diff options
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlxsw')
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/minimal.c | 2 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/reg.h | 152 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 121 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 14 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c | 30 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c | 60 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c | 15 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c | 567 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 874 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h | 3 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c | 61 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h | 1 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c | 52 | ||||
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/trap.h | 9 |
14 files changed, 1246 insertions, 715 deletions
diff --git a/drivers/net/ethernet/mellanox/mlxsw/minimal.c b/drivers/net/ethernet/mellanox/mlxsw/minimal.c index 2b543911ae00..c4caeeadcba9 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/minimal.c +++ b/drivers/net/ethernet/mellanox/mlxsw/minimal.c @@ -213,8 +213,8 @@ mlxsw_m_port_create(struct mlxsw_m *mlxsw_m, u8 local_port, u8 module) err_register_netdev: mlxsw_m->ports[local_port] = NULL; - free_netdev(dev); err_dev_addr_get: + free_netdev(dev); err_alloc_etherdev: mlxsw_core_port_fini(mlxsw_m->core, local_port); return err; diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index af30e8a76682..dd6685156396 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3477,10 +3477,10 @@ MLXSW_REG_DEFINE(qeec, MLXSW_REG_QEEC_ID, MLXSW_REG_QEEC_LEN); MLXSW_ITEM32(reg, qeec, local_port, 0x00, 16, 8); enum mlxsw_reg_qeec_hr { - MLXSW_REG_QEEC_HIERARCY_PORT, - MLXSW_REG_QEEC_HIERARCY_GROUP, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, - MLXSW_REG_QEEC_HIERARCY_TC, + MLXSW_REG_QEEC_HR_PORT, + MLXSW_REG_QEEC_HR_GROUP, + MLXSW_REG_QEEC_HR_SUBGROUP, + MLXSW_REG_QEEC_HR_TC, }; /* reg_qeec_element_hierarchy @@ -3563,8 +3563,8 @@ MLXSW_ITEM32(reg, qeec, min_shaper_rate, 0x0C, 0, 28); */ MLXSW_ITEM32(reg, qeec, mase, 0x10, 31, 1); -/* A large max rate will disable the max shaper. */ -#define MLXSW_REG_QEEC_MAS_DIS 200000000 /* Kbps */ +/* The largest max shaper value possible to disable the shaper. */ +#define MLXSW_REG_QEEC_MAS_DIS ((1u << 31) - 1) /* Kbps */ /* reg_qeec_max_shaper_rate * Max shaper information rate. @@ -3602,6 +3602,21 @@ MLXSW_ITEM32(reg, qeec, dwrr, 0x18, 15, 1); */ MLXSW_ITEM32(reg, qeec, dwrr_weight, 0x18, 0, 8); +/* reg_qeec_max_shaper_bs + * Max shaper burst size + * Burst size is 2^max_shaper_bs * 512 bits + * For Spectrum-1: Range is: 5..25 + * For Spectrum-2: Range is: 11..25 + * Reserved when ptps = 1 + * Access: RW + */ +MLXSW_ITEM32(reg, qeec, max_shaper_bs, 0x1C, 0, 6); + +#define MLXSW_REG_QEEC_HIGHEST_SHAPER_BS 25 +#define MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP1 5 +#define MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP2 11 +#define MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP3 5 + static inline void mlxsw_reg_qeec_pack(char *payload, u8 local_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index) @@ -3618,8 +3633,7 @@ static inline void mlxsw_reg_qeec_ptps_pack(char *payload, u8 local_port, { MLXSW_REG_ZERO(qeec, payload); mlxsw_reg_qeec_local_port_set(payload, local_port); - mlxsw_reg_qeec_element_hierarchy_set(payload, - MLXSW_REG_QEEC_HIERARCY_PORT); + mlxsw_reg_qeec_element_hierarchy_set(payload, MLXSW_REG_QEEC_HR_PORT); mlxsw_reg_qeec_ptps_set(payload, ptps); } @@ -3749,6 +3763,38 @@ mlxsw_reg_qpdsm_prio_pack(char *payload, unsigned short prio, u8 dscp) mlxsw_reg_qpdsm_prio_entry_color2_dscp_set(payload, prio, dscp); } +/* QPDP - QoS Port DSCP to Priority Mapping Register + * ------------------------------------------------- + * This register controls the port default Switch Priority and Color. The + * default Switch Priority and Color are used for frames where the trust state + * uses default values. All member ports of a LAG should be configured with the + * same default values. + */ +#define MLXSW_REG_QPDP_ID 0x4007 +#define MLXSW_REG_QPDP_LEN 0x8 + +MLXSW_REG_DEFINE(qpdp, MLXSW_REG_QPDP_ID, MLXSW_REG_QPDP_LEN); + +/* reg_qpdp_local_port + * Local Port. Supported for data packets from CPU port. + * Access: Index + */ +MLXSW_ITEM32(reg, qpdp, local_port, 0x00, 16, 8); + +/* reg_qpdp_switch_prio + * Default port Switch Priority (default 0) + * Access: RW + */ +MLXSW_ITEM32(reg, qpdp, switch_prio, 0x04, 0, 4); + +static inline void mlxsw_reg_qpdp_pack(char *payload, u8 local_port, + u8 switch_prio) +{ + MLXSW_REG_ZERO(qpdp, payload); + mlxsw_reg_qpdp_local_port_set(payload, local_port); + mlxsw_reg_qpdp_switch_prio_set(payload, switch_prio); +} + /* QPDPM - QoS Port DSCP to Priority Mapping Register * -------------------------------------------------- * This register controls the mapping from DSCP field to @@ -5482,6 +5528,7 @@ enum mlxsw_reg_htgt_discard_trap_group { MLXSW_REG_HTGT_DISCARD_TRAP_GROUP_BASE = MLXSW_REG_HTGT_TRAP_GROUP_MAX, MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS, MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS, + MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS, }; /* reg_htgt_trap_group @@ -10109,6 +10156,92 @@ static inline void mlxsw_reg_tigcr_pack(char *payload, bool ttlc, u8 ttl_uc) mlxsw_reg_tigcr_ttl_uc_set(payload, ttl_uc); } +/* TIEEM - Tunneling IPinIP Encapsulation ECN Mapping Register + * ----------------------------------------------------------- + * The TIEEM register maps ECN of the IP header at the ingress to the + * encapsulation to the ECN of the underlay network. + */ +#define MLXSW_REG_TIEEM_ID 0xA812 +#define MLXSW_REG_TIEEM_LEN 0x0C + +MLXSW_REG_DEFINE(tieem, MLXSW_REG_TIEEM_ID, MLXSW_REG_TIEEM_LEN); + +/* reg_tieem_overlay_ecn + * ECN of the IP header in the overlay network. + * Access: Index + */ +MLXSW_ITEM32(reg, tieem, overlay_ecn, 0x04, 24, 2); + +/* reg_tineem_underlay_ecn + * ECN of the IP header in the underlay network. + * Access: RW + */ +MLXSW_ITEM32(reg, tieem, underlay_ecn, 0x04, 16, 2); + +static inline void mlxsw_reg_tieem_pack(char *payload, u8 overlay_ecn, + u8 underlay_ecn) +{ + MLXSW_REG_ZERO(tieem, payload); + mlxsw_reg_tieem_overlay_ecn_set(payload, overlay_ecn); + mlxsw_reg_tieem_underlay_ecn_set(payload, underlay_ecn); +} + +/* TIDEM - Tunneling IPinIP Decapsulation ECN Mapping Register + * ----------------------------------------------------------- + * The TIDEM register configures the actions that are done in the + * decapsulation. + */ +#define MLXSW_REG_TIDEM_ID 0xA813 +#define MLXSW_REG_TIDEM_LEN 0x0C + +MLXSW_REG_DEFINE(tidem, MLXSW_REG_TIDEM_ID, MLXSW_REG_TIDEM_LEN); + +/* reg_tidem_underlay_ecn + * ECN field of the IP header in the underlay network. + * Access: Index + */ +MLXSW_ITEM32(reg, tidem, underlay_ecn, 0x04, 24, 2); + +/* reg_tidem_overlay_ecn + * ECN field of the IP header in the overlay network. + * Access: Index + */ +MLXSW_ITEM32(reg, tidem, overlay_ecn, 0x04, 16, 2); + +/* reg_tidem_eip_ecn + * Egress IP ECN. ECN field of the IP header of the packet which goes out + * from the decapsulation. + * Access: RW + */ +MLXSW_ITEM32(reg, tidem, eip_ecn, 0x04, 8, 2); + +/* reg_tidem_trap_en + * Trap enable: + * 0 - No trap due to decap ECN + * 1 - Trap enable with trap_id + * Access: RW + */ +MLXSW_ITEM32(reg, tidem, trap_en, 0x08, 28, 4); + +/* reg_tidem_trap_id + * Trap ID. Either DECAP_ECN0 or DECAP_ECN1. + * Reserved when trap_en is '0'. + * Access: RW + */ +MLXSW_ITEM32(reg, tidem, trap_id, 0x08, 0, 9); + +static inline void mlxsw_reg_tidem_pack(char *payload, u8 underlay_ecn, + u8 overlay_ecn, u8 eip_ecn, + bool trap_en, u16 trap_id) +{ + MLXSW_REG_ZERO(tidem, payload); + mlxsw_reg_tidem_underlay_ecn_set(payload, underlay_ecn); + mlxsw_reg_tidem_overlay_ecn_set(payload, overlay_ecn); + mlxsw_reg_tidem_eip_ecn_set(payload, eip_ecn); + mlxsw_reg_tidem_trap_en_set(payload, trap_en); + mlxsw_reg_tidem_trap_id_set(payload, trap_id); +} + /* SBPR - Shared Buffer Pools Register * ----------------------------------- * The SBPR configures and retrieves the shared buffer pools and configuration. @@ -10581,6 +10714,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = { MLXSW_REG(qeec), MLXSW_REG(qrwe), MLXSW_REG(qpdsm), + MLXSW_REG(qpdp), MLXSW_REG(qpdpm), MLXSW_REG(qtctm), MLXSW_REG(qpsc), @@ -10652,6 +10786,8 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = { MLXSW_REG(tndem), MLXSW_REG(tnpc), MLXSW_REG(tigcr), + MLXSW_REG(tieem), + MLXSW_REG(tidem), MLXSW_REG(sbpr), MLXSW_REG(sbcm), MLXSW_REG(sbpm), diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 8ed15199eb4f..7358b5bc7eb6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -45,11 +45,9 @@ #include "spectrum_ptp.h" #include "../mlxfw/mlxfw.h" -#define MLXSW_SP_FWREV_MINOR_TO_BRANCH(minor) ((minor) / 100) - #define MLXSW_SP1_FWREV_MAJOR 13 #define MLXSW_SP1_FWREV_MINOR 2000 -#define MLXSW_SP1_FWREV_SUBMINOR 2308 +#define MLXSW_SP1_FWREV_SUBMINOR 2714 #define MLXSW_SP1_FWREV_CAN_RESET_MINOR 1702 static const struct mlxsw_fw_rev mlxsw_sp1_fw_rev = { @@ -66,7 +64,7 @@ static const struct mlxsw_fw_rev mlxsw_sp1_fw_rev = { #define MLXSW_SP2_FWREV_MAJOR 29 #define MLXSW_SP2_FWREV_MINOR 2000 -#define MLXSW_SP2_FWREV_SUBMINOR 2308 +#define MLXSW_SP2_FWREV_SUBMINOR 2714 static const struct mlxsw_fw_rev mlxsw_sp2_fw_rev = { .major = MLXSW_SP2_FWREV_MAJOR, @@ -197,6 +195,10 @@ struct mlxsw_sp_ptp_ops { u64 *data, int data_index); }; +struct mlxsw_sp_span_ops { + u32 (*buffsize_get)(int mtu, u32 speed); +}; + static int mlxsw_sp_component_query(struct mlxfw_dev *mlxfw_dev, u16 component_index, u32 *p_max_size, u8 *p_align_bits, u16 *p_max_write_size) @@ -423,13 +425,12 @@ static int mlxsw_sp_fw_rev_validate(struct mlxsw_sp *mlxsw_sp) rev->major, req_rev->major); return -EINVAL; } - if (MLXSW_SP_FWREV_MINOR_TO_BRANCH(rev->minor) == - MLXSW_SP_FWREV_MINOR_TO_BRANCH(req_rev->minor) && - mlxsw_core_fw_rev_minor_subminor_validate(rev, req_rev)) + if (mlxsw_core_fw_rev_minor_subminor_validate(rev, req_rev)) return 0; - dev_info(mlxsw_sp->bus_info->dev, "The firmware version %d.%d.%d is incompatible with the driver\n", - rev->major, rev->minor, rev->subminor); + dev_err(mlxsw_sp->bus_info->dev, "The firmware version %d.%d.%d is incompatible with the driver (required >= %d.%d.