diff options
Diffstat (limited to 'drivers/net/ethernet/intel/ice')
28 files changed, 7975 insertions, 1993 deletions
diff --git a/drivers/net/ethernet/intel/ice/Makefile b/drivers/net/ethernet/intel/ice/Makefile index e5d6f684437e..2d140ba83781 100644 --- a/drivers/net/ethernet/intel/ice/Makefile +++ b/drivers/net/ethernet/intel/ice/Makefile @@ -17,3 +17,4 @@ ice-y := ice_main.o \ ice_txrx.o \ ice_ethtool.o ice-$(CONFIG_PCI_IOV) += ice_virtchnl_pf.o ice_sriov.o +ice-$(CONFIG_DCB) += ice_dcb.o ice_dcb_lib.o diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 89440775aea1..9ee6b55553c0 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -34,6 +34,7 @@ #include "ice_devids.h" #include "ice_type.h" #include "ice_txrx.h" +#include "ice_dcb.h" #include "ice_switch.h" #include "ice_common.h" #include "ice_sched.h" @@ -42,10 +43,28 @@ extern const char ice_drv_ver[]; #define ICE_BAR0 0 -#define ICE_DFLT_NUM_DESC 128 #define ICE_REQ_DESC_MULTIPLE 32 -#define ICE_MIN_NUM_DESC ICE_REQ_DESC_MULTIPLE +#define ICE_MIN_NUM_DESC 64 #define ICE_MAX_NUM_DESC 8160 +#define ICE_DFLT_MIN_RX_DESC 512 +/* if the default number of Rx descriptors between ICE_MAX_NUM_DESC and the + * number of descriptors to fill up an entire page is greater than or equal to + * ICE_DFLT_MIN_RX_DESC set it based on page size, otherwise set it to + * ICE_DFLT_MIN_RX_DESC + */ +#define ICE_DFLT_NUM_RX_DESC \ + min_t(u16, ICE_MAX_NUM_DESC, \ + max_t(u16, ALIGN(PAGE_SIZE / sizeof(union ice_32byte_rx_desc), \ + ICE_REQ_DESC_MULTIPLE), \ + ICE_DFLT_MIN_RX_DESC)) +/* set default number of Tx descriptors to the minimum between ICE_MAX_NUM_DESC + * and the number of descriptors to fill up an entire page + */ +#define ICE_DFLT_NUM_TX_DESC min_t(u16, ICE_MAX_NUM_DESC, \ + ALIGN(PAGE_SIZE / \ + sizeof(struct ice_tx_desc), \ + ICE_REQ_DESC_MULTIPLE)) + #define ICE_DFLT_TRAFFIC_CLASS BIT(0) #define ICE_INT_NAME_STR_LEN (IFNAMSIZ + 16) #define ICE_ETHTOOL_FWVER_LEN 32 @@ -71,6 +90,8 @@ extern const char ice_drv_ver[]; #define ICE_MAX_QS_PER_VF 256 #define ICE_MIN_QS_PER_VF 1 #define ICE_DFLT_QS_PER_VF 4 +#define ICE_NONQ_VECS_VF 1 +#define ICE_MAX_SCATTER_QS_PER_VF 16 #define ICE_MAX_BASE_QS_PER_VF 16 #define ICE_MAX_INTR_PER_VF 65 #define ICE_MIN_INTR_PER_VF (ICE_MIN_QS_PER_VF + 1) @@ -114,6 +135,23 @@ extern const char ice_drv_ver[]; #define ice_for_each_q_vector(vsi, i) \ for ((i) = 0; (i) < (vsi)->num_q_vectors; (i)++) +#define ICE_UCAST_PROMISC_BITS (ICE_PROMISC_UCAST_TX | ICE_PROMISC_MCAST_TX | \ + ICE_PROMISC_UCAST_RX | ICE_PROMISC_MCAST_RX) + +#define ICE_UCAST_VLAN_PROMISC_BITS (ICE_PROMISC_UCAST_TX | \ + ICE_PROMISC_MCAST_TX | \ + ICE_PROMISC_UCAST_RX | \ + ICE_PROMISC_MCAST_RX | \ + ICE_PROMISC_VLAN_TX | \ + ICE_PROMISC_VLAN_RX) + +#define ICE_MCAST_PROMISC_BITS (ICE_PROMISC_MCAST_TX | ICE_PROMISC_MCAST_RX) + +#define ICE_MCAST_VLAN_PROMISC_BITS (ICE_PROMISC_MCAST_TX | \ + ICE_PROMISC_MCAST_RX | \ + ICE_PROMISC_VLAN_TX | \ + ICE_PROMISC_VLAN_RX) + struct ice_tc_info { u16 qoffset; u16 qcount_tx; @@ -123,18 +161,18 @@ struct ice_tc_info { struct ice_tc_cfg { u8 numtc; /* Total number of enabled TCs */ - u8 ena_tc; /* TX map */ + u8 ena_tc; /* Tx map */ struct ice_tc_info tc_info[ICE_MAX_TRAFFIC_CLASS]; }; struct ice_res_tracker { u16 num_entries; - u16 search_hint; + u16 end; u16 list[1]; }; struct ice_qs_cfg { - struct mutex *qs_mutex; /* will be assgined to &pf->avail_q_mutex */ + struct mutex *qs_mutex; /* will be assigned to &pf->avail_q_mutex */ unsigned long *pf_map; unsigned long pf_map_size; unsigned int q_count; @@ -151,6 +189,7 @@ struct ice_sw { }; enum ice_state { + __ICE_TESTING, __ICE_DOWN, __ICE_NEEDS_RESTART, __ICE_PREPARED_FOR_RESET, /* set by driver when prepared */ @@ -213,8 +252,7 @@ struct ice_vsi { u32 rx_buf_failed; u32 rx_page_failed; int num_q_vectors; - int sw_base_vector; /* Irq base for OS reserved vectors */ - int hw_base_vector; /* HW (absolute) index of a vector */ + int base_vector; /* IRQ base for OS reserved vectors */ enum ice_vsi_type type; u16 vsi_num; /* HW (absolute) index of this VSI */ u16 idx; /* software index in pf->vsi[] */ @@ -224,6 +262,8 @@ struct ice_vsi { s16 vf_id; /* VF ID for SR-IOV VSIs */ + u16 ethtype; /* Ethernet protocol for pause frame */ + /* RSS config */ u16 rss_table_size; /* HW RSS table size */ u16 rss_size; /* Allocated RSS queues */ @@ -244,9 +284,10 @@ struct ice_vsi { struct list_head tmp_sync_list; /* MAC filters to be synced */ struct list_head tmp_unsync_list; /* MAC filters to be unsynced */ - u8 irqs_ready; - u8 current_isup; /* Sync 'link up' logging */ - u8 stat_offsets_loaded; + u8 irqs_ready:1; + u8 current_isup:1; /* Sync 'link up' logging */ + u8 stat_offsets_loaded:1; + u8 vlan_ena:1; /* queue information */ u8 tx_mapping_mode; /* ICE_MAP_MODE_[CONTIG|SCATTER] */ @@ -257,26 +298,34 @@ struct ice_vsi { u16 num_txq; /* Used Tx queues */ u16 alloc_rxq; /* Allocated Rx queues */ u16 num_rxq; /* Used Rx queues */ - u16 num_desc; + u16 num_rx_desc; + u16 num_tx_desc; struct ice_tc_cfg tc_cfg; } ____cacheline_internodealigned_in_smp; /* struct that defines an interrupt vector */ struct ice_q_vector { struct ice_vsi *vsi; - cpumask_t affinity_mask; - struct napi_struct napi; - struct ice_ring_container rx; - struct ice_ring_container tx; - struct irq_affinity_notify affinity_notify; + u16 v_idx; /* index in the vsi->q_vector array. */ - u8 num_ring_tx; /* total number of Tx rings in vector */ + u16 reg_idx; u8 num_ring_rx; /* total number of Rx rings in vector */ - char name[ICE_INT_NAME_STR_LEN]; + u8 num_ring_tx; /* total number of Tx rings in vector */ + u8 itr_countdown; /* when 0 should adjust adaptive ITR */ /* in usecs, need to use ice_intrl_to_usecs_reg() before writing this * value to the device */ u8 intrl; + + struct napi_struct napi; + + struct ice_ring_container rx; + struct ice_ring_container tx; + + cpumask_t affinity_mask; + struct irq_affinity_notify affinity_notify; + + char name[ICE_INT_NAME_STR_LEN]; } ____cacheline_internodealigned_in_smp; enum ice_pf_flags { @@ -285,7 +334,11 @@ enum ice_pf_flags { ICE_FLAG_RSS_ENA, ICE_FLAG_SRIOV_ENA, ICE_FLAG_SRIOV_CAPABLE, + ICE_FLAG_DCB_CAPABLE, + ICE_FLAG_DCB_ENA, ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, + ICE_FLAG_ENABLE_FW_LLDP, + ICE_FLAG_ETHTOOL_CTXT, /* set when ethtool holds RTNL lock */ ICE_PF_FLAGS_NBITS /* must be