diff options
25 files changed, 690 insertions, 239 deletions
diff --git a/Documentation/networking/netconsole.rst b/Documentation/networking/netconsole.rst index d55c2a22ec7a..94c4680fdf3e 100644 --- a/Documentation/networking/netconsole.rst +++ b/Documentation/networking/netconsole.rst @@ -124,7 +124,7 @@ To remove a target:: The interface exposes these parameters of a netconsole target to userspace: - ============== ================================= ============ + =============== ================================= ============ enabled Is this target currently enabled? (read-write) extended Extended mode enabled (read-write) release Prepend kernel release to message (read-write) @@ -135,7 +135,8 @@ The interface exposes these parameters of a netconsole target to userspace: remote_ip Remote agent's IP address (read-write) local_mac Local interface's MAC address (read-only) remote_mac Remote agent's MAC address (read-write) - ============== ================================= ============ + transmit_errors Number of packet send errors (read-only) + =============== ================================= ============ The "enabled" attribute is also used to control whether the parameters of a target can be updated or not -- you can modify the parameters of only diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 1fb285fa0bdb..fe6261b81540 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -3960,20 +3960,27 @@ static struct mvneta_port *mvneta_pcs_to_port(struct phylink_pcs *pcs) return container_of(pcs, struct mvneta_port, phylink_pcs); } -static int mvneta_pcs_validate(struct phylink_pcs *pcs, - unsigned long *supported, - const struct phylink_link_state *state) +static unsigned int mvneta_pcs_inband_caps(struct phylink_pcs *pcs, + phy_interface_t interface) { - /* We only support QSGMII, SGMII, 802.3z and RGMII modes. - * When in 802.3z mode, we must have AN enabled: + /* When operating in an 802.3z mode, we must have AN enabled: * "Bit 2 Field InBandAnEn In-band Auto-Negotiation enable. ... * When <PortType> = 1 (1000BASE-X) this field must be set to 1." + * Therefore, inband is "required". */ - if (phy_interface_mode_is_8023z(state->interface) && - !phylink_test(state->advertising, Autoneg)) - return -EINVAL; + if (phy_interface_mode_is_8023z(interface)) + return LINK_INBAND_ENABLE; - return 0; + /* QSGMII, SGMII and RGMII can be configured to use inband + * signalling of the AN result. Indicate these as "possible". + */ + if (interface == PHY_INTERFACE_MODE_SGMII || + interface == PHY_INTERFACE_MODE_QSGMII || + phy_interface_mode_is_rgmii(interface)) + return LINK_INBAND_DISABLE | LINK_INBAND_ENABLE; + + /* For any other modes, indicate that inband is not supported. */ + return LINK_INBAND_DISABLE; } static void mvneta_pcs_get_state(struct phylink_pcs *pcs, @@ -4071,7 +4078,7 @@ static void mvneta_pcs_an_restart(struct phylink_pcs *pcs) } static const struct phylink_pcs_ops mvneta_phylink_pcs_ops = { - .pcs_validate = mvneta_pcs_validate, + .pcs_inband_caps = mvneta_pcs_inband_caps, .pcs_get_state = mvneta_pcs_get_state, .pcs_config = mvneta_pcs_config, .pcs_an_restart = mvneta_pcs_an_restart, diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c index 571631a30320..f85229a30844 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c @@ -6224,19 +6224,26 @@ static const struct phylink_pcs_ops mvpp2_phylink_xlg_pcs_ops = { .pcs_config = mvpp2_xlg_pcs_config, }; -static int mvpp2_gmac_pcs_validate(struct phylink_pcs *pcs, - unsigned long *supported, - const struct phylink_link_state *state) +static unsigned int mvpp2_gmac_pcs_inband_caps(struct phylink_pcs *pcs, + phy_interface_t interface) { - /* When in 802.3z mode, we must have AN enabled: + /* When operating in an 802.3z mode, we must have AN enabled: * Bit 2 Field InBandAnEn In-band Auto-Negotiation enable. ... * When <PortType> = 1 (1000BASE-X) this field must be set to 1. + * Therefore, inband is "required". */ - if (phy_interface_mode_is_8023z(state->interface) && - !phylink_test(state->advertising, Autoneg)) - return -EINVAL; + if (phy_interface_mode_is_8023z(interface)) + return LINK_INBAND_ENABLE; - return 0; + /* SGMII and RGMII can be configured to use inband signalling of the + * AN result. Indicate these as "possible". + */ + if (interface == PHY_INTERFACE_MODE_SGMII || + phy_interface_mode_is_rgmii(interface)) + return LINK_INBAND_DISABLE | LINK_INBAND_ENABLE; + + /* For any other modes, indicate that inband is not supported. */ + return LINK_INBAND_DISABLE; } static void mvpp2_gmac_pcs_get_state(struct phylink_pcs *pcs, @@ -6343,7 +6350,7 @@ static void mvpp2_gmac_pcs_an_restart(struct phylink_pcs *pcs) } static const struct phylink_pcs_ops mvpp2_phylink_gmac_pcs_ops = { - .pcs_validate = mvpp2_gmac_pcs_validate, + .pcs_inband_caps = mvpp2_gmac_pcs_inband_caps, .pcs_get_state = mvpp2_gmac_pcs_get_state, .pcs_config = mvpp2_gmac_pcs_config, .pcs_an_restart = mvpp2_gmac_pcs_an_restart, diff --git a/drivers/net/ethernet/realtek/r8169.h b/drivers/net/ethernet/realtek/r8169.h index be4c9622618d..8904aae41aca 100644 --- a/drivers/net/ethernet/realtek/r8169.h +++ b/drivers/net/ethernet/realtek/r8169.h @@ -23,7 +23,7 @@ enum mac_version { RTL_GIGA_MAC_VER_08, RTL_GIGA_MAC_VER_09, RTL_GIGA_MAC_VER_10, - RTL_GIGA_MAC_VER_11, + /* support for RTL_GIGA_MAC_VER_11 has been removed */ /* RTL_GIGA_MAC_VER_12 was handled the same as VER_17 */ /* RTL_GIGA_MAC_VER_13 was merged with VER_10 */ RTL_GIGA_MAC_VER_14, diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 739707a7b40f..6934bdee2a91 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -104,7 +104,6 @@ static const struct { [RTL_GIGA_MAC_VER_08] = {"RTL8102e" }, [RTL_GIGA_MAC_VER_09] = {"RTL8102e/RTL8103e" }, [RTL_GIGA_MAC_VER_10] = {"RTL8101e/RTL8100e" }, - [RTL_GIGA_MAC_VER_11] = {"RTL8168b/8111b" }, [RTL_GIGA_MAC_VER_14] = {"RTL8401" }, [RTL_GIGA_MAC_VER_17] = {"RTL8168b/8111b" }, [RTL_GIGA_MAC_VER_18] = {"RTL8168cp/8111cp" }, @@ -623,7 +622,6 @@ struct rtl8169_tc_offsets { enum rtl_flag { RTL_FLAG_TASK_RESET_PENDING, - RTL_FLAG_TASK_RESET_NO_QUEUE_WAKE, RTL_FLAG_TASK_TX_TIMEOUT, RTL_FLAG_MAX }; @@ -2336,7 +2334,7 @@ static enum mac_version rtl8169_get_mac_version(u16 xid, bool gmii) /* 8168B family. */ { 0x7c8, 0x380, RTL_GIGA_MAC_VER_17 }, - /* This one is very old and rare, let's see if anybody complains. + /* This one is very old and rare, support has been removed. * { 0x7c8, 0x300, RTL_GIGA_MAC_VER_11 }, */ @@ -3804,7 +3802,6 @@ static void rtl_hw_config(struct rtl8169_private *tp) [RTL_GIGA_MAC_VER_08] = rtl_hw_start_8102e_3, [RTL_GIGA_MAC_VER_09] = rtl_hw_start_8102e_2, [RTL_GIGA_MAC_VER_10] = NULL, - [RTL_GIGA_MAC_VER_11] = rtl_hw_start_8168b, [RTL_GIGA_MAC_VER_14] = rtl_hw_start_8401, [RTL_GIGA_MAC_VER_17] = rtl_hw_start_8168b, [RTL_GIGA_MAC_VER_18] = rtl_hw_start_8168cp_1, @@ -4680,12 +4677,6 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance) if (status & LinkChg) phy_mac_interrupt(tp->phydev); - if (unlikely(status & RxFIFOOver && - tp->mac_version == RTL_GIGA_MAC_VER_11)) { - netif_stop_queue(tp->dev); - rtl_schedule_task(tp, RTL_FLAG_TASK_RESET_PENDING); - } - rtl_irq_disable(tp); napi_schedule(&tp->napi); out: @@ -4723,8 +4714,6 @@ static void rtl_task(struct work_struct *work) reset: rtl_reset_work(tp); netif_wake_queue(tp->dev); - } else if (test_and_clear_bit(RTL_FLAG_TASK_RESET_NO_QUEUE_WAKE, tp->wk.flags)) { - rtl_reset_work(tp); } } @@ -5103,9 +5092,6 @@ static void rtl_set_irq_mask(struct rtl8169_private *tp) if (tp->mac_version <= RTL_GIGA_MAC_VER_06) tp->irq_mask |= SYSErr | RxFIFOOver; - else if (tp->mac_version == RTL_GIGA_MAC_VER_11) - /* special workaround needed */ - tp->irq_mask |= RxFIFOOver; } static int rtl_alloc_irq(struct rtl8169_private *tp) @@ -5300,7 +5286,6 @@ static int rtl_jumbo_max(struct rtl8169_private *tp) case RTL_GIGA_MAC_VER_02 ... RTL_GIGA_MAC_VER_06: return JUMBO_7K; /* RTL8168b */ - case RTL_GIGA_MAC_VER_11: case RTL_GIGA_MAC_VER_17: return JUMBO_4K; /* RTL8168c */ @@ -5347,13 +5332,6 @@ static bool rtl_aspm_is_safe(struct rtl8169_private *tp) return false; } -static umode_t r8169_hwmon_is_visible(const void *drvdata, - enum hwmon_sensor_types type, - u32 