%d)\n", + rev->major, rev->minor, rev->subminor, req_rev->major, + req_rev->minor, req_rev->subminor); dev_info(mlxsw_sp->bus_info->dev, "Flashing firmware using file %s\n", fw_filename); @@ -1793,6 +1794,10 @@ static int mlxsw_sp_setup_tc(struct net_device *dev, enum tc_setup_type type, return mlxsw_sp_setup_tc_red(mlxsw_sp_port, type_data); case TC_SETUP_QDISC_PRIO: return mlxsw_sp_setup_tc_prio(mlxsw_sp_port, type_data); + case TC_SETUP_QDISC_ETS: + return mlxsw_sp_setup_tc_ets(mlxsw_sp_port, type_data); + case TC_SETUP_QDISC_TBF: + return mlxsw_sp_setup_tc_tbf(mlxsw_sp_port, type_data); default: return -EOPNOTSUPP; } @@ -3536,6 +3541,27 @@ mlxsw_sp_port_speed_by_width_set(struct mlxsw_sp_port *mlxsw_sp_port) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); } +int mlxsw_sp_port_speed_get(struct mlxsw_sp_port *mlxsw_sp_port, u32 *speed) +{ + const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char ptys_pl[MLXSW_REG_PTYS_LEN]; + u32 eth_proto_oper; + int err; + + port_type_speed_ops = mlxsw_sp->port_type_speed_ops; + port_type_speed_ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, + mlxsw_sp_port->local_port, 0, + false); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + if (err) + return err; + port_type_speed_ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, NULL, NULL, + ð_proto_oper); + *speed = port_type_speed_ops->from_ptys_speed(mlxsw_sp, eth_proto_oper); + return 0; +} + int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight) @@ -3553,7 +3579,7 @@ int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, - u8 next_index, u32 maxrate) + u8 next_index, u32 maxrate, u8 burst_size) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; char qeec_pl[MLXSW_REG_QEEC_LEN]; @@ -3562,6 +3588,7 @@ int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port, next_index); mlxsw_reg_qeec_mase_set(qeec_pl, true); mlxsw_reg_qeec_max_shaper_rate_set(qeec_pl, maxrate); + mlxsw_reg_qeec_max_shaper_bs_set(qeec_pl, burst_size); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qeec), qeec_pl); } @@ -3599,26 +3626,25 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port) * one subgroup, which are all member in the same group. */ err = mlxsw_sp_port_ets_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_GROUP, 0, 0, false, - 0); + MLXSW_REG_QEEC_HR_GROUP, 0, 0, false, 0); if (err) return err; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { err = mlxsw_sp_port_ets_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, i, + MLXSW_REG_QEEC_HR_SUBGROUP, i, 0, false, 0); if (err) return err; } for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { err = mlxsw_sp_port_ets_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_TC, i, i, + MLXSW_REG_QEEC_HR_TC, i, i, false, 0); if (err) return err; err = mlxsw_sp_port_ets_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_TC, + MLXSW_REG_QEEC_HR_TC, i + 8, i, true, 100); if (err) @@ -3630,30 +3656,30 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port) * for the initial configuration. */ err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_PORT, 0, 0, - MLXSW_REG_QEEC_MAS_DIS); + MLXSW_REG_QEEC_HR_PORT, 0, 0, + MLXSW_REG_QEEC_MAS_DIS, 0); if (err) return err; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, + MLXSW_REG_QEEC_HR_SUBGROUP, i, 0, - MLXSW_REG_QEEC_MAS_DIS); + MLXSW_REG_QEEC_MAS_DIS, 0); if (err) return err; } for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_TC, + MLXSW_REG_QEEC_HR_TC, i, i, - MLXSW_REG_QEEC_MAS_DIS); + MLXSW_REG_QEEC_MAS_DIS, 0); if (err) return err; err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_TC, + MLXSW_REG_QEEC_HR_TC, i + 8, i, - MLXSW_REG_QEEC_MAS_DIS); + MLXSW_REG_QEEC_MAS_DIS, 0); if (err) return err; } @@ -3661,7 +3687,7 @@ static int mlxsw_sp_port_ets_init(struct mlxsw_sp_port *mlxsw_sp_port) /* Configure the min shaper for multicast TCs. */ for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { err = mlxsw_sp_port_min_bw_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_TC, + MLXSW_REG_QEEC_HR_TC, i + 8, i, MLXSW_REG_QEEC_MIS_MIN); if (err) @@ -3885,6 +3911,8 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, INIT_DELAYED_WORK(&mlxsw_sp_port->ptp.shaper_dw, mlxsw_sp->ptp_ops->shaper_work); + INIT_DELAYED_WORK(&mlxsw_sp_port->span.speed_update_dw, + mlxsw_sp_span_speed_update_work); mlxsw_sp->ports[local_port] = mlxsw_sp_port; err = register_netdev(dev); @@ -3941,6 +3969,7 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; cancel_delayed_work_sync(&mlxsw_sp_port->periodic_hw_stats.update_dw); + cancel_delayed_work_sync(&mlxsw_sp_port->span.speed_update_dw); cancel_delayed_work_sync(&mlxsw_sp_port->ptp.shaper_dw); mlxsw_sp_port_ptp_clear(mlxsw_sp_port); mlxsw_core_port_clear(mlxsw_sp->core, local_port, mlxsw_sp); @@ -4348,6 +4377,7 @@ static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg, netdev_info(mlxsw_sp_port->dev, "link up\n"); netif_carrier_on(mlxsw_sp_port->dev); mlxsw_core_schedule_dw(&mlxsw_sp_port->ptp.shaper_dw, 0); + mlxsw_core_schedule_dw(&mlxsw_sp_port->span.speed_update_dw, 0); } else { netdev_info(mlxsw_sp_port->dev, "link down\n"); netif_carrier_off(mlxsw_sp_port->dev); @@ -4547,10 +4577,16 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = { false), MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV4, TRAP_TO_CPU, ROUTER_EXP, false), MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV6, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(IPIP_DECAP_ERROR, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(DECAP_ECN0, TRAP_TO_CPU, ROUTER_EXP, false), MLXSW_SP_RXL_MARK(IPV4_VRRP, TRAP_TO_CPU, VRRP, false), MLXSW_SP_RXL_MARK(IPV6_VRRP, TRAP_TO_CPU, VRRP, false), + MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_CLASS_E, FORWARD, + ROUTER_EXP, false), + MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_MC_DMAC, FORWARD, + ROUTER_EXP, false), + MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_DIP, FORWARD, + ROUTER_EXP, false), + MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_DIP_LINK_LOCAL, FORWARD, + ROUTER_EXP, false), /* PKT Sample trap */ MLXSW_RXL(mlxsw_sp_rx_listener_sample_func, PKT_SAMPLE, MIRROR_TO_CPU, false, SP_IP2ME, DISCARD), @@ -4889,6 +4925,33 @@ static const struct mlxsw_sp_ptp_ops mlxsw_sp2_ptp_ops = { .get_stats = mlxsw_sp2_get_stats, }; +static u32 mlxsw_sp1_span_buffsize_get(int mtu, u32 speed) +{ + return mtu * 5 / 2; +} + +static const struct mlxsw_sp_span_ops mlxsw_sp1_span_ops = { + .buffsize_get = mlxsw_sp1_span_buffsize_get, +}; + +#define MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR 38 + +static u32 mlxsw_sp2_span_buffsize_get(int mtu, u32 speed) +{ + return 3 * mtu + MLXSW_SP2_SPAN_EG_MIRROR_BUFFER_FACTOR * speed / 1000; +} + +static const struct mlxsw_sp_span_ops mlxsw_sp2_span_ops = { + .buffsize_get = mlxsw_sp2_span_buffsize_get, +}; + +u32 mlxsw_sp_span_buffsize_get(struct mlxsw_sp *mlxsw_sp, int mtu, u32 speed) +{ + u32 buffsize = mlxsw_sp->span_ops->buffsize_get(speed, mtu); + + return mlxsw_sp_bytes_cells(mlxsw_sp, buffsize) + 1; +} + static int mlxsw_sp_netdevice_event(struct notifier_block *unused, unsigned long event, void *ptr); @@ -5110,8 +5173,10 @@ static int mlxsw_sp1_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->sb_vals = &mlxsw_sp1_sb_vals; mlxsw_sp->port_type_speed_ops = &mlxsw_sp1_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp1_ptp_ops; + mlxsw_sp->span_ops = &mlxsw_sp1_span_ops; mlxsw_sp->listeners = mlxsw_sp1_listener; mlxsw_sp->listeners_count = ARRAY_SIZE(mlxsw_sp1_listener); + mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP1; return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack); } @@ -5135,6 +5200,8 @@ static int mlxsw_sp2_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals; mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops; + mlxsw_sp->span_ops = &mlxsw_sp2_span_ops; + mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP2; return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack); } @@ -5156,6 +5223,8 @@ static int mlxsw_sp3_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->sb_vals = &mlxsw_sp2_sb_vals; mlxsw_sp->port_type_speed_ops = &mlxsw_sp2_port_type_speed_ops; mlxsw_sp->ptp_ops = &mlxsw_sp2_ptp_ops; + mlxsw_sp->span_ops = &mlxsw_sp2_span_ops; + mlxsw_sp->lowest_shaper_bs = MLXSW_REG_QEEC_LOWEST_SHAPER_BS_SP3; return mlxsw_sp_init(mlxsw_core, mlxsw_bus_info, extack); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 347bec9d1ecf..a0f1f9dceec5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -140,6 +140,7 @@ struct mlxsw_sp_sb_vals; struct mlxsw_sp_port_type_speed_ops; struct mlxsw_sp_ptp_state; struct mlxsw_sp_ptp_ops; +struct mlxsw_sp_span_ops; struct mlxsw_sp_port_mapping { u8 module; @@ -185,8 +186,10 @@ struct mlxsw_sp { const struct mlxsw_sp_sb_vals *sb_vals; const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops; const struct mlxsw_sp_ptp_ops *ptp_ops; + const struct mlxsw_sp_span_ops *span_ops; const struct mlxsw_listener *listeners; size_t listeners_count; + u32 lowest_shaper_bs; }; static inline struct mlxsw_sp_upper * @@ -292,6 +295,9 @@ struct mlxsw_sp_port { struct mlxsw_sp_ptp_port_stats stats; } ptp; u8 split_base_local_port; + struct { + struct delayed_work speed_update_dw; + } span; }; struct mlxsw_sp_port_type_speed_ops { @@ -471,6 +477,7 @@ extern struct notifier_block mlxsw_sp_switchdev_notifier; /* spectrum.c */ void mlxsw_sp_rx_listener_no_mark_func(struct sk_buff *skb, u8 local_port, void *priv); +int mlxsw_sp_port_speed_get(struct mlxsw_sp_port *mlxsw_sp_port, u32 *speed); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight); @@ -481,7 +488,7 @@ int __mlxsw_sp_port_headroom_set(struct mlxsw_sp_port *mlxsw_sp_port, int mtu, struct ieee_pfc *my_pfc); int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, - u8 next_index, u32 maxrate); + u8 next_index, u32 maxrate, u8 burst_size); enum mlxsw_reg_spms_state mlxsw_sp_stp_spms_state(u8 stp_state); int mlxsw_sp_port_vid_stp_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid, u8 state); @@ -501,6 +508,7 @@ int mlxsw_sp_flow_counter_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int *p_counter_index); void mlxsw_sp_flow_counter_free(struct mlxsw_sp *mlxsw_sp, unsigned int counter_index); +u32 mlxsw_sp_span_buffsize_get(struct mlxsw_sp *mlxsw_sp, int mtu, u32 speed); bool mlxsw_sp_port_dev_check(const struct net_device *dev); struct mlxsw_sp *mlxsw_sp_lower_get(struct net_device *dev); struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find(struct net_device *dev); @@ -852,6 +860,10 @@ int mlxsw_sp_setup_tc_red(struct mlxsw_sp_port *mlxsw_sp_port, struct tc_red_qopt_offload *p); int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port, struct tc_prio_qopt_offload *p); +int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_ets_qopt_offload *p); +int mlxsw_sp_setup_tc_tbf(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_tbf_qopt_offload *p); /* spectrum_fid.c */ bool mlxsw_sp_fid_is_dummy(struct mlxsw_sp *mlxsw_sp, u16 fid_index); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c index 21296fa7f7fb..49a72a8f1f57 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dcb.c @@ -160,7 +160,7 @@ static int __mlxsw_sp_dcbnl_ieee_setets(struct mlxsw_sp_port *mlxsw_sp_port, u8 weight = ets->tc_tx_bw[i]; err = mlxsw_sp_port_ets_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, i, + MLXSW_REG_QEEC_HR_SUBGROUP, i, 0, dwrr, weight); if (err) { netdev_err(dev, "Failed to link subgroup ETS element %d to group\n", @@ -198,7 +198,7 @@ err_port_ets_set: u8 weight = my_ets->tc_tx_bw[i]; err = mlxsw_sp_port_ets_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, i, + MLXSW_REG_QEEC_HR_SUBGROUP, i, 0, dwrr, weight); } return err; @@ -369,6 +369,17 @@ err_update_qrwe: } static int +mlxsw_sp_port_dcb_app_update_qpdp(struct mlxsw_sp_port *mlxsw_sp_port, + u8 default_prio) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char qpdp_pl[MLXSW_REG_QPDP_LEN]; + + mlxsw_reg_qpdp_pack(qpdp_pl, mlxsw_sp_port->local_port, default_prio); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qpdp), qpdp_pl); +} + +static int mlxsw_sp_port_dcb_app_update_qpdpm(struct mlxsw_sp_port *mlxsw_sp_port, struct dcb_ieee_app_dscp_map *map) { @@ -405,6 +416,12 @@ static int mlxsw_sp_port_dcb_app_update(struct mlxsw_sp_port *mlxsw_sp_port) int err; default_prio = mlxsw_sp_port_dcb_app_default_prio(mlxsw_sp_port); + err = mlxsw_sp_port_dcb_app_update_qpdp(mlxsw_sp_port, default_prio); + if (err) { + netdev_err(mlxsw_sp_port->dev, "Couldn't configure port default priority\n"); + return err; + } + have_dscp = mlxsw_sp_port_dcb_app_prio_dscp_map(mlxsw_sp_port, &prio_map); @@ -507,9 +524,9 @@ static int mlxsw_sp_dcbnl_ieee_setmaxrate(struct net_device *dev, for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { err = mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, + MLXSW_REG_QEEC_HR_SUBGROUP, i, 0, - maxrate->tc_maxrate[i]); + maxrate->tc_maxrate[i], 0); if (err) { netdev_err(dev, "Failed to set maxrate for TC %d\n", i); goto err_port_ets_maxrate_set; @@ -523,8 +540,9 @@ static int mlxsw_sp_dcbnl_ieee_setmaxrate(struct net_device *dev, err_port_ets_maxrate_set: for (i--; i >= 0; i--) mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, - MLXSW_REG_QEEC_HIERARCY_SUBGROUP, - i, 0, my_maxrate->tc_maxrate[i]); + MLXSW_REG_QEEC_HR_SUBGROUP, + i, 0, + my_maxrate->tc_maxrate[i], 0); return err; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c index 6400cd644b7a..a8525992528f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c @@ -3,8 +3,10 @@ #include <net/ip_tunnels.h> #include <net/ip6_tunnel.h> +#include <net/inet_ecn.h> #include "spectrum_ipip.h" +#include "reg.h" struct ip_tunnel_parm mlxsw_sp_ipip_netdev_parms4(const struct net_device *ol_dev) @@ -338,3 +340,61 @@ static const struct mlxsw_sp_ipip_ops mlxsw_sp_ipip_gre4_ops = { const struct mlxsw_sp_ipip_ops *mlxsw_sp_ipip_ops_arr[] = { [MLXSW_SP_IPIP_TYPE_GRE4] = &mlxsw_sp_ipip_gre4_ops, }; + +static int mlxsw_sp_ipip_ecn_encap_init_one(struct mlxsw_sp *mlxsw_sp, + u8 inner_ecn, u8 outer_ecn) +{ + char