last */ }; @@ -294,10 +347,12 @@ struct ice_pf { /* OS reserved IRQ details */ struct msix_entry *msix_entries; - struct ice_res_tracker *sw_irq_tracker; - - /* HW reserved Interrupts for this PF */ - struct ice_res_tracker *hw_irq_tracker; + struct ice_res_tracker *irq_tracker; + /* First MSIX vector used by SR-IOV VFs. Calculated by subtracting the + * number of MSIX vectors needed for all SR-IOV VFs from the number of + * MSIX vectors allowed on this PF. + */ + u16 sriov_base_vector; struct ice_vsi **vsi; /* VSIs created by the driver */ struct ice_sw *first_sw; /* first switch created by firmware */ @@ -319,13 +374,11 @@ struct ice_pf { struct mutex sw_mutex; /* lock for protecting VSI alloc flow */ u32 msg_enable; u32 hw_csum_rx_error; - u32 sw_oicr_idx; /* Other interrupt cause SW vector index */ + u32 oicr_idx; /* Other interrupt cause MSIX vector index */ u32 num_avail_sw_msix; /* remaining MSIX SW vectors left unclaimed */ - u32 hw_oicr_idx; /* Other interrupt cause vector HW index */ - u32 num_avail_hw_msix; /* remaining HW MSIX vectors left unclaimed */ u32 num_lan_msix; /* Total MSIX vectors for base driver */ - u16 num_lan_tx; /* num lan Tx queues setup */ - u16 num_lan_rx; /* num lan Rx queues setup */ + u16 num_lan_tx; /* num LAN Tx queues setup */ + u16 num_lan_rx; /* num LAN Rx queues setup */ u16 q_left_tx; /* remaining num Tx queues left unclaimed */ u16 q_left_rx; /* remaining num Rx queues left unclaimed */ u16 next_vsi; /* Next free slot in pf->vsi[] - 0-based! */ @@ -338,11 +391,15 @@ struct ice_pf { struct ice_hw_port_stats stats; struct ice_hw_port_stats stats_prev; struct ice_hw hw; - u8 stat_prev_loaded; /* has previous stats been loaded */ + u8 stat_prev_loaded:1; /* has previous stats been loaded */ +#ifdef CONFIG_DCB + u16 dcbx_cap; +#endif /* CONFIG_DCB */ u32 tx_timeout_count; unsigned long tx_timeout_last_recovery; u32 tx_timeout_recovery_level; char int_name[ICE_INT_NAME_STR_LEN]; + u32 sw_int_count; }; struct ice_netdev_priv { @@ -351,15 +408,16 @@ struct ice_netdev_priv { /** * ice_irq_dynamic_ena - Enable default interrupt generation settings - * @hw: pointer to hw struct - * @vsi: pointer to vsi struct, can be NULL + * @hw: pointer to HW struct + * @vsi: pointer to VSI struct, can be NULL * @q_vector: pointer to q_vector, can be NULL */ -static inline void ice_irq_dynamic_ena(struct ice_hw *hw, struct ice_vsi *vsi, - struct ice_q_vector *q_vector) +static inline void +ice_irq_dynamic_ena(struct ice_hw *hw, struct ice_vsi *vsi, + struct ice_q_vector *q_vector) { - u32 vector = (vsi && q_vector) ? vsi->hw_base_vector + q_vector->v_idx : - ((struct ice_pf *)hw->back)->hw_oicr_idx; + u32 vector = (vsi && q_vector) ? q_vector->reg_idx : + ((struct ice_pf *)hw->back)->oicr_idx; int itr = ICE_ITR_NONE; u32 val; @@ -374,19 +432,42 @@ static inline void ice_irq_dynamic_ena(struct ice_hw *hw, struct ice_vsi *vsi, wr32(hw, GLINT_DYN_CTL(vector), val); } -static inline void ice_vsi_set_tc_cfg(struct ice_vsi *vsi) +/** + * ice_find_vsi_by_type - Find and return VSI of a given type + * @pf: PF to search for VSI + * @type: Value indicating type of VSI we are looking for + */ +static inline struct ice_vsi * +ice_find_vsi_by_type(struct ice_pf *pf, enum ice_vsi_type type) { - vsi->tc_cfg.ena_tc = ICE_DFLT_TRAFFIC_CLASS; - vsi->tc_cfg.numtc = 1; + int i; + + for (i = 0; i < pf->num_alloc_vsi; i++) { + struct ice_vsi *vsi = pf->vsi[i]; + + if (vsi && vsi->type == type) + return vsi; + } + + return NULL; } +int ice_vsi_setup_tx_rings(struct ice_vsi *vsi); +int ice_vsi_setup_rx_rings(struct ice_vsi *vsi); void ice_set_ethtool_ops(struct net_device *netdev); int ice_up(struct ice_vsi *vsi); int ice_down(struct ice_vsi *vsi); +int ice_vsi_cfg(struct ice_vsi *vsi); +struct ice_vsi *ice_lb_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi); int ice_set_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size); int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size); void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size); void ice_print_link_msg(struct ice_vsi *vsi, bool isup); -void ice_napi_del(struct ice_vsi *vsi); +#ifdef CONFIG_DCB +int ice_pf_ena_all_vsi(struct ice_pf *pf, bool locked); +void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked); +#endif /* CONFIG_DCB */ +int ice_open(struct net_device *netdev); +int ice_stop(struct net_device *netdev); #endif /* _ICE_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h index 242c78469181..765e3c2ed045 100644 --- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h +++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h @@ -35,8 +35,8 @@ struct ice_aqc_get_ver { /* Queue Shutdown (direct 0x0003) */ struct ice_aqc_q_shutdown { -#define ICE_AQC_DRIVER_UNLOADING BIT(0) __le32 driver_unloading; +#define ICE_AQC_DRIVER_UNLOADING BIT(0) u8 reserved[12]; }; @@ -62,7 +62,7 @@ struct ice_aqc_req_res { #define ICE_AQ_RES_NVM_WRITE_DFLT_TIMEOUT_MS 180000 #define ICE_AQ_RES_CHNG_LOCK_DFLT_TIMEOUT_MS 1000 #define ICE_AQ_RES_GLBL_LOCK_DFLT_TIMEOUT_MS 3000 - /* For SDP: pin id of the SDP */ + /* For SDP: pin ID of the SDP */ __le32 res_number; /* Status is only used for ICE_AQC_RES_ID_GLBL_LOCK */ __le16 status; @@ -120,11 +120,9 @@ struct ice_aqc_manage_mac_read { #define ICE_AQC_MAN_MAC_WOL_ADDR_VALID BIT(7) #define ICE_AQC_MAN_MAC_READ_S 4 #define ICE_AQC_MAN_MAC_READ_M (0xF << ICE_AQC_MAN_MAC_READ_S) - u8 lport_num; - u8 lport_num_valid; -#define ICE_AQC_MAN_MAC_PORT_NUM_IS_VALID BIT(0) + u8 rsvd[2]; u8 num_addr; /* Used in response */ - u8 reserved[3]; + u8 rsvd1[3]; __le32 addr_high; __le32 addr_low; }; @@ -140,7 +138,7 @@ struct ice_aqc_manage_mac_read_resp { /* Manage MAC address, write command - direct (0x0108) */ struct ice_aqc_manage_mac_write { - u8 port_num; + u8 rsvd; u8 flags; #define ICE_AQC_MAN_MAC_WR_MC_MAG_EN BIT(0) #define ICE_AQC_MAN_MAC_WR_WOL_LAA_PFR_KEEP BIT(1) @@ -747,6 +745,32 @@ struct ice_aqc_delete_elem { __le32 teid[1]; }; +/* Query Port ETS (indirect 0x040E) + * + * This indirect command is used to query port TC node configuration. + */ +struct ice_aqc_query_port_ets { + __le32 port_teid; + __le32 reserved; + __le32 addr_high; + __le32 addr_low; +}; + +struct ice_aqc_port_ets_elem { + u8 tc_valid_bits; + u8 reserved[3]; + /* 3 bits for UP per TC 0-7, 4th byte reserved */ + __le32 up2tc; + u8 tc_bw_share[8]; + __le32 