attr, int channel) -{ - return 0444; -} - static int r8169_hwmon_read(struct device *dev, enum hwmon_sensor_types type, u32 attr, int channel, long *val) { @@ -5370,7 +5348,7 @@ static int r8169_hwmon_read(struct device *dev, enum hwmon_sensor_types type, } static const struct hwmon_ops r8169_hwmon_ops = { - .is_visible = r8169_hwmon_is_visible, + .visible = 0444, .read = r8169_hwmon_read, }; diff --git a/drivers/net/ethernet/realtek/r8169_phy_config.c b/drivers/net/ethernet/realtek/r8169_phy_config.c index 5307c6ff4e25..b28b30390e84 100644 --- a/drivers/net/ethernet/realtek/r8169_phy_config.c +++ b/drivers/net/ethernet/realtek/r8169_phy_config.c @@ -276,15 +276,6 @@ static void rtl8169sce_hw_phy_config(struct rtl8169_private *tp, rtl_writephy_batch(phydev, phy_reg_init); } -static void rtl8168bb_hw_phy_config(struct rtl8169_private *tp, - struct phy_device *phydev) -{ - phy_write(phydev, 0x1f, 0x0001); - phy_set_bits(phydev, 0x16, BIT(0)); - phy_write(phydev, 0x10, 0xf41b); - phy_write(phydev, 0x1f, 0x0000); -} - static void rtl8168bef_hw_phy_config(struct rtl8169_private *tp, struct phy_device *phydev) { @@ -1136,7 +1127,6 @@ void r8169_hw_phy_config(struct rtl8169_private *tp, struct phy_device *phydev, [RTL_GIGA_MAC_VER_08] = rtl8102e_hw_phy_config, [RTL_GIGA_MAC_VER_09] = rtl8102e_hw_phy_config, [RTL_GIGA_MAC_VER_10] = NULL, - [RTL_GIGA_MAC_VER_11] = rtl8168bb_hw_phy_config, [RTL_GIGA_MAC_VER_14] = rtl8401_hw_phy_config, [RTL_GIGA_MAC_VER_17] = rtl8168bef_hw_phy_config, [RTL_GIGA_MAC_VER_18] = rtl8168cp_1_hw_phy_config, diff --git a/drivers/net/ethernet/realtek/rtase/rtase.h b/drivers/net/ethernet/realtek/rtase/rtase.h index dbc3f92eebc4..2bbfcad613ab 100644 --- a/drivers/net/ethernet/realtek/rtase/rtase.h +++ b/drivers/net/ethernet/realtek/rtase/rtase.h @@ -13,6 +13,7 @@ #define RTASE_HW_VER_906X_7XA 0x00800000 #define RTASE_HW_VER_906X_7XC 0x04000000 #define RTASE_HW_VER_907XD_V1 0x04800000 +#define RTASE_HW_VER_907XD_VA 0x08000000 #define RTASE_RX_DMA_BURST_256 4 #define RTASE_TX_DMA_BURST_UNLIMITED 7 diff --git a/drivers/net/ethernet/realtek/rtase/rtase_main.c b/drivers/net/ethernet/realtek/rtase/rtase_main.c index de7f11232593..6106aa5333bc 100644 --- a/drivers/net/ethernet/realtek/rtase/rtase_main.c +++ b/drivers/net/ethernet/realtek/rtase/rtase_main.c @@ -1725,6 +1725,7 @@ static int rtase_get_settings(struct net_device *dev, cmd->base.speed = SPEED_5000; break; case RTASE_HW_VER_907XD_V1: + case RTASE_HW_VER_907XD_VA: cmd->base.speed = SPEED_10000; break; } @@ -1993,6 +1994,7 @@ static int rtase_check_mac_version_valid(struct rtase_private *tp) case RTASE_HW_VER_906X_7XA: case RTASE_HW_VER_906X_7XC: case RTASE_HW_VER_907XD_V1: + case RTASE_HW_VER_907XD_VA: ret = 0; break; } diff --git a/drivers/net/mctp/mctp-i2c.c b/drivers/net/mctp/mctp-i2c.c index d2b3f5a59141..e3dcdeacc12c 100644 --- a/drivers/net/mctp/mctp-i2c.c +++ b/drivers/net/mctp/mctp-i2c.c @@ -177,8 +177,7 @@ static struct mctp_i2c_client *mctp_i2c_new_client(struct i2c_client *client) return mcli; err: if (mcli) { - if (mcli->client) - i2c_unregister_device(mcli->client); + i2c_unregister_device(mcli->client); kfree(mcli); } return ERR_PTR(rc); diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index 4ea44a2f48f7..f422a2f666ef 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -36,6 +36,7 @@ #include <linux/inet.h> #include <linux/configfs.h> #include <linux/etherdevice.h> +#include <linux/u64_stats_sync.h> #include <linux/utsname.h> #include <linux/rtnetlink.h> @@ -90,6 +91,12 @@ static DEFINE_MUTEX(target_cleanup_list_lock); */ static struct console netconsole_ext; +struct netconsole_target_stats { + u64_stats_t xmit_drop_count; + u64_stats_t enomem_count; + struct u64_stats_sync syncp; +}; + /** * struct netconsole_target - Represents a configured netconsole target. * @list: Links this target into the target_list. @@ -97,6 +104,7 @@ static struct console netconsole_ext; * @userdata_group: Links to the userdata configfs hierarchy * @userdata_complete: Cached, formatted string of append * @userdata_length: String length of userdata_complete + * @stats: Packet send stats for the target. Used for debugging. * @enabled: On / off knob to enable / disable target. * Visible from userspace (read-write). * We maintain a strict 1:1 correspondence between this and @@ -124,6 +132,7 @@ struct netconsole_target { char userdata_complete[MAX_USERDATA_ENTRY_LENGTH * MAX_USERDATA_ITEMS]; size_t userdata_length; #endif + struct netconsole_target_stats stats; bool enabled; bool extended; bool release; @@ -262,6 +271,7 @@ static void netconsole_process_cleanups_core(void) * | remote_ip * | local_mac * | remote_mac + * | transmit_errors * | userdata/ * | <key>/ * | value @@ -371,6 +381,21 @@ static ssize_t remote_mac_show(struct config_item *item, char *buf) return sysfs_emit(buf, "%pM\n", to_target(item)->np.remote_mac); } +static ssize_t transmit_errors_show(struct config_item *item, char *buf) +{ + struct netconsole_target *nt = to_target(item); + u64 xmit_drop_count, enomem_count; + unsigned int start; + + do { + start = u64_stats_fetch_begin(&nt->stats.syncp); + xmit_drop_count = u64_stats_read(&nt->stats.xmit_drop_count); + enomem_count = u64_stats_read(&nt->stats.enomem_count); + } while (u64_stats_fetch_retry(&nt->stats.syncp, start)); + + return sysfs_emit(buf, "%llu\n", xmit_drop_count + enomem_count); +} + /* * This one is special -- targets created through the configfs interface * are not enabled (and the corresponding netpoll activated) by default. @@ -842,6 +867,7 @@ CONFIGFS_ATTR(, remote_ip); CONFIGFS_ATTR_RO(, local_mac); CONFIGFS_ATTR(, remote_mac); CONFIGFS_ATTR(, release); +CONFIGFS_ATTR_RO(, transmit_errors); static struct configfs_attribute *netconsole_target_attrs[] = { &attr_enabled, @@ -854,6 +880,7 @@ static struct configfs_attribute *netconsole_target_attrs[] = { &attr_remote_ip, &attr_local_mac, &attr_remote_mac, + &attr_transmit_errors, NULL, }; @@ -1058,6 +1085,33 @@ static struct notifier_block netconsole_netdev_notifier = { .notifier_call = netconsole_netdev_event, }; +/** + * send_udp - Wrapper for netpoll_send_udp that counts errors + * @nt: target to send message to + * @msg: message to send + * @len: length of message + * + * Calls netpoll_send_udp and classifies the return value. If an error + * occurred it increments statistics in nt->stats accordingly. + * Only calls netpoll_send_udp if CONFIG_NETCONSOLE_DYNAMIC is disabled. + */ +static void send_udp(struct netconsole_target *nt, const char *msg, int len) +{ + int result = netpoll_send_udp(&nt->np, msg, len); + + if (IS_ENABLED(CONFIG_NETCONSOLE_DYNAMIC)) { + if (result == NET_XMIT_DROP) { + u64_stats_update_begin(&nt->stats.syncp); + u64_stats_inc(&nt->stats.xmit_drop_count); + u64_stats_update_end(&nt->stats.syncp); + } else if (result == -ENOMEM) { + u64_stats_update_begin(&nt->stats.syncp); + u64_stats_inc(&nt->stats.enomem_count); + u64_stats_update_end(&nt->stats.syncp); + } + } +} + static void send_msg_no_fragmentation(struct netconsole_target *nt, const char *msg, int msg_len, @@ -1085,7 +1139,7 @@ static void send_msg_no_fragmentation(struct netconsole_target *nt, MAX_PRINT_CHUNK - msg_len, "%s", userdata); - netpoll_send_udp(&nt->np, buf, msg_len); + send_udp(nt, buf, msg_len); } static void append_release(char *buf) @@ -1178,7 +1232,7 @@ static void send_fragmented_body(struct netconsole_target *nt, char *buf, this_offset += this_chunk; } - netpoll_send_udp(&nt->np, buf, this_header + this_offset); + send_udp(nt, buf, this_header + this_offset); offset += this_offset; } } @@ -1288,7 +1342,7 @@ static void write_msg(struct console *con, const char *msg, unsigned int len) tmp = msg; for (left = len; left;) { frag = min(left, MAX_PRINT_CHUNK); - netpoll_send_udp(&nt->np, tmp, frag); + send_udp(nt, tmp, frag); tmp += frag; left -= frag; } diff --git a/drivers/net/phy/bcm84881.c b/drivers/net/phy/bcm84881.c index 97da3aee4942..47405bded677 100644 --- a/drivers/net/phy/bcm84881.c +++ b/drivers/net/phy/bcm84881.c @@ -235,11 +235,21 @@ static int bcm84881_read_status(struct phy_device *phydev) return genphy_c45_read_mdix(phydev); } +/* The Broadcom BCM84881 in the Methode