diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-07 18:49:45 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-07 18:49:45 +0000 |
commit | 2c3c1048746a4622d8c89a29670120dc8fab93c4 (patch) | |
tree | 848558de17fb3008cdf4d861b01ac7781903ce39 /drivers/net/ethernet/intel/igbvf | |
parent | Initial commit. (diff) | |
download | linux-2c3c1048746a4622d8c89a29670120dc8fab93c4.tar.xz linux-2c3c1048746a4622d8c89a29670120dc8fab93c4.zip |
Adding upstream version 6.1.76.upstream/6.1.76upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'drivers/net/ethernet/intel/igbvf')
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/Makefile | 13 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/defines.h | 98 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/ethtool.c | 463 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/igbvf.h | 299 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/mbx.c | 336 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/mbx.h | 55 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/netdev.c | 2980 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/regs.h | 84 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/vf.c | 430 | ||||
-rw-r--r-- | drivers/net/ethernet/intel/igbvf/vf.h | 243 |
10 files changed, 5001 insertions, 0 deletions
diff --git a/drivers/net/ethernet/intel/igbvf/Makefile b/drivers/net/ethernet/intel/igbvf/Makefile new file mode 100644 index 000000000..afd3e36ea --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/Makefile @@ -0,0 +1,13 @@ +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2009 - 2018 Intel Corporation. +# +# Makefile for the Intel(R) 82576 VF ethernet driver +# + +obj-$(CONFIG_IGBVF) += igbvf.o + +igbvf-objs := vf.o \ + mbx.o \ + ethtool.o \ + netdev.o + diff --git a/drivers/net/ethernet/intel/igbvf/defines.h b/drivers/net/ethernet/intel/igbvf/defines.h new file mode 100644 index 000000000..4437f8324 --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/defines.h @@ -0,0 +1,98 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 1999 - 2018 Intel Corporation. */ + +#ifndef _E1000_DEFINES_H_ +#define _E1000_DEFINES_H_ + +/* Number of Transmit and Receive Descriptors must be a multiple of 8 */ +#define REQ_TX_DESCRIPTOR_MULTIPLE 8 +#define REQ_RX_DESCRIPTOR_MULTIPLE 8 + +/* IVAR valid bit */ +#define E1000_IVAR_VALID 0x80 + +/* Receive Descriptor bit definitions */ +#define E1000_RXD_STAT_DD 0x01 /* Descriptor Done */ +#define E1000_RXD_STAT_EOP 0x02 /* End of Packet */ +#define E1000_RXD_STAT_IXSM 0x04 /* Ignore checksum */ +#define E1000_RXD_STAT_VP 0x08 /* IEEE VLAN Packet */ +#define E1000_RXD_STAT_UDPCS 0x10 /* UDP xsum calculated */ +#define E1000_RXD_STAT_TCPCS 0x20 /* TCP xsum calculated */ +#define E1000_RXD_STAT_IPCS 0x40 /* IP xsum calculated */ +#define E1000_RXD_ERR_SE 0x02 /* Symbol Error */ +#define E1000_RXD_SPC_VLAN_MASK 0x0FFF /* VLAN ID is in lower 12 bits */ + +#define E1000_RXDEXT_STATERR_LB 0x00040000 +#define E1000_RXDEXT_STATERR_CE 0x01000000 +#define E1000_RXDEXT_STATERR_SE 0x02000000 +#define E1000_RXDEXT_STATERR_SEQ 0x04000000 +#define E1000_RXDEXT_STATERR_CXE 0x10000000 +#define E1000_RXDEXT_STATERR_TCPE 0x20000000 +#define E1000_RXDEXT_STATERR_IPE 0x40000000 +#define E1000_RXDEXT_STATERR_RXE 0x80000000 + +/* Same mask, but for extended and packet split descriptors */ +#define E1000_RXDEXT_ERR_FRAME_ERR_MASK ( \ + E1000_RXDEXT_STATERR_CE | \ + E1000_RXDEXT_STATERR_SE | \ + E1000_RXDEXT_STATERR_SEQ | \ + E1000_RXDEXT_STATERR_CXE | \ + E1000_RXDEXT_STATERR_RXE) + +/* Device Control */ +#define E1000_CTRL_RST 0x04000000 /* Global reset */ + +/* Device Status */ +#define E1000_STATUS_FD 0x00000001 /* Full duplex.0=half,1=full */ +#define E1000_STATUS_LU 0x00000002 /* Link up.0=no,1=link */ +#define E1000_STATUS_TXOFF 0x00000010 /* transmission paused */ +#define E1000_STATUS_SPEED_10 0x00000000 /* Speed 10Mb/s */ +#define E1000_STATUS_SPEED_100 0x00000040 /* Speed 100Mb/s */ +#define E1000_STATUS_SPEED_1000 0x00000080 /* Speed 1000Mb/s */ + +#define SPEED_10 10 +#define SPEED_100 100 +#define SPEED_1000 1000 +#define HALF_DUPLEX 1 +#define FULL_DUPLEX 2 + +/* Transmit Descriptor bit definitions */ +#define E1000_TXD_POPTS_IXSM 0x01 /* Insert IP checksum */ +#define E1000_TXD_POPTS_TXSM 0x02 /* Insert TCP/UDP checksum */ +#define E1000_TXD_CMD_DEXT 0x20000000 /* Desc extension (0 = legacy) */ +#define E1000_TXD_STAT_DD 0x00000001 /* Desc Done */ + +#define MAX_JUMBO_FRAME_SIZE 0x3F00 +#define MAX_STD_JUMBO_FRAME_SIZE 9216 + +/* 802.1q VLAN Packet Size */ +#define VLAN_TAG_SIZE 4 /* 802.3ac tag (not DMA'd) */ + +/* Error Codes */ +#define E1000_SUCCESS 0 +#define E1000_ERR_CONFIG 3 +#define E1000_ERR_MAC_INIT 5 +#define E1000_ERR_MBX 15 + +/* SRRCTL bit definitions */ +#define E1000_SRRCTL_BSIZEPKT_SHIFT 10 /* Shift _right_ */ +#define E1000_SRRCTL_BSIZEHDRSIZE_MASK 0x00000F00 +#define E1000_SRRCTL_BSIZEHDRSIZE_SHIFT 2 /* Shift _left_ */ +#define E1000_SRRCTL_DESCTYPE_ADV_ONEBUF 0x02000000 +#define E1000_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS 0x0A000000 +#define E1000_SRRCTL_DESCTYPE_MASK 0x0E000000 +#define E1000_SRRCTL_DROP_EN 0x80000000 + +#define E1000_SRRCTL_BSIZEPKT_MASK 0x0000007F +#define E1000_SRRCTL_BSIZEHDR_MASK 0x00003F00 + +/* Additional Descriptor Control definitions */ +#define E1000_TXDCTL_QUEUE_ENABLE 0x02000000 /* Enable specific Tx Que */ +#define E1000_RXDCTL_QUEUE_ENABLE 0x02000000 /* Enable specific Rx Que */ + +/* Direct Cache Access (DCA) definitions */ +#define E1000_DCA_TXCTRL_TX_WB_RO_EN BIT(11) /* Tx Desc writeback RO bit */ + +#define E1000_VF_INIT_TIMEOUT 200 /* Number of retries to clear RSTI */ + +#endif /* _E1000_DEFINES_H_ */ diff --git a/drivers/net/ethernet/intel/igbvf/ethtool.c b/drivers/net/ethernet/intel/igbvf/ethtool.c new file mode 100644 index 000000000..83b97989a --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/ethtool.c @@ -0,0 +1,463 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +/* ethtool support for igbvf */ + +#include <linux/netdevice.h> +#include <linux/ethtool.h> +#include <linux/pci.h> +#include <linux/vmalloc.h> +#include <linux/delay.h> + +#include "igbvf.h" +#include <linux/if_vlan.h> + +struct igbvf_stats { + char stat_string[ETH_GSTRING_LEN]; + int sizeof_stat; + int stat_offset; + int base_stat_offset; +}; + +#define IGBVF_STAT(current, base) \ + sizeof(((struct igbvf_adapter *)0)->current), \ + offsetof(struct igbvf_adapter, current), \ + offsetof(struct igbvf_adapter, base) + +static const struct igbvf_stats igbvf_gstrings_stats[] = { + { "rx_packets", IGBVF_STAT(stats.gprc, stats.base_gprc) }, + { "tx_packets", IGBVF_STAT(stats.gptc, stats.base_gptc) }, + { "rx_bytes", IGBVF_STAT(stats.gorc, stats.base_gorc) }, + { "tx_bytes", IGBVF_STAT(stats.gotc, stats.base_gotc) }, + { "multicast", IGBVF_STAT(stats.mprc, stats.base_mprc) }, + { "lbrx_bytes", IGBVF_STAT(stats.gorlbc, stats.base_gorlbc) }, + { "lbrx_packets", IGBVF_STAT(stats.gprlbc, stats.base_gprlbc) }, + { "tx_restart_queue", IGBVF_STAT(restart_queue, zero_base) }, + { "rx_long_byte_count", IGBVF_STAT(stats.gorc, stats.base_gorc) }, + { "rx_csum_offload_good", IGBVF_STAT(hw_csum_good, zero_base) }, + { "rx_csum_offload_errors", IGBVF_STAT(hw_csum_err, zero_base) }, + { "rx_header_split", IGBVF_STAT(rx_hdr_split, zero_base) }, + { "alloc_rx_buff_failed", IGBVF_STAT(alloc_rx_buff_failed, zero_base) }, +}; + +#define IGBVF_GLOBAL_STATS_LEN ARRAY_SIZE(igbvf_gstrings_stats) + +static const char igbvf_gstrings_test[][ETH_GSTRING_LEN] = { + "Link test (on/offline)" +}; + +#define IGBVF_TEST_LEN ARRAY_SIZE(igbvf_gstrings_test) + +static int igbvf_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *cmd) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + u32 status; + + ethtool_link_ksettings_zero_link_mode(cmd, supported); + ethtool_link_ksettings_add_link_mode(cmd, supported, 1000baseT_Full); + ethtool_link_ksettings_zero_link_mode(cmd, advertising); + ethtool_link_ksettings_add_link_mode(cmd, advertising, 1000baseT_Full); + + cmd->base.port = -1; + + status = er32(STATUS); + if (status & E1000_STATUS_LU) { + if (status & E1000_STATUS_SPEED_1000) + cmd->base.speed = SPEED_1000; + else if (status & E1000_STATUS_SPEED_100) + cmd->base.speed = SPEED_100; + else + cmd->base.speed = SPEED_10; + + if (status & E1000_STATUS_FD) + cmd->base.duplex = DUPLEX_FULL; + else + cmd->base.duplex = DUPLEX_HALF; + } else { + cmd->base.speed = SPEED_UNKNOWN; + cmd->base.duplex = DUPLEX_UNKNOWN; + } + + cmd->base.autoneg = AUTONEG_DISABLE; + + return 0; +} + +static int igbvf_set_link_ksettings(struct net_device *netdev, + const struct ethtool_link_ksettings *cmd) +{ + return -EOPNOTSUPP; +} + +static void igbvf_get_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ +} + +static int igbvf_set_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ + return -EOPNOTSUPP; +} + +static u32 igbvf_get_msglevel(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + return adapter->msg_enable; +} + +static void igbvf_set_msglevel(struct net_device *netdev, u32 data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + adapter->msg_enable = data; +} + +static int igbvf_get_regs_len(struct net_device *netdev) +{ +#define IGBVF_REGS_LEN 8 + return IGBVF_REGS_LEN * sizeof(u32); +} + +static void igbvf_get_regs(struct net_device *netdev, + struct ethtool_regs *regs, void *p) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + u32 *regs_buff = p; + + memset(p, 0, IGBVF_REGS_LEN * sizeof(u32)); + + regs->version = (1u << 24) | + (adapter->pdev->revision << 16) | + adapter->pdev->device; + + regs_buff[0] = er32(CTRL); + regs_buff[1] = er32(STATUS); + + regs_buff[2] = er32(RDLEN(0)); + regs_buff[3] = er32(RDH(0)); + regs_buff[4] = er32(RDT(0)); + + regs_buff[5] = er32(TDLEN(0)); + regs_buff[6] = er32(TDH(0)); + regs_buff[7] = er32(TDT(0)); +} + +static int igbvf_get_eeprom_len(struct net_device *netdev) +{ + return 0; +} + +static int igbvf_get_eeprom(struct net_device *netdev, + struct ethtool_eeprom *eeprom, u8 *bytes) +{ + return -EOPNOTSUPP; +} + +static int igbvf_set_eeprom(struct net_device *netdev, + struct ethtool_eeprom *eeprom, u8 *bytes) +{ + return -EOPNOTSUPP; +} + +static void igbvf_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *drvinfo) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + strscpy(drvinfo->driver, igbvf_driver_name, sizeof(drvinfo->driver)); + strscpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); +} + +static void igbvf_get_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring, + struct kernel_ethtool_ringparam *kernel_ring, + struct netlink_ext_ack *extack) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct igbvf_ring *rx_ring = adapter->rx_ring; + + ring->rx_max_pending = IGBVF_MAX_RXD; + ring->tx_max_pending = IGBVF_MAX_TXD; + ring->rx_pending = rx_ring->count; + ring->tx_pending = tx_ring->count; +} + +static int igbvf_set_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring, + struct kernel_ethtool_ringparam *kernel_ring, + struct netlink_ext_ack *extack) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct igbvf_ring *temp_ring; + int err = 0; + u32 new_rx_count, new_tx_count; + + if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending)) + return -EINVAL; + + new_rx_count = max_t(u32, ring->rx_pending, IGBVF_MIN_RXD); + new_rx_count = min_t(u32, new_rx_count, IGBVF_MAX_RXD); + new_rx_count = ALIGN(new_rx_count, REQ_RX_DESCRIPTOR_MULTIPLE); + + new_tx_count = max_t(u32, ring->tx_pending, IGBVF_MIN_TXD); + new_tx_count = min_t(u32, new_tx_count, IGBVF_MAX_TXD); + new_tx_count = ALIGN(new_tx_count, REQ_TX_DESCRIPTOR_MULTIPLE); + + if ((new_tx_count == adapter->tx_ring->count) && + (new_rx_count == adapter->rx_ring->count)) { + /* nothing to do */ + return 0; + } + + while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state)) + usleep_range(1000, 2000); + + if (!netif_running(adapter->netdev)) { + adapter->tx_ring->count = new_tx_count; + adapter->rx_ring->count = new_rx_count; + goto clear_reset; + } + + temp_ring = vmalloc(sizeof(struct igbvf_ring)); + if (!temp_ring) { + err = -ENOMEM; + goto clear_reset; + } + + igbvf_down(adapter); + + /* We can't just free everything and then setup again, + * because the ISRs in MSI-X mode get passed pointers + * to the Tx and Rx ring structs. + */ + if (new_tx_count != adapter->tx_ring->count) { + memcpy(temp_ring, adapter->tx_ring, sizeof(struct igbvf_ring)); + + temp_ring->count = new_tx_count; + err = igbvf_setup_tx_resources(adapter, temp_ring); + if (err) + goto err_setup; + + igbvf_free_tx_resources(adapter->tx_ring); + + memcpy(adapter->tx_ring, temp_ring, sizeof(struct igbvf_ring)); + } + + if (new_rx_count != adapter->rx_ring->count) { + memcpy(temp_ring, adapter->rx_ring, sizeof(struct igbvf_ring)); + + temp_ring->count = new_rx_count; + err = igbvf_setup_rx_resources(adapter, temp_ring); + if (err) + goto err_setup; + + igbvf_free_rx_resources(adapter->rx_ring); + + memcpy(adapter->rx_ring, temp_ring, sizeof(struct igbvf_ring)); + } +err_setup: + igbvf_up(adapter); + vfree(temp_ring); +clear_reset: + clear_bit(__IGBVF_RESETTING, &adapter->state); + return err; +} + +static int igbvf_link_test(struct igbvf_adapter *adapter, u64 *data) +{ + struct e1000_hw *hw = &adapter->hw; + *data = 0; + + spin_lock_bh(&hw->mbx_lock); + + hw->mac.ops.check_for_link(hw); + + spin_unlock_bh(&hw->mbx_lock); + + if (!(er32(STATUS) & E1000_STATUS_LU)) + *data = 1; + + return *data; +} + +static void igbvf_diag_test(struct net_device *netdev, + struct ethtool_test *eth_test, u64 *data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + set_bit(__IGBVF_TESTING, &adapter->state); + + /* Link test performed before hardware reset so autoneg doesn't + * interfere with test result + */ + if (igbvf_link_test(adapter, &data[0])) + eth_test->flags |= ETH_TEST_FL_FAILED; + + clear_bit(__IGBVF_TESTING, &adapter->state); + msleep_interruptible(4 * 1000); +} + +static void igbvf_get_wol(struct net_device *netdev, + struct ethtool_wolinfo *wol) +{ + wol->supported = 0; + wol->wolopts = 0; +} + +static int igbvf_set_wol(struct net_device *netdev, + struct ethtool_wolinfo *wol) +{ + return -EOPNOTSUPP; +} + +static int igbvf_get_coalesce(struct net_device *netdev, + struct ethtool_coalesce *ec, + struct kernel_ethtool_coalesce *kernel_coal, + struct netlink_ext_ack *extack) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (adapter->requested_itr <= 3) + ec->rx_coalesce_usecs = adapter->requested_itr; + else + ec->rx_coalesce_usecs = adapter->current_itr >> 2; + + return 0; +} + +static int igbvf_set_coalesce(struct net_device *netdev, + struct ethtool_coalesce *ec, + struct kernel_ethtool_coalesce *kernel_coal, + struct netlink_ext_ack *extack) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + if ((ec->rx_coalesce_usecs >= IGBVF_MIN_ITR_USECS) && + (ec->rx_coalesce_usecs <= IGBVF_MAX_ITR_USECS)) { + adapter->current_itr = ec->rx_coalesce_usecs << 2; + adapter->requested_itr = 1000000000 / + (adapter->current_itr * 256); + } else if ((ec->rx_coalesce_usecs == 3) || + (ec->rx_coalesce_usecs == 2)) { + adapter->current_itr = IGBVF_START_ITR; + adapter->requested_itr = ec->rx_coalesce_usecs; + } else if (ec->rx_coalesce_usecs == 0) { + /* The user's desire is to turn off interrupt throttling + * altogether, but due to HW limitations, we can't do that. + * Instead we set a very small value in EITR, which would + * allow ~967k interrupts per second, but allow the adapter's + * internal clocking to still function properly. + */ + adapter->current_itr = 4; + adapter->requested_itr = 1000000000 / + (adapter->current_itr * 256); + } else { + return -EINVAL; + } + + writel(adapter->current_itr, + hw->hw_addr + adapter->rx_ring->itr_register); + + return 0; +} + +static int igbvf_nway_reset(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (netif_running(netdev)) + igbvf_reinit_locked(adapter); + return 0; +} + +static void igbvf_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, + u64 *data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + int i; + + igbvf_update_stats(adapter); + for (i = 0; i < IGBVF_GLOBAL_STATS_LEN; i++) { + char *p = (char *)adapter + + igbvf_gstrings_stats[i].stat_offset; + char *b = (char *)adapter + + igbvf_gstrings_stats[i].base_stat_offset; + data[i] = ((igbvf_gstrings_stats[i].sizeof_stat == + sizeof(u64)) ? (*(u64 *)p - *(u64 *)b) : + (*(u32 *)p - *(u32 *)b)); + } +} + +static int igbvf_get_sset_count(struct net_device *dev, int stringset) +{ + switch (stringset) { + case ETH_SS_TEST: + return IGBVF_TEST_LEN; + case ETH_SS_STATS: + return IGBVF_GLOBAL_STATS_LEN; + default: + return -EINVAL; + } +} + +static void igbvf_get_strings(struct net_device *netdev, u32 stringset, + u8 *data) +{ + u8 *p = data; + int i; + + switch (stringset) { + case ETH_SS_TEST: + memcpy(data, *igbvf_gstrings_test, sizeof(igbvf_gstrings_test)); + break; + case ETH_SS_STATS: + for (i = 0; i < IGBVF_GLOBAL_STATS_LEN; i++) { + memcpy(p, igbvf_gstrings_stats[i].stat_string, + ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + break; + } +} + +static const struct ethtool_ops igbvf_ethtool_ops = { + .