diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index e0758ddd2e22..e99be9f696c3 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -111,6 +111,7 @@ #define I40E_OEM_VER_PATCH_MASK 0xff #define I40E_OEM_VER_BUILD_SHIFT 8 #define I40E_OEM_VER_SHIFT 24 +#define I40E_PHY_DEBUG_PORT BIT(4) /* The values in here are decimal coded as hex as is the case in the NVM map*/ #define I40E_CURRENT_NVM_VERSION_HI 0x2 @@ -355,6 +356,7 @@ struct i40e_pf { #define I40E_FLAG_NO_DCB_SUPPORT BIT_ULL(45) #define I40E_FLAG_USE_SET_LLDP_MIB BIT_ULL(46) #define I40E_FLAG_STOP_FW_LLDP BIT_ULL(47) +#define I40E_FLAG_HAVE_10GBASET_PHY BIT_ULL(48) #define I40E_FLAG_PF_MAC BIT_ULL(50) /* tracks features that get auto disabled by errors */ @@ -440,6 +442,7 @@ struct i40e_pf { u32 ioremap_len; u32 fd_inv; + u16 phy_led_val; }; struct i40e_mac_filter { diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h index bb7ecbbdf948..8d5c65ab6267 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h @@ -146,6 +146,8 @@ enum i40e_admin_queue_opc { i40e_aqc_opc_set_port_parameters = 0x0203, i40e_aqc_opc_get_switch_resource_alloc = 0x0204, i40e_aqc_opc_set_switch_config = 0x0205, + i40e_aqc_opc_rx_ctl_reg_read = 0x0206, + i40e_aqc_opc_rx_ctl_reg_write = 0x0207, i40e_aqc_opc_add_vsi = 0x0210, i40e_aqc_opc_update_vsi_parameters = 0x0211, @@ -696,6 +698,20 @@ struct i40e_aqc_set_switch_config { I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config); +/* Read Receive control registers (direct 0x0206) + * Write Receive control registers (direct 0x0207) + * used for accessing Rx control registers that can be + * slow and need special handling when under high Rx load + */ +struct i40e_aqc_rx_ctl_reg_read_write { + __le32 reserved1; + __le32 address; + __le32 reserved2; + __le32 value; +}; + +I40E_CHECK_CMD_LENGTH(i40e_aqc_rx_ctl_reg_read_write); + /* Add VSI (indirect 0x0210) * this indirect command uses struct i40e_aqc_vsi_properties_data * as the indirect buffer (128 bytes) diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index f9239330f49d..4596294c2ab1 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -1239,7 +1239,13 @@ i40e_status i40e_pf_reset(struct i40e_hw *hw) grst_del = (rd32(hw, I40E_GLGEN_RSTCTL) & I40E_GLGEN_RSTCTL_GRSTDEL_MASK) >> I40E_GLGEN_RSTCTL_GRSTDEL_SHIFT; - for (cnt = 0; cnt < grst_del + 10; cnt++) { + + /* It can take upto 15 secs for GRST steady state. + * Bump it to 16 secs max to be safe. + */ + grst_del = grst_del * 20; + + for (cnt = 0; cnt < grst_del; cnt++) { reg = rd32(hw, I40E_GLGEN_RSTAT); if (!(reg & I40E_GLGEN_RSTAT_DEVSTATE_MASK)) break; @@ -1887,6 +1893,32 @@ i40e_status i40e_aq_set_phy_int_mask(struct i40e_hw *hw, return status; } +/** + * i40e_aq_set_phy_debug + * @hw: pointer to the hw struct + * @cmd_flags: debug command flags + * @cmd_details: pointer to command details structure or NULL + * + * Reset the external PHY. + **/ +enum i40e_status_code i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_set_phy_debug *cmd = + (struct i40e_aqc_set_phy_debug *)&desc.params.raw; + enum i40e_status_code status; + + i40e_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_set_phy_debug); + + cmd->command_flags = cmd_flags; + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + /** * i40e_aq_add_vsi * @hw: pointer to the hw struct @@ -3850,7 +3882,7 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw, return ret; /* Read the PF Queue Filter control register */ - val = rd32(hw, I40E_PFQF_CTL_0); + val = i40e_read_rx_ctl(hw, I40E_PFQF_CTL_0); /* Program required PE hash buckets for the PF */ val &= ~I40E_PFQF_CTL_0_PEHSIZE_MASK; @@ -3887,7 +3919,7 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw, if (settings->enable_macvlan) val |= I40E_PFQF_CTL_0_MACVLAN_ENA_MASK; - wr32(hw, I40E_PFQF_CTL_0, val); + i40e_write_rx_ctl(hw, I40E_PFQF_CTL_0, val); return 0; } @@ -4214,3 +4246,454 @@ i40e_status i40e_aq_configure_partition_bw(struct i40e_hw *hw, return status; } + +/** + * i40e_read_phy_register + * @hw: pointer to the HW structure + * @page: registers page number + * @reg: register address in the page + * @phy_adr: PHY address on MDIO interface + * @value: PHY register value + * + * Reads specified PHY register value + **/ +i40e_status i40e_read_phy_register(struct i40e_hw *hw, + u8 page, u16 reg, u8 phy_addr, + u16 *value) +{ + i40e_status status = I40E_ERR_TIMEOUT; + u32 command = 0; + u16 retry = 1000; + u8 port_num = hw->func_caps.mdio_port_num; + + command = (reg << I40E_GLGEN_MSCA_MDIADD_SHIFT) | + (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) | + (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) | + (I40E_MDIO_OPCODE_ADDRESS) | + (I40E_MDIO_STCODE) | + (I40E_GLGEN_MSCA_MDICMD_MASK) | + (I40E_GLGEN_MSCA_MDIINPROGEN_MASK); + wr32(hw, I40E_GLGEN_MSCA(port_num), command); + do { + command = rd32(hw, I40E_GLGEN_MSCA(port_num)); + if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) { + status = 0; + break; + } + usleep_range(10, 20); + retry--; + } while (retry); + + if (status) { + i40e_debug(hw, I40E_DEBUG_PHY, + "PHY: Can't write command to external PHY.\n"); + goto phy_read_end; + } + + command = (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) | + (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) | + (I40E_MDIO_OPCODE_READ) | + (I40E_MDIO_STCODE) | + (I40E_GLGEN_MSCA_MDICMD_MASK) | + (I40E_GLGEN_MSCA_MDIINPROGEN_MASK); + status = I40E_ERR_TIMEOUT; + retry = 1000; + wr32(hw, I40E_GLGEN_MSCA(port_num), command); + do { + command = rd32(hw, I40E_GLGEN_MSCA(port_num)); + if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) { + status = 0; + break; + } + usleep_range(10, 20); + retry--; + } while (retry); + + if (!status) { + command = rd32(hw, I40E_GLGEN_MSRWD(port_num)); + *value = (command & I40E_GLGEN_MSRWD_MDIRDDATA_MASK) >> + I40E_GLGEN_MSRWD_MDIRDDATA_SHIFT; + } else { + i40e_debug(hw, I40E_DEBUG_PHY, + "PHY: Can't read register value from external PHY.\n"); + } + +phy_read_end: + return status; +} + +/** + * i40e_write_phy_register + * @hw: pointer to the HW structure + * @page: registers page number + * @reg: register address in the page + * @phy_adr: PHY address on MDIO interface + * @value: PHY register value + * + * Writes value to specified PHY register + **/ +i40e_status i40e_write_phy_register(struct i40e_hw *hw, + u8 page, u16 reg, u8 phy_addr, + u16 value) +{ + i40e_status status = I40E_ERR_TIMEOUT; + u32 command = 0; + u16 retry = 1000; + u8 port_num = hw->func_caps.mdio_port_num; + + command = (reg << I40E_GLGEN_MSCA_MDIADD_SHIFT) | + (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) | + (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) | + (I40E_MDIO_OPCODE_ADDRESS) | + (I40E_MDIO_STCODE) | + (I40E_GLGEN_MSCA_MDICMD_MASK) | + (I40E_GLGEN_MSCA_MDIINPROGEN_MASK); + wr32(hw, I40E_GLGEN_MSCA(port_num), command); + do { + command = rd32(hw, I40E_GLGEN_MSCA(port_num)); + if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) { + status = 0; + break; + } + usleep_range(10, 20); + retry--; + } while (retry); + if (status) { + i40e_debug(hw, I40E_DEBUG_PHY, + "PHY: Can't write command to external PHY.\n"); + goto phy_write_end; + } + + command = value << I40E_GLGEN_MSRWD_MDIWRDATA_SHIFT; + wr32(hw, I40E_GLGEN_MSRWD(port_num), command); + + command = (page << I40E_GLGEN_MSCA_DEVADD_SHIFT) | + (phy_addr << I40E_GLGEN_MSCA_PHYADD_SHIFT) | + (I40E_MDIO_OPCODE_WRITE) | + (I40E_MDIO_STCODE) | + (I40E_GLGEN_MSCA_MDICMD_MASK) | + (I40E_GLGEN_MSCA_MDIINPROGEN_MASK); + status = I40E_ERR_TIMEOUT; + retry = 1000; + wr32(hw, I40E_GLGEN_MSCA(port_num), command); + do { + command = rd32(hw, I40E_GLGEN_MSCA(port_num)); + if (!