X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=drivers%2Fnet%2Fe1000%2Fe1000_main.c;h=97e71a4fe8eb70e91db07df7e52432b1287aa17c;hb=43bc926fffd92024b46cafaf7350d669ba9ca884;hp=137226d98d4760f8d1a72b536ffa530effcbd541;hpb=cee37fe97739d85991964371c1f3a745c00dd236;p=linux-2.6.git diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 137226d98..97e71a4fe 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -29,21 +29,54 @@ #include "e1000.h" /* Change Log - * 6.0.44+ 2/15/05 - * o applied Anton's patch to resolve tx hang in hardware - * o Applied Andrew Mortons patch - e1000 stops working after resume + * 7.0.33 3-Feb-2006 + * o Added another fix for the pass false carrier bit + * 7.0.32 24-Jan-2006 + * o Need to rebuild with noew version number for the pass false carrier + * fix in e1000_hw.c + * 7.0.30 18-Jan-2006 + * o fixup for tso workaround to disable it for pci-x + * o fix mem leak on 82542 + * o fixes for 10 Mb/s connections and incorrect stats + * 7.0.28 01/06/2006 + * o hardware workaround to only set "speed mode" bit for 1G link. + * 7.0.26 12/23/2005 + * o wake on lan support modified for device ID 10B5 + * o fix dhcp + vlan issue not making it to the iAMT firmware + * 7.0.24 12/9/2005 + * o New hardware support for the Gigabit NIC embedded in the south bridge + * o Fixes to the recycling logic (skb->tail) from IBM LTC + * 6.3.9 12/16/2005 + * o incorporate fix for recycled skbs from IBM LTC + * 6.3.7 11/18/2005 + * o Honor eeprom setting for enabling/disabling Wake On Lan + * 6.3.5 11/17/2005 + * o Fix memory leak in rx ring handling for PCI Express adapters + * 6.3.4 11/8/05 + * o Patch from Jesper Juhl to remove redundant NULL checks for kfree + * 6.3.2 9/20/05 + * o Render logic that sets/resets DRV_LOAD as inline functions to + * avoid code replication. If f/w is AMT then set DRV_LOAD only when + * network interface is open. + * o Handle DRV_LOAD set/reset in cases where AMT uses VLANs. + * o Adjust PBA partioning for Jumbo frames using MTU size and not + * rx_buffer_len + * 6.3.1 9/19/05 + * o Use adapter->tx_timeout_factor in Tx Hung Detect logic + * (e1000_clean_tx_irq) + * o Support for 8086:10B5 device (Quad Port) */ char e1000_driver_name[] = "e1000"; -char e1000_driver_string[] = "Intel(R) PRO/1000 Network Driver"; +static char e1000_driver_string[] = "Intel(R) PRO/1000 Network Driver"; #ifndef CONFIG_E1000_NAPI #define DRIVERNAPI #else #define DRIVERNAPI "-NAPI" #endif -#define DRV_VERSION "6.0.54-k2"DRIVERNAPI +#define DRV_VERSION "7.0.33-k2"DRIVERNAPI char e1000_driver_version[] = DRV_VERSION; -char e1000_copyright[] = "Copyright (c) 1999-2004 Intel Corporation."; +static char e1000_copyright[] = "Copyright (c) 1999-2005 Intel Corporation."; /* e1000_pci_tbl - PCI Device ID Table * @@ -78,6 +111,9 @@ static struct pci_device_id e1000_pci_tbl[] = { INTEL_E1000_ETHERNET_DEVICE(0x1026), INTEL_E1000_ETHERNET_DEVICE(0x1027), INTEL_E1000_ETHERNET_DEVICE(0x1028), + INTEL_E1000_ETHERNET_DEVICE(0x105E), + INTEL_E1000_ETHERNET_DEVICE(0x105F), + INTEL_E1000_ETHERNET_DEVICE(0x1060), INTEL_E1000_ETHERNET_DEVICE(0x1075), INTEL_E1000_ETHERNET_DEVICE(0x1076), INTEL_E1000_ETHERNET_DEVICE(0x1077), @@ -86,25 +122,32 @@ static struct pci_device_id e1000_pci_tbl[] = { INTEL_E1000_ETHERNET_DEVICE(0x107A), INTEL_E1000_ETHERNET_DEVICE(0x107B), INTEL_E1000_ETHERNET_DEVICE(0x107C), + INTEL_E1000_ETHERNET_DEVICE(0x107D), + INTEL_E1000_ETHERNET_DEVICE(0x107E), + INTEL_E1000_ETHERNET_DEVICE(0x107F), INTEL_E1000_ETHERNET_DEVICE(0x108A), INTEL_E1000_ETHERNET_DEVICE(0x108B), INTEL_E1000_ETHERNET_DEVICE(0x108C), + INTEL_E1000_ETHERNET_DEVICE(0x1096), + INTEL_E1000_ETHERNET_DEVICE(0x1098), INTEL_E1000_ETHERNET_DEVICE(0x1099), + INTEL_E1000_ETHERNET_DEVICE(0x109A), + INTEL_E1000_ETHERNET_DEVICE(0x10B5), + INTEL_E1000_ETHERNET_DEVICE(0x10B9), /* required last entry */ {0,} }; MODULE_DEVICE_TABLE(pci, e1000_pci_tbl); -int e1000_up(struct e1000_adapter *adapter); -void e1000_down(struct e1000_adapter *adapter); -void e1000_reset(struct e1000_adapter *adapter); -int e1000_set_spd_dplx(struct e1000_adapter *adapter, uint16_t spddplx); -int e1000_setup_tx_resources(struct e1000_adapter *adapter); -int e1000_setup_rx_resources(struct e1000_adapter *adapter); -void e1000_free_tx_resources(struct e1000_adapter *adapter); -void e1000_free_rx_resources(struct e1000_adapter *adapter); -void e1000_update_stats(struct e1000_adapter *adapter); +static int e1000_setup_tx_resources(struct e1000_adapter *adapter, + struct e1000_tx_ring *txdr); +static int e1000_setup_rx_resources(struct e1000_adapter *adapter, + struct e1000_rx_ring *rxdr); +static void e1000_free_tx_resources(struct e1000_adapter *adapter, + struct e1000_tx_ring *tx_ring); +static void e1000_free_rx_resources(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring); /* Local Function Prototypes */ @@ -112,14 +155,19 @@ static int e1000_init_module(void); static void e1000_exit_module(void); static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent); static void __devexit e1000_remove(struct pci_dev *pdev); +static int e1000_alloc_queues(struct e1000_adapter *adapter); static int e1000_sw_init(struct e1000_adapter *adapter); static int e1000_open(struct net_device *netdev); static int e1000_close(struct net_device *netdev); static void e1000_configure_tx(struct e1000_adapter *adapter); static void e1000_configure_rx(struct e1000_adapter *adapter); static void e1000_setup_rctl(struct e1000_adapter *adapter); -static void e1000_clean_tx_ring(struct e1000_adapter *adapter); -static void e1000_clean_rx_ring(struct e1000_adapter *adapter); +static void e1000_clean_all_tx_rings(struct e1000_adapter *adapter); +static void e1000_clean_all_rx_rings(struct e1000_adapter *adapter); +static void e1000_clean_tx_ring(struct e1000_adapter *adapter, + struct e1000_tx_ring *tx_ring); +static void e1000_clean_rx_ring(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring); static void e1000_set_multi(struct net_device *netdev); static void e1000_update_phy_info(unsigned long data); static void e1000_watchdog(unsigned long data); @@ -130,27 +178,35 @@ static struct net_device_stats * e1000_get_stats(struct net_device *netdev); static int e1000_change_mtu(struct net_device *netdev, int new_mtu); static int e1000_set_mac(struct net_device *netdev, void *p); static irqreturn_t e1000_intr(int irq, void *data, struct pt_regs *regs); -static boolean_t e1000_clean_tx_irq(struct e1000_adapter *adapter); +static boolean_t e1000_clean_tx_irq(struct e1000_adapter *adapter, + struct e1000_tx_ring *tx_ring); #ifdef CONFIG_E1000_NAPI -static int e1000_clean(struct net_device *netdev, int *budget); +static int e1000_clean(struct net_device *poll_dev, int *budget); static boolean_t e1000_clean_rx_irq(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, int *work_done, int work_to_do); static boolean_t e1000_clean_rx_irq_ps(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, int *work_done, int work_to_do); #else -static boolean_t e1000_clean_rx_irq(struct e1000_adapter *adapter); -static boolean_t e1000_clean_rx_irq_ps(struct e1000_adapter *adapter); +static boolean_t e1000_clean_rx_irq(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring); +static boolean_t e1000_clean_rx_irq_ps(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring); #endif -static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter); -static void e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter); +static void e1000_alloc_rx_buffers(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, + int cleaned_count); +static void e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, + int cleaned_count); static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd); static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd); -void e1000_set_ethtool_ops(struct net_device *netdev); static void e1000_enter_82542_rst(struct e1000_adapter *adapter); static void e1000_leave_82542_rst(struct e1000_adapter *adapter); static void e1000_tx_timeout(struct net_device *dev); -static void e1000_tx_timeout_task(struct net_device *dev); +static void e1000_reset_task(struct net_device *dev); static void e1000_smartspeed(struct e1000_adapter *adapter); static inline int e1000_82547_fifo_workaround(struct e1000_adapter *adapter, struct sk_buff *skb); @@ -160,26 +216,17 @@ static void e1000_vlan_rx_add_vid(struct net_device *netdev, uint16_t vid); static void e1000_vlan_rx_kill_vid(struct net_device *netdev, uint16_t vid); static void e1000_restore_vlan(struct e1000_adapter *adapter); -static int e1000_notify_reboot(struct notifier_block *, unsigned long event, void *ptr); -static int e1000_suspend(struct pci_dev *pdev, uint32_t state); #ifdef CONFIG_PM +static int e1000_suspend(struct pci_dev *pdev, pm_message_t state); static int e1000_resume(struct pci_dev *pdev); #endif +static void e1000_shutdown(struct pci_dev *pdev); #ifdef CONFIG_NET_POLL_CONTROLLER /* for netdump / net console */ static void e1000_netpoll (struct net_device *netdev); #endif -struct notifier_block e1000_notifier_reboot = { - .notifier_call = e1000_notify_reboot, - .next = NULL, - .priority = 0 -}; - -/* Exported from other modules */ - -extern void e1000_check_options(struct e1000_adapter *adapter); static struct pci_driver e1000_driver = { .name = e1000_driver_name, @@ -189,8 +236,9 @@ static struct pci_driver e1000_driver = { /* Power Managment Hooks */ #ifdef CONFIG_PM .suspend = e1000_suspend, - .resume = e1000_resume + .resume = e1000_resume, #endif + .shutdown = e1000_shutdown }; MODULE_AUTHOR("Intel Corporation, "); @@ -219,9 +267,7 @@ e1000_init_module(void) printk(KERN_INFO "%s\n", e1000_copyright); ret = pci_module_init(&e1000_driver); - if(ret >= 0) { - register_reboot_notifier(&e1000_notifier_reboot); - } + return ret; } @@ -237,7 +283,6 @@ module_init(e1000_init_module); static void __exit e1000_exit_module(void) { - unregister_reboot_notifier(&e1000_notifier_reboot); pci_unregister_driver(&e1000_driver); } @@ -265,47 +310,117 @@ e1000_irq_disable(struct e1000_adapter *adapter) static inline void e1000_irq_enable(struct e1000_adapter *adapter) { - if(likely(atomic_dec_and_test(&adapter->irq_sem))) { + if (likely(atomic_dec_and_test(&adapter->irq_sem))) { E1000_WRITE_REG(&adapter->hw, IMS, IMS_ENABLE_MASK); E1000_WRITE_FLUSH(&adapter->hw); } } -void + +static void e1000_update_mng_vlan(struct e1000_adapter *adapter) { struct net_device *netdev = adapter->netdev; uint16_t vid = adapter->hw.mng_cookie.vlan_id; uint16_t old_vid = adapter->mng_vlan_id; - if(adapter->vlgrp) { - if(!adapter->vlgrp->vlan_devices[vid]) { - if(adapter->hw.mng_cookie.status & + if (adapter->vlgrp) { + if (!adapter->vlgrp->vlan_devices[vid]) { + if (adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT) { e1000_vlan_rx_add_vid(netdev, vid); adapter->mng_vlan_id = vid; } else adapter->mng_vlan_id = E1000_MNG_VLAN_NONE; - - if((old_vid != (uint16_t)E1000_MNG_VLAN_NONE) && - (vid != old_vid) && + + if ((old_vid != (uint16_t)E1000_MNG_VLAN_NONE) && + (vid != old_vid) && !adapter->vlgrp->vlan_devices[old_vid]) e1000_vlan_rx_kill_vid(netdev, old_vid); - } + } else + adapter->mng_vlan_id = vid; + } +} + +/** + * e1000_release_hw_control - release control of the h/w to f/w + * @adapter: address of board private structure + * + * e1000_release_hw_control resets {CTRL_EXT|FWSM}:DRV_LOAD bit. + * For ASF and Pass Through versions of f/w this means that the + * driver is no longer loaded. For AMT version (only with 82573) i + * of the f/w this means that the netowrk i/f is closed. + * + **/ + +static inline void +e1000_release_hw_control(struct e1000_adapter *adapter) +{ + uint32_t ctrl_ext; + uint32_t swsm; + + /* Let firmware taken over control of h/w */ + switch (adapter->hw.mac_type) { + case e1000_82571: + case e1000_82572: + ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT); + E1000_WRITE_REG(&adapter->hw, CTRL_EXT, + ctrl_ext & ~E1000_CTRL_EXT_DRV_LOAD); + break; + case e1000_82573: + swsm = E1000_READ_REG(&adapter->hw, SWSM); + E1000_WRITE_REG(&adapter->hw, SWSM, + swsm & ~E1000_SWSM_DRV_LOAD); + default: + break; + } +} + +/** + * e1000_get_hw_control - get control of the h/w from f/w + * @adapter: address of board private structure + * + * e1000_get_hw_control sets {CTRL_EXT|FWSM}:DRV_LOAD bit. + * For ASF and Pass Through versions of f/w this means that + * the driver is loaded. For AMT version (only with 82573) + * of the f/w this means that the netowrk i/f is open. + * + **/ + +static inline void +e1000_get_hw_control(struct e1000_adapter *adapter) +{ + uint32_t ctrl_ext; + uint32_t swsm; + /* Let firmware know the driver has taken over */ + switch (adapter->hw.mac_type) { + case e1000_82571: + case e1000_82572: + ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT); + E1000_WRITE_REG(&adapter->hw, CTRL_EXT, + ctrl_ext | E1000_CTRL_EXT_DRV_LOAD); + break; + case e1000_82573: + swsm = E1000_READ_REG(&adapter->hw, SWSM); + E1000_WRITE_REG(&adapter->hw, SWSM, + swsm | E1000_SWSM_DRV_LOAD); + break; + default: + break; } } - + int e1000_up(struct e1000_adapter *adapter) { struct net_device *netdev = adapter->netdev; - int err; + int i, err; /* hardware has been reset, we need to reload some things */ /* Reset the PHY if it was previously powered down */ - if(adapter->hw.media_type == e1000_media_type_copper) { + if (adapter->hw.media_type == e1000_media_type_copper) { uint16_t mii_reg; e1000_read_phy_reg(&adapter->hw, PHY_CTRL, &mii_reg); - if(mii_reg & MII_CR_POWER_DOWN) + if (mii_reg & MII_CR_POWER_DOWN) e1000_phy_reset(&adapter->hw); } @@ -316,19 +431,26 @@ e1000_up(struct e1000_adapter *adapter) e1000_configure_tx(adapter); e1000_setup_rctl(adapter); e1000_configure_rx(adapter); - adapter->alloc_rx_buf(adapter); + /* call E1000_DESC_UNUSED which always leaves + * at least 1 descriptor unused to make sure + * next_to_use != next_to_clean */ + for (i = 0; i < adapter->num_rx_queues; i++) { + struct e1000_rx_ring *ring = &adapter->rx_ring[i]; + adapter->alloc_rx_buf(adapter, ring, + E1000_DESC_UNUSED(ring)); + } #ifdef CONFIG_PCI_MSI - if(adapter->hw.mac_type > e1000_82547_rev_2) { + if (adapter->hw.mac_type > e1000_82547_rev_2) { adapter->have_msi = TRUE; - if((err = pci_enable_msi(adapter->pdev))) { + if ((err = pci_enable_msi(adapter->pdev))) { DPRINTK(PROBE, ERR, "Unable to allocate MSI interrupt Error: %d\n", err); adapter->have_msi = FALSE; } } #endif - if((err = request_irq(adapter->pdev->irq, &e1000_intr, + if ((err = request_irq(adapter->pdev->irq, &e1000_intr, SA_SHIRQ | SA_SAMPLE_RANDOM, netdev->name, netdev))) { DPRINTK(PROBE, ERR, @@ -336,6 +458,8 @@ e1000_up(struct e1000_adapter *adapter) return err; } + adapter->tx_queue_len = netdev->tx_queue_len; + mod_timer(&adapter->watchdog_timer, jiffies); #ifdef CONFIG_E1000_NAPI @@ -350,11 +474,14 @@ void e1000_down(struct e1000_adapter *adapter) { struct net_device *netdev = adapter->netdev; + boolean_t mng_mode_enabled = (adapter->hw.mac_type >= e1000_82571) && + e1000_check_mng_mode(&adapter->hw); e1000_irq_disable(adapter); + free_irq(adapter->pdev->irq, netdev); #ifdef CONFIG_PCI_MSI - if(adapter->hw.mac_type > e1000_82547_rev_2 && + if (adapter->hw.mac_type > e1000_82547_rev_2 && adapter->have_msi == TRUE) pci_disable_msi(adapter->pdev); #endif @@ -365,22 +492,26 @@ e1000_down(struct e1000_adapter *adapter) #ifdef CONFIG_E1000_NAPI netif_poll_disable(netdev); #endif + netdev->tx_queue_len = adapter->tx_queue_len; adapter->link_speed = 0; adapter->link_duplex = 0; netif_carrier_off(netdev); netif_stop_queue(netdev); e1000_reset(adapter); - e1000_clean_tx_ring(adapter); - e1000_clean_rx_ring(adapter); - - /* If WoL is not enabled - * and management mode is not IAMT - * Power down the PHY so no link is implied when interface is down */ - if(!adapter->wol && adapter->hw.mac_type >= e1000_82540 && + e1000_clean_all_tx_rings(adapter); + e1000_clean_all_rx_rings(adapter); + + /* Power down the PHY so no link is implied when interface is down * + * The PHY cannot be powered down if any of the following is TRUE * + * (a) WoL is enabled + * (b) AMT is active + * (c) SoL/IDER session is active */ + if (!adapter->wol && adapter->hw.mac_type >= e1000_82540 && adapter->hw.media_type == e1000_media_type_copper && - !e1000_check_mng_mode(&adapter->hw) && - !(E1000_READ_REG(&adapter->hw, MANC) & E1000_MANC_SMBUS_EN)) { + !