port_eir_prof_id; + __le32 port_cir_prof_id; + /* 3 bits per Node priority to TC 0-7, 4th byte reserved */ + __le32 tc_node_prio; +#define ICE_TC_NODE_PRIO_S 0x4 + u8 reserved1[4]; + __le32 tc_node_teid[8]; /* Used for response, reserved in command */ +}; + /* Query Scheduler Resource Allocation (indirect 0x0412) * This indirect command retrieves the scheduler resources allocated by * EMP Firmware to the given PF. @@ -894,6 +918,8 @@ struct ice_aqc_get_phy_caps_data { #define ICE_AQC_PHY_EN_LINK BIT(3) #define ICE_AQC_PHY_AN_MODE BIT(4) #define ICE_AQC_GET_PHY_EN_MOD_QUAL BIT(5) +#define ICE_AQC_PHY_EN_AUTO_FEC BIT(7) +#define ICE_AQC_PHY_CAPS_MASK ICE_M(0xff, 0) u8 low_power_ctrl; #define ICE_AQC_PHY_EN_D3COLD_LOW_POWER_AUTONEG BIT(0) __le16 eee_cap; @@ -906,6 +932,7 @@ struct ice_aqc_get_phy_caps_data { #define ICE_AQC_PHY_EEE_EN_40GBASE_KR4 BIT(6) __le16 eeer_value; u8 phy_id_oui[4]; /* PHY/Module ID connected on the port */ + u8 phy_fw_ver[8]; u8 link_fec_options; #define ICE_AQC_PHY_FEC_10G_KR_40G_KR4_EN BIT(0) #define ICE_AQC_PHY_FEC_10G_KR_40G_KR4_REQ BIT(1) @@ -914,6 +941,8 @@ struct ice_aqc_get_phy_caps_data { #define ICE_AQC_PHY_FEC_25G_RS_544_REQ BIT(4) #define ICE_AQC_PHY_FEC_25G_RS_CLAUSE91_EN BIT(6) #define ICE_AQC_PHY_FEC_25G_KR_CLAUSE74_EN BIT(7) +#define ICE_AQC_PHY_FEC_MASK ICE_M(0xdf, 0) + u8 rsvd1; /* Byte 35 reserved */ u8 extended_compliance_code; #define ICE_MODULE_TYPE_TOTAL_BYTE 3 u8 module_type[ICE_MODULE_TYPE_TOTAL_BYTE]; @@ -928,13 +957,14 @@ struct ice_aqc_get_phy_caps_data { #define ICE_AQC_MOD_TYPE_BYTE2_SFP_PLUS 0xA0 #define ICE_AQC_MOD_TYPE_BYTE2_QSFP_PLUS 0x86 u8 qualified_module_count; + u8 rsvd2[7]; /* Bytes 47:41 reserved */ #define ICE_AQC_QUAL_MOD_COUNT_MAX 16 struct { u8 v_oui[3]; - u8 rsvd1; + u8 rsvd3; u8 v_part[16]; __le32 v_rev; - __le64 rsvd8; + __le64 rsvd4; } qual_modules[ICE_AQC_QUAL_MOD_COUNT_MAX]; }; @@ -953,8 +983,9 @@ struct ice_aqc_set_phy_cfg_data { __le64 phy_type_low; /* Use values from ICE_PHY_TYPE_LOW_* */ __le64 phy_type_high; /* Use values from ICE_PHY_TYPE_HIGH_* */ u8 caps; -#define ICE_AQ_PHY_ENA_TX_PAUSE_ABILITY BIT(0) -#define ICE_AQ_PHY_ENA_RX_PAUSE_ABILITY BIT(1) +#define ICE_AQ_PHY_ENA_VALID_MASK ICE_M(0xef, 0) +#define ICE_AQ_PHY_ENA_TX_PAUSE_ABILITY BIT(0) +#define ICE_AQ_PHY_ENA_RX_PAUSE_ABILITY BIT(1) #define ICE_AQ_PHY_ENA_LOW_POWER BIT(2) #define ICE_AQ_PHY_ENA_LINK BIT(3) #define ICE_AQ_PHY_ENA_AUTO_LINK_UPDT BIT(5) @@ -1023,7 +1054,7 @@ struct ice_aqc_get_link_status_data { u8 ext_info; #define ICE_AQ_LINK_PHY_TEMP_ALARM BIT(0) #define ICE_AQ_LINK_EXCESSIVE_ERRORS BIT(1) /* Excessive Link Errors */ - /* Port TX Suspended */ + /* Port Tx Suspended */ #define ICE_AQ_LINK_TX_S 2 #define ICE_AQ_LINK_TX_M (0x03 << ICE_AQ_LINK_TX_S) #define ICE_AQ_LINK_TX_ACTIVE 0 @@ -1035,6 +1066,7 @@ struct ice_aqc_get_link_status_data { #define ICE_AQ_LINK_25G_KR_FEC_EN BIT(0) #define ICE_AQ_LINK_25G_RS_528_FEC_EN BIT(1) #define ICE_AQ_LINK_25G_RS_544_FEC_EN BIT(2) +#define ICE_AQ_FEC_MASK ICE_M(0x7, 0) /* Pacing Config */ #define ICE_AQ_CFG_PACING_S 3 #define ICE_AQ_CFG_PACING_M (0xF << ICE_AQ_CFG_PACING_S) @@ -1085,6 +1117,14 @@ struct ice_aqc_set_event_mask { u8 reserved1[6]; }; +/* Set MAC Loopback command (direct 0x0620) */ +struct ice_aqc_set_mac_lb { + u8 lb_mode; +#define ICE_AQ_MAC_LB_EN BIT(0) +#define ICE_AQ_MAC_LB_OSC_CLK BIT(1) + u8 reserved[15]; +}; + /* Set Port Identification LED (direct, 0x06E9) */ struct ice_aqc_set_port_id_led { u8 lport_num; @@ -1118,10 +1158,21 @@ struct ice_aqc_nvm { __le32 addr_low; }; +/* NVM Checksum Command (direct, 0x0706) */ +struct ice_aqc_nvm_checksum { + u8 flags; +#define ICE_AQC_NVM_CHECKSUM_VERIFY BIT(0) +#define ICE_AQC_NVM_CHECKSUM_RECALC BIT(1) + u8 rsvd; + __le16 checksum; /* Used only by response */ +#define ICE_AQC_NVM_CHECKSUM_CORRECT 0xBABA + u8 rsvd2[12]; +}; + /** - * Send to PF command (indirect 0x0801) id is only used by PF + * Send to PF command (indirect 0x0801) ID is only used by PF * - * Send to VF command (indirect 0x0802) id is only used by PF + * Send to VF command (indirect 0x0802) ID is only used by PF * */ struct ice_aqc_pf_vf_msg { @@ -1131,6 +1182,126 @@ struct ice_aqc_pf_vf_msg { __le32 addr_low; }; +/* Get LLDP MIB (indirect 0x0A00) + * Note: This is also used by the LLDP MIB Change Event (0x0A01) + * as the format is the same. + */ +struct ice_aqc_lldp_get_mib { + u8 type; +#define ICE_AQ_LLDP_MIB_TYPE_S 0 +#define ICE_AQ_LLDP_MIB_TYPE_M (0x3 << ICE_AQ_LLDP_MIB_TYPE_S) +#define ICE_AQ_LLDP_MIB_LOCAL 0 +#define ICE_AQ_LLDP_MIB_REMOTE 1 +#define ICE_AQ_LLDP_MIB_LOCAL_AND_REMOTE 2 +#define ICE_AQ_LLDP_BRID_TYPE_S 2 +#define ICE_AQ_LLDP_BRID_TYPE_M (0x3 << ICE_AQ_LLDP_BRID_TYPE_S) +#define ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID 0 +#define ICE_AQ_LLDP_BRID_TYPE_NON_TPMR 1 +/* Tx pause flags in the 0xA01 event use ICE_AQ_LLDP_TX_* */ +#define ICE_AQ_LLDP_TX_S 0x4 +#define ICE_AQ_LLDP_TX_M (0x03 << ICE_AQ_LLDP_TX_S) +#define ICE_AQ_LLDP_TX_ACTIVE 0 +#define ICE_AQ_LLDP_TX_SUSPENDED 1 +#define ICE_AQ_LLDP_TX_FLUSHED 3 +/* The following bytes are reserved for the Get LLDP MIB command (0x0A00) + * and in the LLDP MIB Change Event (0x0A01). They are valid for the + * Get LLDP MIB (0x0A00) response only. + */ + u8 reserved1; + __le16 local_len; + __le16 remote_len; + u8 reserved2[2]; + __le32 addr_high; + __le32 addr_low; +}; + +/* Configure LLDP MIB Change Event (direct 0x0A01) */ +/* For MIB Change Event use ice_aqc_lldp_get_mib structure above */ +struct ice_aqc_lldp_set_mib_change { + u8 command; +#define ICE_AQ_LLDP_MIB_UPDATE_ENABLE 0x0 +#define ICE_AQ_LLDP_MIB_UPDATE_DIS 0x1 + u8 reserved[15]; +}; + +/* Stop LLDP (direct 0x0A05) */ +struct ice_aqc_lldp_stop { + u8 command; +#define ICE_AQ_LLDP_AGENT_STATE_MASK BIT(0) +#define ICE_AQ_LLDP_AGENT_STOP 0x0 +#define ICE_AQ_LLDP_AGENT_SHUTDOWN ICE_AQ_LLDP_AGENT_STATE_MASK +#define ICE_AQ_LLDP_AGENT_PERSIST_DIS BIT(1) + u8 reserved[15]; +}; + +/* Start LLDP (direct 0x0A06) */ +struct ice_aqc_lldp_start { + u8 command; +#define ICE_AQ_LLDP_AGENT_START BIT(0) +#define ICE_AQ_LLDP_AGENT_PERSIST_ENA BIT(1) + u8 reserved[15]; +}; + +/* Get CEE DCBX Oper Config (0x0A07) + * The command uses the generic descriptor struct and + * returns the struct below as an indirect response. + */ +struct ice_aqc_get_cee_dcb_cfg_resp { + u8 oper_num_tc; + u8 