(command & I40E_GLGEN_MSCA_MDICMD_MASK)) { + status = 0; + break; + } + usleep_range(10, 20); + retry--; + } while (retry); + +phy_write_end: + return status; +} + +/** + * i40e_get_phy_address + * @hw: pointer to the HW structure + * @dev_num: PHY port num that address we want + * @phy_addr: Returned PHY address + * + * Gets PHY address for current port + **/ +u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num) +{ + u8 port_num = hw->func_caps.mdio_port_num; + u32 reg_val = rd32(hw, I40E_GLGEN_MDIO_I2C_SEL(port_num)); + + return (u8)(reg_val >> ((dev_num + 1) * 5)) & 0x1f; +} + +/** + * i40e_blink_phy_led + * @hw: pointer to the HW structure + * @time: time how long led will blinks in secs + * @interval: gap between LED on and off in msecs + * + * Blinks PHY link LED + **/ +i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw, + u32 time, u32 interval) +{ + i40e_status status = 0; + u32 i; + u16 led_ctl; + u16 gpio_led_port; + u16 led_reg; + u16 led_addr = I40E_PHY_LED_PROV_REG_1; + u8 phy_addr = 0; + u8 port_num; + + i = rd32(hw, I40E_PFGEN_PORTNUM); + port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK); + phy_addr = i40e_get_phy_address(hw, port_num); + + for (gpio_led_port = 0; gpio_led_port < 3; gpio_led_port++, + led_addr++) { + status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, &led_reg); + if (status) + goto phy_blinking_end; + led_ctl = led_reg; + if (led_reg & I40E_PHY_LED_LINK_MODE_MASK) { + led_reg = 0; + status = i40e_write_phy_register(hw, + I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, + led_reg); + if (status) + goto phy_blinking_end; + break; + } + } + + if (time > 0 && interval > 0) { + for (i = 0; i < time * 1000; i += interval) { + status = i40e_read_phy_register(hw, + I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, + &led_reg); + if (status) + goto restore_config; + if (led_reg & I40E_PHY_LED_MANUAL_ON) + led_reg = 0; + else + led_reg = I40E_PHY_LED_MANUAL_ON; + status = i40e_write_phy_register(hw, + I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, + led_reg); + if (status) + goto restore_config; + msleep(interval); + } + } + +restore_config: + status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE, led_addr, + phy_addr, led_ctl); + +phy_blinking_end: + return status; +} + +/** + * i40e_led_get_phy - return current on/off mode + * @hw: pointer to the hw struct + * @led_addr: address of led register to use + * @val: original value of register to use + * + **/ +i40e_status i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr, + u16 *val) +{ + i40e_status status = 0; + u16 gpio_led_port; + u8 phy_addr = 0; + u16 reg_val; + u16 temp_addr; + u8 port_num; + u32 i; + + temp_addr = I40E_PHY_LED_PROV_REG_1; + i = rd32(hw, I40E_PFGEN_PORTNUM); + port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK); + phy_addr = i40e_get_phy_address(hw, port_num); + + for (gpio_led_port = 0; gpio_led_port < 3; gpio_led_port++, + temp_addr++) { + status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE, + temp_addr, phy_addr, ®_val); + if (status) + return status; + *val = reg_val; + if (reg_val & I40E_PHY_LED_LINK_MODE_MASK) { + *led_addr = temp_addr; + break; + } + } + return status; +} + +/** + * i40e_led_set_phy + * @hw: pointer to the HW structure + * @on: true or false + * @mode: original val plus bit for set or ignore + * Set led's on or off when controlled by the PHY + * + **/ +i40e_status i40e_led_set_phy(struct i40e_hw *hw, bool on, + u16 led_addr, u32 mode) +{ + i40e_status status = 0; + u16 led_ctl = 0; + u16 led_reg = 0; + u8 phy_addr = 0; + u8 port_num; + u32 i; + + i = rd32(hw, I40E_PFGEN_PORTNUM); + port_num = (u8)(i & I40E_PFGEN_PORTNUM_PORT_NUM_MASK); + phy_addr = i40e_get_phy_address(hw, port_num); + + status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE, led_addr, + phy_addr, &led_reg); + if (status) + return status; + led_ctl = led_reg; + if (led_reg & I40E_PHY_LED_LINK_MODE_MASK) { + led_reg = 0; + status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, led_reg); + if (status) + return status; + } + status = i40e_read_phy_register(hw, I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, &led_reg); + if (status) + goto restore_config; + if (on) + led_reg = I40E_PHY_LED_MANUAL_ON; + else + led_reg = 0; + status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, led_reg); + if (status) + goto restore_config; + if (mode & I40E_PHY_LED_MODE_ORIG) { + led_ctl = (mode & I40E_PHY_LED_MODE_MASK); + status = i40e_write_phy_register(hw, + I40E_PHY_COM_REG_PAGE, + led_addr, phy_addr, led_ctl); + } + return status; +restore_config: + status = i40e_write_phy_register(hw, I40E_PHY_COM_REG_PAGE, led_addr, + phy_addr, led_ctl); + return status; +} + +/** + * i40e_aq_rx_ctl_read_register - use FW to read from an Rx control register + * @hw: pointer to the hw struct + * @reg_addr: register address + * @reg_val: ptr to register value + * @cmd_details: pointer to command details structure or NULL + * + * Use the firmware to read the Rx control register, + * especially useful if the Rx unit is under heavy pressure + **/ +i40e_status i40e_aq_rx_ctl_read_register(struct i40e_hw *hw, + u32 reg_addr, u32 *reg_val, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_rx_ctl_reg_read_write *cmd_resp = + (struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw; + i40e_status status; + + if (!reg_val) + return I40E_ERR_PARAM; + + i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_rx_ctl_reg_read); + + cmd_resp->address = cpu_to_le32(reg_addr); + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + if (status == 0) + *reg_val = le32_to_cpu(cmd_resp->value); + + return status; +} + +/** + * i40e_read_rx_ctl - read from an Rx control register + * @hw: pointer to the hw struct + * @reg_addr: register address + **/ +u32 i40e_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr) +{ + i40e_status status = 0; + bool use_register; + int retry = 5; + u32 val = 0; + + use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5); + if (!use_register) { +do_retry: + status = i40e_aq_rx_ctl_read_register(hw, reg_addr, &val, NULL); + if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) { + usleep_range(1000, 2000); + retry--; + goto do_retry; + } + } + + /* if the AQ access failed, try the old-fashioned way */ + if (status || use_register) + val = rd32(hw, reg_addr); + + return val; +} + +/** + * i40e_aq_rx_ctl_write_register + * @hw: pointer to the hw struct + * @reg_addr: register address + * @reg_val: register value + * @cmd_details: pointer to command details structure or NULL + * + * Use the firmware to write to an Rx control register, + * especially useful if the Rx unit is under heavy pressure + **/ +i40e_status i40e_aq_rx_ctl_write_register(struct i40e_hw *hw, + u32 reg_addr, u32 reg_val, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_rx_ctl_reg_read_write *cmd = + (struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw; + i40e_status status; + + i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_rx_ctl_reg_write); + + cmd->address = cpu_to_le32(reg_addr); + cmd->value = cpu_to_le32(reg_val); + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + +/** + * i40e_write_rx_ctl - write to an Rx control register + * @hw: pointer to the hw struct + * @reg_addr: register address + * @reg_val: register value + **/ +void i40e_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val) +{ + i40e_status status = 0; + bool use_register; + int retry = 5; + + use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5); + if (!use_register) { +do_retry: + status = i40e_aq_rx_ctl_write_register(hw, reg_addr, + reg_val, NULL); + if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) { + usleep_range(1000, 2000); + retry--; + goto do_retry; + } + } + + /* if the AQ access failed, try the old-fashioned way */ + if (status || use_register) + wr32(hw, reg_addr, reg_val); +} diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 9fc7546bfc9b..a85bc94cdf87 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -1826,28 +1826,52 @@ static int i40e_set_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state) { struct i40e_netdev_priv *np = netdev_priv(netdev); + i40e_status ret = 0; struct i40e_pf *pf = np->vsi->back; struct i40e_hw *hw = &pf->hw; int blink_freq = 2; + u16 temp_status; switch (state) { case ETHTOOL_ID_ACTIVE: - pf->led_status = i40e_led_get(hw); + if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY)) { + pf->led_status = i40e_led_get(hw); + } else { + i40e_aq_set_phy_debug(hw, I40E_PHY_DEBUG_PORT, NULL); + ret = i40e_led_get_phy(hw, &temp_status, + &pf->phy_led_val); + pf->led_status = temp_status; + } return blink_freq; case ETHTOOL_ID_ON: - i40e_led_set(hw, 0xF, false); + if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY)) + i40e_led_set(hw, 0xf, false); + else + ret = i40e_led_set_phy(hw, true, pf->led_status, 0); break; case ETHTOOL_ID_OFF: - i40e_led_set(hw, 0x0, false); + if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY)) + i40e_led_set(hw, 0x0, false); + else + ret = i40e_led_set_phy(hw, false, pf->led_status, 0); break; case ETHTOOL_ID_INACTIVE: - i40e_led_set(hw, pf->led_status, false); + if (!(pf->flags & I40E_FLAG_HAVE_10GBASET_PHY)) { + i40e_led_set(hw, false, pf->led_status); + } else { + ret = i40e_led_set_phy(hw, false, pf->led_status, + (pf->phy_led_val | + I40E_PHY_LED_MODE_ORIG)); + i40e_aq_set_phy_debug(hw, 0, NULL); + } break; default: break; } - - return 0; + if (ret) + return -ENOENT; + else + return 0; } /* NOTE: i40e hardware uses a conversion factor of 2 for Interrupt @@ -2157,8 +2181,8 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc) { struct i40e_hw *hw = &pf->hw; - u64 hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) | - ((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32); + u64 hena = (u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0)) | + ((u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)) << 32); /* RSS does not support anything other than hashing * to queues on src and dst IPs and ports @@ -2267,8 +2291,8 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc) return -EINVAL; } - wr32(hw, I40E_PFQF_HENA(0), (u32)hena); - wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32)); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (u32)hena); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32)); i40e_flush(hw); /* Save setting for future output/update */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c index 7c66ce416ec7..8ad162c16f61 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_fcoe.c +++ b/drivers/net/ethernet/intel/i40e/i40e_fcoe.c @@ -295,11 +295,11 @@ void i40e_init_pf_fcoe(struct i40e_pf *pf) } /* enable FCoE hash filter */ - val = rd32(hw, I40E_PFQF_HENA(1)); + val = i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)); val |= BIT(I40E_FILTER_PCTYPE_FCOE_OX - 32); val |= BIT(I40E_FILTER_PCTYPE_FCOE_RX - 32); val &= I40E_PFQF_HENA_PTYPE_ENA_MASK; - wr32(hw, I40E_PFQF_HENA(1), val); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), val); /* enable flag */ pf->flags |= I40E_FLAG_FCOE_ENABLED; @@ -317,11 +317,11 @@ void i40e_init_pf_fcoe(struct i40e_pf *pf) pf->filter_settings.fcoe_cntx_num = I40E_DMA_CNTX_SIZE_4K; /* Setup max frame with FCoE_MTU plus L2 overheads */ - val = rd32(hw, I40E_GLFCOE_RCTL); + val = i40e_read_rx_ctl(hw, I40E_GLFCOE_RCTL); val &= ~I40E_GLFCOE_RCTL_MAX_SIZE_MASK; val |= ((FCOE_MTU + ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN) << I40E_GLFCOE_RCTL_MAX_SIZE_SHIFT); - wr32(hw, I40E_GLFCOE_RCTL, val); + i40e_write_rx_ctl(hw, I40E_GLFCOE_RCTL, val); dev_info(&pf->pdev->dev, "FCoE is supported.\n"); } @@ -1359,16 +1359,32 @@ static netdev_tx_t i40e_fcoe_xmit_frame(struct sk_buff *skb, struct i40e_ring *tx_ring = vsi->tx_rings[skb->queue_mapping]; struct i40e_tx_buffer *first; u32 tx_flags = 0; + int fso, count; u8 hdr_len = 0; u8 sof = 0; u8 eof = 0; - int fso; if (i40e_fcoe_set_skb_header(skb)) goto out_drop; - if (!i40e_xmit_descriptor_count(skb, tx_ring)) + count = i40e_xmit_descriptor_count(skb); + if (i40e_chk_linearize(skb, count)) { + if (__skb_linearize(skb)) + goto out_drop; + count = TXD_USE_COUNT(skb->len); + tx_ring->tx_stats.tx_linearize++; + } + + /* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD, + * + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD, + * + 4 desc gap to avoid the cache line where head is, + * + 1 desc for context descriptor, + * otherwise try next time + */ + if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) { + tx_ring->tx_stats.tx_busy++; return NETDEV_TX_BUSY; + } /* prepare the xmit flags */ if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags)) diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 2f2b2d714f63..3c8d8c4491f8 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -46,7 +46,7 @@ static const char i40e_driver_string[] = #define DRV_VERSION_MAJOR 1 #define DRV_VERSION_MINOR 4 -#define DRV_VERSION_BUILD 15 +#define DRV_VERSION_BUILD 25 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) DRV_KERN @@ -3929,6 +3929,9 @@ static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable) else rx_reg &= ~I40E_QRX_ENA_QENA_REQ_MASK; wr32(hw, I40E_QRX_ENA(pf_q), rx_reg); + /* No waiting for the Tx queue to disable */ + if (!enable && test_bit(__I40E_PORT_TX_SUSPENDED, &pf->state)) + continue; /* wait for the change to finish */ ret = i40e_pf_rxq_wait(pf, pf_q, enable); @@ -4287,12 +4290,12 @@ static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf) #ifdef CONFIG_I40E_DCB /** - * i40e_vsi_wait_txq_disabled - Wait for VSI's queues to be disabled + * i40e_vsi_wait_queues_disabled - Wait for VSI's queues to be disabled * @vsi: the VSI being configured * - * This function waits for the given VSI's Tx queues to be disabled. + * This function waits for the given VSI's queues to be disabled. **/ -static int i40e_vsi_wait_txq_disabled(struct i40e_vsi *vsi) +static int i40e_vsi_wait_queues_disabled(struct i40e_vsi *vsi) { struct i40e_pf *pf = vsi->back; int i, pf_q, ret; @@ -4309,24 +4312,36 @@ static int i40e_vsi_wait_txq_disabled(struct i40e_vsi *vsi) } } + pf_q = vsi->base_queue; + for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) { + /* Check and wait for the disable status of the queue */ + ret = i40e_pf_rxq_wait(pf, pf_q, false); + if (ret) { + dev_info(&pf->pdev->dev, + "VSI seid %d Rx ring %d disable timeout\n", + vsi->seid, pf_q); + return ret; + } + } + return 0; } /** - * i40e_pf_wait_txq_disabled - Wait for all queues of PF VSIs to be disabled + * i40e_pf_wait_queues_disabled - Wait for all queues of PF VSIs to be disabled * @pf: the PF * - * This function waits for the Tx queues to be in disabled state for all the + * This function waits for the queues to be in disabled state for all the * VSIs that are managed by this PF. **/ -static int i40e_pf_wait_txq_disabled(struct i40e_pf *pf) +static int i40e_pf_wait_queues_disabled(struct i40e_pf *pf) { int v, ret = 0; for (v = 0; v < pf->hw.func_caps.num_vsis; v++) { /* No need to wait for FCoE VSI queues */ if (pf->vsi[v] && pf->vsi[v]->type != I40E_VSI_FCOE) { - ret = i40e_vsi_wait_txq_disabled(pf->vsi[v]); + ret = i40e_vsi_wait_queues_disabled(pf->vsi[v]); if (ret) break; } @@ -5726,8 +5741,8 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf, if (ret) goto exit; - /* Wait for the PF's Tx queues to be disabled */ - ret = i40e_pf_wait_txq_disabled(pf); + /* Wait for the PF's queues to be disabled */ + ret = i40e_pf_wait_queues_disabled(pf); if (ret) { /* Schedule PF reset to recover */ set_bit(__I40E_PF_RESET_REQUESTED, &pf->state); @@ -7094,12 +7109,13 @@ static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf) ret = i40e_aq_del_udp_tunnel(hw, i, NULL); if (ret) { - dev_info(&pf->pdev->dev, - "%s vxlan port %d, index %d failed, err %s aq_err %s\n", - port ? "add" : "delete", - ntohs(port), i, - i40e_stat_str(&pf->hw, ret), - i40e_aq_str(&pf->hw, + dev_dbg(&pf->pdev->dev, + "%s %s port %d, index %d failed, err %s aq_err %s\n", + pf->udp_ports[i].type ? "vxlan" : "geneve", + port ? "add" : "delete", + ntohs(port), i, + i40e_stat_str(&pf->hw, ret), + i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); pf->udp_ports[i].index = 0; } @@ -8016,7 +8032,7 @@ static int i40e_config_rss_reg(struct i40e_vsi *vsi, const u8 *seed, u32 *seed_dw = (u32 *)seed; for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++) - wr32(hw, I40E_PFQF_HKEY(i), seed_dw[i]); + i40e_write_rx_ctl(hw, I40E_PFQF_HKEY(i), seed_dw[i]); } if (lut) { @@ -8053,7 +8069,7 @@ static int i40e_get_rss_reg(struct i40e_vsi *vsi, u8 *seed, u32 *seed_dw = (u32 *)seed; for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++) - seed_dw[i] = rd32(hw, I40E_PFQF_HKEY(i)); + seed_dw[i] = i40e_read_rx_ctl(hw, I40E_PFQF_HKEY(i)); } if (lut) { u32 *lut_dw = (u32 *)lut; @@ -8136,19 +8152,19 @@ static int i40e_pf_config_rss(struct i40e_pf *pf) int ret; /* By default we enable TCP/UDP with IPv4/IPv6 ptypes */ - hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) | - ((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32); + hena = (u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0)) | + ((u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)) << 32); hena |= i40e_pf_get_default_rss_hena(pf); - wr32(hw, I40E_PFQF_HENA(0), (u32)hena); - wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32)); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (u32)hena); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32)); /* Determine the RSS table size based on the hardware capabilities */ - reg_val = rd32(hw, I40E_PFQF_CTL_0); + reg_val = i40e_read_rx_ctl(hw, I40E_PFQF_CTL_0); reg_val = (pf->rss_table_size == 512) ? (reg_val | I40E_PFQF_CTL_0_HASHLUTSIZE_512) : (reg_val & ~I40E_PFQF_CTL_0_HASHLUTSIZE_512); - wr32(hw, I40E_PFQF_CTL_0, reg_val); + i40e_write_rx_ctl(hw, I40E_PFQF_CTL_0, reg_val); /* Determine the RSS size of the VSI */ if (!vsi->rss_size) @@ -9567,10 +9583,15 @@ vector_setup_out: **/ static struct i40e_vsi *i40e_vsi_reinit_setup(struct i40e_vsi *vsi) { - struct i40e_pf *pf = vsi->back; + struct i40e_pf *pf; u8 enabled_tc; int ret; + if (!vsi) + return NULL; + + pf = vsi->back; + i40e_put_lump(pf->qp_pile, vsi->base_queue, vsi->idx); i40e_vsi_clear_rings(vsi); @@ -11130,6 +11151,10 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) i40e_add_filter_to_drop_tx_flow_control_frames(&pf->hw, pf->main_vsi_seid); + if ((pf->hw.device_id == I40E_DEV_ID_10G_BASE_T) || + (pf->hw.device_id == I40E_DEV_ID_10G_BASE_T4)) + pf->flags |= I40E_FLAG_HAVE_10GBASET_PHY; + /* print a string summarizing features */ i40e_print_features(pf); @@ -11186,10 +11211,11 @@ static void i40e_remove(struct pci_dev *pdev) i40e_ptp_stop(pf); /* Disable RSS in hw */ - wr32(hw, I40E_PFQF_HENA(0), 0); - wr32(hw, I40E_PFQF_HENA(1), 0); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), 0); + i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), 0); /* no more scheduling of any task */ + set_bit(__I40E_SUSPENDED, &pf->state); set_bit(__I40E_DOWN, &pf->state); del_timer_sync(&pf->service_timer); cancel_work_sync(&pf->service_task); diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h index e8deabde82b4..d51eee5bf79a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h +++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h @@ -74,6 +74,12 @@ i40e_status i40e_aq_set_rss_key(struct i40e_hw *hw, u32 i40e_led_get(struct i40e_hw *hw); void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink); +i40e_status i40e_led_set_phy(struct i40e_hw *hw, bool on, + u16 led_addr, u32 mode); +i40e_status i40e_led_get_phy(struct i40e_hw *hw, u16 *led_addr, + u16 *val); +i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw, + u32 time, u32 interval); /* admin send queue commands */ @@ -336,4 +342,19 @@ i40e_status i40e_aq_debug_dump(struct i40e_hw *hw, u8 cluster_id, struct i40e_asq_cmd_details *cmd_details); void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw, u16 vsi_seid); +i40e_status i40e_aq_rx_ctl_read_register(struct i40e_hw *hw, + u32 reg_addr, u32 *reg_val, + struct i40e_asq_cmd_details *cmd_details); +u32 i40e_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr); +i40e_status i40e_aq_rx_ctl_write_register(struct i40e_hw *hw, + u32 reg_addr, u32 reg_val, + struct i40e_asq_cmd_details *cmd_details); +void i40e_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val); +i40e_status i40e_read_phy_register(struct i40e_hw *hw, u8 page, + u16 reg, u8 phy_addr, u16 *value); +i40e_status i40e_write_phy_register(struct i40e_hw *hw, u8 page, + u16 reg, u8 phy_addr, u16 value); +u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num); +i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw, + u32 time, u32 interval); #endif /* _I40E_PROTOTYPE_H_ */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_register.h b/drivers/net/ethernet/intel/i40e/i40e_register.h index dc0402fe3370..86ca27f72f02 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_register.h +++ b/drivers/net/ethernet/intel/i40e/i40e_register.h @@ -2045,6 +2045,14 @@ #define I40E_PRTPM_TLPIC 0x001E43C0 /* Reset: GLOBR */ #define I40E_PRTPM_TLPIC_ETLPIC_SHIFT 0 #define I40E_PRTPM_TLPIC_ETLPIC_MASK I40E_MASK(0xFFFFFFFF, I40E_PRTPM_TLPIC_ETLPIC_SHIFT) +#define I40E_GL_PRS_FVBM(_i) (0x00269760 + ((_i) * 4)) /* _i=0...3 */ /* Reset: CORER */ +#define I40E_GL_PRS_FVBM_MAX_INDEX 3 +#define I40E_GL_PRS_FVBM_FV_BYTE_INDX_SHIFT 0 +#define I40E_GL_PRS_FVBM_FV_BYTE_INDX_MASK I40E_MASK(0x7F, I40E_GL_PRS_FVBM_FV_BYTE_INDX_SHIFT) +#define I40E_GL_PRS_FVBM_RULE_BUS_INDX_SHIFT 8 +#define I40E_GL_PRS_FVBM_RULE_BUS_INDX_MASK I40E_MASK(0x3F, I40E_GL_PRS_FVBM_RULE_BUS_INDX_SHIFT) +#define I40E_GL_PRS_FVBM_MSK_ENA_SHIFT 31 +#define I40E_GL_PRS_FVBM_MSK_ENA_MASK I40E_MASK(0x1, I40E_GL_PRS_FVBM_MSK_ENA_SHIFT) #define I40E_GLRPB_DPSS 0x000AC828 /* Reset: CORER */ #define I40E_GLRPB_DPSS_DPS_TCN_SHIFT 0 #define I40E_GLRPB_DPSS_DPS_TCN_MASK I40E_MASK(0xFFFFF, I40E_GLRPB_DPSS_DPS_TCN_SHIFT) @@ -2216,6 +2224,14 @@ #define I40E_PRTQF_FD_FLXINSET_MAX_INDEX 63 #define I40E_PRTQF_FD_FLXINSET_INSET_SHIFT 0 #define I40E_PRTQF_FD_FLXINSET_INSET_MASK I40E_MASK(0xFF, I40E_PRTQF_FD_FLXINSET_INSET_SHIFT) +#define I40E_PRTQF_FD_INSET(_i, _j) (0x00250000 + ((_i) * 64 + (_j) * 32)) /* _i=0...63, _j=0...1 */ /* Reset: CORER */ +#define I40E_PRTQF_FD_INSET_MAX_INDEX 63 +#define I40E_PRTQF_FD_INSET_INSET_SHIFT 0 +#define I40E_PRTQF_FD_INSET_INSET_MASK I40E_MASK(0xFFFFFFFF, I40E_PRTQF_FD_INSET_INSET_SHIFT) +#define I40E_PRTQF_FD_INSET(_i, _j) (0x00250000 + ((_i) * 64 + (_j) * 32)) /* _i=0...63, _j=0...1 */ /* Reset: CORER */ +#define I40E_PRTQF_FD_INSET_MAX_INDEX 63 +#define I40E_PRTQF_FD_INSET_INSET_SHIFT 0 +#define I40E_PRTQF_FD_INSET_INSET_MASK I40E_MASK(0xFFFFFFFF, I40E_PRTQF_FD_INSET_INSET_SHIFT) #define I40E_PRTQF_FD_MSK(_i, _j) (0x00252000 + ((_i) * 64 + (_j) * 32)) /* _i=0...63, _j=0...1 */ /* Reset: CORER */ #define I40E_PRTQF_FD_MSK_MAX_INDEX 63 #define I40E_PRTQF_FD_MSK_MASK_SHIFT 0 @@ -5155,6 +5171,38 @@ #define I40E_GLQF_FD_PCTYPES_MAX_INDEX 63 #define I40E_GLQF_FD_PCTYPES_FD_PCTYPE_SHIFT 0 #define I40E_GLQF_FD_PCTYPES_FD_PCTYPE_MASK I40E_MASK(0x3F, I40E_GLQF_FD_PCTYPES_FD_PCTYPE_SHIFT) +#define I40E_GLQF_FD_MSK(_i, _j) (0x00267200 + ((_i) * 4 + (_j) * 8)) /* _i=0...1, _j=0...63 */ /* Reset: CORER */ +#define I40E_GLQF_FD_MSK_MAX_INDEX 1 +#define I40E_GLQF_FD_MSK_MASK_SHIFT 0 +#define I40E_GLQF_FD_MSK_MASK_MASK