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS, + .get_drvinfo = igbvf_get_drvinfo, + .get_regs_len = igbvf_get_regs_len, + .get_regs = igbvf_get_regs, + .get_wol = igbvf_get_wol, + .set_wol = igbvf_set_wol, + .get_msglevel = igbvf_get_msglevel, + .set_msglevel = igbvf_set_msglevel, + .nway_reset = igbvf_nway_reset, + .get_link = ethtool_op_get_link, + .get_eeprom_len = igbvf_get_eeprom_len, + .get_eeprom = igbvf_get_eeprom, + .set_eeprom = igbvf_set_eeprom, + .get_ringparam = igbvf_get_ringparam, + .set_ringparam = igbvf_set_ringparam, + .get_pauseparam = igbvf_get_pauseparam, + .set_pauseparam = igbvf_set_pauseparam, + .self_test = igbvf_diag_test, + .get_sset_count = igbvf_get_sset_count, + .get_strings = igbvf_get_strings, + .get_ethtool_stats = igbvf_get_ethtool_stats, + .get_coalesce = igbvf_get_coalesce, + .set_coalesce = igbvf_set_coalesce, + .get_link_ksettings = igbvf_get_link_ksettings, + .set_link_ksettings = igbvf_set_link_ksettings, +}; + +void igbvf_set_ethtool_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &igbvf_ethtool_ops; +} diff --git a/drivers/net/ethernet/intel/igbvf/igbvf.h b/drivers/net/ethernet/intel/igbvf/igbvf.h new file mode 100644 index 000000000..7b83678ba --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/igbvf.h @@ -0,0 +1,299 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +/* Linux PRO/1000 Ethernet Driver main header file */ + +#ifndef _IGBVF_H_ +#define _IGBVF_H_ + +#include <linux/types.h> +#include <linux/timer.h> +#include <linux/io.h> +#include <linux/netdevice.h> +#include <linux/if_vlan.h> + +#include "vf.h" + +/* Forward declarations */ +struct igbvf_info; +struct igbvf_adapter; + +/* Interrupt defines */ +#define IGBVF_START_ITR 488 /* ~8000 ints/sec */ +#define IGBVF_4K_ITR 980 +#define IGBVF_20K_ITR 196 +#define IGBVF_70K_ITR 56 + +enum latency_range { + lowest_latency = 0, + low_latency = 1, + bulk_latency = 2, + latency_invalid = 255 +}; + +/* Interrupt modes, as used by the IntMode parameter */ +#define IGBVF_INT_MODE_LEGACY 0 +#define IGBVF_INT_MODE_MSI 1 +#define IGBVF_INT_MODE_MSIX 2 + +/* Tx/Rx descriptor defines */ +#define IGBVF_DEFAULT_TXD 256 +#define IGBVF_MAX_TXD 4096 +#define IGBVF_MIN_TXD 64 + +#define IGBVF_DEFAULT_RXD 256 +#define IGBVF_MAX_RXD 4096 +#define IGBVF_MIN_RXD 64 + +#define IGBVF_MIN_ITR_USECS 10 /* 100000 irq/sec */ +#define IGBVF_MAX_ITR_USECS 10000 /* 100 irq/sec */ + +/* RX descriptor control thresholds. + * PTHRESH - MAC will consider prefetch if it has fewer than this number of + * descriptors available in its onboard memory. + * Setting this to 0 disables RX descriptor prefetch. + * HTHRESH - MAC will only prefetch if there are at least this many descriptors + * available in host memory. + * If PTHRESH is 0, this should also be 0. + * WTHRESH - RX descriptor writeback threshold - MAC will delay writing back + * descriptors until either it has this many to write back, or the + * ITR timer expires. + */ +#define IGBVF_RX_PTHRESH 16 +#define IGBVF_RX_HTHRESH 8 +#define IGBVF_RX_WTHRESH 1 + +/* this is the size past which hardware will drop packets when setting LPE=0 */ +#define MAXIMUM_ETHERNET_VLAN_SIZE 1522 + +#define IGBVF_FC_PAUSE_TIME 0x0680 /* 858 usec */ + +/* How many Tx Descriptors do we need to call netif_wake_queue ? */ +#define IGBVF_TX_QUEUE_WAKE 32 +/* How many Rx Buffers do we bundle into one write to the hardware ? */ +#define IGBVF_RX_BUFFER_WRITE 16 /* Must be power of 2 */ + +#define AUTO_ALL_MODES 0 +#define IGBVF_EEPROM_APME 0x0400 + +#define IGBVF_MNG_VLAN_NONE (-1) + +#define IGBVF_MAX_MAC_FILTERS 3 + +/* Number of packet split data buffers (not including the header buffer) */ +#define PS_PAGE_BUFFERS (MAX_PS_BUFFERS - 1) + +enum igbvf_boards { + board_vf, + board_i350_vf, +}; + +struct igbvf_queue_stats { + u64 packets; + u64 bytes; +}; + +/* wrappers around a pointer to a socket buffer, + * so a DMA handle can be stored along with the buffer + */ +struct igbvf_buffer { + dma_addr_t dma; + struct sk_buff *skb; + union { + /* Tx */ + struct { + unsigned long time_stamp; + union e1000_adv_tx_desc *next_to_watch; + u16 length; + u16 mapped_as_page; + }; + /* Rx */ + struct { + struct page *page; + u64 page_dma; + unsigned int page_offset; + }; + }; +}; + +union igbvf_desc { + union e1000_adv_rx_desc rx_desc; + union e1000_adv_tx_desc tx_desc; + struct e1000_adv_tx_context_desc tx_context_desc; +}; + +struct igbvf_ring { + struct igbvf_adapter *adapter; /* backlink */ + union igbvf_desc *desc; /* pointer to ring memory */ + dma_addr_t dma; /* phys address of ring */ + unsigned int size; /* length of ring in bytes */ + unsigned int count; /* number of desc. in ring */ + + u16 next_to_use; + u16 next_to_clean; + + u16 head; + u16 tail; + + /* array of buffer information structs */ + struct igbvf_buffer *buffer_info; + struct napi_struct napi; + + char name[IFNAMSIZ + 5]; + u32 eims_value; + u32 itr_val; + enum latency_range itr_range; + u16 itr_register; + int set_itr; + + struct sk_buff *rx_skb_top; + + struct igbvf_queue_stats stats; +}; + +/* board specific private data structure */ +struct igbvf_adapter { + struct timer_list watchdog_timer; + struct timer_list blink_timer; + + struct work_struct reset_task; + struct work_struct watchdog_task; + + const struct igbvf_info *ei; + + unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; + u32 bd_number; + u32 rx_buffer_len; + u32 polling_interval; + u16 mng_vlan_id; + u16 link_speed; + u16 link_duplex; + + spinlock_t tx_queue_lock; /* prevent concurrent tail updates */ + + /* track device up/down/testing state */ + unsigned long state; + + /* Interrupt Throttle Rate */ + u32 requested_itr; /* ints/sec or adaptive */ + u32 current_itr; /* Actual ITR register value, not ints/sec */ + + /* Tx */ + struct igbvf_ring *tx_ring /* One per active queue */ + ____cacheline_aligned_in_smp; + + unsigned int restart_queue; + u32 txd_cmd; + + u32 tx_int_delay; + u32 tx_abs_int_delay; + + unsigned int total_tx_bytes; + unsigned int total_tx_packets; + unsigned int total_rx_bytes; + unsigned int total_rx_packets; + + /* Tx stats */ + u32 tx_timeout_count; + u32 tx_fifo_head; + u32 tx_head_addr; + u32 tx_fifo_size; + u32 tx_dma_failed; + + /* Rx */ + struct igbvf_ring *rx_ring; + + u32 rx_int_delay; + u32 rx_abs_int_delay; + + /* Rx stats */ + u64 hw_csum_err; + u64 hw_csum_good; + u64 rx_hdr_split; + u32 alloc_rx_buff_failed; + u32 rx_dma_failed; + + unsigned int rx_ps_hdr_size; + u32 max_frame_size; + u32 min_frame_size; + + /* OS defined structs */ + struct net_device *netdev; + struct pci_dev *pdev; + spinlock_t stats_lock; /* prevent concurrent stats updates */ + + /* structs defined in e1000_hw.h */ + struct e1000_hw hw; + + /* The VF counters don't clear on read so we have to get a base + * count on driver start up and always subtract that base on + * the first update, thus the flag.. + */ + struct e1000_vf_stats stats; + u64 zero_base; + + struct igbvf_ring test_tx_ring; + struct igbvf_ring test_rx_ring; + u32 test_icr; + + u32 msg_enable; + struct msix_entry *msix_entries; + int int_mode; + u32 eims_enable_mask; + u32 eims_other; + u32 int_counter0; + u32 int_counter1; + + u32 eeprom_wol; + u32 wol; + u32 pba; + + bool fc_autoneg; + + unsigned long led_status; + + unsigned int flags; + unsigned long last_reset; +}; + +struct igbvf_info { + enum e1000_mac_type mac; + unsigned int flags; + u32 pba; + void (*init_ops)(struct e1000_hw *); + s32 (*get_variants)(struct igbvf_adapter *); +}; + +/* hardware capability, feature, and workaround flags */ +#define IGBVF_FLAG_RX_CSUM_DISABLED BIT(0) +#define IGBVF_FLAG_RX_LB_VLAN_BSWAP BIT(1) +#define IGBVF_RX_DESC_ADV(R, i) \ + (&((((R).desc))[i].rx_desc)) +#define IGBVF_TX_DESC_ADV(R, i) \ + (&((((R).desc))[i].tx_desc)) +#define IGBVF_TX_CTXTDESC_ADV(R, i) \ + (&((((R).desc))[i].tx_context_desc)) + +enum igbvf_state_t { + __IGBVF_TESTING, + __IGBVF_RESETTING, + __IGBVF_DOWN +}; + +extern char igbvf_driver_name[]; + +void igbvf_check_options(struct igbvf_adapter *); +void igbvf_set_ethtool_ops(struct net_device *); + +int igbvf_up(struct igbvf_adapter *); +void igbvf_down(struct igbvf_adapter *); +void igbvf_reinit_locked(struct igbvf_adapter *); +int igbvf_setup_rx_resources(struct igbvf_adapter *, struct igbvf_ring *); +int igbvf_setup_tx_resources(struct igbvf_adapter *, struct igbvf_ring *); +void igbvf_free_rx_resources(struct igbvf_ring *); +void igbvf_free_tx_resources(struct igbvf_ring *); +void igbvf_update_stats(struct igbvf_adapter *); + +extern unsigned int copybreak; + +#endif /* _IGBVF_H_ */ diff --git a/drivers/net/ethernet/intel/igbvf/mbx.c b/drivers/net/ethernet/intel/igbvf/mbx.c new file mode 100644 index 000000000..a3cd7ac48 --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/mbx.c @@ -0,0 +1,336 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +#include "mbx.h" + +/** + * e1000_poll_for_msg - Wait for message notification + * @hw: pointer to the HW structure + * + * returns SUCCESS if it successfully received a message notification + **/ +static s32 e1000_poll_for_msg(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; + + if (!mbx->ops.check_for_msg) + goto out; + + while (countdown && mbx->ops.check_for_msg(hw)) { + countdown--; + udelay(mbx->usec_delay); + } + + /* if we failed, all future posted messages fail until reset */ + if (!countdown) + mbx->timeout = 0; +out: + return countdown ? E1000_SUCCESS : -E1000_ERR_MBX; +} + +/** + * e1000_poll_for_ack - Wait for message acknowledgment + * @hw: pointer to the HW structure + * + * returns SUCCESS if it successfully received a message acknowledgment + **/ +static s32 e1000_poll_for_ack(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; + + if (!mbx->ops.check_for_ack) + goto out; + + while (countdown && mbx->ops.check_for_ack(hw)) { + countdown--; + udelay(mbx->usec_delay); + } + + /* if we failed, all future posted messages fail until reset */ + if (!countdown) + mbx->timeout = 0; +out: + return countdown ? E1000_SUCCESS : -E1000_ERR_MBX; +} + +/** + * e1000_read_posted_mbx - Wait for message notification and receive message + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully received a message notification and + * copied it into the receive buffer. + **/ +static s32 e1000_read_posted_mbx(struct e1000_hw *hw, u32 *msg, u16 size) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + s32 ret_val = -E1000_ERR_MBX; + + if (!mbx->ops.read) + goto out; + + ret_val = e1000_poll_for_msg(hw); + + /* if ack received read message, otherwise we timed out */ + if (!ret_val) + ret_val = mbx->ops.read(hw, msg, size); +out: + return ret_val; +} + +/** + * e1000_write_posted_mbx - Write a message to the mailbox, wait for ack + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully copied message into the buffer and + * received an ack to that message within delay * timeout period + **/ +static s32 e1000_write_posted_mbx(struct e1000_hw *hw, u32 *msg, u16 size) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + s32 ret_val = -E1000_ERR_MBX; + + /* exit if we either can't write or there isn't a defined timeout */ + if (!mbx->ops.write || !mbx->timeout) + goto out; + + /* send msg*/ + ret_val = mbx->ops.write(hw, msg, size); + + /* if msg sent wait until we receive an ack */ + if (!ret_val) + ret_val = e1000_poll_for_ack(hw); +out: + return ret_val; +} + +/** + * e1000_read_v2p_mailbox - read v2p mailbox + * @hw: pointer to the HW structure + * + * This function is used to read the v2p mailbox without losing the read to + * clear status bits. + **/ +static u32 e1000_read_v2p_mailbox(struct e1000_hw *hw) +{ + u32 v2p_mailbox = er32(V2PMAILBOX(0)); + + v2p_mailbox |= hw->dev_spec.vf.v2p_mailbox; + hw->dev_spec.vf.v2p_mailbox |= v2p_mailbox & E1000_V2PMAILBOX_R2C_BITS; + + return v2p_mailbox; +} + +/** + * e1000_check_for_bit_vf - Determine if a status bit was set + * @hw: pointer to the HW structure + * @mask: bitmask for bits to be tested and cleared + * + * This function is used to check for the read to clear bits within + * the V2P mailbox. + **/ +static s32 e1000_check_for_bit_vf(struct e1000_hw *hw, u32 mask) +{ + u32 v2p_mailbox = e1000_read_v2p_mailbox(hw); + s32 ret_val = -E1000_ERR_MBX; + + if (v2p_mailbox & mask) + ret_val = E1000_SUCCESS; + + hw->dev_spec.vf.v2p_mailbox &= ~mask; + + return ret_val; +} + +/** + * e1000_check_for_msg_vf - checks to see if the PF has sent mail + * @hw: pointer to the HW structure + * + * returns SUCCESS if the PF has set the Status bit or else ERR_MBX + **/ +static s32 e1000_check_for_msg_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + if (!e1000_check_for_bit_vf(hw, E1000_V2PMAILBOX_PFSTS)) { + ret_val = E1000_SUCCESS; + hw->mbx.stats.reqs++; + } + + return ret_val; +} + +/** + * e1000_check_for_ack_vf - checks to see if the PF has ACK'd + * @hw: pointer to the HW structure + * + * returns SUCCESS if the PF has set the ACK bit or else ERR_MBX + **/ +static s32 e1000_check_for_ack_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + if (!e1000_check_for_bit_vf(hw, E1000_V2PMAILBOX_PFACK)) { + ret_val = E1000_SUCCESS; + hw->mbx.stats.acks++; + } + + return ret_val; +} + +/** + * e1000_check_for_rst_vf - checks to see if the PF has reset + * @hw: pointer to the HW structure + * + * returns true if the PF has set the reset done bit or else false + **/ +static s32 e1000_check_for_rst_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + if (!e1000_check_for_bit_vf(hw, (E1000_V2PMAILBOX_RSTD | + E1000_V2PMAILBOX_RSTI))) { + ret_val = E1000_SUCCESS; + hw->mbx.stats.rsts++; + } + + return ret_val; +} + +/** + * e1000_obtain_mbx_lock_vf - obtain mailbox lock + * @hw: pointer to the HW structure + * + * return SUCCESS if we obtained the mailbox lock + **/ +static s32 e1000_obtain_mbx_lock_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + int count = 10; + + do { + /* Take ownership of the buffer */ + ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_VFU); + + /* reserve mailbox for VF use */ + if (e1000_read_v2p_mailbox(hw) & E1000_V2PMAILBOX_VFU) { + ret_val = 0; + break; + } + udelay(1000); + } while (count-- > 0); + + return ret_val; +} + +/** + * e1000_write_mbx_vf - Write a message to the mailbox + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully copied message into the buffer + **/ +static s32 e1000_write_mbx_vf(struct e1000_hw *hw, u32 *msg, u16 size) +{ + s32 err; + u16 i; + + lockdep_assert_held(&hw->mbx_lock); + + /* lock the mailbox to prevent pf/vf race condition */ + err = e1000_obtain_mbx_lock_vf(hw); + if (err) + goto out_no_write; + + /* flush any ack or msg as we are going to overwrite mailbox */ + e1000_check_for_ack_vf(hw); + e1000_check_for_msg_vf(hw); + + /* copy the caller specified message to the mailbox memory buffer */ + for (i = 0; i < size; i++) + array_ew32(VMBMEM(0), i, msg[i]); + + /* update stats */ + hw->mbx.stats.msgs_tx++; + + /* Drop VFU and interrupt the PF to tell it a message has been sent */ + ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_REQ); + +out_no_write: + return err; +} + +/** + * e1000_read_mbx_vf - Reads a message from the inbox intended for VF + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully read message from buffer + **/ +static s32 e1000_read_mbx_vf(struct e1000_hw *hw, u32 *msg, u16 size) +{ + s32 err; + u16 i; + + lockdep_assert_held(&hw->mbx_lock); + + /* lock the mailbox to prevent pf/vf race condition */ + err = e1000_obtain_mbx_lock_vf(hw); + if (err) + goto out_no_read; + + /* copy the message from the mailbox memory buffer */ + for (i = 0; i < size; i++) + msg[i] = array_er32(VMBMEM(0), i); + + /* Acknowledge receipt and release mailbox, then we're done */ + ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_ACK); + + /* update stats */ + hw->mbx.stats.msgs_rx++; + +out_no_read: + return err; +} + +/** + * e1000_init_mbx_params_vf - set initial values for VF mailbox + * @hw: pointer to the HW structure + * + * Initializes the hw->mbx struct to correct values for VF mailbox + */ +s32 e1000_init_mbx_params_vf(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + + /* start mailbox as timed out and let the reset_hw call set the timeout + * value to being communications + */ + mbx->timeout = 0; + mbx->usec_delay = E1000_VF_MBX_INIT_DELAY; + + mbx->size = E1000_VFMAILBOX_SIZE; + + mbx->ops.read = e1000_read_mbx_vf; + mbx->ops.write = e1000_write_mbx_vf; + mbx->ops.read_posted = e1000_read_posted_mbx; + mbx->ops.write_posted = e1000_write_posted_mbx; + mbx->ops.check_for_msg = e1000_check_for_msg_vf; + mbx->ops.check_for_ack = e1000_check_for_ack_vf; + mbx->ops.check_for_rst = e1000_check_for_rst_vf; + + mbx->stats.msgs_tx = 0; + mbx->stats.msgs_rx = 0; + mbx->stats.reqs = 0; + mbx->stats.acks = 0; + mbx->stats.rsts = 0; + + return E1000_SUCCESS; +} diff --git a/drivers/net/ethernet/intel/igbvf/mbx.h b/drivers/net/ethernet/intel/igbvf/mbx.h new file mode 100644 index 000000000..e5b31818d --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/mbx.h @@ -0,0 +1,55 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 1999 - 2018 Intel Corporation. */ + +#ifndef _E1000_MBX_H_ +#define _E1000_MBX_H_ + +#include "vf.h" + +#define E1000_V2PMAILBOX_REQ 0x00000001 /* Request for PF Ready bit */ +#define E1000_V2PMAILBOX_ACK 0x00000002 /* Ack PF message received */ +#define E1000_V2PMAILBOX_VFU 0x00000004 /* VF owns the mailbox buffer */ +#define E1000_V2PMAILBOX_PFU 0x00000008 /* PF owns the mailbox buffer */ +#define E1000_V2PMAILBOX_PFSTS 0x00000010 /* PF wrote a message in the MB */ +#define E1000_V2PMAILBOX_PFACK 0x00000020 /* PF ack the previous VF msg */ +#define E1000_V2PMAILBOX_RSTI 0x00000040 /* PF has reset indication */ +#define E1000_V2PMAILBOX_RSTD 0x00000080 /* PF has indicated reset done */ +#define E1000_V2PMAILBOX_R2C_BITS 0x000000B0 /* All read to clear bits */ + +#define E1000_VFMAILBOX_SIZE 16 /* 16 32 bit words - 64 bytes */ + +/* If it's a E1000_VF_* msg then it originates in the VF and is sent to the + * PF. The reverse is true if it is E1000_PF_*. + * Message ACK's are the value or'd with 0xF0000000 + */ +/* Messages below or'd with this are the ACK */ +#define E1000_VT_MSGTYPE_ACK 0x80000000 +/* Messages below or'd with this are the NACK */ +#define E1000_VT_MSGTYPE_NACK 0x40000000 +/* Indicates that VF is still clear to send requests */ +#define E1000_VT_MSGTYPE_CTS 0x20000000 + +/* We have a total wait time of 1s for vf mailbox posted messages */ +#define E1000_VF_MBX_INIT_TIMEOUT 2000 /* retry count for mbx timeout */ +#define E1000_VF_MBX_INIT_DELAY 500 /* usec delay between retries */ + +#define E1000_VT_MSGINFO_SHIFT 16 +/* bits 23:16 are used for exra info for certain messages */ +#define E1000_VT_MSGINFO_MASK (0xFF << E1000_VT_MSGINFO_SHIFT) + +#define E1000_VF_RESET 0x01 /* VF requests reset */ +#define E1000_VF_SET_MAC_ADDR 0x02 /* VF requests PF to set MAC addr */ +/* VF requests PF to clear all unicast MAC filters */ +#define E1000_VF_MAC_FILTER_CLR (0x01 << E1000_VT_MSGINFO_SHIFT) +/* VF requests PF to add unicast MAC filter */ +#define E1000_VF_MAC_FILTER_ADD (0x02 << E1000_VT_MSGINFO_SHIFT) +#define E1000_VF_SET_MULTICAST 0x03 /* VF requests PF to set MC addr */ +#define E1000_VF_SET_VLAN 0x04 /* VF requests PF to set VLAN */ +#define E1000_VF_SET_LPE 0x05 /* VF requests PF to set VMOLR.LPE */ + +#define E1000_PF_CONTROL_MSG 0x0100 /* PF control message */ + +void e1000_init_mbx_ops_generic(struct e1000_hw *hw); +s32 e1000_init_mbx_params_vf(struct e1000_hw *); + +#endif /* _E1000_MBX_H_ */ diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c new file mode 100644 index 000000000..72cb1b56e --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -0,0 +1,2980 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/vmalloc.h> +#include <linux/pagemap.h> +#include <linux/delay.h> +#include <linux/netdevice.h> +#include <linux/tcp.h> +#include <linux/ipv6.h> +#include <linux/slab.h> +#include <net/checksum.h> +#include <net/ip6_checksum.h> +#include <linux/mii.h> +#include <linux/ethtool.h> +#include <linux/if_vlan.h> +#include <linux/prefetch.h> +#include <linux/sctp.h> + +#include "igbvf.h" + +char igbvf_driver_name[] = "igbvf"; +static const char igbvf_driver_string[] = + "Intel(R) Gigabit Virtual Function Network Driver"; +static const char igbvf_copyright[] = + "Copyright (c) 2009 - 2012 Intel Corporation."; + +#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK) +static int debug = -1; +module_param(debug, int, 0); +MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)"); + +static int igbvf_poll(struct napi_struct *napi, int budget); +static void igbvf_reset(struct igbvf_adapter *); +static void igbvf_set_interrupt_capability(struct igbvf_adapter *); +static void igbvf_reset_interrupt_capability(struct igbvf_adapter *); + +static struct igbvf_info igbvf_vf_info = { + .mac = e1000_vfadapt, + .flags = 0, + .pba = 10, + .init_ops = e1000_init_function_pointers_vf, +}; + +static struct igbvf_info igbvf_i350_vf_info = { + .mac = e1000_vfadapt_i350, + .flags = 0, + .pba = 10, + .init_ops = e1000_init_function_pointers_vf, +}; + +static const struct igbvf_info *igbvf_info_tbl[] = { + [board_vf] = &igbvf_vf_info, + [board_i350_vf] = &igbvf_i350_vf_info, +}; + +/** + * igbvf_desc_unused - calculate if we have unused descriptors + * @ring: address of receive ring structure + **/ +static int igbvf_desc_unused(struct igbvf_ring *ring) +{ + if (ring->next_to_clean > ring->next_to_use) + return ring->next_to_clean - ring->next_to_use - 1; + + return ring->count + ring->next_to_clean - ring->next_to_use - 1; +} + +/** + * igbvf_receive_skb - helper function to handle Rx indications + * @adapter: board private structure + * @netdev: pointer to netdev struct + * @skb: skb to indicate to stack + * @status: descriptor status field as written by hardware + * @vlan: descriptor vlan field as written by hardware (no le/be conversion) + * @skb: pointer to sk_buff to be indicated to stack + **/ +static void igbvf_receive_skb(struct igbvf_adapter *adapter, + struct net_device *netdev, + struct sk_buff *skb, + u32 status, __le16 vlan) +{ + u16 vid; + + if (status & E1000_RXD_STAT_VP) { + if ((adapter->flags & IGBVF_FLAG_RX_LB_VLAN_BSWAP) && + (status & E1000_RXDEXT_STATERR_LB)) + vid = be16_to_cpu((__force __be16)vlan) & E1000_RXD_SPC_VLAN_MASK; + else + vid = le16_to_cpu(vlan) & E1000_RXD_SPC_VLAN_MASK; + if (test_bit(vid, adapter->active_vlans)) + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vid); + } + + napi_gro_receive(&adapter->rx_ring->napi, skb); +} + +static inline void igbvf_rx_checksum_adv(struct igbvf_adapter *adapter, + u32 status_err, struct sk_buff *skb) +{ + skb_checksum_none_assert(skb); + + /* Ignore Checksum bit is set or checksum is disabled through ethtool */ + if ((status_err & E1000_RXD_STAT_IXSM) || + (adapter->flags & IGBVF_FLAG_RX_CSUM_DISABLED)) + return; + + /* TCP/UDP checksum error bit is set */ + if (status_err & + (E1000_RXDEXT_STATERR_TCPE | E1000_RXDEXT_STATERR_IPE)) { + /* let the stack verify checksum errors */ + adapter->hw_csum_err++; + return; + } + + /* It must be a TCP or UDP packet with a valid checksum */ + if (status_err & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + adapter->hw_csum_good++; +} + +/** + * igbvf_alloc_rx_buffers - Replace used receive buffers; packet split + * @rx_ring: address of ring structure to repopulate + * @cleaned_count: number of buffers to repopulate + **/ +static void igbvf_alloc_rx_buffers(struct igbvf_ring *rx_ring, + int cleaned_count) +{ + struct igbvf_adapter *adapter = rx_ring->adapter; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + union e1000_adv_rx_desc *rx_desc; + struct igbvf_buffer *buffer_info; + struct sk_buff *skb; + unsigned int i; + int bufsz; + + i = rx_ring->next_to_use; + buffer_info = &rx_ring->buffer_info[i]; + + if (adapter->rx_ps_hdr_size) + bufsz = adapter->rx_ps_hdr_size; + else + bufsz = adapter->rx_buffer_len; + + while (cleaned_count--) { + rx_desc = IGBVF_RX_DESC_ADV(*rx_ring, i); + + if (adapter->rx_ps_hdr_size && !buffer_info->page_dma) { + if (!buffer_info->page) { + buffer_info->page = alloc_page(GFP_ATOMIC); + if (!buffer_info->page) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + buffer_info->page_offset = 0; + } else { + buffer_info->page_offset ^= PAGE_SIZE / 2; + } + buffer_info->page_dma = + dma_map_page(&pdev->dev, buffer_info->page, + buffer_info->page_offset, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + if (dma_mapping_error(&pdev->dev, + buffer_info->page_dma)) { + __free_page(buffer_info->page); + buffer_info->page = NULL; + dev_err(&pdev->dev, "RX DMA map failed\n"); + break; + } + } + + if (!buffer_info->skb) { + skb = netdev_alloc_skb_ip_align(netdev, bufsz); + if (!skb) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + + buffer_info->skb = skb; + buffer_info->dma = dma_map_single(&pdev->dev, skb->data, + bufsz, + DMA_FROM_DEVICE); + if (dma_mapping_error(&pdev->dev, buffer_info->dma)) { + dev_kfree_skb(buffer_info->skb); + buffer_info->skb = NULL; + dev_err(&pdev->dev, "RX DMA map failed\n"); + goto no_buffers; + } + } + /* Refresh the desc even if buffer_addrs didn't change because + * each write-back erases this info. + */ + if (adapter->rx_ps_hdr_size) { + rx_desc->read.pkt_addr = + cpu_to_le64(buffer_info->page_dma); + rx_desc->read.hdr_addr = cpu_to_le64(buffer_info->dma); + } else { + rx_desc->read.pkt_addr = cpu_to_le64(buffer_info->dma); + rx_desc->read.hdr_addr = 0; + } + + i++; + if (i == rx_ring->count) + i = 0; + buffer_info = &rx_ring->buffer_info[i]; + } + +no_buffers: + if (rx_ring->next_to_use != i) { + rx_ring->next_to_use = i; + if (i == 0) + i = (rx_ring->count - 1); + else + 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(); + writel(i, adapter->hw.hw_addr + rx_ring->tail); + } +} + +/** + * igbvf_clean_rx_irq - Send received data up the network stack; legacy + * @adapter: board private structure + * @work_done: output parameter used to indicate completed work + * @work_to_do: input parameter setting limit of work + * + * the return value indicates whether actual cleaning was done, there + * is no guarantee that everything was cleaned + **/ +static bool igbvf_clean_rx_irq(struct igbvf_adapter *adapter, + int *work_done, int work_to_do) +{ + struct igbvf_ring *rx_ring = adapter->rx_ring; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + union e1000_adv_rx_desc *rx_desc, *next_rxd; + struct igbvf_buffer *buffer_info, *next_buffer; + struct sk_buff *skb; + bool cleaned = false; + int cleaned_count = 0; + unsigned int total_bytes = 0, total_packets = 0; + unsigned int i; + u32 length, hlen, staterr; + + i = rx_ring->next_to_clean; + rx_desc = IGBVF_RX_DESC_ADV(*rx_ring, i); + staterr = le32_to_cpu(rx_desc->wb.upper.status_error); + + while (staterr & E1000_RXD_STAT_DD) { + if (*work_done >= work_to_do) + break; + (*work_done)++; + rmb(); /* read descriptor and rx_buffer_info after status DD */ + + buffer_info = &rx_ring->buffer_info[i]; + + /* HW will not DMA in data larger than the given buffer, even + * if it parses the (NFS, of course) header to be larger. In + * that case, it fills the header buffer and spills the rest + * into the page. + */ + hlen = (le16_to_cpu(rx_desc->wb.lower.lo_dword.hs_rss.hdr_info) + & E1000_RXDADV_HDRBUFLEN_MASK) >> + E1000_RXDADV_HDRBUFLEN_SHIFT; + if (hlen > adapter->rx_ps_hdr_size) + hlen = adapter->rx_ps_hdr_size; + + length = le16_to_cpu(rx_desc->wb.upper.length); + cleaned = true; + cleaned_count++; + + skb = buffer_info->skb; + prefetch(skb->data - NET_IP_ALIGN); + buffer_info->skb = NULL; + if (!adapter->rx_ps_hdr_size) { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_buffer_len, + DMA_FROM_DEVICE); + buffer_info->dma = 0; + skb_put(skb, length); + goto send_up; + } + + if (!skb_shinfo(skb)->nr_frags) { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_ps_hdr_size, + DMA_FROM_DEVICE); + buffer_info->dma = 0; + skb_put(skb, hlen); + } + + if (length) { + dma_unmap_page(&pdev->dev, buffer_info->page_dma, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + buffer_info->page_dma = 0; + + skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, + buffer_info->page, + buffer_info->page_offset, + length); + + if ((adapter->rx_buffer_len > (PAGE_SIZE / 2)) || + (page_count(buffer_info->page) != 1)) + buffer_info->page = NULL; + else + get_page(buffer_info->page); + + skb->len += length; + skb->data_len += length; + skb->truesize += PAGE_SIZE / 2; + } +send_up: + i++; + if (i == rx_ring->count) + i = 0; + next_rxd = IGBVF_RX_DESC_ADV(*rx_ring, i); + prefetch(next_rxd); + next_buffer = &rx_ring->buffer_info[i]; + + if (!(staterr & E1000_RXD_STAT_EOP)) { + buffer_info->skb = next_buffer->skb; + buffer_info->dma = next_buffer->dma; + next_buffer->skb = skb; + next_buffer->dma = 0; + goto next_desc; + } + + if (staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) { + dev_kfree_skb_irq(skb); + goto next_desc; + } + + total_bytes += skb->len; + total_packets++; + + igbvf_rx_checksum_adv(adapter, staterr, skb); + + skb->protocol = eth_type_trans(skb, netdev); + + igbvf_receive_skb(adapter, netdev, skb, staterr, + rx_desc->wb.upper.vlan); + +next_desc: + rx_desc->wb.upper.status_error = 0; + + /* return some buffers to hardware, one at a time is too slow */ + if (cleaned_count >= IGBVF_RX_BUFFER_WRITE) { + igbvf_alloc_rx_buffers(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.upper.status_error); + } + + rx_ring->next_to_clean = i; + cleaned_count = igbvf_desc_unused(rx_ring); + + if (cleaned_count) + igbvf_alloc_rx_buffers(rx_ring, cleaned_count); + + adapter->total_rx_packets += total_packets; + adapter->total_rx_bytes += total_bytes; + netdev->stats.rx_bytes += total_bytes; + netdev->stats.rx_packets += total_packets; + return cleaned; +} + +static void igbvf_put_txbuf(struct igbvf_adapter *adapter, + struct igbvf_buffer *buffer_info) +{ + if (buffer_info->dma) { + if (buffer_info->mapped_as_page) + dma_unmap_page(&adapter->pdev->dev, + buffer_info->dma, + buffer_info->length, + DMA_TO_DEVICE); + else + dma_unmap_single(&adapter->pdev->dev, + buffer_info->dma, + buffer_info->length, + DMA_TO_DEVICE); + buffer_info->dma = 0; + } + if (buffer_info->skb) { + dev_kfree_skb_any(buffer_info->skb); + buffer_info->skb = NULL; + } + buffer_info->time_stamp = 0; +} + +/** + * igbvf_setup_tx_resources - allocate Tx resources (Descriptors) + * @adapter: board private structure + * @tx_ring: ring being initialized + * + * Return 0 on success, negative on failure + **/ +int igbvf_setup_tx_resources(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring) +{ + struct pci_dev *pdev = adapter->pdev; + int size; + + size = sizeof(struct igbvf_buffer) * tx_ring->count; + tx_ring->buffer_info = vzalloc(size); + if (!tx_ring->buffer_info) + goto err; + + /* round up to nearest 4K */ + tx_ring->size = tx_ring->count * sizeof(union e1000_adv_tx_desc); + tx_ring->size = ALIGN(tx_ring->size, 4096); + + tx_ring->desc = dma_alloc_coherent(&pdev->dev, tx_ring->size, + &tx_ring->dma, GFP_KERNEL); + if (!tx_ring->desc) + goto err; + + tx_ring->adapter = adapter; + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + + return 0; +err: + vfree(tx_ring->buffer_info); + dev_err(&adapter->pdev->dev, + "Unable to allocate memory for the transmit descriptor ring\n"); + return -ENOMEM; +} + +/** + * igbvf_setup_rx_resources - allocate Rx resources (Descriptors) + * @adapter: board private structure + * @rx_ring: ring being initialized + * + * Returns 0 on success, negative on failure + **/ +int igbvf_setup_rx_resources(struct igbvf_adapter *adapter, + struct igbvf_ring *rx_ring) +{ + struct pci_dev *pdev = adapter->pdev; + int size, desc_len; + + size = sizeof(struct igbvf_buffer) * rx_ring->count; + rx_ring->buffer_info = vzalloc(size); + if (!rx_ring->buffer_info) + goto err; + + desc_len = sizeof(union e1000_adv_rx_desc); + + /* Round up to nearest 4K */ + rx_ring->size = rx_ring->count * desc_len; + rx_ring->size = ALIGN(rx_ring->size, 4096); + + rx_ring->desc = dma_alloc_coherent(&pdev->dev, rx_ring->size, + &rx_ring->dma, GFP_KERNEL); + if (!rx_ring->desc) + goto err; + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + + rx_ring->adapter = adapter; + + return 0; + +err: + vfree(rx_ring->buffer_info); + rx_ring->buffer_info = NULL; + dev_err(&adapter->pdev->dev, + "Unable to allocate memory for the receive descriptor ring\n"); + return -ENOMEM; +} + +/** + * igbvf_clean_tx_ring - Free Tx Buffers + * @tx_ring: ring to be cleaned + **/ +static void igbvf_clean_tx_ring(struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = tx_ring->adapter; + struct igbvf_buffer *buffer_info; + unsigned long size; + unsigned int i; + + if (!tx_ring->buffer_info) + return; + + /* Free all the Tx ring sk_buffs */ + for (i = 0; i < tx_ring->count; i++) { + buffer_info = &tx_ring->buffer_info[i]; + igbvf_put_txbuf(adapter, buffer_info); + } + + size = sizeof(struct igbvf_buffer) * tx_ring->count; + memset(tx_ring->buffer_info, 0, size); + + /* Zero out the descriptor ring */ + memset(tx_ring->desc, 0, tx_ring->size); + + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + + writel(0, adapter->hw.hw_addr + tx_ring->head); + writel(0, adapter->hw.hw_addr + tx_ring->tail); +} + +/** + * igbvf_free_tx_resources - Free Tx Resources per Queue + * @tx_ring: ring to free resources from + * + * Free all transmit software resources + **/ +void igbvf_free_tx_resources(struct igbvf_ring *tx_ring) +{ + struct pci_dev *pdev = tx_ring->adapter->pdev; + + igbvf_clean_tx_ring(tx_ring); + + vfree(tx_ring->buffer_info); + tx_ring->buffer_info = NULL; + + dma_free_coherent(&pdev->dev, tx_ring->size, tx_ring->desc, + tx_ring->dma); + + tx_ring->desc = NULL; +} + +/** + * igbvf_clean_rx_ring - Free Rx Buffers per Queue + * @rx_ring: ring structure pointer to free buffers from + **/ +static void igbvf_clean_rx_ring(struct igbvf_ring *rx_ring) +{ + struct igbvf_adapter *adapter = rx_ring->adapter; + struct igbvf_buffer *buffer_info; + struct