DM7052 is unable to provide a SGMII + * or 802.3z control word, so inband will not work. + */ +static unsigned int bcm84881_inband_caps(struct phy_device *phydev, + phy_interface_t interface) +{ + return LINK_INBAND_DISABLE; +} + static struct phy_driver bcm84881_drivers[] = { { .phy_id = 0xae025150, .phy_id_mask = 0xfffffff0, .name = "Broadcom BCM84881", + .inband_caps = bcm84881_inband_caps, .config_init = bcm84881_config_init, .probe = bcm84881_probe, .get_features = bcm84881_get_features, diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index cd50cd6a7f75..b885bc0fe6e0 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -717,6 +717,48 @@ static int marvell_config_aneg_fiber(struct phy_device *phydev) return genphy_check_and_restart_aneg(phydev, changed); } +static unsigned int m88e1111_inband_caps(struct phy_device *phydev, + phy_interface_t interface) +{ + /* In 1000base-X and SGMII modes, the inband mode can be changed + * through the Fibre page BMCR ANENABLE bit. + */ + if (interface == PHY_INTERFACE_MODE_1000BASEX || + interface == PHY_INTERFACE_MODE_SGMII) + return LINK_INBAND_DISABLE | LINK_INBAND_ENABLE | + LINK_INBAND_BYPASS; + + return 0; +} + +static int m88e1111_config_inband(struct phy_device *phydev, unsigned int modes) +{ + u16 extsr, bmcr; + int err; + + if (phydev->interface != PHY_INTERFACE_MODE_1000BASEX && + phydev->interface != PHY_INTERFACE_MODE_SGMII) + return -EINVAL; + + if (modes == LINK_INBAND_BYPASS) + extsr = MII_M1111_HWCFG_SERIAL_AN_BYPASS; + else + extsr = 0; + + if (modes == LINK_INBAND_DISABLE) + bmcr = 0; + else + bmcr = BMCR_ANENABLE; + + err = phy_modify(phydev, MII_M1111_PHY_EXT_SR, + MII_M1111_HWCFG_SERIAL_AN_BYPASS, extsr); + if (err < 0) + return extsr; + + return phy_modify_paged(phydev, MII_MARVELL_FIBER_PAGE, MII_BMCR, + BMCR_ANENABLE, bmcr); +} + static int m88e1111_config_aneg(struct phy_device *phydev) { int extsr = phy_read(phydev, MII_M1111_PHY_EXT_SR); @@ -3677,6 +3719,8 @@ static struct phy_driver marvell_drivers[] = { .name = "Marvell 88E1112", /* PHY_GBIT_FEATURES */ .probe = marvell_probe, + .inband_caps = m88e1111_inband_caps, + .config_inband = m88e1111_config_inband, .config_init = m88e1112_config_init, .config_aneg = marvell_config_aneg, .config_intr = marvell_config_intr, @@ -3698,6 +3742,8 @@ static struct phy_driver marvell_drivers[] = { /* PHY_GBIT_FEATURES */ .flags = PHY_POLL_CABLE_TEST, .probe = marvell_probe, + .inband_caps = m88e1111_inband_caps, + .config_inband = m88e1111_config_inband, .config_init = m88e1111gbe_config_init, .config_aneg = m88e1111_config_aneg, .read_status = marvell_read_status, @@ -3721,6 +3767,8 @@ static struct phy_driver marvell_drivers[] = { .name = "Marvell 88E1111 (Finisar)", /* PHY_GBIT_FEATURES */ .probe = marvell_probe, + .inband_caps = m88e1111_inband_caps, + .config_inband = m88e1111_config_inband, .config_init = m88e1111gbe_config_init, .config_aneg = m88e1111_config_aneg, .read_status = marvell_read_status, diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 0d20b534122b..0c228aa18019 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -1006,6 +1006,59 @@ static int phy_check_link_status(struct phy_device *phydev) } /** + * phy_inband_caps - query which in-band signalling modes are supported + * @phydev: a pointer to a &struct phy_device + * @interface: the interface mode for the PHY + * + * Returns zero if it is unknown what in-band signalling is supported by the + * PHY (e.g. because the PHY driver doesn't implement the method.) Otherwise, + * returns a bit mask of the LINK_INBAND_* values from + * &enum link_inband_signalling to describe which inband modes are supported + * by the PHY for this interface mode. + */ +unsigned int phy_inband_caps(struct phy_device *phydev, + phy_interface_t interface) +{ + if (phydev->drv && phydev->drv->inband_caps) + return phydev->drv->inband_caps(phydev, interface); + + return 0; +} +EXPORT_SYMBOL_GPL(phy_inband_caps); + +/** + * phy_config_inband - configure the desired PHY in-band mode + * @phydev: the phy_device struct + * @modes: in-band modes to configure + * + * Description: disables, enables or enables-with-bypass in-band signalling + * between the PHY and host system. + * + * Returns: zero on success, or negative errno value. + */ +int phy_config_inband(struct phy_device *phydev, unsigned int modes) +{ + int err; + + if (!!(modes & LINK_INBAND_DISABLE) + + !!(modes & LINK_INBAND_ENABLE) + + !!(modes & LINK_INBAND_BYPASS) != 1) + return -EINVAL; + + mutex_lock(&phydev->lock); + if (!phydev->drv) + err = -EIO; + else if (!phydev->drv->config_inband) + err = -EOPNOTSUPP; + else + err = phydev->drv->config_inband(phydev, modes); + mutex_unlock(&phydev->lock); + + return err; +} +EXPORT_SYMBOL(phy_config_inband); + +/** * _phy_start_aneg - start auto-negotiation for this PHY device * @phydev: the phy_device struct * diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index 30a654e98352..95fbc363f9a6 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -56,7 +56,8 @@ struct phylink { struct phy_device *phydev; phy_interface_t link_interface; /* PHY_INTERFACE_xxx */ u8 cfg_link_an_mode; /* MLO_AN_xxx */ - u8 cur_link_an_mode; + u8 req_link_an_mode; /* Requested MLO_AN_xxx mode */ + u8 act_link_an_mode; /* Active MLO_AN_xxx mode */ u8 link_port; /* The current non-phy ethtool port */ __ETHTOOL_DECLARE_LINK_MODE_MASK(supported); @@ -74,6 +75,7 @@ struct phylink { struct mutex state_mutex; struct phylink_link_state phy_state; + unsigned int phy_ib_mode; struct work_struct resolve; unsigned int pcs_neg_mode; unsigned int pcs_state; @@ -174,6 +176,24 @@ static const char *phylink_an_mode_str(unsigned int mode) return mode < ARRAY_SIZE(modestr) ? modestr[mode] : "unknown"; } +static const char *phylink_pcs_mode_str(unsigned int mode) +{ + if (!mode) + return "none"; + + if (mode & PHYLINK_PCS_NEG_OUTBAND) + return "outband"; + + if (mode & PHYLINK_PCS_NEG_INBAND) { + if (mode & PHYLINK_PCS_NEG_ENABLED) + return "inband,an-enabled"; + else + return "inband,an-disabled"; + } + + return "unknown"; +} + static unsigned int phylink_interface_signal_rate(phy_interface_t interface) { switch (interface) { @@ -971,6 +991,15 @@ static void phylink_resolve_an_pause(struct phylink_link_state *state) } } +static unsigned int phylink_pcs_inband_caps(struct phylink_pcs *pcs, + phy_interface_t interface) +{ + if (pcs && pcs->ops->pcs_inband_caps) + return pcs->ops->pcs_inband_caps(pcs, interface); + + return 0; +} + static void phylink_pcs_pre_config(struct phylink_pcs *pcs, phy_interface_t interface) { @@ -1024,6 +1053,24 @@ static void phylink_pcs_link_up(struct phylink_pcs *pcs, unsigned int neg_mode, pcs->ops->pcs_link_up(pcs, neg_mode, interface, speed, duplex); } +/* Query inband for a specific interface mode, asking the MAC for the + * PCS which will be used to handle the interface mode. + */ +static unsigned int phylink_inband_caps(struct phylink *pl, + phy_interface_t interface) +{ + struct phylink_pcs *pcs; + + if (!pl->mac_ops->mac_select_pcs) + return 0; + + pcs = pl->mac_ops->mac_select_pcs(pl->config, interface); + if (!pcs) + return 0; + + return phylink_pcs_inband_caps(pcs, interface); +} + static void phylink_pcs_poll_stop(struct phylink *pl) { if (pl->cfg_link_an_mode == MLO_AN_INBAND) @@ -1065,13 +1112,13 @@ static void phylink_mac_config(struct phylink *pl, phylink_dbg(pl, "%s: mode=%s/%s/%s adv=%*pb pause=%02x\n", - __func__, phylink_an_mode_str(pl->cur_link_an_mode), + __func__, phylink_an_mode_str(pl->act_link_an_mode), phy_modes(st.interface), phy_rate_matching_to_str(st.rate_matching), __ETHTOOL_LINK_MODE_MASK_NBITS, st.advertising, st.pause); - pl->mac_ops->mac_config(pl->config, pl->cur_link_an_mode, &st); + pl->mac_ops->mac_config(pl->config, pl->act_link_an_mode, &st); } static void phylink_pcs_an_restart(struct phylink *pl) @@ -1079,13 +1126,14 @@ static void phylink_pcs_an_restart(struct phylink *pl) if (pl->pcs && linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, pl->link_config.advertising) && phy_interface_mode_is_8023z(pl->link_config.interface) && - phylink_autoneg_inband(pl->cur_link_an_mode)) + phylink_autoneg_inband(pl->act_link_an_mode)) pl->pcs->ops->pcs_an_restart(pl->pcs); } /** * phylink_pcs_neg_mode() - helper