diff options
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx5/core')
85 files changed, 4439 insertions, 1225 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index a22c32aabf11..cd4a1ab0ea78 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -111,6 +111,7 @@ mlx5_core-$(CONFIG_MLX5_SW_STEERING) += steering/dr_domain.o steering/dr_table.o steering/dr_ste_v2.o \ steering/dr_cmd.o steering/dr_fw.o \ steering/dr_action.o steering/fs_dr.o \ + steering/dr_definer.o \ steering/dr_dbg.o lib/smfs.o # # SF device diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index e7a894ba5c3e..d3ca745d107d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -37,7 +37,6 @@ #include <linux/slab.h> #include <linux/delay.h> #include <linux/random.h> -#include <linux/io-mapping.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/eq.h> #include <linux/debugfs.h> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index 66c6a7017695..ddb197970c22 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -46,10 +46,6 @@ mlx5_devlink_info_get(struct devlink *devlink, struct devlink_info_req *req, u32 running_fw, stored_fw; int err; - err = devlink_info_driver_name_put(req, KBUILD_MODNAME); - if (err) - return err; - err = devlink_info_version_fixed_put(req, "fw.psid", dev->board_id); if (err) return err; @@ -318,6 +314,10 @@ static const struct devlink_ops mlx5_devlink_ops = { .rate_node_new = mlx5_esw_devlink_rate_node_new, .rate_node_del = mlx5_esw_devlink_rate_node_del, .rate_leaf_parent_set = mlx5_esw_devlink_rate_parent_set, + .port_fn_roce_get = mlx5_devlink_port_fn_roce_get, + .port_fn_roce_set = mlx5_devlink_port_fn_roce_set, + .port_fn_migratable_get = mlx5_devlink_port_fn_migratable_get, + .port_fn_migratable_set = mlx5_devlink_port_fn_migratable_set, #endif #ifdef CONFIG_MLX5_SF_MANAGER .port_new = mlx5_devlink_sf_port_new, @@ -840,7 +840,7 @@ static const struct devlink_trap_group mlx5_trap_groups_arr[] = { DEVLINK_TRAP_GROUP_GENERIC(L2_DROPS, 0), }; -static int mlx5_devlink_traps_register(struct devlink *devlink) +int mlx5_devlink_traps_register(struct devlink *devlink) { struct mlx5_core_dev *core_dev = devlink_priv(devlink); int err; @@ -862,7 +862,7 @@ err_trap_group: return err; } -static void mlx5_devlink_traps_unregister(struct devlink *devlink) +void mlx5_devlink_traps_unregister(struct devlink *devlink) { devl_traps_unregister(devlink, mlx5_traps_arr, ARRAY_SIZE(mlx5_traps_arr)); devl_trap_groups_unregister(devlink, mlx5_trap_groups_arr, @@ -889,17 +889,11 @@ int mlx5_devlink_register(struct devlink *devlink) if (err) goto max_uc_list_err; - err = mlx5_devlink_traps_register(devlink); - if (err) - goto traps_reg_err; - if (!mlx5_core_is_mp_slave(dev)) devlink_set_features(devlink, DEVLINK_F_RELOAD); return 0; -traps_reg_err: - mlx5_devlink_max_uc_list_param_unregister(devlink); max_uc_list_err: mlx5_devlink_auxdev_params_unregister(devlink); auxdev_reg_err: @@ -910,7 +904,6 @@ auxdev_reg_err: void mlx5_devlink_unregister(struct devlink *devlink) { - mlx5_devlink_traps_unregister(devlink); mlx5_devlink_max_uc_list_param_unregister(devlink); mlx5_devlink_auxdev_params_unregister(devlink); devlink_params_unregister(devlink, mlx5_devlink_params, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h index 30bf4882779b..fd033df24856 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.h @@ -30,6 +30,8 @@ void mlx5_devlink_trap_report(struct mlx5_core_dev *dev, int trap_id, struct sk_ int mlx5_devlink_trap_get_num_active(struct mlx5_core_dev *dev); int mlx5_devlink_traps_get_action(struct mlx5_core_dev *dev, int trap_id, enum devlink_trap_action *action); +int mlx5_devlink_traps_register(struct devlink *devlink); +void mlx5_devlink_traps_unregister(struct devlink *devlink); struct devlink *mlx5_devlink_alloc(struct device *dev); void mlx5_devlink_free(struct devlink *devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c index c5bb79a4fa57..2732128e7a6e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c @@ -228,6 +228,17 @@ const char *parse_fs_hdrs(struct trace_seq *p, return ret; } +static const char +*fs_dest_range_field_to_str(enum mlx5_flow_dest_range_field field) +{ + switch (field) { + case MLX5_FLOW_DEST_RANGE_FIELD_PKT_LEN: + return "packet len"; + default: + return "unknown dest range field"; + } +} + const char *parse_fs_dst(struct trace_seq *p, const struct mlx5_flow_destination *dst, u32 counter_id) @@ -259,6 +270,11 @@ const char *parse_fs_dst(struct trace_seq *p, case MLX5_FLOW_DESTINATION_TYPE_PORT: trace_seq_printf(p, "port\n"); break; + case MLX5_FLOW_DESTINATION_TYPE_RANGE: + trace_seq_printf(p, "field=%s min=%d max=%d\n", + fs_dest_range_field_to_str(dst->range.field), + dst->range.min, dst->range.max); + break; case MLX5_FLOW_DESTINATION_TYPE_NONE: trace_seq_printf(p, "none\n"); break; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 26a23047f1f3..2d77fb8a8a01 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -103,11 +103,11 @@ struct page_pool; * size actually used at runtime, but it's not a problem when calculating static * array sizes. */ -#define MLX5_UMR_MAX_MTT_SPACE \ +#define MLX5_UMR_MAX_FLEX_SPACE \ (ALIGN_DOWN(MLX5_SEND_WQE_MAX_SIZE - sizeof(struct mlx5e_umr_wqe), \ - MLX5_UMR_MTT_ALIGNMENT)) + MLX5_UMR_FLEX_ALIGNMENT)) #define MLX5_MPWRQ_MAX_PAGES_PER_WQE \ - rounddown_pow_of_two(MLX5_UMR_MAX_MTT_SPACE / sizeof(struct mlx5_mtt)) + rounddown_pow_of_two(MLX5_UMR_MAX_FLEX_SPACE / sizeof(struct mlx5_mtt)) #define MLX5E_MAX_RQ_NUM_MTTS \ (ALIGN_DOWN(U16_MAX, 4) * 2) /* Fits into u16 and aligned by WQEBB. */ @@ -160,7 +160,7 @@ struct page_pool; (((wqe_size) - sizeof(struct mlx5e_umr_wqe)) / sizeof(struct mlx5_klm)) #define MLX5E_KLM_ENTRIES_PER_WQE(wqe_size)\ - ALIGN_DOWN(MLX5E_KLM_MAX_ENTRIES_PER_WQE(wqe_size), MLX5_UMR_KLM_ALIGNMENT) + ALIGN_DOWN(MLX5E_KLM_MAX_ENTRIES_PER_WQE(wqe_size), MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT) #define MLX5E_MAX_KLM_PER_WQE(mdev) \ MLX5E_KLM_ENTRIES_PER_WQE(MLX5_SEND_WQE_BB * mlx5e_get_max_sq_aligned_wqebbs(mdev)) @@ -344,6 +344,7 @@ enum { MLX5E_RQ_STATE_CSUM_FULL, /* cqe_csum_full hw bit is set */ MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, /* set when mini_cqe_resp_stride_index cap is used */ MLX5E_RQ_STATE_SHAMPO, /* set when SHAMPO cap is used */ + MLX5E_RQ_STATE_MINI_CQE_ENHANCED, /* set when enhanced mini_cqe_cap is used */ }; struct mlx5e_cq { @@ -370,6 +371,7 @@ struct mlx5e_cq_decomp { u8 mini_arr_idx; u16 left; u16 wqe_counter; + bool last_cqe_title; } ____cacheline_aligned_in_smp; enum mlx5e_dma_map_type { @@ -1243,4 +1245,5 @@ int mlx5e_set_vf_rate(struct net_device *dev, int vf, int min_tx_rate, int max_t int mlx5e_get_vf_config(struct net_device *dev, int vf, struct ifla_vf_info *ivi); int mlx5e_get_vf_stats(struct net_device *dev, int vf, struct ifla_vf_stats *vf_stats); #endif +int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, u32 *mkey); #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c index b69f9d10ccbd..83adaabf59f5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.c @@ -51,13 +51,6 @@ int mlx5e_devlink_port_register(struct mlx5e_priv *priv) return ret; } -void mlx5e_devlink_port_type_eth_set(struct mlx5e_priv *priv) -{ - struct devlink_port *dl_port = mlx5e_devlink_get_dl_port(priv); - - devlink_port_type_eth_set(dl_port, priv->netdev); -} - void mlx5e_devlink_port_unregister(struct mlx5e_priv *priv) { struct devlink_port *dl_port = mlx5e_devlink_get_dl_port(priv); @@ -69,13 +62,3 @@ void mlx5e_devlink_port_unregister(struct mlx5e_priv *priv) if (!(priv->mdev->priv.flags & MLX5_PRIV_FLAGS_MLX5E_LOCKED_FLOW)) devl_unlock(devlink); } - -struct devlink_port *mlx5e_get_devlink_port(struct net_device *dev) -{ - struct mlx5e_priv *priv = netdev_priv(dev); - - if (!netif_device_present(dev)) - return NULL; - - return mlx5e_devlink_get_dl_port(priv); -} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h index 10b50feb9883..4f238d4fff55 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/devlink.h @@ -9,8 +9,6 @@ int mlx5e_devlink_port_register(struct mlx5e_priv *priv); void mlx5e_devlink_port_unregister(struct mlx5e_priv *priv); -void mlx5e_devlink_port_type_eth_set(struct mlx5e_priv *priv); -struct devlink_port *mlx5e_get_devlink_port(struct net_device *dev); static inline struct devlink_port * mlx5e_devlink_get_dl_port(struct mlx5e_priv *priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h index bf2741eb7f9b..379c6dc9a3be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h @@ -84,7 +84,8 @@ enum { MLX5E_ARFS_FT_LEVEL = MLX5E_INNER_TTC_FT_LEVEL + 1, #endif #ifdef CONFIG_MLX5_EN_IPSEC - MLX5E_ACCEL_FS_ESP_FT_LEVEL = MLX5E_INNER_TTC_FT_LEVEL + 1, + MLX5E_ACCEL_FS_POL_FT_LEVEL = MLX5E_INNER_TTC_FT_LEVEL + 1, + MLX5E_ACCEL_FS_ESP_FT_LEVEL, MLX5E_ACCEL_FS_ESP_FT_ERR_LEVEL, #endif }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c index 29dd3a04c154..585bdc8383ee 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c @@ -107,7 +107,7 @@ u8 mlx5e_mpwrq_log_wqe_sz(struct mlx5_core_dev *mdev, u8 page_shift, /* Keep in sync with MLX5_MPWRQ_MAX_PAGES_PER_WQE. */ max_wqe_size = mlx5e_get_max_sq_aligned_wqebbs(mdev) * MLX5_SEND_WQE_BB; max_pages_per_wqe = ALIGN_DOWN(max_wqe_size - sizeof(struct mlx5e_umr_wqe), - MLX5_UMR_MTT_ALIGNMENT) / umr_entry_size; + MLX5_UMR_FLEX_ALIGNMENT) / umr_entry_size; max_log_mpwqe_size = ilog2(max_pages_per_wqe) + page_shift; WARN_ON_ONCE(max_log_mpwqe_size < MLX5E_ORDER2_MAX_PACKET_MTU); @@ -146,7 +146,7 @@ u16 mlx5e_mpwrq_umr_wqe_sz(struct mlx5_core_dev *mdev, u8 page_shift, u16 umr_wqe_sz; umr_wqe_sz = sizeof(struct mlx5e_umr_wqe) + - ALIGN(pages_per_wqe * umr_entry_size, MLX5_UMR_MTT_ALIGNMENT); + ALIGN(pages_per_wqe * umr_entry_size, MLX5_UMR_FLEX_ALIGNMENT); WARN_ON_ONCE(DIV_ROUND_UP(umr_wqe_sz, MLX5_SEND_WQE_DS) > MLX5_WQE_CTRL_DS_MASK); @@ -607,14 +607,6 @@ void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev, params->log_rq_mtu_frames = is_kdump_kernel() ? MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE : MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE; - - mlx5_core_info(mdev, "MLX5E: StrdRq(%d) RqSz(%ld) StrdSz(%ld) RxCqeCmprss(%d)\n", - params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ, - params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ ? - BIT(mlx5e_mpwqe_get_log_rq_size(mdev, params, NULL)) : - BIT(params->log_rq_mtu_frames), - BIT(mlx5e_mpwqe_get_log_stride_size(mdev, params, NULL)), - MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)); } void mlx5e_set_rq_type(struct mlx5_core_dev *mdev, struct mlx5e_params *params) @@ -852,6 +844,10 @@ static void mlx5e_build_rx_cq_param(struct mlx5_core_dev *mdev, if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)) { MLX5_SET(cqc, cqc, mini_cqe_res_format, hw_stridx ? MLX5_CQE_FORMAT_CSUM_STRIDX : MLX5_CQE_FORMAT_CSUM); + MLX5_SET(cqc, cqc, cqe_compression_layout, + MLX5_CAP_GEN(mdev, enhanced_cqe_compression) ? + MLX5_CQE_COMPRESS_LAYOUT_ENHANCED : + MLX5_CQE_COMPRESS_LAYOUT_BASIC); MLX5_SET(cqc, cqc, cqe_comp_en, 1); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h index 034debd140bc..c9be6eb88012 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.h @@ -154,4 +154,18 @@ int mlx5e_build_channel_param(struct mlx5_core_dev *mdev, u16 mlx5e_calc_sq_stop_room(struct mlx5_core_dev *mdev, struct mlx5e_params *params); int mlx5e_validate_params(struct mlx5_core_dev *mdev, struct mlx5e_params *params); +static inline void mlx5e_params_print_info(struct mlx5_core_dev *mdev, + struct mlx5e_params *params) +{ + mlx5_core_info(mdev, "MLX5E: StrdRq(%d) RqSz(%ld) StrdSz(%ld) RxCqeCmprss(%d %s)\n", + params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ, + params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ ? + BIT(mlx5e_mpwqe_get_log_rq_size(mdev, params, NULL)) : + BIT(params->log_rq_mtu_frames), + BIT(mlx5e_mpwqe_get_log_stride_size(mdev, params, NULL)), + MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS), + MLX5_CAP_GEN(mdev, enhanced_cqe_compression) ? + "enhanced" : "basic"); +}; + #endif /* __MLX5_EN_PARAMS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c index fac7e3ff2674..b08339d986d5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c @@ -690,7 +690,6 @@ static bool mlx5e_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb, err = mapping_find(uplink_priv->tunnel_mapping, tun_id, &key); if (err) { - WARN_ON_ONCE(true); netdev_dbg(priv->netdev, "Couldn't find tunnel for tun_id: %d, err: %d\n", tun_id, err); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c index 21aab96357b5..a278f52d52b0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/accept.c @@ -28,4 +28,5 @@ tc_act_parse_accept(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5e_tc_act mlx5e_tc_act_accept = { .can_offload = tc_act_can_offload_accept, .parse_action = tc_act_parse_accept, + .is_terminating_action = true, }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c index 3337241cfd84..eba0c8698926 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c @@ -11,7 +11,7 @@ static struct mlx5e_tc_act *tc_acts_fdb[NUM_FLOW_ACTIONS] = { [FLOW_ACTION_DROP] = &mlx5e_tc_act_drop, [FLOW_ACTION_TRAP] = &mlx5e_tc_act_trap, [FLOW_ACTION_GOTO] = &mlx5e_tc_act_goto, - [FLOW_ACTION_REDIRECT] = &mlx5e_tc_act_mirred, + [FLOW_ACTION_REDIRECT] = &mlx5e_tc_act_redirect, [FLOW_ACTION_MIRRED] = &mlx5e_tc_act_mirred, [FLOW_ACTION_REDIRECT_INGRESS] = &mlx5e_tc_act_redirect_ingress, [FLOW_ACTION_VLAN_PUSH] = &mlx5e_tc_act_vlan, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h index e1570ff056ae..8346557eeaf6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.h @@ -32,6 +32,11 @@ struct mlx5e_tc_act_parse_state { struct mlx5_tc_ct_priv *ct_priv; }; +struct mlx5e_tc_act_branch_ctrl { + enum flow_action_id act_id; + u32 extval; +}; + struct mlx5e_tc_act { bool (*can_offload)(struct mlx5e_tc_act_parse_state *parse_state, const struct flow_action_entry *act, @@ -60,6 +65,12 @@ struct mlx5e_tc_act { int (*stats_action)(struct mlx5e_priv *priv, struct flow_offload_action *fl_act); + + bool (*get_branch_ctrl)(const struct flow_action_entry *act, + struct mlx5e_tc_act_branch_ctrl *cond_true, + struct mlx5e_tc_act_branch_ctrl *cond_false); + + bool is_terminating_action; }; struct mlx5e_tc_flow_action { @@ -81,6 +92,7 @@ extern struct mlx5e_tc_act mlx5e_tc_act_vlan_mangle; extern struct mlx5e_tc_act mlx5e_tc_act_mpls_push; extern struct mlx5e_tc_act mlx5e_tc_act_mpls_pop; extern struct mlx5e_tc_act mlx5e_tc_act_mirred; +extern struct mlx5e_tc_act mlx5e_tc_act_redirect; extern struct mlx5e_tc_act mlx5e_tc_act_mirred_nic; extern struct mlx5e_tc_act mlx5e_tc_act_ct; extern struct mlx5e_tc_act mlx5e_tc_act_sample; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c index dd025a95c439..7d16aeabb119 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/drop.c @@ -27,4 +27,5 @@ tc_act_parse_drop(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5e_tc_act mlx5e_tc_act_drop = { .can_offload = tc_act_can_offload_drop, .parse_action = tc_act_parse_drop, + .is_terminating_action = true, }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/goto.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/goto.c index 25174f68613e..0923e6db2d0a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/goto.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/goto.c @@ -121,4 +121,5 @@ struct mlx5e_tc_act mlx5e_tc_act_goto = { .can_offload = tc_act_can_offload_goto, .parse_action = tc_act_parse_goto, .post_parse = tc_act_post_parse_goto, + .is_terminating_action = true, }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c index 4ac7de3f6afa..78c427b38048 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred.c @@ -334,4 +334,11 @@ tc_act_parse_mirred(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5e_tc_act mlx5e_tc_act_mirred = { .can_offload = tc_act_can_offload_mirred, .parse_action = tc_act_parse_mirred, + .is_terminating_action = false, +}; + +struct mlx5e_tc_act mlx5e_tc_act_redirect = { + .can_offload = tc_act_can_offload_mirred, + .parse_action = tc_act_parse_mirred, + .is_terminating_action = true, }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred_nic.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred_nic.c index 90b4c1b34776..7f409692b18f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred_nic.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/mirred_nic.c @@ -48,4 +48,5 @@ tc_act_parse_mirred_nic(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5e_tc_act mlx5e_tc_act_mirred_nic = { .can_offload = tc_act_can_offload_mirred_nic, .parse_action = tc_act_parse_mirred_nic, + .is_terminating_action = true, }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/police.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/police.c index c8e5ca65bb6e..512d43148922 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/police.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/police.c @@ -3,6 +3,45 @@ #include "act.h" #include "en/tc_priv.h" +#include "fs_core.h" + +static bool police_act_validate_control(enum flow_action_id act_id, + struct netlink_ext_ack *extack) +{ + if (act_id != FLOW_ACTION_PIPE && + act_id != FLOW_ACTION_ACCEPT && + act_id != FLOW_ACTION_JUMP && + act_id != FLOW_ACTION_DROP) { + NL_SET_ERR_MSG_MOD(extack, + "Offload not supported when conform-exceed action is not pipe, ok, jump or drop"); + return false; + } + + return true; +} + +static int police_act_validate(const struct flow_action_entry *act, + struct netlink_ext_ack *extack) +{ + if (!police_act_validate_control(act->police.exceed.act_id, extack) || + !police_act_validate_control(act->police.notexceed.act_id, extack)) + return -EOPNOTSUPP; + + if (act->police.peakrate_bytes_ps || + act->police.avrate || act->police.overhead) { + NL_SET_ERR_MSG_MOD(extack, + "Offload not supported when peakrate/avrate/overhead is configured"); + return -EOPNOTSUPP; + } + + if (act->police.rate_pkt_ps) { + NL_SET_ERR_MSG_MOD(extack, + "QoS offload not support packets per second"); + return -EOPNOTSUPP; + } + + return 0; +} static bool tc_act_can_offload_police(struct mlx5e_tc_act_parse_state *parse_state, @@ -10,14 +49,10 @@ tc_act_can_offload_police(struct mlx5e_tc_act_parse_state *parse_state, int act_index, struct mlx5_flow_attr *attr) { - if (act->police.notexceed.act_id != FLOW_ACTION_PIPE && - act->police.notexceed.act_id != FLOW_ACTION_ACCEPT) { - NL_SET_ERR_MSG_MOD(parse_state->extack, - "Offload not supported when conform action is not pipe or ok"); - return false; - } - if (mlx5e_policer_validate(parse_state->flow_action, act, - parse_state->extack)) + int err; + + err = police_act_validate(act, parse_state->extack); + if (err) return false; return !!mlx5e_get_flow_meters(parse_state->flow->priv->mdev); @@ -37,6 +72,8 @@ fill_meter_params_from_act(const struct flow_action_entry *act, params->mode = MLX5_RATE_LIMIT_PPS; params->rate = act->police.rate_pkt_ps; params->burst = act->police.burst_pkt; + } else if (act->police.mtu) { + params->mtu = act->police.mtu; } else { return -EOPNOTSUPP; } @@ -50,14 +87,25 @@ tc_act_parse_police(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5e_priv *priv, struct mlx5_flow_attr *attr) { + enum mlx5_flow_namespace_type ns = mlx5e_get_flow_namespace(parse_state->flow); + struct mlx5e_flow_meter_params *params = &attr->meter_attr.params; int err; - err = fill_meter_params_from_act(act, &attr->meter_attr.params); + err = fill_meter_params_from_act(act, params); if (err) return err; - attr->action |= MLX5_FLOW_CONTEXT_ACTION_EXECUTE_ASO; - attr->exe_aso_type = MLX5_EXE_ASO_FLOW_METER; + if (params->mtu) { + if (!(mlx5_fs_get_capabilities(priv->mdev, ns) & + MLX5_FLOW_STEERING_CAP_MATCH_RANGES)) + return -EOPNOTSUPP; + + attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + attr->flags |= MLX5_ATTR_FLAG_MTU; + } else { + attr->action |= MLX5_FLOW_CONTEXT_ACTION_EXECUTE_ASO; + attr->exe_aso_type = MLX5_EXE_ASO_FLOW_METER; + } return 0; } @@ -79,7 +127,7 @@ tc_act_police_offload(struct mlx5e_priv *priv, struct mlx5e_flow_meter_handle *meter; int err = 0; - err = mlx5e_policer_validate(&fl_act->action, act, fl_act->extack); + err = police_act_validate(act, fl_act->extack); if (err) return err; @@ -147,6 +195,19 @@ tc_act_police_stats(struct mlx5e_priv *priv, return 0; } +static bool +tc_act_police_get_branch_ctrl(const struct flow_action_entry *act, + struct mlx5e_tc_act_branch_ctrl *cond_true, + struct mlx5e_tc_act_branch_ctrl *cond_false) +{ + cond_true->act_id = act->police.notexceed.act_id; + cond_true->extval = act->police.notexceed.extval; + + cond_false->act_id = act->police.exceed.act_id; + cond_false->extval = act->police.exceed.extval; + return true; +} + struct mlx5e_tc_act mlx5e_tc_act_police = { .can_offload = tc_act_can_offload_police, .parse_action = tc_act_parse_police, @@ -154,4 +215,5 @@ struct mlx5e_tc_act mlx5e_tc_act_police = { .offload_action = tc_act_police_offload, .destroy_action = tc_act_police_destroy, .stats_action = tc_act_police_stats, + .get_branch_ctrl = tc_act_police_get_branch_ctrl, }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c index 53b270f652b9..915ce201aeb2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/trap.c @@ -3,6 +3,7 @@ #include "act.h" #include "en/tc_priv.h" +#include "eswitch.h" static bool tc_act_can_offload_trap(struct mlx5e_tc_act_parse_state *parse_state, @@ -10,13 +11,6 @@ tc_act_can_offload_trap(struct mlx5e_tc_act_parse_state *parse_state, int act_index, struct mlx5_flow_attr *attr) { - struct netlink_ext_ack *extack = parse_state->extack; - - if (parse_state->flow_action->num_entries != 1) { - NL_SET_ERR_MSG_MOD(extack, "action trap is supported as a sole action only"); - return false; - } - return true; } @@ -27,7 +21,7 @@ tc_act_parse_trap(struct mlx5e_tc_act_parse_state *parse_state, struct mlx5_flow_attr *attr) { attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - attr->flags |= MLX5_ATTR_FLAG_SLOW_PATH; + attr->dest_ft = mlx5_eswitch_get_slow_fdb(priv->mdev->priv.eswitch); return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.c index be74e1403328..78af8a3175bf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.c @@ -162,7 +162,6 @@ mlx5e_tc_meter_modify(struct mlx5_core_dev *mdev, MLX5_ACCESS_ASO_OPC_MOD_FLOW_METER); aso_ctrl = &aso_wqe->aso_ctrl; - memset(aso_ctrl, 0, sizeof(*aso_ctrl)); aso_ctrl->data_mask_mode = MLX5_ASO_DATA_MASK_MODE_BYTEWISE_64BYTE << 6; aso_ctrl->condition_1_0_operand = MLX5_ASO_ALWAYS_TRUE | MLX5_ASO_ALWAYS_TRUE << 4; @@ -241,7 +240,7 @@ mlx5e_flow_meter_destroy_aso_obj(struct mlx5_core_dev *mdev, u32 obj_id) } static struct mlx5e_flow_meter_handle * -__mlx5e_flow_meter_alloc(struct mlx5e_flow_meters *flow_meters) +__mlx5e_flow_meter_alloc(struct mlx5e_flow_meters *flow_meters, bool alloc_aso) { struct mlx5_core_dev *mdev = flow_meters->mdev; struct mlx5e_flow_meter_aso_obj *meters_obj; @@ -257,16 +256,19 @@ __mlx5e_flow_meter_alloc(struct mlx5e_flow_meters *flow_meters) counter = mlx5_fc_create(mdev, true); if (IS_ERR(counter)) { err = PTR_ERR(counter); - goto err_red_counter; + goto err_drop_counter; } - meter->red_counter = counter; + meter->drop_counter = counter; counter = mlx5_fc_create(mdev, true); if (IS_ERR(counter)) { err = PTR_ERR(counter); - goto err_green_counter; + goto err_act_counter; } - meter->green_counter = counter; + meter->act_counter = counter; + + if (!alloc_aso) + goto no_aso; meters_obj = list_first_entry_or_null(&flow_meters->partial_list, struct mlx5e_flow_meter_aso_obj, @@ -300,11 +302,12 @@ __mlx5e_flow_meter_alloc(struct mlx5e_flow_meters *flow_meters) } bitmap_set(meters_obj->meters_map, pos, 1); - meter->flow_meters = flow_meters; meter->meters_obj = meters_obj; meter->obj_id = meters_obj->base_id + pos / 2; meter->idx = pos % 2; +no_aso: + meter->flow_meters = flow_meters; mlx5_core_dbg(mdev, "flow meter allocated, obj_id=0x%x, index=%d\n", meter->obj_id, meter->idx); @@ -313,10 +316,10 @@ __mlx5e_flow_meter_alloc(struct mlx5e_flow_meters *flow_meters) err_mem: mlx5e_flow_meter_destroy_aso_obj(mdev, id); err_create: - mlx5_fc_destroy(mdev, meter->green_counter); -err_green_counter: - mlx5_fc_destroy(mdev, meter->red_counter); -err_red_counter: + mlx5_fc_destroy(mdev, meter->act_counter); +err_act_counter: + mlx5_fc_destroy(mdev, meter->drop_counter); +err_drop_counter: kfree(meter); return ERR_PTR(err); } @@ -329,8 +332,11 @@ __mlx5e_flow_meter_free(struct mlx5e_flow_meter_handle *meter) struct mlx5e_flow_meter_aso_obj *meters_obj; int n, pos; - mlx5_fc_destroy(mdev, meter->green_counter); - mlx5_fc_destroy(mdev, meter->red_counter); + mlx5_fc_destroy(mdev, meter->act_counter); + mlx5_fc_destroy(mdev, meter->drop_counter); + + if (meter->params.mtu) + goto out_no_aso; meters_obj = meter->meters_obj; pos = (meter->obj_id - meters_obj->base_id) * 2 + meter->idx; @@ -345,6 +351,7 @@ __mlx5e_flow_meter_free(struct mlx5e_flow_meter_handle *meter) list_add(&meters_obj->entry, &flow_meters->partial_list); } +out_no_aso: mlx5_core_dbg(mdev, "flow meter freed, obj_id=0x%x, index=%d\n", meter->obj_id, meter->idx); kfree(meter); @@ -409,12 +416,13 @@ mlx5e_tc_meter_alloc(struct mlx5e_flow_meters *flow_meters, { struct mlx5e_flow_meter_handle *meter; - meter = __mlx5e_flow_meter_alloc(flow_meters); + meter = __mlx5e_flow_meter_alloc(flow_meters, !params->mtu); if (IS_ERR(meter)) return meter; hash_add(flow_meters->hashtbl, &meter->hlist, params->index); meter->params.index = params->index; + meter->params.mtu = params->mtu; meter->refcnt++; return meter; @@ -575,8 +583,8 @@ mlx5e_tc_meter_get_stats(struct mlx5e_flow_meter_handle *meter, u64 bytes1, packets1, lastuse1; u64 bytes2, packets2, lastuse2; - mlx5_fc_query_cached(meter->green_counter, &bytes1, &packets1, &lastuse1); - mlx5_fc_query_cached(meter->red_counter, &bytes2, &packets2, &lastuse2); + mlx5_fc_query_cached(meter->act_counter, &bytes1, &packets1, &lastuse1); + mlx5_fc_query_cached(meter->drop_counter, &bytes2, &packets2, &lastuse2); *bytes = bytes1 + bytes2; *packets = packets1 + packets2; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.h index 6de6e8a16327..9b795cd106bb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/meter.h @@ -20,6 +20,7 @@ struct mlx5e_flow_meter_params { u32 index; u64 rate; u64 burst; + u32 mtu; }; struct mlx5e_flow_meter_handle { @@ -32,8 +33,8 @@ struct mlx5e_flow_meter_handle { struct hlist_node hlist; struct mlx5e_flow_meter_params params; - struct mlx5_fc *green_counter; - struct mlx5_fc *red_counter; + struct mlx5_fc *act_counter; + struct mlx5_fc *drop_counter; }; struct mlx5e_meter_attr { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.c index 8b77e822810e..8d7d761482d2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.c @@ -8,23 +8,56 @@ #define MLX5_PACKET_COLOR_BITS MLX5_REG_MAPPING_MBITS(PACKET_COLOR_TO_REG) #define MLX5_PACKET_COLOR_MASK MLX5_REG_MAPPING_MASK(PACKET_COLOR_TO_REG) -struct mlx5e_post_meter_priv { +struct mlx5e_post_meter_rate_table { + struct mlx5_flow_table *ft; + struct mlx5_flow_group *fg; + struct mlx5_flow_handle *green_rule; + struct mlx5_flow_attr *green_attr; + struct mlx5_flow_handle *red_rule; + struct mlx5_flow_attr *red_attr; +}; + +struct mlx5e_post_meter_mtu_table { struct mlx5_flow_table *ft; struct mlx5_flow_group *fg; - struct mlx5_flow_handle *fwd_green_rule; - struct mlx5_flow_handle *drop_red_rule; + struct mlx5_flow_handle *rule; + struct mlx5_flow_attr *attr; +}; + +struct mlx5e_post_meter_mtu_tables { + struct mlx5e_post_meter_mtu_table green_table; + struct mlx5e_post_meter_mtu_table red_table; +}; + +struct mlx5e_post_meter_priv { + enum mlx5e_post_meter_type type; + union { + struct mlx5e_post_meter_rate_table rate_steering_table; + struct mlx5e_post_meter_mtu_tables mtu_tables; + }; }; struct mlx5_flow_table * mlx5e_post_meter_get_ft(struct mlx5e_post_meter_priv *post_meter) { - return post_meter->ft; + return post_meter->rate_steering_table.ft; } -static int +struct mlx5_flow_table * +mlx5e_post_meter_get_mtu_true_ft(struct mlx5e_post_meter_priv *post_meter) +{ + return post_meter->mtu_tables.green_table.ft; +} + +struct mlx5_flow_table * +mlx5e_post_meter_get_mtu_false_ft(struct mlx5e_post_meter_priv *post_meter) +{ + return post_meter->mtu_tables.red_table.ft; +} + +static struct mlx5_flow_table * mlx5e_post_meter_table_create(struct mlx5e_priv *priv, - enum mlx5_flow_namespace_type ns_type, - struct mlx5e_post_meter_priv *post_meter) + enum mlx5_flow_namespace_type ns_type) { struct mlx5_flow_table_attr ft_attr = {}; struct mlx5_flow_namespace *root_ns; @@ -32,7 +65,7 @@ mlx5e_post_meter_table_create(struct mlx5e_priv *priv, root_ns = mlx5_get_flow_namespace(priv->mdev, ns_type); if (!root_ns) { mlx5_core_warn(priv->mdev, "Failed to get namespace for flow meter\n"); - return -EOPNOTSUPP; + return ERR_PTR(-EOPNOTSUPP); } ft_attr.flags = MLX5_FLOW_TABLE_UNMANAGED; @@ -40,19 +73,14 @@ mlx5e_post_meter_table_create(struct mlx5e_priv *priv, ft_attr.max_fte = 2; ft_attr.level = 1; - post_meter->ft = mlx5_create_flow_table(root_ns, &ft_attr); - if (IS_ERR(post_meter->ft)) { - mlx5_core_warn(priv->mdev, "Failed to create post_meter table\n"); - return PTR_ERR(post_meter->ft); - } - - return 0; + return mlx5_create_flow_table(root_ns, &ft_attr); } static int -mlx5e_post_meter_fg_create(struct mlx5e_priv *priv, - struct mlx5e_post_meter_priv *post_meter) +mlx5e_post_meter_rate_fg_create(struct mlx5e_priv *priv, + struct mlx5e_post_meter_priv *post_meter) { + struct mlx5e_post_meter_rate_table *table = &post_meter->rate_steering_table; int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); void *misc2, *match_criteria; u32 *flow_group_in; @@ -71,25 +99,58 @@ mlx5e_post_meter_fg_create(struct mlx5e_priv *priv, MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); - post_meter->fg = mlx5_create_flow_group(post_meter->ft, flow_group_in); - if (IS_ERR(post_meter->fg)) { + table->fg = mlx5_create_flow_group(table->ft, flow_group_in); + if (IS_ERR(table->fg)) { mlx5_core_warn(priv->mdev, "Failed to create post_meter flow group\n"); - err = PTR_ERR(post_meter->fg); + err = PTR_ERR(table->fg); } kvfree(flow_group_in); return err; } +static struct mlx5_flow_handle * +mlx5e_post_meter_add_rule(struct mlx5e_priv *priv, + struct mlx5e_post_meter_priv *post_meter, + struct mlx5_flow_spec *spec, + struct mlx5_flow_attr *attr, + struct mlx5_fc *act_counter, + struct mlx5_fc *drop_counter) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_flow_handle *ret; + + attr->action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_DROP) + attr->counter = drop_counter; + else + attr->counter = act_counter; + + attr->flags |= MLX5_ATTR_FLAG_NO_IN_PORT; + attr->outer_match_level = MLX5_MATCH_NONE; + attr->chain = 0; + attr->prio = 0; + + ret = mlx5_eswitch_add_offloaded_rule(esw, spec, attr); + + /* We did not create the counter, so we can't delete it. + * Avoid freeing the counter when the attr is deleted in free_branching_attr + */ + attr->action &= ~MLX5_FLOW_CONTEXT_ACTION_COUNT; + + return ret; +} + static int -mlx5e_post_meter_rules_create(struct mlx5e_priv *priv, - struct mlx5e_post_meter_priv *post_meter, - struct mlx5e_post_act *post_act, - struct mlx5_fc *green_counter, - struct mlx5_fc *red_counter) -{ - struct mlx5_flow_destination dest[2] = {}; - struct mlx5_flow_act flow_act = {}; +mlx5e_post_meter_rate_rules_create(struct mlx5e_priv *priv, + struct mlx5e_post_meter_priv *post_meter, + struct mlx5e_post_act *post_act, + struct mlx5_fc *act_counter, + struct mlx5_fc *drop_counter, + struct mlx5_flow_attr *green_attr, + struct mlx5_flow_attr *red_attr) +{ + struct mlx5e_post_meter_rate_table *table = &post_meter->rate_steering_table; struct mlx5_flow_handle *rule; struct mlx5_flow_spec *spec; int err; @@ -100,72 +161,242 @@ mlx5e_post_meter_rules_create(struct mlx5e_priv *priv, mlx5e_tc_match_to_reg_match(spec, PACKET_COLOR_TO_REG, MLX5_FLOW_METER_COLOR_RED, MLX5_PACKET_COLOR_MASK); - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP | - MLX5_FLOW_CONTEXT_ACTION_COUNT; - flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; - dest[0].