(E1000_READ_REG(&adapter->hw, MANC) & E1000_MANC_SMBUS_EN) && + !mng_mode_enabled && + !e1000_check_phy_reset_block(&adapter->hw)) { uint16_t mii_reg; e1000_read_phy_reg(&adapter->hw, PHY_CTRL, &mii_reg); mii_reg |= MII_CR_POWER_DOWN; @@ -392,10 +523,8 @@ e1000_down(struct e1000_adapter *adapter) void e1000_reset(struct e1000_adapter *adapter) { - struct net_device *netdev = adapter->netdev; uint32_t pba, manc; uint16_t fc_high_water_mark = E1000_FC_HIGH_DIFF; - uint16_t fc_low_water_mark = E1000_FC_LOW_DIFF; /* Repartition Pba for greater than 9k mtu * To take effect CTRL.RST is required. @@ -406,6 +535,11 @@ e1000_reset(struct e1000_adapter *adapter) case e1000_82547_rev_2: pba = E1000_PBA_30K; break; + case e1000_82571: + case e1000_82572: + case e1000_80003es2lan: + pba = E1000_PBA_38K; + break; case e1000_82573: pba = E1000_PBA_12K; break; @@ -414,19 +548,12 @@ e1000_reset(struct e1000_adapter *adapter) break; } - if((adapter->hw.mac_type != e1000_82573) && - (adapter->rx_buffer_len > E1000_RXBUFFER_8192)) { + if ((adapter->hw.mac_type != e1000_82573) && + (adapter->netdev->mtu > E1000_RXBUFFER_8192)) pba -= 8; /* allocate more FIFO for Tx */ - /* send an XOFF when there is enough space in the - * Rx FIFO to hold one extra full size Rx packet - */ - fc_high_water_mark = netdev->mtu + ENET_HEADER_SIZE + - ETHERNET_FCS_SIZE + 1; - fc_low_water_mark = fc_high_water_mark + 8; - } - if(adapter->hw.mac_type == e1000_82547) { + if (adapter->hw.mac_type == e1000_82547) { adapter->tx_fifo_head = 0; adapter->tx_head_addr = pba << E1000_TX_HEAD_ADDR_SHIFT; adapter->tx_fifo_size = @@ -437,19 +564,24 @@ e1000_reset(struct e1000_adapter *adapter) E1000_WRITE_REG(&adapter->hw, PBA, pba); /* flow control settings */ - adapter->hw.fc_high_water = (pba << E1000_PBA_BYTES_SHIFT) - - fc_high_water_mark; - adapter->hw.fc_low_water = (pba << E1000_PBA_BYTES_SHIFT) - - fc_low_water_mark; - adapter->hw.fc_pause_time = E1000_FC_PAUSE_TIME; + /* Set the FC high water mark to 90% of the FIFO size. + * Required to clear last 3 LSB */ + fc_high_water_mark = ((pba * 9216)/10) & 0xFFF8; + + adapter->hw.fc_high_water = fc_high_water_mark; + adapter->hw.fc_low_water = fc_high_water_mark - 8; + if (adapter->hw.mac_type == e1000_80003es2lan) + adapter->hw.fc_pause_time = 0xFFFF; + else + adapter->hw.fc_pause_time = E1000_FC_PAUSE_TIME; adapter->hw.fc_send_xon = 1; adapter->hw.fc = adapter->hw.original_fc; /* Allow time for pending master requests to run */ e1000_reset_hw(&adapter->hw); - if(adapter->hw.mac_type >= e1000_82544) + if (adapter->hw.mac_type >= e1000_82544) E1000_WRITE_REG(&adapter->hw, WUC, 0); - if(e1000_init_hw(&adapter->hw)) + if (e1000_init_hw(&adapter->hw)) DPRINTK(PROBE, ERR, "Hardware Error\n"); e1000_update_mng_vlan(adapter); /* Enable h/w to recognize an 802.1Q VLAN Ethernet packet */ @@ -483,32 +615,32 @@ e1000_probe(struct pci_dev *pdev, struct net_device *netdev; struct e1000_adapter *adapter; unsigned long mmio_start, mmio_len; - uint32_t swsm; static int cards_found = 0; + static int e1000_ksp3_port_a = 0; /* global ksp3 port a indication */ int i, err, pci_using_dac; uint16_t eeprom_data; uint16_t eeprom_apme_mask = E1000_EEPROM_APME; - if((err = pci_enable_device(pdev))) + if ((err = pci_enable_device(pdev))) return err; - if(!(err = pci_set_dma_mask(pdev, DMA_64BIT_MASK))) { + if (!(err = pci_set_dma_mask(pdev, DMA_64BIT_MASK))) { pci_using_dac = 1; } else { - if((err = pci_set_dma_mask(pdev, DMA_32BIT_MASK))) { + if ((err = pci_set_dma_mask(pdev, DMA_32BIT_MASK))) { E1000_ERR("No usable DMA configuration, aborting\n"); return err; } pci_using_dac = 0; } - if((err = pci_request_regions(pdev, e1000_driver_name))) + if ((err = pci_request_regions(pdev, e1000_driver_name))) return err; pci_set_master(pdev); netdev = alloc_etherdev(sizeof(struct e1000_adapter)); - if(!netdev) { + if (!netdev) { err = -ENOMEM; goto err_alloc_etherdev; } @@ -517,7 +649,7 @@ e1000_probe(struct pci_dev *pdev, SET_NETDEV_DEV(netdev, &pdev->dev); pci_set_drvdata(pdev, netdev); - adapter = netdev->priv; + adapter = netdev_priv(netdev); adapter->netdev = netdev; adapter->pdev = pdev; adapter->hw.back = adapter; @@ -527,15 +659,15 @@ e1000_probe(struct pci_dev *pdev, mmio_len = pci_resource_len(pdev, BAR_0); adapter->hw.hw_addr = ioremap(mmio_start, mmio_len); - if(!adapter->hw.hw_addr) { + if (!adapter->hw.hw_addr) { err = -EIO; goto err_ioremap; } - for(i = BAR_1; i <= BAR_5; i++) { - if(pci_resource_len(pdev, i) == 0) + for (i = BAR_1; i <= BAR_5; i++) { + if (pci_resource_len(pdev, i) == 0) continue; - if(pci_resource_flags(pdev, i) & IORESOURCE_IO) { + if (pci_resource_flags(pdev, i) & IORESOURCE_IO) { adapter->hw.io_base = pci_resource_start(pdev, i); break; } @@ -572,13 +704,22 @@ e1000_probe(struct pci_dev *pdev, /* setup the private structure */ - if((err = e1000_sw_init(adapter))) + if ((err = e1000_sw_init(adapter))) goto err_sw_init; - if((err = e1000_check_phy_reset_block(&adapter->hw))) + if ((err = e1000_check_phy_reset_block(&adapter->hw))) DPRINTK(PROBE, INFO, "PHY reset is blocked due to SOL/IDER session.\n"); - if(adapter->hw.mac_type >= e1000_82543) { + /* if ksp3, indicate if it's port a being setup */ + if (pdev->device == E1000_DEV_ID_82546GB_QUAD_COPPER_KSP3 && + e1000_ksp3_port_a == 0) + adapter->ksp3_port_a = 1; + e1000_ksp3_port_a++; + /* Reset for multiple KP3 adapters */ + if (e1000_ksp3_port_a == 4) + e1000_ksp3_port_a = 0; + + if (adapter->hw.mac_type >= e1000_82543) { netdev->features = NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_TX | @@ -587,16 +728,16 @@ e1000_probe(struct pci_dev *pdev, } #ifdef NETIF_F_TSO - if((adapter->hw.mac_type >= e1000_82544) && + if ((adapter->hw.mac_type >= e1000_82544) && (adapter->hw.mac_type != e1000_82547)) netdev->features |= NETIF_F_TSO; #ifdef NETIF_F_TSO_IPV6 - if(adapter->hw.mac_type > e1000_82547_rev_2) + if (adapter->hw.mac_type > e1000_82547_rev_2) netdev->features |= NETIF_F_TSO_IPV6; #endif #endif - if(pci_using_dac) + if (pci_using_dac) netdev->features |= NETIF_F_HIGHDMA; /* hard_start_xmit is safe against parallel locking */ @@ -604,14 +745,14 @@ e1000_probe(struct pci_dev *pdev, adapter->en_mng_pt = e1000_enable_mng_pass_thru(&adapter->hw); - /* before reading the EEPROM, reset the controller to + /* before reading the EEPROM, reset the controller to * put the device in a known good starting state */ - + e1000_reset_hw(&adapter->hw); /* make sure the EEPROM is good */ - if(e1000_validate_eeprom_checksum(&adapter->hw) < 0) { + if (e1000_validate_eeprom_checksum(&adapter->hw) < 0) { DPRINTK(PROBE, ERR, "The EEPROM Checksum Is Not Valid\n"); err = -EIO; goto err_eeprom; @@ -619,11 +760,12 @@ e1000_probe(struct pci_dev *pdev, /* copy the MAC address out of the EEPROM */ - if(e1000_read_mac_addr(&adapter->hw)) + if (e1000_read_mac_addr(&adapter->hw)) DPRINTK(PROBE, ERR, "EEPROM Read Error\n"); memcpy(netdev->dev_addr, adapter->hw.mac_addr, netdev->addr_len); + memcpy(netdev->perm_addr, adapter->hw.mac_addr, netdev->addr_len); - if(!is_valid_ether_addr(netdev->dev_addr)) { + if (!is_valid_ether_addr(netdev->perm_addr)) { DPRINTK(PROBE, ERR, "Invalid MAC Address\n"); err = -EIO; goto err_eeprom; @@ -648,8 +790,8 @@ e1000_probe(struct pci_dev *pdev, adapter->phy_info_timer.function = &e1000_update_phy_info; adapter->phy_info_timer.data = (unsigned long) adapter; - INIT_WORK(&adapter->tx_timeout_task, - (void (*)(void *))e1000_tx_timeout_task, netdev); + INIT_WORK(&adapter->reset_task, + (void (*)(void *))e1000_reset_task, netdev); /* we're going to reset, so assume we have no link for now */ @@ -663,7 +805,7 @@ e1000_probe(struct pci_dev *pdev, * enable the ACPI Magic Packet filter */ - switch(adapter->hw.mac_type) { + switch (adapter->hw.mac_type) { case e1000_82542_rev2_0: case e1000_82542_rev2_1: case e1000_82543: @@ -675,8 +817,9 @@ e1000_probe(struct pci_dev *pdev, break; case e1000_82546: case e1000_82546_rev_3: - if((E1000_READ_REG(&adapter->hw, STATUS) & E1000_STATUS_FUNC_1) - && (adapter->hw.media_type == e1000_media_type_copper)) { + case e1000_82571: + case e1000_80003es2lan: + if (E1000_READ_REG(&adapter->hw, STATUS) & E1000_STATUS_FUNC_1){ e1000_read_eeprom(&adapter->hw, EEPROM_INIT_CONTROL3_PORT_B, 1, &eeprom_data); break; @@ -687,25 +830,42 @@ e1000_probe(struct pci_dev *pdev, EEPROM_INIT_CONTROL3_PORT_A, 1, &eeprom_data); break; } - if(eeprom_data & eeprom_apme_mask) + if (eeprom_data & eeprom_apme_mask) adapter->wol |= E1000_WUFC_MAG; + /* print bus type/speed/width info */ + { + struct e1000_hw *hw = &adapter->hw; + DPRINTK(PROBE, INFO, "(PCI%s:%s:%s) ", + ((hw->bus_type == e1000_bus_type_pcix) ? "-X" : + (hw->bus_type == e1000_bus_type_pci_express ? " Express":"")), + ((hw->bus_speed == e1000_bus_speed_2500) ? "2.5Gb/s" : + (hw->bus_speed == e1000_bus_speed_133) ? "133MHz" : + (hw->bus_speed == e1000_bus_speed_120) ? "120MHz" : + (hw->bus_speed == e1000_bus_speed_100) ? "100MHz" : + (hw->bus_speed == e1000_bus_speed_66) ? "66MHz" : "33MHz"), + ((hw->bus_width == e1000_bus_width_64) ? "64-bit" : + (hw->bus_width == e1000_bus_width_pciex_4) ? "Width x4" : + (hw->bus_width == e1000_bus_width_pciex_1) ? "Width x1" : + "32-bit")); + } + + for (i = 0; i < 6; i++) + printk("%2.2x%c", netdev->dev_addr[i], i == 5 ? '\n' : ':'); + /* reset the hardware with the new settings */ e1000_reset(adapter); - /* Let firmware know the driver has taken over */ - switch(adapter->hw.mac_type) { - case e1000_82573: - swsm = E1000_READ_REG(&adapter->hw, SWSM); - E1000_WRITE_REG(&adapter->hw, SWSM, - swsm | E1000_SWSM_DRV_LOAD); - break; - default: - break; - } + /* If the controller is 82573 and f/w is AMT, do not set + * DRV_LOAD until the interface is up. For all other cases, + * let the f/w know that the h/w is now under the control + * of the driver. */ + if (adapter->hw.mac_type != e1000_82573 || + !e1000_check_mng_mode(&adapter->hw)) + e1000_get_hw_control(adapter); strcpy(netdev->name, "eth%d"); - if((err = register_netdev(netdev))) + if ((err = register_netdev(netdev))) goto err_register; DPRINTK(PROBE, INFO, "Intel(R) PRO/1000 Network Connection\n"); @@ -738,36 +898,42 @@ static void __devexit e1000_remove(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); - struct e1000_adapter *adapter = netdev->priv; - uint32_t manc, swsm; + struct e1000_adapter *adapter = netdev_priv(netdev); + uint32_t manc; +#ifdef CONFIG_E1000_NAPI + int i; +#endif flush_scheduled_work(); - if(adapter->hw.mac_type >= e1000_82540 && + if (adapter->hw.mac_type >= e1000_82540 && adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); - if(manc & E1000_MANC_SMBUS_EN) { + if (manc & E1000_MANC_SMBUS_EN) { manc |= E1000_MANC_ARP_EN; E1000_WRITE_REG(&adapter->hw, MANC, manc); } } - switch(adapter->hw.mac_type) { - case e1000_82573: - swsm = E1000_READ_REG(&adapter->hw, SWSM); - E1000_WRITE_REG(&adapter->hw, SWSM, - swsm & ~E1000_SWSM_DRV_LOAD); - break; - - default: - break; - } + /* Release control of h/w to f/w. If f/w is AMT enabled, this + * would have already happened in close and is redundant. */ + e1000_release_hw_control(adapter); unregister_netdev(netdev); +#ifdef CONFIG_E1000_NAPI + for (i = 0; i < adapter->num_rx_queues; i++) + dev_put(&adapter->polling_netdev[i]); +#endif - if(!e1000_check_phy_reset_block(&adapter->hw)) + if (!e1000_check_phy_reset_block(&adapter->hw)) e1000_phy_hw_reset(&adapter->hw); + kfree(adapter->tx_ring); + kfree(adapter->rx_ring); +#ifdef CONFIG_E1000_NAPI + kfree(adapter->polling_netdev); +#endif + iounmap(adapter->hw.hw_addr); pci_release_regions(pdev); @@ -791,6 +957,9 @@ e1000_sw_init(struct e1000_adapter *adapter) struct e1000_hw *hw = &adapter->hw; struct net_device *netdev = adapter->netdev; struct pci_dev *pdev = adapter->pdev; +#ifdef CONFIG_E1000_NAPI + int i; +#endif /* PCI config space info */ @@ -811,19 +980,19 @@ e1000_sw_init(struct e1000_adapter *adapter) /* identify the MAC */ - if(e1000_set_mac_type(hw)) { + if (e1000_set_mac_type(hw)) { DPRINTK(PROBE, ERR, "Unknown MAC Type\n"); return -EIO; } /* initialize eeprom parameters */ - if(e1000_init_eeprom_params(hw)) { + if (e1000_init_eeprom_params(hw)) { E1000_ERR("EEPROM initialization failed\n"); return -EIO; } - switch(hw->mac_type) { + switch (hw->mac_type) { default: break; case e1000_82541: @@ -842,19 +1011,79 @@ e1000_sw_init(struct e1000_adapter *adapter) /* Copper options */ - if(hw->media_type == e1000_media_type_copper) { + if (hw->media_type == e1000_media_type_copper) { hw->mdix = AUTO_ALL_MODES; hw->disable_polarity_correction = FALSE; hw->master_slave = E1000_MASTER_SLAVE; } + adapter->num_tx_queues = 1; + adapter->num_rx_queues = 1; + + if (e1000_alloc_queues(adapter)) { + DPRINTK(PROBE, ERR, "Unable to allocate memory for queues\n"); + return -ENOMEM; + } + +#ifdef CONFIG_E1000_NAPI + for (i = 0; i < adapter->num_rx_queues; i++) { + adapter->polling_netdev[i].priv = adapter; + adapter->polling_netdev[i].poll = &e1000_clean; + adapter->polling_netdev[i].weight = 64; + dev_hold(&adapter->polling_netdev[i]); + set_bit(__LINK_STATE_START, &adapter->polling_netdev[i].state); + } + spin_lock_init(&adapter->tx_queue_lock); +#endif + atomic_set(&adapter->irq_sem, 1); spin_lock_init(&adapter->stats_lock); - spin_lock_init(&adapter->tx_lock); return 0; } +/** + * e1000_alloc_queues - Allocate memory for all rings + * @adapter: board private structure to initialize + * + * We allocate one ring per queue at run-time since we don't know the + * number of queues at compile-time. The polling_netdev array is + * intended for Multiqueue, but should work fine with a single queue. + **/ + +static int __devinit +e1000_alloc_queues(struct e1000_adapter *adapter) +{ + int size; + + size = sizeof(struct e1000_tx_ring) * adapter->num_tx_queues; + adapter->tx_ring = kmalloc(size, GFP_KERNEL); + if (!adapter->tx_ring) + return -ENOMEM; + memset(adapter->tx_ring, 0, size); + + size = sizeof(struct e1000_rx_ring) * adapter->num_rx_queues; + adapter->rx_ring = kmalloc(size, GFP_KERNEL); + if (!adapter->rx_ring) { + kfree(adapter->tx_ring); + return -ENOMEM; + } + memset(adapter->rx_ring, 0, size); + +#ifdef CONFIG_E1000_NAPI + size = sizeof(struct net_device) * adapter->num_rx_queues; + adapter->polling_netdev = kmalloc(size, GFP_KERNEL); + if (!adapter->polling_netdev) { + kfree(adapter->tx_ring); + kfree(adapter->rx_ring); + return -ENOMEM; + } + memset(adapter->polling_netdev, 0, size); +#endif + + return E1000_SUCCESS; +} + /** * e1000_open - Called when a network interface is made active * @netdev: network interface device structure @@ -871,33 +1100,39 @@ e1000_sw_init(struct e1000_adapter *adapter) static int e1000_open(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); int err; /* allocate transmit descriptors */ - if((err = e1000_setup_tx_resources(adapter))) + if ((err = e1000_setup_all_tx_resources(adapter))) goto err_setup_tx; /* allocate receive descriptors */ - if((err = e1000_setup_rx_resources(adapter))) + if ((err = e1000_setup_all_rx_resources(adapter))) goto err_setup_rx; - if((err = e1000_up(adapter))) + if ((err = e1000_up(adapter))) goto err_up; adapter->mng_vlan_id = E1000_MNG_VLAN_NONE; - if((adapter->hw.mng_cookie.status & + if ((adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT)) { e1000_update_mng_vlan(adapter); } + /* If AMT is enabled, let the firmware know that the network + * interface is now open */ + if (adapter->hw.mac_type == e1000_82573 && + e1000_check_mng_mode(&adapter->hw)) + e1000_get_hw_control(adapter); + return E1000_SUCCESS; err_up: - e1000_free_rx_resources(adapter); + e1000_free_all_rx_resources(adapter); err_setup_rx: - e1000_free_tx_resources(adapter); + e1000_free_all_tx_resources(adapter); err_setup_tx: e1000_reset(adapter); @@ -919,17 +1154,24 @@ err_setup_tx: static int e1000_close(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); e1000_down(adapter); - e1000_free_tx_resources(adapter); - e1000_free_rx_resources(adapter); + e1000_free_all_tx_resources(adapter); + e1000_free_all_rx_resources(adapter); - if((adapter->hw.mng_cookie.status & + if ((adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT)) { e1000_vlan_rx_kill_vid(netdev, adapter->mng_vlan_id); } + + /* If AMT is enabled, let the firmware know that the network + * interface is now closed */ + if (adapter->hw.mac_type == e1000_82573 && + e1000_check_mng_mode(&adapter->hw)) + e1000_release_hw_control(adapter); + return 0; } @@ -959,20 +1201,22 @@ e1000_check_64k_bound(struct e1000_adapter *adapter, /** * e1000_setup_tx_resources - allocate Tx resources (Descriptors) * @adapter: board private structure + * @txdr: tx descriptor ring (for a specific queue) to setup * * Return 0 on success, negative on failure **/ -int -e1000_setup_tx_resources(struct e1000_adapter *adapter) +static int +e1000_setup_tx_resources(struct e1000_adapter *adapter, + struct e1000_tx_ring *txdr) { - struct e1000_desc_ring *txdr = &adapter->tx_ring; struct