tieem_pl[MLXSW_REG_TIEEM_LEN]; + + mlxsw_reg_tieem_pack(tieem_pl, inner_ecn, outer_ecn); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(tieem), tieem_pl); +} + +int mlxsw_sp_ipip_ecn_encap_init(struct mlxsw_sp *mlxsw_sp) +{ + int i; + + /* Iterate over inner ECN values */ + for (i = INET_ECN_NOT_ECT; i <= INET_ECN_CE; i++) { + u8 outer_ecn = INET_ECN_encapsulate(0, i); + int err; + + err = mlxsw_sp_ipip_ecn_encap_init_one(mlxsw_sp, i, outer_ecn); + if (err) + return err; + } + + return 0; +} + +static int mlxsw_sp_ipip_ecn_decap_init_one(struct mlxsw_sp *mlxsw_sp, + u8 inner_ecn, u8 outer_ecn) +{ + char tidem_pl[MLXSW_REG_TIDEM_LEN]; + bool trap_en, set_ce = false; + u8 new_inner_ecn; + + trap_en = __INET_ECN_decapsulate(outer_ecn, inner_ecn, &set_ce); + new_inner_ecn = set_ce ? INET_ECN_CE : inner_ecn; + + mlxsw_reg_tidem_pack(tidem_pl, outer_ecn, inner_ecn, new_inner_ecn, + trap_en, trap_en ? MLXSW_TRAP_ID_DECAP_ECN0 : 0); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(tidem), tidem_pl); +} + +int mlxsw_sp_ipip_ecn_decap_init(struct mlxsw_sp *mlxsw_sp) +{ + int i, j, err; + + /* Iterate over inner ECN values */ + for (i = INET_ECN_NOT_ECT; i <= INET_ECN_CE; i++) { + /* Iterate over outer ECN values */ + for (j = INET_ECN_NOT_ECT; j <= INET_ECN_CE; j++) { + err = mlxsw_sp_ipip_ecn_decap_init_one(mlxsw_sp, i, j); + if (err) + return err; + } + } + + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c index ec2ff3d7f41c..34f7c3501b08 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c @@ -920,6 +920,7 @@ static int mlxsw_sp_ptp_get_message_types(const struct hwtstamp_config *config, egr_types = 0xff; break; case HWTSTAMP_TX_ONESTEP_SYNC: + case HWTSTAMP_TX_ONESTEP_P2P: return -ERANGE; } @@ -1015,27 +1016,17 @@ mlxsw_sp1_ptp_port_shaper_set(struct mlxsw_sp_port *mlxsw_sp_port, bool enable) static int mlxsw_sp1_ptp_port_shaper_check(struct mlxsw_sp_port *mlxsw_sp_port) { - const struct mlxsw_sp_port_type_speed_ops *port_type_speed_ops; - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char ptys_pl[MLXSW_REG_PTYS_LEN]; - u32 eth_proto_oper, speed; bool ptps = false; int err, i; + u32 speed; if (!mlxsw_sp1_ptp_hwtstamp_enabled(mlxsw_sp_port)) return mlxsw_sp1_ptp_port_shaper_set(mlxsw_sp_port, false); - port_type_speed_ops = mlxsw_sp->port_type_speed_ops; - port_type_speed_ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, - mlxsw_sp_port->local_port, 0, - false); - err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + err = mlxsw_sp_port_speed_get(mlxsw_sp_port, &speed); if (err) return err; - port_type_speed_ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, NULL, NULL, - ð_proto_oper); - speed = port_type_speed_ops->from_ptys_speed(mlxsw_sp, eth_proto_oper); for (i = 0; i < MLXSW_SP1_PTP_SHAPER_PARAMS_LEN; i++) { if (mlxsw_sp1_ptp_shaper_params[i].ethtool_speed == speed) { ptps = true; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c index 0124bfe1963b..79a2801d59f6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_qdisc.c @@ -18,6 +18,8 @@ enum mlxsw_sp_qdisc_type { MLXSW_SP_QDISC_NO_QDISC, MLXSW_SP_QDISC_RED, MLXSW_SP_QDISC_PRIO, + MLXSW_SP_QDISC_ETS, + MLXSW_SP_QDISC_TBF, }; struct mlxsw_sp_qdisc_ops { @@ -226,6 +228,70 @@ mlxsw_sp_qdisc_bstats_per_priority_get(struct mlxsw_sp_port_xstats *xstats, } } +static void +mlxsw_sp_qdisc_collect_tc_stats(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + u64 *p_tx_bytes, u64 *p_tx_packets, + u64 *p_drops, u64 *p_backlog) +{ + u8 tclass_num = mlxsw_sp_qdisc->tclass_num; + struct mlxsw_sp_port_xstats *xstats; + u64 tx_bytes, tx_packets; + + xstats = &mlxsw_sp_port->periodic_hw_stats.xstats; + mlxsw_sp_qdisc_bstats_per_priority_get(xstats, + mlxsw_sp_qdisc->prio_bitmap, + &tx_packets, &tx_bytes); + + *p_tx_packets += tx_packets; + *p_tx_bytes += tx_bytes; + *p_drops += xstats->wred_drop[tclass_num] + + mlxsw_sp_xstats_tail_drop(xstats, tclass_num); + *p_backlog += mlxsw_sp_xstats_backlog(xstats, tclass_num); +} + +static void +mlxsw_sp_qdisc_update_stats(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + u64 tx_bytes, u64 tx_packets, + u64 drops, u64 backlog, + struct tc_qopt_offload_stats *stats_ptr) +{ + struct mlxsw_sp_qdisc_stats *stats_base = &mlxsw_sp_qdisc->stats_base; + + tx_bytes -= stats_base->tx_bytes; + tx_packets -= stats_base->tx_packets; + drops -= stats_base->drops; + backlog -= stats_base->backlog; + + _bstats_update(stats_ptr->bstats, tx_bytes, tx_packets); + stats_ptr->qstats->drops += drops; + stats_ptr->qstats->backlog += mlxsw_sp_cells_bytes(mlxsw_sp, backlog); + + stats_base->backlog += backlog; + stats_base->drops += drops; + stats_base->tx_bytes += tx_bytes; + stats_base->tx_packets += tx_packets; +} + +static void +mlxsw_sp_qdisc_get_tc_stats(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + struct tc_qopt_offload_stats *stats_ptr) +{ + u64 tx_packets = 0; + u64 tx_bytes = 0; + u64 backlog = 0; + u64 drops = 0; + + mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + &tx_bytes, &tx_packets, + &drops, &backlog); + mlxsw_sp_qdisc_update_stats(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc, + tx_bytes, tx_packets, drops, backlog, + stats_ptr); +} + static int mlxsw_sp_tclass_congestion_enable(struct mlxsw_sp_port *mlxsw_sp_port, int tclass_num, u32 min, u32 max, @@ -356,19 +422,28 @@ mlxsw_sp_qdisc_red_replace(struct mlxsw_sp_port *mlxsw_sp_port, } static void -mlxsw_sp_qdisc_red_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, - void *params) +mlxsw_sp_qdisc_leaf_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + struct gnet_stats_queue *qstats) { - struct tc_red_qopt_offload_params *p = params; u64 backlog; backlog = mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc->stats_base.backlog); - p->qstats->backlog -= backlog; + qstats->backlog -= backlog; mlxsw_sp_qdisc->stats_base.backlog = 0; } +static void +mlxsw_sp_qdisc_red_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_red_qopt_offload_params *p = params; + + mlxsw_sp_qdisc_leaf_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, p->qstats); +} + static int mlxsw_sp_qdisc_get_red_xstats(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, @@ -402,41 +477,21 @@ mlxsw_sp_qdisc_get_red_stats(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, struct tc_qopt_offload_stats *stats_ptr) { - u64 tx_bytes, tx_packets, overlimits, drops, backlog; u8 tclass_num = mlxsw_sp_qdisc->tclass_num; struct mlxsw_sp_qdisc_stats *stats_base; struct mlxsw_sp_port_xstats *xstats; + u64 overlimits; xstats = &mlxsw_sp_port->periodic_hw_stats.xstats; stats_base = &mlxsw_sp_qdisc->stats_base; - mlxsw_sp_qdisc_bstats_per_priority_get(xstats, - mlxsw_sp_qdisc->prio_bitmap, - &tx_packets, &tx_bytes); - tx_bytes = tx_bytes - stats_base->tx_bytes; - tx_packets = tx_packets - stats_base->tx_packets; - + mlxsw_sp_qdisc_get_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, stats_ptr); overlimits = xstats->wred_drop[tclass_num] + xstats->ecn - stats_base->overlimits; - drops = xstats->wred_drop[tclass_num] + - mlxsw_sp_xstats_tail_drop(xstats, tclass_num) - - stats_base->drops; - backlog = mlxsw_sp_xstats_backlog(xstats, tclass_num); - _bstats_update(stats_ptr->bstats, tx_bytes, tx_packets); stats_ptr->qstats->overlimits += overlimits; - stats_ptr->qstats->drops += drops; - stats_ptr->qstats->backlog += - mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, - backlog) - - mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, - stats_base->backlog); - - stats_base->backlog = backlog; - stats_base->drops += drops; stats_base->overlimits += overlimits; - stats_base->tx_bytes += tx_bytes; - stats_base->tx_packets += tx_packets; + return 0; } @@ -486,15 +541,215 @@ int mlxsw_sp_setup_tc_red(struct mlxsw_sp_port *mlxsw_sp_port, } } +static void +mlxsw_sp_setup_tc_qdisc_leaf_clean_stats(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) +{ + u64 backlog_cells = 0; + u64 tx_packets = 0; + u64 tx_bytes = 0; + u64 drops = 0; + + mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + &tx_bytes, &tx_packets, + &drops, &backlog_cells); + + mlxsw_sp_qdisc->stats_base.tx_packets = tx_packets; + mlxsw_sp_qdisc->stats_base.tx_bytes = tx_bytes; + mlxsw_sp_qdisc->stats_base.drops = drops; + mlxsw_sp_qdisc->stats_base.backlog = 0; +} + static int -mlxsw_sp_qdisc_prio_destroy(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) +mlxsw_sp_qdisc_tbf_destroy(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) +{ + struct mlxsw_sp_qdisc *root_qdisc = mlxsw_sp_port->root_qdisc; + + if (root_qdisc != mlxsw_sp_qdisc) + root_qdisc->stats_base.backlog -= + mlxsw_sp_qdisc->stats_base.backlog; + + return mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, + MLXSW_REG_QEEC_HR_SUBGROUP, + mlxsw_sp_qdisc->tclass_num, 0, + MLXSW_REG_QEEC_MAS_DIS, 0); +} + +static int +mlxsw_sp_qdisc_tbf_bs(struct mlxsw_sp_port *mlxsw_sp_port, + u32 max_size, u8 *p_burst_size) +{ + /* TBF burst size is configured in bytes. The ASIC burst size value is + * ((2 ^ bs) * 512 bits. Convert the TBF bytes to 512-bit units. + */ + u32 bs512 = max_size / 64; + u8 bs = fls(bs512); + + if (!bs) + return -EINVAL; + --bs; + + /* Demand a power of two. */ + if ((1 << bs) != bs512) + return -EINVAL; + + if (bs < mlxsw_sp_port->mlxsw_sp->lowest_shaper_bs || + bs > MLXSW_REG_QEEC_HIGHEST_SHAPER_BS) + return -EINVAL; + + *p_burst_size = bs; + return 0; +} + +static u32 +mlxsw_sp_qdisc_tbf_max_size(u8 bs) +{ + return (1U << bs) * 64; +} + +static u64 +mlxsw_sp_qdisc_tbf_rate_kbps(struct tc_tbf_qopt_offload_replace_params *p) +{ + /* TBF interface is in bytes/s, whereas Spectrum ASIC is configured in + * Kbits/s. + */ + return p->rate.rate_bytes_ps / 1000 * 8; +} + +static int +mlxsw_sp_qdisc_tbf_check_params(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_tbf_qopt_offload_replace_params *p = params; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + u64 rate_kbps = mlxsw_sp_qdisc_tbf_rate_kbps(p); + u8 burst_size; + int err; + + if (rate_kbps >= MLXSW_REG_QEEC_MAS_DIS) { + dev_err(mlxsw_sp_port->mlxsw_sp->bus_info->dev, + "spectrum: TBF: rate of %lluKbps must be below %u\n", + rate_kbps, MLXSW_REG_QEEC_MAS_DIS); + return -EINVAL; + } + + err = mlxsw_sp_qdisc_tbf_bs(mlxsw_sp_port, p->max_size, &burst_size); + if (err) { + u8 highest_shaper_bs = MLXSW_REG_QEEC_HIGHEST_SHAPER_BS; + + dev_err(mlxsw_sp->bus_info->dev, + "spectrum: TBF: invalid burst size of %u, must be a power of two between %u and %u", + p->max_size, + mlxsw_sp_qdisc_tbf_max_size(mlxsw_sp->lowest_shaper_bs), + mlxsw_sp_qdisc_tbf_max_size(highest_shaper_bs)); + return -EINVAL; + } + + return 0; +} + +static int +mlxsw_sp_qdisc_tbf_replace(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_tbf_qopt_offload_replace_params *p = params; + u64 rate_kbps = mlxsw_sp_qdisc_tbf_rate_kbps(p); + u8 burst_size; + int err; + + err = mlxsw_sp_qdisc_tbf_bs(mlxsw_sp_port, p->max_size, &burst_size); + if (WARN_ON_ONCE(err)) + /* check_params above was supposed to reject this value. */ + return -EINVAL; + + /* Configure subgroup shaper, so that both UC and MC traffic is subject + * to shaping. That is unlike RED, however UC queue lengths are going to + * be different than MC ones due to different pool and quota + * configurations, so the configuration is not applicable. For shaper on + * the other hand, subjecting the overall stream to the configured + * shaper makes sense. Also note that that is what we do for + * ieee_setmaxrate(). + */ + return mlxsw_sp_port_ets_maxrate_set(mlxsw_sp_port, + MLXSW_REG_QEEC_HR_SUBGROUP, + mlxsw_sp_qdisc->tclass_num, 0, + rate_kbps, burst_size); +} + +static void +mlxsw_sp_qdisc_tbf_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_tbf_qopt_offload_replace_params *p = params; + + mlxsw_sp_qdisc_leaf_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, p->qstats); +} + +static int +mlxsw_sp_qdisc_get_tbf_stats(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + struct tc_qopt_offload_stats *stats_ptr) +{ + mlxsw_sp_qdisc_get_tc_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + stats_ptr); + return 0; +} + +static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_tbf = { + .type = MLXSW_SP_QDISC_TBF, + .check_params = mlxsw_sp_qdisc_tbf_check_params, + .replace = mlxsw_sp_qdisc_tbf_replace, + .unoffload = mlxsw_sp_qdisc_tbf_unoffload, + .destroy = mlxsw_sp_qdisc_tbf_destroy, + .get_stats = mlxsw_sp_qdisc_get_tbf_stats, + .clean_stats = mlxsw_sp_setup_tc_qdisc_leaf_clean_stats, +}; + +int mlxsw_sp_setup_tc_tbf(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_tbf_qopt_offload *p) +{ + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc; + + mlxsw_sp_qdisc = mlxsw_sp_qdisc_find(mlxsw_sp_port, p->parent, false); + if (!mlxsw_sp_qdisc) + return -EOPNOTSUPP; + + if (p->command == TC_TBF_REPLACE) + return mlxsw_sp_qdisc_replace(mlxsw_sp_port, p->handle, + mlxsw_sp_qdisc, + &mlxsw_sp_qdisc_ops_tbf, + &p->replace_params); + + if (!mlxsw_sp_qdisc_compare(mlxsw_sp_qdisc, p->handle, + MLXSW_SP_QDISC_TBF)) + return -EOPNOTSUPP; + + switch (p->command) { + case TC_TBF_DESTROY: + return mlxsw_sp_qdisc_destroy(mlxsw_sp_port, mlxsw_sp_qdisc); + case TC_TBF_STATS: + return mlxsw_sp_qdisc_get_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + &p->stats); + default: + return -EOPNOTSUPP; + } +} + +static int +__mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port) { int i; for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { mlxsw_sp_port_prio_tc_set(mlxsw_sp_port, i, MLXSW_SP_PORT_DEFAULT_TCLASS); + mlxsw_sp_port_ets_set(mlxsw_sp_port, + MLXSW_REG_QEEC_HR_SUBGROUP, + i, 0, false, 0); mlxsw_sp_qdisc_destroy(mlxsw_sp_port, &mlxsw_sp_port->tclass_qdiscs[i]); mlxsw_sp_port->tclass_qdiscs[i].prio_bitmap = 0; @@ -504,36 +759,58 @@ mlxsw_sp_qdisc_prio_destroy(struct mlxsw_sp_port *mlxsw_sp_port, } static int -mlxsw_sp_qdisc_prio_check_params(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, - void *params) +mlxsw_sp_qdisc_prio_destroy(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) { - struct tc_prio_qopt_offload_params *p = params; + return __mlxsw_sp_qdisc_ets_destroy(mlxsw_sp_port); +} - if (p->bands > IEEE_8021QAZ_MAX_TCS) +static int +__mlxsw_sp_qdisc_ets_check_params(unsigned int nbands) +{ + if (nbands > IEEE_8021QAZ_MAX_TCS) return -EOPNOTSUPP; return 0; } static int -mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, - void *params) +mlxsw_sp_qdisc_prio_check_params(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) { struct tc_prio_qopt_offload_params *p = params; + + return __mlxsw_sp_qdisc_ets_check_params(p->bands); +} + +static int +__mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, + unsigned int nbands, + const unsigned int *quanta, + const unsigned int *weights, + const u8 *priomap) +{ struct mlxsw_sp_qdisc *child_qdisc; int tclass, i, band, backlog; u8 old_priomap; int err; - for (band = 0; band < p->bands; band++) { + for (band = 0; band < nbands; band++) { tclass = MLXSW_SP_PRIO_BAND_TO_TCLASS(band); child_qdisc = &mlxsw_sp_port->tclass_qdiscs[tclass]; old_priomap = child_qdisc->prio_bitmap; child_qdisc->prio_bitmap = 0; + + err = mlxsw_sp_port_ets_set(mlxsw_sp_port, + MLXSW_REG_QEEC_HR_SUBGROUP, + tclass, 0, !!quanta[band], + weights[band]); + if (err) + return err; + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { - if (p->priomap[i] == band) { + if (priomap[i] == band) { child_qdisc->prio_bitmap |= BIT(i); if (BIT(i) & old_priomap) continue; @@ -556,21 +833,46 @@ mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port, child_qdisc = &mlxsw_sp_port->tclass_qdiscs[tclass]; child_qdisc->prio_bitmap = 0; mlxsw_sp_qdisc_destroy(mlxsw_sp_port, child_qdisc); + mlxsw_sp_port_ets_set(mlxsw_sp_port, + MLXSW_REG_QEEC_HR_SUBGROUP, + tclass, 0, false, 0); } return 0; } +static int +mlxsw_sp_qdisc_prio_replace(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_prio_qopt_offload_params *p = params; + unsigned int zeroes[TCQ_ETS_MAX_BANDS] = {0}; + + return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, p->bands, + zeroes, zeroes, p->priomap); +} + +static void +__mlxsw_sp_qdisc_ets_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + struct gnet_stats_queue *qstats) +{ + u64 backlog; + + backlog = mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, + mlxsw_sp_qdisc->stats_base.backlog); + qstats->backlog -= backlog; +} + static void mlxsw_sp_qdisc_prio_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, void *params) { struct tc_prio_qopt_offload_params *p = params; - u64 backlog; - backlog = mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, - mlxsw_sp_qdisc->stats_base.backlog); - p->qstats->backlog -= backlog; + __mlxsw_sp_qdisc_ets_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, + p->qstats); } static int @@ -578,37 +880,23 @@ mlxsw_sp_qdisc_get_prio_stats(struct mlxsw_sp_port *mlxsw_sp_port, struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, struct tc_qopt_offload_stats *stats_ptr) { - u64 tx_bytes, tx_packets, drops = 0, backlog = 0; - struct mlxsw_sp_qdisc_stats *stats_base; - struct mlxsw_sp_port_xstats *xstats; - struct rtnl_link_stats64 *stats; + struct mlxsw_sp_qdisc *tc_qdisc; + u64 tx_packets = 0; + u64 tx_bytes = 0; + u64 backlog = 0; + u64 drops = 0; int i; - xstats = &mlxsw_sp_port->periodic_hw_stats.xstats; - stats = &mlxsw_sp_port->periodic_hw_stats.stats; - stats_base = &mlxsw_sp_qdisc->stats_base; - - tx_bytes = stats->tx_bytes - stats_base->tx_bytes; - tx_packets = stats->tx_packets - stats_base->tx_packets; - for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { - drops += mlxsw_sp_xstats_tail_drop(xstats, i); - drops += xstats->wred_drop[i]; - backlog += mlxsw_sp_xstats_backlog(xstats, i); + tc_qdisc = &mlxsw_sp_port->tclass_qdiscs[i]; + mlxsw_sp_qdisc_collect_tc_stats(mlxsw_sp_port, tc_qdisc, + &tx_bytes, &tx_packets, + &drops, &backlog); } - drops = drops - stats_base->drops; - _bstats_update(stats_ptr->bstats, tx_bytes, tx_packets); - stats_ptr->qstats->drops += drops; - stats_ptr->qstats->backlog += - mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, - backlog) - - mlxsw_sp_cells_bytes(mlxsw_sp_port->mlxsw_sp, - stats_base->backlog); - stats_base->backlog = backlog; - stats_base->drops += drops; - stats_base->tx_bytes += tx_bytes; - stats_base->tx_packets += tx_packets; + mlxsw_sp_qdisc_update_stats(mlxsw_sp_port->mlxsw_sp, mlxsw_sp_qdisc, + tx_bytes, tx_packets, drops, backlog, + stats_ptr); return 0; } @@ -647,27 +935,93 @@ static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_prio = { .clean_stats = mlxsw_sp_setup_tc_qdisc_prio_clean_stats, }; -/* Grafting is not supported in mlxsw. It will result in un-offloading of the - * grafted qdisc as well as the qdisc in the qdisc new location. - * (However, if the graft is to the location where the qdisc is already at, it - * will be ignored completely and won't cause un-offloading). +static int +mlxsw_sp_qdisc_ets_check_params(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_ets_qopt_offload_replace_params *p = params; + + return __mlxsw_sp_qdisc_ets_check_params(p->bands); +} + +static int +mlxsw_sp_qdisc_ets_replace(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_ets_qopt_offload_replace_params *p = params; + + return __mlxsw_sp_qdisc_ets_replace(mlxsw_sp_port, p->bands, + p->quanta, p->weights, p->priomap); +} + +static void +mlxsw_sp_qdisc_ets_unoffload(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + void *params) +{ + struct tc_ets_qopt_offload_replace_params *p = params; + + __mlxsw_sp_qdisc_ets_unoffload(mlxsw_sp_port, mlxsw_sp_qdisc, + p->qstats); +} + +static int +mlxsw_sp_qdisc_ets_destroy(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc) +{ + return __mlxsw_sp_qdisc_ets_destroy(mlxsw_sp_port); +} + +static struct mlxsw_sp_qdisc_ops mlxsw_sp_qdisc_ops_ets = { + .type = MLXSW_SP_QDISC_ETS, + .check_params = mlxsw_sp_qdisc_ets_check_params, + .replace = mlxsw_sp_qdisc_ets_replace, + .unoffload = mlxsw_sp_qdisc_ets_unoffload, + .destroy = mlxsw_sp_qdisc_ets_destroy, + .get_stats = mlxsw_sp_qdisc_get_prio_stats, + .clean_stats = mlxsw_sp_setup_tc_qdisc_prio_clean_stats, +}; + +/* Linux allows linking of Qdiscs to arbitrary classes (so long as the resulting + * graph is free of cycles). These operations do not change the parent handle + * though, which means it can be incomplete (if there is more than one class + * where the Qdisc in question is grafted) or outright wrong (if the Qdisc was + * linked to a different class and then removed from the original class). + * + * E.g. consider this sequence of operations: + * + * # tc qdisc add dev swp1 root handle 1: prio + * # tc qdisc add dev swp1 parent 1:3 handle 13: red limit 1000000 avpkt 10000 + * RED: set bandwidth to 10Mbit + * # tc qdisc link dev swp1 handle 13: parent 1:2 + * + * At this point, both 1:2 and 1:3 have the same RED Qdisc instance as their + * child. But RED will still only claim that 1:3 is its parent. If it's removed + * from that band, its only parent will be 1:2, but it will continue to claim + * that it is in fact 1:3. + * + * The notification for child Qdisc replace (e.g. TC_RED_REPLACE) comes before + * the notification for parent graft (e.g. TC_PRIO_GRAFT). We take the replace + * notification to offload the child Qdisc, based on its parent handle, and use + * the graft operation to validate that the class where the child is actually + * grafted corresponds to the parent handle. If the two don't match, we + * unoffload the child. */ static int -mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, - struct tc_prio_qopt_offload_graft_params *p) +__mlxsw_sp_qdisc_ets_graft(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + u8 band, u32 child_handle) { - int tclass_num = MLXSW_SP_PRIO_BAND_TO_TCLASS(p->band); + int tclass_num = MLXSW_SP_PRIO_BAND_TO_TCLASS(band); struct mlxsw_sp_qdisc *old_qdisc; - /* Check if the grafted qdisc is already in its "new" location. If so - - * nothing needs to be done. - */ - if (p->band < IEEE_8021QAZ_MAX_TCS && - mlxsw_sp_port->tclass_qdiscs[tclass_num].handle == p->child_handle) + if (band < IEEE_8021QAZ_MAX_TCS && + mlxsw_sp_port->tclass_qdiscs[tclass_num].handle == child_handle) return 0; - if (!p->child_handle) { + if (!child_handle) { /* This is an invisible FIFO replacing the original Qdisc. * Ignore it--the original Qdisc's destroy will follow. */ @@ -678,7 +1032,7 @@ mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port, * unoffload it. */ old_qdisc = mlxsw_sp_qdisc_find_by_handle(mlxsw_sp_port, - p->child_handle); + child_handle); if (old_qdisc) mlxsw_sp_qdisc_destroy(mlxsw_sp_port, old_qdisc); @@ -687,6 +1041,15 @@ mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port, return -EOPNOTSUPP; } +static int +mlxsw_sp_qdisc_prio_graft(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc, + struct tc_prio_qopt_offload_graft_params *p) +{ + return __mlxsw_sp_qdisc_ets_graft(mlxsw_sp_port, mlxsw_sp_qdisc, + p->band, p->child_handle); +} + int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port, struct tc_prio_qopt_offload *p) { @@ -720,6 +1083,40 @@ int mlxsw_sp_setup_tc_prio(struct mlxsw_sp_port *mlxsw_sp_port, } } +int mlxsw_sp_setup_tc_ets(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_ets_qopt_offload *p) +{ + struct mlxsw_sp_qdisc *mlxsw_sp_qdisc; + + mlxsw_sp_qdisc = mlxsw_sp_qdisc_find(mlxsw_sp_port, p->parent, true); + if (!mlxsw_sp_qdisc) + return -EOPNOTSUPP; + + if (p->command == TC_ETS_REPLACE) + return mlxsw_sp_qdisc_replace(mlxsw_sp_port, p->handle, + mlxsw_sp_qdisc, + &mlxsw_sp_qdisc_ops_ets, + &p->replace_params); + + if (!mlxsw_sp_qdisc_compare(mlxsw_sp_qdisc, p->handle, + MLXSW_SP_QDISC_ETS)) + return -EOPNOTSUPP; + + switch (p->command) { + case TC_ETS_DESTROY: + return mlxsw_sp_qdisc_destroy(mlxsw_sp_port, mlxsw_sp_qdisc); + case TC_ETS_STATS: + return mlxsw_sp_qdisc_get_stats(mlxsw_sp_port, mlxsw_sp_qdisc, + &p->stats); + case TC_ETS_GRAFT: + return __mlxsw_sp_qdisc_ets_graft(mlxsw_sp_port, mlxsw_sp_qdisc, + p->graft_params.band, + p->graft_params.child_handle); + default: + return -EOPNOTSUPP; + } +} + int mlxsw_sp_tc_qdisc_init(struct mlxsw_sp_port *mlxsw_sp_port) { struct mlxsw_sp_qdisc *mlxsw_sp_qdisc; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 8290e82240fc..ce707723f8cf 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -382,9 +382,10 @@ enum mlxsw_sp_fib_entry_type { }; struct mlxsw_sp_nexthop_group; +struct mlxsw_sp_fib_entry; struct mlxsw_sp_fib_node { - struct list_head entry_list; + struct mlxsw_sp_fib_entry *fib_entry; struct list_head list; struct rhash_head