I40E_MASK(0xFFFF, I40E_GLQF_FD_MSK_MASK_SHIFT) +#define I40E_GLQF_FD_MSK_OFFSET_SHIFT 16 +#define I40E_GLQF_FD_MSK_OFFSET_MASK I40E_MASK(0x3F, I40E_GLQF_FD_MSK_OFFSET_SHIFT) +#define I40E_GLQF_HASH_INSET(_i, _j) (0x00267600 + ((_i) * 4 + (_j) * 8)) /* _i=0...1, _j=0...63 */ /* Reset: CORER */ +#define I40E_GLQF_HASH_INSET_MAX_INDEX 1 +#define I40E_GLQF_HASH_INSET_INSET_SHIFT 0 +#define I40E_GLQF_HASH_INSET_INSET_MASK I40E_MASK(0xFFFFFFFF, I40E_GLQF_HASH_INSET_INSET_SHIFT) +#define I40E_GLQF_HASH_MSK(_i, _j) (0x00267A00 + ((_i) * 4 + (_j) * 8)) /* _i=0...1, _j=0...63 */ /* Reset: CORER */ +#define I40E_GLQF_HASH_MSK_MAX_INDEX 1 +#define I40E_GLQF_HASH_MSK_MASK_SHIFT 0 +#define I40E_GLQF_HASH_MSK_MASK_MASK I40E_MASK(0xFFFF, I40E_GLQF_HASH_MSK_MASK_SHIFT) +#define I40E_GLQF_HASH_MSK_OFFSET_SHIFT 16 +#define I40E_GLQF_HASH_MSK_OFFSET_MASK I40E_MASK(0x3F, I40E_GLQF_HASH_MSK_OFFSET_SHIFT) +#define I40E_GLQF_ORT(_i) (0x00268900 + ((_i) * 4)) /* _i=0...63 */ /* Reset: CORER */ +#define I40E_GLQF_ORT_MAX_INDEX 63 +#define I40E_GLQF_ORT_PIT_INDX_SHIFT 0 +#define I40E_GLQF_ORT_PIT_INDX_MASK I40E_MASK(0x1F, I40E_GLQF_ORT_PIT_INDX_SHIFT) +#define I40E_GLQF_ORT_FIELD_CNT_SHIFT 5 +#define I40E_GLQF_ORT_FIELD_CNT_MASK I40E_MASK(0x3, I40E_GLQF_ORT_FIELD_CNT_SHIFT) +#define I40E_GLQF_ORT_FLX_PAYLOAD_SHIFT 7 +#define I40E_GLQF_ORT_FLX_PAYLOAD_MASK I40E_MASK(0x1, I40E_GLQF_ORT_FLX_PAYLOAD_SHIFT) +#define I40E_GLQF_PIT(_i) (0x00268C80 + ((_i) * 4)) /* _i=0...23 */ /* Reset: CORER */ +#define I40E_GLQF_PIT_MAX_INDEX 23 +#define I40E_GLQF_PIT_SOURCE_OFF_SHIFT 0 +#define I40E_GLQF_PIT_SOURCE_OFF_MASK I40E_MASK(0x1F, I40E_GLQF_PIT_SOURCE_OFF_SHIFT) +#define I40E_GLQF_PIT_FSIZE_SHIFT 5 +#define I40E_GLQF_PIT_FSIZE_MASK I40E_MASK(0x1F, I40E_GLQF_PIT_FSIZE_SHIFT) +#define I40E_GLQF_PIT_DEST_OFF_SHIFT 10 +#define I40E_GLQF_PIT_DEST_OFF_MASK I40E_MASK(0x3F, I40E_GLQF_PIT_DEST_OFF_SHIFT) #define I40E_GLQF_FDEVICTENA(_i) (0x00270384 + ((_i) * 4)) /* _i=0...1 */ /* Reset: CORER */ #define I40E_GLQF_FDEVICTENA_MAX_INDEX 1 #define I40E_GLQF_FDEVICTENA_GLQF_FDEVICTENA_SHIFT 0 diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index 1d3afa7dda18..cb52f39d514a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -2576,7 +2576,7 @@ static void i40e_create_tx_ctx(struct i40e_ring *tx_ring, * * Returns -EBUSY if a stop is needed, else 0 **/ -static inline int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) { netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); /* Memory barrier before checking head and tail */ @@ -2593,77 +2593,71 @@ static inline int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) } /** - * i40e_maybe_stop_tx - 1st level check for tx stop conditions - * @tx_ring: the ring to be checked - * @size: the size buffer we want to assure is available - * - * Returns 0 if stop is not needed - **/ -#ifdef I40E_FCOE -inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) -#else -static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) -#endif -{ - if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) - return 0; - return __i40e_maybe_stop_tx(tx_ring, size); -} - -/** - * i40e_chk_linearize - Check if there are more than 8 fragments per packet + * __i40e_chk_linearize - Check if there are more than 8 fragments per packet * @skb: send buffer - * @tx_flags: collected send information * * Note: Our HW can't scatter-gather more than 8 fragments to build * a packet on the wire and so we need to figure out the cases where we * need to linearize the skb. **/ -static bool i40e_chk_linearize(struct sk_buff *skb, u32 tx_flags) +bool __i40e_chk_linearize(struct sk_buff *skb) { - struct skb_frag_struct *frag; - bool linearize = false; - unsigned int size = 0; - u16 num_frags; - u16 gso_segs; + const struct skb_frag_struct *frag, *stale; + int gso_size, nr_frags, sum; - num_frags = skb_shinfo(skb)->nr_frags; - gso_segs = skb_shinfo(skb)->gso_segs; + /* check to see if TSO is enabled, if so we may get a repreive */ + gso_size = skb_shinfo(skb)->gso_size; + if (unlikely(!gso_size)) + return true; - if (tx_flags & (I40E_TX_FLAGS_TSO | I40E_TX_FLAGS_FSO)) { - u16 j = 0; + /* no need to check if number of frags is less than 8 */ + nr_frags = skb_shinfo(skb)->nr_frags; + if (nr_frags < I40E_MAX_BUFFER_TXD) + return false; - if (num_frags < (I40E_MAX_BUFFER_TXD)) - goto linearize_chk_done; - /* try the simple math, if we have too many frags per segment */ - if (DIV_ROUND_UP((num_frags + gso_segs), gso_segs) > - I40E_MAX_BUFFER_TXD) { - linearize = true; - goto linearize_chk_done; - } - frag = &skb_shinfo(skb)->frags[0]; - /* we might still have more fragments per segment */ - do { - size += skb_frag_size(frag); - frag++; j++; - if ((size >= skb_shinfo(skb)->gso_size) && - (j < I40E_MAX_BUFFER_TXD)) { - size = (size % skb_shinfo(skb)->gso_size); - j = (size) ? 1 : 0; - } - if (j == I40E_MAX_BUFFER_TXD) { - linearize = true; - break; - } - num_frags--; - } while (num_frags); - } else { - if (num_frags >= I40E_MAX_BUFFER_TXD) - linearize = true; + /* We need to walk through the list and validate that each group + * of 6 fragments totals at least gso_size. However we don't need + * to perform such validation on the first or last 6 since the first + * 6 cannot inherit any data from a descriptor before them, and the + * last 6 cannot inherit any data from a descriptor after them. + */ + nr_frags -= I40E_MAX_BUFFER_TXD - 1; + frag = &skb_shinfo(skb)->frags[0]; + + /* Initialize size to the negative value of gso_size minus 1. We + * use this as the worst case scenerio in which the frag ahead + * of us only provides one byte which is why we are limited to 6 + * descriptors for a single transmit as the header and previous + * fragment are already consuming 2 descriptors. + */ + sum = 1 - gso_size; + + /* Add size of frags 1 through 5 to create our initial sum */ + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + + /* Walk through fragments adding latest fragment, testing it, and + * then removing stale fragments from the sum. + */ + stale = &skb_shinfo(skb)->frags[0]; + for (;;) { + sum += skb_frag_size(++frag); + + /* if sum is negative we failed to make sufficient progress */ + if (sum < 0) + return true; + + /* use pre-decrement to avoid processing last fragment */ + if (!--nr_frags) + break; + + sum -= skb_frag_size(++stale); } -linearize_chk_done: - return linearize; + return false; } /** @@ -2869,43 +2863,6 @@ dma_error: tx_ring->next_to_use = i; } -/** - * i40e_xmit_descriptor_count - calculate number of tx descriptors needed - * @skb: send buffer - * @tx_ring: ring to send buffer on - * - * Returns number of data descriptors needed for this skb. Returns 0 to indicate - * there is not enough descriptors available in this ring since we need at least - * one descriptor. - **/ -#ifdef I40E_FCOE -inline int i40e_xmit_descriptor_count(struct sk_buff *skb, - struct i40e_ring *tx_ring) -#else -static inline int i40e_xmit_descriptor_count(struct sk_buff *skb, - struct i40e_ring *tx_ring) -#endif -{ - unsigned int f; - int count = 0; - - /* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD, - * + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD, - * + 4 desc gap to avoid the cache line where head is, - * + 1 desc for context descriptor, - * otherwise try next time - */ - for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) - count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); - - count += TXD_USE_COUNT(skb_headlen(skb)); - if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) { - tx_ring->tx_stats.tx_busy++; - return 0; - } - return count; -} - /** * i40e_xmit_frame_ring - Sends buffer on Tx ring * @skb: send buffer @@ -2924,14 +2881,30 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, __be16 protocol; u32 