to determine PCS inband mode - * @mode: one of %MLO_AN_FIXED, %MLO_AN_PHY, %MLO_AN_INBAND. + * @pl: a pointer to a &struct phylink returned from phylink_create() + * @pcs: a pointer to &struct phylink_pcs * @interface: interface mode to be used * @advertising: adertisement ethtool link mode mask * @@ -1102,11 +1150,21 @@ static void phylink_pcs_an_restart(struct phylink *pl) * Note: this is for cases where the PCS itself is involved in negotiation * (e.g. Clause 37, SGMII and similar) not Clause 73. */ -static unsigned int phylink_pcs_neg_mode(unsigned int mode, - phy_interface_t interface, - const unsigned long *advertising) +static void phylink_pcs_neg_mode(struct phylink *pl, struct phylink_pcs *pcs, + phy_interface_t interface, + const unsigned long *advertising) { - unsigned int neg_mode; + unsigned int pcs_ib_caps = 0; + unsigned int phy_ib_caps = 0; + unsigned int neg_mode, mode; + enum { + INBAND_CISCO_SGMII, + INBAND_BASEX, + } type; + + mode = pl->req_link_an_mode; + + pl->phy_ib_mode = 0; switch (interface) { case PHY_INTERFACE_MODE_SGMII: @@ -1119,10 +1177,7 @@ static unsigned int phylink_pcs_neg_mode(unsigned int mode, * inband communication. Note: there exist PHYs that run * with SGMII but do not send the inband data. */ - if (!phylink_autoneg_inband(mode)) - neg_mode = PHYLINK_PCS_NEG_OUTBAND; - else - neg_mode = PHYLINK_PCS_NEG_INBAND_ENABLED; + type = INBAND_CISCO_SGMII; break; case PHY_INTERFACE_MODE_1000BASEX: @@ -1133,21 +1188,143 @@ static unsigned int phylink_pcs_neg_mode(unsigned int mode, * as well, but drivers may not support this, so may * need to override this. */ - if (!phylink_autoneg_inband(mode)) + type = INBAND_BASEX; + break; + + default: + pl->pcs_neg_mode = PHYLINK_PCS_NEG_NONE; + pl->act_link_an_mode = mode; + return; + } + + if (pcs) + pcs_ib_caps = phylink_pcs_inband_caps(pcs, interface); + + if (pl->phydev) + phy_ib_caps = phy_inband_caps(pl->phydev, interface); + + phylink_dbg(pl, "interface %s inband modes: pcs=%02x phy=%02x\n", + phy_modes(interface), pcs_ib_caps, phy_ib_caps); + + if (!phylink_autoneg_inband(mode)) { + bool pcs_ib_only = false; + bool phy_ib_only = false; + + if (pcs_ib_caps && pcs_ib_caps != LINK_INBAND_DISABLE) { + /* PCS supports reporting in-band capabilities, and + * supports more than disable mode. + */ + if (pcs_ib_caps & LINK_INBAND_DISABLE) + neg_mode = PHYLINK_PCS_NEG_OUTBAND; + else if (pcs_ib_caps & LINK_INBAND_ENABLE) + pcs_ib_only = true; + } + + if (phy_ib_caps && phy_ib_caps != LINK_INBAND_DISABLE) { + /* PHY supports in-band capabilities, and supports + * more than disable mode. + */ + if (phy_ib_caps & LINK_INBAND_DISABLE) + pl->phy_ib_mode = LINK_INBAND_DISABLE; + else if (phy_ib_caps & LINK_INBAND_BYPASS) + pl->phy_ib_mode = LINK_INBAND_BYPASS; + else if (phy_ib_caps & LINK_INBAND_ENABLE) + phy_ib_only = true; + } + + /* If either the PCS or PHY requires inband to be enabled, + * this is an invalid configuration. Provide a diagnostic + * message for this case, but don't try to force the issue. + */ + if (pcs_ib_only || phy_ib_only) + phylink_warn(pl, + "firmware wants %s mode, but %s%s%s requires inband\n", + phylink_an_mode_str(mode), + pcs_ib_only ? "PCS" : "", + pcs_ib_only && phy_ib_only ? " and " : "", + phy_ib_only ? "PHY" : ""); + + neg_mode = PHYLINK_PCS_NEG_OUTBAND; + } else if (type == INBAND_CISCO_SGMII || pl->phydev) { + /* For SGMII modes which are designed to be used with PHYs, or + * Base-X with a PHY, we try to use in-band mode where-ever + * possible. However, there are some PHYs e.g. BCM84881 which + * do not support in-band. + */ + const unsigned int inband_ok = LINK_INBAND_ENABLE | + LINK_INBAND_BYPASS; + const unsigned int outband_ok = LINK_INBAND_DISABLE | + LINK_INBAND_BYPASS; + /* PCS PHY + * D E D E + * 0 0 0 0 no information inband enabled + * 1 0 0 0 pcs doesn't support outband + * 0 1 0 0 pcs required inband enabled + * 1 1 0 0 pcs optional inband enabled + * 0 0 1 0 phy doesn't support outband + * 1 0 1 0 pcs+phy doesn't support outband + * 0 1 1 0 pcs required, phy doesn't support, invalid + * 1 1 1 0 pcs optional, phy doesn't support, outband + * 0 0 0 1 phy required inband enabled + * 1 0 0 1 pcs doesn't support, phy required, invalid + * 0 1 0 1 pcs+phy required inband enabled + * 1 1 0 1 pcs optional, phy required inband enabled + * 0 0 1 1 phy optional inband enabled + * 1 0 1 1 pcs doesn't support, phy optional, outband + * 0 1 1 1 pcs required, phy optional inband enabled + * 1 1 1 1 pcs+phy optional inband enabled + */ + if ((!pcs_ib_caps || pcs_ib_caps & inband_ok) && + (!phy_ib_caps || phy_ib_caps & inband_ok)) { + /* In-band supported or unknown at both ends. Enable + * in-band mode with or without bypass at the PHY. + */ + if (phy_ib_caps & LINK_INBAND_ENABLE) + pl->phy_ib_mode = LINK_INBAND_ENABLE; + else if (phy_ib_caps & LINK_INBAND_BYPASS) + pl->phy_ib_mode = LINK_INBAND_BYPASS; + + neg_mode = PHYLINK_PCS_NEG_INBAND_ENABLED; + } else if ((!pcs_ib_caps || pcs_ib_caps & outband_ok) && + (!phy_ib_caps || phy_ib_caps & outband_ok)) { + /* Either in-band not supported at at least one end. + * In-band bypass at the other end is possible. + */ + if (phy_ib_caps & LINK_INBAND_DISABLE) + pl->phy_ib_mode = LINK_INBAND_DISABLE; + else if (phy_ib_caps & LINK_INBAND_BYPASS) + pl->phy_ib_mode = LINK_INBAND_BYPASS; + neg_mode = PHYLINK_PCS_NEG_OUTBAND; + if (pl->phydev) + mode = MLO_AN_PHY; + } else { + /* invalid */ + phylink_warn(pl, "%s: incompatible in-band capabilities, trying in-band", + phy_modes(interface)); + neg_mode = PHYLINK_PCS_NEG_INBAND_ENABLED; + } + } else { + /* For Base-X without a PHY */ + if (pcs_ib_caps == LINK_INBAND_DISABLE) + /* If the PCS doesn't support inband, then inband must + * be disabled. + */ + neg_mode = PHYLINK_PCS_NEG_INBAND_DISABLED; + else if (pcs_ib_caps == LINK_INBAND_ENABLE) + /* If the PCS requires inband, then inband must always + * be enabled. + */ + neg_mode = PHYLINK_PCS_NEG_INBAND_ENABLED; else if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, advertising)) neg_mode = PHYLINK_PCS_NEG_INBAND_ENABLED; else neg_mode = PHYLINK_PCS_NEG_INBAND_DISABLED; - break; - - default: - neg_mode = PHYLINK_PCS_NEG_NONE; - break; } - return neg_mode; + pl->pcs_neg_mode = neg_mode; + pl->act_link_an_mode = mode; } static void phylink_major_config(struct phylink *pl, bool restart, @@ -1159,11 +1336,9 @@ static void phylink_major_config(struct phylink *pl, bool restart, unsigned int neg_mode; int err; - phylink_dbg(pl, "major config %s\n", phy_modes(state->interface)); - - pl->pcs_neg_mode = phylink_pcs_neg_mode(pl->cur_link_an_mode, - state->interface, - state->advertising); + phylink_dbg(pl, "major config, requested %s/%s\n", + phylink_an_mode_str(pl->req_link_an_mode), + phy_modes(state->interface)); if (pl->mac_ops->mac_select_pcs) { pcs = pl->mac_ops->mac_select_pcs(pl->config, state->interface); @@ -1177,10 +1352,17 @@ static void phylink_major_config(struct phylink *pl, bool restart, pcs_changed = pl->pcs != pcs; } + phylink_pcs_neg_mode(pl, pcs, state->interface, state->advertising); + + phylink_dbg(pl, "major config, active %s/%s/%s\n", + phylink_an_mode_str(pl->act_link_an_mode), + phylink_pcs_mode_str(pl->pcs_neg_mode), + phy_modes(state->interface)); + phylink_pcs_poll_stop(pl); if (pl->mac_ops->mac_prepare) { - err = pl->mac_ops->mac_prepare(pl->config, pl->cur_link_an_mode, + err = pl->mac_ops->mac_prepare(pl->config, pl->act_link_an_mode, state->interface); if (err < 0) { phylink_err(pl, "mac_prepare failed: %pe\n", @@ -1214,7 +1396,7 @@ static void phylink_major_config(struct phylink *pl, bool restart, if (pl->pcs_state == PCS_STATE_STARTING || pcs_changed) phylink_pcs_enable(pl->pcs); - neg_mode = pl->cur_link_an_mode; + neg_mode = pl->act_link_an_mode; if (pl->pcs && pl->pcs->neg_mode) neg_mode = pl->pcs_neg_mode; @@ -1230,13 +1412,20 @@ static void phylink_major_config(struct phylink *pl, bool restart, phylink_pcs_an_restart(pl); if (pl->mac_ops->mac_finish) { - err = pl->mac_ops->mac_finish(pl->config, pl->cur_link_an_mode, + err = pl->mac_ops->mac_finish(pl->config, pl->act_link_an_mode, state->interface); if (err < 0) phylink_err(pl, "mac_finish failed: %pe\n", ERR_PTR(err)); } + if (pl->phydev && pl->phy_ib_mode) { + err = phy_config_inband(pl->phydev, pl->phy_ib_mode); + if (err < 0) + phylink_err(pl, "phy_config_inband: %pe\n", + ERR_PTR(err)); + } + if (pl->sfp_bus) { rate_kbd = phylink_interface_signal_rate(state->interface); if (rate_kbd) @@ -1261,17 +1450,16 @@ static int phylink_change_inband_advert(struct phylink *pl) return 0; phylink_dbg(pl, "%s: mode=%s/%s adv=%*pb pause=%02x\n", __func__, - phylink_an_mode_str(pl->cur_link_an_mode), + phylink_an_mode_str(pl->req_link_an_mode), phy_modes(pl->link_config.interface), __ETHTOOL_LINK_MODE_MASK_NBITS, pl->link_config.advertising, pl->link_config.pause); /* Recompute the PCS neg mode */ - pl->pcs_neg_mode = phylink_pcs_neg_mode(pl->cur_link_an_mode, - pl->link_config.interface, - pl->link_config.advertising); + phylink_pcs_neg_mode(pl, pl->pcs, pl->link_config.interface, + pl->link_config.advertising); - neg_mode = pl->cur_link_an_mode; + neg_mode = pl->act_link_an_mode; if (pl->pcs->neg_mode) neg_mode = pl->pcs_neg_mode; @@ -1336,7 +1524,7 @@ static void phylink_mac_initial_config(struct phylink *pl, bool force_restart) { struct phylink_link_state link_state; - switch (pl->cur_link_an_mode) { + switch (pl->req_link_an_mode) { case MLO_AN_PHY: link_state = pl->phy_state; break; @@ -1410,14 +1598,14 @@ static void phylink_link_up(struct phylink *pl, pl->cur_interface = link_state.interface; - neg_mode = pl->cur_link_an_mode; + neg_mode = pl->act_link_an_mode; if (pl->pcs && pl->pcs->neg_mode) neg_mode = pl->pcs_neg_mode; phylink_pcs_link_up(pl->pcs, neg_mode, pl->cur_interface, speed, duplex); - pl->mac_ops->mac_link_up(pl->config, pl->phydev, pl->cur_link_an_mode, + pl->mac_ops->mac_link_up(pl->config, pl->phydev, pl->act_link_an_mode, pl->cur_interface, speed, duplex, !!(link_state.pause & MLO_PAUSE_TX), rx_pause); @@ -1437,7 +1625,7 @@ static void phylink_link_down(struct phylink *pl) if (ndev) netif_carrier_off(ndev); - pl->mac_ops->mac_link_down(pl->config, pl->cur_link_an_mode, + pl->mac_ops->mac_link_down(pl->config, pl->act_link_an_mode, pl->cur_interface); phylink_info(pl, "Link is Down\n"); } @@ -1463,10 +1651,10 @@ static void phylink_resolve(struct work_struct *w) } else if (pl->link_failed) { link_state.link = false; retrigger = true; - } else if (pl->cur_link_an_mode == MLO_AN_FIXED) { + } else if (pl->act_link_an_mode == MLO_AN_FIXED) { phylink_get_fixed_state(pl, &link_state); mac_config = link_state.link; - } else if (pl->cur_link_an_mode == MLO_AN_PHY) { + } else if (pl->act_link_an_mode == MLO_AN_PHY) { link_state = pl->phy_state; mac_config = link_state.link; } else { @@ -1520,7 +1708,7 @@ static void phylink_resolve(struct work_struct *w) } } - if (pl->cur_link_an_mode != MLO_AN_FIXED) + if (pl->act_link_an_mode != MLO_AN_FIXED) phylink_apply_manual_flow(pl, &link_state); if (mac_config) { @@ -1644,7 +1832,7 @@ int phylink_set_fixed_link(struct phylink *pl, pl->link_config.an_complete = 1; pl->cfg_link_an_mode = MLO_AN_FIXED; - pl->cur_link_an_mode = pl->cfg_link_an_mode; + pl->req_link_an_mode = pl->cfg_link_an_mode; return 0; } @@ -1732,7 +1920,7 @@ struct phylink *phylink_create(struct phylink_config *config, } } - pl->cur_link_an_mode = pl->cfg_link_an_mode; + pl->req_link_an_mode = pl->cfg_link_an_mode; ret = phylink_register_sfp(pl, fwnode); if (ret < 0) { @@ -2189,7 +2377,7 @@ void phylink_start(struct phylink *pl) ASSERT_RTNL(); phylink_info(pl, "configuring for %s/%s link mode\n", - phylink_an_mode_str(pl->cur_link_an_mode), + phylink_an_mode_str(pl->req_link_an_mode), phy_modes(pl->link_config.interface)); /* Always set the carrier off */ @@ -2474,7 +2662,7 @@ int phylink_ethtool_ksettings_get(struct phylink *pl, linkmode_copy(kset->link_modes.supported, pl->supported); - switch (pl->cur_link_an_mode) { + switch (pl->act_link_an_mode) { case MLO_AN_FIXED: /* We are using fixed settings. Report these as the * current link settings - and note that these also @@ -2505,6 +2693,26 @@ int phylink_ethtool_ksettings_get(struct phylink *pl, } EXPORT_SYMBOL_GPL(phylink_ethtool_ksettings_get); +static bool phylink_validate_pcs_inband_autoneg(struct phylink *pl, + phy_interface_t interface, + unsigned long *adv) +{ + unsigned int inband = phylink_inband_caps(pl, interface); + unsigned int mask; + + /* If the PCS doesn't implement inband support, be permissive. */ + if (!inband) + return true; + + if (linkmode_test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, adv)) + mask = LINK_INBAND_ENABLE; + else + mask = LINK_INBAND_DISABLE; + + /* Check whether the PCS implements the required mode */ + return !!(inband & mask); +} + /** * phylink_ethtool_ksettings_set() - set the link settings * @pl: a pointer to a &struct phylink returned from phylink_create() @@ -2566,7 +2774,7 @@ int phylink_ethtool_ksettings_set(struct phylink *pl, /* If we have a fixed link, refuse to change link parameters. * If the link parameters match, accept them but do nothing. */ - if (pl->cur_link_an_mode == MLO_AN_FIXED) { + if (pl->req_link_an_mode == MLO_AN_FIXED) { if (s->speed != pl->link_config.speed || s->duplex != pl->link_config.duplex) return -EINVAL; @@ -2582,7 +2790,7 @@ int phylink_ethtool_ksettings_set(struct phylink *pl, * is our default case) but do not allow the advertisement to * be changed. If the advertisement matches, simply return. */ - if (pl->cur_link_an_mode == MLO_AN_FIXED) { + if (pl->req_link_an_mode == MLO_AN_FIXED) { if (!linkmode_equal(config.advertising, pl->link_config.advertising)) return -EINVAL; @@ -2617,7 +2825,7 @@ int phylink_ethtool_ksettings_set(struct phylink *pl, linkmode_copy(support, pl->supported); if (phylink_validate(pl, support, &config)) { phylink_err(pl, "validation of %s/%s with support %*pb failed\n", - phylink_an_mode_str(pl->cur_link_an_mode), + phylink_an_mode_str(pl->req_link_an_mode), phy_modes(config.interface), __ETHTOOL_LINK_MODE_MASK_NBITS, support); return -EINVAL; @@ -2635,6 +2843,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl, phylink_is_empty_linkmode(config.advertising)) return -EINVAL; + /* Validate the autonegotiation state. We don't have a PHY in this + * situation, so the PCS is the media-facing entity. + */ + if (!phylink_validate_pcs_inband_autoneg(pl, config.interface, + config.advertising)) + return -EINVAL; + mutex_lock(&pl->state_mutex); pl->link_config.speed = config.speed; pl->link_config.duplex = config.duplex; @@ -2717,7 +2932,7 @@ int phylink_ethtool_set_pauseparam(struct phylink *pl, ASSERT_RTNL(); - if (pl->cur_link_an_mode == MLO_AN_FIXED) + if (pl->req_link_an_mode == MLO_AN_FIXED) return -EOPNOTSUPP; if (!phylink_test(pl->supported, Pause) && @@ -2981,7 +3196,7 @@ static int phylink_mii_read(struct phylink *pl, unsigned int phy_id, struct phylink_link_state state; int val = 0xffff; - switch (pl->cur_link_an_mode) { + switch (pl->act_link_an_mode) { case MLO_AN_FIXED: if (phy_id == 0) { phylink_get_fixed_state(pl, &state); @@ -3006,7 +3221,7 @@ static int phylink_mii_read(struct phylink *pl, unsigned int phy_id, static int phylink_mii_write(struct phylink *pl, unsigned int phy_id, unsigned int reg, unsigned int val) { - switch (pl->cur_link_an_mode) { + switch (pl->act_link_an_mode) { case MLO_AN_FIXED: break; @@ -3176,10 +3391,11 @@ static phy_interface_t phylink_choose_sfp_interface(struct phylink *pl, return interface; } -static void phylink_sfp_set_config(struct phylink *pl, u8 mode, +static void phylink_sfp_set_config(struct phylink *pl, unsigned long *supported, struct phylink_link_state *state) { + u8 mode = MLO_AN_INBAND; bool changed = false; phylink_dbg(pl, "requesting link mode %s/%s with support %*pb\n", @@ -3196,9 +3412,9 @@ static void phylink_sfp_set_config(struct phylink *pl, u8 mode, changed = true; } - if (pl->cur_link_an_mode != mode || + if (pl->req_link_an_mode != mode || pl->link_config.interface != state->interface) { - pl->cur_link_an_mode = mode; + pl->req_link_an_mode = mode; pl->link_config.interface = state->interface; changed = true; @@ -3213,8 +3429,7 @@ static void phylink_sfp_set_config(struct phylink *pl, u8 