg_idx = fcoe_rx_i; + adapter->tx_ring[f->mask + i]->reg_idx = fcoe_tx_i; + } + return true; } #endif /* IXGBE_FCOE */ @@ -4471,65 +4690,55 @@ static void ixgbe_cache_ring_register(struct ixgbe_adapter *adapter) **/ static int ixgbe_alloc_queues(struct ixgbe_adapter *adapter) { - int i; - int orig_node = adapter->node; + int rx = 0, tx = 0, nid = adapter->node; - for (i = 0; i < adapter->num_tx_queues; i++) { - struct ixgbe_ring *ring = adapter->tx_ring[i]; - if (orig_node == -1) { - int cur_node = next_online_node(adapter->node); - if (cur_node == MAX_NUMNODES) - cur_node = first_online_node; - adapter->node = cur_node; - } - ring = kzalloc_node(sizeof(struct ixgbe_ring), GFP_KERNEL, - adapter->node); + if (nid < 0 || !node_online(nid)) + nid = first_online_node; + + for (; tx < adapter->num_tx_queues; tx++) { + struct ixgbe_ring *ring; + + ring = kzalloc_node(sizeof(*ring), GFP_KERNEL, nid); if (!ring) - ring = kzalloc(sizeof(struct ixgbe_ring), GFP_KERNEL); + ring = kzalloc(sizeof(*ring), GFP_KERNEL); if (!ring) - goto err_tx_ring_allocation; + goto err_allocation; ring->count = adapter->tx_ring_count; - ring->queue_index = i; - ring->numa_node = adapter->node; + ring->queue_index = tx; + ring->numa_node = nid; + ring->dev = &adapter->pdev->dev; + ring->netdev = adapter->netdev; - adapter->tx_ring[i] = ring; + adapter->tx_ring[tx] = ring; } - /* Restore the adapter's original node */ - adapter->node = orig_node; + for (; rx < adapter->num_rx_queues; rx++) { + struct ixgbe_ring *ring; - for (i = 0; i < adapter->num_rx_queues; i++) { - struct ixgbe_ring *ring = adapter->rx_ring[i]; - if (orig_node == -1) { - int cur_node = next_online_node(adapter->node); - if (cur_node == MAX_NUMNODES) - cur_node = first_online_node; - adapter->node = cur_node; - } - ring = kzalloc_node(sizeof(struct ixgbe_ring), GFP_KERNEL, - adapter->node); + ring = kzalloc_node(sizeof(*ring), GFP_KERNEL, nid); if (!ring) - ring = kzalloc(sizeof(struct ixgbe_ring), GFP_KERNEL); + ring = kzalloc(sizeof(*ring), GFP_KERNEL); if (!ring) - goto err_rx_ring_allocation; + goto err_allocation; ring->count = adapter->rx_ring_count; - ring->queue_index = i; - ring->numa_node = adapter->node; + ring->queue_index = rx; + ring->numa_node = nid; + ring->dev = &adapter->pdev->dev; + ring->netdev = adapter->netdev; - adapter->rx_ring[i] = ring; + adapter->rx_ring[rx] = ring; } - /* Restore the adapter's original node */ - adapter->node = orig_node; - ixgbe_cache_ring_register(adapter); return 0; -err_rx_ring_allocation: - for (i = 0; i < adapter->num_tx_queues; i++) - kfree(adapter->tx_ring[i]); -err_tx_ring_allocation: +err_allocation: + while (tx) + kfree(adapter->tx_ring[--tx]); + + while (rx) + kfree(adapter->rx_ring[--rx]); return -ENOMEM; } @@ -4751,6 +4960,11 @@ err_set_interrupt: return err; } +static void ring_free_rcu(struct rcu_head *head) +{ + kfree(container_of(head, struct ixgbe_ring, rcu)); +} + /** * ixgbe_clear_interrupt_scheme - Clear the current interrupt scheme settings * @adapter: board private structure to clear interrupt scheme on @@ -4767,7 +4981,12 @@ void ixgbe_clear_interrupt_scheme(struct ixgbe_adapter *adapter) adapter->tx_ring[i] = NULL; } for (i = 0; i < adapter->num_rx_queues; i++) { - kfree(adapter->rx_ring[i]); + struct ixgbe_ring *ring = adapter->rx_ring[i]; + + /* ixgbe_get_stats64() might access this ring, we must wait + * a grace period before freeing it. + */ + call_rcu(&ring->rcu, ring_free_rcu); adapter->rx_ring[i] = NULL; } @@ -4847,6 +5066,7 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter) int j; struct tc_configuration *tc; #endif + int max_frame = dev->mtu + ETH_HLEN + ETH_FCS_LEN; /* PCI config space info */ @@ -4861,11 +5081,14 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter) adapter->ring_feature[RING_F_RSS].indices = rss; adapter->flags |= IXGBE_FLAG_RSS_ENABLED; adapter->ring_feature[RING_F_DCB].indices = IXGBE_MAX_DCB_INDICES; - if (hw->mac.type == ixgbe_mac_82598EB) { + switch (hw->mac.type) { + case ixgbe_mac_82598EB: if (hw->device_id == IXGBE_DEV_ID_82598AT) adapter->flags |= IXGBE_FLAG_FAN_FAIL_CAPABLE; adapter->max_msix_q_vectors = MAX_MSIX_Q_VECTORS_82598; - } else if (hw->mac.type == ixgbe_mac_82599EB) { + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: adapter->max_msix_q_vectors = MAX_MSIX_Q_VECTORS_82599; adapter->flags2 |= IXGBE_FLAG2_RSC_CAPABLE; adapter->flags2 |= IXGBE_FLAG2_RSC_ENABLED; @@ -4894,6 +5117,9 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter) adapter->fcoe.up = IXGBE_FCOE_DEFTC; #endif #endif /* IXGBE_FCOE */ + break; + default: + break; } #ifdef CONFIG_IXGBE_DCB @@ -4923,8 +5149,8 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter) #ifdef CONFIG_DCB adapter->last_lfc_mode = hw->fc.current_mode; #endif - hw->fc.high_water = IXGBE_DEFAULT_FCRTH; - hw->fc.low_water = IXGBE_DEFAULT_FCRTL; + hw->fc.high_water = FC_HIGH_WATER(max_frame); + hw->fc.low_water = FC_LOW_WATER(max_frame); hw->fc.pause_time = IXGBE_DEFAULT_FCPAUSE; hw->fc.send_xon = true; hw->fc.disable_fc_autoneg = false; @@ -4962,30 +5188,27 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter) /** * ixgbe_setup_tx_resources - allocate Tx resources (Descriptors) - * @adapter: board private structure * @tx_ring: tx descriptor ring (for a specific queue) to setup * * Return 0 on success, negative on failure **/ -int ixgbe_setup_tx_resources(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring) +int ixgbe_setup_tx_resources(struct ixgbe_ring *tx_ring) { - struct pci_dev *pdev = adapter->pdev; + struct device *dev = tx_ring->dev; int size; size = sizeof(struct ixgbe_tx_buffer) * tx_ring->count; - tx_ring->tx_buffer_info = vmalloc_node(size, tx_ring->numa_node); + tx_ring->tx_buffer_info = vzalloc_node(size, tx_ring->numa_node); if (!tx_ring->tx_buffer_info) - tx_ring->tx_buffer_info = vmalloc(size); + tx_ring->tx_buffer_info = vzalloc(size); if (!tx_ring->tx_buffer_info) goto err; - memset(tx_ring->tx_buffer_info, 0, size); /* round up to nearest 4K */ tx_ring->size = tx_ring->count * sizeof(union ixgbe_adv_tx_desc); tx_ring->size = ALIGN(tx_ring->size, 4096); - tx_ring->desc = dma_alloc_coherent(&pdev->dev, tx_ring->size, + tx_ring->desc = dma_alloc_coherent(dev, tx_ring->size, &tx_ring->dma, GFP_KERNEL); if (!tx_ring->desc) goto err; @@ -4998,7 +5221,7 @@ int ixgbe_setup_tx_resources(struct ixgbe_adapter *adapter, err: vfree(tx_ring->tx_buffer_info); tx_ring->tx_buffer_info = NULL; - e_err(probe, "Unable to allocate memory for the Tx descriptor ring\n"); + dev_err(dev, "Unable to allocate memory for the Tx descriptor ring\n"); return -ENOMEM; } @@ -5017,7 +5240,7 @@ static int ixgbe_setup_all_tx_resources(struct ixgbe_adapter *adapter) int i, err = 0; for (i = 0; i < adapter->num_tx_queues; i++) { - err = ixgbe_setup_tx_resources(adapter, adapter->tx_ring[i]); + err = ixgbe_setup_tx_resources(adapter->tx_ring[i]); if (!err) continue; e_err(probe, "Allocation for Tx Queue %u failed\n", i); @@ -5029,48 +5252,40 @@ static int ixgbe_setup_all_tx_resources(struct ixgbe_adapter *adapter) /** * ixgbe_setup_rx_resources - allocate Rx resources (Descriptors) - * @adapter: board private structure * @rx_ring: rx descriptor ring (for a specific queue) to setup * * Returns 0 on success, negative on failure **/ -int ixgbe_setup_rx_resources(struct ixgbe_adapter *adapter, - struct ixgbe_ring *rx_ring) +int ixgbe_setup_rx_resources(struct ixgbe_ring *rx_ring) { - struct pci_dev *pdev = adapter->pdev; + struct device *dev = rx_ring->dev; int size; size = sizeof(struct ixgbe_rx_buffer) * rx_ring->count; - rx_ring->rx_buffer_info = vmalloc_node(size, adapter->node); + rx_ring->rx_buffer_info = vzalloc_node(size, rx_ring->numa_node); if (!rx_ring->rx_buffer_info) - rx_ring->rx_buffer_info = vmalloc(size); - if (!rx_ring->rx_buffer_info) { - e_err(probe, "vmalloc allocation failed for the Rx " - "descriptor ring\n"); - goto alloc_failed; - } - memset(rx_ring->rx_buffer_info, 0, size); + rx_ring->rx_buffer_info = vzalloc(size); + if (!rx_ring->rx_buffer_info) + goto err; /* Round up to nearest 4K */ rx_ring->size = rx_ring->count * sizeof(union ixgbe_adv_rx_desc); rx_ring->size = ALIGN(rx_ring->size, 4096); - rx_ring->desc = dma_alloc_coherent(&pdev->dev, rx_ring->size, + rx_ring->desc = dma_alloc_coherent(dev, rx_ring->size, &rx_ring->dma, GFP_KERNEL); - if (!rx_ring->desc) { - e_err(probe, "Memory allocation failed for the Rx " - "descriptor ring\n"); - vfree(rx_ring->rx_buffer_info); - goto alloc_failed; - } + if (!rx_ring->desc) + goto err; rx_ring->next_to_clean = 0; rx_ring->next_to_use = 0; return 0; - -alloc_failed: +err: + vfree(rx_ring->rx_buffer_info); + rx_ring->rx_buffer_info = NULL; + dev_err(dev, "Unable to allocate memory for the Rx descriptor ring\n"); return -ENOMEM; } @@ -5084,13 +5299,12 @@ alloc_failed: * * Return 0 on success, negative on failure **/ - static int ixgbe_setup_all_rx_resources(struct ixgbe_adapter *adapter) { int i, err = 0; for (i = 0; i < adapter->num_rx_queues; i++) { - err = ixgbe_setup_rx_resources(adapter, adapter->rx_ring[i]); + err = ixgbe_setup_rx_resources(adapter->rx_ring[i]); if (!err) continue; e_err(probe, "Allocation for Rx Queue %u failed\n", i); @@ -5102,23 +5316,23 @@ static int ixgbe_setup_all_rx_resources(struct ixgbe_adapter *adapter) /** * ixgbe_free_tx_resources - Free Tx Resources per Queue - * @adapter: board private structure * @tx_ring: Tx descriptor ring for a specific queue * * Free all transmit software