ht_node; struct mlxsw_sp_fib *fib; @@ -397,7 +398,6 @@ struct mlxsw_sp_fib_entry_decap { }; struct mlxsw_sp_fib_entry { - struct list_head list; struct mlxsw_sp_fib_node *fib_node; enum mlxsw_sp_fib_entry_type type; struct list_head nexthop_group_node; @@ -1162,7 +1162,6 @@ mlxsw_sp_router_ip2me_fib_entry_find(struct mlxsw_sp *mlxsw_sp, u32 tb_id, const union mlxsw_sp_l3addr *addr, enum mlxsw_sp_fib_entry_type type) { - struct mlxsw_sp_fib_entry *fib_entry; struct mlxsw_sp_fib_node *fib_node; unsigned char addr_prefix_len; struct mlxsw_sp_fib *fib; @@ -1191,15 +1190,10 @@ mlxsw_sp_router_ip2me_fib_entry_find(struct mlxsw_sp *mlxsw_sp, u32 tb_id, fib_node = mlxsw_sp_fib_node_lookup(fib, addrp, addr_len, addr_prefix_len); - if (!fib_node || list_empty(&fib_node->entry_list)) + if (!fib_node || fib_node->fib_entry->type != type) return NULL; - fib_entry = list_first_entry(&fib_node->entry_list, - struct mlxsw_sp_fib_entry, list); - if (fib_entry->type != type) - return NULL; - - return fib_entry; + return fib_node->fib_entry; } /* Given an IPIP entry, find the corresponding decap route. */ @@ -1209,7 +1203,6 @@ mlxsw_sp_ipip_entry_find_decap(struct mlxsw_sp *mlxsw_sp, { static struct mlxsw_sp_fib_node *fib_node; const struct mlxsw_sp_ipip_ops *ipip_ops; - struct mlxsw_sp_fib_entry *fib_entry; unsigned char saddr_prefix_len; union mlxsw_sp_l3addr saddr; struct mlxsw_sp_fib *ul_fib; @@ -1244,15 +1237,11 @@ mlxsw_sp_ipip_entry_find_decap(struct mlxsw_sp *mlxsw_sp, fib_node = mlxsw_sp_fib_node_lookup(ul_fib, saddrp, saddr_len, saddr_prefix_len); - if (!fib_node || list_empty(&fib_node->entry_list)) + if (!fib_node || + fib_node->fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_TRAP) return NULL; - fib_entry = list_first_entry(&fib_node->entry_list, - struct mlxsw_sp_fib_entry, list); - if (fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_TRAP) - return NULL; - - return fib_entry; + return fib_node->fib_entry; } static struct mlxsw_sp_ipip_entry * @@ -3231,10 +3220,6 @@ mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp, return 0; } -static bool -mlxsw_sp_fib_node_entry_is_first(const struct mlxsw_sp_fib_node *fib_node, - const struct mlxsw_sp_fib_entry *fib_entry); - static int mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop_group *nh_grp) @@ -3243,9 +3228,6 @@ mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp, int err; list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) { - if (!mlxsw_sp_fib_node_entry_is_first(fib_entry->fib_node, - fib_entry)) - continue; err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); if (err) return err; @@ -3253,24 +3235,6 @@ mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp, return 0; } -static void -mlxsw_sp_fib_entry_offload_refresh(struct mlxsw_sp_fib_entry *fib_entry, - enum mlxsw_reg_ralue_op op, int err); - -static void -mlxsw_sp_nexthop_fib_entries_refresh(struct mlxsw_sp_nexthop_group *nh_grp) -{ - enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_WRITE; - struct mlxsw_sp_fib_entry *fib_entry; - - list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) { - if (!mlxsw_sp_fib_node_entry_is_first(fib_entry->fib_node, - fib_entry)) - continue; - mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0); - } -} - static void mlxsw_sp_adj_grp_size_round_up(u16 *p_adj_grp_size) { /* Valid sizes for an adjacency group are: @@ -3374,6 +3338,73 @@ mlxsw_sp_nexthop_group_rebalance(struct mlxsw_sp_nexthop_group *nh_grp) } } +static struct mlxsw_sp_nexthop * +mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp, + const struct mlxsw_sp_rt6 *mlxsw_sp_rt6); + +static void +mlxsw_sp_nexthop4_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + int i; + + for (i = 0; i < nh_grp->count; i++) { + struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i]; + + if (nh->offloaded) + nh->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD; + else + nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD; + } +} + +static void +__mlxsw_sp_nexthop6_group_offload_refresh(struct mlxsw_sp_nexthop_group *nh_grp, + struct mlxsw_sp_fib6_entry *fib6_entry) +{ + struct mlxsw_sp_rt6 *mlxsw_sp_rt6; + + list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) { + struct fib6_nh *fib6_nh = mlxsw_sp_rt6->rt->fib6_nh; + struct mlxsw_sp_nexthop *nh; + + nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6); + if (nh && nh->offloaded) + fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD; + else + fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD; + } +} + +static void +mlxsw_sp_nexthop6_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_fib6_entry *fib6_entry; + + /* Unfortunately, in IPv6 the route and the nexthop are described by + * the same struct, so we need to iterate over all the routes using the + * nexthop group and set / clear the offload indication for them. + */ + list_for_each_entry(fib6_entry, &nh_grp->fib_list, + common.nexthop_group_node) + __mlxsw_sp_nexthop6_group_offload_refresh(nh_grp, fib6_entry); +} + +static void +mlxsw_sp_nexthop_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + switch (mlxsw_sp_nexthop_group_type(nh_grp)) { + case AF_INET: + mlxsw_sp_nexthop4_group_offload_refresh(mlxsw_sp, nh_grp); + break; + case AF_INET6: + mlxsw_sp_nexthop6_group_offload_refresh(mlxsw_sp, nh_grp); + break; + } +} + static void mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop_group *nh_grp) @@ -3447,6 +3478,8 @@ mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, goto set_trap; } + mlxsw_sp_nexthop_group_offload_refresh(mlxsw_sp, nh_grp); + if (!old_adj_index_valid) { /* The trap was set for fib entries, so we have to call * fib entry update to unset it and use adjacency index. @@ -3468,9 +3501,6 @@ mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, goto set_trap; } - /* Offload state within the group changed, so update the flags. */ - mlxsw_sp_nexthop_fib_entries_refresh(nh_grp); - return; set_trap: @@ -3483,6 +3513,7 @@ set_trap: err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp); if (err) dev_warn(mlxsw_sp->bus_info->dev, "Failed to set traps for fib entries.\n"); + mlxsw_sp_nexthop_group_offload_refresh(mlxsw_sp, nh_grp); if (old_adj_index_valid) mlxsw_sp_kvdl_free(mlxsw_sp, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ, nh_grp->ecmp_size, nh_grp->adj_index); @@ -3845,7 +3876,7 @@ static void mlxsw_sp_nexthop4_event(struct mlxsw_sp *mlxsw_sp, key.fib_nh = fib_nh; nh = mlxsw_sp_nexthop_lookup(mlxsw_sp, key); - if (WARN_ON_ONCE(!nh)) + if (!nh) return; switch (event) { @@ -4065,131 +4096,128 @@ mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp, } static void -mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry) +mlxsw_sp_fib4_entry_hw_flags_set(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { - struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group; - int i; - - if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL || - fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE || - fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP || - fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP) { - nh_grp->nexthops->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD; - return; - } - - for (i = 0; i < nh_grp->count; i++) { - struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i]; + struct fib_info *fi = mlxsw_sp_nexthop4_group_fi(fib_entry->nh_group); + u32 *p_dst = (u32 *) fib_entry->fib_node->key.addr; + int dst_len = fib_entry->fib_node->key.prefix_len; + struct mlxsw_sp_fib4_entry *fib4_entry; + struct fib_rt_info fri; + bool should_offload; - if (nh->offloaded) - nh->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD; - else - nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD; - } + should_offload = mlxsw_sp_fib_entry_should_offload(fib_entry); + fib4_entry = container_of(fib_entry, struct mlxsw_sp_fib4_entry, + common); + fri.fi = fi; + fri.tb_id = fib4_entry->tb_id; + fri.dst = cpu_to_be32(*p_dst); + fri.dst_len = dst_len; + fri.tos = fib4_entry->tos; + fri.type = fib4_entry->type; + fri.offload = should_offload; + fri.trap = !should_offload; + fib_alias_hw_flags_set(mlxsw_sp_net(mlxsw_sp), &fri); } static void -mlxsw_sp_fib4_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry) +mlxsw_sp_fib4_entry_hw_flags_clear(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { - struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group; - int i; - - if (!list_is_singular(&nh_grp->fib_list)) - return; - - for (i = 0; i < nh_grp->count; i++) { - struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i]; + struct fib_info *fi = mlxsw_sp_nexthop4_group_fi(fib_entry->nh_group); + u32 *p_dst = (u32 *) fib_entry->fib_node->key.addr; + int dst_len = fib_entry->fib_node->key.prefix_len; + struct mlxsw_sp_fib4_entry *fib4_entry; + struct fib_rt_info fri; - nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD; - } + fib4_entry = container_of(fib_entry, struct mlxsw_sp_fib4_entry, + common); + fri.fi = fi; + fri.tb_id = fib4_entry->tb_id; + fri.dst = cpu_to_be32(*p_dst); + fri.dst_len = dst_len; + fri.tos = fib4_entry->tos; + fri.type = fib4_entry->type; + fri.offload = false; + fri.trap = false; + fib_alias_hw_flags_set(mlxsw_sp_net(mlxsw_sp), &fri); } static void -mlxsw_sp_fib6_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry) +mlxsw_sp_fib6_entry_hw_flags_set(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { struct mlxsw_sp_fib6_entry *fib6_entry; struct mlxsw_sp_rt6 *mlxsw_sp_rt6; + bool should_offload; + + should_offload = mlxsw_sp_fib_entry_should_offload(fib_entry); + /* In IPv6 a multipath route is represented using multiple routes, so + * we need to set the flags on all of them. + */ fib6_entry = container_of(fib_entry, struct mlxsw_sp_fib6_entry, common); - - if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL || - fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE) { - list_first_entry(&fib6_entry->rt6_list, struct mlxsw_sp_rt6, - list)->rt->fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD; - return; - } - - list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) { - struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group; - struct fib6_nh *fib6_nh = mlxsw_sp_rt6->rt->fib6_nh; - struct mlxsw_sp_nexthop *nh; - - nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6); - if (nh && nh->offloaded) - fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD; - else - fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD; - } + list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) + fib6_info_hw_flags_set(mlxsw_sp_rt6->rt, should_offload, + !should_offload); } static void -mlxsw_sp_fib6_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry) +mlxsw_sp_fib6_entry_hw_flags_clear(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { struct mlxsw_sp_fib6_entry *fib6_entry; struct mlxsw_sp_rt6 *mlxsw_sp_rt6; fib6_entry = container_of(fib_entry, struct mlxsw_sp_fib6_entry, common); - list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) { - struct fib6_info *rt = mlxsw_sp_rt6->rt; - - rt->fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD; - } + list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) + fib6_info_hw_flags_set(mlxsw_sp_rt6->rt, false, false); } -static void mlxsw_sp_fib_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry) +static void +mlxsw_sp_fib_entry_hw_flags_set(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { switch (fib_entry->fib_node->fib->proto) { case MLXSW_SP_L3_PROTO_IPV4: - mlxsw_sp_fib4_entry_offload_set(fib_entry); + mlxsw_sp_fib4_entry_hw_flags_set(mlxsw_sp, fib_entry); break; case MLXSW_SP_L3_PROTO_IPV6: - mlxsw_sp_fib6_entry_offload_set(fib_entry); + mlxsw_sp_fib6_entry_hw_flags_set(mlxsw_sp, fib_entry); break; } } static void -mlxsw_sp_fib_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry) +mlxsw_sp_fib_entry_hw_flags_clear(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { switch (fib_entry->fib_node->fib->proto) { case MLXSW_SP_L3_PROTO_IPV4: - mlxsw_sp_fib4_entry_offload_unset(fib_entry); + mlxsw_sp_fib4_entry_hw_flags_clear(mlxsw_sp, fib_entry); break; case MLXSW_SP_L3_PROTO_IPV6: - mlxsw_sp_fib6_entry_offload_unset(fib_entry); + mlxsw_sp_fib6_entry_hw_flags_clear(mlxsw_sp, fib_entry); break; } } static void -mlxsw_sp_fib_entry_offload_refresh(struct