oper_prio_tc[4]; + u8 oper_tc_bw[8]; + u8 oper_pfc_en; + __le16 oper_app_prio; +#define ICE_AQC_CEE_APP_FCOE_S 0 +#define ICE_AQC_CEE_APP_FCOE_M (0x7 << ICE_AQC_CEE_APP_FCOE_S) +#define ICE_AQC_CEE_APP_ISCSI_S 3 +#define ICE_AQC_CEE_APP_ISCSI_M (0x7 << ICE_AQC_CEE_APP_ISCSI_S) +#define ICE_AQC_CEE_APP_FIP_S 8 +#define ICE_AQC_CEE_APP_FIP_M (0x7 << ICE_AQC_CEE_APP_FIP_S) + __le32 tlv_status; +#define ICE_AQC_CEE_PG_STATUS_S 0 +#define ICE_AQC_CEE_PG_STATUS_M (0x7 << ICE_AQC_CEE_PG_STATUS_S) +#define ICE_AQC_CEE_PFC_STATUS_S 3 +#define ICE_AQC_CEE_PFC_STATUS_M (0x7 << ICE_AQC_CEE_PFC_STATUS_S) +#define ICE_AQC_CEE_FCOE_STATUS_S 8 +#define ICE_AQC_CEE_FCOE_STATUS_M (0x7 << ICE_AQC_CEE_FCOE_STATUS_S) +#define ICE_AQC_CEE_ISCSI_STATUS_S 11 +#define ICE_AQC_CEE_ISCSI_STATUS_M (0x7 << ICE_AQC_CEE_ISCSI_STATUS_S) +#define ICE_AQC_CEE_FIP_STATUS_S 16 +#define ICE_AQC_CEE_FIP_STATUS_M (0x7 << ICE_AQC_CEE_FIP_STATUS_S) + u8 reserved[12]; +}; + +/* Set Local LLDP MIB (indirect 0x0A08) + * Used to replace the local MIB of a given LLDP agent. e.g. DCBX + */ +struct ice_aqc_lldp_set_local_mib { + u8 type; +#define SET_LOCAL_MIB_TYPE_DCBX_M BIT(0) +#define SET_LOCAL_MIB_TYPE_LOCAL_MIB 0 +#define SET_LOCAL_MIB_TYPE_CEE_M BIT(1) +#define SET_LOCAL_MIB_TYPE_CEE_WILLING 0 +#define SET_LOCAL_MIB_TYPE_CEE_NON_WILLING SET_LOCAL_MIB_TYPE_CEE_M + u8 reserved0; + __le16 length; + u8 reserved1[4]; + __le32 addr_high; + __le32 addr_low; +}; + +/* Stop/Start LLDP Agent (direct 0x0A09) + * Used for stopping/starting specific LLDP agent. e.g. DCBX. + * The same structure is used for the response, with the command field + * being used as the status field. + */ +struct ice_aqc_lldp_stop_start_specific_agent { + u8 command; +#define ICE_AQC_START_STOP_AGENT_M BIT(0) +#define ICE_AQC_START_STOP_AGENT_STOP_DCBX 0 +#define ICE_AQC_START_STOP_AGENT_START_DCBX ICE_AQC_START_STOP_AGENT_M + u8 reserved[15]; +}; + /* Get/Set RSS key (indirect 0x0B04/0x0B02) */ struct ice_aqc_get_set_rss_key { #define ICE_AQC_GSET_RSS_KEY_VSI_VALID BIT(15) @@ -1144,6 +1315,9 @@ struct ice_aqc_get_set_rss_key { #define ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE 0x28 #define ICE_AQC_GET_SET_RSS_KEY_DATA_HASH_KEY_SIZE 0xC +#define ICE_GET_SET_RSS_KEY_EXTEND_KEY_SIZE \ + (ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE + \ + ICE_AQC_GET_SET_RSS_KEY_DATA_HASH_KEY_SIZE) struct ice_aqc_get_set_rss_keys { u8 standard_rss_key[ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE]; @@ -1185,7 +1359,7 @@ struct ice_aqc_get_set_rss_lut { __le32 addr_low; }; -/* Add TX LAN Queues (indirect 0x0C30) */ +/* Add Tx LAN Queues (indirect 0x0C30) */ struct ice_aqc_add_txqs { u8 num_qgrps; u8 reserved[3]; @@ -1194,7 +1368,7 @@ struct ice_aqc_add_txqs { __le32 addr_low; }; -/* This is the descriptor of each queue entry for the Add TX LAN Queues +/* This is the descriptor of each queue entry for the Add Tx LAN Queues * command (0x0C30). Only used within struct ice_aqc_add_tx_qgrp. */ struct ice_aqc_add_txqs_perq { @@ -1206,7 +1380,7 @@ struct ice_aqc_add_txqs_perq { struct ice_aqc_txsched_elem info; }; -/* The format of the command buffer for Add TX LAN Queues (0x0C30) +/* The format of the command buffer for Add Tx LAN Queues (0x0C30) * is an array of the following structs. Please note that the length of * each struct ice_aqc_add_tx_qgrp is variable due * to the variable number of queues in each group! @@ -1218,7 +1392,7 @@ struct ice_aqc_add_tx_qgrp { struct ice_aqc_add_txqs_perq txqs[1]; }; -/* Disable TX LAN Queues (indirect 0x0C31) */ +/* Disable Tx LAN Queues (indirect 0x0C31) */ struct ice_aqc_dis_txqs { u8 cmd_type; #define ICE_AQC_Q_DIS_CMD_S 0 @@ -1240,7 +1414,7 @@ struct ice_aqc_dis_txqs { __le32 addr_low; }; -/* The buffer for Disable TX LAN Queues (indirect 0x0C31) +/* The buffer for Disable Tx LAN Queues (indirect 0x0C31) * contains the following structures, arrayed one after the * other. * Note: Since the q_id is 16 bits wide, if the @@ -1387,8 +1561,16 @@ struct ice_aq_desc { struct ice_aqc_get_topo get_topo; struct ice_aqc_sched_elem_cmd sched_elem_cmd; struct ice_aqc_query_txsched_res query_sched_res; + struct ice_aqc_query_port_ets port_ets; struct ice_aqc_nvm nvm; + struct ice_aqc_nvm_checksum nvm_checksum; struct ice_aqc_pf_vf_msg virt; + struct ice_aqc_lldp_get_mib lldp_get_mib; + struct ice_aqc_lldp_set_mib_change lldp_set_event; + struct ice_aqc_lldp_stop lldp_stop; + struct ice_aqc_lldp_start lldp_start; + struct ice_aqc_lldp_set_local_mib lldp_set_mib; + struct ice_aqc_lldp_stop_start_specific_agent lldp_agent_ctrl; struct ice_aqc_get_set_rss_lut get_set_rss_lut; struct ice_aqc_get_set_rss_key get_set_rss_key; struct ice_aqc_add_txqs add_txqs; @@ -1397,6 +1579,7 @@ struct ice_aq_desc { struct ice_aqc_add_update_free_vsi_resp add_update_free_vsi_res; struct ice_aqc_fw_logging fw_logging; struct ice_aqc_get_clear_fw_log get_clear_fw_log; + struct ice_aqc_set_mac_lb set_mac_lb; struct ice_aqc_alloc_free_res_cmd sw_res_ctrl; struct ice_aqc_set_event_mask set_event_mask; struct ice_aqc_get_link_status get_link_status; @@ -1421,6 +1604,8 @@ struct ice_aq_desc { /* error codes */ enum ice_aq_err { ICE_AQ_RC_OK = 0, /* Success */ + ICE_AQ_RC_EPERM = 1, /* Operation not permitted */ + ICE_AQ_RC_ENOENT = 2, /* No such element */ ICE_AQ_RC_ENOMEM = 9, /* Out of memory */ ICE_AQ_RC_EBUSY = 12, /* Device or resource busy */ ICE_AQ_RC_EEXIST = 13, /* Object already exists */ @@ -1473,6 +1658,7 @@ enum ice_adminq_opc { ice_aqc_opc_get_sched_elems = 0x0404, ice_aqc_opc_suspend_sched_elems = 0x0409, ice_aqc_opc_resume_sched_elems = 0x040A, + ice_aqc_opc_query_port_ets = 0x040E, ice_aqc_opc_delete_sched_elems = 0x040F, ice_aqc_opc_query_sched_res = 0x0412, @@ -1482,14 +1668,24 @@ enum ice_adminq_opc { ice_aqc_opc_restart_an = 0x0605, ice_aqc_opc_get_link_status = 0x0607, ice_aqc_opc_set_event_mask = 0x0613, + ice_aqc_opc_set_mac_lb = 0x0620, ice_aqc_opc_set_port_id_led = 0x06E9, /* NVM commands */ ice_aqc_opc_nvm_read = 0x0701, + ice_aqc_opc_nvm_checksum = 0x0706, /* PF/VF mailbox commands */ ice_mbx_opc_send_msg_to_pf = 0x0801, ice_mbx_opc_send_msg_to_vf = 0x0802, + /* LLDP commands */ + ice_aqc_opc_lldp_get_mib = 0x0A00, + ice_aqc_opc_lldp_set_mib_change = 0x0A01, + ice_aqc_opc_lldp_stop = 0x0A05, + ice_aqc_opc_lldp_start = 0x0A06, + ice_aqc_opc_get_cee_dcb_cfg = 