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; - dest[0].counter_id = mlx5_fc_id(red_counter); - - rule = mlx5_add_flow_rules(post_meter->ft, spec, &flow_act, dest, 1); + red_attr->ft = post_meter->rate_steering_table.ft; + rule = mlx5e_post_meter_add_rule(priv, post_meter, spec, red_attr, + act_counter, drop_counter); if (IS_ERR(rule)) { - mlx5_core_warn(priv->mdev, "Failed to create post_meter flow drop rule\n"); + mlx5_core_warn(priv->mdev, "Failed to create post_meter exceed rule\n"); err = PTR_ERR(rule); goto err_red; } - post_meter->drop_red_rule = rule; + table->red_rule = rule; + table->red_attr = red_attr; mlx5e_tc_match_to_reg_match(spec, PACKET_COLOR_TO_REG, MLX5_FLOW_METER_COLOR_GREEN, MLX5_PACKET_COLOR_MASK); - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | - MLX5_FLOW_CONTEXT_ACTION_COUNT; - dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest[0].ft = mlx5e_tc_post_act_get_ft(post_act); - dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; - dest[1].counter_id = mlx5_fc_id(green_counter); - - rule = mlx5_add_flow_rules(post_meter->ft, spec, &flow_act, dest, 2); + green_attr->ft = post_meter->rate_steering_table.ft; + rule = mlx5e_post_meter_add_rule(priv, post_meter, spec, green_attr, + act_counter, drop_counter); if (IS_ERR(rule)) { - mlx5_core_warn(priv->mdev, "Failed to create post_meter flow fwd rule\n"); + mlx5_core_warn(priv->mdev, "Failed to create post_meter notexceed rule\n"); err = PTR_ERR(rule); goto err_green; } - post_meter->fwd_green_rule = rule; + table->green_rule = rule; + table->green_attr = green_attr; kvfree(spec); return 0; err_green: - mlx5_del_flow_rules(post_meter->drop_red_rule); + mlx5_del_flow_rules(table->red_rule); err_red: kvfree(spec); return err; } static void -mlx5e_post_meter_rules_destroy(struct mlx5e_post_meter_priv *post_meter) +mlx5e_post_meter_rate_rules_destroy(struct mlx5_eswitch *esw, + struct mlx5e_post_meter_priv *post_meter) { - mlx5_del_flow_rules(post_meter->drop_red_rule); - mlx5_del_flow_rules(post_meter->fwd_green_rule); + struct mlx5e_post_meter_rate_table *rate_table = &post_meter->rate_steering_table; + + mlx5_eswitch_del_offloaded_rule(esw, rate_table->red_rule, rate_table->red_attr); + mlx5_eswitch_del_offloaded_rule(esw, rate_table->green_rule, rate_table->green_attr); } static void -mlx5e_post_meter_fg_destroy(struct mlx5e_post_meter_priv *post_meter) +mlx5e_post_meter_rate_fg_destroy(struct mlx5e_post_meter_priv *post_meter) { - mlx5_destroy_flow_group(post_meter->fg); + mlx5_destroy_flow_group(post_meter->rate_steering_table.fg); } static void -mlx5e_post_meter_table_destroy(struct mlx5e_post_meter_priv *post_meter) +mlx5e_post_meter_rate_table_destroy(struct mlx5e_post_meter_priv *post_meter) +{ + mlx5_destroy_flow_table(post_meter->rate_steering_table.ft); +} + +static void +mlx5e_post_meter_mtu_rules_destroy(struct mlx5e_post_meter_priv *post_meter) +{ + struct mlx5e_post_meter_mtu_tables *mtu_tables = &post_meter->mtu_tables; + + mlx5_del_flow_rules(mtu_tables->green_table.rule); + mlx5_del_flow_rules(mtu_tables->red_table.rule); +} + +static void +mlx5e_post_meter_mtu_fg_destroy(struct mlx5e_post_meter_priv *post_meter) +{ + struct mlx5e_post_meter_mtu_tables *mtu_tables = &post_meter->mtu_tables; + + mlx5_destroy_flow_group(mtu_tables->green_table.fg); + mlx5_destroy_flow_group(mtu_tables->red_table.fg); +} + +static void +mlx5e_post_meter_mtu_table_destroy(struct mlx5e_post_meter_priv *post_meter) +{ + struct mlx5e_post_meter_mtu_tables *mtu_tables = &post_meter->mtu_tables; + + mlx5_destroy_flow_table(mtu_tables->green_table.ft); + mlx5_destroy_flow_table(mtu_tables->red_table.ft); +} + +static int +mlx5e_post_meter_rate_create(struct mlx5e_priv *priv, + enum mlx5_flow_namespace_type ns_type, + struct mlx5e_post_act *post_act, + struct mlx5_fc *act_counter, + struct mlx5_fc *drop_counter, + struct mlx5e_post_meter_priv *post_meter, + struct mlx5_flow_attr *green_attr, + struct mlx5_flow_attr *red_attr) +{ + struct mlx5_flow_table *ft; + int err; + + post_meter->type = MLX5E_POST_METER_RATE; + + ft = mlx5e_post_meter_table_create(priv, ns_type); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + mlx5_core_warn(priv->mdev, "Failed to create post_meter table\n"); + goto err_ft; + } + + post_meter->rate_steering_table.ft = ft; + + err = mlx5e_post_meter_rate_fg_create(priv, post_meter); + if (err) + goto err_fg; + + err = mlx5e_post_meter_rate_rules_create(priv, post_meter, post_act, + act_counter, drop_counter, + green_attr, red_attr); + if (err) + goto err_rules; + + return 0; + +err_rules: + mlx5e_post_meter_rate_fg_destroy(post_meter); +err_fg: + mlx5e_post_meter_rate_table_destroy(post_meter); +err_ft: + return err; +} + +static int +mlx5e_post_meter_create_mtu_table(struct mlx5e_priv *priv, + enum mlx5_flow_namespace_type ns_type, + struct mlx5e_post_meter_mtu_table *table) { - mlx5_destroy_flow_table(post_meter->ft); + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *fg; + u32 *flow_group_in; + int err; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + table->ft = mlx5e_post_meter_table_create(priv, ns_type); + if (IS_ERR(table->ft)) { + err = PTR_ERR(table->ft); + goto err_ft; + } + + /* create miss group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); + fg = mlx5_create_flow_group(table->ft, flow_group_in); + if (IS_ERR(fg)) { + err = PTR_ERR(fg); + goto err_miss_grp; + } + table->fg = fg; + + kvfree(flow_group_in); + return 0; + +err_miss_grp: + mlx5_destroy_flow_table(table->ft); +err_ft: + kvfree(flow_group_in); + return err; +} + +static int +mlx5e_post_meter_mtu_create(struct mlx5e_priv *priv, + enum mlx5_flow_namespace_type ns_type, + struct mlx5e_post_act *post_act, + struct mlx5_fc *act_counter, + struct mlx5_fc *drop_counter, + struct mlx5e_post_meter_priv *post_meter, + struct mlx5_flow_attr *green_attr, + struct mlx5_flow_attr *red_attr) +{ + struct mlx5e_post_meter_mtu_tables *mtu_tables = &post_meter->mtu_tables; + static struct mlx5_flow_spec zero_spec = {}; + struct mlx5_flow_handle *rule; + int err; + + post_meter->type = MLX5E_POST_METER_MTU; + + err = mlx5e_post_meter_create_mtu_table(priv, ns_type, &mtu_tables->green_table); + if (err) + goto err_green_ft; + + green_attr->ft = mtu_tables->green_table.ft; + rule = mlx5e_post_meter_add_rule(priv, post_meter, &zero_spec, green_attr, + act_counter, drop_counter); + if (IS_ERR(rule)) { + mlx5_core_warn(priv->mdev, "Failed to create post_meter conform rule\n"); + err = PTR_ERR(rule); + goto err_green_rule; + } + mtu_tables->green_table.rule = rule; + mtu_tables->green_table.attr = green_attr; + + err = mlx5e_post_meter_create_mtu_table(priv, ns_type, &mtu_tables->red_table); + if (err) + goto err_red_ft; + + red_attr->ft = mtu_tables->red_table.ft; + rule = mlx5e_post_meter_add_rule(priv, post_meter, &zero_spec, red_attr, + act_counter, drop_counter); + if (IS_ERR(rule)) { + mlx5_core_warn(priv->mdev, "Failed to create post_meter exceed rule\n"); + err = PTR_ERR(rule); + goto err_red_rule; + } + mtu_tables->red_table.rule = rule; + mtu_tables->red_table.attr = red_attr; + + return 0; + +err_red_rule: + mlx5_destroy_flow_table(mtu_tables->red_table.ft); +err_red_ft: + mlx5_del_flow_rules(mtu_tables->green_table.rule); +err_green_rule: + mlx5_destroy_flow_table(mtu_tables->green_table.ft); +err_green_ft: + return err; } struct mlx5e_post_meter_priv * mlx5e_post_meter_init(struct mlx5e_priv *priv, enum mlx5_flow_namespace_type ns_type, struct mlx5e_post_act *post_act, - struct mlx5_fc *green_counter, - struct mlx5_fc *red_counter) + enum mlx5e_post_meter_type type, + struct mlx5_fc *act_counter, + struct mlx5_fc *drop_counter, + struct mlx5_flow_attr *branch_true, + struct mlx5_flow_attr *branch_false) { struct mlx5e_post_meter_priv *post_meter; int err; @@ -174,36 +405,55 @@ mlx5e_post_meter_init(struct mlx5e_priv *priv, if (!post_meter) return ERR_PTR(-ENOMEM); - err = mlx5e_post_meter_table_create(priv, ns_type, post_meter); - if (err) - goto err_ft; - - err = mlx5e_post_meter_fg_create(priv, post_meter); - if (err) - goto err_fg; + switch (type) { + case MLX5E_POST_METER_MTU: + err = mlx5e_post_meter_mtu_create(priv, ns_type, post_act, + act_counter, drop_counter, post_meter, + branch_true, branch_false); + break; + case MLX5E_POST_METER_RATE: + err = mlx5e_post_meter_rate_create(priv, ns_type, post_act, + act_counter, drop_counter, post_meter, + branch_true, branch_false); + break; + default: + err = -EOPNOTSUPP; + } - err = mlx5e_post_meter_rules_create(priv, post_meter, post_act, green_counter, - red_counter); if (err) - goto err_rules; + goto err; return post_meter; -err_rules: - mlx5e_post_meter_fg_destroy(post_meter); -err_fg: - mlx5e_post_meter_table_destroy(post_meter); -err_ft: +err: kfree(post_meter); return ERR_PTR(err); } +static void +mlx5e_post_meter_rate_destroy(struct mlx5_eswitch *esw, struct mlx5e_post_meter_priv *post_meter) +{ + mlx5e_post_meter_rate_rules_destroy(esw, post_meter); + mlx5e_post_meter_rate_fg_destroy(post_meter); + mlx5e_post_meter_rate_table_destroy(post_meter); +} + +static void +mlx5e_post_meter_mtu_destroy(struct mlx5e_post_meter_priv *post_meter) +{ + mlx5e_post_meter_mtu_rules_destroy(post_meter); + mlx5e_post_meter_mtu_fg_destroy(post_meter); + mlx5e_post_meter_mtu_table_destroy(post_meter); +} + void -mlx5e_post_meter_cleanup(struct mlx5e_post_meter_priv *post_meter) +mlx5e_post_meter_cleanup(struct mlx5_eswitch *esw, struct mlx5e_post_meter_priv *post_meter) { - mlx5e_post_meter_rules_destroy(post_meter); - mlx5e_post_meter_fg_destroy(post_meter); - mlx5e_post_meter_table_destroy(post_meter); + if (post_meter->type == MLX5E_POST_METER_RATE) + mlx5e_post_meter_rate_destroy(esw, post_meter); + else + mlx5e_post_meter_mtu_destroy(post_meter); + kfree(post_meter); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.h index 34d0e4b9fc7a..e013b77186b2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/post_meter.h @@ -14,16 +14,49 @@ struct mlx5e_post_meter_priv; +enum mlx5e_post_meter_type { + MLX5E_POST_METER_RATE = 0, + MLX5E_POST_METER_MTU +}; + +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + struct mlx5_flow_table * mlx5e_post_meter_get_ft(struct mlx5e_post_meter_priv *post_meter); +struct mlx5_flow_table * +mlx5e_post_meter_get_mtu_true_ft(struct mlx5e_post_meter_priv *post_meter); + +struct mlx5_flow_table * +mlx5e_post_meter_get_mtu_false_ft(struct mlx5e_post_meter_priv *post_meter); + struct mlx5e_post_meter_priv * mlx5e_post_meter_init(struct mlx5e_priv *priv, enum mlx5_flow_namespace_type ns_type, struct mlx5e_post_act *post_act, - struct mlx5_fc *green_counter, - struct mlx5_fc *red_counter); + enum mlx5e_post_meter_type type, + struct mlx5_fc *act_counter, + struct mlx5_fc *drop_counter, + struct mlx5_flow_attr *branch_true, + struct mlx5_flow_attr *branch_false); + void -mlx5e_post_meter_cleanup(struct mlx5e_post_meter_priv *post_meter); +mlx5e_post_meter_cleanup(struct mlx5_eswitch *esw, struct mlx5e_post_meter_priv *post_meter); + +#else /* CONFIG_MLX5_CLS_ACT */ + +static inline struct mlx5_flow_table * +mlx5e_post_meter_get_mtu_true_ft(struct mlx5e_post_meter_priv *post_meter) +{ + return NULL; +} + +static inline struct mlx5_flow_table * +mlx5e_post_meter_get_mtu_false_ft(struct mlx5e_post_meter_priv *post_meter) +{ + return NULL; +} + +#endif #endif /* __MLX5_EN_POST_METER_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c index 864ce0c393e6..a69849e0deed 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c @@ -1774,35 +1774,42 @@ mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft) /* We translate the tc filter with CT action to the following HW model: * - * +---------------------+ - * + ft prio (tc chain) + - * + original match + - * +---------------------+ - * | set chain miss mapping - * | set fte_id - * | set tunnel_id - * | do decap - * v - * +---------------------+ - * + pre_ct/pre_ct_nat + if matches +-------------------------+ - * + zone+nat match +---------------->+ post_act (see below) + - * +---------------------+ set zone +-------------------------+ - * | set zone - * v - * +--------------------+ - * + CT (nat or no nat) + - * + tuple + zone match + - * +--------------------+ - * | set mark - * | set labels_id - * | set established - * | set zone_restore - * | do nat (if needed) - * v - * +--------------+ - * + post_act + original filter actions - * + fte_id match +------------------------> - * +--------------+ + * +---------------------+ + * + ft prio (tc chain) + + * + original match + + * +---------------------+ + * | set chain miss mapping + * | set fte_id + * | set tunnel_id + * | do decap + * | + * +-------------+ + * | Chain 0 | + * | optimization| + * | v + * | +---------------------+ + * | + pre_ct/pre_ct_nat + if matches +----------------------+ + * | + zone+nat match +---------------->+ post_act (see below) + + * | +---------------------+ set zone +----------------------+ + * | | + * +-------------+ set zone + * | + * v + * +--------------------+ + * + CT (nat or no nat) + + * + tuple + zone match + + * +--------------------+ + * | set mark + * | set labels_id + * | set established + * | set zone_restore + * | do nat (if needed) + * v + * +--------------+ + * + post_act + original filter actions + * + fte_id match +------------------------> + * +--------------+ + * */ static struct mlx5_flow_handle * __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv, @@ -1818,6 +1825,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_flow *ct_flow; int chain_mapping = 0, err; struct mlx5_ct_ft *ft; + u16 zone; ct_flow = kzalloc(sizeof(*ct_flow), GFP_KERNEL); if (!ct_flow) { @@ -1884,6 +1892,25 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv, } } + /* Change original rule point to ct table + * Chain 0 sets the zone and jumps to ct table + * Other chains jump to pre_ct table to align with act_ct cached logic + */ + pre_ct_attr->dest_chain = 0; + if (!attr->chain) { + zone = ft->zone & MLX5_CT_ZONE_MASK; + err = mlx5e_tc_match_to_reg_set(priv->mdev, pre_mod_acts, ct_priv->ns_type, + ZONE_TO_REG, zone); + if (err) { + ct_dbg("Failed to set zone register mapping"); + goto err_mapping; + } + + pre_ct_attr->dest_ft = nat ? ct_priv->ct_nat : ct_priv->ct; + } else { + pre_ct_attr->dest_ft = nat ? ft->pre_ct_nat.ft : ft->pre_ct.ft; + } + mod_hdr = mlx5_modify_header_alloc(priv->mdev, ct_priv->ns_type, pre_mod_acts->num_actions, pre_mod_acts->actions); @@ -1893,10 +1920,6 @@ __mlx5_tc_ct_flow_offload(struct mlx5_tc_ct_priv *ct_priv, goto err_mapping; } pre_ct_attr->modify_hdr = mod_hdr; - - /* Change original rule point to ct table */ - pre_ct_attr->dest_chain = 0; - pre_ct_attr->dest_ft = nat ? ft->pre_ct_nat.ft : ft->pre_ct.ft; ct_flow->pre_ct_rule = mlx5_tc_rule_insert(priv, orig_spec, pre_ct_attr); if (IS_ERR(ct_flow->pre_ct_rule)) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h index 2e42d7c5451e..2b7fd1c0e643 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_priv.h @@ -211,8 +211,4 @@ struct mlx5e_flow_meters *mlx5e_get_flow_meters(struct mlx5_core_dev *dev); void *mlx5e_get_match_headers_value(u32 flags, struct mlx5_flow_spec *spec); void *mlx5e_get_match_headers_criteria(u32 flags, struct mlx5_flow_spec *spec); -int mlx5e_policer_validate(const struct flow_action *action, - const struct flow_action_entry *act, - struct netlink_ext_ack *extack); - #endif /* __MLX5_EN_TC_PRIV_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c index a715601865d3..bb9023957f74 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c @@ -45,55 +45,9 @@ static struct mlx5e_ipsec_sa_entry *to_ipsec_sa_entry(struct xfrm_state *x) return (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; } -struct xfrm_state *mlx5e_ipsec_sadb_rx_lookup(struct mlx5e_ipsec *ipsec, - unsigned int handle) +static struct mlx5e_ipsec_pol_entry *to_ipsec_pol_entry(struct xfrm_policy *x) { - struct mlx5e_ipsec_sa_entry *sa_entry; - struct xfrm_state *ret = NULL; - - rcu_read_lock(); - hash_for_each_possible_rcu(ipsec->sadb_rx, sa_entry, hlist, handle) - if (sa_entry->handle == handle) { - ret = sa_entry->x; - xfrm_state_hold(ret); - break; - } - rcu_read_unlock(); - - return ret; -} - -static int mlx5e_ipsec_sadb_rx_add(struct mlx5e_ipsec_sa_entry *sa_entry) -{ - unsigned int handle = sa_entry->ipsec_obj_id; - struct mlx5e_ipsec *ipsec = sa_entry->ipsec; - struct mlx5e_ipsec_sa_entry *_sa_entry; - unsigned long flags; - - rcu_read_lock(); - hash_for_each_possible_rcu(ipsec->sadb_rx, _sa_entry, hlist, handle) - if (_sa_entry->handle == handle) { - rcu_read_unlock(); - return -EEXIST; - } - rcu_read_unlock(); - - spin_lock_irqsave(&ipsec->sadb_rx_lock, flags); - sa_entry->handle = handle; - hash_add_rcu(ipsec->sadb_rx, &sa_entry->hlist, sa_entry->handle); - spin_unlock_irqrestore(&ipsec->sadb_rx_lock, flags); - - return 0; -} - -static void mlx5e_ipsec_sadb_rx_del(struct mlx5e_ipsec_sa_entry *sa_entry) -{ - struct mlx5e_ipsec *ipsec = sa_entry->ipsec; - unsigned long flags; - - spin_lock_irqsave(&ipsec->sadb_rx_lock, flags); - hash_del_rcu(&sa_entry->hlist); - spin_unlock_irqrestore(&ipsec->sadb_rx_lock, flags); + return (struct mlx5e_ipsec_pol_entry *)x->xdo.offload_handle; } static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry) @@ -129,9 +83,33 @@ static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry) return false; } -static void -mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, - struct mlx5_accel_esp_xfrm_attrs *attrs) +static void mlx5e_ipsec_init_limits(struct mlx5e_ipsec_sa_entry *sa_entry, + struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + struct xfrm_state *x = sa_entry->x; + + attrs->hard_packet_limit = x->lft.hard_packet_limit; + if (x->lft.soft_packet_limit == XFRM_INF) + return; + + /* Hardware decrements hard_packet_limit counter through + * the operation. While fires an event when soft_packet_limit + * is reached. It emans that we need substitute the numbers + * in order to properly count soft limit. + * + * As an example: + * XFRM user sets soft limit is 2 and hard limit is 9 and + * expects to see soft event after 2 packets and hard event + * after 9 packets. In our case, the hard limit will be set + * to 9 and soft limit is comparator to 7 so user gets the + * soft event after 2 packeta + */ + attrs->soft_packet_limit = + x->lft.hard_packet_limit - x->lft.soft_packet_limit; +} + +void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, + struct mlx5_accel_esp_xfrm_attrs *attrs) { struct xfrm_state *x = sa_entry->x; struct aes_gcm_keymat *aes_gcm = &attrs->aes_gcm; @@ -157,33 +135,31 @@ mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, memcpy(&aes_gcm->salt, x->aead->alg_key + key_len, sizeof(aes_gcm->salt)); + attrs->authsize = crypto_aead_authsize(aead) / 4; /* in dwords */ + /* iv len */ aes_gcm->icv_len = x->aead->alg_icv_len; /* esn */ if (sa_entry->esn_state.trigger) { - attrs->flags |= MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED; + attrs->esn_trigger = true; attrs->esn = sa_entry->esn_state.esn; - if (sa_entry->esn_state.overlap) - attrs->flags |= MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP; + attrs->esn_overlap = sa_entry->esn_state.overlap; + attrs->replay_window = x->replay_esn->replay_window; } - /* action */ - attrs->action = (x->xso.dir == XFRM_DEV_OFFLOAD_OUT) ? - MLX5_ACCEL_ESP_ACTION_ENCRYPT : - MLX5_ACCEL_ESP_ACTION_DECRYPT; - /* flags */ - attrs->flags |= (x->props.mode == XFRM_MODE_TRANSPORT) ? - MLX5_ACCEL_ESP_FLAGS_TRANSPORT : - MLX5_ACCEL_ESP_FLAGS_TUNNEL; - + attrs->dir = x->xso.dir; /* spi */ attrs->spi = be32_to_cpu(x->id.spi); /* source , destination ips */ memcpy(&attrs->saddr, x->props.saddr.a6, sizeof(attrs->saddr)); memcpy(&attrs->daddr, x->id.daddr.a6, sizeof(attrs->daddr)); - attrs->is_ipv6 = (x->props.family != AF_INET); + attrs->family = x->props.family; + attrs->type = x->xso.type; + attrs->reqid = x->props.reqid; + + mlx5e_ipsec_init_limits(sa_entry, attrs); } static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) @@ -215,11 +191,6 @@ static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) netdev_info(netdev, "Only IPv4/6 xfrm states may be offloaded\n"); return -EINVAL; } - if (x->props.mode != XFRM_MODE_TRANSPORT && - x->props.mode != XFRM_MODE_TUNNEL) { - dev_info(&netdev->dev, "Only transport and tunnel xfrm states may be offloaded\n"); - return -EINVAL; - } if (x->id.proto != IPPROTO_ESP) { netdev_info(netdev, "Only ESP xfrm state may be offloaded\n"); return -EINVAL; @@ -253,6 +224,67 @@ static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) netdev_info(netdev, "Cannot offload xfrm states with geniv other than seqiv\n"); return -EINVAL; } + switch (x->xso.type) { + case XFRM_DEV_OFFLOAD_CRYPTO: + if (!(mlx5_ipsec_device_caps(priv->mdev) & + MLX5_IPSEC_CAP_CRYPTO)) { + netdev_info(netdev, "Crypto offload is not supported\n"); + return -EINVAL; + } + + if (x->props.mode != XFRM_MODE_TRANSPORT && + x->props.mode != XFRM_MODE_TUNNEL) { + netdev_info(netdev, "Only transport and tunnel xfrm states may be offloaded\n"); + return -EINVAL; + } + break; + case XFRM_DEV_OFFLOAD_PACKET: + if (!(mlx5_ipsec_device_caps(priv->mdev) & + MLX5_IPSEC_CAP_PACKET_OFFLOAD)) { + netdev_info(netdev, "Packet offload is not supported\n"); + return -EINVAL; + } + + if (x->props.mode != XFRM_MODE_TRANSPORT) { + netdev_info(netdev, "Only transport xfrm states may be offloaded in packet mode\n"); + return -EINVAL; + } + + if (x->replay_esn && x->replay_esn->replay_window != 32 && + x->replay_esn->replay_window != 64 && + x->replay_esn->replay_window != 128 && + x->replay_esn->replay_window != 256) { + netdev_info(netdev, + "Unsupported replay window size %u\n", + x->replay_esn->replay_window); + return -EINVAL; + } + + if (!x->props.reqid) { + netdev_info(netdev, "Cannot offload without reqid\n"); + return -EINVAL; + } + + if (x->lft.hard_byte_limit != XFRM_INF || + x->lft.soft_byte_limit != XFRM_INF) { + netdev_info(netdev, + "Device doesn't support limits in bytes\n"); + return -EINVAL; + } + + if (x->lft.soft_packet_limit >= x->lft.hard_packet_limit && + x->lft.hard_packet_limit != XFRM_INF) { + /* XFRM stack doesn't prevent such configuration :(. */ + netdev_info(netdev, + "Hard packet limit must be greater than soft one\n"); + return -EINVAL; + } + break; + default: + netdev_info(netdev, "Unsupported xfrm offload type %d\n", + x->xso.type); + return -EINVAL; + } return 0; } @@ -270,6 +302,7 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) { struct mlx5e_ipsec_sa_entry *sa_entry = NULL; struct net_device *netdev = x->xso.real_dev; + struct mlx5e_ipsec *ipsec; struct mlx5e_priv *priv; int err; @@ -277,6 +310,7 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) if (!priv->ipsec) return -EOPNOTSUPP; + ipsec = priv->ipsec; err = mlx5e_xfrm_validate_state(x); if (err) return err; @@ -288,7 +322,7 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) } sa_entry->x = x; - sa_entry->ipsec = priv->ipsec; + sa_entry->ipsec = ipsec; /* check esn */ mlx5e_ipsec_update_esn_state(sa_entry); @@ -299,25 +333,29 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) if (err) goto err_xfrm; - err = mlx5e_accel_ipsec_fs_add_rule(priv, sa_entry); + err = mlx5e_accel_ipsec_fs_add_rule(sa_entry); if (err) goto err_hw_ctx; - if (x->xso.dir == XFRM_DEV_OFFLOAD_IN) { - err = mlx5e_ipsec_sadb_rx_add(sa_entry); - if (err) - goto err_add_rule; - } else { + /* We use *_bh() variant because xfrm_timer_handler(), which runs + * in softirq context, can reach our state delete logic and we need + * xa_erase_bh() there. + */ + err = xa_insert_bh(&ipsec->sadb, sa_entry->ipsec_obj_id, sa_entry, + GFP_KERNEL); + if (err) + goto err_add_rule; + + if (x->xso.dir == XFRM_DEV_OFFLOAD_OUT) sa_entry->set_iv_op = (x->props.flags & XFRM_STATE_ESN) ? mlx5e_ipsec_set_iv_esn : mlx5e_ipsec_set_iv; - } INIT_WORK(&sa_entry->modify_work.work, _update_xfrm_state); x->xso.offload_handle = (unsigned long)sa_entry; - goto out; + return 0; err_add_rule: - mlx5e_accel_ipsec_fs_del_rule(priv, sa_entry); + mlx5e_accel_ipsec_fs_del_rule(sa_entry); err_hw_ctx: mlx5_ipsec_free_sa_ctx(sa_entry); err_xfrm: @@ -329,59 +367,69 @@ out: static void mlx5e_xfrm_del_state(struct xfrm_state *x) { struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); + struct mlx5e_ipsec *ipsec = sa_entry->ipsec; + struct mlx5e_ipsec_sa_entry *old; - if (x->xso.dir == XFRM_DEV_OFFLOAD_IN) - mlx5e_ipsec_sadb_rx_del(sa_entry); + old = xa_erase_bh(&ipsec->sadb, sa_entry->ipsec_obj_id); + WARN_ON(old != sa_entry); } static void mlx5e_xfrm_free_state(struct xfrm_state *x) { struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); - struct mlx5e_priv *priv = netdev_priv(x->xso.dev); cancel_work_sync(&sa_entry->modify_work.work); - mlx5e_accel_ipsec_fs_del_rule(priv, sa_entry); + mlx5e_accel_ipsec_fs_del_rule(sa_entry); mlx5_ipsec_free_sa_ctx(sa_entry); kfree(sa_entry); } -int mlx5e_ipsec_init(struct mlx5e_priv *priv) +void mlx5e_ipsec_init(struct mlx5e_priv *priv) { struct mlx5e_ipsec *ipsec; - int ret; + int ret = -ENOMEM; if (!mlx5_ipsec_device_caps(priv->mdev)) { netdev_dbg(priv->netdev, "Not an IPSec offload device\n"); - return 0; + return; } ipsec = kzalloc(sizeof(*ipsec), GFP_KERNEL); if (!ipsec) - return -ENOMEM; + return; - hash_init(ipsec->sadb_rx); - spin_lock_init(&ipsec->sadb_rx_lock); + xa_init_flags(&ipsec->sadb, XA_FLAGS_ALLOC); ipsec->mdev = priv->mdev; ipsec->wq = alloc_ordered_workqueue("mlx5e_ipsec: %s", 0, priv->netdev->name); - if (!ipsec->wq) { - ret = -ENOMEM; + if (!ipsec->wq) goto err_wq; + + if (mlx5_ipsec_device_caps(priv->mdev) & + MLX5_IPSEC_CAP_PACKET_OFFLOAD) { + ret = mlx5e_ipsec_aso_init(ipsec); + if (ret) + goto err_aso; } ret = mlx5e_accel_ipsec_fs_init(ipsec); if (ret) goto err_fs_init; + ipsec->fs = priv->fs; priv->ipsec = ipsec; netdev_dbg(priv->netdev, "IPSec attached to netdevice\n"); - return 0; + return; err_fs_init: + if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD) + mlx5e_ipsec_aso_cleanup(ipsec); +err_aso: destroy_workqueue(ipsec->wq); err_wq: kfree(ipsec); - return (ret != -EOPNOTSUPP) ? ret : 0; + mlx5_core_err(priv->mdev, "IPSec initialization failed, %d\n", ret); + return; } void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv) @@ -392,6 +440,8 @@ void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv) return; mlx5e_accel_ipsec_fs_cleanup(ipsec); + if (mlx5_ipsec_device_caps(priv->mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD) + mlx5e_ipsec_aso_cleanup(ipsec); destroy_workqueue(ipsec->wq); kfree(ipsec); priv->ipsec = NULL; @@ -427,6 +477,122 @@ static void mlx5e_xfrm_advance_esn_state(struct xfrm_state *x) queue_work(sa_entry->ipsec->wq, &modify_work->work); } +static void mlx5e_xfrm_update_curlft(struct xfrm_state *x) +{ + struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); + int err; + + lockdep_assert_held(&x->lock); + + if (sa_entry->attrs.soft_packet_limit == XFRM_INF) + /* Limits are not configured, as soft limit + * must be lowever than hard limit. + */ + return; + + err = mlx5e_ipsec_aso_query(sa_entry, NULL); + if (err) + return; + + mlx5e_ipsec_aso_update_curlft(sa_entry, &x->curlft.packets); +} + +static int mlx5e_xfrm_validate_policy(struct xfrm_policy *x) +{ + struct net_device *netdev = x->xdo.real_dev; + + if (x->type != XFRM_POLICY_TYPE_MAIN) { + netdev_info(netdev, "Cannot offload non-main policy types\n"); + return -EINVAL; + } + + /* Please pay attention that we support only one template */ + if (x->xfrm_nr > 1) { + netdev_info(netdev, "Cannot offload more than one template\n"); + return -EINVAL; + } + + if (x->xdo.dir != XFRM_DEV_OFFLOAD_IN && + x->xdo.dir != XFRM_DEV_OFFLOAD_OUT) { + netdev_info(netdev, "Cannot offload forward policy\n"); + return -EINVAL; + } + + if (!x->xfrm_vec[0].reqid) { + netdev_info(netdev, "Cannot offload policy without reqid\n"); + return -EINVAL; + } + + if (x->xdo.type != XFRM_DEV_OFFLOAD_PACKET) { + netdev_info(netdev, "Unsupported xfrm offload type\n"); + return -EINVAL; + } + + return 0; +} + +static void +mlx5e_ipsec_build_accel_pol_attrs(struct mlx5e_ipsec_pol_entry *pol_entry, + struct mlx5_accel_pol_xfrm_attrs *attrs) +{ + struct xfrm_policy *x = pol_entry->x; + struct xfrm_selector *sel; + + sel = &x->selector; + memset(attrs, 0, sizeof(*attrs)); + + memcpy(&attrs->saddr, sel->saddr.a6, sizeof(attrs->saddr)); + memcpy(&attrs->daddr, sel->daddr.a6, sizeof(attrs->daddr)); + attrs->family = sel->family; + attrs->dir = x->xdo.dir; + attrs->action = x->action; + attrs->type = XFRM_DEV_OFFLOAD_PACKET; + attrs->reqid = x->xfrm_vec[0].reqid; +} + +static int mlx5e_xfrm_add_policy(struct xfrm_policy *x) +{ + struct net_device *netdev = x->xdo.real_dev; + struct mlx5e_ipsec_pol_entry *pol_entry; + struct mlx5e_priv *priv; + int err; + + priv = netdev_priv(netdev); + if (!priv->ipsec) + return -EOPNOTSUPP; + + err = mlx5e_xfrm_validate_policy(x); + if (err) + return err; + + pol_entry = kzalloc(sizeof(*pol_entry), GFP_KERNEL); + if (!pol_entry) + return -ENOMEM; + + pol_entry->x = x; + pol_entry->ipsec = priv->ipsec; + + mlx5e_ipsec_build_accel_pol_attrs(pol_entry, &pol_entry->attrs); + err = mlx5e_accel_ipsec_fs_add_pol(pol_entry); + if (err) + goto err_fs; + + x->xdo.offload_handle = (unsigned long)pol_entry; + return 0; + +err_fs: + kfree(pol_entry); + return err; +} + +static void mlx5e_xfrm_free_policy(struct xfrm_policy *x) +{ + struct mlx5e_ipsec_pol_entry *pol_entry = to_ipsec_pol_entry(x); + + mlx5e_accel_ipsec_fs_del_pol(pol_entry); + kfree(pol_entry); +} + static const struct xfrmdev_ops mlx5e_ipsec_xfrmdev_ops = { .xdo_dev_state_add = mlx5e_xfrm_add_state, .xdo_dev_state_delete = mlx5e_xfrm_del_state, @@ -435,6 +601,18 @@ static const struct xfrmdev_ops mlx5e_ipsec_xfrmdev_ops = { .xdo_dev_state_advance_esn = mlx5e_xfrm_advance_esn_state, }; +static const struct xfrmdev_ops mlx5e_ipsec_packet_xfrmdev_ops = { + .xdo_dev_state_add = mlx5e_xfrm_add_state, + .xdo_dev_state_delete = mlx5e_xfrm_del_state, + .xdo_dev_state_free = mlx5e_xfrm_free_state, + .xdo_dev_offload_ok = mlx5e_ipsec_offload_ok, + .xdo_dev_state_advance_esn = mlx5e_xfrm_advance_esn_state, + + .xdo_dev_state_update_curlft = mlx5e_xfrm_update_curlft, + .xdo_dev_policy_add = mlx5e_xfrm_add_policy, + .xdo_dev_policy_free = mlx5e_xfrm_free_policy, +}; + void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -444,7 +622,12 @@ void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) return; mlx5_core_info(mdev, "mlx5e: IPSec ESP acceleration enabled\n"); - netdev->xfrmdev_ops = &mlx5e_ipsec_xfrmdev_ops; + + if (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD) + netdev->xfrmdev_ops = &mlx5e_ipsec_packet_xfrmdev_ops; + else + netdev->xfrmdev_ops = &mlx5e_ipsec_xfrmdev_ops; + netdev->features |= NETIF_F_HW_ESP; netdev->hw_enc_features |= NETIF_F_HW_ESP; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h index 16bcceec16c4..a92e19c4c499 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h @@ -34,27 +34,14 @@ #ifndef __MLX5E_IPSEC_H__ #define __MLX5E_IPSEC_H__ -#ifdef CONFIG_MLX5_EN_IPSEC - #include <linux/mlx5/device.h> #include <net/xfrm.h> #include <linux/idr.h> +#include "lib/aso.h" #define MLX5E_IPSEC_SADB_RX_BITS 10 #define MLX5E_IPSEC_ESN_SCOPE_MID 0x80000000L -enum mlx5_accel_esp_flags { - MLX5_ACCEL_ESP_FLAGS_TUNNEL = 0, /* Default */ - MLX5_ACCEL_ESP_FLAGS_TRANSPORT = 1UL << 0, - MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED = 1UL << 1, - MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP = 1UL << 2, -}; - -enum mlx5_accel_esp_action { - MLX5_ACCEL_ESP_ACTION_DECRYPT, - MLX5_ACCEL_ESP_ACTION_ENCRYPT, -}; - struct aes_gcm_keymat { u64 seq_iv; @@ -66,7 +53,6 @@ struct aes_gcm_keymat { }; struct mlx5_accel_esp_xfrm_attrs { - enum mlx5_accel_esp_action action; u32 esn; u32 spi; u32 flags; @@ -82,16 +68,37 @@ struct mlx5_accel_esp_xfrm_attrs { __be32 a6[4]; } daddr; - u8 is_ipv6; + u8 dir : 2; + u8 esn_overlap : 1; + u8 esn_trigger : 1; + u8 type : 2; + u8 family; + u32 replay_window; + u32 authsize; + u32 reqid; + u64 hard_packet_limit; + u64 soft_packet_limit; }; enum mlx5_ipsec_cap { MLX5_IPSEC_CAP_CRYPTO = 1 << 0, MLX5_IPSEC_CAP_ESN = 1 << 1, + MLX5_IPSEC_CAP_PACKET_OFFLOAD = 1 << 2, }; struct mlx5e_priv; +struct mlx5e_ipsec_hw_stats { + u64 ipsec_rx_pkts; + u64 ipsec_rx_bytes; + u64 ipsec_rx_drop_pkts; + u64 ipsec_rx_drop_bytes; + u64 ipsec_tx_pkts; + u64 ipsec_tx_bytes; + u64 ipsec_tx_drop_pkts; + u64 ipsec_tx_drop_bytes; +}; + struct mlx5e_ipsec_sw_stats { atomic64_t ipsec_rx_drop_sp_alloc; atomic64_t ipsec_rx_drop_sadb_miss; @@ -102,17 +109,38 @@ struct mlx5e_ipsec_sw_stats { atomic64_t ipsec_tx_drop_trailer; }; -struct mlx5e_accel_fs_esp; +struct mlx5e_ipsec_rx; struct mlx5e_ipsec_tx; +struct mlx5e_ipsec_work { + struct work_struct work; + struct mlx5e_ipsec *ipsec; + u32 id; +}; + +struct mlx5e_ipsec_aso { + u8 ctx[MLX5_ST_SZ_BYTES(ipsec_aso)]; + dma_addr_t dma_addr; + struct mlx5_aso *aso; + /* IPsec ASO caches data on every query call, + * so in nested calls, we can use this boolean to save + * recursive calls to mlx5e_ipsec_aso_query() + */ + u8 use_cache : 1; +}; + struct mlx5e_ipsec { struct mlx5_core_dev *mdev; - DECLARE_HASHTABLE(sadb_rx, MLX5E_IPSEC_SADB_RX_BITS); - spinlock_t sadb_rx_lock; /* Protects sadb_rx */ + struct xarray sadb; struct mlx5e_ipsec_sw_stats sw_stats; + struct mlx5e_ipsec_hw_stats hw_stats; struct workqueue_struct *wq; - struct mlx5e_accel_fs_esp *rx_fs; - struct mlx5e_ipsec_tx *tx_fs; + struct mlx5e_flow_steering *fs; + struct mlx5e_ipsec_rx *rx_ipv4; + struct mlx5e_ipsec_rx *rx_ipv6; + struct mlx5e_ipsec_tx *tx; + struct mlx5e_ipsec_aso *aso; + struct notifier_block nb; }; struct mlx5e_ipsec_esn_state { @@ -123,7 +151,8 @@ struct mlx5e_ipsec_esn_state { struct mlx5e_ipsec_rule { struct mlx5_flow_handle *rule; - struct mlx5_modify_hdr *set_modify_hdr; + struct mlx5_modify_hdr *modify_hdr; + struct mlx5_pkt_reformat *pkt_reformat; }; struct mlx5e_ipsec_modify_state_work { @@ -132,9 +161,7 @@ struct mlx5e_ipsec_modify_state_work { }; struct mlx5e_ipsec_sa_entry { - struct hlist_node hlist; /* Item in SADB_RX hashtable */ struct mlx5e_ipsec_esn_state esn_state; - unsigned int handle; /* Handle in SADB_RX */ struct xfrm_state *x; struct mlx5e_ipsec *ipsec; struct mlx5_accel_esp_xfrm_attrs attrs; @@ -146,19 +173,43 @@ struct mlx5e_ipsec_sa_entry { struct mlx5e_ipsec_modify_state_work modify_work; }; -int mlx5e_ipsec_init(struct mlx5e_priv *priv); +struct mlx5_accel_pol_xfrm_attrs { + union { + __be32 a4; + __be32 a6[4]; + } saddr; + + union { + __be32 a4; + __be32 a6[4]; + } daddr; + + u8 family; + u8 action; + u8 type : 2; + u8 dir : 2; + u32 reqid; +}; + +struct mlx5e_ipsec_pol_entry { + struct xfrm_policy *x; + struct mlx5e_ipsec *ipsec; + struct mlx5e_ipsec_rule ipsec_rule; + struct mlx5_accel_pol_xfrm_attrs attrs; +}; + +#ifdef CONFIG_MLX5_EN_IPSEC + +void mlx5e_ipsec_init(struct mlx5e_priv *priv); void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv); void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv); -struct xfrm_state *mlx5e_ipsec_sadb_rx_lookup(struct mlx5e_ipsec *dev, - unsigned int handle); - void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec); int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec); -int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_priv *priv, - struct mlx5e_ipsec_sa_entry *sa_entry); -void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_priv *priv, - struct mlx5e_ipsec_sa_entry *sa_entry); +int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry); +void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry); +int mlx5e_accel_ipsec_fs_add_pol(struct mlx5e_ipsec_pol_entry *pol_entry); +void mlx5e_accel_ipsec_fs_del_pol(struct mlx5e_ipsec_pol_entry *pol_entry); int mlx5_ipsec_create_sa_ctx(struct mlx5e_ipsec_sa_entry *sa_entry); void mlx5_ipsec_free_sa_ctx(struct mlx5e_ipsec_sa_entry *sa_entry); @@ -168,15 +219,33 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev); void mlx5_accel_esp_modify_xfrm(struct mlx5e_ipsec_sa_entry *sa_entry, const struct mlx5_accel_esp_xfrm_attrs *attrs); +int mlx5e_ipsec_aso_init(struct mlx5e_ipsec *ipsec); +void mlx5e_ipsec_aso_cleanup(struct mlx5e_ipsec *ipsec); + +int mlx5e_ipsec_aso_query(struct mlx5e_ipsec_sa_entry *sa_entry, + struct mlx5_wqe_aso_ctrl_seg *data); +void mlx5e_ipsec_aso_update_curlft(struct mlx5e_ipsec_sa_entry *sa_entry, + u64 *packets); + +void mlx5e_accel_ipsec_fs_read_stats(struct mlx5e_priv *priv, + void *ipsec_stats); + +void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, + struct mlx5_accel_esp_xfrm_attrs *attrs); static inline struct mlx5_core_dev * mlx5e_ipsec_sa2dev(struct mlx5e_ipsec_sa_entry *sa_entry) { return sa_entry->ipsec->mdev; } + +static inline struct mlx5_core_dev * +mlx5e_ipsec_pol2dev(struct mlx5e_ipsec_pol_entry *pol_entry) +{ + return pol_entry->ipsec->mdev; +} #else -static inline int mlx5e_ipsec_init(struct mlx5e_priv *priv) +static inline void mlx5e_ipsec_init(struct mlx5e_priv *priv) { - return 0; } static inline void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c index b859e4a4c744..9f19f4b59a70 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c @@ -9,53 +9,67 @@ #define NUM_IPSEC_FTE BIT(15) -enum accel_fs_esp_type { - ACCEL_FS_ESP4, - ACCEL_FS_ESP6, - ACCEL_FS_ESP_NUM_TYPES, +struct mlx5e_ipsec_fc { + struct mlx5_fc *cnt; + struct mlx5_fc *drop; }; -struct mlx5e_ipsec_rx_err { - struct mlx5_flow_table *ft; - struct mlx5_flow_handle *rule; - struct mlx5_modify_hdr *copy_modify_hdr; +struct mlx5e_ipsec_ft { + struct mutex mutex; /* Protect changes to this struct */ + struct mlx5_flow_table *pol; + struct mlx5_flow_table *sa; + struct mlx5_flow_table *status; + u32 refcnt; }; -struct mlx5e_accel_fs_esp_prot { - struct mlx5_flow_table *ft; - struct mlx5_flow_group *miss_group; - struct mlx5_flow_handle *miss_rule; - struct mlx5_flow_destination default_dest; - struct mlx5e_ipsec_rx_err rx_err; - u32 refcnt; - struct mutex prot_mutex; /* protect ESP4/ESP6 protocol */ +struct mlx5e_ipsec_miss { + struct mlx5_flow_group *group; + struct mlx5_flow_handle *rule; }; -struct mlx5e_accel_fs_esp { - struct mlx5e_accel_fs_esp_prot fs_prot[ACCEL_FS_ESP_NUM_TYPES]; +struct mlx5e_ipsec_rx { + struct mlx5e_ipsec_ft ft; + struct mlx5e_ipsec_miss pol; + struct mlx5e_ipsec_miss sa; + struct mlx5e_ipsec_rule status; + struct mlx5e_ipsec_fc *fc; }; struct mlx5e_ipsec_tx { + struct mlx5e_ipsec_ft ft; + struct mlx5e_ipsec_miss pol; struct mlx5_flow_namespace *ns; - struct mlx5_flow_table *ft; - struct mutex mutex; /* Protect IPsec TX steering */ - u32 refcnt; + struct mlx5e_ipsec_fc *fc; }; /* IPsec RX flow steering */ -static enum mlx5_traffic_types fs_esp2tt(enum accel_fs_esp_type i) +static enum mlx5_traffic_types family2tt(u32 family) { - if (i == ACCEL_FS_ESP4) + if (family == AF_INET) return MLX5_TT_IPV4_IPSEC_ESP; return MLX5_TT_IPV6_IPSEC_ESP; } -static int rx_err_add_rule(struct mlx5e_priv *priv, - struct mlx5e_accel_fs_esp_prot *fs_prot, - struct mlx5e_ipsec_rx_err *rx_err) +static struct mlx5_flow_table *ipsec_ft_create(struct mlx5_flow_namespace *ns, + int level, int prio, + int max_num_groups) +{ + struct mlx5_flow_table_attr ft_attr = {}; + + ft_attr.autogroup.num_reserved_entries = 1; + ft_attr.autogroup.max_num_groups = max_num_groups; + ft_attr.max_fte = NUM_IPSEC_FTE; + ft_attr.level = level; + ft_attr.prio = prio; + + return mlx5_create_auto_grouped_flow_table(ns, &ft_attr); +} + +static int ipsec_status_rule(struct mlx5_core_dev *mdev, + struct mlx5e_ipsec_rx *rx, + struct mlx5_flow_destination *dest) { u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {}; - struct mlx5_core_dev *mdev = priv->mdev; struct mlx5_flow_act flow_act = {}; struct mlx5_modify_hdr *modify_hdr; struct mlx5_flow_handle *fte; @@ -79,26 +93,26 @@ static int rx_err_add_rule(struct mlx5e_priv *priv, if (IS_ERR(modify_hdr)) { err = PTR_ERR(modify_hdr); - netdev_err(priv->netdev, - "fail to alloc ipsec copy modify_header_id err=%d\n", err); + mlx5_core_err(mdev, + "fail to alloc ipsec copy modify_header_id err=%d\n", err); goto out_spec; } /* create fte */ flow_act.action = MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_COUNT; flow_act.modify_hdr = modify_hdr; - fte = mlx5_add_flow_rules(rx_err->ft, spec, &flow_act, - &fs_prot->default_dest, 1); + fte = mlx5_add_flow_rules(rx->ft.status, spec, &flow_act, dest, 2); if (IS_ERR(fte)) { err = PTR_ERR(fte); - netdev_err(priv->netdev, "fail to add ipsec rx err copy rule err=%d\n", err); + mlx5_core_err(mdev, "fail to add ipsec rx err copy rule err=%d\n", err); goto out; } kvfree(spec); - rx_err->rule = fte; - rx_err->copy_modify_hdr = modify_hdr; + rx->status.rule = fte; + rx->status.modify_hdr = modify_hdr; return 0; out: @@ -108,13 +122,12 @@ out_spec: return err; } -static int rx_fs_create(struct mlx5e_priv *priv, - struct mlx5e_accel_fs_esp_prot *fs_prot) +static int ipsec_miss_create(struct mlx5_core_dev *mdev, + struct mlx5_flow_table *ft, + struct mlx5e_ipsec_miss *miss, + struct mlx5_flow_destination *dest) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct mlx5_flow_table *ft = fs_prot->ft; - struct mlx5_flow_group *miss_group; - struct mlx5_flow_handle *miss_rule; MLX5_DECLARE_FLOW_ACT(flow_act); struct mlx5_flow_spec *spec; u32 *flow_group_in; @@ -130,450 +143,888 @@ static int rx_fs_create(struct mlx5e_priv *priv, /* Create miss_group */ MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ft->max_fte - 1); MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ft->max_fte - 1); - miss_group = mlx5_create_flow_group(ft, flow_group_in); - if (IS_ERR(miss_group)) { - err = PTR_ERR(miss_group); - netdev_err(priv->netdev, "fail to create ipsec rx miss_group err=%d\n", err); + miss->group = mlx5_create_flow_group(ft, flow_group_in); + if (IS_ERR(miss->group)) { + err = PTR_ERR(miss->group); + mlx5_core_err(mdev, "fail to create IPsec miss_group err=%d\n", + err); goto out; } - fs_prot->miss_group = miss_group; /* Create miss rule */ - miss_rule = mlx5_add_flow_rules(ft, spec, &flow_act, &fs_prot->default_dest, 1); - if (IS_ERR(miss_rule)) { - mlx5_destroy_flow_group(fs_prot->miss_group); - err = PTR_ERR(miss_rule); - netdev_err(priv->netdev, "fail to create ipsec rx miss_rule err=%d\n", err); + miss->rule = mlx5_add_flow_rules(ft, spec, &flow_act, dest, 1); + if (IS_ERR(miss->rule)) { + mlx5_destroy_flow_group(miss->group); + err = PTR_ERR(miss->rule); + mlx5_core_err(mdev, "fail to create IPsec miss_rule err=%d\n", + err); goto out; } - fs_prot->miss_rule = miss_rule; out: kvfree(flow_group_in); kvfree(spec); return err; } -static void rx_destroy(struct mlx5e_priv *priv, enum accel_fs_esp_type type) +static void rx_destroy(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_rx *rx) { - struct mlx5e_accel_fs_esp_prot *fs_prot; - struct mlx5e_accel_fs_esp *accel_esp; - - accel_esp = priv->ipsec->rx_fs; - - /* The netdev unreg already happened, so all offloaded rule are already removed */ - fs_prot = &accel_esp->fs_prot[type]; + mlx5_del_flow_rules(rx->pol.rule); + mlx5_destroy_flow_group(rx->pol.group); + mlx5_destroy_flow_table(rx->ft.pol); - mlx5_del_flow_rules(fs_prot->miss_rule); - mlx5_destroy_flow_group(fs_prot->miss_group); - mlx5_destroy_flow_table(fs_prot->ft); + mlx5_del_flow_rules(rx->sa.rule); + mlx5_destroy_flow_group(rx->sa.group); + mlx5_destroy_flow_table(rx->ft.sa); - mlx5_del_flow_rules(fs_prot->rx_err.rule); - mlx5_modify_header_dealloc(priv->mdev, fs_prot->rx_err.copy_modify_hdr); - mlx5_destroy_flow_table(fs_prot->rx_err.ft); + mlx5_del_flow_rules(rx->status.rule); + mlx5_modify_header_dealloc(mdev, rx->status.modify_hdr); + mlx5_destroy_flow_table(rx->ft.status); } -static int rx_create(struct mlx5e_priv *priv, enum accel_fs_esp_type type) +static int rx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec, + struct mlx5e_ipsec_rx *rx, u32 family) { - struct mlx5_flow_namespace *ns = mlx5e_fs_get_ns(priv->fs, false); - struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(priv->fs, false); - struct mlx5_flow_table_attr ft_attr = {}; - struct mlx5e_accel_fs_esp_prot *fs_prot; - struct mlx5e_accel_fs_esp *accel_esp; + struct mlx5_flow_namespace *ns = mlx5e_fs_get_ns(ipsec->fs, false); + struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false); + struct mlx5_flow_destination dest[2]; struct mlx5_flow_table *ft; int err; - accel_esp = priv->ipsec->rx_fs; - fs_prot = &accel_esp->fs_prot[type]; - fs_prot->default_dest = - mlx5_ttc_get_default_dest(ttc, fs_esp2tt(type)); - - ft_attr.max_fte = 1; - ft_attr.autogroup.max_num_groups = 1; - ft_attr.level = MLX5E_ACCEL_FS_ESP_FT_ERR_LEVEL; - ft_attr.prio = MLX5E_NIC_PRIO; - ft = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); + ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_ESP_FT_ERR_LEVEL, + MLX5E_NIC_PRIO, 1); if (IS_ERR(ft)) return PTR_ERR(ft); - fs_prot->rx_err.ft = ft; - err = rx_err_add_rule(priv, fs_prot, &fs_prot->rx_err); + rx->ft.status = ft; + + dest[0] = mlx5_ttc_get_default_dest(ttc, family2tt(family)); + dest[1].