pci_dev *pdev = adapter->pdev; + unsigned long size; + unsigned int i; + + if (!rx_ring->buffer_info) + return; + + /* Free all the Rx ring sk_buffs */ + for (i = 0; i < rx_ring->count; i++) { + buffer_info = &rx_ring->buffer_info[i]; + if (buffer_info->dma) { + if (adapter->rx_ps_hdr_size) { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_ps_hdr_size, + DMA_FROM_DEVICE); + } else { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_buffer_len, + DMA_FROM_DEVICE); + } + buffer_info->dma = 0; + } + + if (buffer_info->skb) { + dev_kfree_skb(buffer_info->skb); + buffer_info->skb = NULL; + } + + if (buffer_info->page) { + if (buffer_info->page_dma) + dma_unmap_page(&pdev->dev, + buffer_info->page_dma, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + put_page(buffer_info->page); + buffer_info->page = NULL; + buffer_info->page_dma = 0; + buffer_info->page_offset = 0; + } + } + + size = sizeof(struct igbvf_buffer) * rx_ring->count; + memset(rx_ring->buffer_info, 0, size); + + /* Zero out the descriptor ring */ + memset(rx_ring->desc, 0, rx_ring->size); + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + + writel(0, adapter->hw.hw_addr + rx_ring->head); + writel(0, adapter->hw.hw_addr + rx_ring->tail); +} + +/** + * igbvf_free_rx_resources - Free Rx Resources + * @rx_ring: ring to clean the resources from + * + * Free all receive software resources + **/ + +void igbvf_free_rx_resources(struct igbvf_ring *rx_ring) +{ + struct pci_dev *pdev = rx_ring->adapter->pdev; + + igbvf_clean_rx_ring(rx_ring); + + vfree(rx_ring->buffer_info); + rx_ring->buffer_info = NULL; + + dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc, + rx_ring->dma); + rx_ring->desc = NULL; +} + +/** + * igbvf_update_itr - update the dynamic ITR value based on statistics + * @adapter: pointer to adapter + * @itr_setting: current adapter->itr + * @packets: the number of packets during this measurement interval + * @bytes: the number of bytes during this measurement interval + * + * Stores a new ITR value based on packets and byte counts during the last + * interrupt. The advantage of per interrupt computation is faster updates + * and more accurate ITR for the current traffic pattern. Constants in this + * function were computed based on theoretical maximum wire speed and thresholds + * were set based on testing data as well as attempting to minimize response + * time while increasing bulk throughput. + **/ +static enum latency_range igbvf_update_itr(struct igbvf_adapter *adapter, + enum latency_range itr_setting, + int packets, int bytes) +{ + enum latency_range retval = itr_setting; + + if (packets == 0) + goto update_itr_done; + + switch (itr_setting) { + case lowest_latency: + /* handle TSO and jumbo frames */ + if (bytes/packets > 8000) + retval = bulk_latency; + else if ((packets < 5) && (bytes > 512)) + retval = low_latency; + break; + case low_latency: /* 50 usec aka 20000 ints/s */ + if (bytes > 10000) { + /* this if handles the TSO accounting */ + if (bytes/packets > 8000) + retval = bulk_latency; + else if ((packets < 10) || ((bytes/packets) > 1200)) + retval = bulk_latency; + else if ((packets > 35)) + retval = lowest_latency; + } else if (bytes/packets > 2000) { + retval = bulk_latency; + } else if (packets <= 2 && bytes < 512) { + retval = lowest_latency; + } + break; + case bulk_latency: /* 250 usec aka 4000 ints/s */ + if (bytes > 25000) { + if (packets > 35) + retval = low_latency; + } else if (bytes < 6000) { + retval = low_latency; + } + break; + default: + break; + } + +update_itr_done: + return retval; +} + +static int igbvf_range_to_itr(enum latency_range current_range) +{ + int new_itr; + + switch (current_range) { + /* counts and packets in update_itr are dependent on these numbers */ + case lowest_latency: + new_itr = IGBVF_70K_ITR; + break; + case low_latency: + new_itr = IGBVF_20K_ITR; + break; + case bulk_latency: + new_itr = IGBVF_4K_ITR; + break; + default: + new_itr = IGBVF_START_ITR; + break; + } + return new_itr; +} + +static void igbvf_set_itr(struct igbvf_adapter *adapter) +{ + u32 new_itr; + + adapter->tx_ring->itr_range = + igbvf_update_itr(adapter, + adapter->tx_ring->itr_val, + adapter->total_tx_packets, + adapter->total_tx_bytes); + + /* conservative mode (itr 3) eliminates the lowest_latency setting */ + if (adapter->requested_itr == 3 && + adapter->tx_ring->itr_range == lowest_latency) + adapter->tx_ring->itr_range = low_latency; + + new_itr = igbvf_range_to_itr(adapter->tx_ring->itr_range); + + if (new_itr != adapter->tx_ring->itr_val) { + u32 current_itr = adapter->tx_ring->itr_val; + /* this attempts to bias the interrupt rate towards Bulk + * by adding intermediate steps when interrupt rate is + * increasing + */ + new_itr = new_itr > current_itr ? + min(current_itr + (new_itr >> 2), new_itr) : + new_itr; + adapter->tx_ring->itr_val = new_itr; + + adapter->tx_ring->set_itr = 1; + } + + adapter->rx_ring->itr_range = + igbvf_update_itr(adapter, adapter->rx_ring->itr_val, + adapter->total_rx_packets, + adapter->total_rx_bytes); + if (adapter->requested_itr == 3 && + adapter->rx_ring->itr_range == lowest_latency) + adapter->rx_ring->itr_range = low_latency; + + new_itr = igbvf_range_to_itr(adapter->rx_ring->itr_range); + + if (new_itr != adapter->rx_ring->itr_val) { + u32 current_itr = adapter->rx_ring->itr_val; + + new_itr = new_itr > current_itr ? + min(current_itr + (new_itr >> 2), new_itr) : + new_itr; + adapter->rx_ring->itr_val = new_itr; + + adapter->rx_ring->set_itr = 1; + } +} + +/** + * igbvf_clean_tx_irq - Reclaim resources after transmit completes + * @tx_ring: ring structure to clean descriptors from + * + * returns true if ring is completely cleaned + **/ +static bool igbvf_clean_tx_irq(struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = tx_ring->adapter; + struct net_device *netdev = adapter->netdev; + struct igbvf_buffer *buffer_info; + struct sk_buff *skb; + union e1000_adv_tx_desc *tx_desc, *eop_desc; + unsigned int total_bytes = 0, total_packets = 0; + unsigned int i, count = 0; + bool cleaned = false; + + i = tx_ring->next_to_clean; + buffer_info = &tx_ring->buffer_info[i]; + eop_desc = buffer_info->next_to_watch; + + do { + /* if next_to_watch is not set then there is no work pending */ + if (!eop_desc) + break; + + /* prevent any other reads prior to eop_desc */ + smp_rmb(); + + /* if DD is not set pending work has not been completed */ + if (!(eop_desc->wb.status & cpu_to_le32(E1000_TXD_STAT_DD))) + break; + + /* clear next_to_watch to prevent false hangs */ + buffer_info->next_to_watch = NULL; + + for (cleaned = false; !cleaned; count++) { + tx_desc = IGBVF_TX_DESC_ADV(*tx_ring, i); + cleaned = (tx_desc == eop_desc); + skb = buffer_info->skb; + + if (skb) { + unsigned int segs, bytecount; + + /* gso_segs is currently only valid for tcp */ + segs = skb_shinfo(skb)->gso_segs ?: 1; + /* multiply data chunks by size of headers */ + bytecount = ((segs - 1) * skb_headlen(skb)) + + skb->len; + total_packets += segs; + total_bytes += bytecount; + } + + igbvf_put_txbuf(adapter, buffer_info); + tx_desc->wb.status = 0; + + i++; + if (i == tx_ring->count) + i = 0; + + buffer_info = &tx_ring->buffer_info[i]; + } + + eop_desc = buffer_info->next_to_watch; + } while (count < tx_ring->count); + + tx_ring->next_to_clean = i; + + if (unlikely(count && netif_carrier_ok(netdev) && + igbvf_desc_unused(tx_ring) >= IGBVF_TX_QUEUE_WAKE)) { + /* Make sure that anybody stopping the queue after this + * sees the new next_to_clean. + */ + smp_mb(); + if (netif_queue_stopped(netdev) && + !(test_bit(__IGBVF_DOWN, &adapter->state))) { + netif_wake_queue(netdev); + ++adapter->restart_queue; + } + } + + netdev->stats.tx_bytes += total_bytes; + netdev->stats.tx_packets += total_packets; + return count < tx_ring->count; +} + +static irqreturn_t igbvf_msix_other(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + adapter->int_counter1++; + + hw->mac.get_link_status = 1; + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + ew32(EIMS, adapter->eims_other); + + return IRQ_HANDLED; +} + +static irqreturn_t igbvf_intr_msix_tx(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + + if (tx_ring->set_itr) { + writel(tx_ring->itr_val, + adapter->hw.hw_addr + tx_ring->itr_register); + adapter->tx_ring->set_itr = 0; + } + + adapter->total_tx_bytes = 0; + adapter->total_tx_packets = 0; + + /* auto mask will automatically re-enable the interrupt when we write + * EICS + */ + if (!igbvf_clean_tx_irq(tx_ring)) + /* Ring was not completely cleaned, so fire another interrupt */ + ew32(EICS, tx_ring->eims_value); + else + ew32(EIMS, tx_ring->eims_value); + + return IRQ_HANDLED; +} + +static irqreturn_t igbvf_intr_msix_rx(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + + adapter->int_counter0++; + + /* Write the ITR value calculated at the end of the + * previous interrupt. + */ + if (adapter->rx_ring->set_itr) { + writel(adapter->rx_ring->itr_val, + adapter->hw.hw_addr + adapter->rx_ring->itr_register); + adapter->rx_ring->set_itr = 0; + } + + if (napi_schedule_prep(&adapter->rx_ring->napi)) { + adapter->total_rx_bytes = 0; + adapter->total_rx_packets = 0; + __napi_schedule(&adapter->rx_ring->napi); + } + + return IRQ_HANDLED; +} + +#define IGBVF_NO_QUEUE -1 + +static void igbvf_assign_vector(struct igbvf_adapter *adapter, int rx_queue, + int tx_queue, int msix_vector) +{ + struct e1000_hw *hw = &adapter->hw; + u32 ivar, index; + + /* 82576 uses a table-based method for assigning vectors. + * Each queue has a single entry in the table to which we write + * a vector number along with a "valid" bit. Sadly, the layout + * of the table is somewhat counterintuitive. + */ + if (rx_queue > IGBVF_NO_QUEUE) { + index = (rx_queue >> 1); + ivar = array_er32(IVAR0, index); + if (rx_queue & 0x1) { + /* vector goes into third byte of register */ + ivar = ivar & 0xFF00FFFF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 16; + } else { + /* vector goes into low byte of register */ + ivar = ivar & 0xFFFFFF00; + ivar |= msix_vector | E1000_IVAR_VALID; + } + adapter->rx_ring[rx_queue].eims_value = BIT(msix_vector); + array_ew32(IVAR0, index, ivar); + } + if (tx_queue > IGBVF_NO_QUEUE) { + index = (tx_queue >> 1); + ivar = array_er32(IVAR0, index); + if (tx_queue & 0x1) { + /* vector goes into high byte of register */ + ivar = ivar & 0x00FFFFFF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 24; + } else { + /* vector goes into second byte of register */ + ivar = ivar & 0xFFFF00FF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 8; + } + adapter->tx_ring[tx_queue].eims_value = BIT(msix_vector); + array_ew32(IVAR0, index, ivar); + } +} + +/** + * igbvf_configure_msix - Configure MSI-X hardware + * @adapter: board private structure + * + * igbvf_configure_msix sets up the hardware to properly + * generate MSI-X interrupts. + **/ +static void igbvf_configure_msix(struct igbvf_adapter *adapter) +{ + u32 tmp; + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct igbvf_ring *rx_ring = adapter->rx_ring; + int vector = 0; + + adapter->eims_enable_mask = 0; + + igbvf_assign_vector(adapter, IGBVF_NO_QUEUE, 0, vector++); + adapter->eims_enable_mask |= tx_ring->eims_value; + writel(tx_ring->itr_val, hw->hw_addr + tx_ring->itr_register); + igbvf_assign_vector(adapter, 0, IGBVF_NO_QUEUE, vector++); + adapter->eims_enable_mask |= rx_ring->eims_value; + writel(rx_ring->itr_val, hw->hw_addr + rx_ring->itr_register); + + /* set vector for other causes, i.e. link changes */ + + tmp = (vector++ | E1000_IVAR_VALID); + + ew32(IVAR_MISC, tmp); + + adapter->eims_enable_mask = GENMASK(vector - 1, 0); + adapter->eims_other = BIT(vector - 1); + e1e_flush(); +} + +static void igbvf_reset_interrupt_capability(struct igbvf_adapter *adapter) +{ + if (adapter->msix_entries) { + pci_disable_msix(adapter->pdev); + kfree(adapter->msix_entries); + adapter->msix_entries = NULL; + } +} + +/** + * igbvf_set_interrupt_capability - set MSI or MSI-X if supported + * @adapter: board private structure + * + * Attempt to configure interrupts using the best available + * capabilities of the hardware and kernel. + **/ +static void igbvf_set_interrupt_capability(struct igbvf_adapter *adapter) +{ + int err = -ENOMEM; + int i; + + /* we allocate 3 vectors, 1 for Tx, 1 for Rx, one for PF messages */ + adapter->msix_entries = kcalloc(3, sizeof(struct msix_entry), + GFP_KERNEL); + if (adapter->msix_entries) { + for (i = 0; i < 3; i++) + adapter->msix_entries[i].entry = i; + + err = pci_enable_msix_range(adapter->pdev, + adapter->msix_entries, 3, 3); + } + + if (err < 0) { + /* MSI-X failed */ + dev_err(&adapter->pdev->dev, + "Failed to initialize MSI-X interrupts.\n"); + igbvf_reset_interrupt_capability(adapter); + } +} + +/** + * igbvf_request_msix - Initialize MSI-X interrupts + * @adapter: board private structure + * + * igbvf_request_msix allocates MSI-X vectors and requests interrupts from the + * kernel. + **/ +static int igbvf_request_msix(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int err = 0, vector = 0; + + if (strlen(netdev->name) < (IFNAMSIZ - 5)) { + sprintf(adapter->tx_ring->name, "%s-tx-0", netdev->name); + sprintf(adapter->rx_ring->name, "%s-rx-0", netdev->name); + } else { + memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ); + memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ); + } + + err = request_irq(adapter->msix_entries[vector].vector, + igbvf_intr_msix_tx, 0, adapter->tx_ring->name, + netdev); + if (err) + goto out; + + adapter->tx_ring->itr_register = E1000_EITR(vector); + adapter->tx_ring->itr_val = adapter->current_itr; + vector++; + + err = request_irq(adapter->msix_entries[vector].vector, + igbvf_intr_msix_rx, 0, adapter->rx_ring->name, + netdev); + if (err) + goto free_irq_tx; + + adapter->rx_ring->itr_register = E1000_EITR(vector); + adapter->rx_ring->itr_val = adapter->current_itr; + vector++; + + err = request_irq(adapter->msix_entries[vector].vector, + igbvf_msix_other, 0, netdev->name, netdev); + if (err) + goto free_irq_rx; + + igbvf_configure_msix(adapter); + return 0; +free_irq_rx: + free_irq(adapter->msix_entries[--vector].vector, netdev); +free_irq_tx: + free_irq(adapter->msix_entries[--vector].vector, netdev); +out: + return err; +} + +/** + * igbvf_alloc_queues - Allocate memory for all rings + * @adapter: board private structure to initialize + **/ +static int igbvf_alloc_queues(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + + adapter->tx_ring = kzalloc(sizeof(struct igbvf_ring), GFP_KERNEL); + if (!adapter->tx_ring) + return -ENOMEM; + + adapter->rx_ring = kzalloc(sizeof(struct igbvf_ring), GFP_KERNEL); + if (!adapter->rx_ring) { + kfree(adapter->tx_ring); + return -ENOMEM; + } + + netif_napi_add(netdev, &adapter->rx_ring->napi, igbvf_poll); + + return 0; +} + +/** + * igbvf_request_irq - initialize interrupts + * @adapter: board private structure + * + * Attempts to configure interrupts using the best available + * capabilities of the hardware and kernel. + **/ +static int igbvf_request_irq(struct igbvf_adapter *adapter) +{ + int err = -1; + + /* igbvf supports msi-x only */ + if (adapter->msix_entries) + err = igbvf_request_msix(adapter); + + if (!err) + return err; + + dev_err(&adapter->pdev->dev, + "Unable to allocate interrupt, Error: %d\n", err); + + return err; +} + +static void igbvf_free_irq(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int vector; + + if (adapter->msix_entries) { + for (vector = 0; vector < 3; vector++) + free_irq(adapter->msix_entries[vector].vector, netdev); + } +} + +/** + * igbvf_irq_disable - Mask off interrupt generation on the NIC + * @adapter: board private structure + **/ +static void igbvf_irq_disable(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + ew32(EIMC, ~0); + + if (adapter->msix_entries) + ew32(EIAC, 0); +} + +/** + * igbvf_irq_enable - Enable default interrupt generation settings + * @adapter: board private structure + **/ +static void igbvf_irq_enable(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + ew32(EIAC, adapter->eims_enable_mask); + ew32(EIAM, adapter->eims_enable_mask); + ew32(EIMS, adapter->eims_enable_mask); +} + +/** + * igbvf_poll - NAPI Rx polling callback + * @napi: struct associated with this polling callback + * @budget: amount of packets driver is allowed to process this poll + **/ +static int igbvf_poll(struct napi_struct *napi, int budget) +{ + struct igbvf_ring *rx_ring = container_of(napi, struct igbvf_ring, napi); + struct igbvf_adapter *adapter = rx_ring->adapter; + struct e1000_hw *hw = &adapter->hw; + int work_done = 0; + + igbvf_clean_rx_irq(adapter, &work_done, budget); + + if (work_done == budget) + return budget; + + /* Exit the polling mode, but don't re-enable interrupts if stack might + * poll us due to busy-polling + */ + if (likely(napi_complete_done(napi, work_done))) { + if (adapter->requested_itr & 3) + igbvf_set_itr(adapter); + + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + ew32(EIMS, adapter->rx_ring->eims_value); + } + + return work_done; +} + +/** + * igbvf_set_rlpml - set receive large packet maximum length + * @adapter: board private structure + * + * Configure the maximum size of packets that will be received + */ +static void igbvf_set_rlpml(struct igbvf_adapter *adapter) +{ + int max_frame_size; + struct e1000_hw *hw = &adapter->hw; + + max_frame_size = adapter->max_frame_size + VLAN_TAG_SIZE; + + spin_lock_bh(&hw->mbx_lock); + + e1000_rlpml_set_vf(hw, max_frame_size); + + spin_unlock_bh(&hw->mbx_lock); +} + +static int igbvf_vlan_rx_add_vid(struct net_device *netdev, + __be16 proto, u16 vid) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + spin_lock_bh(&hw->mbx_lock); + + if (hw->mac.ops.set_vfta(hw, vid, true)) { + dev_warn(&adapter->pdev->dev, "Vlan id %d\n is not added", vid); + spin_unlock_bh(&hw->mbx_lock); + return -EINVAL; + } + + spin_unlock_bh(&hw->mbx_lock); + + set_bit(vid, adapter->active_vlans); + return 0; +} + +static int igbvf_vlan_rx_kill_vid(struct net_device *netdev, + __be16 proto, u16 vid) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + spin_lock_bh(&hw->mbx_lock); + + if (hw->mac.ops.set_vfta(hw, vid, false)) { + dev_err(&adapter->pdev->dev, + "Failed to remove vlan id %d\n", vid); + spin_unlock_bh(&hw->mbx_lock); + return -EINVAL; + } + + spin_unlock_bh(&hw->mbx_lock); + + clear_bit(vid, adapter->active_vlans); + return 0; +} + +static void igbvf_restore_vlan(struct igbvf_adapter *adapter) +{ + u16 vid; + + for_each_set_bit(vid, adapter->active_vlans, VLAN_N_VID) + igbvf_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), vid); +} + +/** + * igbvf_configure_tx - Configure Transmit Unit after Reset + * @adapter: board private structure + * + * Configure the Tx unit of the MAC after a reset. + **/ +static void igbvf_configure_tx(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + u64 tdba; + u32 txdctl, dca_txctrl; + + /* disable transmits */ + txdctl = er32(TXDCTL(0)); + ew32(TXDCTL(0), txdctl & ~E1000_TXDCTL_QUEUE_ENABLE); + e1e_flush(); + msleep(10); + + /* Setup the HW Tx Head and Tail descriptor pointers */ + ew32(TDLEN(0), tx_ring->count * sizeof(union e1000_adv_tx_desc)); + tdba = tx_ring->dma; + ew32(TDBAL(0), (tdba & DMA_BIT_MASK(32))); + ew32(TDBAH(0), (tdba >> 32)); + ew32(TDH(0), 0); + ew32(TDT(0), 0); + tx_ring->head = E1000_TDH(0); + tx_ring->tail = E1000_TDT(0); + + /* Turn off Relaxed Ordering on head write-backs. The writebacks + * MUST be delivered in order or it will completely screw up + * our bookkeeping. + */ + dca_txctrl = er32(DCA_TXCTRL(0)); + dca_txctrl &= ~E1000_DCA_TXCTRL_TX_WB_RO_EN; + ew32(DCA_TXCTRL(0), dca_txctrl); + + /* enable transmits */ + txdctl |= E1000_TXDCTL_QUEUE_ENABLE; + ew32(TXDCTL(0), txdctl); + + /* Setup Transmit Descriptor Settings for eop descriptor */ + adapter->txd_cmd = E1000_ADVTXD_DCMD_EOP | E1000_ADVTXD_DCMD_IFCS; + + /* enable Report Status bit */ + adapter->txd_cmd |= E1000_ADVTXD_DCMD_RS; +} + +/** + * igbvf_setup_srrctl - configure the receive control registers + * @adapter: Board private structure + **/ +static void igbvf_setup_srrctl(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + u32 srrctl = 0; + + srrctl &= ~(E1000_SRRCTL_DESCTYPE_MASK | + E1000_SRRCTL_BSIZEHDR_MASK | + E1000_SRRCTL_BSIZEPKT_MASK); + + /* Enable queue drop to avoid head of line blocking */ + srrctl |= E1000_SRRCTL_DROP_EN; + + /* Setup buffer sizes */ + srrctl |= ALIGN(adapter->rx_buffer_len, 1024) >> + E1000_SRRCTL_BSIZEPKT_SHIFT; + + if (adapter->rx_buffer_len < 2048) { + adapter->rx_ps_hdr_size = 0; + srrctl |= E1000_SRRCTL_DESCTYPE_ADV_ONEBUF; + } else { + adapter->rx_ps_hdr_size = 128; + srrctl |= adapter->rx_ps_hdr_size << + E1000_SRRCTL_BSIZEHDRSIZE_SHIFT; + srrctl |= E1000_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS; + } + + ew32(SRRCTL(0), srrctl); +} + +/** + * igbvf_configure_rx - Configure Receive Unit after Reset + * @adapter: board private structure + * + * Configure the Rx unit of the MAC after a reset. + **/ +static void igbvf_configure_rx(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *rx_ring = adapter->rx_ring; + u64 rdba; + u32 rxdctl; + + /* disable receives */ + rxdctl = er32(RXDCTL(0)); + ew32(RXDCTL(0), rxdctl & ~E1000_RXDCTL_QUEUE_ENABLE); + e1e_flush(); + msleep(10); + + /* Setup the HW Rx Head and Tail Descriptor Pointers and + * the Base and Length of the Rx Descriptor Ring + */ + rdba = rx_ring->dma; + ew32(RDBAL(0), (rdba & DMA_BIT_MASK(32))); + ew32(RDBAH(0), (rdba >> 32)); + ew32(RDLEN(0), rx_ring->count * sizeof(union e1000_adv_rx_desc)); + rx_ring->head = E1000_RDH(0); + rx_ring->tail = E1000_RDT(0); + ew32(RDH(0), 0); + ew32(RDT(0), 0); + + rxdctl |= E1000_RXDCTL_QUEUE_ENABLE; + rxdctl &= 0xFFF00000; + rxdctl |= IGBVF_RX_PTHRESH; + rxdctl |= IGBVF_RX_HTHRESH << 8; + rxdctl |= IGBVF_RX_WTHRESH << 16; + + igbvf_set_rlpml(adapter); + + /* enable receives */ + ew32(RXDCTL(0), rxdctl); +} + +/** + * igbvf_set_multi - Multicast and Promiscuous mode set + * @netdev: network interface device structure + * + * The set_multi entry point is called whenever the multicast address + * list or the network interface flags are updated. This routine is + * responsible for configuring the hardware for proper multicast, + * promiscuous mode, and all-multi behavior. + **/ +static void igbvf_set_multi(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct netdev_hw_addr *ha; + u8 *mta_list = NULL; + int i; + + if (!netdev_mc_empty(netdev)) { + mta_list = kmalloc_array(netdev_mc_count(netdev), ETH_ALEN, + GFP_ATOMIC); + if (!mta_list) + return; + } + + /* prepare a packed array of only addresses. */ + i = 0; + netdev_for_each_mc_addr(ha, netdev) + memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN); + + spin_lock_bh(&hw->mbx_lock); + + hw->mac.ops.update_mc_addr_list(hw, mta_list, i, 0, 0); + + spin_unlock_bh(&hw->mbx_lock); + kfree(mta_list); +} + +/** + * igbvf_set_uni - Configure unicast MAC filters + * @netdev: network interface device structure + * + * This routine is responsible for configuring the hardware for proper + * unicast filters. + **/ +static int igbvf_set_uni(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + if (netdev_uc_count(netdev) > IGBVF_MAX_MAC_FILTERS) { + pr_err("Too many unicast filters - No Space\n"); + return -ENOSPC; + } + + spin_lock_bh(&hw->mbx_lock); + + /* Clear all unicast MAC filters */ + hw->mac.ops.set_uc_addr(hw, E1000_VF_MAC_FILTER_CLR, NULL); + + spin_unlock_bh(&hw->mbx_lock); + + if (!netdev_uc_empty(netdev)) { + struct netdev_hw_addr *ha; + + /* Add MAC filters one by one */ + netdev_for_each_uc_addr(ha, netdev) { + spin_lock_bh(&hw->mbx_lock); + + hw->mac.ops.set_uc_addr(hw, E1000_VF_MAC_FILTER_ADD, + ha->addr); + + spin_unlock_bh(&hw->mbx_lock); + udelay(200); + } + } + + return 0; +} + +static void igbvf_set_rx_mode(struct net_device *netdev) +{ + igbvf_set_multi(netdev); + igbvf_set_uni(netdev); +} + +/** + * igbvf_configure - configure the hardware for Rx and Tx + * @adapter: private board structure + **/ +static void igbvf_configure(struct igbvf_adapter *adapter) +{ + igbvf_set_rx_mode(adapter->netdev); + + igbvf_restore_vlan(adapter); + + igbvf_configure_tx(adapter); + igbvf_setup_srrctl(adapter); + igbvf_configure_rx(adapter); + igbvf_alloc_rx_buffers(adapter->rx_ring, + igbvf_desc_unused(adapter->rx_ring)); +} + +/* igbvf_reset - bring the hardware into a known good state + * @adapter: private board structure + * + * This function boots the hardware and enables some settings that + * require a configuration cycle of the hardware - those cannot be + * set/changed during runtime. After reset the device needs to be + * properly configured for Rx, Tx etc. + */ +static void igbvf_reset(struct igbvf_adapter *adapter) +{ + struct e1000_mac_info *mac = &adapter->hw.mac; + struct net_device *netdev = adapter->netdev; + struct e1000_hw *hw = &adapter->hw; + + spin_lock_bh(&hw->mbx_lock); + + /* Allow time for pending master requests to run */ + if (mac->ops.reset_hw(hw)) + dev_info(&adapter->pdev->dev, "PF still resetting\n"); + + mac->ops.init_hw(hw); + + spin_unlock_bh(&hw->mbx_lock); + + if (is_valid_ether_addr(adapter->hw.mac.addr)) { + eth_hw_addr_set(netdev, adapter->hw.mac.addr); + memcpy(netdev->perm_addr, adapter->hw.mac.addr, + netdev->addr_len); + } + + adapter->last_reset = jiffies; +} + +int igbvf_up(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + /* hardware has been reset, we need to reload some things */ + igbvf_configure(adapter); + + clear_bit(__IGBVF_DOWN, &adapter->state); + + napi_enable(&adapter->rx_ring->napi); + if (adapter->msix_entries) + igbvf_configure_msix(adapter); + + /* Clear any pending interrupts. */ + er32(EICR); + igbvf_irq_enable(adapter); + + /* start the watchdog */ + hw->mac.get_link_status = 1; + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + return 0; +} + +void igbvf_down(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + struct e1000_hw *hw = &adapter->hw; + u32 rxdctl, txdctl; + + /* signal that we're down so the interrupt handler does not + * reschedule our watchdog timer + */ + set_bit(__IGBVF_DOWN, &adapter->state); + + /* disable receives in the hardware */ + rxdctl = er32(RXDCTL(0)); + ew32(RXDCTL(0), rxdctl & ~E1000_RXDCTL_QUEUE_ENABLE); + + netif_carrier_off(netdev); + netif_stop_queue(netdev); + + /* disable transmits in the hardware */ + txdctl = er32(TXDCTL(0)); + ew32(TXDCTL(0), txdctl & ~E1000_TXDCTL_QUEUE_ENABLE); + + /* flush both disables and wait for them to finish */ + e1e_flush(); + msleep(10); + + napi_disable(&adapter->rx_ring->napi); + + igbvf_irq_disable(adapter); + + del_timer_sync(&adapter->watchdog_timer); + + /* record the stats before reset*/ + igbvf_update_stats(adapter); + + adapter->link_speed = 0; + adapter->link_duplex = 0; + + igbvf_reset(adapter); + igbvf_clean_tx_ring(adapter->tx_ring); + igbvf_clean_rx_ring(adapter->rx_ring); +} + +void igbvf_reinit_locked(struct igbvf_adapter *adapter) +{ + might_sleep(); + while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state)) + usleep_range(1000, 2000); + igbvf_down(adapter); + igbvf_up(adapter); + clear_bit(__IGBVF_RESETTING, &adapter->state); +} + +/** + * igbvf_sw_init - Initialize general software structures (struct igbvf_adapter) + * @adapter: board private structure to initialize + * + * igbvf_sw_init initializes the Adapter private data structure. + * Fields are initialized based on PCI device information and + * OS network device settings (MTU size). + **/ +static int igbvf_sw_init(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + s32 rc; + + adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN; + adapter->rx_ps_hdr_size = 0; + adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN; + adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN; + + adapter->tx_int_delay = 8; + adapter->tx_abs_int_delay = 32; + adapter->rx_int_delay = 0; + adapter->rx_abs_int_delay = 8; + adapter->requested_itr = 3; + adapter->current_itr = IGBVF_START_ITR; + + /* Set various function pointers */ + adapter->ei->init_ops(&adapter->hw); + + rc = adapter->hw.mac.ops.init_params(&adapter->hw); + if (rc) + return rc; + + rc = adapter->hw.mbx.ops.init_params(&adapter->hw); + if (rc) + return rc; + + igbvf_set_interrupt_capability(adapter); + + if (igbvf_alloc_queues(adapter)) + return -ENOMEM; + + spin_lock_init(&adapter->tx_queue_lock); + + /* Explicitly disable IRQ since the NIC can be in any state. */ + igbvf_irq_disable(adapter); + + spin_lock_init(&adapter->stats_lock); + spin_lock_init(&adapter->hw.mbx_lock); + + set_bit(__IGBVF_DOWN, &adapter->state); + return 0; +} + +static void igbvf_initialize_last_counter_stats(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + adapter->stats.last_gprc = er32(VFGPRC); + adapter->stats.last_gorc = er32(VFGORC); + adapter->stats.last_gptc = er32(VFGPTC); + adapter->stats.last_gotc = er32(VFGOTC); + adapter->stats.last_mprc = er32(VFMPRC); + adapter->stats.last_gotlbc = er32(VFGOTLBC); + adapter->stats.last_gptlbc = er32(VFGPTLBC); + adapter->stats.last_gorlbc = er32(VFGORLBC); + adapter->stats.last_gprlbc = er32(VFGPRLBC); + + adapter->stats.base_gprc = er32(VFGPRC); + adapter->stats.base_gorc = er32(VFGORC); + adapter->stats.base_gptc = er32(VFGPTC); + adapter->stats.base_gotc = er32(VFGOTC); + adapter->stats.base_mprc = er32(VFMPRC); + adapter->stats.base_gotlbc = er32(VFGOTLBC); + adapter->stats.base_gptlbc = er32(VFGPTLBC); + adapter->stats.base_gorlbc = er32(VFGORLBC); + adapter->stats.base_gprlbc = er32(VFGPRLBC); +} + +/** + * igbvf_open - Called when a network interface is made active + * @netdev: network interface device structure + * + * Returns 0 on success, negative value on failure + * + * The open entry point is called when a network interface is made + * active by the system (IFF_UP). At this point all resources needed + * for transmit and receive operations are allocated, the interrupt + * handler is registered with the OS, the watchdog timer is started, + * and the stack is notified that the interface is ready. + **/ +static int igbvf_open(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + int err; + + /* disallow open during test */ + if (test_bit(__IGBVF_TESTING, &adapter->state)) + return -EBUSY; + + /* allocate transmit descriptors */ + err = igbvf_setup_tx_resources(adapter, adapter->tx_ring); + if (err) + goto err_setup_tx; + + /* allocate receive descriptors */ + err = igbvf_setup_rx_resources(adapter, adapter->rx_ring); + if (err) + goto err_setup_rx; + + /* before we allocate an interrupt, we must be ready to handle it. + * Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt + * as soon as we call pci_request_irq, so we have to setup our + * clean_rx handler before we do so. + */ + igbvf_configure(adapter); + + err = igbvf_request_irq(adapter); + if (err) + goto err_req_irq; + + /* From here on the code is the same as igbvf_up() */ + clear_bit(__IGBVF_DOWN, &adapter->state); + + napi_enable(&adapter->rx_ring->napi); + + /* clear any pending interrupts */ + er32(EICR); + + igbvf_irq_enable(adapter); + + /* start the watchdog */ + hw->mac.get_link_status = 1; + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + return 0; + +err_req_irq: + igbvf_free_rx_resources(adapter->rx_ring); +err_setup_rx: + igbvf_free_tx_resources(adapter->tx_ring); +err_setup_tx: + igbvf_reset(adapter); + + return err; +} + +/** + * igbvf_close - Disables a network interface + * @netdev: network interface device structure + * + * Returns 0, this is not allowed to fail + * + * The close entry point is called when an interface is de-activated + * by the OS. The hardware is still under the drivers control, but + * needs to be disabled. A global MAC reset is issued to stop the + * hardware, and all transmit and receive resources are freed. + **/ +static int igbvf_close(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + WARN_ON(test_bit(__IGBVF_RESETTING, &adapter->state)); + igbvf_down(adapter); + + igbvf_free_irq(adapter); + + igbvf_free_tx_resources(adapter->tx_ring); + igbvf_free_rx_resources(adapter->rx_ring); + + return 0; +} + +/** + * igbvf_set_mac - Change the Ethernet Address of the NIC + * @netdev: network interface device structure + * @p: pointer to an address structure + * + * Returns 0 on success, negative on failure + **/ +static int igbvf_set_mac(struct net_device *netdev, void *p) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct sockaddr *addr = p; + + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(hw->mac.addr, addr->sa_data, netdev->addr_len); + + spin_lock_bh(&hw->mbx_lock); + + hw->mac.ops.rar_set(hw, hw->mac.addr, 0); + + spin_unlock_bh(&hw->mbx_lock); + + if (!ether_addr_equal(addr->sa_data, hw->mac.addr)) + return -EADDRNOTAVAIL; + + eth_hw_addr_set(netdev, addr->sa_data); + + return 0; +} + +#define UPDATE_VF_COUNTER(reg, name) \ +{ \ + u32 current_counter = er32(reg); \ + if (current_counter < adapter->stats.last_##name) \ + adapter->stats.name += 0x100000000LL; \ + adapter->stats.last_##name = current_counter; \ + adapter->stats.name &= 0xFFFFFFFF00000000LL; \ + adapter->stats.name |= current_counter; \ +} + +/** + * igbvf_update_stats - Update the board statistics counters + * @adapter: board private structure +**/ +void igbvf_update_stats(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct pci_dev *pdev = adapter->pdev; + + /* Prevent stats update while adapter is being reset, link is down + * or if the pci connection is down. + */ + if (adapter->link_speed == 0) + return; + + if (test_bit(__IGBVF_RESETTING, &adapter->state)) + return; + + if (pci_channel_offline(pdev)) + return; + + UPDATE_VF_COUNTER(VFGPRC, gprc); + UPDATE_VF_COUNTER(VFGORC, gorc); + UPDATE_VF_COUNTER(VFGPTC, gptc); + UPDATE_VF_COUNTER(VFGOTC, gotc); + UPDATE_VF_COUNTER(VFMPRC, mprc); + UPDATE_VF_COUNTER(VFGOTLBC, gotlbc); + UPDATE_VF_COUNTER(VFGPTLBC, gptlbc); + UPDATE_VF_COUNTER(VFGORLBC, gorlbc); + UPDATE_VF_COUNTER(VFGPRLBC, gprlbc); + + /* Fill out the OS statistics structure */ + adapter->netdev->stats.multicast = adapter->stats.mprc; +} + +static void igbvf_print_link_info(struct igbvf_adapter *adapter) +{ + dev_info(&adapter->pdev->dev, "Link is Up %d Mbps %s Duplex\n", + adapter->link_speed, + adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half"); +} + +static bool igbvf_has_link(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + s32 ret_val = E1000_SUCCESS; + bool link_active; + + /* If interface is down, stay link down */ + if (test_bit(__IGBVF_DOWN, &adapter->state)) + return false; + + spin_lock_bh(&hw->mbx_lock); + + ret_val = hw->mac.ops.check_for_link(hw); + + spin_unlock_bh(&hw->mbx_lock); + + link_active = !hw->mac.get_link_status; + + /* if check for link returns error we will need to reset */ + if (ret_val && time_after(jiffies, adapter->last_reset + (10 * HZ))) + schedule_work(&adapter->reset_task); + + return link_active; +} + +/** + * igbvf_watchdog - Timer Call-back + * @t: timer list pointer containing private struct + **/ +static void igbvf_watchdog(struct timer_list *t) +{ + struct igbvf_adapter *adapter = from_timer(adapter, t, watchdog_timer); + + /* Do the rest outside of interrupt context */ + schedule_work(&adapter->watchdog_task); +} + +static void igbvf_watchdog_task(struct work_struct *work) +{ + struct igbvf_adapter *adapter = container_of(work, + struct igbvf_adapter, + watchdog_task); + struct net_device *netdev = adapter->netdev; + struct e1000_mac_info *mac = &adapter->hw.mac; + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct e1000_hw *hw = &adapter->hw; + u32 link; + int tx_pending = 0; + + link = igbvf_has_link(adapter); + + if (link) { + if (!netif_carrier_ok(netdev)) { + mac->ops.get_link_up_info(&adapter->hw, + &adapter->link_speed, + &adapter->link_duplex); + igbvf_print_link_info(adapter); + + netif_carrier_on(netdev); + netif_wake_queue(netdev); + } + } else { + if (netif_carrier_ok(netdev)) { + adapter->link_speed = 0; + adapter->link_duplex = 0; + dev_info(&adapter->pdev->dev, "Link is Down\n"); + netif_carrier_off(netdev); + netif_stop_queue(netdev); + } + } + + if (netif_carrier_ok(netdev)) { + igbvf_update_stats(adapter); + } else { + tx_pending = (igbvf_desc_unused(tx_ring) + 1 < + tx_ring->count); + if (tx_pending) { + /* 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). + */ + adapter->tx_timeout_count++; + schedule_work(&adapter->reset_task); + } + } + + /* Cause software interrupt to ensure Rx ring is cleaned */ + ew32(EICS, adapter->rx_ring->eims_value); + + /* Reset the timer */ + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + mod_timer(&adapter->watchdog_timer, + round_jiffies(jiffies + (2 * HZ))); +} + +#define IGBVF_TX_FLAGS_CSUM 0x00000001 +#define IGBVF_TX_FLAGS_VLAN 0x00000002 +#define IGBVF_TX_FLAGS_TSO 0x00000004 +#define IGBVF_TX_FLAGS_IPV4 0x00000008 +#define IGBVF_TX_FLAGS_VLAN_MASK 0xffff0000 +#define IGBVF_TX_FLAGS_VLAN_SHIFT 16 + +static void igbvf_tx_ctxtdesc(struct igbvf_ring *tx_ring, u32 vlan_macip_lens, + u32 type_tucmd, u32 mss_l4len_idx) +{ + struct e1000_adv_tx_context_desc *context_desc; + struct igbvf_buffer *buffer_info; + u16 i = tx_ring->next_to_use; + + context_desc = IGBVF_TX_CTXTDESC_ADV(*tx_ring, i); + buffer_info = &tx_ring->buffer_info[i]; + + i++; + tx_ring->next_to_use = (i < tx_ring->count) ? i : 0; + + /* set bits to identify this as an advanced context descriptor */ + type_tucmd |= E1000_TXD_CMD_DEXT | E1000_ADVTXD_DTYP_CTXT; + + context_desc->vlan_macip_lens = cpu_to_le32(vlan_macip_lens); + context_desc->seqnum_seed = 0; + context_desc->type_tucmd_mlhl = cpu_to_le32(type_tucmd); + context_desc->mss_l4len_idx = cpu_to_le32(mss_l4len_idx); + + buffer_info->time_stamp = jiffies; + buffer_info->dma = 0; +} + +static int igbvf_tso(struct igbvf_ring *tx_ring, + struct sk_buff *skb, u32 tx_flags, u8 *hdr_len) +{ + u32 vlan_macip_lens, type_tucmd, mss_l4len_idx; + union { + struct iphdr *v4; + struct ipv6hdr *v6; + unsigned char *hdr; + } ip; + union { + struct tcphdr *tcp; + unsigned char *hdr; + } l4; + u32 paylen, l4_offset; + int err; + + if (skb->ip_summed != CHECKSUM_PARTIAL) + return 0; + + if (!skb_is_gso(skb)) + return 0; + + err = skb_cow_head(skb, 0); + if (err < 0) + return err; + + ip.hdr = skb_network_header(skb); + l4.hdr = skb_checksum_start(skb); + + /* ADV DTYP TUCMD MKRLOC/ISCSIHEDLEN */ + type_tucmd = E1000_ADVTXD_TUCMD_L4T_TCP; + + /* initialize outer IP header fields */ + if (ip.v4->version == 4) { + unsigned char *csum_start = skb_checksum_start(skb); + unsigned char *trans_start = ip.hdr + (ip.v4->ihl * 4); + + /* IP header will have to cancel out any data that + * is not a part of the outer IP header + */ + ip.v4->check = csum_fold(csum_partial(trans_start, + csum_start - trans_start, + 0)); + type_tucmd |= E1000_ADVTXD_TUCMD_IPV4; + + ip.v4->tot_len = 0; + } else { + ip.v6->payload_len = 0; + } + + /* determine offset of inner transport header */ + l4_offset = l4.hdr - skb->data; + + /* compute length of segmentation header */ + *hdr_len = (l4.tcp->doff * 4) + l4_offset; + + /* remove payload length from inner checksum */ + paylen = skb->len - l4_offset; + csum_replace_by_diff(&l4.tcp->check, (__force __wsum)htonl(paylen)); + + /* MSS L4LEN IDX */ + mss_l4len_idx = (*hdr_len - l4_offset) << E1000_ADVTXD_L4LEN_SHIFT; + mss_l4len_idx |= skb_shinfo(skb)->gso_size << E1000_ADVTXD_MSS_SHIFT; + + /* VLAN MACLEN IPLEN */ + vlan_macip_lens = l4.hdr - ip.hdr; + vlan_macip_lens |= (ip.hdr - skb->data) << E1000_ADVTXD_MACLEN_SHIFT; + vlan_macip_lens |= tx_flags & IGBVF_TX_FLAGS_VLAN_MASK; + + igbvf_tx_ctxtdesc(tx_ring, vlan_macip_lens, type_tucmd, mss_l4len_idx); + + return 1; +} + +static bool igbvf_tx_csum(struct igbvf_ring *tx_ring, struct sk_buff *skb, + u32 tx_flags, __be16 protocol) +{ + u32 vlan_macip_lens = 0; + u32 type_tucmd = 0; + + if (skb->ip_summed != CHECKSUM_PARTIAL) { +csum_failed: + if (!(tx_flags & IGBVF_TX_FLAGS_VLAN)) + return false; + goto no_csum; + } + + switch (skb->csum_offset) { + case offsetof(struct tcphdr, check): + type_tucmd = E1000_ADVTXD_TUCMD_L4T_TCP; + fallthrough; + case offsetof(struct udphdr, check): + break; + case offsetof(struct sctphdr, checksum): + /* validate that this is actually an SCTP request */ + if (skb_csum_is_sctp(skb)) { + type_tucmd = E1000_ADVTXD_TUCMD_L4T_SCTP; + break; + } + fallthrough; + default: + skb_checksum_help(skb); + goto csum_failed; + } + + vlan_macip_lens = skb_checksum_start_offset(skb) - + skb_network_offset(skb); +no_csum: + vlan_macip_lens |= skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT; + vlan_macip_lens |= tx_flags & IGBVF_TX_FLAGS_VLAN_MASK; + + igbvf_tx_ctxtdesc(tx_ring, vlan_macip_lens, type_tucmd, 0); + return true; +} + +static int igbvf_maybe_stop_tx(struct net_device *netdev, int size) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + /* there is enough descriptors then we don't need to worry */ + if (igbvf_desc_unused(adapter->tx_ring) >= size) + return 0; + + netif_stop_queue(netdev); + + /* Herbert's original patch had: + * smp_mb__after_netif_stop_queue(); + * but since that doesn't exist yet, just open code it. + */ + smp_mb(); + + /* We need to check again just in case room has been made available */ + if (igbvf_desc_unused(adapter->tx_ring) < size) + return -EBUSY; + + netif_wake_queue(netdev); + + ++adapter->restart_queue; + return 0; +} + +#define IGBVF_MAX_TXD_PWR 16 +#define IGBVF_MAX_DATA_PER_TXD (1u << IGBVF_MAX_TXD_PWR) + +static inline int igbvf_tx_map_adv(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring, + struct sk_buff *skb) +{ + struct igbvf_buffer *buffer_info; + struct pci_dev *pdev = adapter->pdev; + unsigned int len = skb_headlen(skb); + unsigned int count = 0, i; + unsigned int f; + + i = tx_ring->next_to_use; + + buffer_info = &tx_ring->buffer_info[i]; + BUG_ON(len >= IGBVF_MAX_DATA_PER_TXD); + buffer_info->length = len; + /* set time_stamp *before* dma to help avoid a possible race */ + buffer_info->time_stamp = jiffies; + buffer_info->mapped_as_page = false; + buffer_info->dma = dma_map_single(&pdev->dev, skb->data, len, + DMA_TO_DEVICE); + if (dma_mapping_error(&pdev->dev, buffer_info->dma)) + goto dma_error; + + for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) { + const skb_frag_t *frag; + + count++; + i++; + if (i == tx_ring->count) + i = 0; + + frag = &skb_shinfo(skb)->frags[f]; + len = skb_frag_size(frag); + + buffer_info = &tx_ring->buffer_info[i]; + BUG_ON(len >= IGBVF_MAX_DATA_PER_TXD); + buffer_info->length = len; + buffer_info->time_stamp = jiffies; + buffer_info->mapped_as_page = true; + buffer_info->dma = skb_frag_dma_map(&pdev->dev, frag, 0, len, + DMA_TO_DEVICE); + if (dma_mapping_error(&pdev->dev, buffer_info->dma)) + goto dma_error; + } + + tx_ring->buffer_info[i].skb = skb; + + return ++count; + +dma_error: + dev_err(&pdev->dev, "TX DMA map failed\n"); + + /* clear timestamp and dma mappings for failed buffer_info mapping */ + buffer_info->dma = 0; + buffer_info->time_stamp = 0; + buffer_info->length = 0; + buffer_info->mapped_as_page = false; + if (count) + count--; + + /* clear timestamp and dma mappings for remaining portion of packet */ + while (count--) { + if (i == 0) + i += tx_ring->count; + i--; + buffer_info = &tx_ring->buffer_info[i]; + igbvf_put_txbuf(adapter, buffer_info); + } + + return 0; +} + +static inline void igbvf_tx_queue_adv(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring, + int tx_flags, int count, + unsigned int first, u32 paylen, + u8 hdr_len) +{ + union e1000_adv_tx_desc *tx_desc = NULL; + struct igbvf_buffer *buffer_info; + u32 olinfo_status = 0, cmd_type_len; + unsigned int i; + + cmd_type_len = (E1000_ADVTXD_DTYP_DATA | E1000_ADVTXD_DCMD_IFCS | + E1000_ADVTXD_DCMD_DEXT); + + if (tx_flags & IGBVF_TX_FLAGS_VLAN) + cmd_type_len |= E1000_ADVTXD_DCMD_VLE; + + if (tx_flags & IGBVF_TX_FLAGS_TSO) { + cmd_type_len |= E1000_ADVTXD_DCMD_TSE; + + /* insert tcp checksum */ + olinfo_status |= E1000_TXD_POPTS_TXSM << 8; + + /* insert ip checksum */ + if (tx_flags & IGBVF_TX_FLAGS_IPV4) + olinfo_status |= E1000_TXD_POPTS_IXSM << 8; + + } else if (tx_flags & IGBVF_TX_FLAGS_CSUM) { + olinfo_status |= E1000_TXD_POPTS_TXSM << 8; + } + + olinfo_status |= ((paylen - hdr_len) << E1000_ADVTXD_PAYLEN_SHIFT); + + i = tx_ring->next_to_use; + while (count--) { + buffer_info = &tx_ring->buffer_info[i]; + tx_desc = IGBVF_TX_DESC_ADV(*tx_ring, i); + tx_desc->read.buffer_addr = cpu_to_le64(buffer_info->dma); + tx_desc->read.cmd_type_len = + cpu_to_le32(cmd_type_len | buffer_info->length); + tx_desc->read.olinfo_status = cpu_to_le32(olinfo_status); + i++; + if (i == tx_ring->count) + i = 0; + } + + tx_desc->read.cmd_type_len |= cpu_to_le32(adapter->txd_cmd); + /* 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(); + + tx_ring->buffer_info[first].next_to_watch = tx_desc; + tx_ring->next_to_use = i; + writel(i, adapter->hw.hw_addr + tx_ring->tail); +} + +static netdev_tx_t igbvf_xmit_frame_ring_adv(struct sk_buff *skb, + struct net_device *netdev, + struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + unsigned int first, tx_flags = 0; + u8 hdr_len = 0; + int count = 0; + int tso = 0; + __be16 protocol = vlan_get_protocol(skb); + + if (test_bit(__IGBVF_DOWN, &adapter->state)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + if (skb->len <= 0) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + /* need: count + 4 desc gap to keep tail from touching + * + 2 desc gap to keep tail from touching head, + * + 1 desc for skb->data, + * + 1 desc for context descriptor, + * head, otherwise try next time + */ + if (igbvf_maybe_stop_tx(netdev, skb_shinfo(skb)->nr_frags + 4)) { + /* this is a hard error */ + return NETDEV_TX_BUSY; + } + + if (skb_vlan_tag_present(skb)) { + tx_flags |= IGBVF_TX_FLAGS_VLAN; + tx_flags |= (skb_vlan_tag_get(skb) << + IGBVF_TX_FLAGS_VLAN_SHIFT); + } + + if (protocol == htons(ETH_P_IP)) + tx_flags |= IGBVF_TX_FLAGS_IPV4; + + first = tx_ring->next_to_use; + + tso = igbvf_tso(tx_ring, skb, tx_flags, &hdr_len); + if (unlikely(tso < 0)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + if (tso) + tx_flags |= IGBVF_TX_FLAGS_TSO; + else if (igbvf_tx_csum(tx_ring, skb, tx_flags, protocol) && + (skb->ip_summed == CHECKSUM_PARTIAL)) + tx_flags |= IGBVF_TX_FLAGS_CSUM; + + /* count reflects descriptors mapped, if 0 then mapping error + * has occurred and we need to rewind the descriptor queue + */ + count = igbvf_tx_map_adv(adapter, tx_ring, skb); + + if (count) { + igbvf_tx_queue_adv(adapter, tx_ring, tx_flags, count, + first, skb->len, hdr_len); + /* Make sure there is space in the ring for the next send. */ + igbvf_maybe_stop_tx(netdev, MAX_SKB_FRAGS + 4); + } else { + dev_kfree_skb_any(skb); + tx_ring->buffer_info[first].time_stamp = 0; + tx_ring->next_to_use = first; + } + + return NETDEV_TX_OK; +} + +static netdev_tx_t igbvf_xmit_frame(struct sk_buff *skb, + struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct igbvf_ring *tx_ring; + + if (test_bit(__IGBVF_DOWN, &adapter->state)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + tx_ring = &adapter->tx_ring[0]; + + return igbvf_xmit_frame_ring_adv(skb, netdev, tx_ring); +} + +/** + * igbvf_tx_timeout - Respond to a Tx Hang + * @netdev: network interface device structure + * @txqueue: queue timing out (unused) + **/ +static void igbvf_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + /* Do the reset outside of interrupt context */ + adapter->tx_timeout_count++; + schedule_work(&adapter->reset_task); +} + +static void igbvf_reset_task(struct work_struct *work) +{ + struct igbvf_adapter *adapter; + + adapter = container_of(work, struct igbvf_adapter, reset_task); + + igbvf_reinit_locked(adapter); +} + +/** + * igbvf_change_mtu - Change the Maximum Transfer Unit + * @netdev: network interface device structure + * @new_mtu: new value for maximum frame size + * + * Returns 0 on success, negative on failure + **/ +static int igbvf_change_mtu(struct net_device *netdev, int new_mtu) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN; + + while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state)) + usleep_range(1000, 2000); + /* igbvf_down has a dependency on max_frame_size */ + adapter->max_frame_size = max_frame; + if (netif_running(netdev)) + igbvf_down(adapter); + + /* NOTE: netdev_alloc_skb reserves 16 bytes, and typically NET_IP_ALIGN + * means we reserve 2 more, this pushes us to allocate from the next + * larger slab size. + * i.e. RXBUFFER_2048 --> size-4096 slab + * However with the new *_jumbo_rx* routines, jumbo receives will use + * fragmented skbs + */ + + if (max_frame <= 1024) + adapter->rx_buffer_len = 1024; + else if (max_frame <= 2048) + adapter->rx_buffer_len = 2048; + else +#if (PAGE_SIZE / 2) > 16384 + adapter->rx_buffer_len = 16384; +#else + adapter->rx_buffer_len = PAGE_SIZE / 2; +#endif + + /* adjust allocation if LPE protects us, and we aren't using SBP */ + if ((max_frame == ETH_FRAME_LEN + ETH_FCS_LEN) || + (max_frame == ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN)) + adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN + + ETH_FCS_LEN; + + netdev_dbg(netdev, "changing MTU from %d to %d\n", + netdev->mtu, new_mtu); + netdev->mtu = new_mtu; + + if (netif_running(netdev)) + igbvf_up(adapter); + else + igbvf_reset(adapter); + + clear_bit(__IGBVF_RESETTING, &adapter->state); + + return 0; +} + +static int igbvf_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +{ + switch (cmd) { + default: + return -EOPNOTSUPP; + } +} + +static int igbvf_suspend(struct device *dev_d) +{ + struct net_device *netdev = dev_get_drvdata(dev_d); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + netif_device_detach(netdev); + + if (netif_running(netdev)) { + WARN_ON(test_bit(__IGBVF_RESETTING, &adapter->state)); + igbvf_down(adapter); + igbvf_free_irq(adapter); + } + + return 0; +} + +static int __maybe_unused igbvf_resume(struct device *dev_d) +{ + struct pci_dev *pdev = to_pci_dev(dev_d); + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + u32 err; + + pci_set_master(pdev); + + if (netif_running(netdev)) { + err = igbvf_request_irq(adapter); + if (err) + return err; + } + + igbvf_reset(adapter); + + if (netif_running(netdev)) + igbvf_up(adapter); + + netif_device_attach(netdev); + + return 0; +} + +static void igbvf_shutdown(struct pci_dev *pdev) +{ + igbvf_suspend(&pdev->dev); +} + +#ifdef CONFIG_NET_POLL_CONTROLLER +/* Polling 'interrupt' - used by things like netconsole to send skbs + * without having to re-enable interrupts. It's not called while + * the interrupt routine is executing. + */ +static void igbvf_netpoll(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + disable_irq(adapter->pdev->irq); + + igbvf_clean_tx_irq(adapter->tx_ring); + + enable_irq(adapter->pdev->irq); +} +#endif + +/** + * igbvf_io_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current pci connection state + * + * This function is called after a PCI bus error affecting + * this device has been detected. + */ +static pci_ers_result_t igbvf_io_error_detected(struct pci_dev *pdev, + pci_channel_state_t state) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + netif_device_detach(netdev); + + if (state == pci_channel_io_perm_failure) + return PCI_ERS_RESULT_DISCONNECT; + + if (netif_running(netdev)) + igbvf_down(adapter); + pci_disable_device(pdev); + + /* Request a slot reset. */ + return PCI_ERS_RESULT_NEED_RESET; +} + +/** + * igbvf_io_slot_reset - called after the pci bus has been reset. + * @pdev: Pointer to PCI device + * + * Restart the card from scratch, as if from a cold-boot. Implementation + * resembles the first-half of the igbvf_resume routine. + */ +static pci_ers_result_t igbvf_io_slot_reset(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (pci_enable_device_mem(pdev)) { + dev_err(&pdev->dev, + "Cannot re-enable PCI device after reset.