mlxsw_sp_fib_entry *fib_entry, - enum mlxsw_reg_ralue_op op, int err) +mlxsw_sp_fib_entry_hw_flags_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) { switch (op) { - case MLXSW_REG_RALUE_OP_WRITE_DELETE: - return mlxsw_sp_fib_entry_offload_unset(fib_entry); case MLXSW_REG_RALUE_OP_WRITE_WRITE: - if (err) - return; - if (mlxsw_sp_fib_entry_should_offload(fib_entry)) - mlxsw_sp_fib_entry_offload_set(fib_entry); - else - mlxsw_sp_fib_entry_offload_unset(fib_entry); - return; + mlxsw_sp_fib_entry_hw_flags_set(mlxsw_sp, fib_entry); + break; + case MLXSW_REG_RALUE_OP_WRITE_DELETE: + mlxsw_sp_fib_entry_hw_flags_clear(mlxsw_sp, fib_entry); + break; default: - return; + break; } } @@ -4416,7 +4444,10 @@ static int mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp, { int err = __mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, op); - mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, err); + if (err) + return err; + + mlxsw_sp_fib_entry_hw_flags_refresh(mlxsw_sp, fib_entry, op); return err; } @@ -4491,6 +4522,19 @@ mlxsw_sp_fib4_entry_type_set(struct mlxsw_sp *mlxsw_sp, } } +static void +mlxsw_sp_fib4_entry_type_unset(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + switch (fib_entry->type) { + case MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP: + mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, fib_entry); + break; + default: + break; + } +} + static struct mlxsw_sp_fib4_entry * mlxsw_sp_fib4_entry_create(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_node *fib_node, @@ -4523,6 +4567,7 @@ mlxsw_sp_fib4_entry_create(struct mlxsw_sp *mlxsw_sp, return fib4_entry; err_nexthop4_group_get: + mlxsw_sp_fib4_entry_type_unset(mlxsw_sp, fib_entry); err_fib4_entry_type_set: kfree(fib4_entry); return ERR_PTR(err); @@ -4532,6 +4577,7 @@ static void mlxsw_sp_fib4_entry_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib4_entry *fib4_entry) { mlxsw_sp_nexthop4_group_put(mlxsw_sp, &fib4_entry->common); + mlxsw_sp_fib4_entry_type_unset(mlxsw_sp, &fib4_entry->common); kfree(fib4_entry); } @@ -4555,15 +4601,14 @@ mlxsw_sp_fib4_entry_lookup(struct mlxsw_sp *mlxsw_sp, if (!fib_node) return NULL; - list_for_each_entry(fib4_entry, &fib_node->entry_list, common.list) { - if (fib4_entry->tb_id == fen_info->tb_id && - fib4_entry->tos == fen_info->tos && - fib4_entry->type == fen_info->type && - mlxsw_sp_nexthop4_group_fi(fib4_entry->common.nh_group) == - fen_info->fi) { - return fib4_entry; - } - } + fib4_entry = container_of(fib_node->fib_entry, + struct mlxsw_sp_fib4_entry, common); + if (fib4_entry->tb_id == fen_info->tb_id && + fib4_entry->tos == fen_info->tos && + fib4_entry->type == fen_info->type && + mlxsw_sp_nexthop4_group_fi(fib4_entry->common.nh_group) == + fen_info->fi) + return fib4_entry; return NULL; } @@ -4611,7 +4656,6 @@ mlxsw_sp_fib_node_create(struct mlxsw_sp_fib *fib, const void *addr, if (!fib_node) return NULL; - INIT_LIST_HEAD(&fib_node->entry_list); list_add(&fib_node->list, &fib->node_list); memcpy(fib_node->key.addr, addr, addr_len); fib_node->key.prefix_len = prefix_len; @@ -4622,18 +4666,9 @@ mlxsw_sp_fib_node_create(struct mlxsw_sp_fib *fib, const void *addr, static void mlxsw_sp_fib_node_destroy(struct mlxsw_sp_fib_node *fib_node) { list_del(&fib_node->list); - WARN_ON(!list_empty(&fib_node->entry_list)); kfree(fib_node); } -static bool -mlxsw_sp_fib_node_entry_is_first(const struct mlxsw_sp_fib_node *fib_node, - const struct mlxsw_sp_fib_entry *fib_entry) -{ - return list_first_entry(&fib_node->entry_list, - struct mlxsw_sp_fib_entry, list) == fib_entry; -} - static int mlxsw_sp_fib_lpm_tree_link(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_node *fib_node) { @@ -4773,200 +4808,48 @@ static void mlxsw_sp_fib_node_put(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_vr *vr = fib_node->fib->vr; - if (!list_empty(&fib_node->entry_list)) + if (fib_node->fib_entry) return; mlxsw_sp_fib_node_fini(mlxsw_sp, fib_node); mlxsw_sp_fib_node_destroy(fib_node); mlxsw_sp_vr_put(mlxsw_sp, vr); } -static struct mlxsw_sp_fib4_entry * -mlxsw_sp_fib4_node_entry_find(const struct mlxsw_sp_fib_node *fib_node, - const struct mlxsw_sp_fib4_entry *new4_entry) -{ - struct mlxsw_sp_fib4_entry *fib4_entry; - - list_for_each_entry(fib4_entry, &fib_node->entry_list, common.list) { - if (fib4_entry->tb_id > new4_entry->tb_id) - continue; - if (fib4_entry->tb_id != new4_entry->tb_id) - break; - if (fib4_entry->tos > new4_entry->tos) - continue; - if (fib4_entry->prio >= new4_entry->prio || - fib4_entry->tos < new4_entry->tos) - return fib4_entry; - } - - return NULL; -} - -static int -mlxsw_sp_fib4_node_list_append(struct mlxsw_sp_fib4_entry *fib4_entry, - struct mlxsw_sp_fib4_entry *new4_entry) -{ - struct mlxsw_sp_fib_node *fib_node; - - if (WARN_ON(!fib4_entry)) - return -EINVAL; - - fib_node = fib4_entry->common.fib_node; - list_for_each_entry_from(fib4_entry, &fib_node->entry_list, - common.list) { - if (fib4_entry->tb_id != new4_entry->tb_id || - fib4_entry->tos != new4_entry->tos || - fib4_entry->prio != new4_entry->prio) - break; - } - - list_add_tail(&new4_entry->common.list, &fib4_entry->common.list); - return 0; -} - -static int -mlxsw_sp_fib4_node_list_insert(struct mlxsw_sp_fib4_entry *new4_entry, - bool replace, bool append) -{ - struct mlxsw_sp_fib_node *fib_node = new4_entry->common.fib_node; - struct mlxsw_sp_fib4_entry *fib4_entry; - - fib4_entry = mlxsw_sp_fib4_node_entry_find(fib_node, new4_entry); - - if (append) - return mlxsw_sp_fib4_node_list_append(fib4_entry, new4_entry); - if (replace && WARN_ON(!fib4_entry)) - return -EINVAL; - - /* Insert new entry before replaced one, so that we can later - * remove the second. - */ - if (fib4_entry) { - list_add_tail(&new4_entry->common.list, - &fib4_entry->common.list); - } else { - struct mlxsw_sp_fib4_entry *last; - - list_for_each_entry(last, &fib_node->entry_list, common.list) { - if (new4_entry->tb_id > last->tb_id) - break; - fib4_entry = last; - } - - if (fib4_entry) - list_add(&new4_entry->common.list, - &fib4_entry->common.list); - else - list_add(&new4_entry->common.list, - &fib_node->entry_list); - } - - return 0; -} - -static void -mlxsw_sp_fib4_node_list_remove(struct mlxsw_sp_fib4_entry *fib4_entry) -{ - list_del(&fib4_entry->common.list); -} - -static int mlxsw_sp_fib_node_entry_add(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib_entry *fib_entry) -{ - struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node; - - if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry)) - return 0; - - /* To prevent packet loss, overwrite the previously offloaded - * entry. - */ - if (!list_is_singular(&fib_node->entry_list)) { - enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_DELETE; - struct mlxsw_sp_fib_entry *n = list_next_entry(fib_entry, list); - - mlxsw_sp_fib_entry_offload_refresh(n, op, 0); - } - - return mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); -} - -static void mlxsw_sp_fib_node_entry_del(struct mlxsw_sp *mlxsw_sp, +static int mlxsw_sp_fib_node_entry_link(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_entry *fib_entry) { struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node; - - if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry)) - return; - - /* Promote the next entry by overwriting the deleted entry */ - if (!list_is_singular(&fib_node->entry_list)) { - struct mlxsw_sp_fib_entry *n = list_next_entry(fib_entry, list); - enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_DELETE; - - mlxsw_sp_fib_entry_update(mlxsw_sp, n); - mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0); - return; - } - - mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry); -} - -static int mlxsw_sp_fib4_node_entry_link(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib4_entry *fib4_entry, - bool replace, bool append) -{ int err; - err = mlxsw_sp_fib4_node_list_insert(fib4_entry, replace, append); - if (err) - return err; + fib_node->fib_entry = fib_entry; - err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib4_entry->common); + err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); if (err) - goto err_fib_node_entry_add; + goto err_fib_entry_update; return 0; -err_fib_node_entry_add: - mlxsw_sp_fib4_node_list_remove(fib4_entry); +err_fib_entry_update: + fib_node->fib_entry = NULL; return err; } static void -mlxsw_sp_fib4_node_entry_unlink(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib4_entry *fib4_entry) -{ - mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib4_entry->common); - mlxsw_sp_fib4_node_list_remove(fib4_entry); - - if (fib4_entry->common.type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP) - mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, &fib4_entry->common); -} - -static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib4_entry *fib4_entry, - bool replace) +mlxsw_sp_fib_node_entry_unlink(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) { - struct mlxsw_sp_fib_node *fib_node = fib4_entry->common.fib_node; - struct mlxsw_sp_fib4_entry *replaced; - - if (!replace) - return; - - /* We inserted the new entry before replaced one */ - replaced = list_next_entry(fib4_entry, common.list); + struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node; - mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, replaced); - mlxsw_sp_fib4_entry_destroy(mlxsw_sp, replaced); - mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); + mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry); + fib_node->fib_entry = NULL; } static int -mlxsw_sp_router_fib4_add(struct mlxsw_sp *mlxsw_sp, - const struct fib_entry_notifier_info *fen_info, - bool replace, bool append) +mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp, + const struct fib_entry_notifier_info *fen_info) { - struct mlxsw_sp_fib4_entry *fib4_entry; + struct mlxsw_sp_fib4_entry *fib4_entry, *fib4_replaced; + struct mlxsw_sp_fib_entry *replaced; struct mlxsw_sp_fib_node *fib_node; int err; @@ -4989,18 +4872,26 @@ mlxsw_sp_router_fib4_add(struct mlxsw_sp *mlxsw_sp, goto err_fib4_entry_create; } - err = mlxsw_sp_fib4_node_entry_link(mlxsw_sp, fib4_entry, replace, - append); + replaced = fib_node->fib_entry; + err = mlxsw_sp_fib_node_entry_link(mlxsw_sp, &fib4_entry->common); if (err) { dev_warn(mlxsw_sp->bus_info->dev, "Failed to link FIB entry to node\n"); - goto err_fib4_node_entry_link; + goto err_fib_node_entry_link; } - mlxsw_sp_fib4_entry_replace(mlxsw_sp, fib4_entry, replace); + /* Nothing to replace */ + if (!replaced) + return 0; + + mlxsw_sp_fib_entry_hw_flags_clear(mlxsw_sp, replaced); + fib4_replaced = container_of(replaced, struct mlxsw_sp_fib4_entry, + common); + mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_replaced); return 0; -err_fib4_node_entry_link: +err_fib_node_entry_link: + fib_node->fib_entry = replaced; mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry); err_fib4_entry_create: mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); @@ -5021,7 +4912,7 @@ static void mlxsw_sp_router_fib4_del(struct mlxsw_sp *mlxsw_sp, return; fib_node = fib4_entry->common.fib_node; - mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, fib4_entry); + mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib4_entry->common); mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry); mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); } @@ -5083,13 +4974,6 @@ static void mlxsw_sp_rt6_destroy(struct mlxsw_sp_rt6 *mlxsw_sp_rt6) kfree(mlxsw_sp_rt6); } -static bool mlxsw_sp_fib6_rt_can_mp(const struct fib6_info *rt) -{ - /* RTF_CACHE routes are ignored */ - return !