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[1].counter_id = mlx5_fc_id(rx->fc->cnt); + err = ipsec_status_rule(mdev, rx, dest); if (err) goto err_add; /* Create FT */ - ft_attr.max_fte = NUM_IPSEC_FTE; - ft_attr.level = MLX5E_ACCEL_FS_ESP_FT_LEVEL; - ft_attr.prio = MLX5E_NIC_PRIO; - ft_attr.autogroup.num_reserved_entries = 1; - ft_attr.autogroup.max_num_groups = 1; - ft = mlx5_create_auto_grouped_flow_table(ns, &ft_attr); + ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_ESP_FT_LEVEL, MLX5E_NIC_PRIO, + 2); if (IS_ERR(ft)) { err = PTR_ERR(ft); goto err_fs_ft; } - fs_prot->ft = ft; + rx->ft.sa = ft; - err = rx_fs_create(priv, fs_prot); + err = ipsec_miss_create(mdev, rx->ft.sa, &rx->sa, dest); if (err) goto err_fs; + ft = ipsec_ft_create(ns, MLX5E_ACCEL_FS_POL_FT_LEVEL, MLX5E_NIC_PRIO, + 2); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + goto err_pol_ft; + } + rx->ft.pol = ft; + memset(dest, 0x00, 2 * sizeof(*dest)); + dest[0].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest[0].ft = rx->ft.sa; + err = ipsec_miss_create(mdev, rx->ft.pol, &rx->pol, dest); + if (err) + goto err_pol_miss; + return 0; +err_pol_miss: + mlx5_destroy_flow_table(rx->ft.pol); +err_pol_ft: + mlx5_del_flow_rules(rx->sa.rule); + mlx5_destroy_flow_group(rx->sa.group); err_fs: - mlx5_destroy_flow_table(fs_prot->ft); + mlx5_destroy_flow_table(rx->ft.sa); err_fs_ft: - mlx5_del_flow_rules(fs_prot->rx_err.rule); - mlx5_modify_header_dealloc(priv->mdev, fs_prot->rx_err.copy_modify_hdr); + mlx5_del_flow_rules(rx->status.rule); + mlx5_modify_header_dealloc(mdev, rx->status.modify_hdr); err_add: - mlx5_destroy_flow_table(fs_prot->rx_err.ft); + mlx5_destroy_flow_table(rx->ft.status); return err; } -static int rx_ft_get(struct mlx5e_priv *priv, enum accel_fs_esp_type type) +static struct mlx5e_ipsec_rx *rx_ft_get(struct mlx5_core_dev *mdev, + struct mlx5e_ipsec *ipsec, u32 family) { - struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(priv->fs, false); - struct mlx5e_accel_fs_esp_prot *fs_prot; + struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false); struct mlx5_flow_destination dest = {}; - struct mlx5e_accel_fs_esp *accel_esp; + struct mlx5e_ipsec_rx *rx; int err = 0; - accel_esp = priv->ipsec->rx_fs; - fs_prot = &accel_esp->fs_prot[type]; - mutex_lock(&fs_prot->prot_mutex); - if (fs_prot->refcnt) + if (family == AF_INET) + rx = ipsec->rx_ipv4; + else + rx = ipsec->rx_ipv6; + + mutex_lock(&rx->ft.mutex); + if (rx->ft.refcnt) goto skip; /* create FT */ - err = rx_create(priv, type); + err = rx_create(mdev, ipsec, rx, family); if (err) goto out; /* connect */ dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest.ft = fs_prot->ft; - mlx5_ttc_fwd_dest(ttc, fs_esp2tt(type), &dest); + dest.ft = rx->ft.pol; + mlx5_ttc_fwd_dest(ttc, family2tt(family), &dest); skip: - fs_prot->refcnt++; + rx->ft.refcnt++; out: - mutex_unlock(&fs_prot->prot_mutex); - return err; + mutex_unlock(&rx->ft.mutex); + if (err) + return ERR_PTR(err); + return rx; } -static void rx_ft_put(struct mlx5e_priv *priv, enum accel_fs_esp_type type) +static void rx_ft_put(struct mlx5_core_dev *mdev, struct mlx5e_ipsec *ipsec, + u32 family) { - struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(priv->fs, false); - struct mlx5e_accel_fs_esp_prot *fs_prot; - struct mlx5e_accel_fs_esp *accel_esp; + struct mlx5_ttc_table *ttc = mlx5e_fs_get_ttc(ipsec->fs, false); + struct mlx5e_ipsec_rx *rx; + + if (family == AF_INET) + rx = ipsec->rx_ipv4; + else + rx = ipsec->rx_ipv6; - accel_esp = priv->ipsec->rx_fs; - fs_prot = &accel_esp->fs_prot[type]; - mutex_lock(&fs_prot->prot_mutex); - fs_prot->refcnt--; - if (fs_prot->refcnt) + mutex_lock(&rx->ft.mutex); + rx->ft.refcnt--; + if (rx->ft.refcnt) goto out; /* disconnect */ - mlx5_ttc_fwd_default_dest(ttc, fs_esp2tt(type)); + mlx5_ttc_fwd_default_dest(ttc, family2tt(family)); /* remove FT */ - rx_destroy(priv, type); + rx_destroy(mdev, rx); out: - mutex_unlock(&fs_prot->prot_mutex); + mutex_unlock(&rx->ft.mutex); } /* IPsec TX flow steering */ -static int tx_create(struct mlx5e_priv *priv) +static int tx_create(struct mlx5_core_dev *mdev, struct mlx5e_ipsec_tx *tx) { - struct mlx5_flow_table_attr ft_attr = {}; - struct mlx5e_ipsec *ipsec = priv->ipsec; + struct mlx5_flow_destination dest = {}; struct mlx5_flow_table *ft; int err; - ft_attr.max_fte = NUM_IPSEC_FTE; - ft_attr.autogroup.max_num_groups = 1; - ft = mlx5_create_auto_grouped_flow_table(ipsec->tx_fs->ns, &ft_attr); + ft = ipsec_ft_create(tx->ns, 1, 0, 4); + if (IS_ERR(ft)) + return PTR_ERR(ft); + + tx->ft.sa = ft; + + ft = ipsec_ft_create(tx->ns, 0, 0, 2); if (IS_ERR(ft)) { err = PTR_ERR(ft); - netdev_err(priv->netdev, "fail to create ipsec tx ft err=%d\n", err); - return err; + goto err_pol_ft; } - ipsec->tx_fs->ft = ft; + tx->ft.pol = ft; + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest.ft = tx->ft.sa; + err = ipsec_miss_create(mdev, tx->ft.pol, &tx->pol, &dest); + if (err) + goto err_pol_miss; return 0; + +err_pol_miss: + mlx5_destroy_flow_table(tx->ft.pol); +err_pol_ft: + mlx5_destroy_flow_table(tx->ft.sa); + return err; } -static int tx_ft_get(struct mlx5e_priv *priv) +static struct mlx5e_ipsec_tx *tx_ft_get(struct mlx5_core_dev *mdev, + struct mlx5e_ipsec *ipsec) { - struct mlx5e_ipsec_tx *tx_fs = priv->ipsec->tx_fs; + struct mlx5e_ipsec_tx *tx = ipsec->tx; int err = 0; - mutex_lock(&tx_fs->mutex); - if (tx_fs->refcnt) + mutex_lock(&tx->ft.mutex); + if (tx->ft.refcnt) goto skip; - err = tx_create(priv); + err = tx_create(mdev, tx); if (err) goto out; skip: - tx_fs->refcnt++; + tx->ft.refcnt++; out: - mutex_unlock(&tx_fs->mutex); - return err; + mutex_unlock(&tx->ft.mutex); + if (err) + return ERR_PTR(err); + return tx; } -static void tx_ft_put(struct mlx5e_priv *priv) +static void tx_ft_put(struct mlx5e_ipsec *ipsec) { - struct mlx5e_ipsec_tx *tx_fs = priv->ipsec->tx_fs; + struct mlx5e_ipsec_tx *tx = ipsec->tx; - mutex_lock(&tx_fs->mutex); - tx_fs->refcnt--; - if (tx_fs->refcnt) + mutex_lock(&tx->ft.mutex); + tx->ft.refcnt--; + if (tx->ft.refcnt) goto out; - mlx5_destroy_flow_table(tx_fs->ft); + mlx5_del_flow_rules(tx->pol.rule); + mlx5_destroy_flow_group(tx->pol.group); + mlx5_destroy_flow_table(tx->ft.pol); + mlx5_destroy_flow_table(tx->ft.sa); out: - mutex_unlock(&tx_fs->mutex); + mutex_unlock(&tx->ft.mutex); } -static void setup_fte_common(struct mlx5_accel_esp_xfrm_attrs *attrs, - u32 ipsec_obj_id, - struct mlx5_flow_spec *spec, - struct mlx5_flow_act *flow_act) +static void setup_fte_addr4(struct mlx5_flow_spec *spec, __be32 *saddr, + __be32 *daddr) { - u8 ip_version = attrs->is_ipv6 ? 6 : 4; - - spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS | MLX5_MATCH_MISC_PARAMETERS; + spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; - /* ip_version */ MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_version); - MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_version, ip_version); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_version, 4); + + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), saddr, 4); + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), daddr, 4); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4); +} - /* Non fragmented */ - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.frag); - MLX5_SET(fte_match_param, spec->match_value, outer_headers.frag, 0); +static void setup_fte_addr6(struct mlx5_flow_spec *spec, __be32 *saddr, + __be32 *daddr) +{ + spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_version); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_version, 6); + + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), saddr, 16); + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), daddr, 16); + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), 0xff, 16); + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), 0xff, 16); +} + +static void setup_fte_esp(struct mlx5_flow_spec *spec) +{ /* ESP header */ + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_protocol); MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_protocol, IPPROTO_ESP); +} +static void setup_fte_spi(struct mlx5_flow_spec *spec, u32 spi) +{ /* SPI number */ + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, misc_parameters.outer_esp_spi); + MLX5_SET(fte_match_param, spec->match_value, misc_parameters.outer_esp_spi, spi); +} + +static void setup_fte_no_frags(struct mlx5_flow_spec *spec) +{ + /* Non fragmented */ + spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS; + + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.frag); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.frag, 0); +} + +static void setup_fte_reg_a(struct mlx5_flow_spec *spec) +{ + /* Add IPsec indicator in metadata_reg_a */ + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; + + MLX5_SET(fte_match_param, spec->match_criteria, + misc_parameters_2.metadata_reg_a, MLX5_ETH_WQE_FT_META_IPSEC); MLX5_SET(fte_match_param, spec->match_value, - misc_parameters.outer_esp_spi, attrs->spi); - - if (ip_version == 4) { - memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, - outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), - &attrs->saddr.a4, 4); - memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, - outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), - &attrs->daddr.a4, 4); - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, - outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4); - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, - outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4); - } else { - memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, - outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), - &attrs->saddr.a6, 16); - memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, - outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), - &attrs->daddr.a6, 16); - memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, - outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), - 0xff, 16); - memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, - outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), - 0xff, 16); - } + misc_parameters_2.metadata_reg_a, MLX5_ETH_WQE_FT_META_IPSEC); +} - flow_act->crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC; - flow_act->crypto.obj_id = ipsec_obj_id; - flow_act->flags |= FLOW_ACT_NO_APPEND; +static void setup_fte_reg_c0(struct mlx5_flow_spec *spec, u32 reqid) +{ + /* Pass policy check before choosing this SA */ + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; + + MLX5_SET(fte_match_param, spec->match_criteria, + misc_parameters_2.metadata_reg_c_0, reqid); + MLX5_SET(fte_match_param, spec->match_value, + misc_parameters_2.metadata_reg_c_0, reqid); } -static int rx_add_rule(struct mlx5e_priv *priv, - struct mlx5e_ipsec_sa_entry *sa_entry) +static int setup_modify_header(struct mlx5_core_dev *mdev, u32 val, u8 dir, + struct mlx5_flow_act *flow_act) { u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {}; - struct mlx5e_ipsec_rule *ipsec_rule = &sa_entry->ipsec_rule; + enum mlx5_flow_namespace_type ns_type; + struct mlx5_modify_hdr *modify_hdr; + + MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET); + switch (dir) { + case XFRM_DEV_OFFLOAD_IN: + MLX5_SET(set_action_in, action, field, + MLX5_ACTION_IN_FIELD_METADATA_REG_B); + ns_type = MLX5_FLOW_NAMESPACE_KERNEL; + break; + case XFRM_DEV_OFFLOAD_OUT: + MLX5_SET(set_action_in, action, field, + MLX5_ACTION_IN_FIELD_METADATA_REG_C_0); + ns_type = MLX5_FLOW_NAMESPACE_EGRESS; + break; + default: + return -EINVAL; + } + + MLX5_SET(set_action_in, action, data, val); + MLX5_SET(set_action_in, action, offset, 0); + MLX5_SET(set_action_in, action, length, 32); + + modify_hdr = mlx5_modify_header_alloc(mdev, ns_type, 1, action); + if (IS_ERR(modify_hdr)) { + mlx5_core_err(mdev, "Failed to allocate modify_header %ld\n", + PTR_ERR(modify_hdr)); + return PTR_ERR(modify_hdr); + } + + flow_act->modify_hdr = modify_hdr; + flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + return 0; +} + +static int setup_pkt_reformat(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm_attrs *attrs, + struct mlx5_flow_act *flow_act) +{ + enum mlx5_flow_namespace_type ns_type = MLX5_FLOW_NAMESPACE_EGRESS; + struct mlx5_pkt_reformat_params reformat_params = {}; + struct mlx5_pkt_reformat *pkt_reformat; + u8 reformatbf[16] = {}; + __be32 spi; + + if (attrs->dir == XFRM_DEV_OFFLOAD_IN) { + reformat_params.type = MLX5_REFORMAT_TYPE_DEL_ESP_TRANSPORT; + ns_type = MLX5_FLOW_NAMESPACE_KERNEL; + goto cmd; + } + + if (attrs->family == AF_INET) + reformat_params.type = + MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV4; + else + reformat_params.type = + MLX5_REFORMAT_TYPE_ADD_ESP_TRANSPORT_OVER_IPV6; + + /* convert to network format */ + spi = htonl(attrs->spi); + memcpy(reformatbf, &spi, 4); + + reformat_params.param_0 = attrs->authsize; + reformat_params.size = sizeof(reformatbf); + reformat_params.data = &reformatbf; + +cmd: + pkt_reformat = + mlx5_packet_reformat_alloc(mdev, &reformat_params, ns_type); + if (IS_ERR(pkt_reformat)) + return PTR_ERR(pkt_reformat); + + flow_act->pkt_reformat = pkt_reformat; + flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT; + return 0; +} + +static int rx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry) +{ struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs; - u32 ipsec_obj_id = sa_entry->ipsec_obj_id; - struct mlx5_modify_hdr *modify_hdr = NULL; - struct mlx5e_accel_fs_esp_prot *fs_prot; + struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry); + struct mlx5e_ipsec *ipsec = sa_entry->ipsec; struct mlx5_flow_destination dest = {}; - struct mlx5e_accel_fs_esp *accel_esp; struct mlx5_flow_act flow_act = {}; struct mlx5_flow_handle *rule; - enum accel_fs_esp_type type; struct mlx5_flow_spec *spec; - int err = 0; + struct mlx5e_ipsec_rx *rx; + int err; - accel_esp = priv->ipsec->rx_fs; - type = attrs->is_ipv6 ? ACCEL_FS_ESP6 : ACCEL_FS_ESP4; - fs_prot = &accel_esp->fs_prot[type]; - - err = rx_ft_get(priv, type); - if (err) - return err; + rx = rx_ft_get(mdev, ipsec, attrs->family); + if (IS_ERR(rx)) + return PTR_ERR(rx); spec = kvzalloc(sizeof(*spec), GFP_KERNEL); if (!spec) { err = -ENOMEM; - goto out_err; + goto err_alloc; } - setup_fte_common(attrs, ipsec_obj_id, spec, &flow_act); + if (attrs->family == AF_INET) + setup_fte_addr4(spec, &attrs->saddr.a4, &attrs->daddr.a4); + else + setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6); - /* Set bit[31] ipsec marker */ - /* Set bit[23-0] ipsec_obj_id */ - MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET); - MLX5_SET(set_action_in, action, field, MLX5_ACTION_IN_FIELD_METADATA_REG_B); - MLX5_SET(set_action_in, action, data, (ipsec_obj_id | BIT(31))); - MLX5_SET(set_action_in, action, offset, 0); - MLX5_SET(set_action_in, action, length, 32); + setup_fte_spi(spec, attrs->spi); + setup_fte_esp(spec); + setup_fte_no_frags(spec); - modify_hdr = mlx5_modify_header_alloc(priv->mdev, MLX5_FLOW_NAMESPACE_KERNEL, - 1, action); - if (IS_ERR(modify_hdr)) { - err = PTR_ERR(modify_hdr); - netdev_err(priv->netdev, - "fail to alloc ipsec set modify_header_id err=%d\n", err); - modify_hdr = NULL; - goto out_err; + err = setup_modify_header(mdev, sa_entry->ipsec_obj_id | BIT(31), + XFRM_DEV_OFFLOAD_IN, &flow_act); + if (err) + goto err_mod_header; + + switch (attrs->type) { + case XFRM_DEV_OFFLOAD_PACKET: + err = setup_pkt_reformat(mdev, attrs, &flow_act); + if (err) + goto err_pkt_reformat; + break; + default: + break; } - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | - MLX5_FLOW_CONTEXT_ACTION_CRYPTO_DECRYPT | - MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC; + flow_act.crypto.obj_id = sa_entry->ipsec_obj_id; + flow_act.flags |= FLOW_ACT_NO_APPEND; + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_CRYPTO_DECRYPT; dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - flow_act.modify_hdr = modify_hdr; - dest.ft = fs_prot->rx_err.ft; - rule = mlx5_add_flow_rules(fs_prot->ft, spec, &flow_act, &dest, 1); + dest.ft = rx->ft.status; + rule = mlx5_add_flow_rules(rx->ft.sa, spec, &flow_act, &dest, 1); if (IS_ERR(rule)) { err = PTR_ERR(rule); - netdev_err(priv->netdev, "fail to add ipsec rule attrs->action=0x%x, err=%d\n", - attrs->action, err); - goto out_err; + mlx5_core_err(mdev, "fail to add RX ipsec rule err=%d\n", err); + goto err_add_flow; } + kvfree(spec); - ipsec_rule->rule = rule; - ipsec_rule->set_modify_hdr = modify_hdr; - goto out; - -out_err: - if (modify_hdr) - mlx5_modify_header_dealloc(priv->mdev, modify_hdr); - rx_ft_put(priv, type); + sa_entry->ipsec_rule.rule = rule; + sa_entry->ipsec_rule.modify_hdr = flow_act.modify_hdr; + sa_entry->ipsec_rule.pkt_reformat = flow_act.pkt_reformat; + return 0; -out: +err_add_flow: + if (flow_act.pkt_reformat) + mlx5_packet_reformat_dealloc(mdev, flow_act.pkt_reformat); +err_pkt_reformat: + mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr); +err_mod_header: kvfree(spec); +err_alloc: + rx_ft_put(mdev, ipsec, attrs->family); return err; } -static int tx_add_rule(struct mlx5e_priv *priv, - struct mlx5e_ipsec_sa_entry *sa_entry) +static int tx_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry) { + struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs; + struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry); + struct mlx5e_ipsec *ipsec = sa_entry->ipsec; + struct mlx5_flow_destination dest = {}; struct mlx5_flow_act flow_act = {}; struct mlx5_flow_handle *rule; struct mlx5_flow_spec *spec; + struct mlx5e_ipsec_tx *tx; int err = 0; - err = tx_ft_get(priv); - if (err) - return err; + tx = tx_ft_get(mdev, ipsec); + if (IS_ERR(tx)) + return PTR_ERR(tx); spec = kvzalloc(sizeof(*spec), GFP_KERNEL); if (!spec) { err = -ENOMEM; - goto out; + goto err_alloc; } - setup_fte_common(&sa_entry->attrs, sa_entry->ipsec_obj_id, spec, - &flow_act); + if (attrs->family == AF_INET) + setup_fte_addr4(spec, &attrs->saddr.a4, &attrs->daddr.a4); + else + setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6); + + setup_fte_no_frags(spec); + + switch (attrs->type) { + case XFRM_DEV_OFFLOAD_CRYPTO: + setup_fte_spi(spec, attrs->spi); + setup_fte_esp(spec); + setup_fte_reg_a(spec); + break; + case XFRM_DEV_OFFLOAD_PACKET: + setup_fte_reg_c0(spec, attrs->reqid); + err = setup_pkt_reformat(mdev, attrs, &flow_act); + if (err) + goto err_pkt_reformat; + break; + default: + break; + } - /* Add IPsec indicator in metadata_reg_a */ - spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; - MLX5_SET(fte_match_param, spec->match_criteria, misc_parameters_2.metadata_reg_a, - MLX5_ETH_WQE_FT_META_IPSEC); - MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_a, - MLX5_ETH_WQE_FT_META_IPSEC); - - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW | - MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT; - rule = mlx5_add_flow_rules(priv->ipsec->tx_fs->ft, spec, &flow_act, NULL, 0); + flow_act.crypto.type = MLX5_FLOW_CONTEXT_ENCRYPT_DECRYPT_TYPE_IPSEC; + flow_act.crypto.obj_id = sa_entry->ipsec_obj_id; + flow_act.flags |= FLOW_ACT_NO_APPEND; + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_ALLOW | + MLX5_FLOW_CONTEXT_ACTION_CRYPTO_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest.counter_id = mlx5_fc_id(tx->fc->cnt); + rule = mlx5_add_flow_rules(tx->ft.sa, spec, &flow_act, &dest, 1); if (IS_ERR(rule)) { err = PTR_ERR(rule); - netdev_err(priv->netdev, "fail to add ipsec rule attrs->action=0x%x, err=%d\n", - sa_entry->attrs.action, err); - goto out; + mlx5_core_err(mdev, "fail to add TX ipsec rule err=%d\n", err); + goto err_add_flow; } + kvfree(spec); sa_entry->ipsec_rule.rule = rule; + sa_entry->ipsec_rule.pkt_reformat = flow_act.pkt_reformat; + return 0; -out: +err_add_flow: + if (flow_act.pkt_reformat) + mlx5_packet_reformat_dealloc(mdev, flow_act.pkt_reformat); +err_pkt_reformat: kvfree(spec); +err_alloc: + tx_ft_put(ipsec); + return err; +} + +static int tx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry) +{ + struct mlx5_accel_pol_xfrm_attrs *attrs = &pol_entry->attrs; + struct mlx5_core_dev *mdev = mlx5e_ipsec_pol2dev(pol_entry); + struct mlx5_flow_destination dest[2] = {}; + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_handle *rule; + struct mlx5_flow_spec *spec; + struct mlx5e_ipsec_tx *tx; + int err, dstn = 0; + + tx = tx_ft_get(mdev, pol_entry->ipsec); + if (IS_ERR(tx)) + return PTR_ERR(tx); + + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) { + err = -ENOMEM; + goto err_alloc; + } + + if (attrs->family == AF_INET) + setup_fte_addr4(spec, &attrs->saddr.a4, &attrs->daddr.a4); + else + setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6); + + setup_fte_no_frags(spec); + + err = setup_modify_header(mdev, attrs->reqid, XFRM_DEV_OFFLOAD_OUT, + &flow_act); if (err) - tx_ft_put(priv); + goto err_mod_header; + + switch (attrs->action) { + case XFRM_POLICY_ALLOW: + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + break; + case XFRM_POLICY_BLOCK: + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_DROP | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest[dstn].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[dstn].counter_id = mlx5_fc_id(tx->fc->drop); + dstn++; + break; + default: + WARN_ON(true); + err = -EINVAL; + goto err_action; + } + + flow_act.flags |= FLOW_ACT_NO_APPEND; + dest[dstn].ft = tx->ft.sa; + dest[dstn].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dstn++; + rule = mlx5_add_flow_rules(tx->ft.pol, spec, &flow_act, dest, dstn); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + mlx5_core_err(mdev, "fail to add TX ipsec rule err=%d\n", err); + goto err_action; + } + + kvfree(spec); + pol_entry->ipsec_rule.rule = rule; + pol_entry->ipsec_rule.modify_hdr = flow_act.modify_hdr; + return 0; + +err_action: + mlx5_modify_header_dealloc(mdev, flow_act.modify_hdr); +err_mod_header: + kvfree(spec); +err_alloc: + tx_ft_put(pol_entry->ipsec); return err; } -int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_priv *priv, - struct mlx5e_ipsec_sa_entry *sa_entry) +static int rx_add_policy(struct mlx5e_ipsec_pol_entry *pol_entry) { - if (sa_entry->attrs.action == MLX5_ACCEL_ESP_ACTION_ENCRYPT) - return tx_add_rule(priv, sa_entry); + struct mlx5_accel_pol_xfrm_attrs *attrs = &pol_entry->attrs; + struct mlx5_core_dev *mdev = mlx5e_ipsec_pol2dev(pol_entry); + struct mlx5_flow_destination dest[2]; + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_handle *rule; + struct mlx5_flow_spec *spec; + struct mlx5e_ipsec_rx *rx; + int err, dstn = 0; + + rx = rx_ft_get(mdev, pol_entry->ipsec, attrs->family); + if (IS_ERR(rx)) + return PTR_ERR(rx); + + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) { + err = -ENOMEM; + goto err_alloc; + } - return rx_add_rule(priv, sa_entry); + if (attrs->family == AF_INET) + setup_fte_addr4(spec, &attrs->saddr.a4, &attrs->daddr.a4); + else + setup_fte_addr6(spec, attrs->saddr.a6, attrs->daddr.a6); + + setup_fte_no_frags(spec); + + switch (attrs->action) { + case XFRM_POLICY_ALLOW: + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + break; + case XFRM_POLICY_BLOCK: + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT; + dest[dstn].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest[dstn].counter_id = mlx5_fc_id(rx->fc->drop); + dstn++; + break; + default: + WARN_ON(true); + err = -EINVAL; + goto err_action; + } + + flow_act.flags |= FLOW_ACT_NO_APPEND; + dest[dstn].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + dest[dstn].ft = rx->ft.sa; + dstn++; + rule = mlx5_add_flow_rules(rx->ft.pol, spec, &flow_act, dest, dstn); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + mlx5_core_err(mdev, "Fail to add RX IPsec policy rule err=%d\n", err); + goto err_action; + } + + kvfree(spec); + pol_entry->ipsec_rule.rule = rule; + return 0; + +err_action: + kvfree(spec); +err_alloc: + rx_ft_put(mdev, pol_entry->ipsec, attrs->family); + return err; } -void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_priv *priv, - struct mlx5e_ipsec_sa_entry *sa_entry) +static void ipsec_fs_destroy_counters(struct mlx5e_ipsec *ipsec) +{ + struct mlx5e_ipsec_rx *rx_ipv4 = ipsec->rx_ipv4; + struct mlx5_core_dev *mdev = ipsec->mdev; + struct mlx5e_ipsec_tx *tx = ipsec->tx; + + mlx5_fc_destroy(mdev, tx->fc->drop); + mlx5_fc_destroy(mdev, tx->fc->cnt); + kfree(tx->fc); + mlx5_fc_destroy(mdev, rx_ipv4->fc->drop); + mlx5_fc_destroy(mdev, rx_ipv4->fc->cnt); + kfree(rx_ipv4->fc); +} + +static int ipsec_fs_init_counters(struct mlx5e_ipsec *ipsec) +{ + struct mlx5e_ipsec_rx *rx_ipv4 = ipsec->rx_ipv4; + struct mlx5e_ipsec_rx *rx_ipv6 = ipsec->rx_ipv6; + struct mlx5_core_dev *mdev = ipsec->mdev; + struct mlx5e_ipsec_tx *tx = ipsec->tx; + struct mlx5e_ipsec_fc *fc; + struct mlx5_fc *counter; + int err; + + fc = kzalloc(sizeof(*rx_ipv4->fc), GFP_KERNEL); + if (!fc) + return -ENOMEM; + + /* Both IPv4 and IPv6 point to same flow counters struct. */ + rx_ipv4->fc = fc; + rx_ipv6->fc = fc; + counter = mlx5_fc_create(mdev, false); + if (IS_ERR(counter)) { + err = PTR_ERR(counter); + goto err_rx_cnt; + } + + fc->cnt = counter; + counter = mlx5_fc_create(mdev, false); + if (IS_ERR(counter)) { + err = PTR_ERR(counter); + goto err_rx_drop; + } + + fc->drop = counter; + fc = kzalloc(sizeof(*tx->fc), GFP_KERNEL); + if (!fc) { + err = -ENOMEM; + goto err_tx_fc; + } + + tx->fc = fc; + counter = mlx5_fc_create(mdev, false); + if (IS_ERR(counter)) { + err = PTR_ERR(counter); + goto err_tx_cnt; + } + + fc->cnt = counter; + counter = mlx5_fc_create(mdev, false); + if (IS_ERR(counter)) { + err = PTR_ERR(counter); + goto err_tx_drop; + } + + fc->drop = counter; + return 0; + +err_tx_drop: + mlx5_fc_destroy(mdev, tx->fc->cnt); +err_tx_cnt: + kfree(tx->fc); +err_tx_fc: + mlx5_fc_destroy(mdev, rx_ipv4->fc->drop); +err_rx_drop: + mlx5_fc_destroy(mdev, rx_ipv4->fc->cnt); +err_rx_cnt: + kfree(rx_ipv4->fc); + return err; +} + +void mlx5e_accel_ipsec_fs_read_stats(struct mlx5e_priv *priv, void *ipsec_stats) +{ + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_ipsec *ipsec = priv->ipsec; + struct mlx5e_ipsec_hw_stats *stats; + struct mlx5e_ipsec_fc *fc; + + stats = (struct mlx5e_ipsec_hw_stats *)ipsec_stats; + + stats->ipsec_rx_pkts = 0; + stats->ipsec_rx_bytes = 0; + stats->ipsec_rx_drop_pkts = 0; + stats->ipsec_rx_drop_bytes = 0; + stats->ipsec_tx_pkts = 0; + stats->ipsec_tx_bytes = 0; + stats->ipsec_tx_drop_pkts = 0; + stats->ipsec_tx_drop_bytes = 0; + + fc = ipsec->rx_ipv4->fc; + mlx5_fc_query(mdev, fc->cnt, &stats->ipsec_rx_pkts, &stats->ipsec_rx_bytes); + mlx5_fc_query(mdev, fc->drop, &stats->ipsec_rx_drop_pkts, + &stats->ipsec_rx_drop_bytes); + + fc = ipsec->tx->fc; + mlx5_fc_query(mdev, fc->cnt, &stats->ipsec_tx_pkts, &stats->ipsec_tx_bytes); + mlx5_fc_query(mdev, fc->drop, &stats->ipsec_tx_drop_pkts, + &stats->ipsec_tx_drop_bytes); +} + +int mlx5e_accel_ipsec_fs_add_rule(struct mlx5e_ipsec_sa_entry *sa_entry) +{ + if (sa_entry->attrs.dir == XFRM_DEV_OFFLOAD_OUT) + return tx_add_rule(sa_entry); + + return rx_add_rule(sa_entry); +} + +void mlx5e_accel_ipsec_fs_del_rule(struct mlx5e_ipsec_sa_entry *sa_entry) { struct mlx5e_ipsec_rule *ipsec_rule = &sa_entry->ipsec_rule; struct mlx5_core_dev *mdev = mlx5e_ipsec_sa2dev(sa_entry); mlx5_del_flow_rules(ipsec_rule->rule); - if (sa_entry->attrs.action == MLX5_ACCEL_ESP_ACTION_ENCRYPT) { - tx_ft_put(priv); + if (ipsec_rule->pkt_reformat) + mlx5_packet_reformat_dealloc(mdev, ipsec_rule->pkt_reformat); + + if (sa_entry->attrs.dir == XFRM_DEV_OFFLOAD_OUT) { + tx_ft_put(sa_entry->ipsec); return; } - mlx5_modify_header_dealloc(mdev, ipsec_rule->set_modify_hdr); - rx_ft_put(priv, - sa_entry->attrs.is_ipv6 ? ACCEL_FS_ESP6 : ACCEL_FS_ESP4); + mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr); + rx_ft_put(mdev, sa_entry->ipsec, sa_entry->attrs.family); } -void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec) +int mlx5e_accel_ipsec_fs_add_pol(struct mlx5e_ipsec_pol_entry *pol_entry) { - struct mlx5e_accel_fs_esp_prot *fs_prot; - struct mlx5e_accel_fs_esp *accel_esp; - enum accel_fs_esp_type i; + if (pol_entry->attrs.dir == XFRM_DEV_OFFLOAD_OUT) + return tx_add_policy(pol_entry); - if (!ipsec->rx_fs) - return; + return rx_add_policy(pol_entry); +} + +void mlx5e_accel_ipsec_fs_del_pol(struct mlx5e_ipsec_pol_entry *pol_entry) +{ + struct mlx5e_ipsec_rule *ipsec_rule = &pol_entry->ipsec_rule; + struct mlx5_core_dev *mdev = mlx5e_ipsec_pol2dev(pol_entry); - mutex_destroy(&ipsec->tx_fs->mutex); - WARN_ON(ipsec->tx_fs->refcnt); - kfree(ipsec->tx_fs); + mlx5_del_flow_rules(ipsec_rule->rule); - accel_esp = ipsec->rx_fs; - for (i = 0; i < ACCEL_FS_ESP_NUM_TYPES; i++) { - fs_prot = &accel_esp->fs_prot[i]; - mutex_destroy(&fs_prot->prot_mutex); - WARN_ON(fs_prot->refcnt); + if (pol_entry->attrs.dir == XFRM_DEV_OFFLOAD_IN) { + rx_ft_put(mdev, pol_entry->ipsec, pol_entry->attrs.family); + return; } - kfree(ipsec->rx_fs); + + mlx5_modify_header_dealloc(mdev, ipsec_rule->modify_hdr); + tx_ft_put(pol_entry->ipsec); +} + +void mlx5e_accel_ipsec_fs_cleanup(struct mlx5e_ipsec *ipsec) +{ + if (!ipsec->tx) + return; + + ipsec_fs_destroy_counters(ipsec); + mutex_destroy(&ipsec->tx->ft.mutex); + WARN_ON(ipsec->tx->ft.refcnt); + kfree(ipsec->tx); + + mutex_destroy(&ipsec->rx_ipv4->ft.mutex); + WARN_ON(ipsec->rx_ipv4->ft.refcnt); + kfree(ipsec->rx_ipv4); + + mutex_destroy(&ipsec->rx_ipv6->ft.mutex); + WARN_ON(ipsec->rx_ipv6->ft.refcnt); + kfree(ipsec->rx_ipv6); } int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec) { - struct mlx5e_accel_fs_esp_prot *fs_prot; - struct mlx5e_accel_fs_esp *accel_esp; struct mlx5_flow_namespace *ns; - enum accel_fs_esp_type i; int err = -ENOMEM; ns = mlx5_get_flow_namespace(ipsec->mdev, @@ -581,26 +1032,34 @@ int mlx5e_accel_ipsec_fs_init(struct mlx5e_ipsec *ipsec) if (!ns) return -EOPNOTSUPP; - ipsec->tx_fs = kzalloc(sizeof(*ipsec->tx_fs), GFP_KERNEL); - if (!ipsec->tx_fs) + ipsec->tx = kzalloc(sizeof(*ipsec->tx), GFP_KERNEL); + if (!ipsec->tx) return -ENOMEM; - ipsec->rx_fs = kzalloc(sizeof(*ipsec->rx_fs), GFP_KERNEL); - if (!ipsec->rx_fs) - goto err_rx; + ipsec->rx_ipv4 = kzalloc(sizeof(*ipsec->rx_ipv4), GFP_KERNEL); + if (!ipsec->rx_ipv4) + goto err_rx_ipv4; - mutex_init(&ipsec->tx_fs->mutex); - ipsec->tx_fs->ns = ns; + ipsec->rx_ipv6 = kzalloc(sizeof(*ipsec->rx_ipv6), GFP_KERNEL); + if (!ipsec->rx_ipv6) + goto err_rx_ipv6; - accel_esp = ipsec->rx_fs; - for (i = 0; i < ACCEL_FS_ESP_NUM_TYPES; i++) { - fs_prot = &accel_esp->fs_prot[i]; - mutex_init(&fs_prot->prot_mutex); - } + err = ipsec_fs_init_counters(ipsec); + if (err) + goto err_counters; + + mutex_init(&ipsec->tx->ft.mutex); + mutex_init(&ipsec->rx_ipv4->ft.mutex); + mutex_init(&ipsec->rx_ipv6->ft.mutex); + ipsec->tx->ns = ns; return 0; -err_rx: - kfree(ipsec->tx_fs); +err_counters: + kfree(ipsec->rx_ipv6); +err_rx_ipv6: + kfree(ipsec->rx_ipv4); +err_rx_ipv4: + kfree(ipsec->tx); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c index 792724ce7336..8e3614218fc4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_offload.c @@ -2,9 +2,14 @@ /* Copyright (c) 2017, Mellanox Technologies inc. All rights reserved. */ #include "mlx5_core.h" +#include "en.h" #include "ipsec.h" #include "lib/mlx5.h" +enum { + MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET, +}; + u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev) { u32 caps = 0; @@ -31,6 +36,12 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev) MLX5_CAP_ETH(mdev, insert_trailer) && MLX5_CAP_ETH(mdev, swp)) caps |= MLX5_IPSEC_CAP_CRYPTO; + if (MLX5_CAP_IPSEC(mdev, ipsec_full_offload) && + MLX5_CAP_FLOWTABLE_NIC_TX(mdev, reformat_add_esp_trasport) && + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, reformat_del_esp_trasport) && + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, decap)) + caps |= MLX5_IPSEC_CAP_PACKET_OFFLOAD; + if (!caps) return 0; @@ -46,6 +57,52 @@ u32 mlx5_ipsec_device_caps(struct mlx5_core_dev *mdev) } EXPORT_SYMBOL_GPL(mlx5_ipsec_device_caps); +static void mlx5e_ipsec_packet_setup(void *obj, u32 pdn, + struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + void *aso_ctx; + + aso_ctx = MLX5_ADDR_OF(ipsec_obj, obj, ipsec_aso); + if (attrs->esn_trigger) { + MLX5_SET(ipsec_aso, aso_ctx, esn_event_arm, 1); + + if (attrs->dir == XFRM_DEV_OFFLOAD_IN) { + MLX5_SET(ipsec_aso, aso_ctx, window_sz, + attrs->replay_window / 64); + MLX5_SET(ipsec_aso, aso_ctx, mode, + MLX5_IPSEC_ASO_REPLAY_PROTECTION); + } + } + + /* ASO context */ + MLX5_SET(ipsec_obj, obj, ipsec_aso_access_pd, pdn); + MLX5_SET(ipsec_obj, obj, full_offload, 1); + MLX5_SET(ipsec_aso, aso_ctx, valid, 1); + /* MLX5_IPSEC_ASO_REG_C_4_5 is type C register that is used + * in flow steering to perform matching against. Please be + * aware that this register was chosen arbitrary and can't + * be used in other places as long as IPsec packet offload + * active. + */ + MLX5_SET(ipsec_obj, obj, aso_return_reg, MLX5_IPSEC_ASO_REG_C_4_5); + if (attrs->dir == XFRM_DEV_OFFLOAD_OUT) + MLX5_SET(ipsec_aso, aso_ctx, mode, MLX5_IPSEC_ASO_INC_SN); + + if (attrs->hard_packet_limit != XFRM_INF) { + MLX5_SET(ipsec_aso, aso_ctx, remove_flow_pkt_cnt, + lower_32_bits(attrs->hard_packet_limit)); + MLX5_SET(ipsec_aso, aso_ctx, hard_lft_arm, 1); + MLX5_SET(ipsec_aso, aso_ctx, remove_flow_enable, 1); + } + + if (attrs->soft_packet_limit != XFRM_INF) { + MLX5_SET(ipsec_aso, aso_ctx, remove_flow_soft_lft, + lower_32_bits(attrs->soft_packet_limit)); + + MLX5_SET(ipsec_aso, aso_ctx, soft_lft_arm, 1); + } +} + static int mlx5_create_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry) { struct mlx5_accel_esp_xfrm_attrs *attrs = &sa_entry->attrs; @@ -54,6 +111,7 @@ static int mlx5_create_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry) u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)]; u32 in[MLX5_ST_SZ_DW(create_ipsec_obj_in)] = {}; void *obj, *salt_p, *salt_iv_p; + struct mlx5e_hw_objs *res; int err; obj = MLX5_ADDR_OF(create_ipsec_obj_in, in, ipsec_object); @@ -66,11 +124,10 @@ static int mlx5_create_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry) salt_iv_p = MLX5_ADDR_OF(ipsec_obj, obj, implicit_iv); memcpy(salt_iv_p, &aes_gcm->seq_iv, sizeof(aes_gcm->seq_iv)); /* esn */ - if (attrs->flags & MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED) { + if (attrs->esn_trigger) { MLX5_SET(ipsec_obj, obj, esn_en, 1); MLX5_SET(ipsec_obj, obj, esn_msb, attrs->esn); - if (attrs->flags & MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP) - MLX5_SET(ipsec_obj, obj, esn_overlap, 1); + MLX5_SET(ipsec_obj, obj, esn_overlap, attrs->esn_overlap); } MLX5_SET(ipsec_obj, obj, dekn, sa_entry->enc_key_id); @@ -81,6 +138,10 @@ static int mlx5_create_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry) MLX5_SET(general_obj_in_cmd_hdr, in, obj_type, MLX5_GENERAL_OBJECT_TYPES_IPSEC); + res = &mdev->mlx5e_res.hw_objs; + if (attrs->type == XFRM_DEV_OFFLOAD_PACKET) + mlx5e_ipsec_packet_setup(obj, res->pdn, attrs); + err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out)); if (!err) sa_entry->ipsec_obj_id = @@ -152,7 +213,7 @@ static int mlx5_modify_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry, void *obj; int err; - if (!