pci_dev *pdev = adapter->pdev; int size; size = sizeof(struct e1000_buffer) * txdr->count; - txdr->buffer_info = vmalloc(size); - if(!txdr->buffer_info) { + + txdr->buffer_info = vmalloc_node(size, pcibus_to_node(pdev->bus)); + if (!txdr->buffer_info) { DPRINTK(PROBE, ERR, "Unable to allocate memory for the transmit descriptor ring\n"); return -ENOMEM; @@ -985,7 +1229,7 @@ e1000_setup_tx_resources(struct e1000_adapter *adapter) E1000_ROUNDUP(txdr->size, 4096); txdr->desc = pci_alloc_consistent(pdev, txdr->size, &txdr->dma); - if(!txdr->desc) { + if (!txdr->desc) { setup_tx_desc_die: vfree(txdr->buffer_info); DPRINTK(PROBE, ERR, @@ -1001,8 +1245,8 @@ setup_tx_desc_die: "at %p\n", txdr->size, txdr->desc); /* Try again, without freeing the previous */ txdr->desc = pci_alloc_consistent(pdev, txdr->size, &txdr->dma); - if(!txdr->desc) { /* Failed allocation, critical failure */ + if (!txdr->desc) { pci_free_consistent(pdev, txdr->size, olddesc, olddma); goto setup_tx_desc_die; } @@ -1026,10 +1270,40 @@ setup_tx_desc_die: txdr->next_to_use = 0; txdr->next_to_clean = 0; + spin_lock_init(&txdr->tx_lock); return 0; } +/** + * e1000_setup_all_tx_resources - wrapper to allocate Tx resources + * (Descriptors) for all queues + * @adapter: board private structure + * + * If this function returns with an error, then it's possible one or + * more of the rings is populated (while the rest are not). It is the + * callers duty to clean those orphaned rings. + * + * Return 0 on success, negative on failure + **/ + +int +e1000_setup_all_tx_resources(struct e1000_adapter *adapter) +{ + int i, err = 0; + + for (i = 0; i < adapter->num_tx_queues; i++) { + err = e1000_setup_tx_resources(adapter, &adapter->tx_ring[i]); + if (err) { + DPRINTK(PROBE, ERR, + "Allocation for Tx Queue %u failed\n", i); + break; + } + } + + return err; +} + /** * e1000_configure_tx - Configure 8254x Transmit Unit after Reset * @adapter: board private structure @@ -1040,91 +1314,137 @@ setup_tx_desc_die: static void e1000_configure_tx(struct e1000_adapter *adapter) { - uint64_t tdba = adapter->tx_ring.dma; - uint32_t tdlen = adapter->tx_ring.count * sizeof(struct e1000_tx_desc); - uint32_t tctl, tipg; - - E1000_WRITE_REG(&adapter->hw, TDBAL, (tdba & 0x00000000ffffffffULL)); - E1000_WRITE_REG(&adapter->hw, TDBAH, (tdba >> 32)); - - E1000_WRITE_REG(&adapter->hw, TDLEN, tdlen); + uint64_t tdba; + struct e1000_hw *hw = &adapter->hw; + uint32_t tdlen, tctl, tipg, tarc; + uint32_t ipgr1, ipgr2; /* Setup the HW Tx Head and Tail descriptor pointers */ - E1000_WRITE_REG(&adapter->hw, TDH, 0); - E1000_WRITE_REG(&adapter->hw, TDT, 0); + switch (adapter->num_tx_queues) { + case 1: + default: + tdba = adapter->tx_ring[0].dma; + tdlen = adapter->tx_ring[0].count * + sizeof(struct e1000_tx_desc); + E1000_WRITE_REG(hw, TDBAL, (tdba & 0x00000000ffffffffULL)); + E1000_WRITE_REG(hw, TDBAH, (tdba >> 32)); + E1000_WRITE_REG(hw, TDLEN, tdlen); + E1000_WRITE_REG(hw, TDH, 0); + E1000_WRITE_REG(hw, TDT, 0); + adapter->tx_ring[0].tdh = E1000_TDH; + adapter->tx_ring[0].tdt = E1000_TDT; + break; + } /* Set the default values for the Tx Inter Packet Gap timer */ - switch (adapter->hw.mac_type) { + if (hw->media_type == e1000_media_type_fiber || + hw->media_type == e1000_media_type_internal_serdes) + tipg = DEFAULT_82543_TIPG_IPGT_FIBER; + else + tipg = DEFAULT_82543_TIPG_IPGT_COPPER; + + switch (hw->mac_type) { case e1000_82542_rev2_0: case e1000_82542_rev2_1: tipg = DEFAULT_82542_TIPG_IPGT; - tipg |= DEFAULT_82542_TIPG_IPGR1 << E1000_TIPG_IPGR1_SHIFT; - tipg |= DEFAULT_82542_TIPG_IPGR2 << E1000_TIPG_IPGR2_SHIFT; + ipgr1 = DEFAULT_82542_TIPG_IPGR1; + ipgr2 = DEFAULT_82542_TIPG_IPGR2; + break; + case e1000_80003es2lan: + ipgr1 = DEFAULT_82543_TIPG_IPGR1; + ipgr2 = DEFAULT_80003ES2LAN_TIPG_IPGR2; break; default: - if(adapter->hw.media_type == e1000_media_type_fiber || - adapter->hw.media_type == e1000_media_type_internal_serdes) - tipg = DEFAULT_82543_TIPG_IPGT_FIBER; - else - tipg = DEFAULT_82543_TIPG_IPGT_COPPER; - tipg |= DEFAULT_82543_TIPG_IPGR1 << E1000_TIPG_IPGR1_SHIFT; - tipg |= DEFAULT_82543_TIPG_IPGR2 << E1000_TIPG_IPGR2_SHIFT; + ipgr1 = DEFAULT_82543_TIPG_IPGR1; + ipgr2 = DEFAULT_82543_TIPG_IPGR2; + break; } - E1000_WRITE_REG(&adapter->hw, TIPG, tipg); + tipg |= ipgr1 << E1000_TIPG_IPGR1_SHIFT; + tipg |= ipgr2 << E1000_TIPG_IPGR2_SHIFT; + E1000_WRITE_REG(hw, TIPG, tipg); /* Set the Tx Interrupt Delay register */ - E1000_WRITE_REG(&adapter->hw, TIDV, adapter->tx_int_delay); - if(adapter->hw.mac_type >= e1000_82540) - E1000_WRITE_REG(&adapter->hw, TADV, adapter->tx_abs_int_delay); + E1000_WRITE_REG(hw, TIDV, adapter->tx_int_delay); + if (hw->mac_type >= e1000_82540) + E1000_WRITE_REG(hw, TADV, adapter->tx_abs_int_delay); /* Program the Transmit Control Register */ - tctl = E1000_READ_REG(&adapter->hw, TCTL); + tctl = E1000_READ_REG(hw, TCTL); tctl &= ~E1000_TCTL_CT; - tctl |= E1000_TCTL_EN | E1000_TCTL_PSP | + tctl |= E1000_TCTL_PSP | E1000_TCTL_RTLC | (E1000_COLLISION_THRESHOLD << E1000_CT_SHIFT); - E1000_WRITE_REG(&adapter->hw, TCTL, tctl); +#ifdef DISABLE_MULR + /* disable Multiple Reads for debugging */ + tctl &= ~E1000_TCTL_MULR; +#endif - e1000_config_collision_dist(&adapter->hw); + if (hw->mac_type == e1000_82571 || hw->mac_type == e1000_82572) { + tarc = E1000_READ_REG(hw, TARC0); + tarc |= ((1 << 25) | (1 << 21)); + E1000_WRITE_REG(hw, TARC0, tarc); + tarc = E1000_READ_REG(hw, TARC1); + tarc |= (1 << 25); + if (tctl & E1000_TCTL_MULR) + tarc &= ~(1 << 28); + else + tarc |= (1 << 28); + E1000_WRITE_REG(hw, TARC1, tarc); + } else if (hw->mac_type == e1000_80003es2lan) { + tarc = E1000_READ_REG(hw, TARC0); + tarc |= 1; + if (hw->media_type == e1000_media_type_internal_serdes) + tarc |= (1 << 20); + E1000_WRITE_REG(hw, TARC0, tarc); + tarc = E1000_READ_REG(hw, TARC1); + tarc |= 1; + E1000_WRITE_REG(hw, TARC1, tarc); + } + + e1000_config_collision_dist(hw); /* Setup Transmit Descriptor Settings for eop descriptor */ adapter->txd_cmd = E1000_TXD_CMD_IDE | E1000_TXD_CMD_EOP | E1000_TXD_CMD_IFCS; - if(adapter->hw.mac_type < e1000_82543) + if (hw->mac_type < e1000_82543) adapter->txd_cmd |= E1000_TXD_CMD_RPS; else adapter->txd_cmd |= E1000_TXD_CMD_RS; /* Cache if we're 82544 running in PCI-X because we'll * need this to apply a workaround later in the send path. */ - if(adapter->hw.mac_type == e1000_82544 && - adapter->hw.bus_type == e1000_bus_type_pcix) + if (hw->mac_type == e1000_82544 && + hw->bus_type == e1000_bus_type_pcix) adapter->pcix_82544 = 1; + + E1000_WRITE_REG(hw, TCTL, tctl); + } /** * e1000_setup_rx_resources - allocate Rx resources (Descriptors) * @adapter: board private structure + * @rxdr: rx descriptor ring (for a specific queue) to setup * * Returns 0 on success, negative on failure **/ -int -e1000_setup_rx_resources(struct e1000_adapter *adapter) +static int +e1000_setup_rx_resources(struct e1000_adapter *adapter, + struct e1000_rx_ring *rxdr) { - struct e1000_desc_ring *rxdr = &adapter->rx_ring; struct pci_dev *pdev = adapter->pdev; int size, desc_len; size = sizeof(struct e1000_buffer) * rxdr->count; - rxdr->buffer_info = vmalloc(size); - if(!rxdr->buffer_info) { + rxdr->buffer_info = vmalloc_node(size, pcibus_to_node(pdev->bus)); + if (!rxdr->buffer_info) { DPRINTK(PROBE, ERR, "Unable to allocate memory for the receive descriptor ring\n"); return -ENOMEM; @@ -1133,7 +1453,7 @@ e1000_setup_rx_resources(struct e1000_adapter *adapter) size = sizeof(struct e1000_ps_page) * rxdr->count; rxdr->ps_page = kmalloc(size, GFP_KERNEL); - if(!rxdr->ps_page) { + if (!rxdr->ps_page) { vfree(rxdr->buffer_info); DPRINTK(PROBE, ERR, "Unable to allocate memory for the receive descriptor ring\n"); @@ -1143,7 +1463,7 @@ e1000_setup_rx_resources(struct e1000_adapter *adapter) size = sizeof(struct e1000_ps_page_dma) * rxdr->count; rxdr->ps_page_dma = kmalloc(size, GFP_KERNEL); - if(!rxdr->ps_page_dma) { + if (!rxdr->ps_page_dma) { vfree(rxdr->buffer_info); kfree(rxdr->ps_page); DPRINTK(PROBE, ERR, @@ -1152,7 +1472,7 @@ e1000_setup_rx_resources(struct e1000_adapter *adapter) } memset(rxdr->ps_page_dma, 0, size); - if(adapter->hw.mac_type <= e1000_82547_rev_2) + if (adapter->hw.mac_type <= e1000_82547_rev_2) desc_len = sizeof(struct e1000_rx_desc); else desc_len = sizeof(union e1000_rx_desc_packet_split); @@ -1164,13 +1484,13 @@ e1000_setup_rx_resources(struct e1000_adapter *adapter) rxdr->desc = pci_alloc_consistent(pdev, rxdr->size, &rxdr->dma); - if(!rxdr->desc) { + if (!rxdr->desc) { + DPRINTK(PROBE, ERR, + "Unable to allocate memory for the receive descriptor ring\n"); setup_rx_desc_die: vfree(rxdr->buffer_info); kfree(rxdr->ps_page); kfree(rxdr->ps_page_dma); - DPRINTK(PROBE, ERR, - "Unable to allocate memory for the receive descriptor ring\n"); return -ENOMEM; } @@ -1182,9 +1502,12 @@ setup_rx_desc_die: "at %p\n", rxdr->size, rxdr->desc); /* Try again, without freeing the previous */ rxdr->desc = pci_alloc_consistent(pdev, rxdr->size, &rxdr->dma); - if(!rxdr->desc) { /* Failed allocation, critical failure */ + if (!rxdr->desc) { pci_free_consistent(pdev, rxdr->size, olddesc, olddma); + DPRINTK(PROBE, ERR, + "Unable to allocate memory " + "for the receive descriptor ring\n"); goto setup_rx_desc_die; } @@ -1196,10 +1519,7 @@ setup_rx_desc_die: DPRINTK(PROBE, ERR, "Unable to allocate aligned memory " "for the receive descriptor ring\n"); - vfree(rxdr->buffer_info); - kfree(rxdr->ps_page); - kfree(rxdr->ps_page_dma); - return -ENOMEM; + goto setup_rx_desc_die; } else { /* Free old allocation, new allocation was successful */ pci_free_consistent(pdev, rxdr->size, olddesc, olddma); @@ -1213,16 +1533,49 @@ setup_rx_desc_die: return 0; } +/** + * e1000_setup_all_rx_resources - wrapper to allocate Rx resources + * (Descriptors) for all queues + * @adapter: board private structure + * + * If this function returns with an error, then it's possible one or + * more of the rings is populated (while the rest are not). It is the + * callers duty to clean those orphaned rings. + * + * Return 0 on success, negative on failure + **/ + +int +e1000_setup_all_rx_resources(struct e1000_adapter *adapter) +{ + int i, err = 0; + + for (i = 0; i < adapter->num_rx_queues; i++) { + err = e1000_setup_rx_resources(adapter, &adapter->rx_ring[i]); + if (err) { + DPRINTK(PROBE, ERR, + "Allocation for Rx Queue %u failed\n", i); + break; + } + } + + return err; +} + /** * e1000_setup_rctl - configure the receive control registers * @adapter: Board private structure **/ - +#define PAGE_USE_COUNT(S) (((S) >> PAGE_SHIFT) + \ + (((S) & (PAGE_SIZE - 1)) ? 1 : 0)) static void e1000_setup_rctl(struct e1000_adapter *adapter) { uint32_t rctl, rfctl; uint32_t psrctl = 0; +#ifndef CONFIG_E1000_DISABLE_PACKET_SPLIT + uint32_t pages = 0; +#endif rctl = E1000_READ_REG(&adapter->hw, RCTL); @@ -1232,7 +1585,10 @@ e1000_setup_rctl(struct e1000_adapter *adapter) E1000_RCTL_LBM_NO | E1000_RCTL_RDMTS_HALF | (adapter->hw.mc_filter_type << E1000_RCTL_MO_SHIFT); - if(adapter->hw.tbi_compatibility_on == 1) + if (adapter->hw.mac_type > e1000_82543) + rctl |= E1000_RCTL_SECRC; + + if (adapter->hw.tbi_compatibility_on == 1) rctl |= E1000_RCTL_SBP; else rctl &= ~E1000_RCTL_SBP; @@ -1243,7 +1599,7 @@ e1000_setup_rctl(struct e1000_adapter *adapter) rctl |= E1000_RCTL_LPE; /* Setup buffer sizes */ - if(adapter->hw.mac_type == e1000_82573) { + if (adapter->hw.mac_type >= e1000_82571) { /* We can now specify buffers in 1K increments. * BSIZE and BSEX are ignored in this case. */ rctl |= adapter->rx_buffer_len << 0x11; @@ -1268,7 +1624,7 @@ e1000_setup_rctl(struct e1000_adapter *adapter) } } -#ifdef CONFIG_E1000_PACKET_SPLIT +#ifndef CONFIG_E1000_DISABLE_PACKET_SPLIT /* 82571 and greater support packet-split where the protocol * header is placed in skb->data and the packet data is * placed in pages hanging off of skb_shinfo(skb)->nr_frags. @@ -1276,11 +1632,14 @@ e1000_setup_rctl(struct e1000_adapter *adapter) * followed by the page buffers. Therefore, skb->data is * sized to hold the largest protocol header. */ - adapter->rx_ps = (adapter->hw.mac_type > e1000_82547_rev_2) - && (adapter->netdev->mtu - < ((3 * PAGE_SIZE) + adapter->rx_ps_bsize0)); + pages = PAGE_USE_COUNT(adapter->netdev->mtu); + if ((adapter->hw.mac_type > e1000_82547_rev_2) && (pages <= 3) && + PAGE_SIZE <= 16384) + adapter->rx_ps_pages = pages; + else + adapter->rx_ps_pages = 0; #endif - if(adapter->rx_ps) { + if (adapter->rx_ps_pages) { /* Configure extra packet-split registers */ rfctl = E1000_READ_REG(&adapter->hw, RFCTL); rfctl |= E1000_RFCTL_EXTEN; @@ -1289,15 +1648,22 @@ e1000_setup_rctl(struct e1000_adapter *adapter) E1000_WRITE_REG(&adapter->hw, RFCTL, rfctl); rctl |= E1000_RCTL_DTYP_PS | E1000_RCTL_SECRC; - + psrctl |= adapter->rx_ps_bsize0 >> E1000_PSRCTL_BSIZE0_SHIFT; - psrctl |= PAGE_SIZE >> - E1000_PSRCTL_BSIZE1_SHIFT; - psrctl |= PAGE_SIZE << - E1000_PSRCTL_BSIZE2_SHIFT; - psrctl |= PAGE_SIZE << - E1000_PSRCTL_BSIZE3_SHIFT; + + switch (adapter->rx_ps_pages) { + case 3: + psrctl |= PAGE_SIZE << + E1000_PSRCTL_BSIZE3_SHIFT; + case 2: + psrctl |= PAGE_SIZE << + E1000_PSRCTL_BSIZE2_SHIFT; + case 1: + psrctl |= PAGE_SIZE >> + E1000_PSRCTL_BSIZE1_SHIFT; + break; + } E1000_WRITE_REG(&adapter->hw, PSRCTL, psrctl); } @@ -1315,131 +1681,164 @@ e1000_setup_rctl(struct e1000_adapter *adapter) static void e1000_configure_rx(struct e1000_adapter *adapter) { - uint64_t rdba = adapter->rx_ring.dma; - uint32_t rdlen, rctl, rxcsum; + uint64_t rdba; + struct e1000_hw *hw = &adapter->hw; + uint32_t rdlen, rctl, rxcsum, ctrl_ext; - if(adapter->rx_ps) { - rdlen = adapter->rx_ring.count * + if (adapter->rx_ps_pages) { + /* this is a 32 byte descriptor */ + rdlen = adapter->rx_ring[0].count * sizeof(union e1000_rx_desc_packet_split); adapter->clean_rx = e1000_clean_rx_irq_ps; adapter->alloc_rx_buf = e1000_alloc_rx_buffers_ps; } else { - rdlen = adapter->rx_ring.count * sizeof(struct e1000_rx_desc); + rdlen = adapter->rx_ring[0].count * + sizeof(struct e1000_rx_desc); adapter->clean_rx = e1000_clean_rx_irq; adapter->alloc_rx_buf = e1000_alloc_rx_buffers; } /* disable receives while setting up the descriptors */ - rctl = E1000_READ_REG(&adapter->hw, RCTL); - E1000_WRITE_REG(&adapter->hw, RCTL, rctl & ~E1000_RCTL_EN); + rctl = E1000_READ_REG(hw, RCTL); + E1000_WRITE_REG(hw, RCTL, rctl & ~E1000_RCTL_EN); /* set the Receive Delay Timer Register */ - E1000_WRITE_REG(&adapter->hw, RDTR, adapter->rx_int_delay); + E1000_WRITE_REG(hw, RDTR, adapter->rx_int_delay); - if(adapter->hw.mac_type >= e1000_82540) { - E1000_WRITE_REG(&adapter->hw, RADV, adapter->rx_abs_int_delay); - if(adapter->itr > 1) - E1000_WRITE_REG(&adapter->hw, ITR, + if (hw->mac_type >= e1000_82540) { + E1000_WRITE_REG(hw, RADV, adapter->rx_abs_int_delay); + if (adapter->itr > 1) + E1000_WRITE_REG(hw, ITR, 1000000000 / (adapter->itr * 256)); } - /* Setup the Base and Length of the Rx Descriptor Ring */ - E1000_WRITE_REG(&adapter->hw, RDBAL, (rdba & 0x00000000ffffffffULL)); - E1000_WRITE_REG(&adapter->hw, RDBAH, (rdba >> 32)); - - E1000_WRITE_REG(&adapter->hw, RDLEN, rdlen); + if (hw->mac_type >= e1000_82571) { + ctrl_ext = E1000_READ_REG(hw, CTRL_EXT); + /* Reset delay timers after every interrupt */ + ctrl_ext |= E1000_CTRL_EXT_CANC; +#ifdef CONFIG_E1000_NAPI + /* Auto-Mask interrupts upon ICR read. */ + ctrl_ext |= E1000_CTRL_EXT_IAME; +#endif + E1000_WRITE_REG(hw, CTRL_EXT, ctrl_ext); + E1000_WRITE_REG(hw, IAM, ~0); + E1000_WRITE_FLUSH(hw); + } - /* Setup the HW Rx Head and Tail Descriptor Pointers */ - E1000_WRITE_REG(&adapter->hw, RDH, 0); - E1000_WRITE_REG(&adapter->hw, RDT, 0); + /* Setup the HW Rx Head and Tail Descriptor Pointers and + * the Base and Length of the Rx Descriptor Ring */ + switch (adapter->num_rx_queues) { + case 1: + default: + rdba = adapter->rx_ring[0].dma; + E1000_WRITE_REG(hw, RDBAL, (rdba & 0x00000000ffffffffULL)); + E1000_WRITE_REG(hw, RDBAH, (rdba >> 32)); + E1000_WRITE_REG(hw, RDLEN, rdlen); + E1000_WRITE_REG(hw, RDH, 0); + E1000_WRITE_REG(hw, RDT, 0); + adapter->rx_ring[0].rdh = E1000_RDH; + adapter->rx_ring[0].rdt = E1000_RDT; + break; + } /* Enable 82543 Receive Checksum Offload for TCP and UDP */ - if(adapter->hw.mac_type >= e1000_82543) { - rxcsum = E1000_READ_REG(&adapter->hw, RXCSUM); - if(adapter->rx_csum == TRUE) { + if (hw->mac_type >= e1000_82543) { + rxcsum = E1000_READ_REG(hw, RXCSUM); + if (adapter->rx_csum == TRUE) { rxcsum |= E1000_RXCSUM_TUOFL; - /* Enable 82573 IPv4 payload checksum for UDP fragments + /* Enable 82571 IPv4 payload checksum for UDP fragments * Must be used in conjunction with packet-split. */ - if((adapter->hw.mac_type > e1000_82547_rev_2) && - (adapter->rx_ps)) { + if ((hw->mac_type >= e1000_82571) && + (adapter->rx_ps_pages)) { rxcsum |= E1000_RXCSUM_IPPCSE; } } else { rxcsum &= ~E1000_RXCSUM_TUOFL; /* don't need to clear IPPCSE as it defaults to 0 */ } - E1000_WRITE_REG(&adapter->hw, RXCSUM, rxcsum); + E1000_WRITE_REG(hw, RXCSUM, rxcsum); } - if (adapter->hw.mac_type == e1000_82573) - E1000_WRITE_REG(&adapter->hw, ERT, 0x0100); + if (hw->mac_type == e1000_82573) + E1000_WRITE_REG(hw, ERT, 0x0100); /* Enable Receives */ - E1000_WRITE_REG(&adapter->hw, RCTL, rctl); + E1000_WRITE_REG(hw, RCTL, rctl); } /** - * e1000_free_tx_resources - Free Tx Resources + * e1000_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 -e1000_free_tx_resources(struct e1000_adapter *adapter) +static void +e1000_free_tx_resources(struct e1000_adapter *adapter, + struct e1000_tx_ring *tx_ring) { struct pci_dev *pdev = adapter->pdev; - e1000_clean_tx_ring(adapter); + e1000_clean_tx_ring(adapter, tx_ring); - vfree(adapter->tx_ring.buffer_info); - adapter->tx_ring.buffer_info = NULL; + vfree(tx_ring->buffer_info); + tx_ring->buffer_info = NULL; - pci_free_consistent(pdev, adapter->tx_ring.size, - adapter->tx_ring.desc, adapter->tx_ring.dma); + pci_free_consistent(pdev, tx_ring->size, tx_ring->desc, tx_ring->dma); + + tx_ring->desc = NULL; +} + +/** + * e1000_free_all_tx_resources - Free Tx Resources for All Queues + * @adapter: board private structure + * + * Free all transmit software resources + **/ + +void +e1000_free_all_tx_resources(struct e1000_adapter *adapter) +{ + int i; - adapter->tx_ring.desc = NULL; + for (i = 0; i < adapter->num_tx_queues; i++) + e1000_free_tx_resources(adapter, &adapter->tx_ring[i]); } static inline void e1000_unmap_and_free_tx_resource(struct e1000_adapter *adapter, struct e1000_buffer *buffer_info) { - if(buffer_info->dma) { + if (buffer_info->dma) { pci_unmap_page(adapter->pdev, buffer_info->dma, buffer_info->length, PCI_DMA_TODEVICE); - buffer_info->dma = 0; } - if(buffer_info->skb) { + if (buffer_info->skb) dev_kfree_skb_any(buffer_info->skb); - buffer_info->skb = NULL; - } + memset(buffer_info, 0, sizeof(struct e1000_buffer)); } /** * e1000_clean_tx_ring - Free Tx Buffers * @adapter: board private structure + * @tx_ring: ring to be cleaned **/ static void -e1000_clean_tx_ring(struct e1000_adapter *adapter) +e1000_clean_tx_ring(struct e1000_adapter *adapter, + struct e1000_tx_ring *tx_ring) { - struct e1000_desc_ring *tx_ring = &adapter->tx_ring; struct e1000_buffer *buffer_info; unsigned long size; unsigned int i; /* Free all the Tx ring sk_buffs */ - if (likely(adapter->previous_buffer_info.skb != NULL)) { - e1000_unmap_and_free_tx_resource(adapter, - &adapter->previous_buffer_info); - } - - for(i = 0; i < tx_ring->count; i++) { + for (i = 0; i < tx_ring->count; i++) { buffer_info = &tx_ring->buffer_info[i]; e1000_unmap_and_free_tx_resource(adapter, buffer_info); } @@ -1453,25 +1852,41 @@ e1000_clean_tx_ring(struct e1000_adapter *adapter) tx_ring->next_to_use = 0; tx_ring->next_to_clean = 0; + tx_ring->last_tx_tso = 0; - E1000_WRITE_REG(&adapter->hw, TDH, 0); - E1000_WRITE_REG(&adapter->hw, TDT, 0); + writel(0, adapter->hw.hw_addr + tx_ring->tdh); + writel(0, adapter->hw.hw_addr + tx_ring->tdt); +} + +/** + * e1000_clean_all_tx_rings - Free Tx Buffers for all queues + * @adapter: board private structure + **/ + +static void +e1000_clean_all_tx_rings(struct e1000_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_tx_queues; i++) + e1000_clean_tx_ring(adapter, &adapter->tx_ring[i]); } /** * e1000_free_rx_resources - Free Rx Resources * @adapter: board private structure + * @rx_ring: ring to clean the resources from * * Free all receive software resources **/ -void -e1000_free_rx_resources(struct e1000_adapter *adapter) +static void +e1000_free_rx_resources(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring) { - struct e1000_desc_ring *rx_ring = &adapter->rx_ring; struct pci_dev *pdev = adapter->pdev; - e1000_clean_rx_ring(adapter); + e1000_clean_rx_ring(adapter, rx_ring); vfree(rx_ring->buffer_info); rx_ring->buffer_info = NULL; @@ -1486,14 +1901,31 @@ e1000_free_rx_resources(struct e1000_adapter *adapter) } /** - * e1000_clean_rx_ring - Free Rx Buffers + * e1000_free_all_rx_resources - Free Rx Resources for All Queues * @adapter: board private structure + * + * Free all receive software resources + **/ + +void +e1000_free_all_rx_resources(struct e1000_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_rx_queues; i++) + e1000_free_rx_resources(adapter, &adapter->rx_ring[i]); +} + +/** + * e1000_clean_rx_ring - Free Rx Buffers per Queue + * @adapter: board private structure + * @rx_ring: ring to free buffers from **/ static void -e1000_clean_rx_ring(struct e1000_adapter *adapter) +e1000_clean_rx_ring(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring) { - struct e1000_desc_ring *rx_ring = &adapter->rx_ring; struct e1000_buffer *buffer_info; struct e1000_ps_page *ps_page; struct e1000_ps_page_dma *ps_page_dma; @@ -1502,12 +1934,9 @@ e1000_clean_rx_ring(struct e1000_adapter *adapter) unsigned int i, j; /* Free all the Rx ring sk_buffs */ - - for(i = 0; i < rx_ring->count; i++) { + for (i = 0; i < rx_ring->count; i++) { buffer_info = &rx_ring->buffer_info[i]; - if(buffer_info->skb) { - ps_page = &rx_ring->ps_page[i]; - ps_page_dma = &rx_ring->ps_page_dma[i]; + if (buffer_info->skb) { pci_unmap_single(pdev, buffer_info->dma, buffer_info->length, @@ -1515,16 +1944,17 @@ e1000_clean_rx_ring(struct e1000_adapter *adapter) dev_kfree_skb(buffer_info->skb); buffer_info->skb = NULL; - - for(j = 0; j < PS_PAGE_BUFFERS; j++) { - if(!ps_page->ps_page[j]) break; - pci_unmap_single(pdev, - ps_page_dma->ps_page_dma[j], - PAGE_SIZE, PCI_DMA_FROMDEVICE); - ps_page_dma->ps_page_dma[j] = 0; - put_page(ps_page->ps_page[j]); - ps_page->ps_page[j] = NULL; - } + } + ps_page = &rx_ring->ps_page[i]; + ps_page_dma = &rx_ring->ps_page_dma[i]; + for (j = 0; j < adapter->rx_ps_pages; j++) { + if (!ps_page->ps_page[j]) break; + pci_unmap_page(pdev, + ps_page_dma->ps_page_dma[j], + PAGE_SIZE, PCI_DMA_FROMDEVICE); + ps_page_dma->ps_page_dma[j] = 0; + put_page(ps_page->ps_page[j]); + ps_page->ps_page[j] = NULL; } } @@ -1542,8 +1972,22 @@ e1000_clean_rx_ring(struct e1000_adapter *adapter) rx_ring->next_to_clean = 0; rx_ring->next_to_use = 0; - E1000_WRITE_REG(&adapter->hw, RDH, 0); - E1000_WRITE_REG(&adapter->hw, RDT, 0); + writel(0, adapter->hw.hw_addr + rx_ring->rdh); + writel(0, adapter->hw.hw_addr + rx_ring->rdt); +} + +/** + * e1000_clean_all_rx_rings - Free Rx Buffers for all queues + * @adapter: board private structure + **/ + +static void +e1000_clean_all_rx_rings(struct e1000_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_rx_queues; i++) + e1000_clean_rx_ring(adapter, &adapter->rx_ring[i]); } /* The 82542 2.0 (revision 2) needs to have the receive unit in reset @@ -1563,8 +2007,8 @@ e1000_enter_82542_rst(struct e1000_adapter *adapter) E1000_WRITE_FLUSH(&adapter->hw); mdelay(5); - if(netif_running(netdev)) - e1000_clean_rx_ring(adapter); + if (netif_running(netdev)) + e1000_clean_all_rx_rings(adapter); } static void @@ -1579,12 +2023,14 @@ e1000_leave_82542_rst(struct e1000_adapter *adapter) E1000_WRITE_FLUSH(&adapter->hw); mdelay(5); - if(adapter->hw.pci_cmd_word & PCI_COMMAND_INVALIDATE) + if (adapter->hw.pci_cmd_word & PCI_COMMAND_INVALIDATE) e1000_pci_set_mwi(&adapter->hw); - if(netif_running(netdev)) { + if (netif_running(netdev)) { + /* No need to loop, because 82542 supports only 1 queue */ + struct e1000_rx_ring *ring = &adapter->rx_ring[0]; e1000_configure_rx(adapter); - e1000_alloc_rx_buffers(adapter); + adapter->alloc_rx_buf(adapter, ring, E1000_DESC_UNUSED(ring)); } } @@ -1599,15 +2045,15 @@ e1000_leave_82542_rst(struct e1000_adapter *adapter) static int e1000_set_mac(struct net_device *netdev, void *p) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); struct sockaddr *addr = p; - if(!is_valid_ether_addr(addr->sa_data)) + if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; /* 82542 2.0 needs to be in reset to write receive address registers */ - if(adapter->hw.mac_type == e1000_82542_rev2_0) + if (adapter->hw.mac_type == e1000_82542_rev2_0) e1000_enter_82542_rst(adapter); memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); @@ -1615,7 +2061,23 @@ e1000_set_mac(struct net_device *netdev, void *p) e1000_rar_set(&adapter->hw, adapter->hw.mac_addr, 0); - if(adapter->hw.mac_type == e1000_82542_rev2_0) + /* With 82571 controllers, LAA may be overwritten (with the default) + * due to controller reset from the other port. */ + if (adapter->hw.mac_type == e1000_82571) { + /* activate the work around */ + adapter->hw.laa_is_present = 1; + + /* Hold a copy of the LAA in RAR[14] This is done so that + * between the time RAR[0] gets clobbered and the time it + * gets fixed (in e1000_watchdog), the actual LAA is in one + * of the RARs and no incoming packets directed to this port + * are dropped. Eventaully the LAA will be in RAR[0] and + * RAR[14] */ + e1000_rar_set(&adapter->hw, adapter->hw.mac_addr, + E1000_RAR_ENTRIES - 1); + } + + if (adapter->hw.mac_type == e1000_82542_rev2_0) e1000_leave_82542_rst(adapter); return 0; @@ -1634,23 +2096,24 @@ e1000_set_mac(struct net_device *netdev, void *p) static void e1000_set_multi(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; struct dev_mc_list *mc_ptr; - unsigned long flags; uint32_t rctl; uint32_t hash_value; - int i; + int i, rar_entries = E1000_RAR_ENTRIES; - spin_lock_irqsave(&adapter->tx_lock, flags); + /* reserve RAR[14] for LAA over-write work-around */ + if (adapter->hw.mac_type == e1000_82571) + rar_entries--; /* Check for Promiscuous and All Multicast modes */ rctl = E1000_READ_REG(hw, RCTL); - if(netdev->flags & IFF_PROMISC) { + if (netdev->flags & IFF_PROMISC) { rctl |= (E1000_RCTL_UPE | E1000_RCTL_MPE); - } else if(netdev->flags & IFF_ALLMULTI) { + } else if (netdev->flags & IFF_ALLMULTI) { rctl |= E1000_RCTL_MPE; rctl &= ~E1000_RCTL_UPE; } else { @@ -1661,17 +2124,18 @@ e1000_set_multi(struct net_device *netdev) /* 82542 2.0 needs to be in reset to write receive address registers */ - if(hw->mac_type == e1000_82542_rev2_0) + if (hw->mac_type == e1000_82542_rev2_0) e1000_enter_82542_rst(adapter); /* load the first 14 multicast address into the exact filters 1-14 * RAR 0 is used for the station MAC adddress * if there are not 14 addresses, go ahead and clear the filters + * -- with 82571 controllers only 0-13 entries are filled here */ mc_ptr = netdev->mc_list; - for(i = 1; i < E1000_RAR_ENTRIES; i++) { - if(mc_ptr) { + for (i = 1; i < rar_entries; i++) { + if (mc_ptr) { e1000_rar_set(hw, mc_ptr->dmi_addr, i); mc_ptr = mc_ptr->next; } else { @@ -1682,20 +2146,18 @@ e1000_set_multi(struct net_device *netdev) /* clear the old settings from the multicast hash table */ - for(i = 0; i < E1000_NUM_MTA_REGISTERS; i++) + for (i = 0; i < E1000_NUM_MTA_REGISTERS; i++) E1000_WRITE_REG_ARRAY(hw, MTA, i, 0); /* load any remaining addresses into the hash table */ - for(; mc_ptr; mc_ptr = mc_ptr->next) { + for (; mc_ptr; mc_ptr = mc_ptr->next) { hash_value = e1000_hash_mc_addr(hw, mc_ptr->dmi_addr); e1000_mta_set(hw, hash_value); } - if(hw->mac_type == e1000_82542_rev2_0) + if (hw->mac_type == e1000_82542_rev2_0) e1000_leave_82542_rst(adapter); - - spin_unlock_irqrestore(&adapter->tx_lock, flags); } /* Need to wait a few seconds after link up to get diagnostic information from @@ -1720,8 +2182,8 @@ e1000_82547_tx_fifo_stall(unsigned long data) struct net_device *netdev = adapter->netdev; uint32_t tctl; - if(atomic_read(&adapter->tx_fifo_stall)) { - if((E1000_READ_REG(&adapter->hw, TDT) == + if (atomic_read(&adapter->tx_fifo_stall)) { + if ((E1000_READ_REG(&adapter->hw, TDT) == E1000_READ_REG(&adapter->hw, TDH)) && (E1000_READ_REG(&adapter->hw, TDFT) == E1000_READ_REG(&adapter->hw, TDFH)) && @@ -1767,24 +2229,24 @@ static void e1000_watchdog_task(struct e1000_adapter *adapter) { struct net_device *netdev = adapter->netdev; - struct e1000_desc_ring *txdr = &adapter->tx_ring; - uint32_t link; + struct e1000_tx_ring *txdr = adapter->tx_ring; + uint32_t link, tctl; e1000_check_for_link(&adapter->hw); if (adapter->hw.mac_type == e1000_82573) { e1000_enable_tx_pkt_filtering(&adapter->hw); - if(adapter->mng_vlan_id != adapter->hw.mng_cookie.vlan_id) + if (adapter->mng_vlan_id != adapter->hw.mng_cookie.vlan_id) e1000_update_mng_vlan(adapter); - } + } - if((adapter->hw.media_type == e1000_media_type_internal_serdes) && + if ((adapter->hw.media_type == e1000_media_type_internal_serdes) && !(E1000_READ_REG(&adapter->hw, TXCW) & E1000_TXCW_ANE)) link = !adapter->hw.serdes_link_down; else link = E1000_READ_REG(&adapter->hw, STATUS) & E1000_STATUS_LU; - if(link) { - if(!netif_carrier_ok(netdev)) { + if (link) { + if (!netif_carrier_ok(netdev)) { e1000_get_speed_and_duplex(&adapter->hw, &adapter->link_speed, &adapter->link_duplex); @@ -1794,19 +2256,85 @@ e1000_watchdog_task(struct e1000_adapter *adapter) adapter->link_duplex == FULL_DUPLEX ? "Full Duplex" : "Half Duplex"); + /* tweak tx_queue_len according to speed/duplex + * and adjust the timeout factor */ + netdev->tx_queue_len = adapter->tx_queue_len; + adapter->tx_timeout_factor = 1; + adapter->txb2b = 1; + switch (adapter->link_speed) { + case SPEED_10: + adapter->txb2b = 0; + netdev->tx_queue_len = 10; + adapter->tx_timeout_factor = 8; + break; + case SPEED_100: + adapter->txb2b = 0; + netdev->tx_queue_len = 100; + /* maybe add some timeout factor ? */ + break; + } + + if ((adapter->hw.mac_type == e1000_82571 || + adapter->hw.mac_type == e1000_82572) && + adapter->txb2b == 0) { +#define SPEED_MODE_BIT (1 << 21) + uint32_t tarc0; + tarc0 = E1000_READ_REG(&adapter->hw, TARC0); + tarc0 &= ~SPEED_MODE_BIT; + E1000_WRITE_REG(&adapter->hw, TARC0, tarc0); + } + +#ifdef NETIF_F_TSO + /* disable TSO for pcie and 10/100 speeds, to avoid + * some hardware issues */ + if (!adapter->tso_force && + adapter->hw.bus_type == e1000_bus_type_pci_express){ + switch (adapter->link_speed) { + case SPEED_10: + case SPEED_100: + DPRINTK(PROBE,INFO, + "10/100 speed: disabling TSO\n"); + netdev->features &= ~NETIF_F_TSO; + break; + case SPEED_1000: + netdev->features |= NETIF_F_TSO; + break; + default: + /* oops */ + break; + } + } +#endif + + /* enable transmits in the hardware, need to do this + * after setting TARC0 */ + tctl = E1000_READ_REG(&adapter->hw, TCTL); + tctl |= E1000_TCTL_EN; + E1000_WRITE_REG(&adapter->hw, TCTL, tctl); + netif_carrier_on(netdev); netif_wake_queue(netdev); mod_timer(&adapter->phy_info_timer, jiffies + 2 * HZ); adapter->smartspeed = 0; } } else { - if(netif_carrier_ok(netdev)) { + if (netif_carrier_ok(netdev)) { adapter->link_speed = 0; adapter->link_duplex = 0; DPRINTK(LINK, INFO, "NIC Link is Down\n"); netif_carrier_off(netdev); netif_stop_queue(netdev); mod_timer(&adapter->phy_info_timer, jiffies + 2 * HZ); + + /* 80003ES2LAN workaround-- + * For packet buffer work-around on link down event; + * disable receives in the ISR and + * reset device here in the watchdog + */ + if (adapter->hw.mac_type == e1000_80003es2lan) { + /* reset device */ + schedule_work(&adapter->reset_task); + } } e1000_smartspeed(adapter); @@ -1826,23 +2354,24 @@ e1000_watchdog_task(struct e1000_adapter *adapter) e1000_update_adaptive(&adapter->hw); - if(!netif_carrier_ok(netdev)) { - if(E1000_DESC_UNUSED(txdr) + 1 < txdr->count) { + if (!netif_carrier_ok(netdev)) { + if (E1000_DESC_UNUSED(txdr) + 1 < txdr->count) { /* We've lost link, so the controller stops DMA, * but we've got queued Tx work that's never going * to get done, so reset controller to flush Tx. * (Do the reset outside of interrupt context). */ - schedule_work(&adapter->tx_timeout_task); + adapter->tx_timeout_count++; + schedule_work(&adapter->reset_task); } } /* Dynamic mode for Interrupt Throttle Rate (ITR) */ - if(adapter->hw.mac_type >= e1000_82540 && adapter->itr == 1) { + if (adapter->hw.mac_type >= e1000_82540 && adapter->itr == 1) { /* Symmetric Tx/Rx gets a reduced ITR=2000; Total * asymmetrical Tx or Rx gets ITR=8000; everyone * else is between 2000-8000. */ uint32_t goc = (adapter->gotcl + adapter->gorcl) / 10000; - uint32_t dif = (adapter->gotcl > adapter->gorcl ? + uint32_t dif = (adapter->gotcl > adapter->gorcl ? adapter->gotcl - adapter->gorcl : adapter->gorcl - adapter->gotcl) / 10000; uint32_t itr = goc > 0 ? (dif * 6000 / goc + 2000) : 8000; @@ -1855,6 +2384,11 @@ e1000_watchdog_task(struct e1000_adapter *adapter) /* Force detection of hung controller every watchdog period */ adapter->detect_tx_hung = TRUE; + /* With 82571 controllers, LAA may be