0x0A07, + ice_aqc_opc_lldp_set_local_mib = 0x0A08, + ice_aqc_opc_lldp_stop_start_specific_agent = 0x0A09, /* RSS commands */ ice_aqc_opc_set_rss_key = 0x0B02, @@ -1497,12 +1693,13 @@ enum ice_adminq_opc { ice_aqc_opc_get_rss_key = 0x0B04, ice_aqc_opc_get_rss_lut = 0x0B05, - /* TX queue handling commands/events */ + /* Tx queue handling commands/events */ ice_aqc_opc_add_txqs = 0x0C30, ice_aqc_opc_dis_txqs = 0x0C31, /* debug commands */ ice_aqc_opc_fw_logging = 0xFF09, + ice_aqc_opc_fw_logging_info = 0xFF10, }; #endif /* _ICE_ADMINQ_CMD_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 63f003441300..2e0731c1e1a3 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -31,7 +31,7 @@ * @hw: pointer to the HW structure * * This function sets the MAC type of the adapter based on the - * vendor ID and device ID stored in the hw structure. + * vendor ID and device ID stored in the HW structure. */ static enum ice_status ice_set_mac_type(struct ice_hw *hw) { @@ -51,9 +51,6 @@ static enum ice_status ice_set_mac_type(struct ice_hw *hw) */ void ice_dev_onetime_setup(struct ice_hw *hw) { - /* configure Rx - set non pxe mode */ - wr32(hw, GLLAN_RCTL_0, 0x1); - #define MBX_PF_VT_PFALLOC 0x00231E80 /* set VFs per PF */ wr32(hw, MBX_PF_VT_PFALLOC, rd32(hw, PF_VT_PFALLOC_HIF)); @@ -77,7 +74,7 @@ enum ice_status ice_clear_pf_cfg(struct ice_hw *hw) /** * ice_aq_manage_mac_read - manage MAC address read command - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @buf: a virtual buffer to hold the manage MAC read response * @buf_size: Size of the virtual buffer * @cd: pointer to command details structure or NULL @@ -262,7 +259,7 @@ static enum ice_media_type ice_get_media_type(struct ice_port_info *pi) * * Get Link Status (0x607). Returns the link status of the adapter. */ -static enum ice_status +enum ice_status ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, struct ice_link_status *link, struct ice_sq_cd *cd) { @@ -307,6 +304,8 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, hw_link_info->an_info = link_data.an_info; hw_link_info->ext_info = link_data.ext_info; hw_link_info->max_frame_size = le16_to_cpu(link_data.max_frame_size); + hw_link_info->fec_info = link_data.cfg & ICE_AQ_FEC_MASK; + hw_link_info->topo_media_conflict = link_data.topo_media_conflict; hw_link_info->pacing = link_data.cfg & ICE_AQ_CFG_PACING_M; /* update fc info */ @@ -331,7 +330,7 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, /* flag cleared so calling functions don't call AQ again */ pi->phy.get_link_info = false; - return status; + return 0; } /** @@ -358,22 +357,22 @@ static void ice_init_flex_flags(struct ice_hw *hw, enum ice_rxdid prof_id) */ case ICE_RXDID_FLEX_NIC: case ICE_RXDID_FLEX_NIC_2: - ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_PKT_FRG, - ICE_RXFLG_UDP_GRE, ICE_RXFLG_PKT_DSI, - ICE_RXFLG_FIN, idx++); + ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_PKT_FRG, + ICE_FLG_UDP_GRE, ICE_FLG_PKT_DSI, + ICE_FLG_FIN, idx++); /* flex flag 1 is not used for flexi-flag programming, skipping * these four FLG64 bits. */ - ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_SYN, ICE_RXFLG_RST, - ICE_RXFLG_PKT_DSI, ICE_RXFLG_PKT_DSI, idx++); - ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_PKT_DSI, - ICE_RXFLG_PKT_DSI, ICE_RXFLG_EVLAN_x8100, - ICE_RXFLG_EVLAN_x9100, idx++); - ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_VLAN_x8100, - ICE_RXFLG_TNL_VLAN, ICE_RXFLG_TNL_MAC, - ICE_RXFLG_TNL0, idx++); - ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_RXFLG_TNL1, ICE_RXFLG_TNL2, - ICE_RXFLG_PKT_DSI, ICE_RXFLG_PKT_DSI, idx); + ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_SYN, ICE_FLG_RST, + ICE_FLG_PKT_DSI, ICE_FLG_PKT_DSI, idx++); + ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_PKT_DSI, + ICE_FLG_PKT_DSI, ICE_FLG_EVLAN_x8100, + ICE_FLG_EVLAN_x9100, idx++); + ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_VLAN_x8100, + ICE_FLG_TNL_VLAN, ICE_FLG_TNL_MAC, + ICE_FLG_TNL0, idx++); + ICE_PROG_FLG_ENTRY(hw, prof_id, ICE_FLG_TNL1, ICE_FLG_TNL2, + ICE_FLG_PKT_DSI, ICE_FLG_PKT_DSI, idx); break; default: @@ -418,7 +417,7 @@ static void ice_init_flex_flds(struct ice_hw *hw, enum ice_rxdid prof_id) /** * ice_init_fltr_mgmt_struct - initializes filter management list and locks - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct */ static enum ice_status ice_init_fltr_mgmt_struct(struct ice_hw *hw) { @@ -438,7 +437,7 @@ static enum ice_status ice_init_fltr_mgmt_struct(struct ice_hw *hw) /** * ice_cleanup_fltr_mgmt_struct - cleanup filter management list and locks - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct */ static void ice_cleanup_fltr_mgmt_struct(struct ice_hw *hw) { @@ -476,8 +475,51 @@ static void ice_cleanup_fltr_mgmt_struct(struct ice_hw *hw) ICE_FW_LOG_DESC_SIZE(ICE_AQC_FW_LOG_ID_MAX) /** + * ice_get_fw_log_cfg - get FW logging configuration + * @hw: pointer to the HW struct + */ +static enum ice_status ice_get_fw_log_cfg(struct ice_hw *hw) +{ + struct ice_aqc_fw_logging_data *config; + struct ice_aq_desc desc; + enum ice_status status; + u16 size; + + size = ICE_FW_LOG_DESC_SIZE_MAX; + config = devm_kzalloc(ice_hw_to_dev(hw), size, GFP_KERNEL); + if (!config) + return ICE_ERR_NO_MEMORY; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_fw_logging_info); + + desc.flags |= cpu_to_le16(ICE_AQ_FLAG_BUF); + desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD); + + status = ice_aq_send_cmd(hw, &desc, config, size, NULL); + if (!status) { + u16 i; + + /* Save FW logging information into the HW structure */ + for (i = 0; i < ICE_AQC_FW_LOG_ID_MAX; i++) { + u16 v, m, flgs; + + v = le16_to_cpu(config->entry[i]); + m = (v & ICE_AQC_FW_LOG_ID_M) >> ICE_AQC_FW_LOG_ID_S; + flgs = (v & ICE_AQC_FW_LOG_EN_M) >> ICE_AQC_FW_LOG_EN_S; + + if (m < ICE_AQC_FW_LOG_ID_MAX) + hw->fw_log.evnts[m].cur = flgs; + } + } + + devm_kfree(ice_hw_to_dev(hw), config); + + return status; +} + +/** * ice_cfg_fw_log - configure FW logging - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @enable: enable certain FW logging events if true, disable all if false * * This function enables/disables the FW logging via Rx CQ events and a UART @@ -529,6 +571,11 @@ static enum ice_status ice_cfg_fw_log(struct ice_hw *hw, bool enable) (!hw->fw_log.actv_evnts || !ice_check_sq_alive(hw, &hw->adminq))) return 0; + /* Get current FW