(attrs->flags & MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED)) + if (!attrs->esn_trigger) return 0; general_obj_types = MLX5_CAP_GEN_64(mdev, general_obj_types); @@ -183,8 +244,7 @@ static int mlx5_modify_ipsec_obj(struct mlx5e_ipsec_sa_entry *sa_entry, MLX5_MODIFY_IPSEC_BITMASK_ESN_OVERLAP | MLX5_MODIFY_IPSEC_BITMASK_ESN_MSB); MLX5_SET(ipsec_obj, obj, esn_msb, attrs->esn); - if (attrs->flags & MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP) - MLX5_SET(ipsec_obj, obj, esn_overlap, 1); + MLX5_SET(ipsec_obj, obj, esn_overlap, attrs->esn_overlap); /* general object fields set */ MLX5_SET(general_obj_in_cmd_hdr, in, opcode, MLX5_CMD_OP_MODIFY_GENERAL_OBJECT); @@ -203,3 +263,234 @@ void mlx5_accel_esp_modify_xfrm(struct mlx5e_ipsec_sa_entry *sa_entry, memcpy(&sa_entry->attrs, attrs, sizeof(sa_entry->attrs)); } + +static void +mlx5e_ipsec_aso_update_esn(struct mlx5e_ipsec_sa_entry *sa_entry, + const struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + struct mlx5_wqe_aso_ctrl_seg data = {}; + + data.data_mask_mode = MLX5_ASO_DATA_MASK_MODE_BITWISE_64BIT << 6; + data.condition_1_0_operand = MLX5_ASO_ALWAYS_TRUE | MLX5_ASO_ALWAYS_TRUE + << 4; + data.data_offset_condition_operand = MLX5_IPSEC_ASO_REMOVE_FLOW_PKT_CNT_OFFSET; + data.bitwise_data = cpu_to_be64(BIT_ULL(54)); + data.data_mask = data.bitwise_data; + + mlx5e_ipsec_aso_query(sa_entry, &data); +} + +static void mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry, + u32 mode_param) +{ + struct mlx5_accel_esp_xfrm_attrs attrs = {}; + + if (mode_param < MLX5E_IPSEC_ESN_SCOPE_MID) { + sa_entry->esn_state.esn++; + sa_entry->esn_state.overlap = 0; + } else { + sa_entry->esn_state.overlap = 1; + } + + mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &attrs); + mlx5_accel_esp_modify_xfrm(sa_entry, &attrs); + mlx5e_ipsec_aso_update_esn(sa_entry, &attrs); +} + +static void mlx5e_ipsec_handle_event(struct work_struct *_work) +{ + struct mlx5e_ipsec_work *work = + container_of(_work, struct mlx5e_ipsec_work, work); + struct mlx5_accel_esp_xfrm_attrs *attrs; + struct mlx5e_ipsec_sa_entry *sa_entry; + struct mlx5e_ipsec_aso *aso; + struct mlx5e_ipsec *ipsec; + int ret; + + sa_entry = xa_load(&work->ipsec->sadb, work->id); + if (!sa_entry) + goto out; + + ipsec = sa_entry->ipsec; + aso = ipsec->aso; + attrs = &sa_entry->attrs; + + spin_lock(&sa_entry->x->lock); + ret = mlx5e_ipsec_aso_query(sa_entry, NULL); + if (ret) + goto unlock; + + aso->use_cache = true; + if (attrs->esn_trigger && + !MLX5_GET(ipsec_aso, aso->ctx, esn_event_arm)) { + u32 mode_param = MLX5_GET(ipsec_aso, aso->ctx, mode_parameter); + + mlx5e_ipsec_update_esn_state(sa_entry, mode_param); + } + + if (attrs->soft_packet_limit != XFRM_INF) + if (!MLX5_GET(ipsec_aso, aso->ctx, soft_lft_arm) || + !MLX5_GET(ipsec_aso, aso->ctx, hard_lft_arm) || + !MLX5_GET(ipsec_aso, aso->ctx, remove_flow_enable)) + xfrm_state_check_expire(sa_entry->x); + aso->use_cache = false; + +unlock: + spin_unlock(&sa_entry->x->lock); +out: + kfree(work); +} + +static int mlx5e_ipsec_event(struct notifier_block *nb, unsigned long event, + void *data) +{ + struct mlx5e_ipsec *ipsec = container_of(nb, struct mlx5e_ipsec, nb); + struct mlx5_eqe_obj_change *object; + struct mlx5e_ipsec_work *work; + struct mlx5_eqe *eqe = data; + u16 type; + + if (event != MLX5_EVENT_TYPE_OBJECT_CHANGE) + return NOTIFY_DONE; + + object = &eqe->data.obj_change; + type = be16_to_cpu(object->obj_type); + + if (type != MLX5_GENERAL_OBJECT_TYPES_IPSEC) + return NOTIFY_DONE; + + work = kmalloc(sizeof(*work), GFP_ATOMIC); + if (!work) + return NOTIFY_DONE; + + INIT_WORK(&work->work, mlx5e_ipsec_handle_event); + work->ipsec = ipsec; + work->id = be32_to_cpu(object->obj_id); + + queue_work(ipsec->wq, &work->work); + return NOTIFY_OK; +} + +int mlx5e_ipsec_aso_init(struct mlx5e_ipsec *ipsec) +{ + struct mlx5_core_dev *mdev = ipsec->mdev; + struct mlx5e_ipsec_aso *aso; + struct mlx5e_hw_objs *res; + struct device *pdev; + int err; + + aso = kzalloc(sizeof(*ipsec->aso), GFP_KERNEL); + if (!aso) + return -ENOMEM; + + res = &mdev->mlx5e_res.hw_objs; + + pdev = mlx5_core_dma_dev(mdev); + aso->dma_addr = dma_map_single(pdev, aso->ctx, sizeof(aso->ctx), + DMA_BIDIRECTIONAL); + err = dma_mapping_error(pdev, aso->dma_addr); + if (err) + goto err_dma; + + aso->aso = mlx5_aso_create(mdev, res->pdn); + if (IS_ERR(aso->aso)) { + err = PTR_ERR(aso->aso); + goto err_aso_create; + } + + ipsec->nb.notifier_call = mlx5e_ipsec_event; + mlx5_notifier_register(mdev, &ipsec->nb); + + ipsec->aso = aso; + return 0; + +err_aso_create: + dma_unmap_single(pdev, aso->dma_addr, sizeof(aso->ctx), + DMA_BIDIRECTIONAL); +err_dma: + kfree(aso); + return err; +} + +void mlx5e_ipsec_aso_cleanup(struct mlx5e_ipsec *ipsec) +{ + struct mlx5_core_dev *mdev = ipsec->mdev; + struct mlx5e_ipsec_aso *aso; + struct device *pdev; + + aso = ipsec->aso; + pdev = mlx5_core_dma_dev(mdev); + + mlx5_notifier_unregister(mdev, &ipsec->nb); + mlx5_aso_destroy(aso->aso); + dma_unmap_single(pdev, aso->dma_addr, sizeof(aso->ctx), + DMA_BIDIRECTIONAL); + kfree(aso); +} + +static void mlx5e_ipsec_aso_copy(struct mlx5_wqe_aso_ctrl_seg *ctrl, + struct mlx5_wqe_aso_ctrl_seg *data) +{ + if (!data) + return; + + ctrl->data_mask_mode = data->data_mask_mode; + ctrl->condition_1_0_operand = data->condition_1_0_operand; + ctrl->condition_1_0_offset = data->condition_1_0_offset; + ctrl->data_offset_condition_operand = data->data_offset_condition_operand; + ctrl->condition_0_data = data->condition_0_data; + ctrl->condition_0_mask = data->condition_0_mask; + ctrl->condition_1_data = data->condition_1_data; + ctrl->condition_1_mask = data->condition_1_mask; + ctrl->bitwise_data = data->bitwise_data; + ctrl->data_mask = data->data_mask; +} + +int mlx5e_ipsec_aso_query(struct mlx5e_ipsec_sa_entry *sa_entry, + struct mlx5_wqe_aso_ctrl_seg *data) +{ + struct mlx5e_ipsec *ipsec = sa_entry->ipsec; + struct mlx5e_ipsec_aso *aso = ipsec->aso; + struct mlx5_core_dev *mdev = ipsec->mdev; + struct mlx5_wqe_aso_ctrl_seg *ctrl; + struct mlx5e_hw_objs *res; + struct mlx5_aso_wqe *wqe; + u8 ds_cnt; + + lockdep_assert_held(&sa_entry->x->lock); + if (aso->use_cache) + return 0; + + res = &mdev->mlx5e_res.hw_objs; + + memset(aso->ctx, 0, sizeof(aso->ctx)); + wqe = mlx5_aso_get_wqe(aso->aso); + ds_cnt = DIV_ROUND_UP(sizeof(*wqe), MLX5_SEND_WQE_DS); + mlx5_aso_build_wqe(aso->aso, ds_cnt, wqe, sa_entry->ipsec_obj_id, + MLX5_ACCESS_ASO_OPC_MOD_IPSEC); + + ctrl = &wqe->aso_ctrl; + ctrl->va_l = + cpu_to_be32(lower_32_bits(aso->dma_addr) | ASO_CTRL_READ_EN); + ctrl->va_h = cpu_to_be32(upper_32_bits(aso->dma_addr)); + ctrl->l_key = cpu_to_be32(res->mkey); + mlx5e_ipsec_aso_copy(ctrl, data); + + mlx5_aso_post_wqe(aso->aso, false, &wqe->ctrl); + return mlx5_aso_poll_cq(aso->aso, false); +} + +void mlx5e_ipsec_aso_update_curlft(struct mlx5e_ipsec_sa_entry *sa_entry, + u64 *packets) +{ + struct mlx5e_ipsec *ipsec = sa_entry->ipsec; + struct mlx5e_ipsec_aso *aso = ipsec->aso; + u64 hard_cnt; + + hard_cnt = MLX5_GET(ipsec_aso, aso->ctx, remove_flow_pkt_cnt); + /* HW decresases the limit till it reaches zero to fire an avent. + * We need to fix the calculations, so the returned count is a total + * number of passed packets and not how much left. + */ + *packets = sa_entry->attrs.hard_packet_limit - hard_cnt; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c index 6859f1c1a831..eab5bc718771 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c @@ -312,27 +312,31 @@ void mlx5e_ipsec_offload_handle_rx_skb(struct net_device *netdev, struct mlx5_cqe64 *cqe) { u32 ipsec_meta_data = be32_to_cpu(cqe->ft_metadata); - struct mlx5e_priv *priv; + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_ipsec *ipsec = priv->ipsec; + struct mlx5e_ipsec_sa_entry *sa_entry; struct xfrm_offload *xo; - struct xfrm_state *xs; struct sec_path *sp; u32 sa_handle; sa_handle = MLX5_IPSEC_METADATA_HANDLE(ipsec_meta_data); - priv = netdev_priv(netdev); sp = secpath_set(skb); if (unlikely(!sp)) { - atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_sp_alloc); + atomic64_inc(&ipsec->sw_stats.ipsec_rx_drop_sp_alloc); return; } - xs = mlx5e_ipsec_sadb_rx_lookup(priv->ipsec, sa_handle); - if (unlikely(!xs)) { - atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_sadb_miss); + rcu_read_lock(); + sa_entry = xa_load(&ipsec->sadb, sa_handle); + if (unlikely(!sa_entry)) { + rcu_read_unlock(); + atomic64_inc(&ipsec->sw_stats.ipsec_rx_drop_sadb_miss); return; } + xfrm_state_hold(sa_entry->x); + rcu_read_unlock(); - sp->xvec[sp->len++] = xs; + sp->xvec[sp->len++] = sa_entry->x; sp->olen++; xo = xfrm_offload(skb); @@ -349,6 +353,6 @@ void mlx5e_ipsec_offload_handle_rx_skb(struct net_device *netdev, xo->status = CRYPTO_INVALID_PACKET_SYNTAX; break; default: - atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_syndrome); + atomic64_inc(&ipsec->sw_stats.ipsec_rx_drop_syndrome); } } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c index 9de84821dafb..e0e36a09721c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c @@ -37,6 +37,17 @@ #include "en.h" #include "ipsec.h" +static const struct counter_desc mlx5e_ipsec_hw_stats_desc[] = { + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_rx_pkts) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_rx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_rx_drop_pkts) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_rx_drop_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_tx_pkts) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_tx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_tx_drop_pkts) }, + { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_hw_stats, ipsec_tx_drop_bytes) }, +}; + static const struct counter_desc mlx5e_ipsec_sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_sw_stats, ipsec_rx_drop_sp_alloc) }, { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_sw_stats, ipsec_rx_drop_sadb_miss) }, @@ -50,8 +61,48 @@ static const struct counter_desc mlx5e_ipsec_sw_stats_desc[] = { #define MLX5E_READ_CTR_ATOMIC64(ptr, dsc, i) \ atomic64_read((atomic64_t *)((char *)(ptr) + (dsc)[i].offset)) +#define NUM_IPSEC_HW_COUNTERS ARRAY_SIZE(mlx5e_ipsec_hw_stats_desc) #define NUM_IPSEC_SW_COUNTERS ARRAY_SIZE(mlx5e_ipsec_sw_stats_desc) +static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(ipsec_hw) +{ + if (!priv->ipsec) + return 0; + + return NUM_IPSEC_HW_COUNTERS; +} + +static inline MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(ipsec_hw) {} + +static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(ipsec_hw) +{ + unsigned int i; + + if (!priv->ipsec) + return idx; + + for (i = 0; i < NUM_IPSEC_HW_COUNTERS; i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + mlx5e_ipsec_hw_stats_desc[i].format); + + return idx; +} + +static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(ipsec_hw) +{ + int i; + + if (!priv->ipsec) + return idx; + + mlx5e_accel_ipsec_fs_read_stats(priv, &priv->ipsec->hw_stats); + for (i = 0; i < NUM_IPSEC_HW_COUNTERS; i++) + data[idx++] = MLX5E_READ_CTR_ATOMIC64(&priv->ipsec->hw_stats, + mlx5e_ipsec_hw_stats_desc, i); + + return idx; +} + static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(ipsec_sw) { return priv->ipsec ? NUM_IPSEC_SW_COUNTERS : 0; @@ -81,4 +132,5 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(ipsec_sw) return idx; } +MLX5E_DEFINE_STATS_GRP(ipsec_hw, 0); MLX5E_DEFINE_STATS_GRP(ipsec_sw, 0); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c index 2e0335246967..78072bf93f3f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c @@ -125,10 +125,8 @@ mlx5e_get_ktls_tx_priv_ctx(struct tls_context *tls_ctx) /* struct for callback API management */ struct mlx5e_async_ctx { struct mlx5_async_work context; - struct mlx5_async_ctx async_ctx; - struct work_struct work; + struct mlx5_async_ctx *async_ctx; struct mlx5e_ktls_offload_context_tx *priv_tx; - struct completion complete; int err; union { u32 out_create[MLX5_ST_SZ_DW(create_tis_out)]; @@ -136,34 +134,33 @@ struct mlx5e_async_ctx { }; }; -static struct mlx5e_async_ctx *mlx5e_bulk_async_init(struct mlx5_core_dev *mdev, int n) +struct mlx5e_bulk_async_ctx { + struct mlx5_async_ctx async_ctx; + DECLARE_FLEX_ARRAY(struct mlx5e_async_ctx, arr); +}; + +static struct mlx5e_bulk_async_ctx *mlx5e_bulk_async_init(struct mlx5_core_dev *mdev, int n) { - struct mlx5e_async_ctx *bulk_async; + struct mlx5e_bulk_async_ctx *bulk_async; + int sz; int i; - bulk_async = kvcalloc(n, sizeof(struct mlx5e_async_ctx), GFP_KERNEL); + sz = struct_size(bulk_async, arr, n); + bulk_async = kvzalloc(sz, GFP_KERNEL); if (!bulk_async) return NULL; - for (i = 0; i < n; i++) { - struct mlx5e_async_ctx *async = &bulk_async[i]; + mlx5_cmd_init_async_ctx(mdev, &bulk_async->async_ctx); - mlx5_cmd_init_async_ctx(mdev, &async->async_ctx); - init_completion(&async->complete); - } + for (i = 0; i < n; i++) + bulk_async->arr[i].async_ctx = &bulk_async->async_ctx; return bulk_async; } -static void mlx5e_bulk_async_cleanup(struct mlx5e_async_ctx *bulk_async, int n) +static void mlx5e_bulk_async_cleanup(struct mlx5e_bulk_async_ctx *bulk_async) { - int i; - - for (i = 0; i < n; i++) { - struct mlx5e_async_ctx *async = &bulk_async[i]; - - mlx5_cmd_cleanup_async_ctx(&async->async_ctx); - } + mlx5_cmd_cleanup_async_ctx(&bulk_async->async_ctx); kvfree(bulk_async); } @@ -176,12 +173,10 @@ static void create_tis_callback(int status, struct mlx5_async_work *context) if (status) { async->err = status; priv_tx->create_err = 1; - goto out; + return; } priv_tx->tisn = MLX5_GET(create_tis_out, async->out_create, tisn); -out: - complete(&async->complete); } static void destroy_tis_callback(int status, struct mlx5_async_work *context) @@ -190,7 +185,6 @@ static void destroy_tis_callback(int status, struct mlx5_async_work *context) container_of(context, struct mlx5e_async_ctx, context); struct mlx5e_ktls_offload_context_tx *priv_tx = async->priv_tx; - complete(&async->complete); kfree(priv_tx); } @@ -214,7 +208,7 @@ mlx5e_tls_priv_tx_init(struct mlx5_core_dev *mdev, struct mlx5e_tls_sw_stats *sw goto err_out; } else { async->priv_tx = priv_tx; - err = mlx5e_ktls_create_tis_cb(mdev, &async->async_ctx, + err = mlx5e_ktls_create_tis_cb(mdev, async->async_ctx, async->out_create, sizeof(async->out_create), create_tis_callback, &async->context); if (err) @@ -232,13 +226,12 @@ static void mlx5e_tls_priv_tx_cleanup(struct mlx5e_ktls_offload_context_tx *priv struct mlx5e_async_ctx *async) { if (priv_tx->create_err) { - complete(&async->complete); kfree(priv_tx); return; } async->priv_tx = priv_tx; mlx5e_ktls_destroy_tis_cb(priv_tx->mdev, priv_tx->tisn, - &async->async_ctx, + async->async_ctx, async->out_destroy, sizeof(async->out_destroy), destroy_tis_callback, &async->context); } @@ -247,7 +240,7 @@ static void mlx5e_tls_priv_tx_list_cleanup(struct mlx5_core_dev *mdev, struct list_head *list, int size) { struct mlx5e_ktls_offload_context_tx *obj, *n; - struct mlx5e_async_ctx *bulk_async; + struct mlx5e_bulk_async_ctx *bulk_async; int i; bulk_async = mlx5e_bulk_async_init(mdev, size); @@ -256,16 +249,11 @@ static void mlx5e_tls_priv_tx_list_cleanup(struct mlx5_core_dev *mdev, i = 0; list_for_each_entry_safe(obj, n, list, list_node) { - mlx5e_tls_priv_tx_cleanup(obj, &bulk_async[i]); + mlx5e_tls_priv_tx_cleanup(obj, &bulk_async->arr[i]); i++; } - for (i = 0; i < size; i++) { - struct mlx5e_async_ctx *async = &bulk_async[i]; - - wait_for_completion(&async->complete); - } - mlx5e_bulk_async_cleanup(bulk_async, size); + mlx5e_bulk_async_cleanup(bulk_async); } /* Recycling pool API */ @@ -291,7 +279,7 @@ static void create_work(struct work_struct *work) struct mlx5e_tls_tx_pool *pool = container_of(work, struct mlx5e_tls_tx_pool, create_work); struct mlx5e_ktls_offload_context_tx *obj; - struct mlx5e_async_ctx *bulk_async; + struct mlx5e_bulk_async_ctx *bulk_async; LIST_HEAD(local_list); int i, j, err = 0; @@ -300,7 +288,7 @@ static void create_work(struct work_struct *work) return; for (i = 0; i < MLX5E_TLS_TX_POOL_BULK; i++) { - obj = mlx5e_tls_priv_tx_init(pool->mdev, pool->sw_stats, &bulk_async[i]); + obj = mlx5e_tls_priv_tx_init(pool->mdev, pool->sw_stats, &bulk_async->arr[i]); if (IS_ERR(obj)) { err = PTR_ERR(obj); break; @@ -309,14 +297,13 @@ static void create_work(struct work_struct *work) } for (j = 0; j < i; j++) { - struct mlx5e_async_ctx *async = &bulk_async[j]; + struct mlx5e_async_ctx *async = &bulk_async->arr[j]; - wait_for_completion(&async->complete); if (!err && async->err) err = async->err; } atomic64_add(i, &pool->sw_stats->tx_tls_pool_alloc); - mlx5e_bulk_async_cleanup(bulk_async, MLX5E_TLS_TX_POOL_BULK); + mlx5e_bulk_async_cleanup(bulk_async); if (err) goto err_out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c index f900709639f6..9369a580743e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c @@ -186,7 +186,7 @@ static int mlx5e_macsec_aso_reg_mr(struct mlx5_core_dev *mdev, struct mlx5e_macs return err; } - dma_device = &mdev->pdev->dev; + dma_device = mlx5_core_dma_dev(mdev); dma_addr = dma_map_single(dma_device, umr->ctx, sizeof(umr->ctx), DMA_BIDIRECTIONAL); err = dma_mapping_error(dma_device, dma_addr); if (err) { @@ -1299,12 +1299,12 @@ static void macsec_aso_build_wqe_ctrl_seg(struct mlx5e_macsec_aso *macsec_aso, struct mlx5_wqe_aso_ctrl_seg *aso_ctrl, struct mlx5_aso_ctrl_param *param) { + struct mlx5e_macsec_umr *umr = macsec_aso->umr; + memset(aso_ctrl, 0, sizeof(*aso_ctrl)); - if (macsec_aso->umr->dma_addr) { - aso_ctrl->va_l = cpu_to_be32(macsec_aso->umr->dma_addr | ASO_CTRL_READ_EN); - aso_ctrl->va_h = cpu_to_be32((u64)macsec_aso->umr->dma_addr >> 32); - aso_ctrl->l_key = cpu_to_be32(macsec_aso->umr->mkey); - } + aso_ctrl->va_l = cpu_to_be32(umr->dma_addr | ASO_CTRL_READ_EN); + aso_ctrl->va_h = cpu_to_be32((u64)umr->dma_addr >> 32); + aso_ctrl->l_key = cpu_to_be32(umr->mkey); if (!param) return; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index 0ae1865086ff..bed0c2d043e7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -57,7 +57,6 @@ struct mlx5e_arfs_tables { struct arfs_table arfs_tables[ARFS_NUM_TYPES]; /* Protect aRFS rules list */ spinlock_t arfs_lock; - struct list_head rules; int last_filter_id; struct workqueue_struct *wq; }; @@ -376,7 +375,6 @@ int mlx5e_arfs_create_tables(struct mlx5e_flow_steering *fs, return -ENOMEM; spin_lock_init(&arfs->arfs_lock); - INIT_LIST_HEAD(&arfs->rules); arfs->wq = create_singlethread_workqueue("mlx5e_arfs"); if (!arfs->wq) goto err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 1728e197558d..7708acc9b2ab 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -2445,4 +2445,5 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .get_eth_mac_stats = mlx5e_get_eth_mac_stats, .get_eth_ctrl_stats = mlx5e_get_eth_ctrl_stats, .get_rmon_stats = mlx5e_get_rmon_stats, + .get_link_ext_stats = mlx5e_get_link_ext_stats }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 5e41dfdf79c8..8d36e2de53a9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -208,7 +208,7 @@ static u16 mlx5e_mpwrq_umr_octowords(u32 entries, enum mlx5e_mpwrq_umr_mode umr_ u8 umr_entry_size = mlx5e_mpwrq_umr_entry_size(umr_mode); u32 sz; - sz = ALIGN(entries * umr_entry_size, MLX5_UMR_MTT_ALIGNMENT); + sz = ALIGN(entries * umr_entry_size, MLX5_UMR_FLEX_ALIGNMENT); return sz / MLX5_OCTWORD; } @@ -1206,6 +1206,13 @@ int mlx5e_open_rq(struct mlx5e_params *params, struct mlx5e_rq_param *param, MLX5_CAP_GEN(mdev, mini_cqe_resp_stride_index)) __set_bit(MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, &rq->state); + /* For enhanced CQE compression packet processing. decompress + * session according to the enhanced layout. + */ + if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS) && + MLX5_CAP_GEN(mdev, enhanced_cqe_compression)) + __set_bit(MLX5E_RQ_STATE_MINI_CQE_ENHANCED, &rq->state); + return 0; err_destroy_rq: @@ -1896,6 +1903,7 @@ static int mlx5e_alloc_cq_common(struct mlx5e_priv *priv, struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i); cqe->op_own = 0xf1; + cqe->validity_iteration_count = 0xff; } cq->mdev = mdev; @@ -3062,7 +3070,10 @@ int mlx5e_open_locked(struct net_device *netdev) if (err) goto err_clear_state_opened_flag; - priv->profile->update_rx(priv); + err = priv->profile->update_rx(priv); + if (err) + goto err_close_channels; + mlx5e_selq_apply(&priv->selq); mlx5e_activate_priv_channels(priv); mlx5e_apply_traps(priv, true); @@ -3072,6 +3083,8 @@ int mlx5e_open_locked(struct net_device *netdev) mlx5e_queue_update_stats(priv); return 0; +err_close_channels: + mlx5e_close_channels(&priv->channels); err_clear_state_opened_flag: clear_bit(MLX5E_STATE_OPENED, &priv->state); mlx5e_selq_cancel(&priv->selq); @@ -4898,7 +4911,6 @@ const struct net_device_ops mlx5e_netdev_ops = { .ndo_has_offload_stats = mlx5e_has_offload_stats, .ndo_get_offload_stats = mlx5e_get_offload_stats, #endif - .ndo_get_devlink_port = mlx5e_get_devlink_port, }; static u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout) @@ -5226,10 +5238,6 @@ static int mlx5e_nic_init(struct mlx5_core_dev *mdev, } priv->fs = fs; - err = mlx5e_ipsec_init(priv); - if (err) - mlx5_core_err(mdev, "IPSec initialization failed, %d\n", err); - err = mlx5e_ktls_init(priv); if (err) mlx5_core_err(mdev, "TLS initialization failed, %d\n", err); @@ -5242,7 +5250,6 @@ static void mlx5e_nic_cleanup(struct mlx5e_priv *priv) { mlx5e_health_destroy_reporters(priv); mlx5e_ktls_cleanup(priv); - mlx5e_ipsec_cleanup(priv); mlx5e_fs_cleanup(priv->fs); } @@ -5371,6 +5378,7 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) int err; mlx5e_fs_init_l2_addr(priv->fs, netdev); + mlx5e_ipsec_init(priv); err = mlx5e_macsec_init(priv); if (err) @@ -5434,6 +5442,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) mlx5_lag_remove_netdev(mdev, priv->netdev); mlx5_vxlan_reset_to_default(mdev->vxlan); mlx5e_macsec_cleanup(priv); + mlx5e_ipsec_cleanup(priv); } int mlx5e_update_nic_rx(struct mlx5e_priv *priv) @@ -5940,16 +5949,16 @@ static int mlx5e_probe(struct auxiliary_device *adev, goto err_profile_cleanup; } + SET_NETDEV_DEVLINK_PORT(netdev, mlx5e_devlink_get_dl_port(priv)); err = register_netdev(netdev); if (err) { mlx5_core_err(mdev, "register_netdev failed, %d\n", err); goto err_resume; } - mlx5e_devlink_port_type_eth_set(priv); - mlx5e_dcbnl_init_app(priv); mlx5_uplink_netdev_set(mdev, netdev); + mlx5e_params_print_info(mdev, &priv->channels.params); return 0; err_resume: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 794cd8dfe9c9..75b9e1528fd2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -85,18 +85,25 @@ static const struct counter_desc sw_rep_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_bytes) }, }; -struct vport_stats { - u64 vport_rx_packets; - u64 vport_tx_packets; - u64 vport_rx_bytes; - u64 vport_tx_bytes; -}; - static const struct counter_desc vport_rep_stats_desc[] = { - { MLX5E_DECLARE_STAT(struct vport_stats, vport_rx_packets) }, - { MLX5E_DECLARE_STAT(struct vport_stats, vport_rx_bytes) }, - { MLX5E_DECLARE_STAT(struct vport_stats, vport_tx_packets) }, - { MLX5E_DECLARE_STAT(struct vport_stats, vport_tx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, vport_rx_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, vport_rx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, vport_tx_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, vport_tx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, + rx_vport_rdma_unicast_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, rx_vport_rdma_unicast_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, + tx_vport_rdma_unicast_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, tx_vport_rdma_unicast_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, + rx_vport_rdma_multicast_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, + rx_vport_rdma_multicast_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, + tx_vport_rdma_multicast_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_rep_stats, + tx_vport_rdma_multicast_bytes) }, }; #define NUM_VPORT_REP_SW_COUNTERS ARRAY_SIZE(sw_rep_stats_desc) @@ -161,33 +168,80 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(vport_rep) int i; for (i = 0; i < NUM_VPORT_REP_HW_COUNTERS; i++) - data[idx++] = MLX5E_READ_CTR64_CPU(&priv->stats.vf_vport, + data[idx++] = MLX5E_READ_CTR64_CPU(&priv->stats.rep_stats, vport_rep_stats_desc, i); return idx; } static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(vport_rep) { + struct mlx5e_rep_stats *rep_stats = &priv->stats.rep_stats; + int outlen = MLX5_ST_SZ_BYTES(query_vport_counter_out); struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct mlx5e_rep_priv *rpriv = priv->ppriv; struct mlx5_eswitch_rep *rep = rpriv->rep; - struct rtnl_link_stats64 *vport_stats; - struct ifla_vf_stats vf_stats; + u32 *out; int err; - err = mlx5_eswitch_get_vport_stats(esw, rep->vport, &vf_stats); + out = kvzalloc(outlen, GFP_KERNEL); + if (!out) + return; + + err = mlx5_core_query_vport_counter(esw->dev, 1, rep->vport - 1, 0, out); if (err) { netdev_warn(priv->netdev, "vport %d error %d reading stats\n", rep->vport, err); return; } - vport_stats = &priv->stats.vf_vport; + #define MLX5_GET_CTR(p, x) \ + MLX5_GET64(query_vport_counter_out, p, x) /* flip tx/rx as we are reporting the counters for the switch vport */ - vport_stats->rx_packets = vf_stats.tx_packets; - vport_stats->rx_bytes = vf_stats.tx_bytes; - vport_stats->tx_packets = vf_stats.rx_packets; - vport_stats->tx_bytes = vf_stats.rx_bytes; + rep_stats->vport_rx_packets = + MLX5_GET_CTR(out, transmitted_ib_unicast.packets) + + MLX5_GET_CTR(out, transmitted_eth_unicast.packets) + + MLX5_GET_CTR(out, transmitted_ib_multicast.packets) + + MLX5_GET_CTR(out, transmitted_eth_multicast.packets) + + MLX5_GET_CTR(out, transmitted_eth_broadcast.packets); + + rep_stats->vport_tx_packets = + MLX5_GET_CTR(out, received_ib_unicast.packets) + + MLX5_GET_CTR(out, received_eth_unicast.packets) + + MLX5_GET_CTR(out, received_ib_multicast.packets) + + MLX5_GET_CTR(out, received_eth_multicast.packets) + + MLX5_GET_CTR(out, received_eth_broadcast.packets); + + rep_stats->vport_rx_bytes = + MLX5_GET_CTR(out, transmitted_ib_unicast.octets) + + MLX5_GET_CTR(out, transmitted_eth_unicast.octets) + + MLX5_GET_CTR(out, transmitted_ib_multicast.octets) + + MLX5_GET_CTR(out, transmitted_eth_broadcast.octets); + + rep_stats->vport_tx_bytes = + MLX5_GET_CTR(out, received_ib_unicast.octets) + + MLX5_GET_CTR(out, received_eth_unicast.octets) + + MLX5_GET_CTR(out, received_ib_multicast.octets) + + MLX5_GET_CTR(out, received_eth_multicast.octets) + + MLX5_GET_CTR(out, received_eth_broadcast.octets); + + rep_stats->rx_vport_rdma_unicast_packets = + MLX5_GET_CTR(out, transmitted_ib_unicast.packets); + rep_stats->tx_vport_rdma_unicast_packets = + MLX5_GET_CTR(out, received_ib_unicast.packets); + rep_stats->rx_vport_rdma_unicast_bytes = + MLX5_GET_CTR(out, transmitted_ib_unicast.octets); + rep_stats->tx_vport_rdma_unicast_bytes = + MLX5_GET_CTR(out, received_ib_unicast.octets); + rep_stats->rx_vport_rdma_multicast_packets = + MLX5_GET_CTR(out, transmitted_ib_multicast.packets); + rep_stats->tx_vport_rdma_multicast_packets = + MLX5_GET_CTR(out, received_ib_multicast.packets); + rep_stats->rx_vport_rdma_multicast_bytes = + MLX5_GET_CTR(out, transmitted_ib_multicast.octets); + rep_stats->tx_vport_rdma_multicast_bytes = + MLX5_GET_CTR(out, received_ib_multicast.octets); + + kvfree(out); } static void mlx5e_rep_get_strings(struct net_device *dev, @@ -607,15 +661,6 @@ static int mlx5e_rep_change_mtu(struct net_device *netdev, int new_mtu) return mlx5e_change_mtu(netdev, new_mtu, NULL); } -static struct devlink_port *mlx5e_rep_get_devlink_port(struct net_device *netdev) -{ - struct mlx5e_priv *priv = netdev_priv(netdev); - struct mlx5e_rep_priv *rpriv = priv->ppriv; - struct mlx5_core_dev *dev = priv->mdev; - - return mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); -} - static int mlx5e_rep_change_carrier(struct net_device *dev, bool new_carrier) { struct mlx5e_priv *priv = netdev_priv(dev); @@ -644,7 +689,6 @@ static const struct net_device_ops mlx5e_netdev_ops_rep = { .ndo_stop = mlx5e_rep_close, .ndo_start_xmit = mlx5e_xmit, .ndo_setup_tc = mlx5e_rep_setup_tc, - .ndo_get_devlink_port = mlx5e_rep_get_devlink_port, .ndo_get_stats64 = mlx5e_rep_get_stats, .ndo_has_offload_stats = mlx5e_rep_has_offload_stats, .ndo_get_offload_stats = mlx5e_rep_get_offload_stats, @@ -761,7 +805,6 @@ static int mlx5e_init_ul_rep(struct mlx5_core_dev *mdev, struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); - int err; priv->fs = mlx5e_fs_init(priv->profile, mdev, !test_bit(MLX5E_STATE_DESTROYING, &priv->state)); @@ -770,10 +813,6 @@ static int mlx5e_init_ul_rep(struct mlx5_core_dev *mdev, return -ENOMEM; } - err = mlx5e_ipsec_init(priv); - if (err) - mlx5_core_err(mdev, "Uplink rep IPsec initialization failed, %d\n", err); - mlx5e_vxlan_set_netdev_info(priv); mlx5e_build_rep_params(netdev); mlx5e_timestamp_init(priv); @@ -783,7 +822,6 @@ static int mlx5e_init_ul_rep(struct mlx5_core_dev *mdev, static void mlx5e_cleanup_rep(struct mlx5e_priv *priv) { mlx5e_fs_cleanup(priv->fs); - mlx5e_ipsec_cleanup(priv); } static int mlx5e_create_rep_ttc_table(struct mlx5e_priv *priv) @@ -1122,6 +1160,8 @@ static void mlx5e_uplink_rep_enable(struct mlx5e_priv *priv) struct mlx5_core_dev *mdev = priv->mdev; u16 max_mtu; + mlx5e_ipsec_init(priv); + netdev->min_mtu = ETH_MIN_MTU; mlx5_query_port_max_mtu(priv->mdev, &max_mtu, 1); netdev->max_mtu = MLX5E_HW2SW_MTU(&priv->channels.params, max_mtu); @@ -1168,6 +1208,8 @@ static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv) mlx5e_rep_tc_disable(priv); mlx5_lag_remove_netdev(mdev, priv->netdev); mlx5_vxlan_reset_to_default(mdev->vxlan); + + mlx5e_ipsec_cleanup(priv); } static MLX5E_DEFINE_STATS_GRP(sw_rep, 0); @@ -1253,37 +1295,20 @@ mlx5e_vport_uplink_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep * { struct mlx5e_priv *priv = netdev_priv(mlx5_uplink_netdev_get(dev)); struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep); - struct devlink_port *dl_port; - int err; rpriv->netdev = priv->netdev; - - err = mlx5e_netdev_change_profile(priv, &mlx5e_uplink_rep_profile, - rpriv); - if (err) - return err; - - dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); - if (dl_port) - devlink_port_type_eth_set(dl_port, rpriv->netdev); - - return 0; + return mlx5e_netdev_change_profile(priv, &mlx5e_uplink_rep_profile, + rpriv); } static void mlx5e_vport_uplink_rep_unload(struct mlx5e_rep_priv *rpriv) { struct net_device *netdev = rpriv->netdev; - struct devlink_port *dl_port; - struct mlx5_core_dev *dev; struct mlx5e_priv *priv; priv = netdev_priv(netdev); - dev = priv->mdev; - dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); - if (dl_port) - devlink_port_type_clear(dl_port); mlx5e_netdev_attach_nic_profile(priv); } @@ -1326,6 +1351,11 @@ mlx5e_vport_vf_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) goto err_cleanup_profile; } + dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, + rpriv->rep->vport); + if (dl_port) + SET_NETDEV_DEVLINK_PORT(netdev, dl_port); + err = register_netdev(netdev); if (err) { netdev_warn(netdev, @@ -1334,9 +1364,6 @@ mlx5e_vport_vf_rep_load(struct mlx5_core_dev *dev, struct mlx5_eswitch_rep *rep) goto err_detach_netdev; } - dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); - if (dl_port) - devlink_port_type_eth_set(dl_port, netdev); return 0; err_detach_netdev: @@ -1382,8 +1409,6 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep) struct mlx5e_rep_priv *rpriv = mlx5e_rep_to_rep_priv(rep); struct net_device *netdev = rpriv->netdev; struct mlx5e_priv *priv = netdev_priv(netdev); - struct mlx5_core_dev *dev = priv->mdev; - struct devlink_port *dl_port; void *ppriv = priv->ppriv; if (rep->vport == MLX5_VPORT_UPLINK) { @@ -1391,9 +1416,6 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep) goto free_ppriv; } - dl_port = mlx5_esw_offloads_devlink_port(dev->priv.eswitch, rpriv->rep->vport); - if (dl_port) - devlink_port_type_clear(dl_port); unregister_netdev(netdev); mlx5e_detach_netdev(priv); priv->profile->cleanup(priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index a61a43fc8d5c..c8820ab22169 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -89,6 +89,25 @@ static inline void mlx5e_read_cqe_slot(struct mlx5_cqwq *wq, memcpy(data, mlx5_cqwq_get_wqe(wq, ci), sizeof(struct mlx5_cqe64)); } +static void