\n"); + return PCI_ERS_RESULT_DISCONNECT; + } + pci_set_master(pdev); + + igbvf_reset(adapter); + + return PCI_ERS_RESULT_RECOVERED; +} + +/** + * igbvf_io_resume - called when traffic can start flowing again. + * @pdev: Pointer to PCI device + * + * This callback is called when the error recovery driver tells us that + * its OK to resume normal operation. Implementation resembles the + * second-half of the igbvf_resume routine. + */ +static void igbvf_io_resume(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (netif_running(netdev)) { + if (igbvf_up(adapter)) { + dev_err(&pdev->dev, + "can't bring device back up after reset\n"); + return; + } + } + + netif_device_attach(netdev); +} + +static void igbvf_print_device_info(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + + if (hw->mac.type == e1000_vfadapt_i350) + dev_info(&pdev->dev, "Intel(R) I350 Virtual Function\n"); + else + dev_info(&pdev->dev, "Intel(R) 82576 Virtual Function\n"); + dev_info(&pdev->dev, "Address: %pM\n", netdev->dev_addr); +} + +static int igbvf_set_features(struct net_device *netdev, + netdev_features_t features) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (features & NETIF_F_RXCSUM) + adapter->flags &= ~IGBVF_FLAG_RX_CSUM_DISABLED; + else + adapter->flags |= IGBVF_FLAG_RX_CSUM_DISABLED; + + return 0; +} + +#define IGBVF_MAX_MAC_HDR_LEN 127 +#define IGBVF_MAX_NETWORK_HDR_LEN 511 + +static netdev_features_t +igbvf_features_check(struct sk_buff *skb, struct net_device *dev, + netdev_features_t features) +{ + unsigned int network_hdr_len, mac_hdr_len; + + /* Make certain the headers can be described by a context descriptor */ + mac_hdr_len = skb_network_header(skb) - skb->data; + if (unlikely(mac_hdr_len > IGBVF_MAX_MAC_HDR_LEN)) + return features & ~(NETIF_F_HW_CSUM | + NETIF_F_SCTP_CRC | + NETIF_F_HW_VLAN_CTAG_TX | + NETIF_F_TSO | + NETIF_F_TSO6); + + network_hdr_len = skb_checksum_start(skb) - skb_network_header(skb); + if (unlikely(network_hdr_len > IGBVF_MAX_NETWORK_HDR_LEN)) + return features & ~(NETIF_F_HW_CSUM | + NETIF_F_SCTP_CRC | + NETIF_F_TSO | + NETIF_F_TSO6); + + /* We can only support IPV4 TSO in tunnels if we can mangle the + * inner IP ID field, so strip TSO if MANGLEID is not supported. + */ + if (skb->encapsulation && !(features & NETIF_F_TSO_MANGLEID)) + features &= ~NETIF_F_TSO; + + return features; +} + +static const struct net_device_ops igbvf_netdev_ops = { + .ndo_open = igbvf_open, + .ndo_stop = igbvf_close, + .ndo_start_xmit = igbvf_xmit_frame, + .ndo_set_rx_mode = igbvf_set_rx_mode, + .ndo_set_mac_address = igbvf_set_mac, + .ndo_change_mtu = igbvf_change_mtu, + .ndo_eth_ioctl = igbvf_ioctl, + .ndo_tx_timeout = igbvf_tx_timeout, + .ndo_vlan_rx_add_vid = igbvf_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = igbvf_vlan_rx_kill_vid, +#ifdef CONFIG_NET_POLL_CONTROLLER + .ndo_poll_controller = igbvf_netpoll, +#endif + .ndo_set_features = igbvf_set_features, + .ndo_features_check = igbvf_features_check, +}; + +/** + * igbvf_probe - Device Initialization Routine + * @pdev: PCI device information struct + * @ent: entry in igbvf_pci_tbl + * + * Returns 0 on success, negative on failure + * + * igbvf_probe initializes an adapter identified by a pci_dev structure. + * The OS initialization, configuring of the adapter private structure, + * and a hardware reset occur. + **/ +static int igbvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + struct net_device *netdev; + struct igbvf_adapter *adapter; + struct e1000_hw *hw; + const struct igbvf_info *ei = igbvf_info_tbl[ent->driver_data]; + static int cards_found; + int err; + + err = pci_enable_device_mem(pdev); + if (err) + return err; + + err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); + if (err) { + dev_err(&pdev->dev, + "No usable DMA configuration, aborting\n"); + goto err_dma; + } + + err = pci_request_regions(pdev, igbvf_driver_name); + if (err) + goto err_pci_reg; + + pci_set_master(pdev); + + err = -ENOMEM; + netdev = alloc_etherdev(sizeof(struct igbvf_adapter)); + if (!netdev) + goto err_alloc_etherdev; + + SET_NETDEV_DEV(netdev, &pdev->dev); + + pci_set_drvdata(pdev, netdev); + adapter = netdev_priv(netdev); + hw = &adapter->hw; + adapter->netdev = netdev; + adapter->pdev = pdev; + adapter->ei = ei; + adapter->pba = ei->pba; + adapter->flags = ei->flags; + adapter->hw.back = adapter; + adapter->hw.mac.type = ei->mac; + adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); + + /* PCI config space info */ + + hw->vendor_id = pdev->vendor; + hw->device_id = pdev->device; + hw->subsystem_vendor_id = pdev->subsystem_vendor; + hw->subsystem_device_id = pdev->subsystem_device; + hw->revision_id = pdev->revision; + + err = -EIO; + adapter->hw.hw_addr = ioremap(pci_resource_start(pdev, 0), + pci_resource_len(pdev, 0)); + + if (!adapter->hw.hw_addr) + goto err_ioremap; + + if (ei->get_variants) { + err = ei->get_variants(adapter); + if (err) + goto err_get_variants; + } + + /* setup adapter struct */ + err = igbvf_sw_init(adapter); + if (err) + goto err_sw_init; + + /* construct the net_device struct */ + netdev->netdev_ops = &igbvf_netdev_ops; + + igbvf_set_ethtool_ops(netdev); + netdev->watchdog_timeo = 5 * HZ; + strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1); + + adapter->bd_number = cards_found++; + + netdev->hw_features = NETIF_F_SG | + NETIF_F_TSO | + NETIF_F_TSO6 | + NETIF_F_RXCSUM | + NETIF_F_HW_CSUM | + NETIF_F_SCTP_CRC; + +#define IGBVF_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE | \ + NETIF_F_GSO_GRE_CSUM | \ + NETIF_F_GSO_IPXIP4 | \ + NETIF_F_GSO_IPXIP6 | \ + NETIF_F_GSO_UDP_TUNNEL | \ + NETIF_F_GSO_UDP_TUNNEL_CSUM) + + netdev->gso_partial_features = IGBVF_GSO_PARTIAL_FEATURES; + netdev->hw_features |= NETIF_F_GSO_PARTIAL | + IGBVF_GSO_PARTIAL_FEATURES; + + netdev->features = netdev->hw_features | NETIF_F_HIGHDMA; + + netdev->vlan_features |= netdev->features | NETIF_F_TSO_MANGLEID; + netdev->mpls_features |= NETIF_F_HW_CSUM; + netdev->hw_enc_features |= netdev->vlan_features; + + /* set this bit last since it cannot be part of vlan_features */ + netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER | + NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_CTAG_TX; + + /* MTU range: 68 - 9216 */ + netdev->min_mtu = ETH_MIN_MTU; + netdev->max_mtu = MAX_STD_JUMBO_FRAME_SIZE; + + spin_lock_bh(&hw->mbx_lock); + + /*reset the controller to put the device in a known good state */ + err = hw->mac.ops.reset_hw(hw); + if (err) { + dev_info(&pdev->dev, + "PF still in reset state. Is the PF interface up?\n"); + } else { + err = hw->mac.ops.read_mac_addr(hw); + if (err) + dev_info(&pdev->dev, "Error reading MAC address.\n"); + else if (is_zero_ether_addr(adapter->hw.mac.addr)) + dev_info(&pdev->dev, + "MAC address not assigned by administrator.\n"); + eth_hw_addr_set(netdev, adapter->hw.mac.addr); + } + + spin_unlock_bh(&hw->mbx_lock); + + if (!is_valid_ether_addr(netdev->dev_addr)) { + dev_info(&pdev->dev, "Assigning random MAC address.\n"); + eth_hw_addr_random(netdev); + memcpy(adapter->hw.mac.addr, netdev->dev_addr, + netdev->addr_len); + } + + timer_setup(&adapter->watchdog_timer, igbvf_watchdog, 0); + + INIT_WORK(&adapter->reset_task, igbvf_reset_task); + INIT_WORK(&adapter->watchdog_task, igbvf_watchdog_task); + + /* ring size defaults */ + adapter->rx_ring->count = 1024; + adapter->tx_ring->count = 1024; + + /* reset the hardware with the new settings */ + igbvf_reset(adapter); + + /* set hardware-specific flags */ + if (adapter->hw.mac.type == e1000_vfadapt_i350) + adapter->flags |= IGBVF_FLAG_RX_LB_VLAN_BSWAP; + + strcpy(netdev->name, "eth%d"); + err = register_netdev(netdev); + if (err) + goto err_hw_init; + + /* tell the stack to leave us alone until igbvf_open() is called */ + netif_carrier_off(netdev); + netif_stop_queue(netdev); + + igbvf_print_device_info(adapter); + + igbvf_initialize_last_counter_stats(adapter); + + return 0; + +err_hw_init: + netif_napi_del(&adapter->rx_ring->napi); + kfree(adapter->tx_ring); + kfree(adapter->rx_ring); +err_sw_init: + igbvf_reset_interrupt_capability(adapter); +err_get_variants: + iounmap(adapter->hw.hw_addr); +err_ioremap: + free_netdev(netdev); +err_alloc_etherdev: + pci_release_regions(pdev); +err_pci_reg: +err_dma: + pci_disable_device(pdev); + return err; +} + +/** + * igbvf_remove - Device Removal Routine + * @pdev: PCI device information struct + * + * igbvf_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. The could be caused by a + * Hot-Plug event, or because the driver is going to be removed from + * memory. + **/ +static void igbvf_remove(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + /* The watchdog timer may be rescheduled, so explicitly + * disable it from being rescheduled. + */ + set_bit(__IGBVF_DOWN, &adapter->state); + del_timer_sync(&adapter->watchdog_timer); + + cancel_work_sync(&adapter->reset_task); + cancel_work_sync(&adapter->watchdog_task); + + unregister_netdev(netdev); + + igbvf_reset_interrupt_capability(adapter); + + /* it is important to delete the NAPI struct prior to freeing the + * Rx ring so that you do not end up with null pointer refs + */ + netif_napi_del(&adapter->rx_ring->napi); + kfree(adapter->tx_ring); + kfree(adapter->rx_ring); + + iounmap(hw->hw_addr); + if (hw->flash_address) + iounmap(hw->flash_address); + pci_release_regions(pdev); + + free_netdev(netdev); + + pci_disable_device(pdev); +} + +/* PCI Error Recovery (ERS) */ +static const struct pci_error_handlers igbvf_err_handler = { + .error_detected = igbvf_io_error_detected, + .slot_reset = igbvf_io_slot_reset, + .resume = igbvf_io_resume, +}; + +static const struct pci_device_id igbvf_pci_tbl[] = { + { PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_VF), board_vf }, + { PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_VF), board_i350_vf }, + { } /* terminate list */ +}; +MODULE_DEVICE_TABLE(pci, igbvf_pci_tbl); + +static SIMPLE_DEV_PM_OPS(igbvf_pm_ops, igbvf_suspend, igbvf_resume); + +/* PCI Device API Driver */ +static struct pci_driver igbvf_driver = { + .name = igbvf_driver_name, + .id_table = igbvf_pci_tbl, + .probe = igbvf_probe, + .remove = igbvf_remove, + .driver.pm = &igbvf_pm_ops, + .shutdown = igbvf_shutdown, + .err_handler = &igbvf_err_handler +}; + +/** + * igbvf_init_module - Driver Registration Routine + * + * igbvf_init_module is the first routine called when the driver is + * loaded. All it does is register with the PCI subsystem. + **/ +static int __init igbvf_init_module(void) +{ + int ret; + + pr_info("%s\n", igbvf_driver_string); + pr_info("%s\n", igbvf_copyright); + + ret = pci_register_driver(&igbvf_driver); + + return ret; +} +module_init(igbvf_init_module); + +/** + * igbvf_exit_module - Driver Exit Cleanup Routine + * + * igbvf_exit_module is called just before the driver is removed + * from memory. + **/ +static void __exit igbvf_exit_module(void) +{ + pci_unregister_driver(&igbvf_driver); +} +module_exit(igbvf_exit_module); + +MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>"); +MODULE_DESCRIPTION("Intel(R) Gigabit Virtual Function Network Driver"); +MODULE_LICENSE("GPL v2"); + +/* netdev.c */ diff --git a/drivers/net/ethernet/intel/igbvf/regs.h b/drivers/net/ethernet/intel/igbvf/regs.h new file mode 100644 index 000000000..625a309a3 --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/regs.h @@ -0,0 +1,84 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +#ifndef _E1000_REGS_H_ +#define _E1000_REGS_H_ + +#define E1000_CTRL 0x00000 /* Device Control - RW */ +#define E1000_STATUS 0x00008 /* Device Status - RO */ +#define E1000_ITR 0x000C4 /* Interrupt Throttling Rate - RW */ +#define E1000_EICR 0x01580 /* Ext. Interrupt Cause Read - R/clr */ +#define E1000_EITR(_n) (0x01680 + (0x4 * (_n))) +#define E1000_EICS 0x01520 /* Ext. Interrupt Cause Set - W0 */ +#define E1000_EIMS 0x01524 /* Ext. Interrupt Mask Set/Read - RW */ +#define E1000_EIMC 0x01528 /* Ext. Interrupt Mask Clear - WO */ +#define E1000_EIAC 0x0152C /* Ext. Interrupt Auto Clear - RW */ +#define E1000_EIAM 0x01530 /* Ext. Interrupt Ack Auto Clear Mask - RW */ +#define E1000_IVAR0 0x01700 /* Interrupt Vector Allocation (array) - RW */ +#define E1000_IVAR_MISC 0x01740 /* IVAR for "other" causes - RW */ + +/* Convenience macros + * + * Note: "_n" is the queue number of the register to be written to. + * + * Example usage: + * E1000_RDBAL_REG(current_rx_queue) + */ +#define E1000_RDBAL(_n) ((_n) < 4 ? (0x02800 + ((_n) * 0x100)) : \ + (0x0C000 + ((_n) * 0x40))) +#define E1000_RDBAH(_n) ((_n) < 4 ? (0x02804 + ((_n) * 0x100)) : \ + (0x0C004 + ((_n) * 0x40))) +#define E1000_RDLEN(_n) ((_n) < 4 ? (0x02808 + ((_n) * 0x100)) : \ + (0x0C008 + ((_n) * 0x40))) +#define E1000_SRRCTL(_n) ((_n) < 4 ? (0x0280C + ((_n) * 0x100)) : \ + (0x0C00C + ((_n) * 0x40))) +#define E1000_RDH(_n) ((_n) < 4 ? (0x02810 + ((_n) * 0x100)) : \ + (0x0C010 + ((_n) * 0x40))) +#define E1000_RDT(_n) ((_n) < 4 ? (0x02818 + ((_n) * 0x100)) : \ + (0x0C018 + ((_n) * 0x40))) +#define E1000_RXDCTL(_n) ((_n) < 4 ? (0x02828 + ((_n) * 0x100)) : \ + (0x0C028 + ((_n) * 0x40))) +#define E1000_TDBAL(_n) ((_n) < 4 ? (0x03800 + ((_n) * 0x100)) : \ + (0x0E000 + ((_n) * 0x40))) +#define E1000_TDBAH(_n) ((_n) < 4 ? (0x03804 + ((_n) * 0x100)) : \ + (0x0E004 + ((_n) * 0x40))) +#define E1000_TDLEN(_n) ((_n) < 4 ? (0x03808 + ((_n) * 0x100)) : \ + (0x0E008 + ((_n) * 0x40))) +#define E1000_TDH(_n) ((_n) < 4 ? (0x03810 + ((_n) * 0x100)) : \ + (0x0E010 + ((_n) * 0x40))) +#define E1000_TDT(_n) ((_n) < 4 ? (0x03818 + ((_n) * 0x100)) : \ + (0x0E018 + ((_n) * 0x40))) +#define E1000_TXDCTL(_n) ((_n) < 4 ? (0x03828 + ((_n) * 0x100)) : \ + (0x0E028 + ((_n) * 0x40))) +#define E1000_DCA_TXCTRL(_n) (0x03814 + (_n << 8)) +#define E1000_DCA_RXCTRL(_n) (0x02814 + (_n << 8)) +#define E1000_RAL(_i) (((_i) <= 15) ? (0x05400 + ((_i) * 8)) : \ + (0x054E0 + ((_i - 16) * 8))) +#define E1000_RAH(_i) (((_i) <= 15) ? (0x05404 + ((_i) * 8)) : \ + (0x054E4 + ((_i - 16) * 8))) + +/* Statistics registers */ +#define E1000_VFGPRC 0x00F10 +#define E1000_VFGORC 0x00F18 +#define E1000_VFMPRC 0x00F3C +#define E1000_VFGPTC 0x00F14 +#define E1000_VFGOTC 0x00F34 +#define E1000_VFGOTLBC 0x00F50 +#define E1000_VFGPTLBC 0x00F44 +#define E1000_VFGORLBC 0x00F48 +#define E1000_VFGPRLBC 0x00F40 + +/* These act per VF so an array friendly macro is used */ +#define E1000_V2PMAILBOX(_n) (0x00C40 + (4 * (_n))) +#define E1000_VMBMEM(_n) (0x00800 + (64 * (_n))) + +/* Define macros for handling registers */ +#define er32(reg) readl(hw->hw_addr + E1000_##reg) +#define ew32(reg, val) writel((val), hw->hw_addr + E1000_##reg) +#define array_er32(reg, offset) \ + readl(hw->hw_addr + E1000_##reg + (offset << 2)) +#define array_ew32(reg, offset, val) \ + writel((val), hw->hw_addr + E1000_##reg + (offset << 2)) +#define e1e_flush() er32(STATUS) + +#endif diff --git a/drivers/net/ethernet/intel/igbvf/vf.c b/drivers/net/ethernet/intel/igbvf/vf.c new file mode 100644 index 000000000..a47a2e3e5 --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/vf.c @@ -0,0 +1,430 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +#include <linux/etherdevice.h> + +#include "vf.h" + +static s32 e1000_check_for_link_vf(struct e1000_hw *hw); +static s32 e1000_get_link_up_info_vf(struct e1000_hw *hw, u16 *speed, + u16 *duplex); +static s32 e1000_init_hw_vf(struct e1000_hw *hw); +static s32 e1000_reset_hw_vf(struct e1000_hw *hw); + +static void e1000_update_mc_addr_list_vf(struct e1000_hw *hw, u8 *, + u32, u32, u32); +static void e1000_rar_set_vf(struct e1000_hw *, u8 *, u32); +static s32 e1000_read_mac_addr_vf(struct e1000_hw *); +static s32 e1000_set_uc_addr_vf(struct e1000_hw *hw, u32 subcmd, u8 *addr); +static s32 e1000_set_vfta_vf(struct e1000_hw *, u16, bool); + +/** + * e1000_init_mac_params_vf - Inits MAC params + * @hw: pointer to the HW structure + **/ +static s32 e1000_init_mac_params_vf(struct e1000_hw *hw) +{ + struct e1000_mac_info *mac = &hw->mac; + + /* VF's have no MTA Registers - PF feature only */ + mac->mta_reg_count = 128; + /* VF's have no access to RAR entries */ + mac->rar_entry_count = 1; + + /* Function pointers */ + /* reset */ + mac->ops.reset_hw = e1000_reset_hw_vf; + /* hw initialization */ + mac->ops.init_hw = e1000_init_hw_vf; + /* check for link */ + mac->ops.check_for_link = e1000_check_for_link_vf; + /* link info */ + mac->ops.get_link_up_info = e1000_get_link_up_info_vf; + /* multicast address