td_cmd = 0; u8 hdr_len = 0; + int tso, count; int tsyn; - int tso; /* prefetch the data, we'll need it later */ prefetch(skb->data); - if (0 == i40e_xmit_descriptor_count(skb, tx_ring)) + count = i40e_xmit_descriptor_count(skb); + if (i40e_chk_linearize(skb, count)) { + if (__skb_linearize(skb)) + goto out_drop; + count = TXD_USE_COUNT(skb->len); + tx_ring->tx_stats.tx_linearize++; + } + + /* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD, + * + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD, + * + 4 desc gap to avoid the cache line where head is, + * + 1 desc for context descriptor, + * otherwise try next time + */ + if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) { + tx_ring->tx_stats.tx_busy++; return NETDEV_TX_BUSY; + } /* prepare the xmit flags */ if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags)) @@ -2956,27 +2929,22 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, else if (tso) tx_flags |= I40E_TX_FLAGS_TSO; - tsyn = i40e_tsyn(tx_ring, skb, tx_flags, &cd_type_cmd_tso_mss); - - if (tsyn) - tx_flags |= I40E_TX_FLAGS_TSYN; - - if (i40e_chk_linearize(skb, tx_flags)) { - if (skb_linearize(skb)) - goto out_drop; - tx_ring->tx_stats.tx_linearize++; - } - skb_tx_timestamp(skb); - - /* always enable CRC insertion offload */ - td_cmd |= I40E_TX_DESC_CMD_ICRC; - /* Always offload the checksum, since it's in the data descriptor */ tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset, tx_ring, &cd_tunneling); if (tso < 0) goto out_drop; + tsyn = i40e_tsyn(tx_ring, skb, tx_flags, &cd_type_cmd_tso_mss); + + if (tsyn) + tx_flags |= I40E_TX_FLAGS_TSYN; + + skb_tx_timestamp(skb); + + /* always enable CRC insertion offload */ + td_cmd |= I40E_TX_DESC_CMD_ICRC; + i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss, cd_tunneling, cd_l2tag2); diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h index fde5f42524fb..56009709528a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h @@ -331,13 +331,13 @@ int i40e_napi_poll(struct napi_struct *napi, int budget); void i40e_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, struct i40e_tx_buffer *first, u32 tx_flags, const u8 hdr_len, u32 td_cmd, u32 td_offset); -int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size); -int i40e_xmit_descriptor_count(struct sk_buff *skb, struct i40e_ring *tx_ring); int i40e_tx_prepare_vlan_flags(struct sk_buff *skb, struct i40e_ring *tx_ring, u32 *flags); #endif void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector); u32 i40e_get_tx_pending(struct i40e_ring *ring, bool in_sw); +int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size); +bool __i40e_chk_linearize(struct sk_buff *skb); /** * i40e_get_head - Retrieve head from head writeback @@ -352,4 +352,63 @@ static inline u32 i40e_get_head(struct i40e_ring *tx_ring) return le32_to_cpu(*(volatile __le32 *)head); } + +/** + * i40e_xmit_descriptor_count - calculate number of Tx descriptors needed + * @skb: send buffer + * @tx_ring: ring to send buffer on + * + * Returns number of data descriptors needed for this skb. Returns 0 to indicate + * there is not enough descriptors available in this ring since we need at least + * one descriptor. + **/ +static inline int i40e_xmit_descriptor_count(struct sk_buff *skb) +{ + const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0]; + unsigned int nr_frags = skb_shinfo(skb)->nr_frags; + int count = 0, size = skb_headlen(skb); + + for (;;) { + count += TXD_USE_COUNT(size); + + if (!nr_frags--) + break; + + size = skb_frag_size(frag++); + } + + return count; +} + +/** + * i40e_maybe_stop_tx - 1st level check for Tx stop conditions + * @tx_ring: the ring to be checked + * @size: the size buffer we want to assure is available + * + * Returns 0 if stop is not needed + **/ +static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +{ + if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) + return 0; + return __i40e_maybe_stop_tx(tx_ring, size); +} + +/** + * i40e_chk_linearize - Check if there are more than 8 fragments per packet + * @skb: send buffer + * @count: number of buffers used + * + * Note: Our HW can't scatter-gather more than 8 fragments to build + * a packet on the wire and so we need to figure out the cases where we + * need to linearize the skb. + **/ +static inline bool i40e_chk_linearize(struct sk_buff *skb, int count) +{ + /* we can only support up to 8 data buffers for a single send */ + if (likely(count <= I40E_MAX_BUFFER_TXD)) + return false; + + return __i40e_chk_linearize(skb); +} #endif /* _I40E_TXRX_H_ */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h index b59a021b7a69..0a0baf71041b 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_type.h +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h @@ -90,6 +90,22 @@ enum i40e_debug_mask { I40E_DEBUG_ALL = 0xFFFFFFFF }; +#define I40E_MDIO_STCODE 0 +#define I40E_MDIO_OPCODE_ADDRESS 0 +#define I40E_MDIO_OPCODE_WRITE I40E_MASK(1, \ + I40E_GLGEN_MSCA_OPCODE_SHIFT) +#define I40E_MDIO_OPCODE_READ_INC_ADDR I40E_MASK(2, \ + I40E_GLGEN_MSCA_OPCODE_SHIFT) +#define I40E_MDIO_OPCODE_READ I40E_MASK(3, \ + I40E_GLGEN_MSCA_OPCODE_SHIFT) + +#define I40E_PHY_COM_REG_PAGE 0x1E +#define I40E_PHY_LED_LINK_MODE_MASK 0xF0 +#define I40E_PHY_LED_MANUAL_ON 0x100 +#define I40E_PHY_LED_PROV_REG_1 0xC430 +#define I40E_PHY_LED_MODE_MASK 0xFFFF +#define I40E_PHY_LED_MODE_ORIG 0x80000000 + /* These are structs for managing the hardware information and the operations. * The structures of function pointers are filled out at init time when we * know for sure exactly which hardware we're working with. This gives us the diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 5dcd19869a41..acd2693a4e97 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -602,8 +602,8 @@ static void i40e_enable_vf_mappings(struct i40e_vf *vf) * that VF queues be mapped using this method, even when they are * contiguous in real life */ - wr32(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id), - I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK); + i40e_write_rx_ctl(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id), + I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK); /* enable VF vplan_qtable mappings */ reg = I40E_VPLAN_MAPENA_TXRX_ENA_MASK; @@ -630,7 +630,8 @@ static void i40e_enable_vf_mappings(struct i40e_vf *vf) (j * 2) + 1); reg |= qid << 16; } - wr32(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id), reg); + i40e_write_rx_ctl(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id), + reg); } i40e_flush(hw); @@ -2202,6 +2203,8 @@ int i40e_ndo_set_vf_port_vlan(struct net_device *netdev, * and then reloading the VF driver. */ i40e_vc_disable_vf(pf, vf); + /* During reset the VF got a new VSI, so refresh the pointer. */ + vsi = pf->vsi[vf->lan_vsi_idx]; } /* Check for condition where there was already a port VLAN ID diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h index 815e481ccd9c..aad8d6277110 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h @@ -146,6 +146,8 @@ enum i40e_admin_queue_opc { i40e_aqc_opc_set_port_parameters = 0x0203, i40e_aqc_opc_get_switch_resource_alloc = 0x0204, i40e_aqc_opc_set_switch_config = 0x0205, + i40e_aqc_opc_rx_ctl_reg_read = 0x0206, + i40e_aqc_opc_rx_ctl_reg_write = 0x0207, i40e_aqc_opc_add_vsi = 0x0210, i40e_aqc_opc_update_vsi_parameters = 0x0211, @@ -693,6 +695,20 @@ struct i40e_aqc_set_switch_config { I40E_CHECK_CMD_LENGTH(i40e_aqc_set_switch_config); +/* Read Receive control registers (direct 0x0206) + * Write Receive control registers (direct 0x0207) + * used for accessing Rx control registers that can be + * slow and need special handling when under high Rx load + */ +struct i40e_aqc_rx_ctl_reg_read_write { + __le32 reserved1; + __le32 address; + __le32 reserved2; + __le32 value; +}; + +I40E_CHECK_CMD_LENGTH(i40e_aqc_rx_ctl_reg_read_write); + /* Add VSI (indirect 0x0210) * this indirect command uses struct i40e_aqc_vsi_properties_data * as the indirect buffer (128 bytes) diff --git a/drivers/net/ethernet/intel/i40evf/i40e_common.c b/drivers/net/ethernet/intel/i40evf/i40e_common.c index 938783e0baac..771ac6ad8cda 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_common.