overwritten due to controller + * reset from the other port. Set the appropriate LAA in RAR[0] */ + if (adapter->hw.mac_type == e1000_82571 && adapter->hw.laa_is_present) + e1000_rar_set(&adapter->hw, adapter->hw.mac_addr, 0); + /* Reset the timer */ mod_timer(&adapter->watchdog_timer, jiffies + 2 * HZ); } @@ -1867,17 +2401,19 @@ e1000_watchdog_task(struct e1000_adapter *adapter) #define E1000_TX_FLAGS_VLAN_SHIFT 16 static inline int -e1000_tso(struct e1000_adapter *adapter, struct sk_buff *skb) +e1000_tso(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, + struct sk_buff *skb) { #ifdef NETIF_F_TSO struct e1000_context_desc *context_desc; + struct e1000_buffer *buffer_info; unsigned int i; uint32_t cmd_length = 0; uint16_t ipcse = 0, tucse, mss; uint8_t ipcss, ipcso, tucss, tucso, hdr_len; int err; - if(skb_shinfo(skb)->tso_size) { + if (skb_shinfo(skb)->tso_size) { if (skb_header_cloned(skb)) { err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC); if (err) @@ -1886,7 +2422,7 @@ e1000_tso(struct e1000_adapter *adapter, struct sk_buff *skb) hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2)); mss = skb_shinfo(skb)->tso_size; - if(skb->protocol == ntohs(ETH_P_IP)) { + if (skb->protocol == ntohs(ETH_P_IP)) { skb->nh.iph->tot_len = 0; skb->nh.iph->check = 0; skb->h.th->check = @@ -1898,7 +2434,7 @@ e1000_tso(struct e1000_adapter *adapter, struct sk_buff *skb) cmd_length = E1000_TXD_CMD_IP; ipcse = skb->h.raw - skb->data - 1; #ifdef NETIF_F_TSO_IPV6 - } else if(skb->protocol == ntohs(ETH_P_IPV6)) { + } else if (skb->protocol == ntohs(ETH_P_IPV6)) { skb->nh.ipv6h->payload_len = 0; skb->h.th->check = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, @@ -1918,8 +2454,9 @@ e1000_tso(struct e1000_adapter *adapter, struct sk_buff *skb) cmd_length |= (E1000_TXD_CMD_DEXT | E1000_TXD_CMD_TSE | E1000_TXD_CMD_TCP | (skb->len - (hdr_len))); - i = adapter->tx_ring.next_to_use; - context_desc = E1000_CONTEXT_DESC(adapter->tx_ring, i); + i = tx_ring->next_to_use; + context_desc = E1000_CONTEXT_DESC(*tx_ring, i); + buffer_info = &tx_ring->buffer_info[i]; context_desc->lower_setup.ip_fields.ipcss = ipcss; context_desc->lower_setup.ip_fields.ipcso = ipcso; @@ -1931,28 +2468,33 @@ e1000_tso(struct e1000_adapter *adapter, struct sk_buff *skb) context_desc->tcp_seg_setup.fields.hdr_len = hdr_len; context_desc->cmd_and_length = cpu_to_le32(cmd_length); - if(++i == adapter->tx_ring.count) i = 0; - adapter->tx_ring.next_to_use = i; + buffer_info->time_stamp = jiffies; - return 1; + if (++i == tx_ring->count) i = 0; + tx_ring->next_to_use = i; + + return TRUE; } #endif - return 0; + return FALSE; } static inline boolean_t -e1000_tx_csum(struct e1000_adapter *adapter, struct sk_buff *skb) +e1000_tx_csum(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, + struct sk_buff *skb) { struct e1000_context_desc *context_desc; + struct e1000_buffer *buffer_info; unsigned int i; uint8_t css; - if(likely(skb->ip_summed == CHECKSUM_HW)) { + if (likely(skb->ip_summed == CHECKSUM_HW)) { css = skb->h.raw - skb->data; - i = adapter->tx_ring.next_to_use; - context_desc = E1000_CONTEXT_DESC(adapter->tx_ring, i); + i = tx_ring->next_to_use; + buffer_info = &tx_ring->buffer_info[i]; + context_desc = E1000_CONTEXT_DESC(*tx_ring, i); context_desc->upper_setup.tcp_fields.tucss = css; context_desc->upper_setup.tcp_fields.tucso = css + skb->csum; @@ -1960,8 +2502,10 @@ e1000_tx_csum(struct e1000_adapter *adapter, struct sk_buff *skb) context_desc->tcp_seg_setup.data = 0; context_desc->cmd_and_length = cpu_to_le32(E1000_TXD_CMD_DEXT); - if(unlikely(++i == adapter->tx_ring.count)) i = 0; - adapter->tx_ring.next_to_use = i; + buffer_info->time_stamp = jiffies; + + if (unlikely(++i == tx_ring->count)) i = 0; + tx_ring->next_to_use = i; return TRUE; } @@ -1973,11 +2517,10 @@ e1000_tx_csum(struct e1000_adapter *adapter, struct sk_buff *skb) #define E1000_MAX_DATA_PER_TXD (1<tx_ring; struct e1000_buffer *buffer_info; unsigned int len = skb->len; unsigned int offset = 0, size, count = 0, i; @@ -1986,13 +2529,23 @@ e1000_tx_map(struct e1000_adapter *adapter, struct sk_buff *skb, i = tx_ring->next_to_use; - while(len) { + while (len) { buffer_info = &tx_ring->buffer_info[i]; size = min(len, max_per_txd); #ifdef NETIF_F_TSO + /* Workaround for Controller erratum -- + * descriptor for non-tso packet in a linear SKB that follows a + * tso gets written back prematurely before the data is fully + * DMA'd to the controller */ + if (!skb->data_len && tx_ring->last_tx_tso && + !skb_shinfo(skb)->tso_size) { + tx_ring->last_tx_tso = 0; + size -= 4; + } + /* Workaround for premature desc write-backs * in TSO mode. Append 4-byte sentinel desc */ - if(unlikely(mss && !nr_frags && size == len && size > 8)) + if (unlikely(mss && !nr_frags && size == len && size > 8)) size -= 4; #endif /* work-around for errata 10 and it applies @@ -2000,13 +2553,13 @@ e1000_tx_map(struct e1000_adapter *adapter, struct sk_buff *skb, * The fix is to make sure that the first descriptor of a * packet is smaller than 2048 - 16 - 16 (or 2016) bytes */ - if(unlikely((adapter->hw.bus_type == e1000_bus_type_pcix) && + if (unlikely((adapter->hw.bus_type == e1000_bus_type_pcix) && (size > 2015) && count == 0)) size = 2015; - + /* Workaround for potential 82544 hang in PCI-X. Avoid * terminating buffers within evenly-aligned dwords. */ - if(unlikely(adapter->pcix_82544 && + if (unlikely(adapter->pcix_82544 && !((unsigned long)(skb->data + offset + size - 1) & 4) && size > 4)) size -= 4; @@ -2022,29 +2575,29 @@ e1000_tx_map(struct e1000_adapter *adapter, struct sk_buff *skb, len -= size; offset += size; count++; - if(unlikely(++i == tx_ring->count)) i = 0; + if (unlikely(++i == tx_ring->count)) i = 0; } - for(f = 0; f < nr_frags; f++) { + for (f = 0; f < nr_frags; f++) { struct skb_frag_struct *frag; frag = &skb_shinfo(skb)->frags[f]; len = frag->size; offset = frag->page_offset; - while(len) { + while (len) { buffer_info = &tx_ring->buffer_info[i]; size = min(len, max_per_txd); #ifdef NETIF_F_TSO /* Workaround for premature desc write-backs * in TSO mode. Append 4-byte sentinel desc */ - if(unlikely(mss && f == (nr_frags-1) && size == len && size > 8)) + if (unlikely(mss && f == (nr_frags-1) && size == len && size > 8)) size -= 4; #endif /* Workaround for potential 82544 hang in PCI-X. * Avoid terminating buffers within evenly-aligned * dwords. */ - if(unlikely(adapter->pcix_82544 && + if (unlikely(adapter->pcix_82544 && !((unsigned long)(frag->page+offset+size-1) & 4) && size > 4)) size -= 4; @@ -2061,7 +2614,7 @@ e1000_tx_map(struct e1000_adapter *adapter, struct sk_buff *skb, len -= size; offset += size; count++; - if(unlikely(++i == tx_ring->count)) i = 0; + if (unlikely(++i == tx_ring->count)) i = 0; } } @@ -2073,43 +2626,43 @@ e1000_tx_map(struct e1000_adapter *adapter, struct sk_buff *skb, } static inline void -e1000_tx_queue(struct e1000_adapter *adapter, int count, int tx_flags) +e1000_tx_queue(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, + int tx_flags, int count) { - struct e1000_desc_ring *tx_ring = &adapter->tx_ring; struct e1000_tx_desc *tx_desc = NULL; struct e1000_buffer *buffer_info; uint32_t txd_upper = 0, txd_lower = E1000_TXD_CMD_IFCS; unsigned int i; - if(likely(tx_flags & E1000_TX_FLAGS_TSO)) { + if (likely(tx_flags & E1000_TX_FLAGS_TSO)) { txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D | E1000_TXD_CMD_TSE; txd_upper |= E1000_TXD_POPTS_TXSM << 8; - if(likely(tx_flags & E1000_TX_FLAGS_IPV4)) + if (likely(tx_flags & E1000_TX_FLAGS_IPV4)) txd_upper |= E1000_TXD_POPTS_IXSM << 8; } - if(likely(tx_flags & E1000_TX_FLAGS_CSUM)) { + if (likely(tx_flags & E1000_TX_FLAGS_CSUM)) { txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D; txd_upper |= E1000_TXD_POPTS_TXSM << 8; } - if(unlikely(tx_flags & E1000_TX_FLAGS_VLAN)) { + if (unlikely(tx_flags & E1000_TX_FLAGS_VLAN)) { txd_lower |= E1000_TXD_CMD_VLE; txd_upper |= (tx_flags & E1000_TX_FLAGS_VLAN_MASK); } i = tx_ring->next_to_use; - while(count--) { + while (count--) { buffer_info = &tx_ring->buffer_info[i]; tx_desc = E1000_TX_DESC(*tx_ring, i); tx_desc->buffer_addr = cpu_to_le64(buffer_info->dma); tx_desc->lower.data = cpu_to_le32(txd_lower | buffer_info->length); tx_desc->upper.data = cpu_to_le32(txd_upper); - if(unlikely(++i == tx_ring->count)) i = 0; + if (unlikely(++i == tx_ring->count)) i = 0; } tx_desc->lower.data |= cpu_to_le32(adapter->txd_cmd); @@ -2121,7 +2674,7 @@ e1000_tx_queue(struct e1000_adapter *adapter, int count, int tx_flags) wmb(); tx_ring->next_to_use = i; - E1000_WRITE_REG(&adapter->hw, TDT, i); + writel(i, adapter->hw.hw_addr + tx_ring->tdt); } /** @@ -2144,20 +2697,20 @@ e1000_82547_fifo_workaround(struct e1000_adapter *adapter, struct sk_buff *skb) E1000_ROUNDUP(skb_fifo_len, E1000_FIFO_HDR); - if(adapter->link_duplex != HALF_DUPLEX) + if (adapter->link_duplex != HALF_DUPLEX) goto no_fifo_stall_required; - if(atomic_read(&adapter->tx_fifo_stall)) + if (atomic_read(&adapter->tx_fifo_stall)) return 1; - if(skb_fifo_len >= (E1000_82547_PAD_LEN + fifo_space)) { + if (skb_fifo_len >= (E1000_82547_PAD_LEN + fifo_space)) { atomic_set(&adapter->tx_fifo_stall, 1); return 1; } no_fifo_stall_required: adapter->tx_fifo_head += skb_fifo_len; - if(adapter->tx_fifo_head >= adapter->tx_fifo_size) + if (adapter->tx_fifo_head >= adapter->tx_fifo_size) adapter->tx_fifo_head -= adapter->tx_fifo_size; return 0; } @@ -2168,39 +2721,27 @@ e1000_transfer_dhcp_info(struct e1000_adapter *adapter, struct sk_buff *skb) { struct e1000_hw *hw = &adapter->hw; uint16_t length, offset; - if(vlan_tx_tag_present(skb)) { - if(!((vlan_tx_tag_get(skb) == adapter->hw.mng_cookie.vlan_id) && + if (vlan_tx_tag_present(skb)) { + if (!((vlan_tx_tag_get(skb) == adapter->hw.mng_cookie.vlan_id) && ( adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT)) ) return 0; } - if(htons(ETH_P_IP) == skb->protocol) { - const struct iphdr *ip = skb->nh.iph; - if(IPPROTO_UDP == ip->protocol) { - struct udphdr *udp = (struct udphdr *)(skb->h.uh); - if(ntohs(udp->dest) == 67) { - offset = (uint8_t *)udp + 8 - skb->data; - length = skb->len - offset; - - return e1000_mng_write_dhcp_info(hw, - (uint8_t *)udp + 8, length); - } - } - } else if((skb->len > MINIMUM_DHCP_PACKET_SIZE) && (!skb->protocol)) { + if (skb->len > MINIMUM_DHCP_PACKET_SIZE) { struct ethhdr *eth = (struct ethhdr *) skb->data; - if((htons(ETH_P_IP) == eth->h_proto)) { - const struct iphdr *ip = + if ((htons(ETH_P_IP) == eth->h_proto)) { + const struct iphdr *ip = (struct iphdr *)((uint8_t *)skb->data+14); - if(IPPROTO_UDP == ip->protocol) { - struct udphdr *udp = - (struct udphdr *)((uint8_t *)ip + + if (IPPROTO_UDP == ip->protocol) { + struct udphdr *udp = + (struct udphdr *)((uint8_t *)ip + (ip->ihl << 2)); - if(ntohs(udp->dest) == 67) { + if (ntohs(udp->dest) == 67) { offset = (uint8_t *)udp + 8 - skb->data; length = skb->len - offset; return e1000_mng_write_dhcp_info(hw, - (uint8_t *)udp + 8, + (uint8_t *)udp + 8, length); } } @@ -2213,7 +2754,8 @@ e1000_transfer_dhcp_info(struct e1000_adapter *adapter, struct sk_buff *skb) static int e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); + struct e1000_tx_ring *tx_ring; unsigned int first, max_per_txd = E1000_MAX_DATA_PER_TXD; unsigned int max_txd_pwr = E1000_MAX_TXD_PWR; unsigned int tx_flags = 0; @@ -2222,11 +2764,13 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) unsigned int nr_frags = 0; unsigned int mss = 0; int count = 0; - int tso; + int tso; unsigned int f; len -= skb->data_len; - if(unlikely(skb->len <= 0)) { + tx_ring = adapter->tx_ring; + + if (unlikely(skb->len <= 0)) { dev_kfree_skb_any(skb); return NETDEV_TX_OK; } @@ -2239,100 +2783,138 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) * 4 = ceil(buffer len/mss). To make sure we don't * overrun the FIFO, adjust the max buffer len if mss * drops. */ - if(mss) { + if (mss) { + uint8_t hdr_len; max_per_txd = min(mss << 2, max_per_txd); max_txd_pwr = fls(max_per_txd) - 1; + + /* TSO Workaround for 82571/2/3 Controllers -- if skb->data + * points to just header, pull a few bytes of payload from + * frags into skb->data */ + hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2)); + if (skb->data_len && (hdr_len == (skb->len - skb->data_len))) { + switch (adapter->hw.mac_type) { + unsigned int pull_size; + case e1000_82571: + case e1000_82572: + case e1000_82573: + pull_size = min((unsigned int)4, skb->data_len); + if (!__pskb_pull_tail(skb, pull_size)) { + printk(KERN_ERR + "__pskb_pull_tail failed.\n"); + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + len = skb->len - skb->data_len; + break; + default: + /* do nothing */ + break; + } + } } - if((mss) || (skb->ip_summed == CHECKSUM_HW)) + /* reserve a descriptor for the offload context */ + if ((mss) || (skb->ip_summed == CHECKSUM_HW)) count++; count++; #else - if(skb->ip_summed == CHECKSUM_HW) + if (skb->ip_summed == CHECKSUM_HW) + count++; +#endif + +#ifdef NETIF_F_TSO + /* Controller Erratum workaround */ + if (!skb->data_len && tx_ring->last_tx_tso && + !skb_shinfo(skb)->tso_size) count++; #endif + count += TXD_USE_COUNT(len, max_txd_pwr); - if(adapter->pcix_82544) + if (adapter->pcix_82544) count++; - /* work-around for errata 10 and it applies to all controllers + /* work-around for errata 10 and it applies to all controllers * in PCI-X mode, so add one more descriptor to the count */ - if(unlikely((adapter->hw.bus_type == e1000_bus_type_pcix) && + if (unlikely((adapter->hw.bus_type == e1000_bus_type_pcix) && (len > 2015))) count++; nr_frags = skb_shinfo(skb)->nr_frags; - for(f = 0; f < nr_frags; f++) + for (f = 0; f < nr_frags; f++) count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size, max_txd_pwr); - if(adapter->pcix_82544) + if (adapter->pcix_82544) count += nr_frags; - local_irq_save(flags); - if (!spin_trylock(&adapter->tx_lock)) { - /* Collision - tell upper layer to requeue */ - local_irq_restore(flags); - return NETDEV_TX_LOCKED; - } - if(adapter->hw.tx_pkt_filtering && (adapter->hw.mac_type == e1000_82573) ) + + if (adapter->hw.tx_pkt_filtering && + (adapter->hw.mac_type == e1000_82573)) e1000_transfer_dhcp_info(adapter, skb); + local_irq_save(flags); + if (!spin_trylock(&tx_ring->tx_lock)) { + /* Collision - tell upper layer to requeue */ + local_irq_restore(flags); + return NETDEV_TX_LOCKED; + } /* need: count + 2 desc gap to keep tail from touching * head, otherwise try next time */ - if(unlikely(E1000_DESC_UNUSED(&adapter->tx_ring) < count + 2)) { + if (unlikely(E1000_DESC_UNUSED(tx_ring) < count + 2)) { netif_stop_queue(netdev); - spin_unlock_irqrestore(&adapter->tx_lock, flags); + spin_unlock_irqrestore(&tx_ring->tx_lock, flags); return NETDEV_TX_BUSY; } - if(unlikely(adapter->hw.mac_type == e1000_82547)) { - if(unlikely(e1000_82547_fifo_workaround(adapter, skb))) { + if (unlikely(adapter->hw.mac_type == e1000_82547)) { + if (unlikely(e1000_82547_fifo_workaround(adapter, skb))) { netif_stop_queue(netdev); mod_timer(&adapter->tx_fifo_stall_timer, jiffies); - spin_unlock_irqrestore(&adapter->tx_lock, flags); + spin_unlock_irqrestore(&tx_ring->tx_lock, flags); return NETDEV_TX_BUSY; } } - if(unlikely(adapter->vlgrp && vlan_tx_tag_present(skb))) { + if (unlikely(adapter->vlgrp && vlan_tx_tag_present(skb))) { tx_flags |= E1000_TX_FLAGS_VLAN; tx_flags |= (vlan_tx_tag_get(skb) << E1000_TX_FLAGS_VLAN_SHIFT); } - first = adapter->tx_ring.next_to_use; - - tso = e1000_tso(adapter, skb); + first = tx_ring->next_to_use; + + tso = e1000_tso(adapter, tx_ring, skb); if (tso < 0) { dev_kfree_skb_any(skb); - spin_unlock_irqrestore(&adapter->tx_lock, flags); + spin_unlock_irqrestore(&tx_ring->tx_lock, flags); return NETDEV_TX_OK; } - if (likely(tso)) + if (likely(tso)) { + tx_ring->last_tx_tso = 1; tx_flags |= E1000_TX_FLAGS_TSO; - else if(likely(e1000_tx_csum(adapter, skb))) + } else if (likely(e1000_tx_csum(adapter, tx_ring, skb))) tx_flags |= E1000_TX_FLAGS_CSUM; /* Old method was to assume IPv4 packet by default if TSO was enabled. - * 82573 hardware supports TSO capabilities for IPv6 as well... + * 82571 hardware supports TSO capabilities for IPv6 as well... * no longer assume, we must. */ - if(likely(skb->protocol == ntohs(ETH_P_IP))) + if (likely(skb->protocol == ntohs(ETH_P_IP))) tx_flags |= E1000_TX_FLAGS_IPV4; - e1000_tx_queue(adapter, - e1000_tx_map(adapter, skb, first, max_per_txd, nr_frags, mss), - tx_flags); + e1000_tx_queue(adapter, tx_ring, tx_flags, + e1000_tx_map(adapter, tx_ring, skb, first, + max_per_txd, nr_frags, mss)); netdev->trans_start = jiffies; /* Make sure there is space in the ring for the next send. */ - if(unlikely(E1000_DESC_UNUSED(&adapter->tx_ring) < MAX_SKB_FRAGS + 2)) + if (unlikely(E1000_DESC_UNUSED(tx_ring) < MAX_SKB_FRAGS + 2)) netif_stop_queue(netdev); - spin_unlock_irqrestore(&adapter->tx_lock, flags); + spin_unlock_irqrestore(&tx_ring->tx_lock, flags); return NETDEV_TX_OK; } @@ -2344,16 +2926,17 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) static void e1000_tx_timeout(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); /* Do the reset outside of interrupt context */ - schedule_work(&adapter->tx_timeout_task); + adapter->tx_timeout_count++; + schedule_work(&adapter->reset_task); } static void -e1000_tx_timeout_task(struct net_device *netdev) +e1000_reset_task(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); e1000_down(adapter); e1000_up(adapter); @@ -2370,9 +2953,9 @@ e1000_tx_timeout_task(struct net_device *netdev) static struct net_device_stats * e1000_get_stats(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); - e1000_update_stats(adapter); + /* only return the current stats */ return &adapter->net_stats; } @@ -2387,25 +2970,55 @@ e1000_get_stats(struct net_device *netdev) static int e1000_change_mtu(struct net_device *netdev, int new_mtu) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); int max_frame = new_mtu + ENET_HEADER_SIZE + ETHERNET_FCS_SIZE; + uint16_t eeprom_data = 0; - if((max_frame < MINIMUM_ETHERNET_FRAME_SIZE) || - (max_frame > MAX_JUMBO_FRAME_SIZE)) { - DPRINTK(PROBE, ERR, "Invalid MTU setting\n"); - return -EINVAL; + if ((max_frame < MINIMUM_ETHERNET_FRAME_SIZE) || + (max_frame > MAX_JUMBO_FRAME_SIZE)) { + DPRINTK(PROBE, ERR, "Invalid MTU setting\n"); + return -EINVAL; } -#define MAX_STD_JUMBO_FRAME_SIZE 9216 - /* might want this to be bigger enum check... */ - if (adapter->hw.mac_type == e1000_82573 && - max_frame > MAXIMUM_ETHERNET_FRAME_SIZE) { - DPRINTK(PROBE, ERR, "Jumbo Frames not supported " - "on 82573\n"); - return -EINVAL; + /* Adapter-specific max frame size limits. */ + switch (adapter->hw.mac_type) { + case e1000_82542_rev2_0: + case e1000_82542_rev2_1: + if (max_frame > MAXIMUM_ETHERNET_FRAME_SIZE) { + DPRINTK(PROBE, ERR, "Jumbo Frames not supported.