mlx5e_read_enhanced_title_slot(struct mlx5e_rq *rq, + struct mlx5_cqe64 *cqe) +{ + struct mlx5e_cq_decomp *cqd = &rq->cqd; + struct mlx5_cqe64 *title = &cqd->title; + + memcpy(title, cqe, sizeof(struct mlx5_cqe64)); + + if (likely(test_bit(MLX5E_RQ_STATE_MINI_CQE_HW_STRIDX, &rq->state))) + return; + + if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) + cqd->wqe_counter = mpwrq_get_cqe_stride_index(title) + + mpwrq_get_cqe_consumed_strides(title); + else + cqd->wqe_counter = + mlx5_wq_cyc_ctr2ix(&rq->wqe.wq, be16_to_cpu(title->wqe_counter) + 1); +} + static inline void mlx5e_read_title_slot(struct mlx5e_rq *rq, struct mlx5_cqwq *wq, u32 cqcc) @@ -175,6 +194,38 @@ static inline void mlx5e_decompress_cqe_no_hash(struct mlx5e_rq *rq, cqd->title.rss_hash_result = 0; } +static u32 mlx5e_decompress_enhanced_cqe(struct mlx5e_rq *rq, + struct mlx5_cqwq *wq, + struct mlx5_cqe64 *cqe, + int budget_rem) +{ + struct mlx5e_cq_decomp *cqd = &rq->cqd; + u32 cqcc, left; + u32 i; + + left = get_cqe_enhanced_num_mini_cqes(cqe); + /* Here we avoid breaking the cqe compression session in the middle + * in case budget is not sufficient to handle all of it. In this case + * we return work_done == budget_rem to give 'busy' napi indication. + */ + if (unlikely(left > budget_rem)) + return budget_rem; + + cqcc = wq->cc; + cqd->mini_arr_idx = 0; + memcpy(cqd->mini_arr, cqe, sizeof(struct mlx5_cqe64)); + for (i = 0; i < left; i++, cqd->mini_arr_idx++, cqcc++) { + mlx5e_decompress_cqe_no_hash(rq, wq, cqcc); + INDIRECT_CALL_3(rq->handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq, + mlx5e_handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq_shampo, + rq, &cqd->title); + } + wq->cc = cqcc; + rq->stats->cqe_compress_pkts += left; + + return left; +} + static inline u32 mlx5e_decompress_cqes_cont(struct mlx5e_rq *rq, struct mlx5_cqwq *wq, int update_owner_only, @@ -220,7 +271,7 @@ static inline u32 mlx5e_decompress_cqes_start(struct mlx5e_rq *rq, rq, &cqd->title); cqd->mini_arr_idx++; - return mlx5e_decompress_cqes_cont(rq, wq, 1, budget_rem) - 1; + return mlx5e_decompress_cqes_cont(rq, wq, 1, budget_rem); } static inline bool mlx5e_rx_cache_put(struct mlx5e_rq *rq, struct page *page) @@ -542,8 +593,8 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq, int headroom, i; headroom = rq->buff.headroom; - new_entries = klm_entries - (shampo->pi & (MLX5_UMR_KLM_ALIGNMENT - 1)); - entries = ALIGN(klm_entries, MLX5_UMR_KLM_ALIGNMENT); + new_entries = klm_entries - (shampo->pi & (MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT - 1)); + entries = ALIGN(klm_entries, MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT); wqe_bbs = MLX5E_KLM_UMR_WQEBBS(entries); pi = mlx5e_icosq_get_next_pi(sq, wqe_bbs); umr_wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi); @@ -552,7 +603,7 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq, for (i = 0; i < entries; i++, index++) { dma_info = &shampo->info[index]; if (i >= klm_entries || (index < shampo->pi && shampo->pi - index < - MLX5_UMR_KLM_ALIGNMENT)) + MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT)) goto update_klm; header_offset = (index & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) << MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE; @@ -617,8 +668,8 @@ static int mlx5e_alloc_rx_hd_mpwqe(struct mlx5e_rq *rq) if (!klm_entries) return 0; - klm_entries += (shampo->pi & (MLX5_UMR_KLM_ALIGNMENT - 1)); - index = ALIGN_DOWN(shampo->pi, MLX5_UMR_KLM_ALIGNMENT); + klm_entries += (shampo->pi & (MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT - 1)); + index = ALIGN_DOWN(shampo->pi, MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT); entries_before = shampo->hd_per_wq - index; if (unlikely(entries_before < klm_entries)) @@ -676,6 +727,17 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) }; } + /* Pad if needed, in case the value set to ucseg->xlt_octowords + * in mlx5e_build_umr_wqe() needed alignment. + */ + if (rq->mpwqe.pages_per_wqe & (MLX5_UMR_MTT_NUM_ENTRIES_ALIGNMENT - 1)) { + int pad = ALIGN(rq->mpwqe.pages_per_wqe, MLX5_UMR_MTT_NUM_ENTRIES_ALIGNMENT) - + rq->mpwqe.pages_per_wqe; + + memset(&umr_wqe->inline_mtts[rq->mpwqe.pages_per_wqe], 0, + sizeof(*umr_wqe->inline_mtts) * pad); + } + bitmap_zero(wi->xdp_xmit_bitmap, rq->mpwqe.pages_per_wqe); wi->consumed_strides = 0; @@ -2211,45 +2273,102 @@ mpwrq_cqe_out: mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index); } -int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget) +static int mlx5e_rx_cq_process_enhanced_cqe_comp(struct mlx5e_rq *rq, + struct mlx5_cqwq *cqwq, + int budget_rem) { - struct mlx5e_rq *rq = container_of(cq, struct mlx5e_rq, cq); - struct mlx5_cqwq *cqwq = &cq->wq; - struct mlx5_cqe64 *cqe; + struct mlx5_cqe64 *cqe, *title_cqe = NULL; + struct mlx5e_cq_decomp *cqd = &rq->cqd; int work_done = 0; - if (unlikely(!test_bit(MLX5E_RQ_STATE_ENABLED, &rq->state))) - return 0; + cqe = mlx5_cqwq_get_cqe_enahnced_comp(cqwq); + if (!cqe) + return work_done; - if (rq->cqd.left) { - work_done += mlx5e_decompress_cqes_cont(rq, cqwq, 0, budget); - if (work_done >= budget) - goto out; + if (cqd->last_cqe_title && + (mlx5_get_cqe_format(cqe) == MLX5_COMPRESSED)) { + rq->stats->cqe_compress_blks++; + cqd->last_cqe_title = false; } - cqe = mlx5_cqwq_get_cqe(cqwq); - if (!cqe) { - if (unlikely(work_done)) - goto out; - return 0; + do { + if (mlx5_get_cqe_format(cqe) == MLX5_COMPRESSED) { + if (title_cqe) { + mlx5e_read_enhanced_title_slot(rq, title_cqe); + title_cqe = NULL; + rq->stats->cqe_compress_blks++; + } + work_done += + mlx5e_decompress_enhanced_cqe(rq, cqwq, cqe, + budget_rem - work_done); + continue; + } + title_cqe = cqe; + mlx5_cqwq_pop(cqwq); + + INDIRECT_CALL_3(rq->handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq, + mlx5e_handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq_shampo, + rq, cqe); + work_done++; + } while (work_done < budget_rem && + (cqe = mlx5_cqwq_get_cqe_enahnced_comp(cqwq))); + + /* last cqe might be title on next poll bulk */ + if (title_cqe) { + mlx5e_read_enhanced_title_slot(rq, title_cqe); + cqd->last_cqe_title = true; } - do { + return work_done; +} + +static int mlx5e_rx_cq_process_basic_cqe_comp(struct mlx5e_rq *rq, + struct mlx5_cqwq *cqwq, + int budget_rem) +{ + struct mlx5_cqe64 *cqe; + int work_done = 0; + + if (rq->cqd.left) + work_done += mlx5e_decompress_cqes_cont(rq, cqwq, 0, budget_rem); + + while (work_done < budget_rem && (cqe = mlx5_cqwq_get_cqe(cqwq))) { if (mlx5_get_cqe_format(cqe) == MLX5_COMPRESSED) { work_done += mlx5e_decompress_cqes_start(rq, cqwq, - budget - work_done); + budget_rem - work_done); continue; } mlx5_cqwq_pop(cqwq); - INDIRECT_CALL_3(rq->handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq, mlx5e_handle_rx_cqe, mlx5e_handle_rx_cqe_mpwrq_shampo, rq, cqe); - } while ((++work_done < budget) && (cqe = mlx5_cqwq_get_cqe(cqwq))); + work_done++; + } + + return work_done; +} + +int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget) +{ + struct mlx5e_rq *rq = container_of(cq, struct mlx5e_rq, cq); + struct mlx5_cqwq *cqwq = &cq->wq; + int work_done; + + if (unlikely(!test_bit(MLX5E_RQ_STATE_ENABLED, &rq->state))) + return 0; + + if (test_bit(MLX5E_RQ_STATE_MINI_CQE_ENHANCED, &rq->state)) + work_done = mlx5e_rx_cq_process_enhanced_cqe_comp(rq, cqwq, + budget); + else + work_done = mlx5e_rx_cq_process_basic_cqe_comp(rq, cqwq, + budget); + + if (work_done == 0) + return 0; -out: if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state) && rq->hw_gro_data->skb) mlx5e_shampo_flush_skb(rq, NULL, false); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index 03c1841970f1..6687b8136e44 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -1241,6 +1241,23 @@ static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(phy) mlx5_core_access_reg(mdev, in, sz, out, sz, MLX5_REG_PPCNT, 0, 0); } +void mlx5e_get_link_ext_stats(struct net_device *dev, + struct ethtool_link_ext_stats *stats) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + u32 out[MLX5_ST_SZ_DW(ppcnt_reg)] = {}; + u32 in[MLX5_ST_SZ_DW(ppcnt_reg)] = {}; + int sz = MLX5_ST_SZ_BYTES(ppcnt_reg); + + MLX5_SET(ppcnt_reg, in, local_port, 1); + MLX5_SET(ppcnt_reg, in, grp, MLX5_PHYSICAL_LAYER_COUNTERS_GROUP); + mlx5_core_access_reg(priv->mdev, in, sz, out, + MLX5_ST_SZ_BYTES(ppcnt_reg), MLX5_REG_PPCNT, 0, 0); + + stats->link_down_events = MLX5_GET(ppcnt_reg, out, + counter_set.phys_layer_cntrs.link_down_events); +} + static int fec_num_lanes(struct mlx5_core_dev *dev) { u32 out[MLX5_ST_SZ_DW(pmlp_reg)] = {}; @@ -2463,6 +2480,7 @@ mlx5e_stats_grp_t mlx5e_nic_stats_grps[] = { &MLX5E_STATS_GRP(per_prio), &MLX5E_STATS_GRP(pme), #ifdef CONFIG_MLX5_EN_IPSEC + &MLX5E_STATS_GRP(ipsec_hw), &MLX5E_STATS_GRP(ipsec_sw), #endif &MLX5E_STATS_GRP(tls), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 9f781085be47..375752d6546d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -126,6 +126,8 @@ void mlx5e_stats_eth_ctrl_get(struct mlx5e_priv *priv, void mlx5e_stats_rmon_get(struct mlx5e_priv *priv, struct ethtool_rmon_stats *rmon, const struct ethtool_rmon_hist_range **ranges); +void mlx5e_get_link_ext_stats(struct net_device *dev, + struct ethtool_link_ext_stats *stats); /* Concrete NIC Stats */ @@ -461,6 +463,21 @@ struct mlx5e_ptp_cq_stats { u64 resync_event; }; +struct mlx5e_rep_stats { + u64 vport_rx_packets; + u64 vport_tx_packets; + u64 vport_rx_bytes; + u64 vport_tx_bytes; + u64 rx_vport_rdma_unicast_packets; + u64 tx_vport_rdma_unicast_packets; + u64 rx_vport_rdma_unicast_bytes; + u64 tx_vport_rdma_unicast_bytes; + u64 rx_vport_rdma_multicast_packets; + u64 tx_vport_rdma_multicast_packets; + u64 rx_vport_rdma_multicast_bytes; + u64 tx_vport_rdma_multicast_bytes; +}; + struct mlx5e_stats { struct mlx5e_sw_stats sw; struct mlx5e_qcounter_stats qcnt; @@ -469,6 +486,7 @@ struct mlx5e_stats { struct mlx5e_pport_stats pport; struct rtnl_link_stats64 vf_vport; struct mlx5e_pcie_stats pcie; + struct mlx5e_rep_stats rep_stats; }; extern mlx5e_stats_grp_t mlx5e_nic_stats_grps[]; @@ -488,6 +506,7 @@ extern MLX5E_DECLARE_STATS_GRP(per_prio); extern MLX5E_DECLARE_STATS_GRP(pme); extern MLX5E_DECLARE_STATS_GRP(channels); extern MLX5E_DECLARE_STATS_GRP(per_port_buff_congest); +extern MLX5E_DECLARE_STATS_GRP(ipsec_hw); extern MLX5E_DECLARE_STATS_GRP(ipsec_sw); extern MLX5E_DECLARE_STATS_GRP(ptp); extern MLX5E_DECLARE_STATS_GRP(macsec_hw); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index bd9936af4582..9af2aa2922f5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -132,6 +132,15 @@ struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[] = { [PACKET_COLOR_TO_REG] = packet_color_to_reg, }; +struct mlx5e_tc_jump_state { + u32 jump_count; + bool jump_target; + struct mlx5_flow_attr *jumping_attr; + + enum flow_action_id last_id; + u32 last_index; +}; + struct mlx5e_tc_table *mlx5e_tc_table_alloc(void) { struct mlx5e_tc_table *tc; @@ -160,6 +169,7 @@ static struct lock_class_key tc_ht_lock_key; static void mlx5e_put_flow_tunnel_id(struct mlx5e_tc_flow *flow); static void free_flow_post_acts(struct mlx5e_tc_flow *flow); +static void mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr); void mlx5e_tc_match_to_reg_match(struct mlx5_flow_spec *spec, @@ -392,8 +402,9 @@ mlx5_tc_rule_delete(struct mlx5e_priv *priv, static bool is_flow_meter_action(struct mlx5_flow_attr *attr) { - return ((attr->action & MLX5_FLOW_CONTEXT_ACTION_EXECUTE_ASO) && - (attr->exe_aso_type == MLX5_EXE_ASO_FLOW_METER)); + return (((attr->action & MLX5_FLOW_CONTEXT_ACTION_EXECUTE_ASO) && + (attr->exe_aso_type == MLX5_EXE_ASO_FLOW_METER)) || + attr->flags & MLX5_ATTR_FLAG_MTU); } static int @@ -404,6 +415,7 @@ mlx5e_tc_add_flow_meter(struct mlx5e_priv *priv, struct mlx5e_post_meter_priv *post_meter; enum mlx5_flow_namespace_type ns_type; struct mlx5e_flow_meter_handle *meter; + enum mlx5e_post_meter_type type; meter = mlx5e_tc_meter_replace(priv->mdev, &attr->meter_attr.params); if (IS_ERR(meter)) { @@ -412,8 +424,11 @@ mlx5e_tc_add_flow_meter(struct mlx5e_priv *priv, } ns_type = mlx5e_tc_meter_get_namespace(meter->flow_meters); - post_meter = mlx5e_post_meter_init(priv, ns_type, post_act, meter->green_counter, - meter->red_counter); + type = meter->params.mtu ? MLX5E_POST_METER_MTU : MLX5E_POST_METER_RATE; + post_meter = mlx5e_post_meter_init(priv, ns_type, post_act, + type, + meter->act_counter, meter->drop_counter, + attr->branch_true, attr->branch_false); if (IS_ERR(post_meter)) { mlx5_core_err(priv->mdev, "Failed to init post meter\n"); goto err_meter_init; @@ -432,9 +447,9 @@ err_meter_init: } static void -mlx5e_tc_del_flow_meter(struct mlx5_flow_attr *attr) +mlx5e_tc_del_flow_meter(struct mlx5_eswitch *esw, struct mlx5_flow_attr *attr) { - mlx5e_post_meter_cleanup(attr->meter_attr.post_meter); + mlx5e_post_meter_cleanup(esw, attr->meter_attr.post_meter); mlx5e_tc_meter_put(attr->meter_attr.meter); } @@ -495,7 +510,7 @@ mlx5e_tc_rule_unoffload(struct mlx5e_priv *priv, mlx5_eswitch_del_offloaded_rule(esw, rule, attr); if (attr->meter_attr.meter) - mlx5e_tc_del_flow_meter(attr); + mlx5e_tc_del_flow_meter(esw, attr); } int @@ -606,6 +621,12 @@ int mlx5e_get_flow_namespace(struct mlx5e_tc_flow *flow) MLX5_FLOW_NAMESPACE_FDB : MLX5_FLOW_NAMESPACE_KERNEL; } +static struct mlx5_core_dev * +get_flow_counter_dev(struct mlx5e_tc_flow *flow) +{ + return mlx5e_is_eswitch_flow(flow) ? flow->attr->esw_attr->counter_dev : flow->priv->mdev; +} + static struct mod_hdr_tbl * get_mod_hdr_table(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow) { @@ -1060,12 +1081,9 @@ static int mlx5e_hairpin_flow_add(struct mlx5e_priv *priv, hash_hairpin_info(peer_id, match_prio)); mutex_unlock(&tc->hairpin_tbl_lock); - params.log_data_size = 16; - params.log_data_size = min_t(u8, params.log_data_size, - MLX5_CAP_GEN(priv->mdev, log_max_hairpin_wq_data_sz)); - params.log_data_size = max_t(u8, params.log_data_size, - MLX5_CAP_GEN(priv->mdev, log_min_hairpin_wq_data_sz)); - + params.log_data_size = clamp_t(u8, 16, + MLX5_CAP_GEN(priv->mdev, log_min_hairpin_wq_data_sz), + MLX5_CAP_GEN(priv->mdev, log_max_hairpin_wq_data_sz)); params.log_num_packets = params.log_data_size - MLX5_MPWRQ_MIN_LOG_STRIDE_SZ(priv->mdev); params.log_num_packets = min_t(u8, params.log_num_packets, @@ -1722,6 +1740,90 @@ clean_encap_dests(struct mlx5e_priv *priv, } static int +verify_attr_actions(u32 actions, struct netlink_ext_ack *extack) +{ + if (!(actions & + (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_DROP))) { + NL_SET_ERR_MSG_MOD(extack, "Rule must have at least one forward/drop action"); + return -EOPNOTSUPP; + } + + if (!(~actions & + (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_DROP))) { + NL_SET_ERR_MSG_MOD(extack, "Rule cannot support forward+drop action"); + return -EOPNOTSUPP; + } + + if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR && + actions & MLX5_FLOW_CONTEXT_ACTION_DROP) { + NL_SET_ERR_MSG_MOD(extack, "Drop with modify header action is not supported"); + return -EOPNOTSUPP; + } + + return 0; +} + +static int +post_process_attr(struct mlx5e_tc_flow *flow, + struct mlx5_flow_attr *attr, + bool is_post_act_attr, + struct netlink_ext_ack *extack) +{ + struct mlx5_eswitch *esw = flow->priv->mdev->priv.eswitch; + bool vf_tun; + int err = 0; + + err = verify_attr_actions(attr->action, extack); + if (err) + goto err_out; + + err = set_encap_dests(flow->priv, flow, attr, extack, &vf_tun); + if (err) + goto err_out; + + if (mlx5e_is_eswitch_flow(flow)) { + err = mlx5_eswitch_add_vlan_action(esw, attr); + if (err) + goto err_out; + } + + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { + if (vf_tun || is_post_act_attr) { + err = mlx5e_tc_add_flow_mod_hdr(flow->priv, flow, attr); + if (err) + goto err_out; + } else { + err = mlx5e_attach_mod_hdr(flow->priv, flow, attr->parse_attr); + if (err) + goto err_out; + } + } + + if (attr->branch_true && + attr->branch_true->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { + err = mlx5e_tc_add_flow_mod_hdr(flow->priv, flow, attr->branch_true); + if (err) + goto err_out; + } + + if (attr->branch_false && + attr->branch_false->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { + err = mlx5e_tc_add_flow_mod_hdr(flow->priv, flow, attr->branch_false); + if (err) + goto err_out; + } + + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { + err = alloc_flow_attr_counter(get_flow_counter_dev(flow), attr); + if (err) + goto err_out; + } + +err_out: + return err; +} + +static int mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) @@ -1731,7 +1833,6 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, struct mlx5_flow_attr *attr = flow->attr; struct mlx5_esw_flow_attr *esw_attr; u32 max_prio, max_chain; - bool vf_tun; int err = 0; parse_attr = attr->parse_attr; @@ -1821,32 +1922,10 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, esw_attr->int_port = int_port; } - err = set_encap_dests(priv, flow, attr, extack, &vf_tun); + err = post_process_attr(flow, attr, false, extack); if (err) goto err_out; - err = mlx5_eswitch_add_vlan_action(esw, attr); - if (err) - goto err_out; - - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { - if (vf_tun) { - err = mlx5e_tc_add_flow_mod_hdr(priv, flow, attr); - if (err) - goto err_out; - } else { - err = mlx5e_attach_mod_hdr(priv, flow, parse_attr); - if (err) - goto err_out; - } - } - - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { - err = alloc_flow_attr_counter(esw_attr->counter_dev, attr); - if (err) - goto err_out; - } - /* we get here if one of the following takes place: * (1) there's no error * (2) there's an encap action and we don't have valid neigh @@ -1882,6 +1961,16 @@ static bool mlx5_flow_has_geneve_opt(struct mlx5e_tc_flow *flow) return !!geneve_tlv_opt_0_data; } +static void free_branch_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr) +{ + if (!attr) + return; + + mlx5_free_flow_attr(flow, attr); + kvfree(attr->parse_attr); + kfree(attr); +} + static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow) { @@ -1937,6 +2026,8 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, mlx5e_detach_decap(priv, flow); free_flow_post_acts(flow); + free_branch_attr(flow, attr->branch_true); + free_branch_attr(flow, attr->branch_false); if (flow->attr->lag.count) mlx5_lag_del_mpesw_rule(esw->dev); @@ -3510,36 +3601,6 @@ actions_match_supported(struct mlx5e_priv *priv, ct_clear = flow->attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR; ct_flow = flow_flag_test(flow, CT) && !ct_clear; - if (!(actions & - (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_DROP))) { - NL_SET_ERR_MSG_MOD(extack, "Rule must have at least one forward/drop action"); - return false; - } - - if (!(~actions & - (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_DROP))) { - NL_SET_ERR_MSG_MOD(extack, "Rule cannot support forward+drop action"); - return false; - } - - if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR && - actions & MLX5_FLOW_CONTEXT_ACTION_DROP) { - NL_SET_ERR_MSG_MOD(extack, "Drop with modify header action is not supported"); - return false; - } - - if (!(~actions & - (MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | MLX5_FLOW_CONTEXT_ACTION_DROP))) { - NL_SET_ERR_MSG_MOD(extack, "Rule cannot support forward+drop action"); - return false; - } - - if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR && - actions & MLX5_FLOW_CONTEXT_ACTION_DROP) { - NL_SET_ERR_MSG_MOD(extack, "Drop with modify header action is not supported"); - return false; - } - if (actions & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR && !modify_header_match_supported(priv, &parse_attr->spec, flow_action, actions, ct_flow, ct_clear, extack)) @@ -3639,15 +3700,12 @@ mlx5e_clone_flow_attr_for_post_act(struct mlx5_flow_attr *attr, attr2->esw_attr->split_count = 0; } + attr2->branch_true = NULL; + attr2->branch_false = NULL; + attr2->jumping_attr = NULL; return attr2; } -static struct mlx5_core_dev * -get_flow_counter_dev(struct mlx5e_tc_flow *flow) -{ - return mlx5e_is_eswitch_flow(flow) ? flow->attr->esw_attr->counter_dev : flow->priv->mdev; -} - struct mlx5_flow_attr * mlx5e_tc_get_encap_attr(struct mlx5e_tc_flow *flow) { @@ -3683,28 +3741,15 @@ mlx5e_tc_unoffload_flow_post_acts(struct mlx5e_tc_flow *flow) static void free_flow_post_acts(struct mlx5e_tc_flow *flow) { - struct mlx5_core_dev *counter_dev = get_flow_counter_dev(flow); - struct mlx5e_post_act *post_act = get_post_action(flow->priv); struct mlx5_flow_attr *attr, *tmp; - bool vf_tun; list_for_each_entry_safe(attr, tmp, &flow->attrs, list) { if (list_is_last(&attr->list, &flow->attrs)) break; - if (attr->post_act_handle) - mlx5e_tc_post_act_del(post_act, attr->post_act_handle); - - clean_encap_dests(flow->priv, flow, attr, &vf_tun); - - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) - mlx5_fc_destroy(counter_dev, attr->counter); - - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { - mlx5e_mod_hdr_dealloc(&attr->parse_attr->mod_hdr_acts); - if (attr->modify_hdr) - mlx5_modify_header_dealloc(flow->priv->mdev, attr->modify_hdr); - } + mlx5_free_flow_attr(flow, attr); + free_branch_attr(flow, attr->branch_true); + free_branch_attr(flow, attr->branch_false); list_del(&attr->list); kvfree(attr->parse_attr); @@ -3757,7 +3802,6 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) struct mlx5e_post_act *post_act = get_post_action(flow->priv); struct mlx5_flow_attr *attr, *next_attr = NULL; struct mlx5e_post_act_handle *handle; - bool vf_tun; int err; /* This is going in reverse order as needed. @@ -3767,7 +3811,9 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) if (!next_attr) { /* Set counter action on last post act rule. */ attr->action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; - } else { + } + + if (next_attr && !(attr->flags & MLX5_ATTR_FLAG_TERMINATING)) { err = mlx5e_tc_act_set_next_post_act(flow, attr, next_attr); if (err) goto out_free; @@ -3779,26 +3825,14 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) if (list_is_last(&attr->list, &flow->attrs)) break; - err = set_encap_dests(flow->priv, flow, attr, extack, &vf_tun); + err = actions_prepare_mod_hdr_actions(flow->priv, flow, attr, extack); if (err) goto out_free; - err = actions_prepare_mod_hdr_actions(flow->priv, flow, attr, extack); + err = post_process_attr(flow, attr, true, extack); if (err) goto out_free; - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { - err = mlx5e_tc_add_flow_mod_hdr(flow->priv, flow, attr); - if (err) - goto out_free; - } - - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { - err = alloc_flow_attr_counter(get_flow_counter_dev(flow), attr); - if (err) - goto out_free; - } - handle = mlx5e_tc_post_act_add(post_act, attr); if (IS_ERR(handle)) { err = PTR_ERR(handle); @@ -3806,6 +3840,13 @@ alloc_flow_post_acts(struct mlx5e_tc_flow *flow, struct netlink_ext_ack *extack) } attr->post_act_handle = handle; + + if (attr->jumping_attr) { + err = mlx5e_tc_act_set_next_post_act(flow, attr->jumping_attr, attr); + if (err) + goto out_free; + } + next_attr = attr; } @@ -3825,12 +3866,145 @@ out_free: } static int +alloc_branch_attr(struct mlx5e_tc_flow *flow, + struct mlx5e_tc_act_branch_ctrl *cond, + struct mlx5_flow_attr **cond_attr, + u32 *jump_count, + struct netlink_ext_ack *extack) +{ + struct mlx5_flow_attr *attr; + int err = 0; + + *cond_attr = mlx5e_clone_flow_attr_for_post_act(flow->attr, + mlx5e_get_flow_namespace(flow)); + if (!(*cond_attr)) + return -ENOMEM; + + attr = *cond_attr; + + switch (cond->act_id) { + case FLOW_ACTION_DROP: + attr->action |= MLX5_FLOW_CONTEXT_ACTION_DROP; + break; + case FLOW_ACTION_ACCEPT: + case FLOW_ACTION_PIPE: + attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + attr->dest_ft = mlx5e_tc_post_act_get_ft(get_post_action(flow->priv)); + break; + case FLOW_ACTION_JUMP: + if (*jump_count) { + NL_SET_ERR_MSG_MOD(extack, "Cannot offload flows with nested jumps"); + err = -EOPNOTSUPP; + goto out_err; + } + *jump_count = cond->extval; + attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + attr->dest_ft = mlx5e_tc_post_act_get_ft(get_post_action(flow->priv)); + break; + default: + err = -EOPNOTSUPP; + goto out_err; + } + + return err; +out_err: + kfree(*cond_attr); + *cond_attr = NULL; + return err; +} + +static void +dec_jump_count(struct flow_action_entry *act, struct mlx5e_tc_act *tc_act, + struct mlx5_flow_attr *attr, struct mlx5e_priv *priv, + struct mlx5e_tc_jump_state *jump_state) +{ + if (!jump_state->jump_count) + return; + + /* Single tc action can instantiate multiple offload actions (e.g. pedit) + * Jump only over a tc action + */ + if (act->id == jump_state->last_id && act->hw_index == jump_state->last_index) + return; + + jump_state->last_id = act->id; + jump_state->last_index = act->hw_index; + + /* nothing to do for intermediate actions */ + if (--jump_state->jump_count > 1) + return; + + if (jump_state->jump_count == 1) { /* last action in the jump action list */ + + /* create a new attribute after this action */ + jump_state->jump_target = true; + + if (tc_act->is_terminating_action) { /* the branch ends here */ + attr->flags |= MLX5_ATTR_FLAG_TERMINATING; + attr->action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; + } else { /* the branch continues executing the rest of the actions */ + struct mlx5e_post_act *post_act; + + attr->action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + post_act = get_post_action(priv); + attr->dest_ft = mlx5e_tc_post_act_get_ft(post_act); + } + } else if (jump_state->jump_count == 0) { /* first attr after the jump action list */ + /* This is the post action for the jumping attribute (either red or green) + * Use the stored jumping_attr to set the post act id on the jumping attribute + */ + attr->jumping_attr = jump_state->jumping_attr; + } +} + +static int +parse_branch_ctrl(struct flow_action_entry *act, struct mlx5e_tc_act *tc_act, + struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr, + struct mlx5e_tc_jump_state *jump_state, + struct netlink_ext_ack *extack) +{ + struct mlx5e_tc_act_branch_ctrl cond_true, cond_false; + u32 jump_count = jump_state->jump_count; + int err; + + if (!tc_act->get_branch_ctrl) + return 0; + + tc_act->get_branch_ctrl(act, &cond_true, &cond_false); + + err = alloc_branch_attr(flow, &cond_true, + &attr->branch_true, &jump_count, extack); + if (err) + goto out_err; + + if (jump_count) + jump_state->jumping_attr = attr->branch_true; + + err = alloc_branch_attr(flow, &cond_false, + &attr->branch_false, &jump_count, extack); + if (err) + goto err_branch_false; + + if (jump_count && !jump_state->jumping_attr) + jump_state->jumping_attr = attr->branch_false; + + jump_state->jump_count = jump_count; + return 0; + +err_branch_false: + free_branch_attr(flow, attr->branch_true); +out_err: + return err; +} + +static int parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, struct flow_action *flow_action) { struct netlink_ext_ack *extack = parse_state->extack; struct mlx5e_tc_flow_action flow_action_reorder; struct mlx5e_tc_flow *flow = parse_state->flow; + struct mlx5e_tc_jump_state jump_state = {}; struct mlx5_flow_attr *attr = flow->attr; enum mlx5_flow_namespace_type ns_type; struct mlx5e_priv *priv = flow->priv; @@ -3850,6 +4024,7 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, list_add(&attr->list, &flow->attrs); flow_action_for_each(i, _act, &flow_action_reorder) { + jump_state.jump_target = false; act = *_act; tc_act = mlx5e_tc_act_get(act->id, ns_type); if (!tc_act) { @@ -3867,12 +4042,19 @@ parse_tc_actions(struct mlx5e_tc_act_parse_state *parse_state, if (err) goto out_free; + dec_jump_count(act, tc_act, attr, priv, &jump_state); + + err = parse_branch_ctrl(act, tc_act, flow, attr, &jump_state, extack); + if (err) + goto out_free; + parse_state->actions |= attr->action; /* Split attr for multi table act if not the last act. */ - if (tc_act->is_multi_table_act && + if (jump_state.jump_target || + (tc_act->is_multi_table_act && tc_act->is_multi_table_act(priv, act, attr) && - i < flow_action_reorder.num_entries - 1) { + i < flow_action_reorder.num_entries - 1)) { err = mlx5e_tc_act_post_parse(parse_state, flow_action, attr, ns_type); if (err) goto out_free; @@ -3954,6 +4136,10 @@ parse_tc_nic_actions(struct mlx5e_priv *priv, if (err) return err; + err = verify_attr_actions(attr->action, extack); + if (err) + return err; + if (!actions_match_supported(priv, flow_action, parse_state->actions, parse_attr, flow, extack)) return -EOPNOTSUPP; @@ -4191,6 +4377,30 @@ mlx5_alloc_flow_attr(enum mlx5_flow_namespace_type type) return attr; } +static void +mlx5_free_flow_attr(struct mlx5e_tc_flow *flow, struct mlx5_flow_attr *attr) +{ + struct mlx5_core_dev *counter_dev = get_flow_counter_dev(flow); + bool vf_tun; + + if (!attr) + return; + + if (attr->post_act_handle) + mlx5e_tc_post_act_del(get_post_action(flow->priv), attr->post_act_handle); + + clean_encap_dests(flow->priv, flow, attr, &vf_tun); + + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) + mlx5_fc_destroy(counter_dev, attr->counter); + + if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) { + mlx5e_mod_hdr_dealloc(&attr->parse_attr->mod_hdr_acts); + if (attr->modify_hdr) + mlx5_modify_header_dealloc(flow->priv->mdev, attr->modify_hdr); + } +} + static int mlx5e_alloc_flow(struct mlx5e_priv *priv, int attr_size, struct flow_cls_offload *f, unsigned long flow_flags, @@ -4733,10 +4943,17 @@ static int apply_police_params(struct mlx5e_priv *priv, u64 rate, return err; } -int mlx5e_policer_validate(const struct flow_action *action, - const struct flow_action_entry *act, - struct netlink_ext_ack *extack) +static int +tc_matchall_police_validate(const struct flow_action *action, + const struct flow_action_entry *act, + struct netlink_ext_ack *extack) { + if (act->police.notexceed.act_id != FLOW_ACTION_CONTINUE) { + NL_SET_ERR_MSG_MOD(extack, + "Offload not supported when conform action is not continue"); + return -EOPNOTSUPP; + } + if (act->police.exceed.act_id != FLOW_ACTION_DROP) { NL_SET_ERR_MSG_MOD(extack, "Offload not supported when exceed action is not drop"); @@ -4787,13 +5004,7 @@ static int scan_tc_matchall_fdb_actions(struct mlx5e_priv *priv, flow_action_for_each(i, act, flow_action) { switch (act->id) { case FLOW_ACTION_POLICE: - if (act->police.notexceed.act_id != FLOW_ACTION_CONTINUE) { - NL_SET_ERR_MSG_MOD(extack, - "Offload not supported when conform action is not continue"); - return -EOPNOTSUPP; - } - - err = mlx5e_policer_validate(flow_action, act, extack); + err = tc_matchall_police_validate(flow_action, act, extack); if (err) return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h index 48241317a535..50af70ef22f3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h @@ -95,10 +95,13 @@ struct mlx5_flow_attr { */ bool count; } lag; + struct mlx5_flow_attr *branch_true; + struct mlx5_flow_attr *branch_false; + struct mlx5_flow_attr *jumping_attr; /* keep this union last */ union { - struct mlx5_esw_flow_attr esw_attr[0]; - struct mlx5_nic_flow_attr nic_attr[0]; + DECLARE_FLEX_ARRAY(struct mlx5_esw_flow_attr, esw_attr); + DECLARE_FLEX_ARRAY(struct mlx5_nic_flow_attr, nic_attr); }; }; @@ -110,6 +113,8 @@ enum { MLX5_ATTR_FLAG_SAMPLE = BIT(4), MLX5_ATTR_FLAG_ACCEPT = BIT(5), MLX5_ATTR_FLAG_CT = BIT(6), + MLX5_ATTR_FLAG_TERMINATING = BIT(7), + MLX5_ATTR_FLAG_MTU = BIT(8), }; /* Returns true if any of the flags that require skipping further TC/NF processing are set. */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index a0242dc15741..8f7580fec193 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -19,6 +19,7 @@ #include "diag/fw_tracer.h" #include "mlx5_irq.h" #include "devlink.h" +#include "en_accel/ipsec.h" enum { MLX5_EQE_OWNER_INIT_VAL = 0x1, @@ -578,6 +579,10 @@ static void gather_async_events_mask(struct mlx5_core_dev *dev, u64 mask[4]) if (MLX5_CAP_MACSEC(dev, log_max_macsec_offload)) async_event_mask |= (1ull << MLX5_EVENT_TYPE_OBJECT_CHANGE); + if (mlx5_ipsec_device_caps(dev) & MLX5_IPSEC_CAP_PACKET_OFFLOAD) + async_event_mask |= + (1ull << MLX5_EVENT_TYPE_OBJECT_CHANGE); + mask[0] = async_event_mask; if (MLX5_CAP_GEN(dev, event_cap)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c index 4fbff7bcc155..b176648d1343 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c @@ -1722,7 +1722,7 @@ void mlx5_esw_bridge_fdb_update_used(struct net_device *dev, u16 vport_num, u16 entry = mlx5_esw_bridge_fdb_lookup(bridge, fdb_info->addr, fdb_info->vid); if (!entry) { esw_debug(br_offloads->esw->dev, - "FDB entry with specified key not found (MAC=%pM,vid=%u,vport=%u)\n", + "FDB update entry with specified key not found (MAC=%pM,vid=%u,vport=%u)\n", fdb_info->addr, fdb_info->vid, vport_num); return; } @@ -1775,9 +1775,9 @@ void mlx5_esw_bridge_fdb_remove(struct net_device *dev, u16 vport_num, u16 esw_o bridge = port->bridge; entry = mlx5_esw_bridge_fdb_lookup(bridge, fdb_info->addr, fdb_info->vid); if (!entry) { - esw_warn(esw->dev, - "FDB entry with specified key not found (MAC=%pM,vid=%u,vport=%u)\n", - fdb_info->addr, fdb_info->vid, vport_num); + esw_debug(esw->dev, + "FDB remove entry with specified key not found (MAC=%pM,vid=%u,vport=%u)\n", + fdb_info->addr, fdb_info->vid, vport_num); return; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c index 2db13c71e88c..3d0bbcca1cb9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/debugfs.c @@ -12,10 +12,11 @@ enum vnic_diag_counter { MLX5_VNIC_DIAG_CQ_OVERRUN, MLX5_VNIC_DIAG_INVALID_COMMAND, MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND, + MLX5_VNIC_DIAG_RX_STEERING_DISCARD, }; static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_counter counter, - u32 *val) + u64 *val) { u32 out[MLX5_ST_SZ_DW(query_vnic_env_out)] = {}; u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {}; @@ -57,6 +58,10 @@ static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_cou case MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND: *val = MLX5_GET(vnic_diagnostic_statistics, vnic_diag_out, quota_exceeded_command); break; + case MLX5_VNIC_DIAG_RX_STEERING_DISCARD: + *val = MLX5_GET64(vnic_diagnostic_statistics, vnic_diag_out, + nic_receive_steering_discard); + break; } return 0; @@ -65,14 +70,14 @@ static int mlx5_esw_query_vnic_diag(struct mlx5_vport *vport, enum vnic_diag_cou static int __show_vnic_diag(struct seq_file *file, struct mlx5_vport *vport, enum vnic_diag_counter type) { - u32 val = 0; + u64 val = 0; int ret; ret = mlx5_esw_query_vnic_diag(vport, type, &val); if (ret) return ret; - seq_printf(file, "%d\n", val); + seq_printf(file, "%llu\n", val); return 0; } @@ -112,6 +117,11 @@ static int quota_exceeded_command_show(struct seq_file *file, void *priv) return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_QOUTA_EXCEEDED_COMMAND); } +static int rx_steering_discard_show(struct seq_file *file, void *priv) +{ + return __show_vnic_diag(file, file->private, MLX5_VNIC_DIAG_RX_STEERING_DISCARD); +} + DEFINE_SHOW_ATTRIBUTE(total_q_under_processor_handle); DEFINE_SHOW_ATTRIBUTE(send_queue_priority_update_flow); DEFINE_SHOW_ATTRIBUTE(comp_eq_overrun); @@ -119,6 +129,7 @@ DEFINE_SHOW_ATTRIBUTE(async_eq_overrun); DEFINE_SHOW_ATTRIBUTE(cq_overrun); DEFINE_SHOW_ATTRIBUTE(invalid_command); DEFINE_SHOW_ATTRIBUTE(quota_exceeded_command); +DEFINE_SHOW_ATTRIBUTE(rx_steering_discard); void mlx5_esw_vport_debugfs_destroy(struct mlx5_eswitch *esw, u16 vport_num) { @@ -179,4 +190,9 @@ void mlx5_esw_vport_debugfs_create(struct mlx5_eswitch *esw, u16 vport_num, bool if (MLX5_CAP_GEN(esw->dev, quota_exceeded_count)) debugfs_create_file("quota_exceeded_command", 0444, vnic_diag, vport, "a_exceeded_command_fops); + + if (MLX5_CAP_GEN(esw->dev, nic_receive_steering_discard)) + debugfs_create_file("rx_steering_discard", 0444, vnic_diag, vport, + &rx_steering_discard_fops); + } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c index 9bc7be95db54..084a910bb4e7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/devlink_port.c @@ -91,7 +91,7 @@ int mlx5_esw_offloads_devlink_port_register(struct mlx5_eswitch *esw, u16 vport_ if (err) goto reg_err; - err = devl_rate_leaf_create(dl_port, vport); + err = devl_rate_leaf_create(dl_port, vport, NULL); if (err) goto rate_err; @@ -160,7 +160,7 @@ int mlx5_esw_devlink_sf_port_register(struct mlx5_eswitch *esw, struct devlink_p if (err) return err; - err = devl_rate_leaf_create(dl_port, vport); + err = devl_rate_leaf_create(dl_port, vport, NULL); if (err) goto rate_err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 374e3fbdc2cf..527e4bffda8d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -772,6 +772,41 @@ static void esw_vport_cleanup_acl(struct mlx5_eswitch *esw, esw_vport_destroy_offloads_acl_tables(esw, vport); } +static int mlx5_esw_vport_caps_get(struct mlx5_eswitch *esw, struct mlx5_vport *vport) +{ + int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out); + void *query_ctx; + void *hca_caps; + int err; + + if (!MLX5_CAP_GEN(esw->dev, vhca_resource_manager)) + return 0; + + query_ctx = kzalloc(query_out_sz, GFP_KERNEL); + if (!query_ctx) + return -ENOMEM; + + err = mlx5_vport_get_other_func_cap(esw->dev, vport->vport, query_ctx, + MLX5_CAP_GENERAL); + if (err) + goto out_free; + + hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability); + vport->info.roce_enabled = MLX5_GET(cmd_hca_cap, hca_caps, roce); + + memset(query_ctx, 0, query_out_sz); + err = mlx5_vport_get_other_func_cap(esw->dev, vport->vport, query_ctx, + MLX5_CAP_GENERAL_2); + if (err) + goto out_free; + + hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability); + vport->info.mig_enabled = MLX5_GET(cmd_hca_cap_2, hca_caps, migratable); +out_free: + kfree(query_ctx); + return err; +} + static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { u16 vport_num = vport->vport; @@ -785,6 +820,10 @@ static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) if (mlx5_esw_is_manager_vport(esw, vport_num)) return 0; + err = mlx5_esw_vport_caps_get(esw, vport); + if (err) + goto err_caps; + mlx5_modify_vport_admin_state(esw->dev, MLX5_VPORT_STATE_OP_MOD_ESW_VPORT, vport_num, 1, @@ -804,6 +843,10 @@ static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) vport->info.qos, flags); return 0; + +err_caps: + esw_vport_cleanup_acl(esw, vport); + return err; } /* Don't cleanup vport->info, it's needed to restore vport configuration */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 3029bc1c0dd0..5a85a5d32be7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -153,6 +153,8 @@ struct mlx5_vport_info { u8 qos; u8 spoofchk: 1; u8 trusted: 1; + u8 roce_enabled: 1; + u8 mig_enabled: 1; }; /* Vport context events */ @@ -508,7 +510,14 @@ int mlx5_devlink_port_function_hw_addr_get(struct devlink_port *port, int mlx5_devlink_port_function_hw_addr_set(struct devlink_port *port, const u8 *hw_addr, int hw_addr_len, struct netlink_ext_ack *extack); - +int mlx5_devlink_port_fn_roce_get(struct devlink_port *port, bool *is_enabled, + struct netlink_ext_ack *extack); +int mlx5_devlink_port_fn_roce_set(struct devlink_port *port, bool enable, + struct netlink_ext_ack *extack); +int mlx5_devlink_port_fn_migratable_get(struct devlink_port *port, bool *is_enabled, + struct netlink_ext_ack *extack); +int mlx5_devlink_port_fn_migratable_set(struct devlink_port *port, bool enable, + struct netlink_ext_ack *extack); void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type); int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, @@ -744,6 +753,11 @@ static inline int mlx5_eswitch_num_vfs(struct mlx5_eswitch *esw) return 0; } +static inline struct mlx5_flow_table * +mlx5_eswitch_get_slow_fdb(struct mlx5_eswitch *esw) +{ + return esw->fdb_table.offloads.slow_fdb; +} #else /* CONFIG_MLX5_ESWITCH */ /* eswitch API stubs */ static inline int mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 8c6c9bcb3dc3..e455b215c708 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -50,6 +50,7 @@ #include "en/mapping.h" #include "devlink.h" #include "lag/lag.h" +#include "en/tc/post_meter.h" #define mlx5_esw_for_each_rep(esw, i, rep) \ xa_for_each(&((esw)->offloads.vport_reps), i, rep) @@ -202,6 +203,21 @@ esw_cleanup_decap_indir(struct mlx5_eswitch *esw, } static int +esw_setup_mtu_dest(struct mlx5_flow_destination *dest, + struct mlx5e_meter_attr *meter, + int i) +{ + dest[i].type = MLX5_FLOW_DESTINATION_TYPE_RANGE; + dest[i].range.field = MLX5_FLOW_DEST_RANGE_FIELD_PKT_LEN; + dest[i].range.min = 0; + dest[i].range.max = meter->params.mtu; + dest[i].range.hit_ft = mlx5e_post_meter_get_mtu_true_ft(meter->post_meter); + dest[i].range.miss_ft = mlx5e_post_meter_get_mtu_false_ft(meter->post_meter); + + return 0; +} + +static int esw_setup_sampler_dest(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act, u32 sampler_id, @@ -248,7 +264,7 @@ esw_setup_slow_path_dest(struct mlx5_flow_destination *dest, struct mlx5_flow_ac if (MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level)) flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; - dest[i].ft = esw->fdb_table.offloads.slow_fdb; + dest[i].ft = mlx5_eswitch_get_slow_fdb(esw); } static int @@ -479,16 +495,21 @@ esw_setup_dests(struct mlx5_flow_destination *dest, esw_src_port_rewrite_supported(esw)) attr->flags |= MLX5_ATTR_FLAG_SRC_REWRITE; - if (attr->flags & MLX5_ATTR_FLAG_SAMPLE && - !