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_common.c @@ -903,6 +903,131 @@ struct i40e_rx_ptype_decoded i40evf_ptype_lookup[] = { I40E_PTT_UNUSED_ENTRY(255) }; +/** + * i40evf_aq_rx_ctl_read_register - use FW to read from an Rx control register + * @hw: pointer to the hw struct + * @reg_addr: register address + * @reg_val: ptr to register value + * @cmd_details: pointer to command details structure or NULL + * + * Use the firmware to read the Rx control register, + * especially useful if the Rx unit is under heavy pressure + **/ +i40e_status i40evf_aq_rx_ctl_read_register(struct i40e_hw *hw, + u32 reg_addr, u32 *reg_val, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_rx_ctl_reg_read_write *cmd_resp = + (struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw; + i40e_status status; + + if (!reg_val) + return I40E_ERR_PARAM; + + i40evf_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_rx_ctl_reg_read); + + cmd_resp->address = cpu_to_le32(reg_addr); + + status = i40evf_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + if (status == 0) + *reg_val = le32_to_cpu(cmd_resp->value); + + return status; +} + +/** + * i40evf_read_rx_ctl - read from an Rx control register + * @hw: pointer to the hw struct + * @reg_addr: register address + **/ +u32 i40evf_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr) +{ + i40e_status status = 0; + bool use_register; + int retry = 5; + u32 val = 0; + + use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5); + if (!use_register) { +do_retry: + status = i40evf_aq_rx_ctl_read_register(hw, reg_addr, + &val, NULL); + if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) { + usleep_range(1000, 2000); + retry--; + goto do_retry; + } + } + + /* if the AQ access failed, try the old-fashioned way */ + if (status || use_register) + val = rd32(hw, reg_addr); + + return val; +} + +/** + * i40evf_aq_rx_ctl_write_register + * @hw: pointer to the hw struct + * @reg_addr: register address + * @reg_val: register value + * @cmd_details: pointer to command details structure or NULL + * + * Use the firmware to write to an Rx control register, + * especially useful if the Rx unit is under heavy pressure + **/ +i40e_status i40evf_aq_rx_ctl_write_register(struct i40e_hw *hw, + u32 reg_addr, u32 reg_val, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_rx_ctl_reg_read_write *cmd = + (struct i40e_aqc_rx_ctl_reg_read_write *)&desc.params.raw; + i40e_status status; + + i40evf_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_rx_ctl_reg_write); + + cmd->address = cpu_to_le32(reg_addr); + cmd->value = cpu_to_le32(reg_val); + + status = i40evf_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + +/** + * i40evf_write_rx_ctl - write to an Rx control register + * @hw: pointer to the hw struct + * @reg_addr: register address + * @reg_val: register value + **/ +void i40evf_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val) +{ + i40e_status status = 0; + bool use_register; + int retry = 5; + + use_register = (hw->aq.api_maj_ver == 1) && (hw->aq.api_min_ver < 5); + if (!use_register) { +do_retry: + status = i40evf_aq_rx_ctl_write_register(hw, reg_addr, + reg_val, NULL); + if (hw->aq.asq_last_status == I40E_AQ_RC_EAGAIN && retry) { + usleep_range(1000, 2000); + retry--; + goto do_retry; + } + } + + /* if the AQ access failed, try the old-fashioned way */ + if (status || use_register) + wr32(hw, reg_addr, reg_val); +} + /** * i40e_aq_send_msg_to_pf * @hw: pointer to the hardware structure diff --git a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h index cbd9a1b078ab..d89d52109efa 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_prototype.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_prototype.h @@ -103,4 +103,19 @@ i40e_status i40e_aq_add_rem_control_packet_filter(struct i40e_hw *hw, struct i40e_asq_cmd_details *cmd_details); void i40e_add_filter_to_drop_tx_flow_control_frames(struct i40e_hw *hw, u16 vsi_seid); +i40e_status i40evf_aq_rx_ctl_read_register(struct i40e_hw *hw, + u32 reg_addr, u32 *reg_val, + struct i40e_asq_cmd_details *cmd_details); +u32 i40evf_read_rx_ctl(struct i40e_hw *hw, u32 reg_addr); +i40e_status i40evf_aq_rx_ctl_write_register(struct i40e_hw *hw, + u32 reg_addr, u32 reg_val, + struct i40e_asq_cmd_details *cmd_details); +void i40evf_write_rx_ctl(struct i40e_hw *hw, u32 reg_addr, u32 reg_val); +i40e_status i40e_read_phy_register(struct i40e_hw *hw, u8 page, + u16 reg, u8 phy_addr, u16 *value); +i40e_status i40e_write_phy_register(struct i40e_hw *hw, u8 page, + u16 reg, u8 phy_addr, u16 value); +u8 i40e_get_phy_address(struct i40e_hw *hw, u8 dev_num); +i40e_status i40e_blink_phy_link_led(struct i40e_hw *hw, + u32 time, u32 interval); #endif /* _I40E_PROTOTYPE_H_ */ diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c index 6d66fcdc6122..ebcc25c05796 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c @@ -1796,59 +1796,71 @@ static void i40e_create_tx_ctx(struct i40e_ring *tx_ring, } /** - * i40e_chk_linearize - Check if there are more than 8 fragments per packet + * __i40evf_chk_linearize - Check if there are more than 8 fragments per packet * @skb: send buffer - * @tx_flags: collected send information * * Note: Our HW can't scatter-gather more than 8 fragments to build * a packet on the wire and so we need to figure out the cases where we * need to linearize the skb. **/ -static bool i40e_chk_linearize(struct sk_buff *skb, u32 tx_flags) +bool __i40evf_chk_linearize(struct sk_buff *skb) { - struct skb_frag_struct *frag; - bool linearize = false; - unsigned int size = 0; - u16 num_frags; - u16 gso_segs; + const struct skb_frag_struct *frag, *stale; + int gso_size, nr_frags, sum; - num_frags = skb_shinfo(skb)->nr_frags; - gso_segs = skb_shinfo(skb)->gso_segs; + /* check to see if TSO is enabled, if so we may get a repreive */ + gso_size = skb_shinfo(skb)->gso_size; + if (unlikely(!gso_size)) + return true; - if (tx_flags & (I40E_TX_FLAGS_TSO | I40E_TX_FLAGS_FSO)) { - u16 j = 0; + /* no need to check if number of frags is less than 8 */ + nr_frags = skb_shinfo(skb)->nr_frags; + if (nr_frags < I40E_MAX_BUFFER_TXD) + return false; - if (num_frags < (I40E_MAX_BUFFER_TXD)) - goto linearize_chk_done; - /* try the simple math, if we have too many frags per segment */ - if (DIV_ROUND_UP((num_frags + gso_segs), gso_segs) > - I40E_MAX_BUFFER_TXD) { - linearize = true; - goto linearize_chk_done; - } - frag = &skb_shinfo(skb)->frags[0]; - /* we might still have more fragments per segment */ - do { - size += skb_frag_size(frag); - frag++; j++; - if ((size >= skb_shinfo(skb)->gso_size) && - (j < I40E_MAX_BUFFER_TXD)) { - size = (size % skb_shinfo(skb)->gso_size); - j = (size) ? 