\n"); + return -EINVAL; + } + break; + case e1000_82573: + /* only enable jumbo frames if ASPM is disabled completely + * this means both bits must be zero in 0x1A bits 3:2 */ + e1000_read_eeprom(&adapter->hw, EEPROM_INIT_3GIO_3, 1, + &eeprom_data); + if (eeprom_data & EEPROM_WORD1A_ASPM_MASK) { + if (max_frame > MAXIMUM_ETHERNET_FRAME_SIZE) { + DPRINTK(PROBE, ERR, + "Jumbo Frames not supported.\n"); + return -EINVAL; + } + break; + } + /* fall through to get support */ + case e1000_82571: + case e1000_82572: + case e1000_80003es2lan: +#define MAX_STD_JUMBO_FRAME_SIZE 9234 + if (max_frame > MAX_STD_JUMBO_FRAME_SIZE) { + DPRINTK(PROBE, ERR, "MTU > 9216 not supported.\n"); + return -EINVAL; + } + break; + default: + /* Capable of supporting up to MAX_JUMBO_FRAME_SIZE limit. */ + break; } - if(adapter->hw.mac_type > e1000_82547_rev_2) { + + if (adapter->hw.mac_type > e1000_82547_rev_2) { adapter->rx_buffer_len = max_frame; E1000_ROUNDUP(adapter->rx_buffer_len, 1024); } else { @@ -2414,23 +3027,21 @@ e1000_change_mtu(struct net_device *netdev, int new_mtu) DPRINTK(PROBE, ERR, "Jumbo Frames not supported " "on 82542\n"); return -EINVAL; - } else { - if(max_frame <= E1000_RXBUFFER_2048) { + if(max_frame <= E1000_RXBUFFER_2048) adapter->rx_buffer_len = E1000_RXBUFFER_2048; - } else if(max_frame <= E1000_RXBUFFER_4096) { + else if(max_frame <= E1000_RXBUFFER_4096) adapter->rx_buffer_len = E1000_RXBUFFER_4096; - } else if(max_frame <= E1000_RXBUFFER_8192) { + else if(max_frame <= E1000_RXBUFFER_8192) adapter->rx_buffer_len = E1000_RXBUFFER_8192; - } else if(max_frame <= E1000_RXBUFFER_16384) { + else if(max_frame <= E1000_RXBUFFER_16384) adapter->rx_buffer_len = E1000_RXBUFFER_16384; - } } } netdev->mtu = new_mtu; - if(netif_running(netdev)) { + if (netif_running(netdev)) { e1000_down(adapter); e1000_up(adapter); } @@ -2517,7 +3128,7 @@ e1000_update_stats(struct e1000_adapter *adapter) hw->collision_delta = E1000_READ_REG(hw, COLC); adapter->stats.colc += hw->collision_delta; - if(hw->mac_type >= e1000_82543) { + if (hw->mac_type >= e1000_82543) { adapter->stats.algnerrc += E1000_READ_REG(hw, ALGNERRC); adapter->stats.rxerrc += E1000_READ_REG(hw, RXERRC); adapter->stats.tncrs += E1000_READ_REG(hw, TNCRS); @@ -2525,7 +3136,7 @@ e1000_update_stats(struct e1000_adapter *adapter) adapter->stats.tsctc += E1000_READ_REG(hw, TSCTC); adapter->stats.tsctfc += E1000_READ_REG(hw, TSCTFC); } - if(hw->mac_type > e1000_82547_rev_2) { + if (hw->mac_type > e1000_82547_rev_2) { adapter->stats.iac += E1000_READ_REG(hw, IAC); adapter->stats.icrxoc += E1000_READ_REG(hw, ICRXOC); adapter->stats.icrxptc += E1000_READ_REG(hw, ICRXPTC); @@ -2548,15 +3159,17 @@ e1000_update_stats(struct e1000_adapter *adapter) /* Rx Errors */ + /* RLEC on some newer hardware can be incorrect so build + * our own version based on RUC and ROC */ adapter->net_stats.rx_errors = adapter->stats.rxerrc + adapter->stats.crcerrs + adapter->stats.algnerrc + - adapter->stats.rlec + adapter->stats.mpc + + adapter->stats.ruc + adapter->stats.roc + adapter->stats.cexterr; - adapter->net_stats.rx_dropped = adapter->stats.mpc; - adapter->net_stats.rx_length_errors = adapter->stats.rlec; + adapter->net_stats.rx_dropped = 0; + adapter->net_stats.rx_length_errors = adapter->stats.ruc + + adapter->stats.roc; adapter->net_stats.rx_crc_errors = adapter->stats.crcerrs; adapter->net_stats.rx_frame_errors = adapter->stats.algnerrc; - adapter->net_stats.rx_fifo_errors = adapter->stats.mpc; adapter->net_stats.rx_missed_errors = adapter->stats.mpc; /* Tx Errors */ @@ -2571,14 +3184,14 @@ e1000_update_stats(struct e1000_adapter *adapter) /* Phy Stats */ - if(hw->media_type == e1000_media_type_copper) { - if((adapter->link_speed == SPEED_1000) && + if (hw->media_type == e1000_media_type_copper) { + if ((adapter->link_speed == SPEED_1000) && (!e1000_read_phy_reg(hw, PHY_1000T_STATUS, &phy_tmp))) { phy_tmp &= PHY_IDLE_ERROR_COUNT_MASK; adapter->phy_stats.idle_errors += phy_tmp; } - if((hw->mac_type <= e1000_82546) && + if ((hw->mac_type <= e1000_82546) && (hw->phy_type == e1000_phy_m88) && !e1000_read_phy_reg(hw, M88E1000_RX_ERR_CNTR, &phy_tmp)) adapter->phy_stats.receive_errors += phy_tmp; @@ -2598,55 +3211,78 @@ static irqreturn_t e1000_intr(int irq, void *data, struct pt_regs *regs) { struct net_device *netdev = data; - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; - uint32_t icr = E1000_READ_REG(hw, ICR); + uint32_t rctl, icr = E1000_READ_REG(hw, ICR); #ifndef CONFIG_E1000_NAPI - unsigned int i; + int i; +#else + /* Interrupt Auto-Mask...upon reading ICR, + * interrupts are masked. No need for the + * IMC write, but it does mean we should + * account for it ASAP. */ + if (likely(hw->mac_type >= e1000_82571)) + atomic_inc(&adapter->irq_sem); #endif - if(unlikely(!icr)) + if (unlikely(!icr)) { +#ifdef CONFIG_E1000_NAPI + if (hw->mac_type >= e1000_82571) + e1000_irq_enable(adapter); +#endif return IRQ_NONE; /* Not our interrupt */ + } - if(unlikely(icr & (E1000_ICR_RXSEQ | E1000_ICR_LSC))) { + if (unlikely(icr & (E1000_ICR_RXSEQ | E1000_ICR_LSC))) { hw->get_link_status = 1; + /* 80003ES2LAN workaround-- + * For packet buffer work-around on link down event; + * disable receives here in the ISR and + * reset adapter in watchdog + */ + if (netif_carrier_ok(netdev) && + (adapter->hw.mac_type == e1000_80003es2lan)) { + /* disable receives */ + rctl = E1000_READ_REG(hw, RCTL); + E1000_WRITE_REG(hw, RCTL, rctl & ~E1000_RCTL_EN); + } mod_timer(&adapter->watchdog_timer, jiffies); } #ifdef CONFIG_E1000_NAPI - if(likely(netif_rx_schedule_prep(netdev))) { - - /* Disable interrupts and register for poll. The flush - of the posted write is intentionally left out. - */ - + if (unlikely(hw->mac_type < e1000_82571)) { atomic_inc(&adapter->irq_sem); E1000_WRITE_REG(hw, IMC, ~0); - __netif_rx_schedule(netdev); + E1000_WRITE_FLUSH(hw); } + if (likely(netif_rx_schedule_prep(&adapter->polling_netdev[0]))) + __netif_rx_schedule(&adapter->polling_netdev[0]); + else + e1000_irq_enable(adapter); #else /* Writing IMC and IMS is needed for 82547. - Due to Hub Link bus being occupied, an interrupt - de-assertion message is not able to be sent. - When an interrupt assertion message is generated later, - two messages are re-ordered and sent out. - That causes APIC to think 82547 is in de-assertion - state, while 82547 is in assertion state, resulting - in dead lock. Writing IMC forces 82547 into - de-assertion state. - */ - if(hw->mac_type == e1000_82547 || hw->mac_type == e1000_82547_rev_2){ + * Due to Hub Link bus being occupied, an interrupt + * de-assertion message is not able to be sent. + * When an interrupt assertion message is generated later, + * two messages are re-ordered and sent out. + * That causes APIC to think 82547 is in de-assertion + * state, while 82547 is in assertion state, resulting + * in dead lock. Writing IMC forces 82547 into + * de-assertion state. + */ + if (hw->mac_type == e1000_82547 || hw->mac_type == e1000_82547_rev_2) { atomic_inc(&adapter->irq_sem); E1000_WRITE_REG(hw, IMC, ~0); } - for(i = 0; i < E1000_MAX_INTR; i++) - if(unlikely(!adapter->clean_rx(adapter) & - !e1000_clean_tx_irq(adapter))) + for (i = 0; i < E1000_MAX_INTR; i++) + if (unlikely(!adapter->clean_rx(adapter, adapter->rx_ring) & + !e1000_clean_tx_irq(adapter, adapter->tx_ring))) break; - if(hw->mac_type == e1000_82547 || hw->mac_type == e1000_82547_rev_2) + if (hw->mac_type == e1000_82547 || hw->mac_type == e1000_82547_rev_2) e1000_irq_enable(adapter); + #endif return IRQ_HANDLED; @@ -2659,22 +3295,48 @@ e1000_intr(int irq, void *data, struct pt_regs *regs) **/ static int -e1000_clean(struct net_device *netdev, int *budget) +e1000_clean(struct net_device *poll_dev, int *budget) { - struct e1000_adapter *adapter = netdev->priv; - int work_to_do = min(*budget, netdev->quota); - int tx_cleaned; - int work_done = 0; + struct e1000_adapter *adapter; + int work_to_do = min(*budget, poll_dev->quota); + int tx_cleaned = 0, i = 0, work_done = 0; + + /* Must NOT use netdev_priv macro here. */ + adapter = poll_dev->priv; + + /* Keep link state information with original netdev */ + if (!netif_carrier_ok(adapter->netdev)) + goto quit_polling; + + while (poll_dev != &adapter->polling_netdev[i]) { + i++; + BUG_ON(i == adapter->num_rx_queues); + } + + if (likely(adapter->num_tx_queues == 1)) { + /* e1000_clean is called per-cpu. This lock protects + * tx_ring[0] from being cleaned by multiple cpus + * simultaneously. A failure obtaining the lock means + * tx_ring[0] is currently being cleaned anyway. */ + if (spin_trylock(&adapter->tx_queue_lock)) { + tx_cleaned = e1000_clean_tx_irq(adapter, + &adapter->tx_ring[0]); + spin_unlock(&adapter->tx_queue_lock); + } + } else + tx_cleaned = e1000_clean_tx_irq(adapter, &adapter->tx_ring[i]); - tx_cleaned = e1000_clean_tx_irq(adapter); - adapter->clean_rx(adapter, &work_done, work_to_do); + adapter->clean_rx(adapter, &adapter->rx_ring[i], + &work_done, work_to_do); *budget -= work_done; - netdev->quota -= work_done; - - /* If no Tx and no Rx work done, exit the polling mode */ - if ((!tx_cleaned && (work_done == 0)) || !netif_running(netdev)) { - netif_rx_complete(netdev); + poll_dev->quota -= work_done; + + /* If no Tx and not enough Rx work done, exit the polling mode */ + if ((!tx_cleaned && (work_done == 0)) || + !netif_running(adapter->netdev)) { +quit_polling: + netif_rx_complete(poll_dev); e1000_irq_enable(adapter); return 0; } @@ -2689,117 +3351,89 @@ e1000_clean(struct net_device *netdev, int *budget) **/ static boolean_t -e1000_clean_tx_irq(struct e1000_adapter *adapter) +e1000_clean_tx_irq(struct e1000_adapter *adapter, + struct e1000_tx_ring *tx_ring) { - struct e1000_desc_ring *tx_ring = &adapter->tx_ring; struct net_device *netdev = adapter->netdev; struct e1000_tx_desc *tx_desc, *eop_desc; struct e1000_buffer *buffer_info; unsigned int i, eop; +#ifdef CONFIG_E1000_NAPI + unsigned int count = 0; +#endif boolean_t cleaned = FALSE; i = tx_ring->next_to_clean; eop = tx_ring->buffer_info[i].next_to_watch; eop_desc = E1000_TX_DESC(*tx_ring, eop); - while(eop_desc->upper.data & cpu_to_le32(E1000_TXD_STAT_DD)) { - /* Premature writeback of Tx descriptors clear (free buffers - * and unmap pci_mapping) previous_buffer_info */ - if (likely(adapter->previous_buffer_info.skb != NULL)) { - e1000_unmap_and_free_tx_resource(adapter, - &adapter->previous_buffer_info); - } - - for(cleaned = FALSE; !cleaned; ) { + while (eop_desc->upper.data & cpu_to_le32(E1000_TXD_STAT_DD)) { + for (cleaned = FALSE; !cleaned; ) { tx_desc = E1000_TX_DESC(*tx_ring, i); buffer_info = &tx_ring->buffer_info[i]; cleaned = (i == eop); -#ifdef NETIF_F_TSO - if (!(netdev->features & NETIF_F_TSO)) { -#endif - e1000_unmap_and_free_tx_resource(adapter, - buffer_info); -#ifdef NETIF_F_TSO - } else { - if (cleaned) { - memcpy(&adapter->previous_buffer_info, - buffer_info, - sizeof(struct e1000_buffer)); - memset(buffer_info, 0, - sizeof(struct e1000_buffer)); - } else { - e1000_unmap_and_free_tx_resource( - adapter, buffer_info); - } - } -#endif - - tx_desc->buffer_addr = 0; - tx_desc->lower.data = 0; - tx_desc->upper.data = 0; + e1000_unmap_and_free_tx_resource(adapter, buffer_info); + memset(tx_desc, 0, sizeof(struct e1000_tx_desc)); - if(unlikely(++i == tx_ring->count)) i = 0; + if (unlikely(++i == tx_ring->count)) i = 0; } - + + eop = tx_ring->buffer_info[i].next_to_watch; eop_desc = E1000_TX_DESC(*tx_ring, eop); +#ifdef CONFIG_E1000_NAPI +#define E1000_TX_WEIGHT 64 + /* weight of a sort for tx, to avoid endless transmit cleanup */ + if (count++ == E1000_TX_WEIGHT) break; +#endif } tx_ring->next_to_clean = i; - spin_lock(&adapter->tx_lock); + spin_lock(&tx_ring->tx_lock); - if(unlikely(cleaned && netif_queue_stopped(netdev) && + if (unlikely(cleaned && netif_queue_stopped(netdev) && netif_carrier_ok(netdev))) netif_wake_queue(netdev); - spin_unlock(&adapter->tx_lock); - if(adapter->detect_tx_hung) { + spin_unlock(&tx_ring->tx_lock); + if (adapter->detect_tx_hung) { /* Detect a transmit hang in hardware, this serializes the * check with the clearing of time_stamp and movement of i */ adapter->detect_tx_hung = FALSE; - if (tx_ring->buffer_info[i].dma && - time_after(jiffies, tx_ring->buffer_info[i].time_stamp + HZ) + if (tx_ring->buffer_info[eop].dma && + time_after(jiffies, tx_ring->buffer_info[eop].time_stamp + + (adapter->tx_timeout_factor * HZ)) && !(E1000_READ_REG(&adapter->hw, STATUS) & - E1000_STATUS_TXOFF)) { + E1000_STATUS_TXOFF)) { /* detected Tx unit hang */ - i = tx_ring->next_to_clean; - eop = tx_ring->buffer_info[i].next_to_watch; - eop_desc = E1000_TX_DESC(*tx_ring, eop); - DPRINTK(TX_ERR, ERR, "Detected Tx Unit Hang\n" + DPRINTK(DRV, ERR, "Detected Tx Unit Hang\n" + " Tx Queue <%lu>\n" " TDH <%x>\n" " TDT <%x>\n" " next_to_use <%x>\n" " next_to_clean <%x>\n" "buffer_info[next_to_clean]\n" - " dma <%llx>\n" " time_stamp <%lx>\n" " next_to_watch <%x>\n" " jiffies <%lx>\n" " next_to_watch.status <%x>\n", - E1000_READ_REG(&adapter->hw, TDH), - E1000_READ_REG(&adapter->hw, TDT), + (unsigned long)((tx_ring - adapter->tx_ring) / + sizeof(struct e1000_tx_ring)), + readl(adapter->hw.hw_addr + tx_ring->tdh), + readl(adapter->hw.hw_addr + tx_ring->tdt), tx_ring->next_to_use, - i, - tx_ring->buffer_info[i].dma, - tx_ring->buffer_info[i].time_stamp, + tx_ring->next_to_clean, + tx_ring->buffer_info[eop].time_stamp, eop, jiffies, eop_desc->upper.fields.status); netif_stop_queue(netdev); } } -#ifdef NETIF_F_TSO - - if( unlikely(!(eop_desc->upper.data & cpu_to_le32(E1000_TXD_STAT_DD)) && - time_after(jiffies, adapter->previous_buffer_info.time_stamp + HZ))) - e1000_unmap_and_free_tx_resource( - adapter, &adapter->previous_buffer_info); - -#endif return cleaned; } @@ -2821,21 +3455,21 @@ e1000_rx_checksum(struct e1000_adapter *adapter, skb->ip_summed = CHECKSUM_NONE; /* 82543 or newer only */ - if(unlikely(adapter->hw.mac_type < e1000_82543)) return; + if (unlikely(adapter->hw.mac_type < e1000_82543)) return; /* Ignore Checksum bit is set */ - if(unlikely(status & E1000_RXD_STAT_IXSM)) return; + if (unlikely(status & E1000_RXD_STAT_IXSM)) return; /* TCP/UDP checksum error bit is set */ - if(unlikely(errors & E1000_RXD_ERR_TCPE)) { + if (unlikely(errors & E1000_RXD_ERR_TCPE)) { /* let the stack verify checksum errors */ adapter->hw_csum_err++; return; } /* TCP/UDP Checksum has not been calculated */ - if(adapter->hw.mac_type <= e1000_82547_rev_2) { - if(!(status & E1000_RXD_STAT_TCPCS)) + if (adapter->hw.mac_type <= e1000_82547_rev_2) { + if (!