log settings */ + status = ice_get_fw_log_cfg(hw); + if (status) + return status; + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_fw_logging); cmd = &desc.params.fw_logging; @@ -626,7 +673,7 @@ out: /** * ice_output_fw_log - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @desc: pointer to the AQ message descriptor * @buf: pointer to the buffer accompanying the AQ message * @@ -634,20 +681,20 @@ out: */ void ice_output_fw_log(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf) { - ice_debug(hw, ICE_DBG_AQ_MSG, "[ FW Log Msg Start ]\n"); - ice_debug_array(hw, ICE_DBG_AQ_MSG, 16, 1, (u8 *)buf, + ice_debug(hw, ICE_DBG_FW_LOG, "[ FW Log Msg Start ]\n"); + ice_debug_array(hw, ICE_DBG_FW_LOG, 16, 1, (u8 *)buf, le16_to_cpu(desc->datalen)); - ice_debug(hw, ICE_DBG_AQ_MSG, "[ FW Log Msg End ]\n"); + ice_debug(hw, ICE_DBG_FW_LOG, "[ FW Log Msg End ]\n"); } /** * ice_get_itr_intrl_gran - determine int/intrl granularity - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * - * Determines the itr/intrl granularities based on the maximum aggregate + * Determines the ITR/intrl granularities based on the maximum aggregate * bandwidth according to the device's configuration during power-on. */ -static enum ice_status ice_get_itr_intrl_gran(struct ice_hw *hw) +static void ice_get_itr_intrl_gran(struct ice_hw *hw) { u8 max_agg_bw = (rd32(hw, GL_PWR_MODE_CTL) & GL_PWR_MODE_CTL_CAR_MAX_BW_M) >> @@ -664,13 +711,7 @@ static enum ice_status ice_get_itr_intrl_gran(struct ice_hw *hw) hw->itr_gran = ICE_ITR_GRAN_MAX_25; hw->intrl_gran = ICE_INTRL_GRAN_MAX_25; break; - default: - ice_debug(hw, ICE_DBG_INIT, - "Failed to determine itr/intrl granularity\n"); - return ICE_ERR_CFG; } - - return 0; } /** @@ -697,9 +738,7 @@ enum ice_status ice_init_hw(struct ice_hw *hw) if (status) return status; - status = ice_get_itr_intrl_gran(hw); - if (status) - return status; + ice_get_itr_intrl_gran(hw); status = ice_init_all_ctrlq(hw); if (status) @@ -731,7 +770,7 @@ enum ice_status ice_init_hw(struct ice_hw *hw) goto err_unroll_cqinit; } - /* set the back pointer to hw */ + /* set the back pointer to HW */ hw->port_info->hw = hw; /* Initialize port_info struct with switch configuration data */ @@ -823,6 +862,10 @@ err_unroll_cqinit: /** * ice_deinit_hw - unroll initialization operations done by ice_init_hw * @hw: pointer to the hardware structure + * + * This should be called only during nominal operation, not as a result of + * ice_init_hw() failing since ice_init_hw() will take care of unrolling + * applicable initializations if it fails for any reason. */ void ice_deinit_hw(struct ice_hw *hw) { @@ -988,7 +1031,7 @@ enum ice_status ice_reset(struct ice_hw *hw, enum ice_reset_req req) * @ice_rxq_ctx: pointer to the rxq context * @rxq_index: the index of the Rx queue * - * Copies rxq context from dense structure to hw register space + * Copies rxq context from dense structure to HW register space */ static enum ice_status ice_copy_rxq_ctx_to_hw(struct ice_hw *hw, u8 *ice_rxq_ctx, u32 rxq_index) @@ -1001,7 +1044,7 @@ ice_copy_rxq_ctx_to_hw(struct ice_hw *hw, u8 *ice_rxq_ctx, u32 rxq_index) if (rxq_index > QRX_CTRL_MAX_INDEX) return ICE_ERR_PARAM; - /* Copy each dword separately to hw */ + /* Copy each dword separately to HW */ for (i = 0; i < ICE_RXQ_CTX_SIZE_DWORDS; i++) { wr32(hw, QRX_CONTEXT(i, rxq_index), *((u32 *)(ice_rxq_ctx + (i * sizeof(u32))))); @@ -1045,7 +1088,7 @@ static const struct ice_ctx_ele ice_rlan_ctx_info[] = { * @rxq_index: the index of the Rx queue * * Converts rxq context from sparse to dense structure and then writes - * it to hw register space + * it to HW register space */ enum ice_status ice_write_rxq_ctx(struct ice_hw *hw, struct ice_rlan_ctx *rlan_ctx, @@ -1100,8 +1143,9 @@ const struct ice_ctx_ele ice_tlan_ctx_info[] = { * * Dumps debug log about control command with descriptor contents. */ -void ice_debug_cq(struct ice_hw *hw, u32 __maybe_unused mask, void *desc, - void *buf, u16 buf_len) +void +ice_debug_cq(struct ice_hw *hw, u32 __maybe_unused mask, void *desc, void *buf, + u16 buf_len) { struct ice_aq_desc *cq_desc = (struct ice_aq_desc *)desc; u16 len; @@ -1143,7 +1187,7 @@ void ice_debug_cq(struct ice_hw *hw, u32 __maybe_unused mask, void *desc, /** * ice_aq_send_cmd - send FW Admin Queue command to FW Admin Queue - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @desc: descriptor describing the command * @buf: buffer to use for indirect commands (NULL for direct commands) * @buf_size: size of buffer for indirect commands (0 for direct commands) @@ -1160,7 +1204,7 @@ ice_aq_send_cmd(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf, /** * ice_aq_get_fw_ver - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @cd: pointer to command details structure or NULL * * Get the firmware version (0x0001) from the admin queue commands @@ -1194,7 +1238,7 @@ enum ice_status ice_aq_get_fw_ver(struct ice_hw *hw, struct ice_sq_cd *cd) /** * ice_aq_q_shutdown - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @unloading: is the driver unloading itself * * Tell the Firmware that we're shutting down the AdminQ and whether @@ -1217,8 +1261,8 @@ enum ice_status ice_aq_q_shutdown(struct ice_hw *hw, bool unloading) /** * ice_aq_req_res - * @hw: pointer to the hw struct - * @res: resource id + * @hw: pointer to the HW struct + * @res: resource ID * @access: access type * @sdp_number: resource number * @timeout: the maximum time in ms that the driver may hold the resource @@ -1303,8 +1347,8 @@ ice_aq_req_res(struct ice_hw *hw, enum ice_aq_res_ids res, /** * ice_aq_release_res - * @hw: pointer to the hw struct - * @res: resource id + * @hw: pointer to the HW struct + * @res: resource ID * @sdp_number: resource number * @cd: pointer to command details structure or NULL * @@ -1330,7 +1374,7 @@ ice_aq_release_res(struct ice_hw *hw, enum ice_aq_res_ids res, u8 sdp_number, /** * ice_acquire_res * @hw: pointer to the HW structure - * @res: resource id + * @res: resource ID * @access: access type (read or write) * @timeout: timeout in milliseconds * @@ -1392,7 +1436,7 @@ ice_acquire_res_exit: /** * ice_release_res * @hw: pointer to the HW structure - * @res: resource id + * @res: resource ID * * This function will release a resource using the proper Admin Command. */ @@ -1404,7 +1448,7 @@ void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res) status = ice_aq_release_res(hw, res, 0, NULL); /* there are some rare cases when trying to release the resource - * results in an admin Q timeout, so handle them correctly + * results in an admin queue timeout, so handle them correctly */ while ((status == ICE_ERR_AQ_TIMEOUT) && (total_delay < hw->adminq.sq_cmd_timeout)) { @@ -1415,13 +1459,15 @@ void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res) } /** - * ice_get_guar_num_vsi - determine number of guar VSI for a PF - * @hw: pointer to the hw structure + * ice_get_num_per_func - determine number of resources per PF + * @hw: pointer to the HW structure + * @max: value to be evenly split between each PF * * Determine the number of valid functions by going through the bitmap returned - * from parsing capabilities and use this to calculate the number of VSI per PF. + * from parsing capabilities and use this to calculate the number of resources + * per PF based on the max value passed in. */ -static u32 ice_get_guar_num_vsi(struct ice_hw *hw) +static u32 ice_get_num_per_func(struct ice_hw *hw, u32 max) { u8 funcs; @@ -1432,12 +1478,12 @@ static u32 ice_get_guar_num_vsi(struct ice_hw *hw) if (!funcs) return 0; - return ICE_MAX_VSI / funcs; + return max / funcs; } /** * ice_parse_caps - parse function/device capabilities - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @buf: pointer to a buffer containing function/device capability records * @cap_count: number of capability records in the list * @opc: type of capabilities list to parse @@ -1452,6 +1498,7 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count, struct ice_hw_func_caps *func_p = NULL; struct ice_hw_dev_caps *dev_p = NULL; struct ice_hw_common_caps *caps; + char const *prefix; u32 i; if (!buf) @@ -1462,9 +1509,11 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count, if (opc == ice_aqc_opc_list_dev_caps) { dev_p = &hw->dev_caps; caps = &dev_p->common_cap; + prefix = "dev cap"; } else if (opc == ice_aqc_opc_list_func_caps) { func_p = &hw->func_caps; caps = &func_p->common_cap; + prefix = "func cap"; } else { ice_debug(hw, ICE_DBG_INIT, "wrong opcode\n"); return; @@ -1480,28 +1529,29 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count, case ICE_AQC_CAPS_VALID_FUNCTIONS: caps->valid_functions = number; ice_debug(hw, ICE_DBG_INIT, - "HW caps: Valid Functions = %d\n", + "%s: valid functions = %d\n", prefix, caps->valid_functions); break; case ICE_AQC_CAPS_SRIOV: caps->sr_iov_1_1 = (number == 1); ice_debug(hw, ICE_DBG_INIT, - "HW caps: SR-IOV = %d\n", caps->sr_iov_1_1); + "%s: SR-IOV = %d\n", prefix, + caps->sr_iov_1_1); break; case ICE_AQC_CAPS_VF: if (dev_p) { dev_p->num_vfs_exposed = number; ice_debug(hw, ICE_DBG_INIT, - "HW caps: VFs exposed = %d\n", + "%s: VFs exposed = %d\n", prefix, dev_p->num_vfs_exposed); } else if (func_p) { func_p->num_allocd_vfs = number; func_p->vf_base_id = logical_id; ice_debug(hw, ICE_DBG_INIT, - "HW caps: VFs allocated = %d\n", + "%s: VFs allocated = %d\n", prefix, func_p->num_allocd_vfs); ice_debug(hw, ICE_DBG_INIT, - "HW caps: VF base_id = %d\n", + "%s: VF base_id = %d\n", prefix, func_p->vf_base_id); } break; @@ -1509,68 +1559,69 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count, if (dev_p) { dev_p->num_vsi_allocd_to_host = number; ice_debug(hw, ICE_DBG_INIT, - "HW caps: Dev.VSI cnt = %d\n", + "%s: num VSI alloc to host = %d\n", + prefix, dev_p->num_vsi_allocd_to_host); } else if (func_p) { - func_p->guar_num_vsi = ice_get_guar_num_vsi(hw); + func_p->guar_num_vsi = + ice_get_num_per_func(hw, ICE_MAX_VSI); + ice_debug(hw, ICE_DBG_INIT, + "%s: num guaranteed VSI (fw) = %d\n", + prefix, number); ice_debug(hw, ICE_DBG_INIT, - "HW caps: Func.VSI cnt = %d\n", - number); + "%s: num guaranteed VSI = %d\n", + prefix, func_p->guar_num_vsi); } break; case ICE_AQC_CAPS_RSS: caps->rss_table_size = number; caps->rss_table_entry_width = logical_id; ice_debug(hw, ICE_DBG_INIT, - "HW caps: RSS table size = %d\n", + "%s: RSS table size = %d\n", prefix, caps->rss_table_size); ice_debug(hw, ICE_DBG_INIT, - "HW caps: RSS table width = %d\n", + "%s: RSS table width = %d\n", prefix, caps->rss_table_entry_width); break; case ICE_AQC_CAPS_RXQS: caps->num_rxq = number; caps->rxq_first_id = phys_id; ice_debug(hw, ICE_DBG_INIT, - "HW caps: Num Rx Qs = %d\n", caps->num_rxq); + "%s: num Rx queues = %d\n", prefix, + caps->num_rxq); ice_debug(hw, ICE_DBG_INIT, - "HW caps: Rx first queue ID = %d\n", + "%s: Rx first queue ID = %d\n", prefix, caps->rxq_first_id); break; case ICE_AQC_CAPS_TXQS: caps->num_txq = number; caps->txq_first_id = phys_id; ice_debug(hw, ICE_DBG_INIT, - "HW caps: Num Tx Qs = %d\n", caps->num_txq); + "%s: num Tx queues = %d\n", prefix, + caps->num_txq); ice_debug(hw, ICE_DBG_INIT, - "HW caps: Tx first queue ID = %d\n", + "%s: Tx first queue ID = %d\n", prefix, caps->txq_first_id); break; case ICE_AQC_CAPS_MSIX: caps->num_msix_vectors = number; caps->msix_vector_first_id = phys_id; ice_debug(hw, ICE_DBG_INIT, - "HW caps: MSIX vector count = %d\n", + "%s: MSIX vector count = %d\n", prefix, caps->num_msix_vectors); ice_debug(hw, ICE_DBG_INIT, - "HW caps: MSIX first vector index = %d\n", + "%s: MSIX first vector index = %d\n", prefix, caps->msix_vector_first_id); break; case ICE_AQC_CAPS_MAX_MTU: caps->max_mtu = number; - if (dev_p) - ice_debug(hw, ICE_DBG_INIT, - "HW caps: Dev.MaxMTU = %d\n", - caps->max_mtu); - else if (func_p) - ice_debug(hw, ICE_DBG_INIT, - "HW caps: func.MaxMTU = %d\n", - caps->max_mtu); + ice_debug(hw, ICE_DBG_INIT, "%s: max MTU = %d\n", + prefix, caps->max_mtu); break; default: ice_debug(hw, ICE_DBG_INIT, - "HW caps: Unknown capability[%d]: 0x%x\n", i, - cap); + "%s: unknown capability[%d]: 0x%x\n", prefix, + i, cap); break; } } @@ -1578,7 +1629,7 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count, /** * ice_aq_discover_caps - query function/device capabilities - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @buf: a virtual buffer to hold the capabilities * @buf_size: Size of the virtual buffer * @cap_count: cap count needed if AQ err==ENOMEM @@ -1617,8 +1668,8 @@ ice_aq_discover_caps(struct ice_hw *hw, void *buf, u16 buf_size, u32 *cap_count, * @hw: pointer to the hardware structure * @opc: capabilities type to discover - pass in the command opcode */ -static enum ice_status ice_discover_caps(struct ice_hw *hw, - enum