1 : 0; - } - if (j == I40E_MAX_BUFFER_TXD) { - linearize = true; - break; - } - num_frags--; - } while (num_frags); - } else { - if (num_frags >= I40E_MAX_BUFFER_TXD) - linearize = true; + /* We need to walk through the list and validate that each group + * of 6 fragments totals at least gso_size. However we don't need + * to perform such validation on the first or last 6 since the first + * 6 cannot inherit any data from a descriptor before them, and the + * last 6 cannot inherit any data from a descriptor after them. + */ + nr_frags -= I40E_MAX_BUFFER_TXD - 1; + frag = &skb_shinfo(skb)->frags[0]; + + /* Initialize size to the negative value of gso_size minus 1. We + * use this as the worst case scenerio in which the frag ahead + * of us only provides one byte which is why we are limited to 6 + * descriptors for a single transmit as the header and previous + * fragment are already consuming 2 descriptors. + */ + sum = 1 - gso_size; + + /* Add size of frags 1 through 5 to create our initial sum */ + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + sum += skb_frag_size(++frag); + + /* Walk through fragments adding latest fragment, testing it, and + * then removing stale fragments from the sum. + */ + stale = &skb_shinfo(skb)->frags[0]; + for (;;) { + sum += skb_frag_size(++frag); + + /* if sum is negative we failed to make sufficient progress */ + if (sum < 0) + return true; + + /* use pre-decrement to avoid processing last fragment */ + if (!--nr_frags) + break; + + sum -= skb_frag_size(++stale); } -linearize_chk_done: - return linearize; + return false; } /** @@ -1858,7 +1870,7 @@ linearize_chk_done: * * Returns -EBUSY if a stop is needed, else 0 **/ -static inline int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size) { netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); /* Memory barrier before checking head and tail */ @@ -1874,20 +1886,6 @@ static inline int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size) return 0; } -/** - * i40evf_maybe_stop_tx - 1st level check for tx stop conditions - * @tx_ring: the ring to be checked - * @size: the size buffer we want to assure is available - * - * Returns 0 if stop is not needed - **/ -static inline int i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size) -{ - if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) - return 0; - return __i40evf_maybe_stop_tx(tx_ring, size); -} - /** * i40evf_tx_map - Build the Tx descriptor * @tx_ring: ring to send buffer on @@ -2003,7 +2001,7 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, netdev_tx_sent_queue(netdev_get_tx_queue(tx_ring->netdev, tx_ring->queue_index), first->bytecount); - i40evf_maybe_stop_tx(tx_ring, DESC_NEEDED); + i40e_maybe_stop_tx(tx_ring, DESC_NEEDED); /* Algorithm to optimize tail and RS bit setting: * if xmit_more is supported @@ -2085,38 +2083,6 @@ dma_error: tx_ring->next_to_use = i; } -/** - * i40evf_xmit_descriptor_count - calculate number of tx descriptors needed - * @skb: send buffer - * @tx_ring: ring to send buffer on - * - * Returns number of data descriptors needed for this skb. Returns 0 to indicate - * there is not enough descriptors available in this ring since we need at least - * one descriptor. - **/ -static inline int i40evf_xmit_descriptor_count(struct sk_buff *skb, - struct i40e_ring *tx_ring) -{ - unsigned int f; - int count = 0; - - /* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD, - * + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD, - * + 4 desc gap to avoid the cache line where head is, - * + 1 desc for context descriptor, - * otherwise try next time - */ - for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) - count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size); - - count += TXD_USE_COUNT(skb_headlen(skb)); - if (i40evf_maybe_stop_tx(tx_ring, count + 4 + 1)) { - tx_ring->tx_stats.tx_busy++; - return 0; - } - return count; -} - /** * i40e_xmit_frame_ring - Sends buffer on Tx ring * @skb: send buffer @@ -2135,13 +2101,29 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, __be16 protocol; u32 td_cmd = 0; u8 hdr_len = 0; - int tso; + int tso, count; /* prefetch the data, we'll need it later */ prefetch(skb->data); - if (0 == i40evf_xmit_descriptor_count(skb, tx_ring)) + count = i40e_xmit_descriptor_count(skb); + if (i40e_chk_linearize(skb, count)) { + if (__skb_linearize(skb)) + goto out_drop; + count = TXD_USE_COUNT(skb->len); + tx_ring->tx_stats.tx_linearize++; + } + + /* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD, + * + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD, + * + 4 desc gap to avoid the cache line where head is, + * + 1 desc for context descriptor, + * otherwise try next time + */ + if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) { + tx_ring->tx_stats.tx_busy++; return NETDEV_TX_BUSY; + } /* prepare the xmit flags */ if (i40evf_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags)) @@ -2166,22 +2148,17 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, else if (tso) tx_flags |= I40E_TX_FLAGS_TSO; - if (i40e_chk_linearize(skb, tx_flags)) { - if (skb_linearize(skb)) - goto out_drop; - tx_ring->tx_stats.tx_linearize++; - } - skb_tx_timestamp(skb); - - /* always enable CRC insertion offload */ - td_cmd |= I40E_TX_DESC_CMD_ICRC; - /* Always offload the checksum, since it's in the data descriptor */ tso = i40e_tx_enable_csum(skb, &tx_flags, &td_cmd, &td_offset, tx_ring, &cd_tunneling); if (tso < 0) goto out_drop; + skb_tx_timestamp(skb); + + /* always enable CRC insertion offload */ + td_cmd |= I40E_TX_DESC_CMD_ICRC; + i40e_create_tx_ctx(tx_ring, cd_type_cmd_tso_mss, cd_tunneling, cd_l2tag2); diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h index 6ea8701cf066..c1dd8c5c9666 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h @@ -326,6 +326,8 @@ void i40evf_free_rx_resources(struct i40e_ring *rx_ring); int i40evf_napi_poll(struct napi_struct *napi, int budget); void i40evf_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector); u32 i40evf_get_tx_pending(struct i40e_ring *ring, bool in_sw); +int __i40evf_maybe_stop_tx(struct i40e_ring *tx_ring, int size); +bool __i40evf_chk_linearize(struct sk_buff *skb); /** * i40e_get_head - Retrieve head from head writeback @@ -340,4 +342,63 @@ static inline u32 i40e_get_head(struct i40e_ring *tx_ring) return le32_to_cpu(*(volatile __le32 *)head); } + +/** + * i40e_xmit_descriptor_count - calculate number of Tx descriptors needed + * @skb: send buffer + * @tx_ring: ring to send buffer on + * + * Returns number of data descriptors needed for this skb. Returns 0 to indicate + * there is not enough descriptors available in this ring since we need at least + * one descriptor. + **/ +static inline int i40e_xmit_descriptor_count(struct sk_buff *skb) +{ + const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0]; + unsigned int nr_frags = skb_shinfo(skb)->nr_frags; + int count = 0, size = skb_headlen(skb); + + for (;;) { + count += TXD_USE_COUNT(size); + + if (!nr_frags--) + break; + + size = skb_frag_size(frag++); + } + + return count; +} + +/** + * i40e_maybe_stop_tx - 1st level check for Tx stop conditions + * @tx_ring: the ring to be checked + * @size: the size buffer we want to assure is available + * + * Returns 0 if stop is not needed + **/ +static inline int i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) +{ + if (likely(I40E_DESC_UNUSED(tx_ring) >= size)) + return 0; + return __i40evf_maybe_stop_tx(tx_ring, size); +} + +/** + * i40e_chk_linearize - Check if there are more than 8 fragments per packet + * @skb: send buffer + * @count: number of buffers used + * + * Note: Our HW can't scatter-gather more than 8 fragments to build + * a packet on the wire and so we need to figure out the cases where we + * need to linearize the skb. + **/ +static inline bool i40e_chk_linearize(struct sk_buff *skb, int count) +{ + /* we can only support up to 8 data buffers for a single send */ + if (likely(count <= I40E_MAX_BUFFER_TXD)) + return false; + + return __i40evf_chk_linearize(skb); +} #endif /* _I40E_TXRX_H_ */ diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index 3396fe32cc6d..4b70aae2fa84 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -38,7 +38,7 @@ static const char i40evf_driver_string[] = #define DRV_VERSION_MAJOR 1 #define DRV_VERSION_MINOR 4 -#define DRV_VERSION_BUILD 11 +#define DRV_VERSION_BUILD 15 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) \