(status & E1000_RXD_STAT_TCPCS)) return; } else { - if(!(status & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS))) + if (!(status & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS))) return; } /* It must be a TCP or UDP packet with a valid checksum */ @@ -2861,45 +3495,59 @@ e1000_rx_checksum(struct e1000_adapter *adapter, static boolean_t #ifdef CONFIG_E1000_NAPI -e1000_clean_rx_irq(struct e1000_adapter *adapter, int *work_done, - int work_to_do) +e1000_clean_rx_irq(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, + int *work_done, int work_to_do) #else -e1000_clean_rx_irq(struct e1000_adapter *adapter) +e1000_clean_rx_irq(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring) #endif { - struct e1000_desc_ring *rx_ring = &adapter->rx_ring; struct net_device *netdev = adapter->netdev; struct pci_dev *pdev = adapter->pdev; - struct e1000_rx_desc *rx_desc; - struct e1000_buffer *buffer_info; - struct sk_buff *skb; + struct e1000_rx_desc *rx_desc, *next_rxd; + struct e1000_buffer *buffer_info, *next_buffer; unsigned long flags; uint32_t length; uint8_t last_byte; unsigned int i; + int cleaned_count = 0; boolean_t cleaned = FALSE; i = rx_ring->next_to_clean; rx_desc = E1000_RX_DESC(*rx_ring, i); + buffer_info = &rx_ring->buffer_info[i]; - while(rx_desc->status & E1000_RXD_STAT_DD) { - buffer_info = &rx_ring->buffer_info[i]; + while (rx_desc->status & E1000_RXD_STAT_DD) { + struct sk_buff *skb; + u8 status; #ifdef CONFIG_E1000_NAPI - if(*work_done >= work_to_do) + if (*work_done >= work_to_do) break; (*work_done)++; #endif - cleaned = TRUE; + status = rx_desc->status; + skb = buffer_info->skb; + buffer_info->skb = NULL; + + prefetch(skb->data - NET_IP_ALIGN); + if (++i == rx_ring->count) i = 0; + next_rxd = E1000_RX_DESC(*rx_ring, i); + prefetch(next_rxd); + + next_buffer = &rx_ring->buffer_info[i]; + + cleaned = TRUE; + cleaned_count++; pci_unmap_single(pdev, buffer_info->dma, buffer_info->length, PCI_DMA_FROMDEVICE); - skb = buffer_info->skb; length = le16_to_cpu(rx_desc->length); - if(unlikely(!(rx_desc->status & E1000_RXD_STAT_EOP))) { + if (unlikely(!(status & E1000_RXD_STAT_EOP))) { /* All receives must fit into a single buffer */ E1000_DBG("%s: Receive packet consumed multiple" " buffers\n", netdev->name); @@ -2907,9 +3555,9 @@ e1000_clean_rx_irq(struct e1000_adapter *adapter) goto next_desc; } - if(unlikely(rx_desc->errors & E1000_RXD_ERR_FRAME_ERR_MASK)) { + if (unlikely(rx_desc->errors & E1000_RXD_ERR_FRAME_ERR_MASK)) { last_byte = *(skb->data + length - 1); - if(TBI_ACCEPT(&adapter->hw, rx_desc->status, + if (TBI_ACCEPT(&adapter->hw, status, rx_desc->errors, length, last_byte)) { spin_lock_irqsave(&adapter->stats_lock, flags); e1000_tbi_adjust_stats(&adapter->hw, @@ -2924,18 +3572,39 @@ e1000_clean_rx_irq(struct e1000_adapter *adapter) } } - /* Good Receive */ - skb_put(skb, length - ETHERNET_FCS_SIZE); + /* code added for copybreak, this should improve + * performance for small packets with large amounts + * of reassembly being done in the stack */ +#define E1000_CB_LENGTH 256 + if (length < E1000_CB_LENGTH) { + struct sk_buff *new_skb = + dev_alloc_skb(length + NET_IP_ALIGN); + if (new_skb) { + skb_reserve(new_skb, NET_IP_ALIGN); + new_skb->dev = netdev; + memcpy(new_skb->data - NET_IP_ALIGN, + skb->data - NET_IP_ALIGN, + length + NET_IP_ALIGN); + /* save the skb in buffer_info as good */ + buffer_info->skb = skb; + skb = new_skb; + skb_put(skb, length); + } + } else + skb_put(skb, length); + + /* end copybreak code */ /* Receive Checksum Offload */ e1000_rx_checksum(adapter, - (uint32_t)(rx_desc->status) | + (uint32_t)(status) | ((uint32_t)(rx_desc->errors) << 24), - rx_desc->csum, skb); + le16_to_cpu(rx_desc->csum), skb); + skb->protocol = eth_type_trans(skb, netdev); #ifdef CONFIG_E1000_NAPI - if(unlikely(adapter->vlgrp && - (rx_desc->status & E1000_RXD_STAT_VP))) { + if (unlikely(adapter->vlgrp && + (status & E1000_RXD_STAT_VP))) { vlan_hwaccel_receive_skb(skb, adapter->vlgrp, le16_to_cpu(rx_desc->special) & E1000_RXD_SPC_VLAN_MASK); @@ -2943,8 +3612,8 @@ e1000_clean_rx_irq(struct e1000_adapter *adapter) netif_receive_skb(skb); } #else /* CONFIG_E1000_NAPI */ - if(unlikely(adapter->vlgrp && - (rx_desc->status & E1000_RXD_STAT_VP))) { + if (unlikely(adapter->vlgrp && + (status & E1000_RXD_STAT_VP))) { vlan_hwaccel_rx(skb, adapter->vlgrp, le16_to_cpu(rx_desc->special) & E1000_RXD_SPC_VLAN_MASK); @@ -2956,13 +3625,22 @@ e1000_clean_rx_irq(struct e1000_adapter *adapter) next_desc: rx_desc->status = 0; - buffer_info->skb = NULL; - if(unlikely(++i == rx_ring->count)) i = 0; - rx_desc = E1000_RX_DESC(*rx_ring, i); + /* return some buffers to hardware, one at a time is too slow */ + if (unlikely(cleaned_count >= E1000_RX_BUFFER_WRITE)) { + adapter->alloc_rx_buf(adapter, rx_ring, cleaned_count); + cleaned_count = 0; + } + + /* use prefetched values */ + rx_desc = next_rxd; + buffer_info = next_buffer; } rx_ring->next_to_clean = i; - adapter->alloc_rx_buf(adapter); + + cleaned_count = E1000_DESC_UNUSED(rx_ring); + if (cleaned_count) + adapter->alloc_rx_buf(adapter, rx_ring, cleaned_count); return cleaned; } @@ -2974,59 +3652,71 @@ next_desc: static boolean_t #ifdef CONFIG_E1000_NAPI -e1000_clean_rx_irq_ps(struct e1000_adapter *adapter, int *work_done, - int work_to_do) +e1000_clean_rx_irq_ps(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, + int *work_done, int work_to_do) #else -e1000_clean_rx_irq_ps(struct e1000_adapter *adapter) +e1000_clean_rx_irq_ps(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring) #endif { - struct e1000_desc_ring *rx_ring = &adapter->rx_ring; - union e1000_rx_desc_packet_split *rx_desc; + union e1000_rx_desc_packet_split *rx_desc, *next_rxd; struct net_device *netdev = adapter->netdev; struct pci_dev *pdev = adapter->pdev; - struct e1000_buffer *buffer_info; + struct e1000_buffer *buffer_info, *next_buffer; struct e1000_ps_page *ps_page; struct e1000_ps_page_dma *ps_page_dma; struct sk_buff *skb; unsigned int i, j; uint32_t length, staterr; + int cleaned_count = 0; boolean_t cleaned = FALSE; i = rx_ring->next_to_clean; rx_desc = E1000_RX_DESC_PS(*rx_ring, i); - staterr = rx_desc->wb.middle.status_error; + staterr = le32_to_cpu(rx_desc->wb.middle.status_error); - while(staterr & E1000_RXD_STAT_DD) { + while (staterr & E1000_RXD_STAT_DD) { buffer_info = &rx_ring->buffer_info[i]; ps_page = &rx_ring->ps_page[i]; ps_page_dma = &rx_ring->ps_page_dma[i]; #ifdef CONFIG_E1000_NAPI - if(unlikely(*work_done >= work_to_do)) + if (unlikely(*work_done >= work_to_do)) break; (*work_done)++; #endif + skb = buffer_info->skb; + + /* in the packet split case this is header only */ + prefetch(skb->data - NET_IP_ALIGN); + + if (++i == rx_ring->count) i = 0; + next_rxd = E1000_RX_DESC_PS(*rx_ring, i); + prefetch(next_rxd); + + next_buffer = &rx_ring->buffer_info[i]; + cleaned = TRUE; + cleaned_count++; pci_unmap_single(pdev, buffer_info->dma, buffer_info->length, PCI_DMA_FROMDEVICE); - skb = buffer_info->skb; - - if(unlikely(!(staterr & E1000_RXD_STAT_EOP))) { + if (unlikely(!(staterr & E1000_RXD_STAT_EOP))) { E1000_DBG("%s: Packet Split buffers didn't pick up" " the full packet\n", netdev->name); dev_kfree_skb_irq(skb); goto next_desc; } - if(unlikely(staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK)) { + if (unlikely(staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK)) { dev_kfree_skb_irq(skb); goto next_desc; } length = le16_to_cpu(rx_desc->wb.middle.length0); - if(unlikely(!length)) { + if (unlikely(!length)) { E1000_DBG("%s: Last part of the packet spanning" " multiple descriptors\n", netdev->name); dev_kfree_skb_irq(skb); @@ -3036,45 +3726,70 @@ e1000_clean_rx_irq_ps(struct e1000_adapter *adapter) /* Good Receive */ skb_put(skb, length); - for(j = 0; j < PS_PAGE_BUFFERS; j++) { - if(!(length = le16_to_cpu(rx_desc->wb.upper.length[j]))) + { + /* this looks ugly, but it seems compiler issues make it + more efficient than reusing j */ + int l1 = le16_to_cpu(rx_desc->wb.upper.length[0]); + + /* page alloc/put takes too long and effects small packet + * throughput, so unsplit small packets and save the alloc/put*/ + if (l1 && ((length + l1) < E1000_CB_LENGTH)) { + u8 *vaddr; + /* there is no documentation about how to call + * kmap_atomic, so we can't hold the mapping + * very long */ + pci_dma_sync_single_for_cpu(pdev, + ps_page_dma->ps_page_dma[0], + PAGE_SIZE, + PCI_DMA_FROMDEVICE); + vaddr = kmap_atomic(ps_page->ps_page[0], + KM_SKB_DATA_SOFTIRQ); + memcpy(skb->tail, vaddr, l1); + kunmap_atomic(vaddr, KM_SKB_DATA_SOFTIRQ); + pci_dma_sync_single_for_device(pdev, + ps_page_dma->ps_page_dma[0], + PAGE_SIZE, PCI_DMA_FROMDEVICE); + skb_put(skb, l1); + length += l1; + goto copydone; + } /* if */ + } + + for (j = 0; j < adapter->rx_ps_pages; j++) { + if (!(length= le16_to_cpu(rx_desc->wb.upper.length[j]))) break; - pci_unmap_page(pdev, ps_page_dma->ps_page_dma[j], PAGE_SIZE, PCI_DMA_FROMDEVICE); ps_page_dma->ps_page_dma[j] = 0; - skb_shinfo(skb)->frags[j].page = - ps_page->ps_page[j]; + skb_fill_page_desc(skb, j, ps_page->ps_page[j], 0, + length); ps_page->ps_page[j] = NULL; - skb_shinfo(skb)->frags[j].page_offset = 0; - skb_shinfo(skb)->frags[j].size = length; - skb_shinfo(skb)->nr_frags++; skb->len += length; skb->data_len += length; + skb->truesize += length; } +copydone: e1000_rx_checksum(adapter, staterr, - rx_desc->wb.lower.hi_dword.csum_ip.csum, skb); + le16_to_cpu(rx_desc->wb.lower.hi_dword.csum_ip.csum), skb); skb->protocol = eth_type_trans(skb, netdev); -#ifdef HAVE_RX_ZERO_COPY - if(likely(rx_desc->wb.upper.header_status & - E1000_RXDPS_HDRSTAT_HDRSP)) - skb_shinfo(skb)->zero_copy = TRUE; -#endif + if (likely(rx_desc->wb.upper.header_status & + cpu_to_le16(E1000_RXDPS_HDRSTAT_HDRSP))) + adapter->rx_hdr_split++; #ifdef CONFIG_E1000_NAPI - if(unlikely(adapter->vlgrp && (staterr & E1000_RXD_STAT_VP))) { + if (unlikely(adapter->vlgrp && (staterr & E1000_RXD_STAT_VP))) { vlan_hwaccel_receive_skb(skb, adapter->vlgrp, - le16_to_cpu(rx_desc->wb.middle.vlan & - E1000_RXD_SPC_VLAN_MASK)); + le16_to_cpu(rx_desc->wb.middle.vlan) & + E1000_RXD_SPC_VLAN_MASK); } else { netif_receive_skb(skb); } #else /* CONFIG_E1000_NAPI */ - if(unlikely(adapter->vlgrp && (staterr & E1000_RXD_STAT_VP))) { + if (unlikely(adapter->vlgrp && (staterr & E1000_RXD_STAT_VP))) { vlan_hwaccel_rx(skb, adapter->vlgrp, - le16_to_cpu(rx_desc->wb.middle.vlan & - E1000_RXD_SPC_VLAN_MASK)); + le16_to_cpu(rx_desc->wb.middle.vlan) & + E1000_RXD_SPC_VLAN_MASK); } else { netif_rx(skb); } @@ -3082,15 +3797,26 @@ e1000_clean_rx_irq_ps(struct e1000_adapter *adapter) netdev->last_rx = jiffies; next_desc: - rx_desc->wb.middle.status_error &= ~0xFF; + rx_desc->wb.middle.status_error &= cpu_to_le32(~0xFF); buffer_info->skb = NULL; - if(unlikely(++i == rx_ring->count)) i = 0; - rx_desc = E1000_RX_DESC_PS(*rx_ring, i); - staterr = rx_desc->wb.middle.status_error; + /* return some buffers to hardware, one at a time is too slow */ + if (unlikely(cleaned_count >= E1000_RX_BUFFER_WRITE)) { + adapter->alloc_rx_buf(adapter, rx_ring, cleaned_count); + cleaned_count = 0; + } + + /* use prefetched values */ + rx_desc = next_rxd; + buffer_info = next_buffer; + + staterr = le32_to_cpu(rx_desc->wb.middle.status_error); } rx_ring->next_to_clean = i; - adapter->alloc_rx_buf(adapter); + + cleaned_count = E1000_DESC_UNUSED(rx_ring); + if (cleaned_count) + adapter->alloc_rx_buf(adapter, rx_ring, cleaned_count); return cleaned; } @@ -3101,9 +3827,10 @@ next_desc: **/ static void -e1000_alloc_rx_buffers(struct e1000_adapter *adapter) +e1000_alloc_rx_buffers(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, + int cleaned_count) { - struct e1000_desc_ring *rx_ring = &adapter->rx_ring; struct net_device *netdev = adapter->netdev; struct pci_dev *pdev = adapter->pdev; struct e1000_rx_desc *rx_desc; @@ -3115,11 +3842,17 @@ e1000_alloc_rx_buffers(struct e1000_adapter *adapter) i = rx_ring->next_to_use; buffer_info = &rx_ring->buffer_info[i]; - while(!buffer_info->skb) { - skb = dev_alloc_skb(bufsz); + while (cleaned_count--) { + if (!(skb = buffer_info->skb)) + skb = dev_alloc_skb(bufsz); + else { + skb_trim(skb, 0); + goto map_skb; + } - if(unlikely(!skb)) { + if (unlikely(!skb)) { /* Better luck next round */ + adapter->alloc_rx_buff_failed++; break; } @@ -3156,6 +3889,7 @@ e1000_alloc_rx_buffers(struct e1000_adapter *adapter) buffer_info->skb = skb; buffer_info->length = adapter->rx_buffer_len; +map_skb: buffer_info->dma = pci_map_single(pdev, skb->data, adapter->rx_buffer_len, @@ -3181,20 +3915,23 @@ e1000_alloc_rx_buffers(struct e1000_adapter *adapter) rx_desc = E1000_RX_DESC(*rx_ring, i); rx_desc->buffer_addr = cpu_to_le64(buffer_info->dma); - if(unlikely((i & ~(E1000_RX_BUFFER_WRITE - 1)) == i)) { - /* Force memory writes to complete before letting h/w - * know there are new descriptors to fetch. (Only - * applicable for weak-ordered memory model archs, - * such as IA-64). */ - wmb(); - E1000_WRITE_REG(&adapter->hw, RDT, i); - } - - if(unlikely(++i == rx_ring->count)) i = 0; + if (unlikely(++i == rx_ring->count)) + i = 0; buffer_info = &rx_ring->buffer_info[i]; } - rx_ring->next_to_use = i; + if (likely(rx_ring->next_to_use != i)) { + rx_ring->next_to_use = i; + if (unlikely(i-- == 0)) + i = (rx_ring->count - 1); + + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). */ + wmb(); + writel(i, adapter->hw.hw_addr + rx_ring->rdt); + } } /** @@ -3203,9 +3940,10 @@ e1000_alloc_rx_buffers(struct e1000_adapter *adapter) **/ static void -e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter) +e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter, + struct e1000_rx_ring *rx_ring, + int cleaned_count) { - struct e1000_desc_ring *rx_ring = &adapter->rx_ring; struct net_device *netdev = adapter->netdev; struct pci_dev *pdev = adapter->pdev; union e1000_rx_desc_packet_split *rx_desc; @@ -3220,32 +3958,40 @@ e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter) ps_page = &rx_ring->ps_page[i]; ps_page_dma = &rx_ring->ps_page_dma[i]; - while(!buffer_info->skb) { + while (cleaned_count--) { rx_desc = E1000_RX_DESC_PS(*rx_ring, i); - for(j = 0; j < PS_PAGE_BUFFERS; j++) { - if(unlikely(!ps_page->ps_page[j])) { - ps_page->ps_page[j] = - alloc_page(GFP_ATOMIC); - if(unlikely(!ps_page->ps_page[j])) - goto no_buffers; - ps_page_dma->ps_page_dma[j] = - pci_map_page(pdev, - ps_page->ps_page[j], - 0, PAGE_SIZE, - PCI_DMA_FROMDEVICE); - } - /* Refresh the desc even if buffer_addrs didn't - * change because each write-back erases this info. - */ - rx_desc->read.buffer_addr[j+1] = - cpu_to_le64(ps_page_dma->ps_page_dma[j]); + for (j = 0; j < PS_PAGE_BUFFERS; j++) { + if (j < adapter->rx_ps_pages) { + if (likely(!ps_page->ps_page[j])) { + ps_page->ps_page[j] = + alloc_page(GFP_ATOMIC); + if (unlikely(!ps_page->ps_page[j])) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + ps_page_dma->ps_page_dma[j] = + pci_map_page(pdev, + ps_page->ps_page[j], + 0, PAGE_SIZE, + PCI_DMA_FROMDEVICE); + } + /* Refresh the desc even if buffer_addrs didn't + * change because each write-back erases + * this info. + */ + rx_desc->read.buffer_addr[j+1] = + cpu_to_le64(ps_page_dma->ps_page_dma[j]); + } else + rx_desc->read.buffer_addr[j+1] = ~0; } skb = dev_alloc_skb(adapter->rx_ps_bsize0 + NET_IP_ALIGN); - if(unlikely(!skb)) + if (unlikely(!skb)) { + adapter->alloc_rx_buff_failed++; break; + } /* Make buffer alignment 2 beyond a 16 byte boundary * this will result in a 16 byte aligned IP header after @@ -3263,27 +4009,28 @@ e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter) rx_desc->read.buffer_addr[0] = cpu_to_le64(buffer_info->dma); - if(unlikely((i & ~(E1000_RX_BUFFER_WRITE - 1)) == i)) { - /* Force memory writes to complete before letting h/w - * know there are new descriptors to fetch. (Only - * applicable for weak-ordered memory model archs, - * such as IA-64). */ - wmb(); - /* Hardware increments by 16 bytes, but packet split - * descriptors are 32 bytes...so we increment tail - * twice as much. - */ - E1000_WRITE_REG(&adapter->hw, RDT, i<<1); - } - - if(unlikely(++i == rx_ring->count)) i = 0; + if (unlikely(++i == rx_ring->count)) i = 0; buffer_info = &rx_ring->buffer_info[i]; ps_page = &rx_ring->ps_page[i]; ps_page_dma = &rx_ring->ps_page_dma[i]; } no_buffers: - rx_ring->next_to_use = i; + if (likely(rx_ring->next_to_use != i)) { + rx_ring->next_to_use = i; + if (unlikely(i-- == 0)) i = (rx_ring->count - 1); + + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). */ + wmb(); + /* Hardware increments by 16 bytes, but packet split + * descriptors are 32 bytes...so we increment tail + * twice as much. + */ + writel(i<<1, adapter->hw.hw_addr + rx_ring->rdt); + } } /** @@ -3297,24 +4044,24 @@ e1000_smartspeed(struct e1000_adapter *adapter) uint16_t phy_status; uint16_t phy_ctrl; - if((adapter->hw.phy_type != e1000_phy_igp) || !adapter->hw.autoneg || + if ((adapter->hw.phy_type != e1000_phy_igp) || !adapter->hw.autoneg || !