update */ + mac->ops.update_mc_addr_list = e1000_update_mc_addr_list_vf; + /* set mac address */ + mac->ops.rar_set = e1000_rar_set_vf; + /* read mac address */ + mac->ops.read_mac_addr = e1000_read_mac_addr_vf; + /* set mac filter */ + mac->ops.set_uc_addr = e1000_set_uc_addr_vf; + /* set vlan filter table array */ + mac->ops.set_vfta = e1000_set_vfta_vf; + + return E1000_SUCCESS; +} + +/** + * e1000_init_function_pointers_vf - Inits function pointers + * @hw: pointer to the HW structure + **/ +void e1000_init_function_pointers_vf(struct e1000_hw *hw) +{ + hw->mac.ops.init_params = e1000_init_mac_params_vf; + hw->mbx.ops.init_params = e1000_init_mbx_params_vf; +} + +/** + * e1000_get_link_up_info_vf - Gets link info. + * @hw: pointer to the HW structure + * @speed: pointer to 16 bit value to store link speed. + * @duplex: pointer to 16 bit value to store duplex. + * + * Since we cannot read the PHY and get accurate link info, we must rely upon + * the status register's data which is often stale and inaccurate. + **/ +static s32 e1000_get_link_up_info_vf(struct e1000_hw *hw, u16 *speed, + u16 *duplex) +{ + s32 status; + + status = er32(STATUS); + if (status & E1000_STATUS_SPEED_1000) + *speed = SPEED_1000; + else if (status & E1000_STATUS_SPEED_100) + *speed = SPEED_100; + else + *speed = SPEED_10; + + if (status & E1000_STATUS_FD) + *duplex = FULL_DUPLEX; + else + *duplex = HALF_DUPLEX; + + return E1000_SUCCESS; +} + +/** + * e1000_reset_hw_vf - Resets the HW + * @hw: pointer to the HW structure + * + * VF's provide a function level reset. This is done using bit 26 of ctrl_reg. + * This is all the reset we can perform on a VF. + **/ +static s32 e1000_reset_hw_vf(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 timeout = E1000_VF_INIT_TIMEOUT; + u32 ret_val = -E1000_ERR_MAC_INIT; + u32 msgbuf[3]; + u8 *addr = (u8 *)(&msgbuf[1]); + u32 ctrl; + + /* assert VF queue/interrupt reset */ + ctrl = er32(CTRL); + ew32(CTRL, ctrl | E1000_CTRL_RST); + + /* we cannot initialize while the RSTI / RSTD bits are asserted */ + while (!mbx->ops.check_for_rst(hw) && timeout) { + timeout--; + udelay(5); + } + + if (timeout) { + /* mailbox timeout can now become active */ + mbx->timeout = E1000_VF_MBX_INIT_TIMEOUT; + + /* notify PF of VF reset completion */ + msgbuf[0] = E1000_VF_RESET; + mbx->ops.write_posted(hw, msgbuf, 1); + + mdelay(10); + + /* set our "perm_addr" based on info provided by PF */ + ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + if (!ret_val) { + switch (msgbuf[0]) { + case E1000_VF_RESET | E1000_VT_MSGTYPE_ACK: + memcpy(hw->mac.perm_addr, addr, ETH_ALEN); + break; + case E1000_VF_RESET | E1000_VT_MSGTYPE_NACK: + eth_zero_addr(hw->mac.perm_addr); + break; + default: + ret_val = -E1000_ERR_MAC_INIT; + } + } + } + + return ret_val; +} + +/** + * e1000_init_hw_vf - Inits the HW + * @hw: pointer to the HW structure + * + * Not much to do here except clear the PF Reset indication if there is one. + **/ +static s32 e1000_init_hw_vf(struct e1000_hw *hw) +{ + /* attempt to set and restore our mac address */ + e1000_rar_set_vf(hw, hw->mac.addr, 0); + + return E1000_SUCCESS; +} + +/** + * e1000_hash_mc_addr_vf - Generate a multicast hash value + * @hw: pointer to the HW structure + * @mc_addr: pointer to a multicast address + * + * Generates a multicast address hash value which is used to determine + * the multicast filter table array address and new table value. See + * e1000_mta_set_generic() + **/ +static u32 e1000_hash_mc_addr_vf(struct e1000_hw *hw, u8 *mc_addr) +{ + u32 hash_value, hash_mask; + u8 bit_shift = 0; + + /* Register count multiplied by bits per register */ + hash_mask = (hw->mac.mta_reg_count * 32) - 1; + + /* The bit_shift is the number of left-shifts + * where 0xFF would still fall within the hash mask. + */ + while (hash_mask >> bit_shift != 0xFF) + bit_shift++; + + hash_value = hash_mask & (((mc_addr[4] >> (8 - bit_shift)) | + (((u16)mc_addr[5]) << bit_shift))); + + return hash_value; +} + +/** + * e1000_update_mc_addr_list_vf - Update Multicast addresses + * @hw: pointer to the HW structure + * @mc_addr_list: array of multicast addresses to program + * @mc_addr_count: number of multicast addresses to program + * @rar_used_count: the first RAR register free to program + * @rar_count: total number of supported Receive Address Registers + * + * Updates the Receive Address Registers and Multicast Table Array. + * The caller must have a packed mc_addr_list of multicast addresses. + * The parameter rar_count will usually be hw->mac.rar_entry_count + * unless there are workarounds that change this. + **/ +static void e1000_update_mc_addr_list_vf(struct e1000_hw *hw, + u8 *mc_addr_list, u32 mc_addr_count, + u32 rar_used_count, u32 rar_count) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[E1000_VFMAILBOX_SIZE]; + u16 *hash_list = (u16 *)&msgbuf[1]; + u32 hash_value; + u32 cnt, i; + s32 ret_val; + + /* Each entry in the list uses 1 16 bit word. We have 30 + * 16 bit words available in our HW msg buffer (minus 1 for the + * msg type). That's 30 hash values if we pack 'em right. If + * there are more than 30 MC addresses to add then punt the + * extras for now and then add code to handle more than 30 later. + * It would be unusual for a server to request that many multi-cast + * addresses except for in large enterprise network environments. + */ + + cnt = (mc_addr_count > 30) ? 30 : mc_addr_count; + msgbuf[0] = E1000_VF_SET_MULTICAST; + msgbuf[0] |= cnt << E1000_VT_MSGINFO_SHIFT; + + for (i = 0; i < cnt; i++) { + hash_value = e1000_hash_mc_addr_vf(hw, mc_addr_list); + hash_list[i] = hash_value & 0x0FFFF; + mc_addr_list += ETH_ALEN; + } + + ret_val = mbx->ops.write_posted(hw, msgbuf, E1000_VFMAILBOX_SIZE); + if (!ret_val) + mbx->ops.read_posted(hw, msgbuf, 1); +} + +/** + * e1000_set_vfta_vf - Set/Unset vlan filter table address + * @hw: pointer to the HW structure + * @vid: determines the vfta register and bit to set/unset + * @set: if true then set bit, else clear bit + **/ +static s32 e1000_set_vfta_vf(struct e1000_hw *hw, u16 vid, bool set) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[2]; + s32 err; + + msgbuf[0] = E1000_VF_SET_VLAN; + msgbuf[1] = vid; + /* Setting the 8 bit field MSG INFO to true indicates "add" */ + if (set) + msgbuf[0] |= BIT(E1000_VT_MSGINFO_SHIFT); + + mbx->ops.write_posted(hw, msgbuf, 2); + + err = mbx->ops.read_posted(hw, msgbuf, 2); + + msgbuf[0] &= ~E1000_VT_MSGTYPE_CTS; + + /* if nacked the vlan was rejected */ + if (!err && (msgbuf[0] == (E1000_VF_SET_VLAN | E1000_VT_MSGTYPE_NACK))) + err = -E1000_ERR_MAC_INIT; + + return err; +} + +/** + * e1000_rlpml_set_vf - Set the maximum receive packet length + * @hw: pointer to the HW structure + * @max_size: value to assign to max frame size + **/ +void e1000_rlpml_set_vf(struct e1000_hw *hw, u16 max_size) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[2]; + s32 ret_val; + + msgbuf[0] = E1000_VF_SET_LPE; + msgbuf[1] = max_size; + + ret_val = mbx->ops.write_posted(hw, msgbuf, 2); + if (!ret_val) + mbx->ops.read_posted(hw, msgbuf, 1); +} + +/** + * e1000_rar_set_vf - set device MAC address + * @hw: pointer to the HW structure + * @addr: pointer to the receive address + * @index: receive address array register + **/ +static void e1000_rar_set_vf(struct e1000_hw *hw, u8 *addr, u32 index) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[3]; + u8 *msg_addr = (u8 *)(&msgbuf[1]); + s32 ret_val; + + memset(msgbuf, 0, 12); + msgbuf[0] = E1000_VF_SET_MAC_ADDR; + memcpy(msg_addr, addr, ETH_ALEN); + ret_val = mbx->ops.write_posted(hw, msgbuf, 3); + + if (!ret_val) + ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + + msgbuf[0] &= ~E1000_VT_MSGTYPE_CTS; + + /* if nacked the address was rejected, use "perm_addr" */ + if (!ret_val && + (msgbuf[0] == (E1000_VF_SET_MAC_ADDR | E1000_VT_MSGTYPE_NACK))) + e1000_read_mac_addr_vf(hw); +} + +/** + * e1000_read_mac_addr_vf - Read device MAC address + * @hw: pointer to the HW structure + **/ +static s32 e1000_read_mac_addr_vf(struct e1000_hw *hw) +{ + memcpy(hw->mac.addr, hw->mac.perm_addr, ETH_ALEN); + + return E1000_SUCCESS; +} + +/** + * e1000_set_uc_addr_vf - Set or clear unicast filters + * @hw: pointer to the HW structure + * @sub_cmd: add or clear filters + * @addr: pointer to the filter MAC address + **/ +static s32 e1000_set_uc_addr_vf(struct e1000_hw *hw, u32 sub_cmd, u8 *addr) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[3], msgbuf_chk; + u8 *msg_addr = (u8 *)(&msgbuf[1]); + s32 ret_val; + + memset(msgbuf, 0, sizeof(msgbuf)); + msgbuf[0] |= sub_cmd; + msgbuf[0] |= E1000_VF_SET_MAC_ADDR; + msgbuf_chk = msgbuf[0]; + + if (addr) + memcpy(msg_addr, addr, ETH_ALEN); + + ret_val = mbx->ops.write_posted(hw, msgbuf, 3); + + if (!ret_val) + ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + + msgbuf[0] &= ~E1000_VT_MSGTYPE_CTS; + + if (!ret_val) { + msgbuf[0] &= ~E1000_VT_MSGTYPE_CTS; + + if (msgbuf[0] == (msgbuf_chk | E1000_VT_MSGTYPE_NACK)) + return -ENOSPC; + } + + return ret_val; +} + +/** + * e1000_check_for_link_vf - Check for link for a virtual interface + * @hw: pointer to the HW structure + * + * Checks to see if the underlying PF is still talking to the VF and + * if it is then it reports the link state to the hardware, otherwise + * it reports link down and returns an error. + **/ +static s32 e1000_check_for_link_vf(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + struct e1000_mac_info *mac = &hw->mac; + s32 ret_val = E1000_SUCCESS; + u32 in_msg = 0; + + /* We only want to run this if there has been a rst asserted. + * in this case that could mean a link change, device reset, + * or a virtual function reset + */ + + /* If we were hit with a reset or timeout drop the link */ + if (!mbx->ops.check_for_rst(hw) || !mbx->timeout) + mac->get_link_status = true; + + if (!mac->get_link_status) + goto out; + + /* if link status is down no point in checking to see if PF is up */ + if (!(er32(STATUS) & E1000_STATUS_LU)) + goto out; + + /* if the read failed it could just be a mailbox collision, best wait + * until we are called again and don't report an error + */ + if (mbx->ops.read(hw, &in_msg, 1)) + goto out; + + /* if incoming message isn't clear to send we are waiting on response */ + if (!(in_msg & E1000_VT_MSGTYPE_CTS)) { + /* msg is not CTS and is NACK we must have lost CTS status */ + if (in_msg & E1000_VT_MSGTYPE_NACK) + ret_val = -E1000_ERR_MAC_INIT; + goto out; + } + + /* the PF is talking, if we timed out in the past we reinit */ + if (!mbx->timeout) { + ret_val = -E1000_ERR_MAC_INIT; + goto out; + } + + /* if we passed all the tests above then the link is up and we no + * longer need to check for link + */ + mac->get_link_status = false; + +out: + return ret_val; +} + diff --git a/drivers/net/ethernet/intel/igbvf/vf.h b/drivers/net/ethernet/intel/igbvf/vf.h new file mode 100644 index 000000000..ba9bb3132 --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/vf.h @@ -0,0 +1,243 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2009 - 2018 Intel Corporation. */ + +#ifndef _E1000_VF_H_ +#define _E1000_VF_H_ + +#include <linux/pci.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/if_ether.h> + +#include "regs.h" +#include "defines.h" + +struct e1000_hw; + +#define E1000_DEV_ID_82576_VF 0x10CA +#define E1000_DEV_ID_I350_VF 0x1520 +#define E1000_REVISION_0 0 +#define E1000_REVISION_1 1 +#define E1000_REVISION_2 2 +#define E1000_REVISION_3 3 +#define E1000_REVISION_4 4 + +#define E1000_FUNC_0 0 +#define E1000_FUNC_1 1 + +/* Receive Address Register Count + * Number of high/low register pairs in the RAR. The RAR (Receive Address + * Registers) holds the directed and multicast addresses that we monitor. + * These entries are also used for MAC-based filtering. + */ +#define E1000_RAR_ENTRIES_VF 1 + +/* Receive Descriptor - Advanced */ +union e1000_adv_rx_desc { + struct { + __le64 pkt_addr; /* Packet buffer address */ + __le64 hdr_addr; /* Header buffer address */ + } read; + struct { + struct { + union { + __le32 data; + struct { + __le16 pkt_info; /* RSS/Packet type */ + /* Split Header, hdr buffer length */ + __le16 hdr_info; + } hs_rss; + } lo_dword; + union { + __le32 rss; /* RSS Hash */ + struct { + __le16 ip_id; /* IP id */ + __le16 csum; /* Packet Checksum */ + } csum_ip; + } hi_dword; + } lower; + struct { + __le32 status_error; /* ext status/error */ + __le16 length; /* Packet length */ + __le16 vlan; /* VLAN tag */ + } upper; + } wb; /* writeback */ +}; + +#define E1000_RXDADV_HDRBUFLEN_MASK 0x7FE0 +#define E1000_RXDADV_HDRBUFLEN_SHIFT 5 + +/* Transmit Descriptor - Advanced */ +union e1000_adv_tx_desc { + struct { + __le64 buffer_addr; /* Address of descriptor's data buf */ + __le32 cmd_type_len; + __le32 olinfo_status; + } read; + struct { + __le64 rsvd; /* Reserved */ + __le32 nxtseq_seed; + __le32 status; + } wb; +}; + +/* Adv Transmit Descriptor Config Masks */ +#define E1000_ADVTXD_DTYP_CTXT 0x00200000 /* Advanced Context Descriptor */ +#define E1000_ADVTXD_DTYP_DATA 0x00300000 /* Advanced Data Descriptor */ +#define E1000_ADVTXD_DCMD_EOP 0x01000000 /* End of Packet */ +#define E1000_ADVTXD_DCMD_IFCS 0x02000000 /* Insert FCS (Ethernet CRC) */ +#define E1000_ADVTXD_DCMD_RS 0x08000000 /* Report Status */ +#define E1000_ADVTXD_DCMD_DEXT 0x20000000 /* Descriptor extension (1=Adv) */ +#define E1000_ADVTXD_DCMD_VLE 0x40000000 /* VLAN pkt enable */ +#define E1000_ADVTXD_DCMD_TSE 0x80000000 /* TCP Seg enable */ +#define E1000_ADVTXD_PAYLEN_SHIFT 14 /* Adv desc PAYLEN shift */ + +/* Context descriptors */ +struct e1000_adv_tx_context_desc { + __le32 vlan_macip_lens; + __le32 seqnum_seed; + __le32 type_tucmd_mlhl; + __le32 mss_l4len_idx; +}; + +#define E1000_ADVTXD_MACLEN_SHIFT 9 /* Adv ctxt desc mac len shift */ +#define E1000_ADVTXD_TUCMD_IPV4 0x00000400 /* IP Packet Type: 1=IPv4 */ +#define E1000_ADVTXD_TUCMD_L4T_TCP 0x00000800 /* L4 Packet TYPE of TCP */ +#define E1000_ADVTXD_TUCMD_L4T_SCTP 0x00001000 /* L4 packet TYPE of SCTP */ +#define E1000_ADVTXD_L4LEN_SHIFT 8 /* Adv ctxt L4LEN shift */ +#define E1000_ADVTXD_MSS_SHIFT 16 /* Adv ctxt MSS shift */ + +enum e1000_mac_type { + e1000_undefined = 0, + e1000_vfadapt, + e1000_vfadapt_i350, + e1000_num_macs /* List is 1-based, so subtract 1 for true count. */ +}; + +struct e1000_vf_stats { + u64 base_gprc; + u64 base_gptc; + u64 base_gorc; + u64 base_gotc; + u64 base_mprc; + u64 base_gotlbc; + u64 base_gptlbc; + u64 base_gorlbc; + u64 base_gprlbc; + + u32 last_gprc; + u32 last_gptc; + u32 last_gorc; + u32 last_gotc; + u32 last_mprc; + u32 last_gotlbc; + u32 last_gptlbc; + u32 last_gorlbc; + u32 last_gprlbc; + + u64 gprc; + u64 gptc; + u64 gorc; + u64 gotc; + u64 mprc; + u64 gotlbc; + u64 gptlbc; + u64 gorlbc; + u64 gprlbc; +}; + +#include "mbx.h" + +struct e1000_mac_operations { + /* Function pointers for the MAC. */ + s32 (*init_params)(struct e1000_hw *); + s32 (*check_for_link)(struct e1000_hw *); + void (*clear_vfta)(struct e1000_hw *); + s32 (*get_bus_info)(struct e1000_hw *); + s32 (*get_link_up_info)(struct e1000_hw *, u16 *, u16 *); + void (*update_mc_addr_list)(struct e1000_hw *, u8 *, u32, u32, u32); + s32 (*set_uc_addr)(struct e1000_hw *, u32, u8 *); + s32 (*reset_hw)(struct e1000_hw *); + s32 (*init_hw)(struct e1000_hw *); + s32 (*setup_link)(struct e1000_hw *); + void (*write_vfta)(struct e1000_hw *, u32, u32); + void (*mta_set)(struct e1000_hw *, u32); + void (*rar_set)(struct e1000_hw *, u8*, u32); + s32 (*read_mac_addr)(struct e1000_hw *); + s32 (*set_vfta)(struct e1000_hw *, u16, bool); +}; + +struct e1000_mac_info { + struct e1000_mac_operations ops; + u8 addr[6]; + u8 perm_addr[6]; + + enum e1000_mac_type type; + + u16 mta_reg_count; + u16 rar_entry_count; + + bool get_link_status; +}; + +struct e1000_mbx_operations { + s32 (*init_params)(struct e1000_hw *hw); + s32 (*read)(struct e1000_hw *, u32 *, u16); + s32 (*write)(struct e1000_hw *, u32 *, u16); + s32 (*read_posted)(struct e1000_hw *, u32 *, u16); + s32 (*write_posted)(struct e1000_hw *, u32 *, u16); + s32 (*check_for_msg)(struct e1000_hw *); + s32 (*check_for_ack)(struct e1000_hw *); + s32 (*check_for_rst)(struct e1000_hw *); +}; + +struct e1000_mbx_stats { + u32 msgs_tx; + u32 msgs_rx; + + u32 acks; + u32 reqs; + u32 rsts; +}; + +struct e1000_mbx_info { + struct e1000_mbx_operations ops; + struct e1000_mbx_stats stats; + u32 timeout; + u32 usec_delay; + u16 size; +}; + +struct e1000_dev_spec_vf { + u32 vf_number; + u32 v2p_mailbox; +}; + +struct e1000_hw { + void *back; + + u8 __iomem *hw_addr; + u8 __iomem *flash_address; + unsigned long io_base; + + struct e1000_mac_info mac; + struct e1000_mbx_info mbx; + spinlock_t mbx_lock; /* serializes mailbox ops */ + + union { + struct e1000_dev_spec_vf vf; + } dev_spec; + + u16 device_id; + u16 subsystem_vendor_id; + u16 subsystem_device_id; + u16 vendor_id; + + u8 revision_id; +}; + +/* These functions must be implemented by drivers */ +void e1000_rlpml_set_vf(struct e1000_hw *, u16); +void e1000_init_function_pointers_vf(struct e1000_hw *hw); + +#endif /* _E1000_VF_H_ */ |