(attr->flags & MLX5_ATTR_FLAG_SLOW_PATH)) { - esw_setup_sampler_dest(dest, flow_act, attr->sample_attr.sampler_id, *i); - (*i)++; - } else if (attr->flags & MLX5_ATTR_FLAG_SLOW_PATH) { + if (attr->flags & MLX5_ATTR_FLAG_SLOW_PATH) { esw_setup_slow_path_dest(dest, flow_act, esw, *i); (*i)++; + goto out; + } + + if (attr->flags & MLX5_ATTR_FLAG_SAMPLE) { + esw_setup_sampler_dest(dest, flow_act, attr->sample_attr.sampler_id, *i); + (*i)++; } else if (attr->flags & MLX5_ATTR_FLAG_ACCEPT) { esw_setup_accept_dest(dest, flow_act, chains, *i); (*i)++; + } else if (attr->flags & MLX5_ATTR_FLAG_MTU) { + err = esw_setup_mtu_dest(dest, &attr->meter_attr, *i); + (*i)++; } else if (esw_is_indir_table(esw, attr)) { err = esw_setup_indir_table(dest, flow_act, esw, attr, spec, true, i); } else if (esw_is_chain_src_port_rewrite(esw, esw_attr)) { @@ -506,6 +527,7 @@ esw_setup_dests(struct mlx5_flow_destination *dest, } } +out: return err; } @@ -637,6 +659,11 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, goto err_esw_get; } + if (!i) { + kfree(dest); + dest = NULL; + } + if (mlx5_eswitch_termtbl_required(esw, attr, &flow_act, spec)) rule = mlx5_eswitch_add_termtbl_rule(esw, fdb, spec, esw_attr, &flow_act, dest, i); @@ -1046,7 +1073,7 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *on_esw, if (rep->vport == MLX5_VPORT_UPLINK) spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT; - flow_rule = mlx5_add_flow_rules(on_esw->fdb_table.offloads.slow_fdb, + flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(on_esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow_rule)) esw_warn(on_esw->dev, "FDB: Failed to add send to vport rule err %ld\n", @@ -1095,7 +1122,7 @@ mlx5_eswitch_add_send_to_vport_meta_rule(struct mlx5_eswitch *esw, u16 vport_num mlx5_eswitch_get_vport_metadata_for_match(esw, vport_num)); dest.vport.num = vport_num; - flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, + flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow_rule)) esw_warn(esw->dev, "FDB: Failed to add send to vport meta rule vport %d, err %ld\n", @@ -1248,7 +1275,7 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw, esw_set_peer_miss_rule_source_port(esw, peer_dev->priv.eswitch, spec, MLX5_VPORT_PF); - flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, + flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow)) { err = PTR_ERR(flow); @@ -1260,7 +1287,7 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw, if (mlx5_ecpf_vport_exists(esw->dev)) { vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_ECPF); MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_ECPF); - flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, + flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow)) { err = PTR_ERR(flow); @@ -1274,7 +1301,7 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw, peer_dev->priv.eswitch, spec, vport->vport); - flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, + flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow)) { err = PTR_ERR(flow); @@ -1363,7 +1390,7 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) dest.vport.num = esw->manager_vport; flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; - flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, + flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow_rule)) { err = PTR_ERR(flow_rule); @@ -1378,7 +1405,7 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) dmac_v = MLX5_ADDR_OF(fte_match_param, headers_v, outer_headers.dmac_47_16); dmac_v[0] = 0x01; - flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb, + flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw), spec, &flow_act, &dest, 1); if (IS_ERR(flow_rule)) { err = PTR_ERR(flow_rule); @@ -1927,7 +1954,7 @@ send_vport_err: fdb_chains_err: mlx5_destroy_flow_table(esw->fdb_table.offloads.tc_miss_table); tc_miss_table_err: - mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb); + mlx5_destroy_flow_table(mlx5_eswitch_get_slow_fdb(esw)); slow_fdb_err: /* Holds true only as long as DMFS is the default */ mlx5_flow_namespace_set_mode(root_ns, MLX5_FLOW_STEERING_MODE_DMFS); @@ -1938,7 +1965,7 @@ ns_err: static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw) { - if (!esw->fdb_table.offloads.slow_fdb) + if (!mlx5_eswitch_get_slow_fdb(esw)) return; esw_debug(esw->dev, "Destroy offloads FDB Tables\n"); @@ -1954,7 +1981,7 @@ static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw) esw_chains_destroy(esw, esw_chains(esw)); mlx5_destroy_flow_table(esw->fdb_table.offloads.tc_miss_table); - mlx5_destroy_flow_table(esw->fdb_table.offloads.slow_fdb); + mlx5_destroy_flow_table(mlx5_eswitch_get_slow_fdb(esw)); /* Holds true only as long as DMFS is the default */ mlx5_flow_namespace_set_mode(esw->fdb_table.offloads.ns, MLX5_FLOW_STEERING_MODE_DMFS); @@ -3886,7 +3913,7 @@ static int mlx5_esw_query_vport_vhca_id(struct mlx5_eswitch *esw, u16 vport_num, if (!query_ctx) return -ENOMEM; - err = mlx5_vport_get_other_func_cap(esw->dev, vport_num, query_ctx); + err = mlx5_vport_get_other_func_general_cap(esw->dev, vport_num, query_ctx); if (err) goto out_free; @@ -4019,3 +4046,212 @@ int mlx5_devlink_port_function_hw_addr_set(struct devlink_port *port, return mlx5_eswitch_set_vport_mac(esw, vport_num, hw_addr); } + +static struct mlx5_vport * +mlx5_devlink_port_fn_get_vport(struct devlink_port *port, struct mlx5_eswitch *esw) +{ + u16 vport_num; + + if (!MLX5_CAP_GEN(esw->dev, vhca_resource_manager)) + return ERR_PTR(-EOPNOTSUPP); + + vport_num = mlx5_esw_devlink_port_index_to_vport_num(port->index); + if (!is_port_function_supported(esw, vport_num)) + return ERR_PTR(-EOPNOTSUPP); + + return mlx5_eswitch_get_vport(esw, vport_num); +} + +int mlx5_devlink_port_fn_migratable_get(struct devlink_port *port, bool *is_enabled, + struct netlink_ext_ack *extack) +{ + struct mlx5_eswitch *esw; + struct mlx5_vport *vport; + int err = -EOPNOTSUPP; + + esw = mlx5_devlink_eswitch_get(port->devlink); + if (IS_ERR(esw)) + return PTR_ERR(esw); + + if (!MLX5_CAP_GEN(esw->dev, migration)) { + NL_SET_ERR_MSG_MOD(extack, "Device doesn't support migration"); + return err; + } + + vport = mlx5_devlink_port_fn_get_vport(port, esw); + if (IS_ERR(vport)) { + NL_SET_ERR_MSG_MOD(extack, "Invalid port"); + return PTR_ERR(vport); + } + + mutex_lock(&esw->state_lock); + if (vport->enabled) { + *is_enabled = vport->info.mig_enabled; + err = 0; + } + mutex_unlock(&esw->state_lock); + return err; +} + +int mlx5_devlink_port_fn_migratable_set(struct devlink_port *port, bool enable, + struct netlink_ext_ack *extack) +{ + int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out); + struct mlx5_eswitch *esw; + struct mlx5_vport *vport; + void *query_ctx; + void *hca_caps; + int err = -EOPNOTSUPP; + + esw = mlx5_devlink_eswitch_get(port->devlink); + if (IS_ERR(esw)) + return PTR_ERR(esw); + + if (!MLX5_CAP_GEN(esw->dev, migration)) { + NL_SET_ERR_MSG_MOD(extack, "Device doesn't support migration"); + return err; + } + + vport = mlx5_devlink_port_fn_get_vport(port, esw); + if (IS_ERR(vport)) { + NL_SET_ERR_MSG_MOD(extack, "Invalid port"); + return PTR_ERR(vport); + } + + mutex_lock(&esw->state_lock); + if (!vport->enabled) { + NL_SET_ERR_MSG_MOD(extack, "Eswitch vport is disabled"); + goto out; + } + + if (vport->info.mig_enabled == enable) { + err = 0; + goto out; + } + + query_ctx = kzalloc(query_out_sz, GFP_KERNEL); + if (!query_ctx) { + err = -ENOMEM; + goto out; + } + + err = mlx5_vport_get_other_func_cap(esw->dev, vport->vport, query_ctx, + MLX5_CAP_GENERAL_2); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed getting HCA caps"); + goto out_free; + } + + hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability); + memcpy(hca_caps, MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability), + MLX5_UN_SZ_BYTES(hca_cap_union)); + MLX5_SET(cmd_hca_cap_2, hca_caps, migratable, 1); + + err = mlx5_vport_set_other_func_cap(esw->dev, hca_caps, vport->vport, + MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE2); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed setting HCA migratable cap"); + goto out_free; + } + + vport->info.mig_enabled = enable; + +out_free: + kfree(query_ctx); +out: + mutex_unlock(&esw->state_lock); + return err; +} + +int mlx5_devlink_port_fn_roce_get(struct devlink_port *port, bool *is_enabled, + struct netlink_ext_ack *extack) +{ + struct mlx5_eswitch *esw; + struct mlx5_vport *vport; + int err = -EOPNOTSUPP; + + esw = mlx5_devlink_eswitch_get(port->devlink); + if (IS_ERR(esw)) + return PTR_ERR(esw); + + vport = mlx5_devlink_port_fn_get_vport(port, esw); + if (IS_ERR(vport)) { + NL_SET_ERR_MSG_MOD(extack, "Invalid port"); + return PTR_ERR(vport); + } + + mutex_lock(&esw->state_lock); + if (vport->enabled) { + *is_enabled = vport->info.roce_enabled; + err = 0; + } + mutex_unlock(&esw->state_lock); + return err; +} + +int mlx5_devlink_port_fn_roce_set(struct devlink_port *port, bool enable, + struct netlink_ext_ack *extack) +{ + int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out); + struct mlx5_eswitch *esw; + struct mlx5_vport *vport; + int err = -EOPNOTSUPP; + void *query_ctx; + void *hca_caps; + u16 vport_num; + + esw = mlx5_devlink_eswitch_get(port->devlink); + if (IS_ERR(esw)) + return PTR_ERR(esw); + + vport = mlx5_devlink_port_fn_get_vport(port, esw); + if (IS_ERR(vport)) { + NL_SET_ERR_MSG_MOD(extack, "Invalid port"); + return PTR_ERR(vport); + } + vport_num = vport->vport; + + mutex_lock(&esw->state_lock); + if (!vport->enabled) { + NL_SET_ERR_MSG_MOD(extack, "Eswitch vport is disabled"); + goto out; + } + + if (vport->info.roce_enabled == enable) { + err = 0; + goto out; + } + + query_ctx = kzalloc(query_out_sz, GFP_KERNEL); + if (!query_ctx) { + err = -ENOMEM; + goto out; + } + + err = mlx5_vport_get_other_func_cap(esw->dev, vport_num, query_ctx, + MLX5_CAP_GENERAL); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed getting HCA caps"); + goto out_free; + } + + hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability); + memcpy(hca_caps, MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability), + MLX5_UN_SZ_BYTES(hca_cap_union)); + MLX5_SET(cmd_hca_cap, hca_caps, roce, enable); + + err = mlx5_vport_set_other_func_cap(esw->dev, hca_caps, vport_num, + MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed setting HCA roce cap"); + goto out_free; + } + + vport->info.roce_enabled = enable; + +out_free: + kfree(query_ctx); +out: + mutex_unlock(&esw->state_lock); + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c index edd910258314..3a9a6bb9158d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c @@ -210,6 +210,18 @@ static bool mlx5_eswitch_offload_is_uplink_port(const struct mlx5_eswitch *esw, return (port_mask & port_value) == MLX5_VPORT_UPLINK; } +static bool +mlx5_eswitch_is_push_vlan_no_cap(struct mlx5_eswitch *esw, + struct mlx5_flow_act *flow_act) +{ + if (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH && + !(mlx5_fs_get_capabilities(esw->dev, MLX5_FLOW_NAMESPACE_FDB) & + MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX)) + return true; + + return false; +} + bool mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, struct mlx5_flow_attr *attr, @@ -225,10 +237,7 @@ mlx5_eswitch_termtbl_required(struct mlx5_eswitch *esw, (!mlx5_eswitch_offload_is_uplink_port(esw, spec) && !esw_attr->int_port)) return false; - /* push vlan on RX */ - if (flow_act->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH && - !(mlx5_fs_get_capabilities(esw->dev, MLX5_FLOW_NAMESPACE_FDB) & - MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX)) + if (mlx5_eswitch_is_push_vlan_no_cap(esw, flow_act)) return true; /* hairpin */ @@ -252,19 +261,31 @@ mlx5_eswitch_add_termtbl_rule(struct mlx5_eswitch *esw, struct mlx5_flow_act term_tbl_act = {}; struct mlx5_flow_handle *rule = NULL; bool term_table_created = false; + bool is_push_vlan_on_rx; int num_vport_dests = 0; int i, curr_dest; + is_push_vlan_on_rx = mlx5_eswitch_is_push_vlan_no_cap(esw, flow_act); mlx5_eswitch_termtbl_actions_move(flow_act, &term_tbl_act); term_tbl_act.action |= MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; for (i = 0; i < num_dest; i++) { struct mlx5_termtbl_handle *tt; + bool hairpin = false; /* only vport destinations can be terminated */ if (dest[i].type != MLX5_FLOW_DESTINATION_TYPE_VPORT) continue; + if (attr->dests[num_vport_dests].rep && + attr->dests[num_vport_dests].rep->vport == MLX5_VPORT_UPLINK) + hairpin = true; + + if (!is_push_vlan_on_rx && !hairpin) { + num_vport_dests++; + continue; + } + if (attr->dests[num_vport_dests].flags & MLX5_ESW_DEST_ENCAP) { term_tbl_act.action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT; term_tbl_act.pkt_reformat = attr->dests[num_vport_dests].pkt_reformat; @@ -312,6 +333,9 @@ revert_changes: for (curr_dest = 0; curr_dest < num_vport_dests; curr_dest++) { struct mlx5_termtbl_handle *tt = attr->dests[curr_dest].termtbl; + if (!tt) + continue; + attr->dests[curr_dest].termtbl = NULL; /* search for the destination associated with the diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index d53749248fa0..5a85d8c1e797 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -111,8 +111,8 @@ #define ETHTOOL_PRIO_NUM_LEVELS 1 #define ETHTOOL_NUM_PRIOS 11 #define ETHTOOL_MIN_LEVEL (KERNEL_MIN_LEVEL + ETHTOOL_NUM_PRIOS) -/* Promiscuous, Vlan, mac, ttc, inner ttc, {UDP/ANY/aRFS/accel/{esp, esp_err}} */ -#define KERNEL_NIC_PRIO_NUM_LEVELS 7 +/* Promiscuous, Vlan, mac, ttc, inner ttc, {UDP/ANY/aRFS/accel/{esp, esp_err}}, IPsec policy */ +#define KERNEL_NIC_PRIO_NUM_LEVELS 8 #define KERNEL_NIC_NUM_PRIOS 1 /* One more level for tc */ #define KERNEL_MIN_LEVEL (KERNEL_NIC_PRIO_NUM_LEVELS + 1) @@ -133,7 +133,7 @@ #define LAG_MIN_LEVEL (OFFLOADS_MIN_LEVEL + KERNEL_RX_MACSEC_MIN_LEVEL + 1) #define KERNEL_TX_IPSEC_NUM_PRIOS 1 -#define KERNEL_TX_IPSEC_NUM_LEVELS 1 +#define KERNEL_TX_IPSEC_NUM_LEVELS 2 #define KERNEL_TX_IPSEC_MIN_LEVEL (KERNEL_TX_IPSEC_NUM_LEVELS) #define KERNEL_TX_MACSEC_NUM_PRIOS 1 @@ -448,7 +448,8 @@ static bool is_fwd_dest_type(enum mlx5_flow_destination_type type) type == MLX5_FLOW_DESTINATION_TYPE_UPLINK || type == MLX5_FLOW_DESTINATION_TYPE_VPORT || type == MLX5_FLOW_DESTINATION_TYPE_FLOW_SAMPLER || - type == MLX5_FLOW_DESTINATION_TYPE_TIR; + type == MLX5_FLOW_DESTINATION_TYPE_TIR || + type == MLX5_FLOW_DESTINATION_TYPE_RANGE; } static bool check_valid_spec(const struct mlx5_flow_spec *spec) @@ -1578,7 +1579,13 @@ static bool mlx5_flow_dests_cmp(struct mlx5_flow_destination *d1, (d1->type == MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE_NUM && d1->ft_num == d2->ft_num) || (d1->type == MLX5_FLOW_DESTINATION_TYPE_FLOW_SAMPLER && - d1->sampler_id == d2->sampler_id)) + d1->sampler_id == d2->sampler_id) || + (d1->type == MLX5_FLOW_DESTINATION_TYPE_RANGE && + d1->range.field == d2->range.field && + d1->range.hit_ft == d2->range.hit_ft && + d1->range.miss_ft == d2->range.miss_ft && + d1->range.min == d2->range.min && + d1->range.max == d2->range.max)) return true; } @@ -1962,6 +1969,9 @@ _mlx5_add_flow_rules(struct mlx5_flow_table *ft, if (flow_act->fg && ft->autogroup.active) return ERR_PTR(-EINVAL); + if (dest && dest_num <= 0) + return ERR_PTR(-EINVAL); + for (i = 0; i < dest_num; i++) { if (!dest_is_valid(&dest[i], flow_act, ft)) return ERR_PTR(-EINVAL); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 3af50fd04d28..f137a0611b77 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -123,6 +123,7 @@ enum mlx5_flow_steering_mode { enum mlx5_flow_steering_capabilty { MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX = 1UL << 0, MLX5_FLOW_STEERING_CAP_VLAN_POP_ON_TX = 1UL << 1, + MLX5_FLOW_STEERING_CAP_MATCH_RANGES = 1UL << 2, }; struct mlx5_flow_steering { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 4e3a75496dd9..7c5c500fd215 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -561,12 +561,17 @@ static int mlx5i_open(struct net_device *netdev) if (err) goto err_remove_fs_underlay_qp; - epriv->profile->update_rx(epriv); + err = epriv->profile->update_rx(epriv); + if (err) + goto err_close_channels; + mlx5e_activate_priv_channels(epriv); mutex_unlock(&epriv->state_lock); return 0; +err_close_channels: + mlx5e_close_channels(&epriv->channels); err_remove_fs_underlay_qp: mlx5_fs_remove_rx_underlay_qpn(mdev, ipriv->qpn); err_reset_qp: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c index 0227a521d301..4d9c9e49645c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c @@ -221,12 +221,16 @@ static int mlx5i_pkey_open(struct net_device *netdev) mlx5_core_warn(mdev, "opening child channels failed, %d\n", err); goto err_clear_state_opened_flag; } - epriv->profile->update_rx(epriv); + err = epriv->profile->update_rx(epriv); + if (err) + goto err_close_channels; mlx5e_activate_priv_channels(epriv); mutex_unlock(&epriv->state_lock); return 0; +err_close_channels: + mlx5e_close_channels(&epriv->channels); err_clear_state_opened_flag: mlx5e_destroy_tis(mdev, epriv->tisn[0][0]); err_remove_rx_uderlay_qp: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c index 0259a149a64c..d9fcb9ed726f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c @@ -118,13 +118,41 @@ struct mlx5_fib_event_work { }; }; +static struct net_device* +mlx5_lag_get_next_fib_dev(struct mlx5_lag *ldev, + struct fib_info *fi, + struct net_device *current_dev) +{ + struct net_device *fib_dev; + int i, ldev_idx, nhs; + + nhs = fib_info_num_path(fi); + i = 0; + if (current_dev) { + for (; i < nhs; i++) { + fib_dev = fib_info_nh(fi, i)->fib_nh_dev; + if (fib_dev == current_dev) { + i++; + break; + } + } + } + for (; i < nhs; i++) { + fib_dev = fib_info_nh(fi, i)->fib_nh_dev; + ldev_idx = mlx5_lag_dev_get_netdev_idx(ldev, fib_dev); + if (ldev_idx >= 0) + return ldev->pf[ldev_idx].netdev; + } + + return NULL; +} + static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev, unsigned long event, struct fib_entry_notifier_info *fen_info) { + struct net_device *nh_dev0, *nh_dev1; struct fib_info *fi = fen_info->fi; struct lag_mp *mp = &ldev->lag_mp; - struct fib_nh *fib_nh0, *fib_nh1; - unsigned int nhs; /* Handle delete event */ if (event == FIB_EVENT_ENTRY_DEL) { @@ -140,16 +168,25 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev, unsigned long event, fi->fib_priority >= mp->fib.priority) return; + nh_dev0 = mlx5_lag_get_next_fib_dev(ldev, fi, NULL); + nh_dev1 = mlx5_lag_get_next_fib_dev(ldev, fi, nh_dev0); + /* Handle add/replace event */ - nhs = fib_info_num_path(fi); - if (nhs == 1) { - if (__mlx5_lag_is_active(ldev)) { - struct fib_nh *nh = fib_info_nh(fi, 0); - struct net_device *nh_dev = nh->fib_nh_dev; - int i = mlx5_lag_dev_get_netdev_idx(ldev, nh_dev); + if (!nh_dev0) { + if (mp->fib.dst == fen_info->dst && mp->fib.dst_len == fen_info->dst_len) + mp->fib.mfi = NULL; + return; + } - if (i < 0) - return; + if (nh_dev0 == nh_dev1) { + mlx5_core_warn(ldev->pf[MLX5_LAG_P1].dev, + "Multipath offload doesn't support routes with multiple nexthops of the same device"); + return; + } + + if (!nh_dev1) { + if (__mlx5_lag_is_active(ldev)) { + int i = mlx5_lag_dev_get_netdev_idx(ldev, nh_dev0); i++; mlx5_lag_set_port_affinity(ldev, i); @@ -159,21 +196,6 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev, unsigned long event, return; } - if (nhs != 2) - return; - - /* Verify next hops are ports of the same hca */ - fib_nh0 = fib_info_nh(fi, 0); - fib_nh1 = fib_info_nh(fi, 1); - if (!(fib_nh0->fib_nh_dev == ldev->pf[MLX5_LAG_P1].netdev && - fib_nh1->fib_nh_dev == ldev->pf[MLX5_LAG_P2].netdev) && - !(fib_nh0->fib_nh_dev == ldev->pf[MLX5_LAG_P2].netdev && - fib_nh1->fib_nh_dev == ldev->pf[MLX5_LAG_P1].netdev)) { - mlx5_core_warn(ldev->pf[MLX5_LAG_P1].dev, - "Multipath offload require two ports of the same HCA\n"); - return; - } - /* First time we see multipath route */ if (!mp->fib.mfi && !__mlx5_lag_is_active(ldev)) { struct lag_tracker tracker; @@ -268,7 +290,6 @@ static int mlx5_lag_fib_event(struct notifier_block *nb, struct mlx5_fib_event_work *fib_work; struct fib_entry_notifier_info *fen_info; struct fib_nh_notifier_info *fnh_info; - struct net_device *fib_dev; struct fib_info *fi; if (info->family != AF_INET) @@ -285,11 +306,7 @@ static int mlx5_lag_fib_event(struct notifier_block *nb, fi = fen_info->fi; if (fi->nh) return NOTIFY_DONE; - fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev; - if (fib_dev != ldev->pf[MLX5_LAG_P1].netdev && - fib_dev != ldev->pf[MLX5_LAG_P2].netdev) { - return NOTIFY_DONE; - } + fib_work = mlx5_lag_init_fib_work(ldev, event); if (!fib_work) return NOTIFY_DONE; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c index c971ff04dd04..5a80fb7dbbca 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c @@ -334,9 +334,6 @@ err_cq: void mlx5_aso_destroy(struct mlx5_aso *aso) { - if (IS_ERR_OR_NULL(aso)) - return; - mlx5_aso_destroy_sq(aso); mlx5_aso_destroy_cq(&aso->cq); kfree(aso); @@ -356,12 +353,15 @@ void mlx5_aso_build_wqe(struct mlx5_aso *aso, u8 ds_cnt, cseg->general_id = cpu_to_be32(obj_id); } -void *mlx5_aso_get_wqe(struct mlx5_aso *aso) +struct mlx5_aso_wqe *mlx5_aso_get_wqe(struct mlx5_aso *aso) { + struct mlx5_aso_wqe *wqe; u16 pi; pi = mlx5_wq_cyc_ctr2ix(&aso->wq, aso->pc); - return mlx5_wq_cyc_get_wqe(&aso->wq, pi); + wqe = mlx5_wq_cyc_get_wqe(&aso->wq, pi); + memset(wqe, 0, sizeof(*wqe)); + return wqe; } void mlx5_aso_post_wqe(struct mlx5_aso *aso, bool with_data, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.h index 2d40dcf9d42e..afb078bbb8ef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/aso.h @@ -15,6 +15,7 @@ #define MLX5_WQE_CTRL_WQE_OPC_MOD_SHIFT 24 #define MLX5_MACSEC_ASO_DS_CNT (DIV_ROUND_UP(sizeof(struct mlx5_aso_wqe), MLX5_SEND_WQE_DS)) +#define ASO_CTRL_READ_EN BIT(0) struct mlx5_wqe_aso_ctrl_seg { __be32 va_h; __be32 va_l; /* include read_enable */ @@ -71,13 +72,14 @@ enum { }; enum { + MLX5_ACCESS_ASO_OPC_MOD_IPSEC = 0x0, MLX5_ACCESS_ASO_OPC_MOD_FLOW_METER = 0x2, MLX5_ACCESS_ASO_OPC_MOD_MACSEC = 0x5, }; struct mlx5_aso; -void *mlx5_aso_get_wqe(struct mlx5_aso *aso); +struct mlx5_aso_wqe *mlx5_aso_get_wqe(struct mlx5_aso *aso); void mlx5_aso_build_wqe(struct mlx5_aso *aso, u8 ds_cnt, struct mlx5_aso_wqe *aso_wqe, u32 obj_id, u32 opc_mode); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c index d3a9ae80fd30..69cfe60c558a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c @@ -339,35 +339,25 @@ static int mlx5_ptp_adjfreq_real_time(struct mlx5_core_dev *mdev, s32 freq) return mlx5_set_mtutc(mdev, in, sizeof(in)); } -static int mlx5_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta) +static int mlx5_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm) { struct mlx5_clock *clock = container_of(ptp, struct mlx5_clock, ptp_info); struct mlx5_timer *timer = &clock->timer; struct mlx5_core_dev *mdev; unsigned long flags; - int neg_adj = 0; - u32 diff; - u64 adj; + u32 mult; int err; mdev = container_of(clock, struct mlx5_core_dev, clock); - err = mlx5_ptp_adjfreq_real_time(mdev, delta); + err = mlx5_ptp_adjfreq_real_time(mdev, scaled_ppm_to_ppb(scaled_ppm)); if (err) return err; - if (delta < 0) { - neg_adj = 1; - delta = -delta; - } - - adj = timer->nominal_c_mult; - adj *= delta; - diff = div_u64(adj, 1000000000ULL); + mult = (u32)adjust_by_scaled_ppm(timer->nominal_c_mult, scaled_ppm); write_seqlock_irqsave(&clock->lock, flags); timecounter_read(&timer->tc); - timer->cycles.mult = neg_adj ? timer->nominal_c_mult - diff : - timer->nominal_c_mult + diff; + timer->cycles.mult = mult; mlx5_update_clock_info_page(mdev); write_sequnlock_irqrestore(&clock->lock, flags); @@ -697,7 +687,7 @@ static const struct ptp_clock_info mlx5_ptp_clock_info = { .n_per_out = 0, .n_pins = 0, .pps = 0, - .adjfreq = mlx5_ptp_adjfreq, + .adjfine = mlx5_ptp_adjfine, .adjtime = mlx5_ptp_adjtime, .gettimex64 = mlx5_ptp_gettimex, .settime64 = mlx5_ptp_settime, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index e58775a7d955..7f5db13e3550 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -37,7 +37,6 @@ #include <linux/pci.h> #include <linux/dma-mapping.h> #include <linux/slab.h> -#include <linux/io-mapping.h> #include <linux/interrupt.h> #include <linux/delay.h> #include <linux/mlx5/driver.h> @@ -1306,8 +1305,15 @@ static int mlx5_load(struct mlx5_core_dev *dev) mlx5_sf_dev_table_create(dev); + err = mlx5_devlink_traps_register(priv_to_devlink(dev)); + if (err) + goto err_traps_reg; + return 0; +err_traps_reg: + mlx5_sf_dev_table_destroy(dev); + mlx5_sriov_detach(dev); err_sriov: mlx5_lag_remove_mdev(dev); mlx5_ec_cleanup(dev); @@ -1336,6 +1342,7 @@ err_irq_table: static void mlx5_unload(struct mlx5_core_dev *dev) { + mlx5_devlink_traps_unregister(priv_to_devlink(dev)); mlx5_sf_dev_table_destroy(dev); mlx5_sriov_detach(dev); mlx5_eswitch_disable(dev->priv.eswitch); @@ -1580,14 +1587,22 @@ err: return -ENOMEM; } +static int vhca_id_show(struct seq_file *file, void *priv) +{ + struct mlx5_core_dev *dev = file->private; + + seq_printf(file, "0x%x\n", MLX5_CAP_GEN(dev, vhca_id)); + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(vhca_id); + int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) { struct mlx5_priv *priv = &dev->priv; int err; memcpy(&dev->profile, &profile[profile_idx], sizeof(dev->profile)); - INIT_LIST_HEAD(&priv->ctx_list); - spin_lock_init(&priv->ctx_lock); lockdep_register_key(&dev->lock_key); mutex_init(&dev->intf_state_mutex); lockdep_set_class(&dev->intf_state_mutex, &dev->lock_key); @@ -1604,6 +1619,7 @@ int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) priv->numa_node = dev_to_node(mlx5_core_dma_dev(dev)); priv->dbg.dbg_root = debugfs_create_dir(dev_name(dev->device), mlx5_debugfs_root); + debugfs_create_file("vhca_id", 0400, priv->dbg.dbg_root, dev, &vhca_id_fops); INIT_LIST_HEAD(&priv->traps); err = mlx5_tout_init(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index a806e3de7b7c..029305a8b80a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -324,7 +324,10 @@ void mlx5_unload_one_devl_locked(struct mlx5_core_dev *dev); int mlx5_load_one(struct mlx5_core_dev *dev, bool recovery); int mlx5_load_one_devl_locked(struct mlx5_core_dev *dev, bool recovery); -int mlx5_vport_get_other_func_cap(struct mlx5_core_dev *dev, u16 function_id, void *out); +int mlx5_vport_set_other_func_cap(struct mlx5_core_dev *dev, const void *hca_cap, u16 function_id, + u16 opmod); +#define mlx5_vport_get_other_func_general_cap(dev, fid, out) \ + mlx5_vport_get_other_func_cap(dev, fid, out, MLX5_CAP_GENERAL) void mlx5_events_work_enqueue(struct mlx5_core_dev *dev, struct work_struct *work); static inline u32 mlx5_sriov_get_vf_total_msix(struct pci_dev *pdev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c index 662f1d55e30e..6bde18bcd42f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c @@ -4,6 +4,7 @@ #include <linux/interrupt.h> #include <linux/notifier.h> #include <linux/mlx5/driver.h> +#include <linux/mlx5/vport.h> #include "mlx5_core.h" #include "mlx5_irq.h" #include "pci_irq.h" @@ -101,7 +102,7 @@ int mlx5_set_msix_vec_count(struct mlx5_core_dev *dev, int function_id, goto out; } - ret = mlx5_vport_get_other_func_cap(dev, function_id, query_cap); + ret = mlx5_vport_get_other_func_general_cap(dev, function_id, query_cap); if (ret) goto out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c index b1dfad274a39..ee104cf04392 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c @@ -44,6 +44,7 @@ static const char * const action_type_to_str[] = { [DR_ACTION_TYP_INSERT_HDR] = "DR_ACTION_TYP_INSERT_HDR", [DR_ACTION_TYP_REMOVE_HDR] = "DR_ACTION_TYP_REMOVE_HDR", [DR_ACTION_TYP_ASO_FLOW_METER] = "DR_ACTION_TYP_ASO_FLOW_METER", + [DR_ACTION_TYP_RANGE] = "DR_ACTION_TYP_RANGE", [DR_ACTION_TYP_MAX] = "DR_ACTION_UNKNOWN", }; @@ -61,6 +62,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, @@ -79,6 +81,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_DECAP, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_DECAP, @@ -94,6 +97,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_ENCAP, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ENCAP, @@ -103,6 +107,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_MODIFY_HDR, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR, @@ -116,6 +121,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_POP_VLAN, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_POP_VLAN, @@ -129,6 +135,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_PUSH_VLAN] = { [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_PUSH_VLAN, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_PUSH_VLAN, @@ -141,6 +148,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_TAG] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, @@ -159,6 +167,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ASO, }, [DR_ACTION_STATE_TERM] = { @@ -169,6 +178,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_NO_ACTION] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_ENCAP, @@ -183,6 +193,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_DECAP] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_DECAP, [DR_ACTION_TYP_ASO_FLOW_METER] = DR_ACTION_STATE_ASO, @@ -190,6 +201,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_ENCAP] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ENCAP, [DR_ACTION_TYP_ASO_FLOW_METER] = DR_ACTION_STATE_ASO, @@ -197,6 +209,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_MODIFY_HDR] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR, [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_ENCAP, @@ -207,6 +220,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] }, [DR_ACTION_STATE_POP_VLAN] = { [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_POP_VLAN, [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_POP_VLAN, @@ -220,6 +234,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_PUSH_VLAN] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_PUSH_VLAN, [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_PUSH_VLAN, @@ -231,6 +246,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_NON_TERM] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_ENCAP, @@ -250,6 +266,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ASO, [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, }, [DR_ACTION_STATE_TERM] = { [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_TERM, @@ -259,6 +276,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_NO_ACTION] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_TNL_L2_TO_L2] = DR_ACTION_STATE_DECAP, @@ -276,6 +294,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_DECAP] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_DECAP, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR, @@ -291,6 +310,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_QP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ENCAP, @@ -299,6 +319,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_MODIFY_HDR] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR, [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM, @@ -311,6 +332,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_POP_VLAN] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_POP_VLAN, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_POP_VLAN, @@ -324,6 +346,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_PUSH_VLAN] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_PUSH_VLAN, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_PUSH_VLAN, @@ -337,6 +360,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_NON_TERM] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_TNL_L2_TO_L2] = DR_ACTION_STATE_DECAP, @@ -354,6 +378,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_ASO] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ASO, }, @@ -365,6 +390,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_NO_ACTION] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR, @@ -380,6 +406,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_DECAP] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_DECAP, [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, @@ -388,6 +415,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_ENCAP] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ENCAP, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM, @@ -396,6 +424,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_MODIFY_HDR] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_MODIFY_HDR, [DR_ACTION_TYP_L2_TO_TNL_L2] = DR_ACTION_STATE_ENCAP, @@ -407,6 +436,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] }, [DR_ACTION_STATE_POP_VLAN] = { [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_POP_VLAN, [DR_ACTION_TYP_POP_VLAN] = DR_ACTION_STATE_POP_VLAN, @@ -421,6 +451,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_PUSH_VLAN] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_PUSH_VLAN, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_PUSH_VLAN, @@ -433,6 +464,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_STATE_NON_TERM] = { [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_SAMPLER] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_NON_TERM, [DR_ACTION_TYP_MODIFY_HDR] = DR_ACTION_STATE_MODIFY_HDR, @@ -452,6 +484,7 @@ next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] [DR_ACTION_TYP_PUSH_VLAN] = DR_ACTION_STATE_PUSH_VLAN, [DR_ACTION_TYP_DROP] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_FT] = DR_ACTION_STATE_TERM, + [DR_ACTION_TYP_RANGE] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_VPORT] = DR_ACTION_STATE_TERM, [DR_ACTION_TYP_CTR] = DR_ACTION_STATE_ASO, }, @@ -634,6 +667,83 @@ static void dr_action_print_sequence(struct mlx5dr_domain *dmn, actions[i]->action_type); } +static int dr_action_get_dest_fw_tbl_addr(struct mlx5dr_matcher *matcher, + struct mlx5dr_action_dest_tbl *dest_tbl, + bool is_rx_rule, + u64 *final_icm_addr) +{ + struct mlx5dr_cmd_query_flow_table_details output; + struct mlx5dr_domain *dmn = matcher->tbl->dmn; + int ret; + + if (!dest_tbl->fw_tbl.rx_icm_addr) { + ret = mlx5dr_cmd_query_flow_table(dmn->mdev, + dest_tbl->fw_tbl.type, + dest_tbl->fw_tbl.id, + &output); + if (ret) { + mlx5dr_err(dmn, + "Failed mlx5_cmd_query_flow_table ret: %d\n", + ret); + return ret; + } + + dest_tbl->fw_tbl.tx_icm_addr = output.sw_owner_icm_root_1; + dest_tbl->fw_tbl.rx_icm_addr = output.sw_owner_icm_root_0; + } + + *final_icm_addr = is_rx_rule ? dest_tbl->fw_tbl.rx_icm_addr : + dest_tbl->fw_tbl.tx_icm_addr; + return 0; +} + +static int dr_action_get_dest_sw_tbl_addr(struct mlx5dr_matcher *matcher, + struct mlx5dr_action_dest_tbl *dest_tbl, + bool is_rx_rule, + u64 *final_icm_addr) +{ + struct mlx5dr_domain *dmn = matcher->tbl->dmn; + struct mlx5dr_icm_chunk *chunk; + + if (dest_tbl->tbl->dmn != dmn) { + mlx5dr_err(dmn, + "Destination table belongs to a different domain\n"); + return -EINVAL; + } + + if (dest_tbl->tbl->level <= matcher->tbl->level) { + mlx5_core_dbg_once(dmn->mdev, + "Connecting table to a lower/same level destination table\n"); + mlx5dr_dbg(dmn, + "Connecting table at level %d to a destination table at level %d\n", + matcher->tbl->level, + dest_tbl->tbl->level); + } + + chunk = is_rx_rule ? dest_tbl->tbl->rx.s_anchor->chunk : + dest_tbl->tbl->tx.s_anchor->chunk; + + *final_icm_addr = mlx5dr_icm_pool_get_chunk_icm_addr(chunk); + return 0; +} + +static int dr_action_get_dest_tbl_addr(struct mlx5dr_matcher *matcher, + struct mlx5dr_action_dest_tbl *dest_tbl, + bool is_rx_rule, + u64 *final_icm_addr) +{ + if (dest_tbl->is_fw_tbl) + return dr_action_get_dest_fw_tbl_addr(matcher, + dest_tbl, + is_rx_rule, + final_icm_addr); + + return dr_action_get_dest_sw_tbl_addr(matcher, + dest_tbl, + is_rx_rule, + final_icm_addr); +} + #define WITH_VLAN_NUM_HW_ACTIONS 6 int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, @@ -661,8 +771,6 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, action_domain = dr_action_get_action_domain(dmn->type, nic_dmn->type); for (i = 0; i < num_actions; i++) { - struct mlx5dr_action_dest_tbl *dest_tbl; - struct mlx5dr_icm_chunk *chunk; struct mlx5dr_action *action; int max_actions_type = 1; u32 action_type; @@ -676,50 +784,27 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher, break; case DR_ACTION_TYP_FT: dest_action = action; - dest_tbl = action->dest_tbl; - if (!dest_tbl->is_fw_tbl) { - if (dest_tbl->tbl->dmn != dmn) { - mlx5dr_err(dmn, - "Destination table belongs to a different domain\n"); - return -EINVAL; - } - if (dest_tbl->tbl->level <= matcher->tbl->level) { - mlx5_core_dbg_once(dmn->mdev, - "Connecting table to a lower/same level destination table\n"); - mlx5dr_dbg(dmn, - "Connecting table at level %d to a destination table at level %d\n", - matcher->tbl->level, - dest_tbl->tbl->level); - } - chunk = rx_rule ? dest_tbl->tbl->rx.s_anchor->chunk : - dest_tbl->tbl->tx.s_anchor->chunk; - attr.final_icm_addr = mlx5dr_icm_pool_get_chunk_icm_addr(chunk); - } else { - struct mlx5dr_cmd_query_flow_table_details output; - int ret; - - /* get the relevant addresses */ - if (!action->dest_tbl->fw_tbl.rx_icm_addr) { - ret = mlx5dr_cmd_query_flow_table(dmn->mdev, - dest_tbl->fw_tbl.type, - dest_tbl->fw_tbl.id, - &output); - if (!ret) { - dest_tbl->fw_tbl.tx_icm_addr = - output.sw_owner_icm_root_1; - dest_tbl->fw_tbl.rx_icm_addr = - output.sw_owner_icm_root_0; - } else { - mlx5dr_err(dmn, - "Failed mlx5_cmd_query_flow_table ret: %d\n", - ret); - return ret; - } - } - attr.final_icm_addr = rx_rule ? - dest_tbl->fw_tbl.rx_icm_addr : - dest_tbl->fw_tbl.tx_icm_addr; - } + ret = dr_action_get_dest_tbl_addr(matcher, action->dest_tbl, + rx_rule, &attr.final_icm_addr); + if (ret) + return ret; + break; + case DR_ACTION_TYP_RANGE: + ret = dr_action_get_dest_tbl_addr(matcher, + action->range->hit_tbl_action->dest_tbl, + rx_rule, &attr.final_icm_addr); + if (ret) + return ret; + + ret = dr_action_get_dest_tbl_addr(matcher, + action->range->miss_tbl_action->dest_tbl, + rx_rule, &attr.range.miss_icm_addr); + if (ret) + return ret; + + attr.range.definer_id = action->range->definer_id; + attr.range.min = action->range->min; + attr.range.max = action->range->max; break; case DR_ACTION_TYP_QP: mlx5dr_info(dmn, "Domain doesn't support QP\n"); @@ -866,6 +951,7 @@ static unsigned int action_size[DR_ACTION_TYP_MAX] = { [DR_ACTION_TYP_REMOVE_HDR] = sizeof(struct mlx5dr_action_reformat), [DR_ACTION_TYP_SAMPLER] = sizeof(struct mlx5dr_action_sampler), [DR_ACTION_TYP_ASO_FLOW_METER] = sizeof(struct mlx5dr_action_aso_flow_meter), + [DR_ACTION_TYP_RANGE] = sizeof(struct mlx5dr_action_range), }; static struct mlx5dr_action * @@ -933,6 +1019,123 @@ dec_ref: return NULL; } +static void dr_action_range_definer_fill(u16 *format_id, + u8 *dw_selectors, + u8 *byte_selectors, + u8 *match_mask) +{ + int i; + + *format_id = MLX5_IFC_DEFINER_FORMAT_ID_SELECT; + + dw_selectors[0] = MLX5_IFC_DEFINER_FORMAT_OFFSET_OUTER_ETH_PKT_LEN / 4; + + for (i = 1; i < MLX5_IFC_DEFINER_DW_SELECTORS_NUM; i++) + dw_selectors[i] = MLX5_IFC_DEFINER_FORMAT_OFFSET_UNUSED; + + for (i = 0; i < MLX5_IFC_DEFINER_BYTE_SELECTORS_NUM; i++) + byte_selectors[i] = MLX5_IFC_DEFINER_FORMAT_OFFSET_UNUSED; + + MLX5_SET(match_definer_match_mask, match_mask, + match_dw_0, 0xffffUL << 16); +} + +static int dr_action_create_range_definer(struct mlx5dr_action *action) +{ + u8 match_mask[MLX5_FLD_SZ_BYTES(match_definer, match_mask)] = {}; + u8 byte_selectors[MLX5_IFC_DEFINER_BYTE_SELECTORS_NUM] = {}; + u8 dw_selectors[MLX5_IFC_DEFINER_DW_SELECTORS_NUM] = {}; + struct mlx5dr_domain *dmn = action->range->dmn; + u32 definer_id; + u16 format_id; + int ret; + + dr_action_range_definer_fill(&format_id, + dw_selectors, + byte_selectors, + match_mask); + + ret = mlx5dr_definer_get(dmn, format_id, + dw_selectors, byte_selectors, + match_mask, &definer_id); + if (ret) + return ret; + + action->range->definer_id = definer_id; + return 0; +} + +static void dr_action_destroy_range_definer(struct mlx5dr_action *action) +{ + mlx5dr_definer_put(action->range->dmn, action->range->definer_id); +} + +struct mlx5dr_action * +mlx5dr_action_create_dest_match_range(struct mlx5dr_domain *dmn, + u32 field, + struct mlx5_flow_table *hit_ft, + struct mlx5_flow_table *miss_ft, + u32 min, + u32 max) +{ + struct mlx5dr_action *action; + int ret; + + if (!mlx5dr_supp_match_ranges(dmn->mdev)) { + mlx5dr_dbg(dmn, "SELECT definer support is needed for match range\n"); + return NULL; + } + + if (field != MLX5_FLOW_DEST_RANGE_FIELD_PKT_LEN || + min > 0xffff || max > 0xffff) { + mlx5dr_err(dmn, "Invalid match range parameters\n"); + return NULL; + } + + action = dr_action_create_generic(DR_ACTION_TYP_RANGE); + if (!action) + return NULL; + + action->range->hit_tbl_action = + mlx5dr_is_fw_table(hit_ft) ? + mlx5dr_action_create_dest_flow_fw_table(dmn, hit_ft) : + mlx5dr_action_create_dest_table(hit_ft->fs_dr_table.dr_table); + + if (!action->range->hit_tbl_action) + goto free_action; + + action->range->miss_tbl_action = + mlx5dr_is_fw_table(miss_ft) ? + mlx5dr_action_create_dest_flow_fw_table(dmn, miss_ft) : + mlx5dr_action_create_dest_table(miss_ft->fs_dr_table.dr_table); + + if (!action->range->miss_tbl_action) + goto free_hit_tbl_action; + + action->range->min = min; + action->range->max = max; + action->range->dmn = dmn; + + ret = dr_action_create_range_definer(action); + if (ret) + goto free_miss_tbl_action; + + /* No need to increase refcount on domain for this action, + * the hit/miss table actions will do it internally. + */ + + return action; + +free_miss_tbl_action: + mlx5dr_action_destroy(action->range->miss_tbl_action); +free_hit_tbl_action: + mlx5dr_action_destroy(action->range->hit_tbl_action); +free_action: + kfree(action); + + return NULL; +} + struct mlx5dr_action * mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn, struct mlx5dr_action_dest *dests, @@ -1980,6 +2183,11 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action) case DR_ACTION_TYP_ASO_FLOW_METER: refcount_dec(&action->aso->dmn->refcount); break; + case DR_ACTION_TYP_RANGE: + dr_action_destroy_range_definer(action); + mlx5dr_action_destroy(action->range->miss_tbl_action); + mlx5dr_action_destroy(action->range->hit_tbl_action); + break; default: break; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_buddy.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_buddy.c index 7df11a019df9..fe228d948b47 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_buddy.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_buddy.c @@ -15,8 +15,6 @@ int mlx5dr_buddy_init(struct mlx5dr_icm_buddy_mem *buddy, buddy->max_order = max_order; INIT_LIST_HEAD(&buddy->list_node); - INIT_LIST_HEAD(&buddy->used_list); - INIT_LIST_HEAD(&buddy->hot_list); buddy->bitmap = kcalloc(buddy->max_order + 1, sizeof(*buddy->bitmap), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c index 16d65fe4f654..07b6a6dcb92f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c @@ -271,6 +271,13 @@ int mlx5dr_cmd_sync_steering(struct mlx5_core_dev *mdev) { u32 in[MLX5_ST_SZ_DW(sync_steering_in)] = {}; + /* Skip SYNC in case the device is internal error state. + * Besides a device error, this also happens when we're + * in fast teardown + */ + if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) + return 0; + MLX5_SET(sync_steering_in, in, opcode, MLX5_CMD_OP_SYNC_STEERING); return mlx5_cmd_exec_in(mdev, sync_steering, in); @@ -557,6 +564,83 @@ void mlx5dr_cmd_destroy_reformat_ctx(struct mlx5_core_dev *mdev, mlx5_cmd_exec_in(mdev, dealloc_packet_reformat_context, in); } +static void dr_cmd_set_definer_format(void *ptr, u16 format_id, + u8 *dw_selectors, + u8 *byte_selectors) +{ + if (format_id != MLX5_IFC_DEFINER_FORMAT_ID_SELECT) + return; + + MLX5_SET(match_definer, ptr, format_select_dw0, dw_selectors[0]); + MLX5_SET(match_definer, ptr, format_select_dw1, dw_selectors[1]); + MLX5_SET(match_definer, ptr, format_select_dw2, dw_selectors[2]); + MLX5_SET(match_definer, ptr, format_select_dw3, dw_selectors[3]); + MLX5_SET(match_definer, ptr, format_select_dw4, dw_selectors[4]); + MLX5_SET(match_definer, ptr, format_select_dw5, dw_selectors[5]); + MLX5_SET(match_definer, ptr, format_select_dw6, dw_selectors[6]); + MLX5_SET(match_definer, ptr, format_select_dw7, dw_selectors[7]); + MLX5_SET(match_definer, ptr, format_select_dw8, dw_selectors[8]); + + MLX5_SET(match_definer, ptr, format_select_byte0, byte_selectors[0]); + MLX5_SET(match_definer, ptr, format_select_byte1, byte_selectors[1]); + MLX5_SET(match_definer, ptr, format_select_byte2, byte_selectors[2]); + MLX5_SET(match_definer, ptr, format_select_byte3, byte_selectors[3]); + MLX5_SET(match_definer, ptr, format_select_byte4, byte_selectors[4]); + MLX5_SET(match_definer, ptr, format_select_byte5, byte_selectors[5]); + MLX5_SET(match_definer, ptr, format_select_byte6, byte_selectors[6]); + MLX5_SET(match_definer, ptr, format_select_byte7, byte_selectors[7]); +} + +int mlx5dr_cmd_create_definer(struct mlx5_core_dev *mdev, + u16 format_id, + u8 *dw_selectors, + u8 *byte_selectors, + u8 *match_mask, + u32 *definer_id) +{ + u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)] = {}; + u32 in[MLX5_ST_SZ_DW(create_match_definer_in)] = {}; + void *ptr; + int err; + + ptr = MLX5_ADDR_OF(create_match_definer_in, in, + general_obj_in_cmd_hdr); + MLX5_SET(general_obj_in_cmd_hdr, ptr, opcode, + MLX5_CMD_OP_CREATE_GENERAL_OBJECT); + MLX5_SET(general_obj_in_cmd_hdr, ptr, obj_type, + MLX5_OBJ_TYPE_MATCH_DEFINER); + + ptr = MLX5_ADDR_OF(create_match_definer_in, in, obj_context); + MLX5_SET(match_definer, ptr, format_id, format_id); + + dr_cmd_set_definer_format(ptr, format_id, + dw_selectors, byte_selectors); + + ptr = MLX5_ADDR_OF(match_definer, ptr, match_mask); + memcpy(ptr, match_mask, MLX5_FLD_SZ_BYTES(match_definer, match_mask)); + + err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out)); + if (err) + return err; + + *definer_id = MLX5_GET(general_obj_out_cmd_hdr, out, obj_id); + + return 0; +} + +void +mlx5dr_cmd_destroy_definer(struct mlx5_core_dev *mdev, u32 definer_id) +{ + u32 in[MLX5_ST_SZ_DW(general_obj_in_cmd_hdr)] = {}; + u32 out[MLX5_ST_SZ_DW(general_obj_out_cmd_hdr)]; + + MLX5_SET(general_obj_in_cmd_hdr, in, opcode, MLX5_CMD_OP_DESTROY_GENERAL_OBJECT); + MLX5_SET(general_obj_in_cmd_hdr, in, obj_type, MLX5_OBJ_TYPE_MATCH_DEFINER); + MLX5_SET(general_obj_in_cmd_hdr, in, obj_id, definer_id); + + mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out)); +} + int mlx5dr_cmd_query_gid(struct mlx5_core_dev *mdev, u8 vhca_port_num, u16 index, struct mlx5dr_cmd_gid_attr *attr) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c index 7adcf0eec13b..db81d881d38e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_dbg.c @@ -49,7 +49,8 @@ enum dr_dump_rec_type { DR_DUMP_REC_TYPE_ACTION_POP_VLAN = 3413, DR_DUMP_REC_TYPE_ACTION_SAMPLER = 3415, DR_DUMP_REC_TYPE_ACTION_INSERT_HDR = 3420, - DR_DUMP_REC_TYPE_ACTION_REMOVE_HDR = 3421 + DR_DUMP_REC_TYPE_ACTION_REMOVE_HDR = 3421, + DR_DUMP_REC_TYPE_ACTION_MATCH_RANGE = 3425, }; void mlx5dr_dbg_tbl_add(struct mlx5dr_table *tbl) @@ -107,6 +108,8 @@ dr_dump_rule_action_mem(struct seq_file *file, const u64 rule_id, { struct mlx5dr_action *action = action_mem->action; const u64 action_id = DR_DBG_PTR_TO_ID(action); + u64 hit_tbl_ptr, miss_tbl_ptr; + u32 hit_tbl_id, miss_tbl_id; switch (action->action_type) { case DR_ACTION_TYP_DROP: @@ -198,6 +201,30 @@ dr_dump_rule_action_mem(struct seq_file *file, const u64 rule_id, action->sampler->rx_icm_addr, action->sampler->tx_icm_addr); break; + case DR_ACTION_TYP_RANGE: + if (action->range->hit_tbl_action->dest_tbl->is_fw_tbl) { + hit_tbl_id = action->range->hit_tbl_action->dest_tbl->fw_tbl.id; + hit_tbl_ptr = 0; + } else { + hit_tbl_id = action->range->hit_tbl_action->dest_tbl->tbl->table_id; + hit_tbl_ptr = + DR_DBG_PTR_TO_ID(action->range->hit_tbl_action->dest_tbl->tbl); + } + + if (action->range->miss_tbl_action->dest_tbl->is_fw_tbl) { + miss_tbl_id = action->range->miss_tbl_action->dest_tbl->fw_tbl.id; + miss_tbl_ptr = 0; + } else { + miss_tbl_id = action->range->miss_tbl_action->dest_tbl->tbl->table_id; + miss_tbl_ptr = + DR_DBG_PTR_TO_ID(action->range->miss_tbl_action->dest_tbl->tbl); + } + + seq_printf(file, "%d,0x%llx,0x%llx,0x%x,0x%llx,0x%x,0x%llx,0x%x\n", + DR_DUMP_REC_TYPE_ACTION_MATCH_RANGE, action_id, rule_id, + hit_tbl_id, hit_tbl_ptr, miss_tbl_id, miss_tbl_ptr, + action->range->definer_id); + break; default: return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_definer.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_definer.c new file mode 100644 index 000000000000..d5ea97751945 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_definer.c @@ -0,0 +1,151 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +// Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. + +#include "dr_types.h" +#include "dr_ste.h" + +struct dr_definer_object { + u32 id; + u16 format_id; + u8 dw_selectors[MLX5_IFC_DEFINER_DW_SELECTORS_NUM]; + u8 byte_selectors[MLX5_IFC_DEFINER_BYTE_SELECTORS_NUM]; + u8 match_mask[DR_STE_SIZE_MATCH_TAG]; + refcount_t refcount; +}; + +static bool dr_definer_compare(struct dr_definer_object *definer, + u16 format_id, u8 *dw_selectors, + u8 *byte_selectors, u8 *match_mask) +{ + int i; + + if (definer->format_id != format_id) + return false; + + for (i = 0; i < MLX5_IFC_DEFINER_DW_SELECTORS_NUM; i++) + if (definer->dw_selectors[i] != dw_selectors[i]) + return false; + + for (i = 0; i < MLX5_IFC_DEFINER_BYTE_SELECTORS_NUM; i++) + if (definer->byte_selectors[i] != byte_selectors[i]) + return false; + + if (memcmp(definer->match_mask, match_mask, DR_STE_SIZE_MATCH_TAG)) + return false; + + return true; +} + +static struct dr_definer_object * +dr_definer_find_obj(struct mlx5dr_domain *dmn, u16 format_id, + u8 *dw_selectors, u8 *byte_selectors, u8 *match_mask) +{ + struct dr_definer_object *definer_obj; + unsigned long id; + + xa_for_each(&dmn->definers_xa, id, definer_obj) { + if (dr_definer_compare(definer_obj, format_id, + dw_selectors, byte_selectors, + match_mask)) + return definer_obj; + } + + return NULL; +} + +static struct dr_definer_object * +dr_definer_create_obj(struct mlx5dr_domain *dmn, u16 format_id, + u8 *dw_selectors, u8 *byte_selectors, u8 *match_mask) +{ + struct dr_definer_object *definer_obj; + int ret = 0; + + definer_obj = kzalloc(sizeof(*definer_obj), GFP_KERNEL); + if (!definer_obj) + return NULL; + + ret = mlx5dr_cmd_create_definer(dmn->mdev, + format_id, + dw_selectors, + byte_selectors, + match_mask, + &definer_obj->id); + if (ret) + goto err_free_definer_obj; + + /* Definer ID can have 32 bits, but STE format + * supports only definers with 8 bit IDs. + */ + if (definer_obj->id > 0xff) { + mlx5dr_err(dmn, "Unsupported definer ID (%d)\n", definer_obj->id); + goto err_destroy_definer; + } + + definer_obj->format_id = format_id; + memcpy(definer_obj->dw_selectors, dw_selectors, sizeof(definer_obj->dw_selectors)); + memcpy(definer_obj->byte_selectors, byte_selectors, sizeof(definer_obj->byte_selectors)); + memcpy(definer_obj->match_mask, match_mask, sizeof(definer_obj->match_mask)); + + refcount_set(&definer_obj->refcount, 1); + + ret = xa_insert(&dmn->definers_xa, definer_obj->id, definer_obj, GFP_KERNEL); + if (ret) { + mlx5dr_dbg(dmn, "Couldn't insert new definer into xarray (%d)\n", ret); + goto err_destroy_definer; + } + + return definer_obj; + +err_destroy_definer: + mlx5dr_cmd_destroy_definer(dmn->mdev, definer_obj->id); +err_free_definer_obj: + kfree(definer_obj); + + return NULL; +} + +static void dr_definer_destroy_obj(struct mlx5dr_domain *dmn, + struct dr_definer_object *definer_obj) +{ + mlx5dr_cmd_destroy_definer(dmn->mdev, definer_obj->id); + xa_erase(&dmn->definers_xa, definer_obj->id); + kfree(definer_obj); +} + +int mlx5dr_definer_get(struct mlx5dr_domain *dmn, u16 format_id, + u8 *dw_selectors, u8 *byte_selectors, + u8 *match_mask, u32 *definer_id) +{ + struct dr_definer_object *definer_obj; + int ret = 0; + + definer_obj = dr_definer_find_obj(dmn, format_id, dw_selectors, + byte_selectors, match_mask); + if (!definer_obj) { + definer_obj = dr_definer_create_obj(dmn, format_id, + dw_selectors, byte_selectors, + match_mask); + if (!definer_obj) + return -ENOMEM; + } else { + refcount_inc(&definer_obj->refcount); + } + + *definer_id = definer_obj->id; + + return ret; +} + +void mlx5dr_definer_put(struct mlx5dr_domain *dmn, u32 definer_id) +{ + struct dr_definer_object *definer_obj; + + definer_obj = xa_load(&dmn->definers_xa, definer_id); + if (!definer_obj) { + mlx5dr_err(dmn, "Definer ID %d not found\n", definer_id); + return; + } + + if (refcount_dec_and_test(&definer_obj->refcount)) + dr_definer_destroy_obj(dmn, definer_obj); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c index fc6ae49b5ecc..5b8bb2ca31e6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c @@ -56,6 +56,70 @@ int mlx5dr_domain_get_recalc_cs_ft_addr(struct mlx5dr_domain *dmn, return 0; } +static int dr_domain_init_mem_resources(struct mlx5dr_domain *dmn) +{ + int ret; + + dmn->chunks_kmem_cache = kmem_cache_create("mlx5_dr_chunks", + sizeof(struct mlx5dr_icm_chunk), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!dmn->chunks_kmem_cache) { + mlx5dr_err(dmn, "Couldn't create chunks kmem_cache\n"); + return -ENOMEM; + } + + dmn->htbls_kmem_cache = kmem_cache_create("mlx5_dr_htbls", + sizeof(struct mlx5dr_ste_htbl), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!dmn->htbls_kmem_cache) { + mlx5dr_err(dmn, "Couldn't create hash tables kmem_cache\n"); + ret = -ENOMEM; + goto free_chunks_kmem_cache; + } + + dmn->ste_icm_pool = mlx5dr_icm_pool_create(dmn, DR_ICM_TYPE_STE); + if (!dmn->ste_icm_pool) { + mlx5dr_err(dmn, "Couldn't get icm memory\n"); + ret = -ENOMEM; + goto free_htbls_kmem_cache; + } + + dmn->action_icm_pool = mlx5dr_icm_pool_create(dmn, DR_ICM_TYPE_MODIFY_ACTION); + if (!dmn->action_icm_pool) { + mlx5dr_err(dmn, "Couldn't get action icm memory\n"); + ret = -ENOMEM; + goto free_ste_icm_pool; + } + + ret = mlx5dr_send_info_pool_create(dmn); + if (ret) { + mlx5dr_err(dmn, "Couldn't create send info pool\n"); + goto free_action_icm_pool; + } + + return 0; + +free_action_icm_pool: + mlx5dr_icm_pool_destroy(dmn->action_icm_pool); +free_ste_icm_pool: + mlx5dr_icm_pool_destroy(dmn->ste_icm_pool); +free_htbls_kmem_cache: + kmem_cache_destroy(dmn->htbls_kmem_cache); +free_chunks_kmem_cache: + kmem_cache_destroy(dmn->chunks_kmem_cache); + + return ret; +} + +static void dr_domain_uninit_mem_resources(struct mlx5dr_domain *dmn) +{ + mlx5dr_send_info_pool_destroy(dmn); + mlx5dr_icm_pool_destroy(dmn->action_icm_pool); + mlx5dr_icm_pool_destroy(dmn->ste_icm_pool); + kmem_cache_destroy(dmn->htbls_kmem_cache); + kmem_cache_destroy(dmn->chunks_kmem_cache); +} + static int dr_domain_init_resources(struct mlx5dr_domain *dmn) { int ret; @@ -79,32 +143,22 @@ static int dr_domain_init_resources(struct mlx5dr_domain *dmn) goto clean_pd; } - dmn->ste_icm_pool = mlx5dr_icm_pool_create(dmn, DR_ICM_TYPE_STE); - if (!dmn->ste_icm_pool) { - mlx5dr_err(dmn, "Couldn't get icm memory\n"); - ret = -ENOMEM; + ret = dr_domain_init_mem_resources(dmn); + if (ret) { + mlx5dr_err(dmn, "Couldn't create domain memory resources\n"); goto clean_uar; } - dmn->action_icm_pool = mlx5dr_icm_pool_create(dmn, DR_ICM_TYPE_MODIFY_ACTION); - if (!dmn->action_icm_pool) { - mlx5dr_err(dmn, "Couldn't get action icm memory\n"); - ret = -ENOMEM; - goto free_ste_icm_pool; - } - ret = mlx5dr_send_ring_alloc(dmn); if (ret) { mlx5dr_err(dmn, "Couldn't create send-ring\n"); - goto free_action_icm_pool; + goto clean_mem_resources; } return 0; -free_action_icm_pool: - mlx5dr_icm_pool_destroy(dmn->action_icm_pool); -free_ste_icm_pool: - mlx5dr_icm_pool_destroy(dmn->ste_icm_pool); +clean_mem_resources: + dr_domain_uninit_mem_resources(dmn); clean_uar: mlx5_put_uars_page(dmn->mdev, dmn->uar); clean_pd: @@ -116,8 +170,7 @@ clean_pd: static void dr_domain_uninit_resources(struct mlx5dr_domain *dmn) { mlx5dr_send_ring_free(dmn, dmn->send_ring); - mlx5dr_icm_pool_destroy(dmn->action_icm_pool); - mlx5dr_icm_pool_destroy(dmn->ste_icm_pool); + dr_domain_uninit_mem_resources(dmn); mlx5_put_uars_page(dmn->mdev, dmn->uar); mlx5_core_dealloc_pd(dmn->mdev, dmn->pdn); } @@ -372,10 +425,11 @@ mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type) refcount_set(&dmn->refcount, 1); mutex_init(&dmn->info.rx.mutex); mutex_init(&dmn->info.tx.mutex); + xa_init(&dmn->definers_xa); if (dr_domain_caps_init(mdev, dmn)) { mlx5dr_err(dmn, "Failed init domain, no caps\n"); - goto free_domain; + goto def_xa_destroy; } dmn->info.max_log_action_icm_sz = DR_CHUNK_SIZE_4K; @@ -400,7 +454,8 @@ mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type) uninit_caps: dr_domain_caps_uninit(dmn); -free_domain: +def_xa_destroy: + xa_destroy(&dmn->definers_xa); kfree(dmn); return NULL; } @@ -440,6 +495,7 @@ int mlx5dr_domain_destroy(struct mlx5dr_domain *dmn) dr_domain_uninit_csum_recalc_fts(dmn); dr_domain_uninit_resources(dmn); dr_domain_caps_uninit(dmn); + xa_destroy(&dmn->definers_xa); mutex_destroy(&dmn->info.tx.mutex); mutex_destroy(&dmn->info.rx.mutex); kfree(dmn); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c index 4ca67fa24cc6..3eb6719bc8eb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_icm_pool.c @@ -4,14 +4,30 @@ #include "dr_types.h" #define DR_ICM_MODIFY_HDR_ALIGN_BASE 64 +#define DR_ICM_POOL_HOT_MEMORY_FRACTION 4 + +struct mlx5dr_icm_hot_chunk { + struct mlx5dr_icm_buddy_mem *buddy_mem; + unsigned int seg; + enum mlx5dr_icm_chunk_size size; +}; struct mlx5dr_icm_pool { enum mlx5dr_icm_type icm_type; enum mlx5dr_icm_chunk_size max_log_chunk_sz; struct mlx5dr_domain *dmn; + struct kmem_cache *chunks_kmem_cache; + /* memory management */ struct mutex mutex; /* protect the ICM pool and ICM buddy */ struct list_head buddy_mem_list; + + /* Hardware may be accessing this memory but at some future, + * undetermined time, it might cease to do so. + * sync_ste command sets them free. + */ + struct mlx5dr_icm_hot_chunk *hot_chunks_arr; + u32 hot_chunks_num; u64 hot_memory_size; }; @@ -177,46 +193,20 @@ static int dr_icm_buddy_get_ste_size(struct mlx5dr_icm_buddy_mem *buddy) static void dr_icm_chunk_ste_init(struct mlx5dr_icm_chunk *chunk, int offset) { + int num_of_entries = mlx5dr_icm_pool_get_chunk_num_of_entries(chunk); struct mlx5dr_icm_buddy_mem *buddy = chunk->buddy_mem; + int ste_size = dr_icm_buddy_get_ste_size(buddy); int index = offset / DR_STE_SIZE; chunk->ste_arr = &buddy->ste_arr[index]; chunk->miss_list = &buddy->miss_list[index]; - chunk->hw_ste_arr = buddy->hw_ste_arr + - index * dr_icm_buddy_get_ste_size(buddy); -} + chunk->hw_ste_arr = buddy->hw_ste_arr + index * ste_size; -static void dr_icm_chunk_ste_cleanup(struct mlx5dr_icm_chunk *chunk) -{ - int num_of_entries = mlx5dr_icm_pool_get_chunk_num_of_entries(chunk); - struct mlx5dr_icm_buddy_mem *buddy = chunk->buddy_mem; - - memset(chunk->hw_ste_arr, 0, - num_of_entries * dr_icm_buddy_get_ste_size(buddy)); + memset(chunk->hw_ste_arr, 0, num_of_entries * ste_size); memset(chunk->ste_arr, 0, num_of_entries * sizeof(chunk->ste_arr[0])); } -static enum mlx5dr_icm_type -get_chunk_icm_type(struct mlx5dr_icm_chunk *chunk) -{ - return chunk->buddy_mem->pool->icm_type; -} - -static void dr_icm_chunk_destroy(struct mlx5dr_icm_chunk *chunk, - struct mlx5dr_icm_buddy_mem *buddy) -{ - enum mlx5dr_icm_type icm_type = get_chunk_icm_type(chunk); - - buddy->used_memory -= mlx5dr_icm_pool_get_chunk_byte_size(chunk); - list_del(&chunk->chunk_list); - - if (icm_type == DR_ICM_TYPE_STE) - dr_icm_chunk_ste_cleanup(chunk); - - kvfree(chunk); -} - static int dr_icm_buddy_init_ste_cache(struct mlx5dr_icm_buddy_mem *buddy) { int num_of_entries = @@ -296,14 +286,6 @@ free_mr: static void dr_icm_buddy_destroy(struct mlx5dr_icm_buddy_mem *buddy) { - struct mlx5dr_icm_chunk *chunk, *next; - - list_for_each_entry_safe(chunk, next, &buddy->hot_list, chunk_list) - dr_icm_chunk_destroy(chunk, buddy); - - list_for_each_entry_safe(chunk, next, &buddy->used_list, chunk_list) - dr_icm_chunk_destroy(chunk, buddy); - dr_icm_pool_mr_destroy(buddy->icm_mr); mlx5dr_buddy_cleanup(buddy); @@ -314,53 +296,62 @@ static void dr_icm_buddy_destroy(struct mlx5dr_icm_buddy_mem *buddy) kvfree(buddy); } -static struct mlx5dr_icm_chunk * -dr_icm_chunk_create(struct mlx5dr_icm_pool *pool, - enum mlx5dr_icm_chunk_size chunk_size, - struct mlx5dr_icm_buddy_mem *buddy_mem_pool, - unsigned int seg) +static void +dr_icm_chunk_init(struct mlx5dr_icm_chunk *chunk, + struct mlx5dr_icm_pool *pool, + enum mlx5dr_icm_chunk_size chunk_size, + struct mlx5dr_icm_buddy_mem *buddy_mem_pool, + unsigned int seg) { - struct mlx5dr_icm_chunk *chunk; int offset; - chunk = kvzalloc(sizeof(*chunk), GFP_KERNEL); - if (!chunk) - return NULL; - - offset = mlx5dr_icm_pool_dm_type_to_entry_size(pool->icm_type) * seg; - chunk->seg = seg; chunk->size = chunk_size; chunk->buddy_mem = buddy_mem_pool; - if (pool->icm_type == DR_ICM_TYPE_STE) + if (pool->icm_type == DR_ICM_TYPE_STE) { + offset = mlx5dr_icm_pool_dm_type_to_entry_size(pool->icm_type) * seg; dr_icm_chunk_ste_init(chunk, offset); + } buddy_mem_pool->used_memory += mlx5dr_icm_pool_get_chunk_byte_size(chunk); - INIT_LIST_HEAD(&chunk->chunk_list); - - /* chunk now is part of the used_list */ - list_add_tail(&chunk->chunk_list, &buddy_mem_pool->used_list); - - return chunk; } static bool dr_icm_pool_is_sync_required(struct mlx5dr_icm_pool *pool) { int allow_hot_size; - /* sync when hot memory reaches half of the pool size */ + /* sync when hot memory reaches a certain fraction of the pool size */ allow_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz, - pool->icm_type) / 2; + pool->icm_type) / + DR_ICM_POOL_HOT_MEMORY_FRACTION; return pool->hot_memory_size > allow_hot_size; } +static void dr_icm_pool_clear_hot_chunks_arr(struct mlx5dr_icm_pool *pool) +{ + struct mlx5dr_icm_hot_chunk *hot_chunk; + u32 i, num_entries; + + for (i = 0; i < pool->hot_chunks_num; i++) { + hot_chunk = &pool->hot_chunks_arr[i]; + num_entries = mlx5dr_icm_pool_chunk_size_to_entries(hot_chunk->size); + mlx5dr_buddy_free_mem(hot_chunk->buddy_mem, + hot_chunk->seg, ilog2(num_entries)); + hot_chunk->buddy_mem->used_memory -= + mlx5dr_icm_pool_chunk_size_to_byte(hot_chunk->size, + pool->icm_type); + } + + pool->hot_chunks_num = 0; + pool->hot_memory_size = 0; +} + static int dr_icm_pool_sync_all_buddy_pools(struct mlx5dr_icm_pool *pool) { struct mlx5dr_icm_buddy_mem *buddy, *tmp_buddy; - u32 num_entries; int err; err = mlx5dr_cmd_sync_steering(pool->dmn->mdev); @@ -369,16 +360,9 @@ static int dr_icm_pool_sync_all_buddy_pools(struct mlx5dr_icm_pool *pool) return err; } - list_for_each_entry_safe(buddy, tmp_buddy, &pool->buddy_mem_list, list_node) { - struct mlx5dr_icm_chunk *chunk, *tmp_chunk; - - list_for_each_entry_safe(chunk, tmp_chunk, &buddy->hot_list, chunk_list) { - num_entries = mlx5dr_icm_pool_get_chunk_num_of_entries(chunk); - mlx5dr_buddy_free_mem(buddy, chunk->seg, ilog2(num_entries)); - pool->hot_memory_size -= mlx5dr_icm_pool_get_chunk_byte_size(chunk); - dr_icm_chunk_destroy(chunk, buddy); - } + dr_icm_pool_clear_hot_chunks_arr(pool); + list_for_each_entry_safe(buddy, tmp_buddy, &pool->buddy_mem_list, list_node) { if (!buddy->used_memory && pool->icm_type == DR_ICM_TYPE_STE) dr_icm_buddy_destroy(buddy); } @@ -452,10 +436,12 @@ mlx5dr_icm_alloc_chunk(struct mlx5dr_icm_pool *pool, if (ret) goto out; - chunk = dr_icm_chunk_create(pool, chunk_size, buddy, seg); + chunk = kmem_cache_alloc(pool->chunks_kmem_cache, GFP_KERNEL); if (!chunk) goto out_err; + dr_icm_chunk_init(chunk, pool, chunk_size, buddy, seg); + goto out; out_err: @@ -469,12 +455,23 @@ void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk) { struct mlx5dr_icm_buddy_mem *buddy = chunk->buddy_mem; struct mlx5dr_icm_pool *pool = buddy->pool; + struct mlx5dr_icm_hot_chunk *hot_chunk; + struct kmem_cache *chunks_cache; + + chunks_cache = pool->chunks_kmem_cache; - /* move the memory to the waiting list AKA "hot" */ + /* move the chunk to the waiting chunks array, AKA "hot" memory */ mutex_lock(&pool->mutex); - list_move_tail(&chunk->chunk_list, &buddy->hot_list); + pool->hot_memory_size += mlx5dr_icm_pool_get_chunk_byte_size(chunk); + hot_chunk = &pool->hot_chunks_arr[pool->hot_chunks_num++]; + hot_chunk->buddy_mem = chunk->buddy_mem; + hot_chunk->seg = chunk->seg; + hot_chunk->size = chunk->size; + + kmem_cache_free(chunks_cache, chunk); + /* Check if we have chunks that are waiting for sync-ste */ if (dr_icm_pool_is_sync_required(pool)) dr_icm_pool_sync_all_buddy_pools(pool); @@ -482,9 +479,20 @@ void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk) mutex_unlock(&pool->mutex); } +struct mlx5dr_ste_htbl *mlx5dr_icm_pool_alloc_htbl(struct mlx5dr_icm_pool *pool) +{ + return kmem_cache_alloc(pool->dmn->htbls_kmem_cache, GFP_KERNEL); +} + +void mlx5dr_icm_pool_free_htbl(struct mlx5dr_icm_pool *pool, struct mlx5dr_ste_htbl *htbl) +{ + kmem_cache_free(pool->dmn->htbls_kmem_cache, htbl); +} + struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn, enum mlx5dr_icm_type icm_type) { + u32 num_of_chunks, entry_size, max_hot_size; enum mlx5dr_icm_chunk_size max_log_chunk_sz; struct mlx5dr_icm_pool *pool; @@ -500,21 +508,43 @@ struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn, pool->dmn = dmn; pool->icm_type = icm_type; pool->max_log_chunk_sz = max_log_chunk_sz; + pool->chunks_kmem_cache = dmn->chunks_kmem_cache; INIT_LIST_HEAD(&pool->buddy_mem_list); mutex_init(&pool->mutex); + entry_size = mlx5dr_icm_pool_dm_type_to_entry_size(pool->icm_type); + + max_hot_size = mlx5dr_icm_pool_chunk_size_to_byte(pool->max_log_chunk_sz, + pool->icm_type) / + DR_ICM_POOL_HOT_MEMORY_FRACTION; + + num_of_chunks = DIV_ROUND_UP(max_hot_size, entry_size) + 1; + + pool->hot_chunks_arr = kvcalloc(num_of_chunks, + sizeof(struct mlx5dr_icm_hot_chunk), + GFP_KERNEL); + if (!pool->hot_chunks_arr) + goto free_pool; + return pool; + +free_pool: + kvfree(pool); + return NULL; } void mlx5dr_icm_pool_destroy(struct mlx5dr_icm_pool *pool) { struct mlx5dr_icm_buddy_mem *buddy, *tmp_buddy; + dr_icm_pool_clear_hot_chunks_arr(pool); + list_for_each_entry_safe(buddy, tmp_buddy, &pool->buddy_mem_list, list_node) dr_icm_buddy_destroy(buddy); + kvfree(pool->hot_chunks_arr); mutex_destroy(&pool->mutex); kvfree(pool); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c index 91ff19f67695..74cbe53ee9db 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c @@ -3,13 +3,16 @@ #include "dr_types.h" -#define DR_RULE_MAX_STE_CHAIN (DR_RULE_MAX_STES + DR_ACTION_MAX_STES) +#define DR_RULE_MAX_STES_OPTIMIZED 5 +#define DR_RULE_MAX_STE_CHAIN_OPTIMIZED (DR_RULE_MAX_STES_OPTIMIZED + DR_ACTION_MAX_STES) -static int dr_rule_append_to_miss_list(struct mlx5dr_ste_ctx *ste_ctx, +static int dr_rule_append_to_miss_list(struct mlx5dr_domain *dmn, + enum mlx5dr_domain_nic_type nic_type, struct mlx5dr_ste *new_last_ste, struct list_head *miss_list, struct list_head *send_list) { + struct mlx5dr_ste_ctx *ste_ctx = dmn->ste_ctx; struct mlx5dr_ste_send_info *ste_info_last; struct mlx5dr_ste *last_ste; @@ -17,7 +20,7 @@ static int dr_rule_append_to_miss_list(struct mlx5dr_ste_ctx *ste_ctx, last_ste = list_last_entry(miss_list, struct mlx5dr_ste, miss_list_node); WARN_ON(!last_ste); - ste_info_last = kzalloc(sizeof(*ste_info_last), GFP_KERNEL); + ste_info_last = mlx5dr_send_info_alloc(dmn, nic_type); if (!ste_info_last) return -ENOMEM; @@ -32,16 +35,28 @@ static int dr_rule_append_to_miss_list(struct mlx5dr_ste_ctx *ste_ctx, return 0; } +static void dr_rule_set_last_ste_miss_addr(struct mlx5dr_matcher *matcher, + struct mlx5dr_matcher_rx_tx *nic_matcher, + u8 *hw_ste) +{ + struct mlx5dr_ste_ctx *ste_ctx = matcher->tbl->dmn->ste_ctx; + u64 icm_addr; + + if (mlx5dr_ste_is_miss_addr_set(ste_ctx, hw_ste)) + return; + + icm_addr = mlx5dr_icm_pool_get_chunk_icm_addr(nic_matcher->e_anchor->chunk); + mlx5dr_ste_set_miss_addr(ste_ctx, hw_ste, icm_addr); +} + static struct mlx5dr_ste * dr_rule_create_collision_htbl(struct mlx5dr_matcher *matcher, struct mlx5dr_matcher_rx_tx *nic_matcher, u8 *hw_ste) { struct mlx5dr_domain *dmn = matcher->tbl->dmn; - struct mlx5dr_ste_ctx *ste_ctx = dmn->ste_ctx; struct mlx5dr_ste_htbl *new_htbl; struct mlx5dr_ste *ste; - u64 icm_addr; /* Create new table for miss entry */ new_htbl = mlx5dr_ste_htbl_alloc(dmn->ste_icm_pool, @@ -55,8 +70,7 @@ dr_rule_create_collision_htbl(struct mlx5dr_matcher *matcher, /* One and only entry, never grows */ ste = new_htbl->chunk->ste_arr; - icm_addr = mlx5dr_icm_pool_get_chunk_icm_addr(nic_matcher->e_anchor->chunk); - mlx5dr_ste_set_miss_addr(ste_ctx, hw_ste, icm_addr); + dr_rule_set_last_ste_miss_addr(matcher, nic_matcher, hw_ste); mlx5dr_htbl_get(new_htbl); return ste; @@ -120,7 +134,7 @@ dr_rule_handle_one_ste_in_update_list(struct mlx5dr_ste_send_info *ste_info, goto out; out: - kfree(ste_info); + mlx5dr_send_info_free(ste_info); return ret; } @@ -191,8 +205,8 @@ dr_rule_rehash_handle_collision(struct mlx5dr_matcher *matcher, new_ste->htbl->chunk->miss_list = mlx5dr_ste_get_miss_list(col_ste); /* Update the previous from the list */ - ret = dr_rule_append_to_miss_list(dmn->ste_ctx, new_ste, - mlx5dr_ste_get_miss_list(col_ste), + ret = dr_rule_append_to_miss_list(dmn, nic_matcher->nic_tbl->nic_dmn->type, + new_ste, mlx5dr_ste_get_miss_list(col_ste), update_list); if (ret) { mlx5dr_dbg(dmn, "Failed update dup entry\n"); @@ -238,7 +252,6 @@ dr_rule_rehash_copy_ste(struct mlx5dr_matcher *matcher, bool use_update_list = false; u8 hw_ste[DR_STE_SIZE] = {}; struct mlx5dr_ste *new_ste; - u64 icm_addr; int new_idx; u8 sb_idx; @@ -247,9 +260,8 @@ dr_rule_rehash_copy_ste(struct mlx5dr_matcher *matcher, mlx5dr_ste_set_bit_mask(hw_ste, nic_matcher->ste_builder[sb_idx].bit_mask); /* Copy STE control and tag */ - icm_addr = mlx5dr_icm_pool_get_chunk_icm_addr(nic_matcher->e_anchor->chunk); memcpy(hw_ste, mlx5dr_ste_get_hw_ste(cur_ste), DR_STE_SIZE_REDUCED); - mlx5dr_ste_set_miss_addr(dmn->ste_ctx, hw_ste, icm_addr); + dr_rule_set_last_ste_miss_addr(matcher, nic_matcher, hw_ste); new_idx = mlx5dr_ste_calc_hash_index(hw_ste, new_htbl); new_ste = &new_htbl->chunk->ste_arr[new_idx]; @@ -278,7 +290,8 @@ dr_rule_rehash_copy_ste(struct mlx5dr_matcher *matcher, new_htbl->ctrl.num_of_valid_entries++; if (use_update_list) { - ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL); + ste_info = mlx5dr_send_info_alloc(dmn, + nic_matcher->nic_tbl->nic_dmn->type); if (!ste_info) goto err_exit; @@ -357,6 +370,15 @@ static int dr_rule_rehash_copy_htbl(struct mlx5dr_matcher *matcher, update_list); if (err) goto clean_copy; + + /* In order to decrease the number of allocated ste_send_info + * structs, send the current table row now. + */ + err = dr_rule_send_update_list(update_list, matcher->tbl->dmn, false); + if (err) { + mlx5dr_dbg(matcher->tbl->dmn, "Failed updating table to HW\n"); + goto clean_copy; + } } clean_copy: @@ -387,7 +409,8 @@ dr_rule_rehash_htbl(struct mlx5dr_rule *rule, nic_matcher = nic_rule->nic_matcher; nic_dmn = nic_matcher->nic_tbl->nic_dmn; - ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL); + ste_info = mlx5dr_send_info_alloc(dmn, + nic_matcher->nic_tbl->nic_dmn->type); if (!ste_info) return NULL; @@ -473,13 +496,13 @@ free_ste_list: list_for_each_entry_safe(del_ste_info, tmp_ste_info, &rehash_table_send_list, send_list) { list_del(&del_ste_info->send_list); - kfree(del_ste_info); + mlx5dr_send_info_free(del_ste_info); } free_new_htbl: mlx5dr_ste_htbl_free(new_htbl); free_ste_info: - kfree(ste_info); + mlx5dr_send_info_free(ste_info); mlx5dr_info(dmn, "Failed creating rehash table\n"); return NULL; } @@ -512,11 +535,11 @@ dr_rule_handle_collision(struct mlx5dr_matcher *matcher, struct list_head *send_list) { struct mlx5dr_domain *dmn = matcher->tbl->dmn; - struct mlx5dr_ste_ctx *ste_ctx = dmn->ste_ctx; struct mlx5dr_ste_send_info *ste_info; struct mlx5dr_ste *new_ste; - ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL); + ste_info = mlx5dr_send_info_alloc(dmn, + nic_matcher->nic_tbl->nic_dmn->type); if (!ste_info) return NULL; @@ -524,8 +547,8 @@ dr_rule_handle_collision(struct mlx5dr_matcher *matcher, if (!new_ste) goto free_send_info; - if (dr_rule_append_to_miss_list(ste_ctx, new_ste, - miss_list, send_list)) { + if (dr_rule_append_to_miss_list(dmn, nic_matcher->nic_tbl->nic_dmn->type, + new_ste, miss_list, send_list)) { mlx5dr_dbg(dmn, "Failed to update prev miss_list\n"); goto err_exit; } @@ -541,7 +564,7 @@ dr_rule_handle_collision(struct mlx5dr_matcher *matcher, err_exit: mlx5dr_ste_free(new_ste, matcher, nic_matcher); free_send_info: - kfree(ste_info); + mlx5dr_send_info_free(ste_info); return NULL; } @@ -721,8 +744,8 @@ static int dr_rule_handle_action_stes(struct mlx5dr_rule *rule, list_add_tail(&action_ste->miss_list_node, mlx5dr_ste_get_miss_list(action_ste)); - ste_info_arr[k] = kzalloc(sizeof(*ste_info_arr[k]), - GFP_KERNEL); + ste_info_arr[k] = mlx5dr_send_info_alloc(dmn, + nic_matcher->nic_tbl->nic_dmn->type); if (!ste_info_arr[k]) goto err_exit; @@ -759,7 +782,6 @@ static int dr_rule_handle_empty_entry(struct mlx5dr_matcher *matcher, { struct mlx5dr_domain *dmn = matcher->tbl->dmn; struct mlx5dr_ste_send_info *ste_info; - u64 icm_addr; /* Take ref on table, only on first time this ste is used */ mlx5dr_htbl_get(cur_htbl); @@ -767,12 +789,12 @@ static int dr_rule_handle_empty_entry(struct mlx5dr_matcher *matcher, /* new entry -> new branch */ list_add_tail(&ste->miss_list_node, miss_list); - icm_addr = mlx5dr_icm_pool_get_chunk_icm_addr(nic_matcher->e_anchor->chunk); - mlx5dr_ste_set_miss_addr(dmn->ste_ctx, hw_ste, icm_addr); + dr_rule_set_last_ste_miss_addr(matcher, nic_matcher, hw_ste); ste->ste_chain_location = ste_location; - ste_info = kzalloc(sizeof(*ste_info), GFP_KERNEL); + ste_info = mlx5dr_send_info_alloc(dmn, + nic_matcher->nic_tbl->nic_dmn->type); if (!ste_info) goto clean_ste_setting; @@ -793,7 +815,7 @@ static int dr_rule_handle_empty_entry(struct mlx5dr_matcher *matcher, return 0; clean_ste_info: - kfree(ste_info); + mlx5dr_send_info_free(ste_info); clean_ste_setting: list_del_init(&ste->miss_list_node); mlx5dr_htbl_put(cur_htbl); @@ -1089,6 +1111,7 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, size_t num_actions, struct mlx5dr_action *actions[]) { + u8 hw_ste_arr_optimized[DR_RULE_MAX_STE_CHAIN_OPTIMIZED * DR_STE_SIZE] = {}; struct mlx5dr_ste_send_info *ste_info, *tmp_ste_info; struct mlx5dr_matcher *matcher = rule->matcher; struct mlx5dr_domain *dmn = matcher->tbl->dmn; @@ -1098,6 +1121,7 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, struct mlx5dr_ste_htbl *cur_htbl; struct mlx5dr_ste *ste = NULL; LIST_HEAD(send_ste_list); + bool hw_ste_arr_is_opt; u8 *hw_ste_arr = NULL; u32 new_hw_ste_arr_sz; int ret, i; @@ -1109,9 +1133,23 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, rule->flow_source)) return 0; - hw_ste_arr = kzalloc(DR_RULE_MAX_STE_CHAIN * DR_STE_SIZE, GFP_KERNEL); - if (!hw_ste_arr) - return -ENOMEM; + ret = mlx5dr_matcher_select_builders(matcher, + nic_matcher, + dr_rule_get_ipv(¶m->outer), + dr_rule_get_ipv(¶m->inner)); + if (ret) + return ret; + + hw_ste_arr_is_opt = nic_matcher->num_of_builders <= DR_RULE_MAX_STES_OPTIMIZED; + if (likely(hw_ste_arr_is_opt)) { + hw_ste_arr = hw_ste_arr_optimized; + } else { + hw_ste_arr = kzalloc((nic_matcher->num_of_builders + DR_ACTION_MAX_STES) * + DR_STE_SIZE, GFP_KERNEL); + + if (!hw_ste_arr) + return -ENOMEM; + } mlx5dr_domain_nic_lock(nic_dmn); @@ -1119,13 +1157,6 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, if (ret) goto free_hw_ste; - ret = mlx5dr_matcher_select_builders(matcher, - nic_matcher, - dr_rule_get_ipv(¶m->outer), - dr_rule_get_ipv(¶m->inner)); - if (ret) - goto remove_from_nic_tbl; - /* Set the tag values inside the ste array */ ret = mlx5dr_ste_build_ste_arr(matcher, nic_matcher, param, hw_ste_arr); if (ret) @@ -1187,7 +1218,8 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, mlx5dr_domain_nic_unlock(nic_dmn); - kfree(hw_ste_arr); + if (unlikely(!hw_ste_arr_is_opt)) + kfree(hw_ste_arr); return 0; @@ -1196,7 +1228,7 @@ free_rule: /* Clean all ste_info's */ list_for_each_entry_safe(ste_info, tmp_ste_info, &send_ste_list, send_list) { list_del(&ste_info->send_list); - kfree(ste_info); + mlx5dr_send_info_free(ste_info); } remove_from_nic_tbl: @@ -1205,7 +1237,10 @@ remove_from_nic_tbl: free_hw_ste: mlx5dr_domain_nic_unlock(nic_dmn); - kfree(hw_ste_arr); + + if (unlikely(!hw_ste_arr_is_opt)) + kfree(hw_ste_arr); + return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c index ef19a66f5233..a4476cb4c3b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c @@ -7,6 +7,7 @@ #define QUEUE_SIZE 128 #define SIGNAL_PER_DIV_QUEUE 16 #define TH_NUMS_TO_DRAIN 2 +#define DR_SEND_INFO_POOL_SIZE 1000 enum { CQ_OK = 0, CQ_EMPTY = -1, CQ_POLL_ERR = -2 }; @@ -49,6 +50,136 @@ struct dr_qp_init_attr { u8 isolate_vl_tc:1; }; +struct mlx5dr_send_info_pool_obj { + struct mlx5dr_ste_send_info ste_send_info; + struct mlx5dr_send_info_pool *pool; + struct list_head list_node; +}; + +struct mlx5dr_send_info_pool { + struct list_head free_list; +}; + +static int dr_send_info_pool_fill(struct mlx5dr_send_info_pool *pool) +{ + struct mlx5dr_send_info_pool_obj *pool_obj, *tmp_pool_obj; + int i; + + for (i = 0; i < DR_SEND_INFO_POOL_SIZE; i++) { + pool_obj = kzalloc(sizeof(*pool_obj), GFP_KERNEL); + if (!pool_obj) + goto clean_pool; + + pool_obj->pool = pool; + list_add_tail(&pool_obj->list_node, &pool->free_list); + } + + return 0; + +clean_pool: + list_for_each_entry_safe(pool_obj, tmp_pool_obj, &pool->free_list, list_node) { + list_del(&pool_obj->list_node); + kfree(pool_obj); + } + + return -ENOMEM; +} + +static void dr_send_info_pool_destroy(struct mlx5dr_send_info_pool *pool) +{ + struct mlx5dr_send_info_pool_obj *pool_obj, *tmp_pool_obj; + + list_for_each_entry_safe(pool_obj, tmp_pool_obj, &pool->free_list, list_node) { + list_del(&pool_obj->list_node); + kfree(pool_obj); + } + + kfree(pool); +} + +void mlx5dr_send_info_pool_destroy(struct mlx5dr_domain *dmn) +{ + dr_send_info_pool_destroy(dmn->send_info_pool_tx); + dr_send_info_pool_destroy(dmn->send_info_pool_rx); +} + +static struct mlx5dr_send_info_pool *dr_send_info_pool_create(void) +{ + struct mlx5dr_send_info_pool *pool; + int ret; + + pool = kzalloc(sizeof(*pool), GFP_KERNEL); + if (!pool) + return NULL; + + INIT_LIST_HEAD(&pool->free_list); + + ret = dr_send_info_pool_fill(pool); + if (ret) { + kfree(pool); + return NULL; + } + + return pool; +} + +int mlx5dr_send_info_pool_create(struct mlx5dr_domain *dmn) +{ + dmn->send_info_pool_rx = dr_send_info_pool_create(); + if (!dmn->send_info_pool_rx) + return -ENOMEM; + + dmn->send_info_pool_tx = dr_send_info_pool_create(); + if (!dmn->send_info_pool_tx) { + dr_send_info_pool_destroy(dmn->send_info_pool_rx); + return -ENOMEM; + } + + return 0; +} + +struct mlx5dr_ste_send_info +*mlx5dr_send_info_alloc(struct mlx5dr_domain *dmn, + enum mlx5dr_domain_nic_type nic_type) +{ + struct mlx5dr_send_info_pool_obj *pool_obj; + struct mlx5dr_send_info_pool *pool; + int ret; + + pool = nic_type == DR_DOMAIN_NIC_TYPE_RX ? dmn->send_info_pool_rx : + dmn->send_info_pool_tx; + + if (unlikely(list_empty(&pool->free_list))) { + ret = dr_send_info_pool_fill(pool); + if (ret) + return NULL; + } + + pool_obj = list_first_entry_or_null(&pool->free_list, + struct mlx5dr_send_info_pool_obj, + list_node); + + if (likely(pool_obj)) { + list_del_init(&pool_obj->list_node); + } else { + WARN_ONCE(!pool_obj, "Failed getting ste send info obj from pool"); + return NULL; + } + + return &pool_obj->ste_send_info; +} + +void mlx5dr_send_info_free(struct mlx5dr_ste_send_info *ste_send_info) +{ + struct mlx5dr_send_info_pool_obj *pool_obj; + + pool_obj = container_of(ste_send_info, + struct mlx5dr_send_info_pool_obj, + ste_send_info); + + list_add(&pool_obj->list_node, &pool_obj->pool->free_list); +} + static int dr_parse_cqe(struct mlx5dr_cq *dr_cq, struct mlx5_cqe64 *cqe64) { unsigned int idx; @@ -78,8 +209,15 @@ static int dr_cq_poll_one(struct mlx5dr_cq *dr_cq) int err; cqe64 = mlx5_cqwq_get_cqe(&dr_cq->wq); - if (!cqe64) + if (!cqe64) { + if (unlikely(dr_cq->mdev->state == + MLX5_DEVICE_STATE_INTERNAL_ERROR)) { + mlx5_core_dbg_once(dr_cq->mdev, + "Polling CQ while device is shutting down\n"); + return CQ_POLL_ERR; + } return CQ_EMPTY; + } mlx5_cqwq_pop(&dr_cq->wq); err = dr_parse_cqe(dr_cq, cqe64); @@ -833,6 +971,7 @@ static struct mlx5dr_cq *dr_create_cq(struct mlx5_core_dev *mdev, cq->mcq.vector = 0; cq->mcq.uar = uar; + cq->mdev = mdev; return cq; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c index 09ebd3088857..1e15f605df6e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c @@ -90,6 +90,16 @@ static void dr_ste_set_always_miss(struct dr_hw_ste_format *hw_ste) hw_ste->mask[0] = 0; } +bool mlx5dr_ste_is_miss_addr_set(struct mlx5dr_ste_ctx *ste_ctx, + u8 *hw_ste_p) +{ + if (!ste_ctx->is_miss_addr_set) + return false; + + /* check if miss address is already set for this type of STE */ + return ste_ctx->is_miss_addr_set(hw_ste_p); +} + void mlx5dr_ste_set_miss_addr(struct mlx5dr_ste_ctx *ste_ctx, u8 *hw_ste_p, u64 miss_addr) { @@ -491,7 +501,7 @@ struct mlx5dr_ste_htbl *mlx5dr_ste_htbl_alloc(struct mlx5dr_icm_pool *pool, u32 num_entries; int i; - htbl = kzalloc(sizeof(*htbl), GFP_KERNEL); + htbl = mlx5dr_icm_pool_alloc_htbl(pool); if (!htbl) return NULL; @@ -503,6 +513,9 @@ struct mlx5dr_ste_htbl *mlx5dr_ste_htbl_alloc(struct mlx5dr_icm_pool *pool, htbl->lu_type = lu_type; htbl->byte_mask = byte_mask; htbl->refcount = 0; + htbl->pointing_ste = NULL; + htbl->ctrl.num_of_valid_entries = 0; + htbl->ctrl.num_of_collisions = 0; num_entries = mlx5dr_icm_pool_get_chunk_num_of_entries(chunk); for (i = 0; i < num_entries; i++) { @@ -517,17 +530,20 @@ struct mlx5dr_ste_htbl *mlx5dr_ste_htbl_alloc(struct mlx5dr_icm_pool *pool, return htbl; out_free_htbl: - kfree(htbl); + mlx5dr_icm_pool_free_htbl(pool, htbl); return NULL; } int mlx5dr_ste_htbl_free(struct mlx5dr_ste_htbl *htbl) { + struct mlx5dr_icm_pool *pool = htbl->chunk->buddy_mem->pool; + if (htbl->refcount) return -EBUSY; mlx5dr_icm_free_chunk(htbl->chunk); - kfree(htbl); + mlx5dr_icm_pool_free_htbl(pool, htbl); + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h index 17513baff9b0..7075142bcfb6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.h @@ -151,6 +151,7 @@ struct mlx5dr_ste_ctx { bool is_rx, u16 gvmi); void (*set_next_lu_type)(u8 *hw_ste_p, u16 lu_type); u16 (*get_next_lu_type)(u8 *hw_ste_p); + bool (*is_miss_addr_set)(u8 *hw_ste_p); void (*set_miss_addr)(u8 *hw_ste_p, u64 miss_addr); u64 (*get_miss_addr)(u8 *hw_ste_p); void (*set_hit_addr)(u8 *hw_ste_p, u64 icm_addr, u32 ht_size); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c index ee677a5c76be..084145f18084 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.c @@ -13,6 +13,7 @@ enum dr_ste_v1_entry_format { DR_STE_V1_TYPE_BWC_BYTE = 0x0, DR_STE_V1_TYPE_BWC_DW = 0x1, DR_STE_V1_TYPE_MATCH = 0x2, + DR_STE_V1_TYPE_MATCH_RANGES = 0x7, }; /* Lookup type is built from 2B: [ Definer mode 1B ][ Definer index 1B ] */ @@ -267,6 +268,16 @@ static void dr_ste_v1_set_entry_type(u8 *hw_ste_p, u8 entry_type) MLX5_SET(ste_match_bwc_v1, hw_ste_p, entry_format, entry_type); } +bool dr_ste_v1_is_miss_addr_set(u8 *hw_ste_p) +{ + u8 entry_type = MLX5_GET(ste_match_bwc_v1, hw_ste_p, entry_format); + + /* unlike MATCH STE, for MATCH_RANGES STE both hit and miss addresses + * are part of the action, so they both set as part of STE init + */ + return entry_type == DR_STE_V1_TYPE_MATCH_RANGES; +} + void dr_ste_v1_set_miss_addr(u8 *hw_ste_p, u64 miss_addr) { u64 index = miss_addr >> 6; @@ -520,6 +531,27 @@ static void dr_ste_v1_set_aso_flow_meter(u8 *d_action, init_color); } +static void dr_ste_v1_set_match_range_pkt_len(u8 *hw_ste_p, u32 definer_id, + u32 min, u32 max) +{ + MLX5_SET(ste_match_ranges_v1, hw_ste_p, match_definer_ctx_idx, definer_id); + + /* When the STE will be sent, its mask and tags will be swapped in + * dr_ste_v1_prepare_for_postsend(). This, however, is match range STE + * which doesn't have mask, and shouldn't have mask/tag swapped. + * We're using the common utilities functions to send this STE, so need + * to allow for this swapping - place the values in the corresponding + * locations to allow flipping them when writing to ICM. + * + * min/max_value_2 corresponds to match_dw_0 in its definer. + * To allow mask/tag swapping, writing the min/max_2 to min/max_0. + * + * Pkt len is 2 bytes that are stored in the higher section of the DW. + */ + MLX5_SET(ste_match_ranges_v1, hw_ste_p, min_value_0, min << 16); + MLX5_SET(ste_match_ranges_v1, hw_ste_p, max_value_0, max << 16); +} + static void dr_ste_v1_arr_init_next_match(u8 **last_ste, u32 *added_stes, u16 gvmi) @@ -535,6 +567,14 @@ static void dr_ste_v1_arr_init_next_match(u8 **last_ste, memset(action, 0, MLX5_FLD_SZ_BYTES(ste_mask_and_match_v1, action)); } +static void dr_ste_v1_arr_init_next_match_range(u8 **last_ste, + u32 *added_stes, + u16 gvmi) +{ + dr_ste_v1_arr_init_next_match(last_ste, added_stes, gvmi); + dr_ste_v1_set_entry_type(*last_ste, DR_STE_V1_TYPE_MATCH_RANGES); +} + void dr_ste_v1_set_actions_tx(struct mlx5dr_domain *dmn, u8 *action_type_set, u32 actions_caps, @@ -670,6 +710,20 @@ void dr_ste_v1_set_actions_tx(struct mlx5dr_domain *dmn, action += DR_STE_ACTION_DOUBLE_SZ; } + if (action_type_set[DR_ACTION_TYP_RANGE]) { + /* match ranges requires a new STE of its own type */ + dr_ste_v1_arr_init_next_match_range(&last_ste, added_stes, attr->gvmi); + dr_ste_v1_set_miss_addr(last_ste, attr->range.miss_icm_addr); + + /* we do not support setting any action on the match ranges STE */ + action_sz = 0; + + dr_ste_v1_set_match_range_pkt_len(last_ste, + attr->range.definer_id, + attr->range.min, + attr->range.max); + } + dr_ste_v1_set_hit_gvmi(last_ste, attr->hit_gvmi); dr_ste_v1_set_hit_addr(last_ste, attr->final_icm_addr, 1); } @@ -858,6 +912,20 @@ void dr_ste_v1_set_actions_rx(struct mlx5dr_domain *dmn, action += DR_STE_ACTION_DOUBLE_SZ; } + if (action_type_set[DR_ACTION_TYP_RANGE]) { + /* match ranges requires a new STE of its own type */ + dr_ste_v1_arr_init_next_match_range(&last_ste, added_stes, attr->gvmi); + dr_ste_v1_set_miss_addr(last_ste, attr->range.miss_icm_addr); + + /* we do not support setting any action on the match ranges STE */ + action_sz = 0; + + dr_ste_v1_set_match_range_pkt_len(last_ste, + attr->range.definer_id, + attr->range.min, + attr->range.max); + } + dr_ste_v1_set_hit_gvmi(last_ste, attr->hit_gvmi); dr_ste_v1_set_hit_addr(last_ste, attr->final_icm_addr, 1); } @@ -2144,6 +2212,7 @@ static struct mlx5dr_ste_ctx ste_ctx_v1 = { .ste_init = &dr_ste_v1_init, .set_next_lu_type = &dr_ste_v1_set_next_lu_type, .get_next_lu_type = &dr_ste_v1_get_next_lu_type, + .is_miss_addr_set = &dr_ste_v1_is_miss_addr_set, .set_miss_addr = &dr_ste_v1_set_miss_addr, .get_miss_addr = &dr_ste_v1_get_miss_addr, .set_hit_addr = &dr_ste_v1_set_hit_addr, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h index 8a1d49790c6e..b5c0f0f8392f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v1.h @@ -7,6 +7,7 @@ #include "dr_types.h" #include "dr_ste.h" +bool dr_ste_v1_is_miss_addr_set(u8 *hw_ste_p); void dr_ste_v1_set_miss_addr(u8 *hw_ste_p, u64 miss_addr); u64 dr_ste_v1_get_miss_addr(u8 *hw_ste_p); void dr_ste_v1_set_byte_mask(u8 *hw_ste_p, u16 byte_mask); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c index c60fddd125d2..cf1a3c9a1cf4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v2.c @@ -202,6 +202,7 @@ static struct mlx5dr_ste_ctx ste_ctx_v2 = { .ste_init = &dr_ste_v1_init, .set_next_lu_type = &dr_ste_v1_set_next_lu_type, .get_next_lu_type = &dr_ste_v1_get_next_lu_type, + .is_miss_addr_set = &dr_ste_v1_is_miss_addr_set, .set_miss_addr = &dr_ste_v1_set_miss_addr, .get_miss_addr = &dr_ste_v1_get_miss_addr, .set_hit_addr = &dr_ste_v1_set_hit_addr, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c index f68461b13391..69294a66fd7f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c @@ -295,7 +295,7 @@ int mlx5dr_table_destroy(struct mlx5dr_table *tbl) mlx5dr_dbg_tbl_del(tbl); ret = dr_table_destroy_sw_owned_tbl(tbl); if (ret) - return ret; + mlx5dr_err(tbl->dmn, "Failed to destroy sw owned table\n"); dr_table_uninit(tbl); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h index 1777a1e508e7..2b769dcbd453 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h @@ -81,6 +81,7 @@ mlx5dr_icm_next_higher_chunk(enum mlx5dr_icm_chunk_size chunk) enum { DR_STE_SIZE = 64, DR_STE_SIZE_CTRL = 32, + DR_STE_SIZE_MATCH_TAG = 32, DR_STE_SIZE_TAG = 16, DR_STE_SIZE_MASK = 16, DR_STE_SIZE_REDUCED = DR_STE_SIZE - DR_STE_SIZE_MASK, @@ -128,6 +129,7 @@ enum mlx5dr_action_type { DR_ACTION_TYP_REMOVE_HDR, DR_ACTION_TYP_SAMPLER, DR_ACTION_TYP_ASO_FLOW_METER, + DR_ACTION_TYP_RANGE, DR_ACTION_TYP_MAX, }; @@ -146,6 +148,8 @@ struct mlx5dr_cmd_caps; struct mlx5dr_rule_rx_tx; struct mlx5dr_matcher_rx_tx; struct mlx5dr_ste_ctx; +struct mlx5dr_send_info_pool; +struct mlx5dr_icm_hot_chunk; struct mlx5dr_ste { /* refcount: indicates the num of rules that using this ste */ @@ -235,6 +239,7 @@ static inline void mlx5dr_htbl_get(struct mlx5dr_ste_htbl *htbl) /* STE utils */ u32 mlx5dr_ste_calc_hash_index(u8 *hw_ste_p, struct mlx5dr_ste_htbl *htbl); +bool mlx5dr_ste_is_miss_addr_set(struct mlx5dr_ste_ctx *ste_ctx, u8 *hw_ste_p); void mlx5dr_ste_set_miss_addr(struct mlx5dr_ste_ctx *ste_ctx, u8 *hw_ste, u64 miss_addr); void mlx5dr_ste_set_hit_addr(struct mlx5dr_ste_ctx *ste_ctx, @@ -279,6 +284,13 @@ struct mlx5dr_ste_actions_attr { u8 dest_reg_id; u8 init_color; } aso_flow_meter; + + struct { + u64 miss_icm_addr; + u32 definer_id; + u32 min; + u32 max; + } range; }; void mlx5dr_ste_set_actions_rx(struct mlx5dr_ste_ctx *ste_ctx, @@ -912,12 +924,17 @@ struct mlx5dr_domain { refcount_t refcount; struct mlx5dr_icm_pool *ste_icm_pool; struct mlx5dr_icm_pool *action_icm_pool; + struct mlx5dr_send_info_pool *send_info_pool_rx; + struct mlx5dr_send_info_pool *send_info_pool_tx; + struct kmem_cache *chunks_kmem_cache; + struct kmem_cache *htbls_kmem_cache; struct mlx5dr_send_ring *send_ring; struct mlx5dr_domain_info info; struct xarray csum_fts_xa; struct mlx5dr_ste_ctx *ste_ctx; struct list_head dbg_tbl_list; struct mlx5dr_dbg_dump_info dump_info; + struct xarray definers_xa; }; struct mlx5dr_table_rx_tx { @@ -1020,6 +1037,15 @@ struct mlx5dr_action_dest_tbl { }; }; +struct mlx5dr_action_range { + struct mlx5dr_domain *dmn; + struct mlx5dr_action *hit_tbl_action; + struct mlx5dr_action *miss_tbl_action; + u32 definer_id; + u32 min; + u32 max; +}; + struct mlx5dr_action_ctr { u32 ctr_id; u32 offset; @@ -1066,6 +1092,7 @@ struct mlx5dr_action { struct mlx5dr_action_push_vlan *push_vlan; struct mlx5dr_action_flow_tag *flow_tag; struct mlx5dr_action_aso_flow_meter *aso; + struct mlx5dr_action_range *range; }; }; @@ -1105,7 +1132,6 @@ int mlx5dr_rule_get_reverse_rule_members(struct mlx5dr_ste **ste_arr, struct mlx5dr_icm_chunk { struct mlx5dr_icm_buddy_mem *buddy_mem; - struct list_head chunk_list; /* indicates the index of this chunk in the whole memory, * used for deleting the chunk from the buddy @@ -1158,6 +1184,9 @@ u32 mlx5dr_icm_pool_get_chunk_num_of_entries(struct mlx5dr_icm_chunk *chunk); u32 mlx5dr_icm_pool_get_chunk_byte_size(struct mlx5dr_icm_chunk *chunk); u8 *mlx5dr_ste_get_hw_ste(struct mlx5dr_ste *ste); +struct mlx5dr_ste_htbl *mlx5dr_icm_pool_alloc_htbl(struct mlx5dr_icm_pool *pool); +void mlx5dr_icm_pool_free_htbl(struct mlx5dr_icm_pool *pool, struct mlx5dr_ste_htbl *htbl); + static inline int mlx5dr_icm_pool_dm_type_to_entry_size(enum mlx5dr_icm_type icm_type) { @@ -1287,6 +1316,14 @@ int mlx5dr_cmd_create_reformat_ctx(struct mlx5_core_dev *mdev, u32 *reformat_id); void mlx5dr_cmd_destroy_reformat_ctx(struct mlx5_core_dev *mdev, u32 reformat_id); +int mlx5dr_cmd_create_definer(struct mlx5_core_dev *mdev, + u16 format_id, + u8 *dw_selectors, + u8 *byte_selectors, + u8 *match_mask, + u32 *definer_id); +void mlx5dr_cmd_destroy_definer(struct mlx5_core_dev *mdev, + u32 definer_id); struct mlx5dr_cmd_gid_attr { u8 gid[16]; @@ -1404,6 +1441,12 @@ int mlx5dr_send_postsend_formatted_htbl(struct mlx5dr_domain *dmn, int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn, struct mlx5dr_action *action); +int mlx5dr_send_info_pool_create(struct mlx5dr_domain *dmn); +void mlx5dr_send_info_pool_destroy(struct mlx5dr_domain *dmn); +struct mlx5dr_ste_send_info *mlx5dr_send_info_alloc(struct mlx5dr_domain *dmn, + enum mlx5dr_domain_nic_type nic_type); +void mlx5dr_send_info_free(struct mlx5dr_ste_send_info *ste_send_info); + struct mlx5dr_cmd_ft_info { u32 id; u16 vport; @@ -1469,4 +1512,18 @@ int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn, u32 flow_source); void mlx5dr_fw_destroy_md_tbl(struct mlx5dr_domain *dmn, u32 tbl_id, u32 group_id); + +static inline bool mlx5dr_is_fw_table(struct mlx5_flow_table *ft) +{ + return !ft->fs_dr_table.dr_table; +} + +static inline bool mlx5dr_supp_match_ranges(struct mlx5_core_dev *dev) +{ + return (MLX5_CAP_GEN(dev, steering_format_version) >= + MLX5_STEERING_FORMAT_CONNECTX_6DX) && + (MLX5_CAP_GEN_64(dev, match_definer_format_supported) & + (1ULL << MLX5_IFC_DEFINER_FORMAT_ID_SELECT)); +} + #endif /* _DR_TYPES_H_ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c index 13b6d4721e17..984653756779 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c @@ -7,10 +7,11 @@ #include "fs_cmd.h" #include "mlx5dr.h" #include "fs_dr.h" +#include "dr_types.h" -static bool mlx5_dr_is_fw_table(u32 flags) +static bool dr_is_fw_term_table(struct mlx5_flow_table *ft) { - if (flags & MLX5_FLOW_TABLE_TERMINATION) + if (ft->flags & MLX5_FLOW_TABLE_TERMINATION) return true; return false; @@ -69,7 +70,7 @@ static int mlx5_cmd_dr_create_flow_table(struct mlx5_flow_root_namespace *ns, u32 flags; int err; - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->create_flow_table(ns, ft, ft_attr, next_ft); @@ -109,7 +110,7 @@ static int mlx5_cmd_dr_destroy_flow_table(struct mlx5_flow_root_namespace *ns, struct mlx5dr_action *action = ft->fs_dr_table.miss_action; int err; - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->destroy_flow_table(ns, ft); err = mlx5dr_table_destroy(ft->fs_dr_table.dr_table); @@ -134,7 +135,7 @@ static int mlx5_cmd_dr_modify_flow_table(struct mlx5_flow_root_namespace *ns, struct mlx5_flow_table *ft, struct mlx5_flow_table *next_ft) { - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->modify_flow_table(ns, ft, next_ft); return set_miss_action(ns, ft, next_ft); @@ -153,7 +154,7 @@ static int mlx5_cmd_dr_create_flow_group(struct mlx5_flow_root_namespace *ns, match_criteria_enable); struct mlx5dr_match_parameters mask; - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->create_flow_group(ns, ft, in, fg); @@ -178,7 +179,7 @@ static int mlx5_cmd_dr_destroy_flow_group(struct mlx5_flow_root_namespace *ns, struct mlx5_flow_table *ft, struct mlx5_flow_group *fg) { - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->destroy_flow_group(ns, ft, fg); return mlx5dr_matcher_destroy(fg->fs_dr_matcher.dr_matcher); @@ -209,11 +210,22 @@ static struct mlx5dr_action *create_ft_action(struct mlx5dr_domain *domain, { struct mlx5_flow_table *dest_ft = dst->dest_attr.ft; - if (mlx5_dr_is_fw_table(dest_ft->flags)) + if (mlx5dr_is_fw_table(dest_ft)) return mlx5dr_action_create_dest_flow_fw_table(domain, dest_ft); return mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table); } +static struct mlx5dr_action *create_range_action(struct mlx5dr_domain *domain, + struct mlx5_flow_rule *dst) +{ + return mlx5dr_action_create_dest_match_range(domain, + dst->dest_attr.range.field, + dst->dest_attr.range.hit_ft, + dst->dest_attr.range.miss_ft, + dst->dest_attr.range.min, + dst->dest_attr.range.max); +} + static struct mlx5dr_action *create_action_push_vlan(struct mlx5dr_domain *domain, struct mlx5_fs_vlan *vlan) { @@ -260,7 +272,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns, int err = 0; int i; - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->create_fte(ns, ft, group, fte); actions = kcalloc(MLX5_FLOW_CONTEXT_ACTION_MAX, sizeof(*actions), @@ -467,6 +479,15 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns, fs_dr_actions[fs_dr_num_actions++] = tmp_action; term_actions[num_term_actions++].dest = tmp_action; break; + case MLX5_FLOW_DESTINATION_TYPE_RANGE: + tmp_action = create_range_action(domain, dst); + if (!tmp_action) { + err = -ENOMEM; + goto free_actions; + } + fs_dr_actions[fs_dr_num_actions++] = tmp_action; + term_actions[num_term_actions++].dest = tmp_action; + break; default: err = -EOPNOTSUPP; goto free_actions; @@ -702,7 +723,7 @@ static int mlx5_cmd_dr_delete_fte(struct mlx5_flow_root_namespace *ns, int err; int i; - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->delete_fte(ns, ft, fte); err = mlx5dr_rule_destroy(rule->dr_rule); @@ -727,7 +748,7 @@ static int mlx5_cmd_dr_update_fte(struct mlx5_flow_root_namespace *ns, struct fs_fte fte_tmp = {}; int ret; - if (mlx5_dr_is_fw_table(ft->flags)) + if (dr_is_fw_term_table(ft)) return mlx5_fs_cmd_get_fw_cmds()->update_fte(ns, ft, group, modify_mask, fte); /* Backup current dr rule details */ @@ -780,11 +801,19 @@ static int mlx5_cmd_dr_destroy_ns(struct mlx5_flow_root_namespace *ns) static u32 mlx5_cmd_dr_get_capabilities(struct mlx5_flow_root_namespace *ns, enum fs_flow_table_type ft_type) { + u32 steering_caps = 0; + if (ft_type != FS_FT_FDB || MLX5_CAP_GEN(ns->dev, steering_format_version) == MLX5_STEERING_FORMAT_CONNECTX_5) return 0; - return MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX | MLX5_FLOW_STEERING_CAP_VLAN_POP_ON_TX; + steering_caps |= MLX5_FLOW_STEERING_CAP_VLAN_PUSH_ON_RX; + steering_caps |= MLX5_FLOW_STEERING_CAP_VLAN_POP_ON_TX; + + if (mlx5dr_supp_match_ranges(ns->dev)) + steering_caps |= MLX5_FLOW_STEERING_CAP_MATCH_RANGES; + + return steering_caps; } bool mlx5_fs_dr_is_supported(struct mlx5_core_dev *dev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h index 34c2bd17a8b4..790a17d6207f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5_ifc_dr_ste_v1.h @@ -165,6 +165,41 @@ struct mlx5_ifc_ste_mask_and_match_v1_bits { u8 action[0x60]; }; +struct mlx5_ifc_ste_match_ranges_v1_bits { + u8 entry_format[0x8]; + u8 counter_id[0x18]; + + u8 miss_address_63_48[0x10]; + u8 match_definer_ctx_idx[0x8]; + u8 miss_address_39_32[0x8]; + + u8 miss_address_31_6[0x1a]; + u8 reserved_at_5a[0x1]; + u8 match_polarity[0x1]; + u8 reparse[0x1]; + u8 reserved_at_5d[0x3]; + + u8 next_table_base_63_48[0x10]; + u8 hash_definer_ctx_idx[0x8]; + u8 next_table_base_39_32_size[0x8]; + + u8 next_table_base_31_5_size[0x1b]; + u8 hash_type[0x2]; + u8 hash_after_actions[0x1]; + u8 reserved_at_9e[0x2]; + + u8 action[0x60]; + + u8 max_value_0[0x20]; + u8 min_value_0[0x20]; + u8 max_value_1[0x20]; + u8 min_value_1[0x20]; + u8 max_value_2[0x20]; + u8 min_value_2[0x20]; + u8 max_value_3[0x20]; + u8 min_value_3[0x20]; +}; + struct mlx5_ifc_ste_eth_l2_src_v1_bits { u8 reserved_at_0[0x1]; u8 sx_sniffer[0x1]; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h index 226a0d7bb06d..9afd268a2573 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h @@ -140,8 +140,21 @@ mlx5dr_action_create_aso(struct mlx5dr_domain *dmn, u8 init_color, u8 meter_id); +struct mlx5dr_action * +mlx5dr_action_create_dest_match_range(struct mlx5dr_domain *dmn, + u32 field, + struct mlx5_flow_table *hit_ft, + struct mlx5_flow_table *miss_ft, + u32 min, + u32 max); + int mlx5dr_action_destroy(struct mlx5dr_action *action); +int mlx5dr_definer_get(struct mlx5dr_domain *dmn, u16 format_id, + u8 *dw_selectors, u8 *byte_selectors, + u8 *match_mask, u32 *definer_id); +void mlx5dr_definer_put(struct mlx5dr_domain *dmn, u32 definer_id); + static inline bool mlx5dr_is_supported(struct mlx5_core_dev *dev) { @@ -164,16 +177,9 @@ struct mlx5dr_icm_buddy_mem { struct mlx5dr_icm_mr *icm_mr; struct mlx5dr_icm_pool *pool; - /* This is the list of used chunks. HW may be accessing this memory */ - struct list_head used_list; + /* Amount of memory in used chunks - HW may be accessing this memory */ u64 used_memory; - /* Hardware may be accessing this memory but at some future, - * undetermined time, it might cease to do so. - * sync_ste command sets them free. - */ - struct list_head hot_list; - /* Memory optimisation */ struct mlx5dr_ste *ste_arr; struct list_head *miss_list; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/uar.c b/drivers/net/ethernet/mellanox/mlx5/core/uar.c index 8455e79bc44a..1513112ecec8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/uar.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/uar.c @@ -31,7 +31,6 @@ */ #include <linux/kernel.h> -#include <linux/io-mapping.h> #include <linux/mlx5/driver.h> #include "mlx5_core.h" diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index d5c317325030..ba7e3df22413 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -1160,14 +1160,40 @@ u64 mlx5_query_nic_system_image_guid(struct mlx5_core_dev *mdev) } EXPORT_SYMBOL_GPL(mlx5_query_nic_system_image_guid); -int mlx5_vport_get_other_func_cap(struct mlx5_core_dev *dev, u16 function_id, void *out) +int mlx5_vport_get_other_func_cap(struct mlx5_core_dev *dev, u16 function_id, void *out, + u16 opmod) { - u16 opmod = (MLX5_CAP_GENERAL << 1) | (HCA_CAP_OPMOD_GET_MAX & 0x01); u8 in[MLX5_ST_SZ_BYTES(query_hca_cap_in)] = {}; + opmod = (opmod << 1) | (HCA_CAP_OPMOD_GET_MAX & 0x01); MLX5_SET(query_hca_cap_in, in, opcode, MLX5_CMD_OP_QUERY_HCA_CAP); MLX5_SET(query_hca_cap_in, in, op_mod, opmod); MLX5_SET(query_hca_cap_in, in, function_id, function_id); MLX5_SET(query_hca_cap_in, in, other_function, true); return mlx5_cmd_exec_inout(dev, query_hca_cap, in, out); } +EXPORT_SYMBOL_GPL(mlx5_vport_get_other_func_cap); + +int mlx5_vport_set_other_func_cap(struct mlx5_core_dev *dev, const void *hca_cap, + u16 function_id, u16 opmod) +{ + int set_sz = MLX5_ST_SZ_BYTES(set_hca_cap_in); + void *set_hca_cap; + void *set_ctx; + int ret; + + set_ctx = kzalloc(set_sz, GFP_KERNEL); + if (!set_ctx) + return -ENOMEM; + + MLX5_SET(set_hca_cap_in, set_ctx, opcode, MLX5_CMD_OP_SET_HCA_CAP); + MLX5_SET(set_hca_cap_in, set_ctx, op_mod, opmod << 1); + set_hca_cap = MLX5_ADDR_OF(set_hca_cap_in, set_ctx, capability); + memcpy(set_hca_cap, hca_cap, MLX5_ST_SZ_BYTES(cmd_hca_cap)); + MLX5_SET(set_hca_cap_in, set_ctx, function_id, function_id); + MLX5_SET(set_hca_cap_in, set_ctx, other_function, true); + ret = mlx5_cmd_exec_in(dev, set_hca_cap, set_ctx); + + kfree(set_ctx); + return ret; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.h b/drivers/net/ethernet/mellanox/mlx5/core/wq.h index 4d629e5ddbc7..e4ef1d24a3ad 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wq.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.h @@ -243,6 +243,23 @@ static inline struct mlx5_cqe64 *mlx5_cqwq_get_cqe(struct mlx5_cqwq *wq) return cqe; } +static inline +struct mlx5_cqe64 *mlx5_cqwq_get_cqe_enahnced_comp(struct mlx5_cqwq *wq) +{ + u8 sw_validity_iteration_count = mlx5_cqwq_get_wrap_cnt(wq) & 0xff; + u32 ci = mlx5_cqwq_get_ci(wq); + struct mlx5_cqe64 *cqe; + + cqe = mlx5_cqwq_get_wqe(wq, ci); + if (cqe->validity_iteration_count != sw_validity_iteration_count) + return NULL; + + /* ensure cqe content is read after cqe ownership bit/validity byte */ + dma_rmb(); + + return cqe; +} + static inline u32 mlx5_wq_ll_get_size(struct mlx5_wq_ll *wq) { return (u32)wq->fbc.sz_m1 + 1; |