resources **/ -void ixgbe_free_tx_resources(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring) +void ixgbe_free_tx_resources(struct ixgbe_ring *tx_ring) { - struct pci_dev *pdev = adapter->pdev; - - ixgbe_clean_tx_ring(adapter, tx_ring); + ixgbe_clean_tx_ring(tx_ring); vfree(tx_ring->tx_buffer_info); tx_ring->tx_buffer_info = NULL; - dma_free_coherent(&pdev->dev, tx_ring->size, tx_ring->desc, - tx_ring->dma); + /* if not set, then don't free */ + if (!tx_ring->desc) + return; + + dma_free_coherent(tx_ring->dev, tx_ring->size, + tx_ring->desc, tx_ring->dma); tx_ring->desc = NULL; } @@ -5135,28 +5349,28 @@ static void ixgbe_free_all_tx_resources(struct ixgbe_adapter *adapter) for (i = 0; i < adapter->num_tx_queues; i++) if (adapter->tx_ring[i]->desc) - ixgbe_free_tx_resources(adapter, adapter->tx_ring[i]); + ixgbe_free_tx_resources(adapter->tx_ring[i]); } /** * ixgbe_free_rx_resources - Free Rx Resources - * @adapter: board private structure * @rx_ring: ring to clean the resources from * * Free all receive software resources **/ -void ixgbe_free_rx_resources(struct ixgbe_adapter *adapter, - struct ixgbe_ring *rx_ring) +void ixgbe_free_rx_resources(struct ixgbe_ring *rx_ring) { - struct pci_dev *pdev = adapter->pdev; - - ixgbe_clean_rx_ring(adapter, rx_ring); + ixgbe_clean_rx_ring(rx_ring); vfree(rx_ring->rx_buffer_info); rx_ring->rx_buffer_info = NULL; - dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc, - rx_ring->dma); + /* if not set, then don't free */ + if (!rx_ring->desc) + return; + + dma_free_coherent(rx_ring->dev, rx_ring->size, + rx_ring->desc, rx_ring->dma); rx_ring->desc = NULL; } @@ -5173,7 +5387,7 @@ static void ixgbe_free_all_rx_resources(struct ixgbe_adapter *adapter) for (i = 0; i < adapter->num_rx_queues; i++) if (adapter->rx_ring[i]->desc) - ixgbe_free_rx_resources(adapter, adapter->rx_ring[i]); + ixgbe_free_rx_resources(adapter->rx_ring[i]); } /** @@ -5186,6 +5400,7 @@ static void ixgbe_free_all_rx_resources(struct ixgbe_adapter *adapter) static int ixgbe_change_mtu(struct net_device *netdev, int new_mtu) { struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_hw *hw = &adapter->hw; int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN; /* MTU < 68 is an error and causes problems on some kernels */ @@ -5196,6 +5411,9 @@ static int ixgbe_change_mtu(struct net_device *netdev, int new_mtu) /* must set new MTU before calling down or up */ netdev->mtu = new_mtu; + hw->fc.high_water = FC_HIGH_WATER(max_frame); + hw->fc.low_water = FC_LOW_WATER(max_frame); + if (netif_running(netdev)) ixgbe_reinit_locked(adapter); @@ -5291,8 +5509,8 @@ static int ixgbe_close(struct net_device *netdev) #ifdef CONFIG_PM static int ixgbe_resume(struct pci_dev *pdev) { - struct net_device *netdev = pci_get_drvdata(pdev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = pci_get_drvdata(pdev); + struct net_device *netdev = adapter->netdev; u32 err; pci_set_power_state(pdev, PCI_D0); @@ -5323,7 +5541,7 @@ static int ixgbe_resume(struct pci_dev *pdev) IXGBE_WRITE_REG(&adapter->hw, IXGBE_WUS, ~0); if (netif_running(netdev)) { - err = ixgbe_open(adapter->netdev); + err = ixgbe_open(netdev); if (err) return err; } @@ -5336,8 +5554,8 @@ static int ixgbe_resume(struct pci_dev *pdev) static int __ixgbe_shutdown(struct pci_dev *pdev, bool *enable_wake) { - struct net_device *netdev = pci_get_drvdata(pdev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = pci_get_drvdata(pdev); + struct net_device *netdev = adapter->netdev; struct ixgbe_hw *hw = &adapter->hw; u32 ctrl, fctrl; u32 wufc = adapter->wol; @@ -5354,6 +5572,8 @@ static int __ixgbe_shutdown(struct pci_dev *pdev, bool *enable_wake) ixgbe_free_all_rx_resources(adapter); } + ixgbe_clear_interrupt_scheme(adapter); + #ifdef CONFIG_PM retval = pci_save_state(pdev); if (retval) @@ -5380,15 +5600,20 @@ static int __ixgbe_shutdown(struct pci_dev *pdev, bool *enable_wake) IXGBE_WRITE_REG(hw, IXGBE_WUFC, 0); } - if (wufc && hw->mac.type == ixgbe_mac_82599EB) - pci_wake_from_d3(pdev, true); - else + switch (hw->mac.type) { + case ixgbe_mac_82598EB: pci_wake_from_d3(pdev, false); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + pci_wake_from_d3(pdev, !!wufc); + break; + default: + break; + } *enable_wake = !!wufc; - ixgbe_clear_interrupt_scheme(adapter); - ixgbe_release_hw_control(adapter); pci_disable_device(pdev); @@ -5437,10 +5662,12 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) { struct net_device *netdev = adapter->netdev; struct ixgbe_hw *hw = &adapter->hw; + struct ixgbe_hw_stats *hwstats = &adapter->stats; u64 total_mpc = 0; u32 i, missed_rx = 0, mpc, bprc, lxon, lxoff, xon_off_tot; - u64 non_eop_descs = 0, restart_queue = 0; - struct ixgbe_hw_stats *hwstats = &adapter->stats; + u64 non_eop_descs = 0, restart_queue = 0, tx_busy = 0; + u64 alloc_rx_page_failed = 0, alloc_rx_buff_failed = 0; + u64 bytes = 0, packets = 0; if (test_bit(__IXGBE_DOWN, &adapter->state) || test_bit(__IXGBE_RESETTING, &adapter->state)) @@ -5453,21 +5680,41 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) adapter->hw_rx_no_dma_resources += IXGBE_READ_REG(hw, IXGBE_QPRDC(i)); for (i = 0; i < adapter->num_rx_queues; i++) { - rsc_count += adapter->rx_ring[i]->rsc_count; - rsc_flush += adapter->rx_ring[i]->rsc_flush; + rsc_count += adapter->rx_ring[i]->rx_stats.rsc_count; + rsc_flush += adapter->rx_ring[i]->rx_stats.rsc_flush; } adapter->rsc_total_count = rsc_count; adapter->rsc_total_flush = rsc_flush; } + for (i = 0; i < adapter->num_rx_queues; i++) { + struct ixgbe_ring *rx_ring = adapter->rx_ring[i]; + non_eop_descs += rx_ring->rx_stats.non_eop_descs; + alloc_rx_page_failed += rx_ring->rx_stats.alloc_rx_page_failed; + alloc_rx_buff_failed += rx_ring->rx_stats.alloc_rx_buff_failed; + bytes += rx_ring->stats.bytes; + packets += rx_ring->stats.packets; + } + adapter->non_eop_descs = non_eop_descs; + adapter->alloc_rx_page_failed = alloc_rx_page_failed; + adapter->alloc_rx_buff_failed = alloc_rx_buff_failed; + netdev->stats.rx_bytes = bytes; + netdev->stats.rx_packets = packets; + + bytes = 0; + packets = 0; /* gather some stats to the adapter struct that are per queue */ - for (i = 0; i < adapter->num_tx_queues; i++) - restart_queue += adapter->tx_ring[i]->restart_queue; + for (i = 0; i < adapter->num_tx_queues; i++) { + struct ixgbe_ring *tx_ring = adapter->tx_ring[i]; + restart_queue += tx_ring->tx_stats.restart_queue; + tx_busy += tx_ring->tx_stats.tx_busy; + bytes += tx_ring->stats.bytes; + packets += tx_ring->stats.packets; + } adapter->restart_queue = restart_queue; - - for (i = 0; i < adapter->num_rx_queues; i++) - non_eop_descs += adapter->rx_ring[i]->non_eop_descs; - adapter->non_eop_descs = non_eop_descs; + adapter->tx_busy = tx_busy; + netdev->stats.tx_bytes = bytes; + netdev->stats.tx_packets = packets; hwstats->crcerrs += IXGBE_READ_REG(hw, IXGBE_CRCERRS); for (i = 0; i < 8; i++) { @@ -5482,17 +5729,18 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) hwstats->qbtc[i] += IXGBE_READ_REG(hw, IXGBE_QBTC(i)); hwstats->qprc[i] += IXGBE_READ_REG(hw, IXGBE_QPRC(i)); hwstats->qbrc[i] += IXGBE_READ_REG(hw, IXGBE_QBRC(i)); - if (hw->mac.type == ixgbe_mac_82599EB) { - hwstats->pxonrxc[i] += - IXGBE_READ_REG(hw, IXGBE_PXONRXCNT(i)); - hwstats->pxoffrxc[i] += - IXGBE_READ_REG(hw, IXGBE_PXOFFRXCNT(i)); - hwstats->qprdc[i] += IXGBE_READ_REG(hw, IXGBE_QPRDC(i)); - } else { + switch (hw->mac.type) { + case ixgbe_mac_82598EB: hwstats->pxonrxc[i] += IXGBE_READ_REG(hw, IXGBE_PXONRXC(i)); - hwstats->pxoffrxc[i] += - IXGBE_READ_REG(hw, IXGBE_PXOFFRXC(i)); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: + hwstats->pxonrxc[i] += + IXGBE_READ_REG(hw, IXGBE_PXONRXCNT(i)); + break; + default: + break; } hwstats->pxontxc[i] += IXGBE_READ_REG(hw, IXGBE_PXONTXC(i)); hwstats->pxofftxc[i] += IXGBE_READ_REG(hw, IXGBE_PXOFFTXC(i)); @@ -5501,21 +5749,25 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) /* work around hardware counting issue */ hwstats->gprc -= missed_rx; + ixgbe_update_xoff_received(adapter); + /* 82598 hardware only has a 32 bit counter in the high register */ - if (hw->mac.type == ixgbe_mac_82599EB) { - u64 tmp; + switch (hw->mac.type) { + case ixgbe_mac_82598EB: + hwstats->lxonrxc += IXGBE_READ_REG(hw, IXGBE_LXONRXC); + hwstats->gorc += IXGBE_READ_REG(hw, IXGBE_GORCH); + hwstats->gotc += IXGBE_READ_REG(hw, IXGBE_GOTCH); + hwstats->tor += IXGBE_READ_REG(hw, IXGBE_TORH); + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: hwstats->gorc += IXGBE_READ_REG(hw, IXGBE_GORCL); - tmp = IXGBE_READ_REG(hw, IXGBE_GORCH) & 0xF; - /* 4 high bits of GORC */ - hwstats->gorc += (tmp << 32); + IXGBE_READ_REG(hw, IXGBE_GORCH); /* to clear */ hwstats->gotc += IXGBE_READ_REG(hw, IXGBE_GOTCL); - tmp = IXGBE_READ_REG(hw, IXGBE_GOTCH) & 0xF; - /* 4 high bits of GOTC */ - hwstats->gotc += (tmp << 32); + IXGBE_READ_REG(hw, IXGBE_GOTCH); /* to clear */ hwstats->tor += IXGBE_READ_REG(hw, IXGBE_TORL); - IXGBE_READ_REG(hw, IXGBE_TORH); /* to clear */ + IXGBE_READ_REG(hw, IXGBE_TORH); /* to clear */ hwstats->lxonrxc += IXGBE_READ_REG(hw, IXGBE_LXONRXCNT); - hwstats->lxoffrxc += IXGBE_READ_REG(hw, IXGBE_LXOFFRXCNT); hwstats->fdirmatch += IXGBE_READ_REG(hw, IXGBE_FDIRMATCH); hwstats->fdirmiss += IXGBE_READ_REG(hw, IXGBE_FDIRMISS); #ifdef IXGBE_FCOE @@ -5526,12 +5778,9 @@ void ixgbe_update_stats(struct ixgbe_adapter *adapter) hwstats->fcoedwrc += IXGBE_READ_REG(hw, IXGBE_FCOEDWRC); hwstats->fcoedwtc += IXGBE_READ_REG(hw, IXGBE_FCOEDWTC); #endif /* IXGBE_FCOE */ - } else { - hwstats->lxonrxc += IXGBE_READ_REG(hw, IXGBE_LXONRXC); - hwstats->lxoffrxc += IXGBE_READ_REG(hw, IXGBE_LXOFFRXC); - hwstats->gorc += IXGBE_READ_REG(hw, IXGBE_GORCH); - hwstats->gotc += IXGBE_READ_REG(hw, IXGBE_GOTCH); - hwstats->tor += IXGBE_READ_REG(hw, IXGBE_TORH); + break; + default: + break; } bprc = IXGBE_READ_REG(hw, IXGBE_BPRC); hwstats->bprc += bprc; @@ -5704,8 +5953,8 @@ static void ixgbe_fdir_reinit_task(struct work_struct *work) if (ixgbe_reinit_fdir_tables_82599(hw) == 0) { for (i = 0; i < adapter->num_tx_queues; i++) - set_bit(__IXGBE_FDIR_INIT_DONE, - &(adapter->tx_ring[i]->reinit_state)); + set_bit(__IXGBE_TX_FDIR_INIT_DONE, + &(adapter->tx_ring[i]->state)); } else { e_err(probe, "failed to finish FDIR re-initialization, " "ignored adding FDIR ATR filters\n"); @@ -5714,6 +5963,26 @@ static void ixgbe_fdir_reinit_task(struct work_struct *work) netif_tx_start_all_queues(adapter->netdev); } +static void ixgbe_spoof_check(struct ixgbe_adapter *adapter) +{ + u32 ssvpc; + + /* Do not perform spoof check for 82598 */ + if (adapter->hw.mac.type == ixgbe_mac_82598EB) + return; + + ssvpc = IXGBE_READ_REG(&adapter->hw, IXGBE_SSVPC); + + /* + * ssvpc register is cleared on read, if zero then no + * spoofed packets in the last interval. + */ + if (!ssvpc) + return; + + e_warn(drv, "%d Spoofed packets detected\n", ssvpc); +} + static DEFINE_MUTEX(ixgbe_watchdog_lock); /** @@ -5767,17 +6036,27 @@ static void ixgbe_watchdog_task(struct work_struct *work) if (!netif_carrier_ok(netdev)) { bool flow_rx, flow_tx; - if (hw->mac.type == ixgbe_mac_82599EB) { - u32 mflcn = IXGBE_READ_REG(hw, IXGBE_MFLCN); - u32 fccfg = IXGBE_READ_REG(hw, IXGBE_FCCFG); - flow_rx = !!(mflcn & IXGBE_MFLCN_RFCE); - flow_tx = !!(fccfg & IXGBE_FCCFG_TFCE_802_3X); - } else { + switch (hw->mac.type) { + case ixgbe_mac_82598EB: { u32 frctl = IXGBE_READ_REG(hw, IXGBE_FCTRL); u32 rmcs = IXGBE_READ_REG(hw, IXGBE_RMCS); flow_rx = !!(frctl & IXGBE_FCTRL_RFCE); flow_tx = !!(rmcs & IXGBE_RMCS_TFCE_802_3X); } + break; + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: { + u32 mflcn = IXGBE_READ_REG(hw, IXGBE_MFLCN); + u32 fccfg = IXGBE_READ_REG(hw, IXGBE_FCCFG); + flow_rx = !!(mflcn & IXGBE_MFLCN_RFCE); + flow_tx = !!(fccfg & IXGBE_FCCFG_TFCE_802_3X); + } + break; + default: + flow_tx = false; + flow_rx = false; + break; + } e_info(drv, "NIC Link is Up %s, Flow Control: %s\n", (link_speed == IXGBE_LINK_SPEED_10GB_FULL ? @@ -5791,7 +6070,10 @@ static void ixgbe_watchdog_task(struct work_struct *work) netif_carrier_on(netdev); } else { /* Force detection of hung controller */ - adapter->detect_tx_hung = true; + for (i = 0; i < adapter->num_tx_queues; i++) { + tx_ring = adapter->tx_ring[i]; + set_check_for_tx_hang(tx_ring); + } } } else { adapter->link_up = false; @@ -5821,6 +6103,7 @@ static void ixgbe_watchdog_task(struct work_struct *work) } } + ixgbe_spoof_check(adapter); ixgbe_update_stats(adapter); mutex_unlock(&ixgbe_watchdog_lock); } @@ -6003,15 +6286,17 @@ static bool ixgbe_tx_csum(struct ixgbe_adapter *adapter, static int ixgbe_tx_map(struct ixgbe_adapter *adapter, struct ixgbe_ring *tx_ring, struct sk_buff *skb, u32 tx_flags, - unsigned int first) + unsigned int first, const u8 hdr_len) { - struct pci_dev *pdev = adapter->pdev; + struct device *dev = tx_ring->dev; struct ixgbe_tx_buffer *tx_buffer_info; unsigned int len; unsigned int total = skb->len; unsigned int offset = 0, size, count = 0, i; unsigned int nr_frags = skb_shinfo(skb)->nr_frags; unsigned int f; + unsigned int bytecount = skb->len; + u16 gso_segs = 1; i = tx_ring->next_to_use; @@ -6026,10 +6311,10 @@ static int ixgbe_tx_map(struct ixgbe_adapter *adapter, tx_buffer_info->length = size; tx_buffer_info->mapped_as_page = false; - tx_buffer_info->dma = dma_map_single(&pdev->dev, + tx_buffer_info->dma = dma_map_single(dev, skb->data + offset, size, DMA_TO_DEVICE); - if (dma_mapping_error(&pdev->dev, tx_buffer_info->dma)) + if (dma_mapping_error(dev, tx_buffer_info->dma)) goto dma_error; tx_buffer_info->time_stamp = jiffies; tx_buffer_info->next_to_watch = i; @@ -6062,12 +6347,12 @@ static int ixgbe_tx_map(struct ixgbe_adapter *adapter, size = min(len, (uint)IXGBE_MAX_DATA_PER_TXD); tx_buffer_info->length = size; - tx_buffer_info->dma = dma_map_page(&adapter->pdev->dev, + tx_buffer_info->dma = dma_map_page(dev, frag->page, offset, size, DMA_TO_DEVICE); tx_buffer_info->mapped_as_page = true; - if (dma_mapping_error(&pdev->dev, tx_buffer_info->dma)) + if (dma_mapping_error(dev, tx_buffer_info->dma)) goto dma_error; tx_buffer_info->time_stamp = jiffies; tx_buffer_info->next_to_watch = i; @@ -6081,6 +6366,19 @@ static int ixgbe_tx_map(struct ixgbe_adapter *adapter, break; } + if (tx_flags & IXGBE_TX_FLAGS_TSO) + gso_segs = skb_shinfo(skb)->gso_segs; +#ifdef IXGBE_FCOE + /* adjust for FCoE Sequence Offload */ + else if (tx_flags & IXGBE_TX_FLAGS_FSO) + gso_segs = DIV_ROUND_UP(skb->len - hdr_len, + skb_shinfo(skb)->gso_size); +#endif /* IXGBE_FCOE */ + bytecount += (gso_segs - 1) * hdr_len; + + /* multiply data chunks by size of headers */ + tx_ring->tx_buffer_info[i].bytecount = bytecount; + tx_ring->tx_buffer_info[i].gso_segs = gso_segs; tx_ring->tx_buffer_info[i].skb = skb; tx_ring->tx_buffer_info[first].next_to_watch = i; @@ -6102,14 +6400,13 @@ dma_error: i += tx_ring->count; i--; tx_buffer_info = &tx_ring->tx_buffer_info[i]; - ixgbe_unmap_and_free_tx_resource(adapter, tx_buffer_info); + ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer_info); } return 0; } -static void ixgbe_tx_queue(struct ixgbe_adapter *adapter, - struct ixgbe_ring *tx_ring, +static void ixgbe_tx_queue(struct ixgbe_ring *tx_ring, int tx_flags, int count, u32 paylen, u8 hdr_len) { union ixgbe_adv_tx_desc *tx_desc = NULL; @@ -6174,60 +6471,46 @@ static void ixgbe_tx_queue(struct ixgbe_adapter *adapter, wmb(); tx_ring->next_to_use = i; - writel(i, adapter->hw.hw_addr + tx_ring->tail); + writel(i, tx_ring->tail); } static void ixgbe_atr(struct ixgbe_adapter *adapter, struct sk_buff *skb, - int queue, u32 tx_flags, __be16 protocol) + u8 queue, u32 tx_flags, __be16 protocol) { struct ixgbe_atr_input atr_input; - struct tcphdr *th; struct iphdr *iph = ip_hdr(skb); struct ethhdr *eth = (struct ethhdr *)skb->data; - u16 vlan_id, src_port, dst_port, flex_bytes; - u32 src_ipv4_addr, dst_ipv4_addr; - u8 l4type = 0; + struct tcphdr *th; + u16 vlan_id; - /* Right now, we support IPv4 only */ - if (protocol != htons(ETH_P_IP)) - return; - /* check if we're UDP or TCP */ - if (iph->protocol == IPPROTO_TCP) { - th = tcp_hdr(skb); - src_port = th->source; - dst_port = th->dest; - l4type |= IXGBE_ATR_L4TYPE_TCP; - /* l4type IPv4 type is 0, no need to assign */ - } else { - /* Unsupported L4 header, just bail here */ + /* Right now, we support IPv4 w/ TCP only */ + if (protocol != htons(ETH_P_IP) || + iph->protocol != IPPROTO_TCP) return; - } memset(&atr_input, 0, sizeof(struct ixgbe_atr_input)); vlan_id = (tx_flags & IXGBE_TX_FLAGS_VLAN_MASK) >> IXGBE_TX_FLAGS_VLAN_SHIFT; - src_ipv4_addr = iph->saddr; - dst_ipv4_addr = iph->daddr; - flex_bytes = eth->h_proto; + + th = tcp_hdr(skb); ixgbe_atr_set_vlan_id_82599(&atr_input, vlan_id); - ixgbe_atr_set_src_port_82599(&atr_input, dst_port); - ixgbe_atr_set_dst_port_82599(&atr_input, src_port); - ixgbe_atr_set_flex_byte_82599(&atr_input, flex_bytes); - ixgbe_atr_set_l4type_82599(&atr_input, l4type); + ixgbe_atr_set_src_port_82599(&atr_input, th->dest); + ixgbe_atr_set_dst_port_82599(&atr_input, th->source); + ixgbe_atr_set_flex_byte_82599(&atr_input, eth->h_proto); + ixgbe_atr_set_l4type_82599(&atr_input, IXGBE_ATR_L4TYPE_TCP); /* src and dst are inverted, think how the receiver sees them */ - ixgbe_atr_set_src_ipv4_82599(&atr_input, dst_ipv4_addr); - ixgbe_atr_set_dst_ipv4_82599(&atr_input, src_ipv4_addr); + ixgbe_atr_set_src_ipv4_82599(&atr_input, iph->daddr); + ixgbe_atr_set_dst_ipv4_82599(&atr_input, iph->saddr); /* This assumes the Rx queue and Tx queue are bound to the same CPU */ ixgbe_fdir_add_signature_filter_82599(&adapter->hw, &atr_input, queue); } -static int __ixgbe_maybe_stop_tx(struct net_device *netdev, - struct ixgbe_ring *tx_ring, int size) +static int __ixgbe_maybe_stop_tx(struct ixgbe_ring *tx_ring, int size) { - netif_stop_subqueue(netdev, tx_ring->queue_index); + netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); /* Herbert's original patch had: * smp_mb__after_netif_stop_queue(); * but since that doesn't exist yet, just open code it. */ @@ -6239,17 +6522,16 @@ static int __ixgbe_maybe_stop_tx(struct net_device *netdev, return -EBUSY; /* A reprieve! - use start_queue because it doesn't call schedule */ - netif_start_subqueue(netdev, tx_ring->queue_index); - ++tx_ring->restart_queue; + netif_start_subqueue(tx_ring->netdev, tx_ring->queue_index); + ++tx_ring->tx_stats.restart_queue; return 0; } -static int ixgbe_maybe_stop_tx(struct net_device *netdev, - struct ixgbe_ring *tx_ring, int size) +static int ixgbe_maybe_stop_tx(struct ixgbe_ring *tx_ring, int size) { if (likely(IXGBE_DESC_UNUSED(tx_ring) >= size)) return 0; - return __ixgbe_maybe_stop_tx(netdev, tx_ring, size); + return __ixgbe_maybe_stop_tx(tx_ring, size); } static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb) @@ -6294,10 +6576,11 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb) return skb_tx_hash(dev, skb); } -netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct net_device *netdev, +netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct ixgbe_adapter *adapter, struct ixgbe_ring *tx_ring) { + struct net_device *netdev = tx_ring->netdev; struct netdev_queue *txq; unsigned int first; unsigned int tx_flags = 0; @@ -6355,8 +6638,8 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct net_device *netdev for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); - if (ixgbe_maybe_stop_tx(netdev, tx_ring, count)) { - adapter->tx_busy++; + if (ixgbe_maybe_stop_tx(tx_ring, count)) { + tx_ring->tx_stats.tx_busy++; return NETDEV_TX_BUSY; } @@ -6390,14 +6673,14 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct net_device *netdev tx_flags |= IXGBE_TX_FLAGS_CSUM; } - count = ixgbe_tx_map(adapter, tx_ring, skb, tx_flags, first); + count = ixgbe_tx_map(adapter, tx_ring, skb, tx_flags, first, hdr_len); if (count) { /* add the ATR filter if ATR is on */ if (tx_ring->atr_sample_rate) { ++tx_ring->atr_count; if ((tx_ring->atr_count >= tx_ring->atr_sample_rate) && - test_bit(__IXGBE_FDIR_INIT_DONE, - &tx_ring->reinit_state)) { + test_bit(__IXGBE_TX_FDIR_INIT_DONE, + &tx_ring->state)) { ixgbe_atr(adapter, skb, tx_ring->queue_index, tx_flags, protocol); tx_ring->atr_count = 0; @@ -6406,9 +6689,8 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct net_device *netdev txq = netdev_get_tx_queue(netdev, tx_ring->queue_index); txq->tx_bytes += skb->len; txq->tx_packets++; - ixgbe_tx_queue(adapter, tx_ring, tx_flags, count, skb->len, - hdr_len); - ixgbe_maybe_stop_tx(netdev, tx_ring, DESC_NEEDED); + ixgbe_tx_queue(tx_ring, tx_flags, count, skb->len, hdr_len); + ixgbe_maybe_stop_tx(tx_ring, DESC_NEEDED); } else { dev_kfree_skb_any(skb); @@ -6425,7 +6707,7 @@ static netdev_tx_t ixgbe_xmit_frame(struct sk_buff *skb, struct net_device *netd struct ixgbe_ring *tx_ring; tx_ring = adapter->tx_ring[skb->queue_mapping]; - return ixgbe_xmit_frame_ring(skb, netdev, adapter, tx_ring); + return ixgbe_xmit_frame_ring(skb, adapter, tx_ring); } /** @@ -6566,20 +6848,23 @@ static struct rtnl_link_stats64 *ixgbe_get_stats64(struct net_device *netdev, /* accurate rx/tx bytes/packets stats */ dev_txq_stats_fold(netdev, stats); + rcu_read_lock(); for (i = 0; i < adapter->num_rx_queues; i++) { - struct ixgbe_ring *ring = adapter->rx_ring[i]; + struct ixgbe_ring *ring = ACCESS_ONCE(adapter->rx_ring[i]); u64 bytes, packets; unsigned int start; - do { - start = u64_stats_fetch_begin_bh(&ring->syncp); - packets = ring->stats.packets; - bytes = ring->stats.bytes; - } while (u64_stats_fetch_retry_bh(&ring->syncp, start)); - stats->rx_packets += packets; - stats->rx_bytes += bytes; + if (ring) { + do { + start = u64_stats_fetch_begin_bh(&ring->syncp); + packets = ring->stats.packets; + bytes = ring->stats.bytes; + } while (u64_stats_fetch_retry_bh(&ring->syncp, start)); + stats->rx_packets += packets; + stats->rx_bytes += bytes; + } } - + rcu_read_unlock(); /* following stats updated by ixgbe_watchdog_task() */ stats->multicast = netdev->stats.multicast; stats->rx_errors = netdev->stats.rx_errors; @@ -6628,7 +6913,7 @@ static void __devinit ixgbe_probe_vf(struct ixgbe_adapter *adapter, struct ixgbe_hw *hw = &adapter->hw; int err; - if (hw->mac.type != ixgbe_mac_82599EB || !max_vfs) + if (hw->mac.type == ixgbe_mac_82598EB || !max_vfs) return; /* The 82599 supports up to 64 VFs per physical function @@ -6694,11 +6979,12 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, const struct ixgbe_info *ii = ixgbe_info_tbl[ent->driver_data]; static int cards_found; int i, err, pci_using_dac; + u8 part_str[IXGBE_PBANUM_LENGTH]; unsigned int indices = num_possible_cpus(); #ifdef IXGBE_FCOE u16 device_caps; #endif - u32 part_num, eec; + u32 eec; /* Catch broken hardware that put the wrong VF device ID in * the PCIe SR-IOV capability. @@ -6761,8 +7047,8 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, SET_NETDEV_DEV(netdev, &pdev->dev); - pci_set_drvdata(pdev, netdev); adapter = netdev_priv(netdev); + pci_set_drvdata(pdev, adapter); adapter->netdev = netdev; adapter->pdev = pdev; @@ -6785,7 +7071,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, netdev->netdev_ops = &ixgbe_netdev_ops; ixgbe_set_ethtool_ops(netdev); netdev->watchdog_timeo = 5 * HZ; - strcpy(netdev->name, pci_name(pdev)); + strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1); adapter->bd_number = cards_found; @@ -6835,8 +7121,14 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, goto err_sw_init; /* Make it possible the adapter to be woken up via WOL */ - if (adapter->hw.mac.type == ixgbe_mac_82599EB) + switch (adapter->hw.mac.type) { + case ixgbe_mac_82599EB: + case ixgbe_mac_X540: IXGBE_WRITE_REG(&adapter->hw, IXGBE_WUS, ~0); + break; + default: + break; + } /* * If there is a fan on this device and it