mode, phylink_mac_initial_config(pl, false); } -static int phylink_sfp_config_phy(struct phylink *pl, u8 mode, - struct phy_device *phy) +static int phylink_sfp_config_phy(struct phylink *pl, struct phy_device *phy) { __ETHTOOL_DECLARE_LINK_MODE_MASK(support); struct phylink_link_state config; @@ -3258,7 +3473,7 @@ static int phylink_sfp_config_phy(struct phylink *pl, u8 mode, pl->link_port = pl->sfp_port; - phylink_sfp_set_config(pl, mode, support, &config); + phylink_sfp_set_config(pl, support, &config); return 0; } @@ -3314,6 +3529,12 @@ static int phylink_sfp_config_optical(struct phylink *pl) phylink_dbg(pl, "optical SFP: chosen %s interface\n", phy_modes(interface)); + if (!phylink_validate_pcs_inband_autoneg(pl, interface, + config.advertising)) { + phylink_err(pl, "autoneg setting not compatible with PCS"); + return -EINVAL; + } + config.interface = interface; /* Ignore errors if we're expecting a PHY to attach later */ @@ -3327,7 +3548,7 @@ static int phylink_sfp_config_optical(struct phylink *pl) pl->link_port = pl->sfp_port; - phylink_sfp_set_config(pl, MLO_AN_INBAND, pl->sfp_support, &config); + phylink_sfp_set_config(pl, pl->sfp_support, &config); return 0; } @@ -3398,19 +3619,9 @@ static void phylink_sfp_link_up(void *upstream) phylink_enable_and_run_resolve(pl, PHYLINK_DISABLE_LINK); } -/* The Broadcom BCM84881 in the Methode DM7052 is unable to provide a SGMII - * or 802.3z control word, so inband will not work. - */ -static bool phylink_phy_no_inband(struct phy_device *phy) -{ - return phy->is_c45 && phy_id_compare(phy->c45_ids.device_ids[1], - 0xae025150, 0xfffffff0); -} - static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy) { struct phylink *pl = upstream; - u8 mode; /* * This is the new way of dealing with flow control for PHYs, @@ -3421,17 +3632,12 @@ static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy) */ phy_support_asym_pause(phy); - if (phylink_phy_no_inband(phy)) - mode = MLO_AN_PHY; - else - mode = MLO_AN_INBAND; - /* Set the PHY's host supported interfaces */ phy_interface_and(phy->host_interfaces, phylink_sfp_interfaces, pl->config->supported_interfaces); /* Do the initial configuration */ - return phylink_sfp_config_phy(pl, mode, phy); + return phylink_sfp_config_phy(pl, phy); } static void phylink_sfp_disconnect_phy(void *upstream, diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index b34301650c47..f91e50a76efd 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -57,7 +57,7 @@ static inline void netpoll_poll_disable(struct net_device *dev) { return; } static inline void netpoll_poll_enable(struct net_device *dev) { return; } #endif -void netpoll_send_udp(struct netpoll *np, const char *msg, int len); +int netpoll_send_udp(struct netpoll *np, const char *msg, int len); void netpoll_print_options(struct netpoll *np); int netpoll_parse_options(struct netpoll *np, char *opt); int __netpoll_setup(struct netpoll *np, struct net_device *ndev); diff --git a/include/linux/phy.h b/include/linux/phy.h index 563c46205685..61a1bc81f597 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -818,6 +818,24 @@ struct phy_tdr_config { #define PHY_PAIR_ALL -1 /** + * enum link_inband_signalling - in-band signalling modes that are supported + * + * @LINK_INBAND_DISABLE: in-band signalling can be disabled + * @LINK_INBAND_ENABLE: in-band signalling can be enabled without bypass + * @LINK_INBAND_BYPASS: in-band signalling can be enabled with bypass + * + * The possible and required bits can only be used if the valid bit is set. + * If possible is clear, that means inband signalling can not be used. + * Required is only valid when possible is set, and means that inband + * signalling must be used. + */ +enum link_inband_signalling { + LINK_INBAND_DISABLE = BIT(0), + LINK_INBAND_ENABLE = BIT(1), + LINK_INBAND_BYPASS = BIT(2), +}; + +/** * struct phy_plca_cfg - Configuration of the PLCA (Physical Layer Collision * Avoidance) Reconciliation Sublayer. * @@ -957,6 +975,19 @@ struct phy_driver { int (*get_features)(struct phy_device *phydev); /** + * @inband_caps: query whether in-band is supported for the given PHY + * interface mode. Returns a bitmask of bits defined by enum + * link_inband_signalling. + */ + unsigned int (*inband_caps)(struct phy_device *phydev, + phy_interface_t interface); + + /** + * @config_inband: configure in-band mode for the PHY + */ + int (*config_inband)(struct phy_device *phydev, unsigned int modes); + + /** * @get_rate_matching: Get the supported type of rate matching for a * particular phy interface. This is used by phy consumers to determine * whether to advertise lower-speed modes for that interface. It is @@ -1818,6 +1849,9 @@ int phy_config_aneg(struct phy_device *phydev); int _phy_start_aneg(struct phy_device *phydev); int phy_start_aneg(struct phy_device *phydev); int phy_aneg_done(struct phy_device *phydev); +unsigned int phy_inband_caps(struct phy_device *phydev, + phy_interface_t interface); +int phy_config_inband(struct phy_device *phydev, unsigned int modes); int phy_speed_down(struct phy_device *phydev, bool sync); int phy_speed_up(struct phy_device *phydev); bool phy_check_valid(int speed, int duplex, unsigned long *features); diff --git a/include/linux/phylink.h b/include/linux/phylink.h index 5c01048860c4..5462cc6a37dc 100644 --- a/include/linux/phylink.h +++ b/include/linux/phylink.h @@ -419,6 +419,7 @@ struct phylink_pcs { /** * struct phylink_pcs_ops - MAC PCS operations structure. * @pcs_validate: validate the link configuration. + * @pcs_inband_caps: query inband support for interface mode. * @pcs_enable: enable the PCS. * @pcs_disable: disable the PCS. * @pcs_pre_config: pre-mac_config method (for errata) @@ -434,6 +435,8 @@ struct phylink_pcs { struct phylink_pcs_ops { int (*pcs_validate)(struct phylink_pcs *pcs, unsigned long *supported, const struct phylink_link_state *state); + unsigned int (*pcs_inband_caps)(struct phylink_pcs *pcs, + phy_interface_t interface); int (*pcs_enable)(struct phylink_pcs *pcs); void (*pcs_disable)(struct phylink_pcs *pcs); void (*pcs_pre_config)(struct phylink_pcs *pcs, @@ -471,6 +474,20 @@ int pcs_validate(struct phylink_pcs *pcs, unsigned long *supported, const struct phylink_link_state *state); /** + * pcs_inband_caps - query PCS in-band capabilities for interface mode. + * @pcs: a pointer to a &struct phylink_pcs. + * @interface: interface mode to be queried + * + * Returns zero if it is unknown what in-band signalling is supported by the + * PHY (e.g. because the PHY driver doesn't implement the method.) Otherwise, + * returns a bit mask of the LINK_INBAND_* values from + * &enum link_inband_signalling to describe which inband modes are supported + * for this interface mode. + */ +unsigned int pcs_inband_caps(struct phylink_pcs *pcs, + phy_interface_t interface); + +/** * pcs_enable() - enable the PCS. * @pcs: a pointer to a &struct phylink_pcs. */ diff --git a/include/net/dst.h b/include/net/dst.h index 0f303cc60252..08647c99d79c 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -440,6 +440,15 @@ static inline void dst_set_expires(struct dst_entry *dst, int timeout) dst->expires = expires; } +static inline unsigned int dst_dev_overhead(struct dst_entry *dst, + struct sk_buff *skb) +{ + if (likely(dst)) + return LL_RESERVED_SPACE(dst->dev); + + return skb->mac_len; +} + INDIRECT_CALLABLE_DECLARE(int ip6_output(struct net *, struct sock *, struct sk_buff *)); INDIRECT_CALLABLE_DECLARE(int ip_output(struct net *, struct sock *, diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 2e459b9d88eb..6f2647b000b8 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -390,7 +390,7 @@ netdev_tx_t netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) } EXPORT_SYMBOL(netpoll_send_skb); -void netpoll_send_udp(struct netpoll *np, const char *msg, int len) +int netpoll_send_udp(struct netpoll *np, const char *msg, int len) { int total_len, ip_len, udp_len; struct sk_buff *skb; @@ -414,7 +414,7 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) skb = find_skb(np, total_len + np->dev->needed_tailroom, total_len - len); if (!skb) - return; + return -ENOMEM; skb_copy_to_linear_data(skb, msg, len); skb_put(skb, len); @@ -490,7 +490,7 