(rt->fib6_flags & RTF_ADDRCONF) && - rt->fib6_nh->fib_nh_gw_family; -} - static struct fib6_info * mlxsw_sp_fib6_entry_rt(const struct mlxsw_sp_fib6_entry *fib6_entry) { @@ -5097,37 +4981,6 @@ mlxsw_sp_fib6_entry_rt(const struct mlxsw_sp_fib6_entry *fib6_entry) list)->rt; } -static struct mlxsw_sp_fib6_entry * -mlxsw_sp_fib6_node_mp_entry_find(const struct mlxsw_sp_fib_node *fib_node, - const struct fib6_info *nrt, bool replace) -{ - struct mlxsw_sp_fib6_entry *fib6_entry; - - if (!mlxsw_sp_fib6_rt_can_mp(nrt) || replace) - return NULL; - - list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) { - struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry); - - /* RT6_TABLE_LOCAL and RT6_TABLE_MAIN share the same - * virtual router. - */ - if (rt->fib6_table->tb6_id > nrt->fib6_table->tb6_id) - continue; - if (rt->fib6_table->tb6_id != nrt->fib6_table->tb6_id) - break; - if (rt->fib6_metric < nrt->fib6_metric) - continue; - if (rt->fib6_metric == nrt->fib6_metric && - mlxsw_sp_fib6_rt_can_mp(rt)) - return fib6_entry; - if (rt->fib6_metric > nrt->fib6_metric) - break; - } - - return NULL; -} - static struct mlxsw_sp_rt6 * mlxsw_sp_fib6_entry_rt_find(const struct mlxsw_sp_fib6_entry *fib6_entry, const struct fib6_info *rt) @@ -5313,6 +5166,11 @@ static int mlxsw_sp_nexthop6_group_get(struct mlxsw_sp *mlxsw_sp, &nh_grp->fib_list); fib6_entry->common.nh_group = nh_grp; + /* The route and the nexthop are described by the same struct, so we + * need to the update the nexthop offload indication for the new route. + */ + __mlxsw_sp_nexthop6_group_offload_refresh(nh_grp, fib6_entry); + return 0; } @@ -5345,16 +5203,16 @@ mlxsw_sp_nexthop6_group_update(struct mlxsw_sp *mlxsw_sp, * currently associated with it in the device's table is that * of the old group. Start using the new one instead. */ - err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common); + err = mlxsw_sp_fib_entry_update(mlxsw_sp, &fib6_entry->common); if (err) - goto err_fib_node_entry_add; + goto err_fib_entry_update; if (list_empty(&old_nh_grp->fib_list)) mlxsw_sp_nexthop6_group_destroy(mlxsw_sp, old_nh_grp); return 0; -err_fib_node_entry_add: +err_fib_entry_update: mlxsw_sp_nexthop6_group_put(mlxsw_sp, &fib6_entry->common); err_nexthop6_group_get: list_add_tail(&fib6_entry->common.nexthop_group_node, @@ -5519,112 +5377,13 @@ static void mlxsw_sp_fib6_entry_destroy(struct mlxsw_sp *mlxsw_sp, } static struct mlxsw_sp_fib6_entry * -mlxsw_sp_fib6_node_entry_find(const struct mlxsw_sp_fib_node *fib_node, - const struct fib6_info *nrt, bool replace) -{ - struct mlxsw_sp_fib6_entry *fib6_entry, *fallback = NULL; - - list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) { - struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry); - - if (rt->fib6_table->tb6_id > nrt->fib6_table->tb6_id) - continue; - if (rt->fib6_table->tb6_id != nrt->fib6_table->tb6_id) - break; - if (replace && rt->fib6_metric == nrt->fib6_metric) { - if (mlxsw_sp_fib6_rt_can_mp(rt) == - mlxsw_sp_fib6_rt_can_mp(nrt)) - return fib6_entry; - if (mlxsw_sp_fib6_rt_can_mp(nrt)) - fallback = fallback ?: fib6_entry; - } - if (rt->fib6_metric > nrt->fib6_metric) - return fallback ?: fib6_entry; - } - - return fallback; -} - -static int -mlxsw_sp_fib6_node_list_insert(struct mlxsw_sp_fib6_entry *new6_entry, - bool *p_replace) -{ - struct mlxsw_sp_fib_node *fib_node = new6_entry->common.fib_node; - struct fib6_info *nrt = mlxsw_sp_fib6_entry_rt(new6_entry); - struct mlxsw_sp_fib6_entry *fib6_entry; - - fib6_entry = mlxsw_sp_fib6_node_entry_find(fib_node, nrt, *p_replace); - - if (*p_replace && !fib6_entry) - *p_replace = false; - - if (fib6_entry) { - list_add_tail(&new6_entry->common.list, - &fib6_entry->common.list); - } else { - struct mlxsw_sp_fib6_entry *last; - - list_for_each_entry(last, &fib_node->entry_list, common.list) { - struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(last); - - if (nrt->fib6_table->tb6_id > rt->fib6_table->tb6_id) - break; - fib6_entry = last; - } - - if (fib6_entry) - list_add(&new6_entry->common.list, - &fib6_entry->common.list); - else - list_add(&new6_entry->common.list, - &fib_node->entry_list); - } - - return 0; -} - -static void -mlxsw_sp_fib6_node_list_remove(struct mlxsw_sp_fib6_entry *fib6_entry) -{ - list_del(&fib6_entry->common.list); -} - -static int mlxsw_sp_fib6_node_entry_link(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib6_entry *fib6_entry, - bool *p_replace) -{ - int err; - - err = mlxsw_sp_fib6_node_list_insert(fib6_entry, p_replace); - if (err) - return err; - - err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common); - if (err) - goto err_fib_node_entry_add; - - return 0; - -err_fib_node_entry_add: - mlxsw_sp_fib6_node_list_remove(fib6_entry); - return err; -} - -static void -mlxsw_sp_fib6_node_entry_unlink(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib6_entry *fib6_entry) -{ - mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib6_entry->common); - mlxsw_sp_fib6_node_list_remove(fib6_entry); -} - -static struct mlxsw_sp_fib6_entry * mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp, const struct fib6_info *rt) { struct mlxsw_sp_fib6_entry *fib6_entry; struct mlxsw_sp_fib_node *fib_node; struct mlxsw_sp_fib *fib; + struct fib6_info *cmp_rt; struct mlxsw_sp_vr *vr; vr = mlxsw_sp_vr_find(mlxsw_sp, rt->fib6_table->tb6_id); @@ -5638,40 +5397,23 @@ mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp, if (!fib_node) return NULL; - list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) { - struct fib6_info *iter_rt = mlxsw_sp_fib6_entry_rt(fib6_entry); - - if (rt->fib6_table->tb6_id == iter_rt->fib6_table->tb6_id && - rt->fib6_metric == iter_rt->fib6_metric && - mlxsw_sp_fib6_entry_rt_find(fib6_entry, rt)) - return fib6_entry; - } + fib6_entry = container_of(fib_node->fib_entry, + struct mlxsw_sp_fib6_entry, common); + cmp_rt = mlxsw_sp_fib6_entry_rt(fib6_entry); + if (rt->fib6_table->tb6_id == cmp_rt->fib6_table->tb6_id && + rt->fib6_metric == cmp_rt->fib6_metric && + mlxsw_sp_fib6_entry_rt_find(fib6_entry, rt)) + return fib6_entry; return NULL; } -static void mlxsw_sp_fib6_entry_replace(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fib6_entry *fib6_entry, - bool replace) -{ - struct mlxsw_sp_fib_node *fib_node = fib6_entry->common.fib_node; - struct mlxsw_sp_fib6_entry *replaced; - - if (!replace) - return; - - replaced = list_next_entry(fib6_entry, common.list); - - mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, replaced); - mlxsw_sp_fib6_entry_destroy(mlxsw_sp, replaced); - mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); -} - -static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp, - struct fib6_info **rt_arr, - unsigned int nrt6, bool replace) +static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp, + struct fib6_info **rt_arr, + unsigned int nrt6) { - struct mlxsw_sp_fib6_entry *fib6_entry; + struct mlxsw_sp_fib6_entry *fib6_entry, *fib6_replaced; + struct mlxsw_sp_fib_entry *replaced; struct mlxsw_sp_fib_node *fib_node; struct fib6_info *rt = rt_arr[0]; int err; @@ -5693,18 +5435,6 @@ static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp, if (IS_ERR(fib_node)) return PTR_ERR(fib_node); - /* Before creating a new entry, try to append route to an existing - * multipath entry. - */ - fib6_entry = mlxsw_sp_fib6_node_mp_entry_find(fib_node, rt, replace); - if (fib6_entry) { - err = mlxsw_sp_fib6_entry_nexthop_add(mlxsw_sp, fib6_entry, - rt_arr, nrt6); - if (err) - goto err_fib6_entry_nexthop_add; - return 0; - } - fib6_entry = mlxsw_sp_fib6_entry_create(mlxsw_sp, fib_node, rt_arr, nrt6); if (IS_ERR(fib6_entry)) { @@ -5712,17 +5442,70 @@ static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp, goto err_fib6_entry_create; } - err = mlxsw_sp_fib6_node_entry_link(mlxsw_sp, fib6_entry, &replace); + replaced = fib_node->fib_entry; + err = mlxsw_sp_fib_node_entry_link(mlxsw_sp, &fib6_entry->common); if (err) - goto err_fib6_node_entry_link; + goto err_fib_node_entry_link; + + /* Nothing to replace */ + if (!replaced) + return 0; - mlxsw_sp_fib6_entry_replace(mlxsw_sp, fib6_entry, replace); + mlxsw_sp_fib_entry_hw_flags_clear(mlxsw_sp, replaced); + fib6_replaced = container_of(replaced, struct mlxsw_sp_fib6_entry, + common); + mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_replaced); return 0; -err_fib6_node_entry_link: +err_fib_node_entry_link: + fib_node->fib_entry = replaced; mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry); err_fib6_entry_create: + mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); + return err; +} + +static int mlxsw_sp_router_fib6_append(struct mlxsw_sp *mlxsw_sp, + struct fib6_info **rt_arr, + unsigned int nrt6) +{ + struct mlxsw_sp_fib6_entry *fib6_entry; + struct mlxsw_sp_fib_node *fib_node; + struct fib6_info *rt = rt_arr[0]; + int err; + + if (mlxsw_sp->router->aborted) + return 0; + + if (rt->fib6_src.plen) + return -EINVAL; + + if (mlxsw_sp_fib6_rt_should_ignore(rt)) + return 0; + + fib_node = mlxsw_sp_fib_node_get(mlxsw_sp, rt->fib6_table->tb6_id, + &rt->fib6_dst.addr, + sizeof(rt->fib6_dst.addr), + rt->fib6_dst.plen, + MLXSW_SP_L3_PROTO_IPV6); + if (IS_ERR(fib_node)) + return PTR_ERR(fib_node); + + if (WARN_ON_ONCE(!fib_node->fib_entry)) { + mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); + return -EINVAL; + } + + fib6_entry = container_of(fib_node->fib_entry, + struct mlxsw_sp_fib6_entry, common); + err = mlxsw_sp_fib6_entry_nexthop_add(mlxsw_sp, fib6_entry, rt_arr, + nrt6); + if (err) + goto err_fib6_entry_nexthop_add; + + return 0; + err_fib6_entry_nexthop_add: mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); return err; @@ -5762,7 +5545,7 @@ static void mlxsw_sp_router_fib6_del(struct mlxsw_sp *mlxsw_sp, fib_node = fib6_entry->common.fib_node; - mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, fib6_entry); + mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, &fib6_entry->common); mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry); mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); } @@ -5916,39 +5699,25 @@ static int mlxsw_sp_router_set_abort_trap(struct mlxsw_sp *mlxsw_sp) static void mlxsw_sp_fib4_node_flush(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_node *fib_node) { - struct mlxsw_sp_fib4_entry *fib4_entry, *tmp; - - list_for_each_entry_safe(fib4_entry, tmp, &fib_node->entry_list, - common.list) { - bool do_break = &tmp->common.list == &fib_node->entry_list; + struct mlxsw_sp_fib4_entry *fib4_entry; - mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, fib4_entry); - mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry); - mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); - /* Break when entry list is empty and node was freed. - * Otherwise, we'll access freed memory in the next - * iteration. - */ - if (do_break) - break; - } + fib4_entry = container_of(fib_node->fib_entry, + struct mlxsw_sp_fib4_entry, common); + mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, fib_node->fib_entry); + mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry); + mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); } static void mlxsw_sp_fib6_node_flush(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_node *fib_node) { - struct mlxsw_sp_fib6_entry *fib6_entry, *tmp; - - list_for_each_entry_safe(fib6_entry, tmp, &fib_node->entry_list, - common.list) { - bool do_break = &tmp->common.list == &fib_node->entry_list; + struct mlxsw_sp_fib6_entry *fib6_entry; - mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, fib6_entry); - mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry); - mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); - if (do_break) - break; - } + fib6_entry = container_of(fib_node->fib_entry, + struct mlxsw_sp_fib6_entry, common); + mlxsw_sp_fib_node_entry_unlink(mlxsw_sp, fib_node->fib_entry); + mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry); + mlxsw_sp_fib_node_put(mlxsw_sp, fib_node); } static void mlxsw_sp_fib_node_flush(struct mlxsw_sp *mlxsw_sp, @@ -6099,7 +5868,6 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work) struct mlxsw_sp_fib_event_work *fib_work = container_of(work, struct mlxsw_sp_fib_event_work, work); struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp; - bool replace, append; int err; /* Protect internal structures from changes */ @@ -6107,13 +5875,9 @@ static void mlxsw_sp_router_fib4_event_work(struct work_struct *work) mlxsw_sp_span_respin(mlxsw_sp); switch (fib_work->event) { - case FIB_EVENT_ENTRY_REPLACE: /* fall through */ - case FIB_EVENT_ENTRY_APPEND: /* fall through */ - case FIB_EVENT_ENTRY_ADD: - replace = fib_work->event == FIB_EVENT_ENTRY_REPLACE; - append = fib_work->event == FIB_EVENT_ENTRY_APPEND; - err = mlxsw_sp_router_fib4_add(mlxsw_sp, &fib_work->fen_info, - replace, append); + case FIB_EVENT_ENTRY_REPLACE: + err = mlxsw_sp_router_fib4_replace(mlxsw_sp, + &fib_work->fen_info); if (err) mlxsw_sp_router_fib_abort(mlxsw_sp); fib_info_put(fib_work->fen_info.fi); @@ -6138,20 +5902,24 @@ static void mlxsw_sp_router_fib6_event_work(struct work_struct *work) struct mlxsw_sp_fib_event_work *fib_work = container_of(work, struct mlxsw_sp_fib_event_work, work); struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp; - bool replace; int err; rtnl_lock(); mlxsw_sp_span_respin(mlxsw_sp); switch (fib_work->event) { - case FIB_EVENT_ENTRY_REPLACE: /* fall through */ - case FIB_EVENT_ENTRY_ADD: - replace = fib_work->event == FIB_EVENT_ENTRY_REPLACE; - err = mlxsw_sp_router_fib6_add(mlxsw_sp, - fib_work->fib6_work.rt_arr, - fib_work->fib6_work.nrt6, - replace); + case FIB_EVENT_ENTRY_REPLACE: + err = mlxsw_sp_router_fib6_replace(mlxsw_sp, + fib_work->fib6_work.rt_arr, + fib_work->fib6_work.nrt6); + if (err) + mlxsw_sp_router_fib_abort(mlxsw_sp); + mlxsw_sp_router_fib6_work_fini(&fib_work->fib6_work); + break; + case FIB_EVENT_ENTRY_APPEND: + err = mlxsw_sp_router_fib6_append(mlxsw_sp, + fib_work->fib6_work.rt_arr, + fib_work->fib6_work.nrt6); if (err) mlxsw_sp_router_fib_abort(mlxsw_sp); mlxsw_sp_router_fib6_work_fini(&fib_work->fib6_work); @@ -6216,8 +5984,6 @@ static void mlxsw_sp_router_fib4_event(struct mlxsw_sp_fib_event_work *fib_work, switch (fib_work->event) { case FIB_EVENT_ENTRY_REPLACE: /* fall through */ - case FIB_EVENT_ENTRY_APPEND: /* fall through */ - case FIB_EVENT_ENTRY_ADD: /* fall through */ case FIB_EVENT_ENTRY_DEL: fen_info = container_of(info, struct fib_entry_notifier_info, info); @@ -6245,7 +6011,7 @@ static int mlxsw_sp_router_fib6_event(struct mlxsw_sp_fib_event_work *fib_work, switch (fib_work->event) { case FIB_EVENT_ENTRY_REPLACE: /* fall through */ - case FIB_EVENT_ENTRY_ADD: /* fall through */ + case FIB_EVENT_ENTRY_APPEND: /* fall through */ case FIB_EVENT_ENTRY_DEL: fen6_info = container_of(info, struct fib6_entry_notifier_info, info); @@ -6348,9 +6114,9 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb, err = mlxsw_sp_router_fib_rule_event(event, info, router->mlxsw_sp); return notifier_from_errno(err); - case FIB_EVENT_ENTRY_ADD: + case FIB_EVENT_ENTRY_ADD: /* fall through */ case FIB_EVENT_ENTRY_REPLACE: /* fall through */ - case FIB_EVENT_ENTRY_APPEND: /* fall through */ + case FIB_EVENT_ENTRY_APPEND: if (router->aborted) { NL_SET_ERR_MSG_MOD(info->extack, "FIB offload was aborted. Not configuring route"); return notifier_from_errno(-EINVAL); @@ -8025,8 +7791,18 @@ mlxsw_sp_ipip_config_tigcr(struct mlxsw_sp *mlxsw_sp) static int mlxsw_sp_ipips_init(struct mlxsw_sp *mlxsw_sp) { + int err; + mlxsw_sp->router->ipip_ops_arr = mlxsw_sp_ipip_ops_arr; INIT_LIST_HEAD(&mlxsw_sp->router->ipip_list); + + err = mlxsw_sp_ipip_ecn_encap_init(mlxsw_sp); + if (err) + return err; + err = mlxsw_sp_ipip_ecn_decap_init(mlxsw_sp); + if (err) + return err; + return mlxsw_sp_ipip_config_tigcr(mlxsw_sp); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h index cc1de91e8217..c9b94f435cdd 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h @@ -104,4 +104,7 @@ static inline bool mlxsw_sp_l3addr_eq(const union mlxsw_sp_l3addr *addr1, return !memcmp(addr1, addr2, sizeof(*addr1)); } +int mlxsw_sp_ipip_ecn_encap_init(struct mlxsw_sp *mlxsw_sp); +int mlxsw_sp_ipip_ecn_decap_init(struct mlxsw_sp *mlxsw_sp); + #endif /* _MLXSW_ROUTER_H_*/ diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index 200d324e6d99..0cdd7954a085 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -748,33 +748,50 @@ static bool mlxsw_sp_span_is_egress_mirror(struct mlxsw_sp_port *port) return false; } -static int mlxsw_sp_span_mtu_to_buffsize(const struct mlxsw_sp *mlxsw_sp, - int mtu) +static int +mlxsw_sp_span_port_buffsize_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) { - return mlxsw_sp_bytes_cells(mlxsw_sp, mtu * 5 / 2) + 1; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char sbib_pl[MLXSW_REG_SBIB_LEN]; + u32 buffsize; + u32 speed; + int err; + + err = mlxsw_sp_port_speed_get(mlxsw_sp_port, &speed); + if (err) + return err; + if (speed == SPEED_UNKNOWN) + speed = 0; + + buffsize = mlxsw_sp_span_buffsize_get(mlxsw_sp, speed, mtu); + mlxsw_reg_sbib_pack(sbib_pl, mlxsw_sp_port->local_port, buffsize); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); } int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu) { - struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; - char sbib_pl[MLXSW_REG_SBIB_LEN]; - int err; - /* If port is egress mirrored, the shared buffer size should be * updated according to the mtu value */ - if (mlxsw_sp_span_is_egress_mirror(port)) { - u32 buffsize = mlxsw_sp_span_mtu_to_buffsize(mlxsw_sp, mtu); + if (mlxsw_sp_span_is_egress_mirror(port)) + return mlxsw_sp_span_port_buffsize_update(port, mtu); + return 0; +} - mlxsw_reg_sbib_pack(sbib_pl, port->local_port, buffsize); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); - if (err) { - netdev_err(port->dev, "Could not update shared buffer for mirroring\n"); - return err; - } - } +void mlxsw_sp_span_speed_update_work(struct work_struct *work) +{ + struct delayed_work *dwork = to_delayed_work(work); + struct mlxsw_sp_port *mlxsw_sp_port; - return 0; + mlxsw_sp_port = container_of(dwork, struct mlxsw_sp_port, + span.speed_update_dw); + + /* If port is egress mirrored, the shared buffer size should be + * updated according to the speed value. + */ + if (mlxsw_sp_span_is_egress_mirror(mlxsw_sp_port)) + mlxsw_sp_span_port_buffsize_update(mlxsw_sp_port, + mlxsw_sp_port->dev->mtu); } static struct mlxsw_sp_span_inspected_port * @@ -836,15 +853,9 @@ mlxsw_sp_span_inspected_port_add(struct mlxsw_sp_port *port, /* if it is an egress SPAN, bind a shared buffer to it */ if (type == MLXSW_SP_SPAN_EGRESS) { - u32 buffsize = mlxsw_sp_span_mtu_to_buffsize(mlxsw_sp, - port->dev->mtu); - - mlxsw_reg_sbib_pack(sbib_pl, port->local_port, buffsize); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); - if (err) { - netdev_err(port->dev, "Could not create shared buffer for mirroring\n"); + err = mlxsw_sp_span_port_buffsize_update(port, port->dev->mtu); + if (err) return err; - } } if (bind) { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h index 5e04252f2a11..59724335525f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h @@ -74,5 +74,6 @@ void mlxsw_sp_span_entry_invalidate(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_span_entry *span_entry); int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu); +void mlxsw_sp_span_speed_update_work(struct work_struct *work); #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c index e0d7c49ffae0..60205aa3f6a5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c @@ -9,6 +9,20 @@ #include "reg.h" #include "spectrum.h" +/* All driver-specific traps must be documented in + * Documentation/networking/devlink/mlxsw.rst + */ +enum { + DEVLINK_MLXSW_TRAP_ID_BASE = DEVLINK_TRAP_GENERIC_ID_MAX, + DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED, + DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED, +}; + +#define DEVLINK_MLXSW_TRAP_NAME_IRIF_DISABLED \ + "irif_disabled" +#define DEVLINK_MLXSW_TRAP_NAME_ERIF_DISABLED \ + "erif_disabled" + #define MLXSW_SP_TRAP_METADATA DEVLINK_TRAP_METADATA_TYPE_F_IN_PORT static void mlxsw_sp_rx_drop_listener(struct sk_buff *skb, u8 local_port, @@ -21,6 +35,12 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, DEVLINK_TRAP_GROUP_GENERIC(_group_id), \ MLXSW_SP_TRAP_METADATA) +#define MLXSW_SP_TRAP_DRIVER_DROP(_id, _group_id) \ + DEVLINK_TRAP_DRIVER(DROP, DROP, DEVLINK_MLXSW_TRAP_ID_##_id, \ + DEVLINK_MLXSW_TRAP_NAME_##_id, \ + DEVLINK_TRAP_GROUP_GENERIC(_group_id), \ + MLXSW_SP_TRAP_METADATA) + #define MLXSW_SP_TRAP_EXCEPTION(_id, _group_id) \ DEVLINK_TRAP_GENERIC(EXCEPTION, TRAP, _id, \ DEVLINK_TRAP_GROUP_GENERIC(_group_id), \ @@ -58,6 +78,11 @@ static struct devlink_trap mlxsw_sp_traps_arr[] = { MLXSW_SP_TRAP_EXCEPTION(UNRESOLVED_NEIGH, L3_DROPS), MLXSW_SP_TRAP_EXCEPTION(IPV4_LPM_UNICAST_MISS, L3_DROPS), MLXSW_SP_TRAP_EXCEPTION(IPV6_LPM_UNICAST_MISS, L3_DROPS), + MLXSW_SP_TRAP_DRIVER_DROP(IRIF_DISABLED, L3_DROPS), + MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS), + MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS), + MLXSW_SP_TRAP_EXCEPTION(DECAP_ERROR, TUNNEL_DROPS), + MLXSW_SP_TRAP_DROP(OVERLAY_SMAC_MC, TUNNEL_DROPS), }; static struct mlxsw_listener mlxsw_sp_listeners_arr[] = { @@ -90,6 +115,15 @@ static struct mlxsw_listener mlxsw_sp_listeners_arr[] = { TRAP_EXCEPTION_TO_CPU), MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM6, ROUTER_EXP, TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS), + MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS), + MLXSW_SP_RXL_DISCARD(NON_ROUTABLE, L3_DISCARDS), + MLXSW_SP_RXL_EXCEPTION(DECAP_ECN0, ROUTER_EXP, TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(IPIP_DECAP_ERROR, ROUTER_EXP, + TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(DISCARD_DEC_PKT, TUNNEL_DISCARDS, + TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_DISCARD(OVERLAY_SMAC_MC, TUNNEL_DISCARDS), }; /* Mapping between hardware trap and devlink trap. Multiple hardware traps can @@ -123,6 +157,13 @@ static u16 mlxsw_sp_listener_devlink_map[] = { DEVLINK_TRAP_GENERIC_ID_UNRESOLVED_NEIGH, DEVLINK_TRAP_GENERIC_ID_IPV4_LPM_UNICAST_MISS, DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS, + DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED, + DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED, + DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE, + DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, + DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, + DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, + DEVLINK_TRAP_GENERIC_ID_OVERLAY_SMAC_MC, }; static int mlxsw_sp_rx_listener(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, @@ -304,8 +345,9 @@ mlxsw_sp_trap_group_policer_init(struct mlxsw_sp *mlxsw_sp, u32 rate; switch (group->id) { - case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS:/* fall through */ - case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: + case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: /* fall through */ + case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS: /* fall through */ + case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS: policer_id = MLXSW_SP_DISCARD_POLICER_ID; ir_units = MLXSW_REG_QPCR_IR_UNITS_M; is_bytes = false; @@ -342,6 +384,12 @@ __mlxsw_sp_trap_group_init(struct mlxsw_sp *mlxsw_sp, priority = 0; tc = 1; break; + case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS: + group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS; + policer_id = MLXSW_SP_DISCARD_POLICER_ID; + priority = 0; + tc = 1; + break; default: return -EINVAL; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h index 0c1c142bb6b0..12e1fa998d42 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h @@ -67,6 +67,7 @@ enum { MLXSW_TRAP_ID_NVE_ENCAP_ARP = 0xBD, MLXSW_TRAP_ID_ROUTER_ALERT_IPV4 = 0xD6, MLXSW_TRAP_ID_ROUTER_ALERT_IPV6 = 0xD7, + MLXSW_TRAP_ID_DISCARD_NON_ROUTABLE = 0x11A, MLXSW_TRAP_ID_DISCARD_ROUTER2 = 0x130, MLXSW_TRAP_ID_DISCARD_ROUTER3 = 0x131, MLXSW_TRAP_ID_DISCARD_ING_PACKET_SMAC_MC = 0x140, @@ -80,12 +81,20 @@ enum { MLXSW_TRAP_ID_DISCARD_ING_ROUTER_UC_DIP_MC_DMAC = 0x161, MLXSW_TRAP_ID_DISCARD_ING_ROUTER_DIP_LB = 0x162, MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_MC = 0x163, + MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_CLASS_E = 0x164, MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_LB = 0x165, MLXSW_TRAP_ID_DISCARD_ING_ROUTER_CORRUPTED_IP_HDR = 0x167, + MLXSW_TRAP_ID_DISCARD_ING_ROUTER_MC_DMAC = 0x168, + MLXSW_TRAP_ID_DISCARD_ING_ROUTER_SIP_DIP = 0x169, MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_SIP_BC = 0x16A, MLXSW_TRAP_ID_DISCARD_ING_ROUTER_IPV4_DIP_LOCAL_NET = 0x16B, + MLXSW_TRAP_ID_DISCARD_ING_ROUTER_DIP_LINK_LOCAL = 0x16C, + MLXSW_TRAP_ID_DISCARD_ROUTER_IRIF_EN = 0x178, + MLXSW_TRAP_ID_DISCARD_ROUTER_ERIF_EN = 0x179, MLXSW_TRAP_ID_DISCARD_ROUTER_LPM4 = 0x17B, MLXSW_TRAP_ID_DISCARD_ROUTER_LPM6 = 0x17C, + MLXSW_TRAP_ID_DISCARD_DEC_PKT = 0x188, + MLXSW_TRAP_ID_DISCARD_OVERLAY_SMAC_MC = 0x190, MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_RESERVED_SCOPE = 0x1B0, MLXSW_TRAP_ID_DISCARD_IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE = 0x1B1, MLXSW_TRAP_ID_ACL0 = 0x1C0, |