has failed log the @@ -6944,8 +7236,11 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, goto err_eeprom; } - /* power down the optics */ - if (hw->phy.multispeed_fiber) + /* power down the optics for multispeed fiber and 82599 SFP+ fiber */ + if (hw->mac.ops.disable_tx_laser && + ((hw->phy.multispeed_fiber) || + ((hw->mac.ops.get_media_type(hw) == ixgbe_media_type_fiber) && + (hw->mac.type == ixgbe_mac_82599EB)))) hw->mac.ops.disable_tx_laser(hw); init_timer(&adapter->watchdog_timer); @@ -6960,6 +7255,18 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, goto err_sw_init; switch (pdev->device) { + case IXGBE_DEV_ID_82599_SFP: + /* Only this subdevice supports WOL */ + if (pdev->subsystem_device == IXGBE_SUBDEV_ID_82599_SFP) + adapter->wol = (IXGBE_WUFC_MAG | IXGBE_WUFC_EX | + IXGBE_WUFC_MC | IXGBE_WUFC_BC); + break; + case IXGBE_DEV_ID_82599_COMBO_BACKPLANE: + /* All except this subdevice support WOL */ + if (pdev->subsystem_device != IXGBE_SUBDEV_ID_82599_KX4_KR_MEZZ) + adapter->wol = (IXGBE_WUFC_MAG | IXGBE_WUFC_EX | + IXGBE_WUFC_MC | IXGBE_WUFC_BC); + break; case IXGBE_DEV_ID_82599_KX4: adapter->wol = (IXGBE_WUFC_MAG | IXGBE_WUFC_EX | IXGBE_WUFC_MC | IXGBE_WUFC_BC); @@ -6983,16 +7290,17 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev, hw->bus.width == ixgbe_bus_width_pcie_x1 ? "Width x1" : "Unknown"), netdev->dev_addr); - ixgbe_read_pba_num_generic(hw, &part_num); + + err = ixgbe_read_pba_string_generic(hw, part_str, IXGBE_PBANUM_LENGTH); + if (err) + strncpy(part_str, "Unknown", IXGBE_PBANUM_LENGTH); if (ixgbe_is_sfp(hw) && hw->phy.sfp_type != ixgbe_sfp_type_not_present) - e_dev_info("MAC: %d, PHY: %d, SFP+: %d, " - "PBA No: %06x-%03x\n", + e_dev_info("MAC: %d, PHY: %d, SFP+: %d, PBA No: %s\n", hw->mac.type, hw->phy.type, hw->phy.sfp_type, - (part_num >> 8), (part_num & 0xff)); + part_str); else - e_dev_info("MAC: %d, PHY: %d, PBA No: %06x-%03x\n", - hw->mac.type, hw->phy.type, - (part_num >> 8), (part_num & 0xff)); + e_dev_info("MAC: %d, PHY: %d, PBA No: %s\n", + hw->mac.type, hw->phy.type, part_str); if (hw->bus.width <= ixgbe_bus_width_pcie_x4) { e_dev_warn("PCI-Express bandwidth available for this card is " @@ -7085,17 +7393,19 @@ err_dma: **/ static void __devexit ixgbe_remove(struct pci_dev *pdev) { - struct net_device *netdev = pci_get_drvdata(pdev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = pci_get_drvdata(pdev); + struct net_device *netdev = adapter->netdev; set_bit(__IXGBE_DOWN, &adapter->state); - /* clear the module not found bit to make sure the worker won't - * reschedule + + /* + * The timers may be rescheduled, so explicitly disable them + * from being rescheduled. */ clear_bit(__IXGBE_SFP_MODULE_NOT_FOUND, &adapter->state); del_timer_sync(&adapter->watchdog_timer); - del_timer_sync(&adapter->sfp_timer); + cancel_work_sync(&adapter->watchdog_task); cancel_work_sync(&adapter->sfp_task); cancel_work_sync(&adapter->multispeed_fiber_task); @@ -7103,7 +7413,8 @@ static void __devexit ixgbe_remove(struct pci_dev *pdev) if (adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE || adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE) cancel_work_sync(&adapter->fdir_reinit_task); - flush_scheduled_work(); + if (adapter->flags2 & IXGBE_FLAG2_TEMP_SENSOR_CAPABLE) + cancel_work_sync(&adapter->check_overtemp_task); #ifdef CONFIG_IXGBE_DCA if (adapter->flags & IXGBE_FLAG_DCA_ENABLED) { @@ -7156,8 +7467,8 @@ static void __devexit ixgbe_remove(struct pci_dev *pdev) static pci_ers_result_t ixgbe_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state) { - struct net_device *netdev = pci_get_drvdata(pdev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = pci_get_drvdata(pdev); + struct net_device *netdev = adapter->netdev; netif_device_detach(netdev); @@ -7180,8 +7491,7 @@ static pci_ers_result_t ixgbe_io_error_detected(struct pci_dev *pdev, */ static pci_ers_result_t ixgbe_io_slot_reset(struct pci_dev *pdev) { - struct net_device *netdev = pci_get_drvdata(pdev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = pci_get_drvdata(pdev); pci_ers_result_t result; int err; @@ -7219,8 +7529,8 @@ static pci_ers_result_t ixgbe_io_slot_reset(struct pci_dev *pdev) */ static void ixgbe_io_resume(struct pci_dev *pdev) { - struct net_device *netdev = pci_get_drvdata(pdev); - struct ixgbe_adapter *adapter = netdev_priv(netdev); + struct ixgbe_adapter *adapter = pci_get_drvdata(pdev); + struct net_device *netdev = adapter->netdev; if (netif_running(netdev)) { if (ixgbe_up(adapter)) { @@ -7285,6 +7595,7 @@ static void __exit ixgbe_exit_module(void) dca_unregister_notify(&dca_notifier); #endif pci_unregister_driver(&ixgbe_driver); + rcu_barrier(); /* Wait for completion of call_rcu()'s */ } #ifdef CONFIG_IXGBE_DCA |