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) skb->dev = np->dev; - netpoll_send_skb(np, skb); + return (int)netpoll_send_skb(np, skb); } EXPORT_SYMBOL(netpoll_send_udp); @@ -634,7 +634,8 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) goto out; } - if (!rcu_access_pointer(ndev->npinfo)) { + npinfo = rtnl_dereference(ndev->npinfo); + if (!npinfo) { npinfo = kmalloc(sizeof(*npinfo), GFP_KERNEL); if (!npinfo) { err = -ENOMEM; @@ -654,7 +655,6 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) goto free_npinfo; } } else { - npinfo = rtnl_dereference(ndev->npinfo); refcount_inc(&npinfo->refcnt); } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 0065b1996c94..a59204a8d850 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1169,7 +1169,10 @@ alloc_new_skb: /* [!] NOTE: copy will be negative if pagedlen>0 * because then the equation reduces to -fraggap. */ - if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, fraggap, skb) < 0) { + if (copy > 0 && + INDIRECT_CALL_1(getfrag, ip_generic_getfrag, + from, data + transhdrlen, offset, + copy, fraggap, skb) < 0) { err = -EFAULT; kfree_skb(skb); goto error; @@ -1213,8 +1216,9 @@ alloc_new_skb: unsigned int off; off = skb->len; - if (getfrag(from, skb_put(skb, copy), - offset, copy, off, skb) < 0) { + if (INDIRECT_CALL_1(getfrag, ip_generic_getfrag, + from, skb_put(skb, copy), + offset, copy, off, skb) < 0) { __skb_trim(skb, off); err = -EFAULT; goto error; @@ -1252,7 +1256,8 @@ alloc_new_skb: get_page(pfrag->page); } copy = min_t(int, copy, pfrag->size - pfrag->offset); - if (getfrag(from, + if (INDIRECT_CALL_1(getfrag, ip_generic_getfrag, + from, page_address(pfrag->page) + pfrag->offset, offset, copy, skb->len, skb) < 0) goto error_efault; diff --git a/net/ipv6/ioam6_iptunnel.c b/net/ipv6/ioam6_iptunnel.c index 9d8422e350f8..28e5a89dc255 100644 --- a/net/ipv6/ioam6_iptunnel.c +++ b/net/ipv6/ioam6_iptunnel.c @@ -253,14 +253,15 @@ static int ioam6_do_fill(struct net *net, struct sk_buff *skb) } static int ioam6_do_inline(struct net *net, struct sk_buff *skb, - struct ioam6_lwt_encap *tuninfo) + struct ioam6_lwt_encap *tuninfo, + struct dst_entry *cache_dst) { struct ipv6hdr *oldhdr, *hdr; int hdrlen, err; hdrlen = (tuninfo->eh.hdrlen + 1) << 3; - err = skb_cow_head(skb, hdrlen + skb->mac_len); + err = skb_cow_head(skb, hdrlen + dst_dev_overhead(cache_dst, skb)); if (unlikely(err)) return err; @@ -291,7 +292,8 @@ static int ioam6_do_encap(struct net *net, struct sk_buff *skb, struct ioam6_lwt_encap *tuninfo, bool has_tunsrc, struct in6_addr *tunsrc, - struct in6_addr *tundst) + struct in6_addr *tundst, + struct dst_entry *cache_dst) { struct dst_entry *dst = skb_dst(skb); struct ipv6hdr *hdr, *inner_hdr; @@ -300,7 +302,7 @@ static int ioam6_do_encap(struct net *net, struct sk_buff *skb, hdrlen = (tuninfo->eh.hdrlen + 1) << 3; len = sizeof(*hdr) + hdrlen; - err = skb_cow_head(skb, len + skb->mac_len); + err = skb_cow_head(skb, len + dst_dev_overhead(cache_dst, skb)); if (unlikely(err)) return err; @@ -334,7 +336,7 @@ static int ioam6_do_encap(struct net *net, struct sk_buff *skb, static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb) { - struct dst_entry *dst = skb_dst(skb); + struct dst_entry *dst = skb_dst(skb), *cache_dst; struct in6_addr orig_daddr; struct ioam6_lwt *ilwt; int err = -EINVAL; @@ -352,6 +354,10 @@ static int ioam6_output(struct net *net, struct sock *sk, struct sk_buff *skb) orig_daddr = ipv6_hdr(skb)->daddr; + local_bh_disable(); + cache_dst = dst_cache_get(&ilwt->cache); + local_bh_enable(); + switch (ilwt->mode) { case IOAM6_IPTUNNEL_MODE_INLINE: do_inline: @@ -359,7 +365,7 @@ do_inline: if (ipv6_hdr(skb)->nexthdr == NEXTHDR_HOP) goto out; - err = ioam6_do_inline(net, skb, &ilwt->tuninfo); + err = ioam6_do_inline(net, skb, &ilwt->tuninfo, cache_dst); if (unlikely(err)) goto drop; @@ -369,7 +375,7 @@ do_encap: /* Encapsulation (ip6ip6) */ err = ioam6_do_encap(net, skb, &ilwt->tuninfo, ilwt->has_tunsrc, &ilwt->tunsrc, - &ilwt->tundst); + &ilwt->tundst, cache_dst); if (unlikely(err)) goto drop; @@ -387,41 +393,36 @@ do_encap: goto drop; } - err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); - if (unlikely(err)) - goto drop; + if (unlikely(!cache_dst)) { + struct ipv6hdr *hdr = ipv6_hdr(skb); + struct flowi6 fl6; + + memset(&fl6, 0, sizeof(fl6)); + fl6.daddr = hdr->daddr; + fl6.saddr = hdr->saddr; + fl6.flowlabel = ip6_flowinfo(hdr); + fl6.flowi6_mark = skb->mark; + fl6.flowi6_proto = hdr->nexthdr; + + cache_dst = ip6_route_output(net, NULL, &fl6); + if (cache_dst->error) { + err = cache_dst->error; + dst_release(cache_dst); + goto drop; + } - if (!ipv6_addr_equal(&orig_daddr, &ipv6_hdr(skb)->daddr)) { local_bh_disable(); - dst = dst_cache_get(&ilwt->cache); + dst_cache_set_ip6(&ilwt->cache, cache_dst, &fl6.saddr); local_bh_enable(); - if (unlikely(!dst)) { - struct ipv6hdr *hdr = ipv6_hdr(skb); - struct flowi6 fl6; - - memset(&fl6, 0, sizeof(fl6)); - fl6.daddr = hdr->daddr; - fl6.saddr = hdr->saddr; - fl6.flowlabel = ip6_flowinfo(hdr); - fl6.flowi6_mark = skb->mark; - fl6.flowi6_proto = hdr->nexthdr; - - dst = ip6_route_output(net, NULL, &fl6); - if (dst->error) { - err = dst->error; - dst_release(dst); - goto drop; - } - - local_bh_disable(); - dst_cache_set_ip6(&ilwt->cache, dst, &fl6.saddr); - local_bh_enable(); - } + err = skb_cow_head(skb, LL_RESERVED_SPACE(cache_dst->dev)); + if (unlikely(err)) + goto drop; + } + if (!ipv6_addr_equal(&orig_daddr, &ipv6_hdr(skb)->daddr)) { skb_dst_drop(skb); - skb_dst_set(skb, dst); - + skb_dst_set(skb, cache_dst); return dst_output(net, sk, skb); } out: diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index f7b4608bb316..3d672dea9f56 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1697,8 +1697,9 @@ alloc_new_skb: pskb_trim_unique(skb_prev, maxfraglen); } if (copy > 0 && - getfrag(from, data + transhdrlen, offset, - copy, fraggap, skb) < 0) { + INDIRECT_CALL_1(getfrag, ip_generic_getfrag, + from, data + transhdrlen, offset, + copy, fraggap, skb) < 0) { err = -EFAULT; kfree_skb(skb); goto error; @@ -1742,8 +1743,9 @@ alloc_new_skb: unsigned int off; off = skb->len; - if (getfrag(from, skb_put(skb, copy), - offset, copy, off, skb) < 0) { + if (INDIRECT_CALL_1(getfrag, ip_generic_getfrag, + from, skb_put(skb, copy), + offset, copy, off, skb) < 0) { __skb_trim(skb, off); err = -EFAULT; goto error; @@ -1781,7 +1783,8 @@ alloc_new_skb: get_page(pfrag->page); } copy = min_t(int, copy, pfrag->size - pfrag->offset); - if (getfrag(from, + if (INDIRECT_CALL_1(getfrag, ip_generic_getfrag, + from, page_address(pfrag->page) + pfrag->offset, offset, copy, skb->len, skb) < 0) goto error_efault; diff --git a/net/ipv6/rpl_iptunnel.c b/net/ipv6/rpl_iptunnel.c index db3c19a42e1c..7ba22d2f2bfe 100644 --- a/net/ipv6/rpl_iptunnel.c +++ b/net/ipv6/rpl_iptunnel.c @@ -125,7 +125,8 @@ static void rpl_destroy_state(struct lwtunnel_state *lwt) } static int rpl_do_srh_inline(struct sk_buff *skb, const struct rpl_lwt *rlwt, - const struct ipv6_rpl_sr_hdr *srh) + const struct ipv6_rpl_sr_hdr *srh, + struct dst_entry *cache_dst) { struct ipv6_rpl_sr_hdr *isrh, *csrh; const struct ipv6hdr *oldhdr; @@ -153,7 +154,7 @@ static int rpl_do_srh_inline(struct sk_buff *skb, const struct rpl_lwt *rlwt, hdrlen = ((csrh->hdrlen + 1) << 3); - err = skb_cow_head(skb, hdrlen + skb->mac_len); + err = skb_cow_head(skb, hdrlen + dst_dev_overhead(cache_dst, skb)); if (unlikely(err)) { kfree(buf); return err; @@ -186,7 +187,8 @@ static int rpl_do_srh_inline(struct sk_buff *skb, const struct rpl_lwt *rlwt, return 0; } -static int rpl_do_srh(struct sk_buff *skb, const struct rpl_lwt *rlwt) +static int rpl_do_srh(struct sk_buff *skb, const struct rpl_lwt *rlwt, + struct dst_entry *cache_dst) { struct dst_entry *dst = skb_dst(skb); struct rpl_iptunnel_encap *tinfo; @@ -196,7 +198,7 @@ static int rpl_do_srh(struct sk_buff *skb, const struct rpl_lwt *rlwt) tinfo = rpl_encap_lwtunnel(dst->lwtstate); - return rpl_do_srh_inline(skb, rlwt, tinfo->srh); + return rpl_do_srh_inline(skb, rlwt, tinfo->srh, cache_dst); } static int rpl_output(struct net *net, struct sock *sk, struct sk_buff *skb) @@ -208,14 +210,14 @@ static int rpl_output(struct