(adapter->hw.autoneg_advertised & ADVERTISE_1000_FULL)) return; - if(adapter->smartspeed == 0) { + if (adapter->smartspeed == 0) { /* If Master/Slave config fault is asserted twice, * we assume back-to-back */ e1000_read_phy_reg(&adapter->hw, PHY_1000T_STATUS, &phy_status); - if(!(phy_status & SR_1000T_MS_CONFIG_FAULT)) return; + if (!(phy_status & SR_1000T_MS_CONFIG_FAULT)) return; e1000_read_phy_reg(&adapter->hw, PHY_1000T_STATUS, &phy_status); - if(!(phy_status & SR_1000T_MS_CONFIG_FAULT)) return; + if (!(phy_status & SR_1000T_MS_CONFIG_FAULT)) return; e1000_read_phy_reg(&adapter->hw, PHY_1000T_CTRL, &phy_ctrl); - if(phy_ctrl & CR_1000T_MS_ENABLE) { + if (phy_ctrl & CR_1000T_MS_ENABLE) { phy_ctrl &= ~CR_1000T_MS_ENABLE; e1000_write_phy_reg(&adapter->hw, PHY_1000T_CTRL, phy_ctrl); adapter->smartspeed++; - if(!e1000_phy_setup_autoneg(&adapter->hw) && + if (!e1000_phy_setup_autoneg(&adapter->hw) && !e1000_read_phy_reg(&adapter->hw, PHY_CTRL, &phy_ctrl)) { phy_ctrl |= (MII_CR_AUTO_NEG_EN | @@ -3324,12 +4071,12 @@ e1000_smartspeed(struct e1000_adapter *adapter) } } return; - } else if(adapter->smartspeed == E1000_SMARTSPEED_DOWNSHIFT) { + } else if (adapter->smartspeed == E1000_SMARTSPEED_DOWNSHIFT) { /* If still no link, perhaps using 2/3 pair cable */ e1000_read_phy_reg(&adapter->hw, PHY_1000T_CTRL, &phy_ctrl); phy_ctrl |= CR_1000T_MS_ENABLE; e1000_write_phy_reg(&adapter->hw, PHY_1000T_CTRL, phy_ctrl); - if(!e1000_phy_setup_autoneg(&adapter->hw) && + if (!e1000_phy_setup_autoneg(&adapter->hw) && !e1000_read_phy_reg(&adapter->hw, PHY_CTRL, &phy_ctrl)) { phy_ctrl |= (MII_CR_AUTO_NEG_EN | MII_CR_RESTART_AUTO_NEG); @@ -3337,7 +4084,7 @@ e1000_smartspeed(struct e1000_adapter *adapter) } } /* Restart process after E1000_SMARTSPEED_MAX iterations */ - if(adapter->smartspeed++ == E1000_SMARTSPEED_MAX) + if (adapter->smartspeed++ == E1000_SMARTSPEED_MAX) adapter->smartspeed = 0; } @@ -3371,13 +4118,14 @@ e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); struct mii_ioctl_data *data = if_mii(ifr); int retval; uint16_t mii_reg; uint16_t spddplx; + unsigned long flags; - if(adapter->hw.media_type != e1000_media_type_copper) + if (adapter->hw.media_type != e1000_media_type_copper) return -EOPNOTSUPP; switch (cmd) { @@ -3387,9 +4135,13 @@ e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) case SIOCGMIIREG: if (!capable(CAP_NET_ADMIN)) return -EPERM; + spin_lock_irqsave(&adapter->stats_lock, flags); if (e1000_read_phy_reg(&adapter->hw, data->reg_num & 0x1F, - &data->val_out)) + &data->val_out)) { + spin_unlock_irqrestore(&adapter->stats_lock, flags); return -EIO; + } + spin_unlock_irqrestore(&adapter->stats_lock, flags); break; case SIOCSMIIREG: if (!capable(CAP_NET_ADMIN)) @@ -3397,15 +4149,18 @@ e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) if (data->reg_num & ~(0x1F)) return -EFAULT; mii_reg = data->val_in; + spin_lock_irqsave(&adapter->stats_lock, flags); if (e1000_write_phy_reg(&adapter->hw, data->reg_num, - mii_reg)) + mii_reg)) { + spin_unlock_irqrestore(&adapter->stats_lock, flags); return -EIO; - if (adapter->hw.phy_type == e1000_phy_m88) { + } + if (adapter->hw.phy_type == e1000_media_type_copper) { switch (data->reg_num) { case PHY_CTRL: - if(mii_reg & MII_CR_POWER_DOWN) + if (mii_reg & MII_CR_POWER_DOWN) break; - if(mii_reg & MII_CR_AUTO_NEG_EN) { + if (mii_reg & MII_CR_AUTO_NEG_EN) { adapter->hw.autoneg = 1; adapter->hw.autoneg_advertised = 0x2F; } else { @@ -3416,14 +4171,18 @@ e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) else spddplx = SPEED_10; spddplx += (mii_reg & 0x100) - ? FULL_DUPLEX : - HALF_DUPLEX; + ? DUPLEX_FULL : + DUPLEX_HALF; retval = e1000_set_spd_dplx(adapter, spddplx); - if(retval) + if (retval) { + spin_unlock_irqrestore( + &adapter->stats_lock, + flags); return retval; + } } - if(netif_running(adapter->netdev)) { + if (netif_running(adapter->netdev)) { e1000_down(adapter); e1000_up(adapter); } else @@ -3431,16 +4190,19 @@ e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) break; case M88E1000_PHY_SPEC_CTRL: case M88E1000_EXT_PHY_SPEC_CTRL: - if (e1000_phy_reset(&adapter->hw)) + if (e1000_phy_reset(&adapter->hw)) { + spin_unlock_irqrestore( + &adapter->stats_lock, flags); return -EIO; + } break; } } else { switch (data->reg_num) { case PHY_CTRL: - if(mii_reg & MII_CR_POWER_DOWN) + if (mii_reg & MII_CR_POWER_DOWN) break; - if(netif_running(adapter->netdev)) { + if (netif_running(adapter->netdev)) { e1000_down(adapter); e1000_up(adapter); } else @@ -3448,6 +4210,7 @@ e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) break; } } + spin_unlock_irqrestore(&adapter->stats_lock, flags); break; default: return -EOPNOTSUPP; @@ -3461,7 +4224,7 @@ e1000_pci_set_mwi(struct e1000_hw *hw) struct e1000_adapter *adapter = hw->back; int ret_val = pci_set_mwi(adapter->pdev); - if(ret_val) + if (ret_val) DPRINTK(PROBE, ERR, "Error in setting MWI\n"); } @@ -3504,13 +4267,13 @@ e1000_io_write(struct e1000_hw *hw, unsigned long port, uint32_t value) static void e1000_vlan_rx_register(struct net_device *netdev, struct vlan_group *grp) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); uint32_t ctrl, rctl; e1000_irq_disable(adapter); adapter->vlgrp = grp; - if(grp) { + if (grp) { /* enable VLAN tag insert/strip */ ctrl = E1000_READ_REG(&adapter->hw, CTRL); ctrl |= E1000_CTRL_VME; @@ -3532,7 +4295,7 @@ e1000_vlan_rx_register(struct net_device *netdev, struct vlan_group *grp) rctl = E1000_READ_REG(&adapter->hw, RCTL); rctl &= ~E1000_RCTL_VFE; E1000_WRITE_REG(&adapter->hw, RCTL, rctl); - if(adapter->mng_vlan_id != (uint16_t)E1000_MNG_VLAN_NONE) { + if (adapter->mng_vlan_id != (uint16_t)E1000_MNG_VLAN_NONE) { e1000_vlan_rx_kill_vid(netdev, adapter->mng_vlan_id); adapter->mng_vlan_id = E1000_MNG_VLAN_NONE; } @@ -3544,11 +4307,12 @@ e1000_vlan_rx_register(struct net_device *netdev, struct vlan_group *grp) static void e1000_vlan_rx_add_vid(struct net_device *netdev, uint16_t vid) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); uint32_t vfta, index; - if((adapter->hw.mng_cookie.status & - E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT) && - (vid == adapter->mng_vlan_id)) + + if ((adapter->hw.mng_cookie.status & + E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT) && + (vid == adapter->mng_vlan_id)) return; /* add VID to filter table */ index = (vid >> 5) & 0x7F; @@ -3560,20 +4324,24 @@ e1000_vlan_rx_add_vid(struct net_device *netdev, uint16_t vid) static void e1000_vlan_rx_kill_vid(struct net_device *netdev, uint16_t vid) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); uint32_t vfta, index; e1000_irq_disable(adapter); - if(adapter->vlgrp) + if (adapter->vlgrp) adapter->vlgrp->vlan_devices[vid] = NULL; e1000_irq_enable(adapter); - if((adapter->hw.mng_cookie.status & - E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT) && - (vid == adapter->mng_vlan_id)) + if ((adapter->hw.mng_cookie.status & + E1000_MNG_DHCP_COOKIE_STATUS_VLAN_SUPPORT) && + (vid == adapter->mng_vlan_id)) { + /* release control to f/w */ + e1000_release_hw_control(adapter); return; + } + /* remove VID from filter table */ index = (vid >> 5) & 0x7F; vfta = E1000_READ_REG_ARRAY(&adapter->hw, VFTA, index); @@ -3586,10 +4354,10 @@ e1000_restore_vlan(struct e1000_adapter *adapter) { e1000_vlan_rx_register(adapter->netdev, adapter->vlgrp); - if(adapter->vlgrp) { + if (adapter->vlgrp) { uint16_t vid; - for(vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) { - if(!adapter->vlgrp->vlan_devices[vid]) + for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) { + if (!adapter->vlgrp->vlan_devices[vid]) continue; e1000_vlan_rx_add_vid(adapter->netdev, vid); } @@ -3601,7 +4369,14 @@ e1000_set_spd_dplx(struct e1000_adapter *adapter, uint16_t spddplx) { adapter->hw.autoneg = 0; - switch(spddplx) { + /* Fiber NICs only allow 1000 gbps Full duplex */ + if ((adapter->hw.media_type == e1000_media_type_fiber) && + spddplx != (SPEED_1000 + DUPLEX_FULL)) { + DPRINTK(PROBE, ERR, "Unsupported Speed/Duplex configuration\n"); + return -EINVAL; + } + + switch (spddplx) { case SPEED_10 + DUPLEX_HALF: adapter->hw.forced_speed_duplex = e1000_10_half; break; @@ -3626,52 +4401,96 @@ e1000_set_spd_dplx(struct e1000_adapter *adapter, uint16_t spddplx) return 0; } +#ifdef CONFIG_PM +/* Save/restore 16 or 64 dwords of PCI config space depending on which + * bus we're on (PCI(X) vs. PCI-E) + */ +#define PCIE_CONFIG_SPACE_LEN 256 +#define PCI_CONFIG_SPACE_LEN 64 static int -e1000_notify_reboot(struct notifier_block *nb, unsigned long event, void *p) +e1000_pci_save_state(struct e1000_adapter *adapter) { - struct pci_dev *pdev = NULL; + struct pci_dev *dev = adapter->pdev; + int size; + int i; - switch(event) { - case SYS_DOWN: - case SYS_HALT: - case SYS_POWER_OFF: - while((pdev = pci_find_device(PCI_ANY_ID, PCI_ANY_ID, pdev))) { - if(pci_dev_driver(pdev) == &e1000_driver) - e1000_suspend(pdev, 3); - } + if (adapter->hw.mac_type >= e1000_82571) + size = PCIE_CONFIG_SPACE_LEN; + else + size = PCI_CONFIG_SPACE_LEN; + + WARN_ON(adapter->config_space != NULL); + + adapter->config_space = kmalloc(size, GFP_KERNEL); + if (!adapter->config_space) { + DPRINTK(PROBE, ERR, "unable to allocate %d bytes\n", size); + return -ENOMEM; } - return NOTIFY_DONE; + for (i = 0; i < (size / 4); i++) + pci_read_config_dword(dev, i * 4, &adapter->config_space[i]); + return 0; +} + +static void +e1000_pci_restore_state(struct e1000_adapter *adapter) +{ + struct pci_dev *dev = adapter->pdev; + int size; + int i; + + if (adapter->config_space == NULL) + return; + + if (adapter->hw.mac_type >= e1000_82571) + size = PCIE_CONFIG_SPACE_LEN; + else + size = PCI_CONFIG_SPACE_LEN; + for (i = 0; i < (size / 4); i++) + pci_write_config_dword(dev, i * 4, adapter->config_space[i]); + kfree(adapter->config_space); + adapter->config_space = NULL; + return; } +#endif /* CONFIG_PM */ static int -e1000_suspend(struct pci_dev *pdev, uint32_t state) +e1000_suspend(struct pci_dev *pdev, pm_message_t state) { struct net_device *netdev = pci_get_drvdata(pdev); - struct e1000_adapter *adapter = netdev->priv; - uint32_t ctrl, ctrl_ext, rctl, manc, status, swsm; + struct e1000_adapter *adapter = netdev_priv(netdev); + uint32_t ctrl, ctrl_ext, rctl, manc, status; uint32_t wufc = adapter->wol; + int retval = 0; netif_device_detach(netdev); - if(netif_running(netdev)) + if (netif_running(netdev)) e1000_down(adapter); +#ifdef CONFIG_PM + /* Implement our own version of pci_save_state(pdev) because pci- + * express adapters have 256-byte config spaces. */ + retval = e1000_pci_save_state(adapter); + if (retval) + return retval; +#endif + status = E1000_READ_REG(&adapter->hw, STATUS); - if(status & E1000_STATUS_LU) + if (status & E1000_STATUS_LU) wufc &= ~E1000_WUFC_LNKC; - if(wufc) { + if (wufc) { e1000_setup_rctl(adapter); e1000_set_multi(netdev); /* turn on all-multi mode if wake on multicast is enabled */ - if(adapter->wol & E1000_WUFC_MC) { + if (adapter->wol & E1000_WUFC_MC) { rctl = E1000_READ_REG(&adapter->hw, RCTL); rctl |= E1000_RCTL_MPE; E1000_WRITE_REG(&adapter->hw, RCTL, rctl); } - if(adapter->hw.mac_type >= e1000_82540) { + if (adapter->hw.mac_type >= e1000_82540) { ctrl = E1000_READ_REG(&adapter->hw, CTRL); /* advertise wake from D3Cold */ #define E1000_CTRL_ADVD3WUC 0x00100000 @@ -3682,7 +4501,7 @@ e1000_suspend(struct pci_dev *pdev, uint32_t state) E1000_WRITE_REG(&adapter->hw, CTRL, ctrl); } - if(adapter->hw.media_type == e1000_media_type_fiber || + if (adapter->hw.media_type == e1000_media_type_fiber || adapter->hw.media_type == e1000_media_type_internal_serdes) { /* keep the laser running in D3 */ ctrl_ext = E1000_READ_REG(&adapter->hw, CTRL_EXT); @@ -3695,42 +4514,48 @@ e1000_suspend(struct pci_dev *pdev, uint32_t state) E1000_WRITE_REG(&adapter->hw, WUC, E1000_WUC_PME_EN); E1000_WRITE_REG(&adapter->hw, WUFC, wufc); - pci_enable_wake(pdev, 3, 1); - pci_enable_wake(pdev, 4, 1); /* 4 == D3 cold */ + retval = pci_enable_wake(pdev, PCI_D3hot, 1); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 wake\n"); + retval = pci_enable_wake(pdev, PCI_D3cold, 1); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 cold wake\n"); } else { E1000_WRITE_REG(&adapter->hw, WUC, 0); E1000_WRITE_REG(&adapter->hw, WUFC, 0); - pci_enable_wake(pdev, 3, 0); - pci_enable_wake(pdev, 4, 0); /* 4 == D3 cold */ + retval = pci_enable_wake(pdev, PCI_D3hot, 0); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 wake\n"); + retval = pci_enable_wake(pdev, PCI_D3cold, 0); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 cold wake\n"); } - pci_save_state(pdev); - - if(adapter->hw.mac_type >= e1000_82540 && + if (adapter->hw.mac_type >= e1000_82540 && adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); - if(manc & E1000_MANC_SMBUS_EN) { + if (manc & E1000_MANC_SMBUS_EN) { manc |= E1000_MANC_ARP_EN; E1000_WRITE_REG(&adapter->hw, MANC, manc); - pci_enable_wake(pdev, 3, 1); - pci_enable_wake(pdev, 4, 1); /* 4 == D3 cold */ + retval = pci_enable_wake(pdev, PCI_D3hot, 1); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 wake\n"); + retval = pci_enable_wake(pdev, PCI_D3cold, 1); + if (retval) + DPRINTK(PROBE, ERR, + "Error enabling D3 cold wake\n"); } } - switch(adapter->hw.mac_type) { - case e1000_82573: - swsm = E1000_READ_REG(&adapter->hw, SWSM); - E1000_WRITE_REG(&adapter->hw, SWSM, - swsm & ~E1000_SWSM_DRV_LOAD); - break; - default: - break; - } + /* Release control of h/w to f/w. If f/w is AMT enabled, this + * would have already happened in close and is redundant. */ + e1000_release_hw_control(adapter); pci_disable_device(pdev); - state = (state > 0) ? 3 : 0; - pci_set_power_state(pdev, state); + retval = pci_set_power_state(pdev, pci_choose_state(pdev, state)); + if (retval) + DPRINTK(PROBE, ERR, "Error in setting power state\n"); return 0; } @@ -3740,45 +4565,56 @@ static int e1000_resume(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); - struct e1000_adapter *adapter = netdev->priv; - uint32_t manc, ret, swsm; + struct e1000_adapter *adapter = netdev_priv(netdev); + int retval; + uint32_t manc, ret_val; - pci_set_power_state(pdev, 0); - pci_restore_state(pdev); - ret = pci_enable_device(pdev); + retval = pci_set_power_state(pdev, PCI_D0); + if (retval) + DPRINTK(PROBE, ERR, "Error in setting power state\n"); + e1000_pci_restore_state(adapter); + ret_val = pci_enable_device(pdev); pci_set_master(pdev); - pci_enable_wake(pdev, 3, 0); - pci_enable_wake(pdev, 4, 0); /* 4 == D3 cold */ + retval = pci_enable_wake(pdev, PCI_D3hot, 0); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 wake\n"); + retval = pci_enable_wake(pdev, PCI_D3cold, 0); + if (retval) + DPRINTK(PROBE, ERR, "Error enabling D3 cold wake\n"); e1000_reset(adapter); E1000_WRITE_REG(&adapter->hw, WUS, ~0); - if(netif_running(netdev)) + if (netif_running(netdev)) e1000_up(adapter); netif_device_attach(netdev); - if(adapter->hw.mac_type >= e1000_82540 && + if (adapter->hw.mac_type >= e1000_82540 && adapter->hw.media_type == e1000_media_type_copper) { manc = E1000_READ_REG(&adapter->hw, MANC); manc &= ~(E1000_MANC_ARP_EN); E1000_WRITE_REG(&adapter->hw, MANC, manc); } - switch(adapter->hw.mac_type) { - case e1000_82573: - swsm = E1000_READ_REG(&adapter->hw, SWSM); - E1000_WRITE_REG(&adapter->hw, SWSM, - swsm | E1000_SWSM_DRV_LOAD); - break; - default: - break; - } + /* If the controller is 82573 and f/w is AMT, do not set + * DRV_LOAD until the interface is up. For all other cases, + * let the f/w know that the h/w is now under the control + * of the driver. */ + if (adapter->hw.mac_type != e1000_82573 || + !e1000_check_mng_mode(&adapter->hw)) + e1000_get_hw_control(adapter); return 0; } #endif + +static void e1000_shutdown(struct pci_dev *pdev) +{ + e1000_suspend(pdev, PMSG_SUSPEND); +} + #ifdef CONFIG_NET_POLL_CONTROLLER /* * Polling 'interrupt' - used by things like netconsole to send skbs @@ -3788,9 +4624,13 @@ e1000_resume(struct pci_dev *pdev) static void e1000_netpoll(struct net_device *netdev) { - struct e1000_adapter *adapter = netdev->priv; + struct e1000_adapter *adapter = netdev_priv(netdev); disable_irq(adapter->pdev->irq); e1000_intr(adapter->pdev->irq, netdev, NULL); + e1000_clean_tx_irq(adapter, adapter->tx_ring); +#ifndef CONFIG_E1000_NAPI + adapter->clean_rx(adapter, adapter->rx_ring); +#endif enable_irq(adapter->pdev->irq); } #endif