net *net, struct sock *sk, struct sk_buff *skb) rlwt = rpl_lwt_lwtunnel(orig_dst->lwtstate); - err = rpl_do_srh(skb, rlwt); - if (unlikely(err)) - goto drop; - local_bh_disable(); dst = dst_cache_get(&rlwt->cache); local_bh_enable(); + err = rpl_do_srh(skb, rlwt, dst); + if (unlikely(err)) + goto drop; + if (unlikely(!dst)) { struct ipv6hdr *hdr = ipv6_hdr(skb); struct flowi6 fl6; @@ -237,15 +239,15 @@ static int rpl_output(struct net *net, struct sock *sk, struct sk_buff *skb) local_bh_disable(); dst_cache_set_ip6(&rlwt->cache, dst, &fl6.saddr); local_bh_enable(); + + err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); + if (unlikely(err)) + goto drop; } skb_dst_drop(skb); skb_dst_set(skb, dst); - err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); - if (unlikely(err)) - goto drop; - return dst_output(net, sk, skb); drop: @@ -262,29 +264,31 @@ static int rpl_input(struct sk_buff *skb) rlwt = rpl_lwt_lwtunnel(orig_dst->lwtstate); - err = rpl_do_srh(skb, rlwt); - if (unlikely(err)) - goto drop; - local_bh_disable(); dst = dst_cache_get(&rlwt->cache); + local_bh_enable(); + + err = rpl_do_srh(skb, rlwt, dst); + if (unlikely(err)) + goto drop; if (!dst) { ip6_route_input(skb); dst = skb_dst(skb); if (!dst->error) { + local_bh_disable(); dst_cache_set_ip6(&rlwt->cache, dst, &ipv6_hdr(skb)->saddr); + local_bh_enable(); } + + err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); + if (unlikely(err)) + goto drop; } else { skb_dst_drop(skb); skb_dst_set(skb, dst); } - local_bh_enable(); - - err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); - if (unlikely(err)) - goto drop; return dst_input(skb); diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 098632adc9b5..4bf937bfc263 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -124,8 +124,8 @@ static __be32 seg6_make_flowlabel(struct net *net, struct sk_buff *skb, return flowlabel; } -/* encapsulate an IPv6 packet within an outer IPv6 header with a given SRH */ -int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) +static int __seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, + int proto, struct dst_entry *cache_dst) { struct dst_entry *dst = skb_dst(skb); struct net *net = dev_net(dst->dev); @@ -137,7 +137,7 @@ int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) hdrlen = (osrh->hdrlen + 1) << 3; tot_len = hdrlen + sizeof(*hdr); - err = skb_cow_head(skb, tot_len + skb->mac_len); + err = skb_cow_head(skb, tot_len + dst_dev_overhead(cache_dst, skb)); if (unlikely(err)) return err; @@ -197,11 +197,18 @@ int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) return 0; } + +/* encapsulate an IPv6 packet within an outer IPv6 header with a given SRH */ +int seg6_do_srh_encap(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, int proto) +{ + return __seg6_do_srh_encap(skb, osrh, proto, NULL); +} EXPORT_SYMBOL_GPL(seg6_do_srh_encap); /* encapsulate an IPv6 packet within an outer IPv6 header with reduced SRH */ static int seg6_do_srh_encap_red(struct sk_buff *skb, - struct ipv6_sr_hdr *osrh, int proto) + struct ipv6_sr_hdr *osrh, int proto, + struct dst_entry *cache_dst) { __u8 first_seg = osrh->first_segment; struct dst_entry *dst = skb_dst(skb); @@ -230,7 +237,7 @@ static int seg6_do_srh_encap_red(struct sk_buff *skb, tot_len = red_hdrlen + sizeof(struct ipv6hdr); - err = skb_cow_head(skb, tot_len + skb->mac_len); + err = skb_cow_head(skb, tot_len + dst_dev_overhead(cache_dst, skb)); if (unlikely(err)) return err; @@ -317,8 +324,8 @@ out: return 0; } -/* insert an SRH within an IPv6 packet, just after the IPv6 header */ -int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) +static int __seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh, + struct dst_entry *cache_dst) { struct ipv6hdr *hdr, *oldhdr; struct ipv6_sr_hdr *isrh; @@ -326,7 +333,7 @@ int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) hdrlen = (osrh->hdrlen + 1) << 3; - err = skb_cow_head(skb, hdrlen + skb->mac_len); + err = skb_cow_head(skb, hdrlen + dst_dev_overhead(cache_dst, skb)); if (unlikely(err)) return err; @@ -369,9 +376,8 @@ int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) return 0; } -EXPORT_SYMBOL_GPL(seg6_do_srh_inline); -static int seg6_do_srh(struct sk_buff *skb) +static int seg6_do_srh(struct sk_buff *skb, struct dst_entry *cache_dst) { struct dst_entry *dst = skb_dst(skb); struct seg6_iptunnel_encap *tinfo; @@ -384,7 +390,7 @@ static int seg6_do_srh(struct sk_buff *skb) if (skb->protocol != htons(ETH_P_IPV6)) return -EINVAL; - err = seg6_do_srh_inline(skb, tinfo->srh); + err = __seg6_do_srh_inline(skb, tinfo->srh, cache_dst); if (err) return err; break; @@ -402,9 +408,11 @@ static int seg6_do_srh(struct sk_buff *skb) return -EINVAL; if (tinfo->mode == SEG6_IPTUN_MODE_ENCAP) - err = seg6_do_srh_encap(skb, tinfo->srh, proto); + err = __seg6_do_srh_encap(skb, tinfo->srh, + proto, cache_dst); else - err = seg6_do_srh_encap_red(skb, tinfo->srh, proto); + err = seg6_do_srh_encap_red(skb, tinfo->srh, + proto, cache_dst); if (err) return err; @@ -425,11 +433,13 @@ static int seg6_do_srh(struct sk_buff *skb) skb_push(skb, skb->mac_len); if (tinfo->mode == SEG6_IPTUN_MODE_L2ENCAP) - err = seg6_do_srh_encap(skb, tinfo->srh, - IPPROTO_ETHERNET); + err = __seg6_do_srh_encap(skb, tinfo->srh, + IPPROTO_ETHERNET, + cache_dst); else err = seg6_do_srh_encap_red(skb, tinfo->srh, - IPPROTO_ETHERNET); + IPPROTO_ETHERNET, + cache_dst); if (err) return err; @@ -444,6 +454,13 @@ static int seg6_do_srh(struct sk_buff *skb) return 0; } +/* insert an SRH within an IPv6 packet, just after the IPv6 header */ +int seg6_do_srh_inline(struct sk_buff *skb, struct ipv6_sr_hdr *osrh) +{ + return __seg6_do_srh_inline(skb, osrh, NULL); +} +EXPORT_SYMBOL_GPL(seg6_do_srh_inline); + static int seg6_input_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { @@ -458,31 +475,33 @@ static int seg6_input_core(struct net *net, struct sock *sk, struct seg6_lwt *slwt; int err; - err = seg6_do_srh(skb); - if (unlikely(err)) - goto drop; - slwt = seg6_lwt_lwtunnel(orig_dst->lwtstate); local_bh_disable(); dst = dst_cache_get(&slwt->cache); + local_bh_enable(); + + err = seg6_do_srh(skb, dst); + if (unlikely(err)) + goto drop; if (!dst) { ip6_route_input(skb); dst = skb_dst(skb); if (!dst->error) { + local_bh_disable(); dst_cache_set_ip6(&slwt->cache, dst, &ipv6_hdr(skb)->saddr); + local_bh_enable(); } + + err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); + if (unlikely(err)) + goto drop; } else { skb_dst_drop(skb); skb_dst_set(skb, dst); } - local_bh_enable(); - - err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); - if (unlikely(err)) - goto drop; if (static_branch_unlikely(&nf_hooks_lwtunnel_enabled)) return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, @@ -528,16 +547,16 @@ static int seg6_output_core(struct net *net, struct sock *sk, struct seg6_lwt *slwt; int err; - err = seg6_do_srh(skb); - if (unlikely(err)) - goto drop; - slwt = seg6_lwt_lwtunnel(orig_dst->lwtstate); local_bh_disable(); dst = dst_cache_get(&slwt->cache); local_bh_enable(); + err = seg6_do_srh(skb, dst); + if (unlikely(err)) + goto drop; + if (unlikely(!dst)) { struct ipv6hdr *hdr = ipv6_hdr(skb); struct flowi6 fl6; @@ -559,15 +578,15 @@ static int seg6_output_core(struct net *net, struct sock *sk, local_bh_disable(); dst_cache_set_ip6(&slwt->cache, dst, &fl6.saddr); local_bh_enable(); + + err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); + if (unlikely(err)) + goto drop; } skb_dst_drop(skb); skb_dst_set(skb, dst); - err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); - if (unlikely(err)) - goto drop; - if (static_branch_unlikely(&nf_hooks_lwtunnel_enabled)) return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, net, sk, skb, NULL, skb_dst(skb)->dev, dst_output); diff --git a/tools/testing/selftests/net/udpgso_bench.sh b/tools/testing/selftests/net/udpgso_bench.sh index 640bc43452fa..88fa1d53ba2b 100755 --- a/tools/testing/selftests/net/udpgso_bench.sh +++ b/tools/testing/selftests/net/udpgso_bench.sh @@ -92,6 +92,9 @@ run_udp() { echo "udp" run_in_netns ${args} + echo "udp sendmmsg" + run_in_netns ${args} -m + echo "udp gso" run_in_netns ${args} -S 0 |