ice_adminq_opc opc) +static enum ice_status +ice_discover_caps(struct ice_hw *hw, enum ice_adminq_opc opc) { enum ice_status status; u32 cap_count; @@ -1677,7 +1728,7 @@ enum ice_status ice_get_caps(struct ice_hw *hw) /** * ice_aq_manage_mac_write - manage MAC address write command - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @mac_addr: MAC address to be written as LAA/LAA+WoL/Port address * @flags: flags to control write behavior * @cd: pointer to command details structure or NULL @@ -1705,7 +1756,7 @@ ice_aq_manage_mac_write(struct ice_hw *hw, const u8 *mac_addr, u8 flags, /** * ice_aq_clear_pxe_mode - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Tell the firmware that the driver is taking over from PXE (0x0110). */ @@ -1721,7 +1772,7 @@ static enum ice_status ice_aq_clear_pxe_mode(struct ice_hw *hw) /** * ice_clear_pxe_mode - clear pxe operations mode - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Make sure all PXE mode settings are cleared, including things * like descriptor fetch/write-back mode. @@ -1737,10 +1788,10 @@ void ice_clear_pxe_mode(struct ice_hw *hw) * @phy_type_low: lower part of phy_type * @phy_type_high: higher part of phy_type * - * This helper function will convert an entry in phy type structure + * This helper function will convert an entry in PHY type structure * [phy_type_low, phy_type_high] to its corresponding link speed. * Note: In the structure of [phy_type_low, phy_type_high], there should - * be one bit set, as this function will convert one phy type to its + * be one bit set, as this function will convert one PHY type to its * speed. * If no bit gets set, ICE_LINK_SPEED_UNKNOWN will be returned * If more than one bit gets set, ICE_LINK_SPEED_UNKNOWN will be returned @@ -1884,10 +1935,10 @@ void ice_update_phy_type(u64 *phy_type_low, u64 *phy_type_high, u16 link_speeds_bitmap) { - u16 speed = ICE_AQ_LINK_SPEED_UNKNOWN; u64 pt_high; u64 pt_low; int index; + u16 speed; /* We first check with low part of phy_type */ for (index = 0; index <= ICE_PHY_TYPE_LOW_MAX_INDEX; index++) { @@ -1910,7 +1961,7 @@ ice_update_phy_type(u64 *phy_type_low, u64 *phy_type_high, /** * ice_aq_set_phy_cfg - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @lport: logical port number * @cfg: structure with PHY configuration data to be set * @cd: pointer to command details structure or NULL @@ -1929,6 +1980,15 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport, if (!cfg) return ICE_ERR_PARAM; + /* Ensure that only valid bits of cfg->caps can be turned on. */ + if (cfg->caps & ~ICE_AQ_PHY_ENA_VALID_MASK) { + ice_debug(hw, ICE_DBG_PHY, + "Invalid bit is set in ice_aqc_set_phy_cfg_data->caps : 0x%x\n", + cfg->caps); + + cfg->caps &= ICE_AQ_PHY_ENA_VALID_MASK; + } + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_phy_cfg); desc.params.set_phy.lport_num = lport; desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD); @@ -1942,36 +2002,37 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport, */ enum ice_status ice_update_link_info(struct ice_port_info *pi) { - struct ice_aqc_get_phy_caps_data *pcaps; - struct ice_phy_info *phy_info; + struct ice_link_status *li; enum ice_status status; - struct ice_hw *hw; if (!pi) return ICE_ERR_PARAM; - hw = pi->hw; - - pcaps = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*pcaps), GFP_KERNEL); - if (!pcaps) - return ICE_ERR_NO_MEMORY; + li = &pi->phy.link_info; - phy_info = &pi->phy; status = ice_aq_get_link_info(pi, true, NULL, NULL); if (status) - goto out; + return status; + + if (li->link_info & ICE_AQ_MEDIA_AVAILABLE) { + struct ice_aqc_get_phy_caps_data *pcaps; + struct ice_hw *hw; + + hw = pi->hw; + pcaps = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*pcaps), + GFP_KERNEL); + if (!pcaps) + return ICE_ERR_NO_MEMORY; - if (phy_info->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) { status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps, NULL); - if (status) - goto out; + if (!status) + memcpy(li->module_type, &pcaps->module_type, + sizeof(li->module_type)); - memcpy(phy_info->link_info.module_type, &pcaps->module_type, - sizeof(phy_info->link_info.module_type)); + devm_kfree(ice_hw_to_dev(hw), pcaps); } -out: - devm_kfree(ice_hw_to_dev(hw), pcaps); + return status; } @@ -2016,7 +2077,7 @@ ice_set_fc(struct ice_port_info *pi, u8 *aq_failures, bool ena_auto_link_update) if (!pcaps) return ICE_ERR_NO_MEMORY; - /* Get the current phy config */ + /* Get the current PHY config */ status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps, NULL); if (status) { @@ -2027,8 +2088,10 @@ ice_set_fc(struct ice_port_info *pi, u8 *aq_failures, bool ena_auto_link_update) /* clear the old pause settings */ cfg.caps = pcaps->caps & ~(ICE_AQC_PHY_EN_TX_LINK_PAUSE | ICE_AQC_PHY_EN_RX_LINK_PAUSE); + /* set the new capabilities */ cfg.caps |= pause_mask; + /* If the capabilities have changed, then set the new config */ if (cfg.caps != pcaps->caps) { int retry_count, retry_max = 10; @@ -2074,6 +2137,74 @@ out: } /** + * ice_copy_phy_caps_to_cfg - Copy PHY ability data to configuration data + * @caps: PHY ability structure to copy date from + * @cfg: PHY configuration structure to copy data to + * + * Helper function to copy AQC PHY get ability data to PHY set configuration + * data structure + */ +void +ice_copy_phy_caps_to_cfg(struct ice_aqc_get_phy_caps_data *caps, + struct ice_aqc_set_phy_cfg_data *cfg) +{ + if (!caps || !cfg) + return; + + cfg->phy_type_low = caps->phy_type_low; + cfg->phy_type_high = caps->phy_type_high; + cfg->caps = caps->caps; + cfg->low_power_ctrl = caps->low_power_ctrl; + cfg->eee_cap = caps->eee_cap; + cfg->eeer_value = caps->eeer_value; + cfg->link_fec_opt = caps->link_fec_options; +} + +/** + * ice_cfg_phy_fec - Configure PHY FEC data based on FEC mode + * @cfg: PHY configuration data to set FEC mode + * @fec: FEC mode to configure + * + * Caller should copy ice_aqc_get_phy_caps_data.caps ICE_AQC_PHY_EN_AUTO_FEC + * (bit 7) and ice_aqc_get_phy_caps_data.link_fec_options to cfg.caps + * ICE_AQ_PHY_ENA_AUTO_FEC (bit 7) and cfg.link_fec_options before calling. + */ +void +ice_cfg_phy_fec(struct ice_aqc_set_phy_cfg_data *cfg, enum ice_fec_mode fec) +{ + switch (fec) { + case ICE_FEC_BASER: + /* Clear auto FEC and RS bits, and AND BASE-R ability + * bits and OR request bits. + */ + cfg->caps &= ~ICE_AQC_PHY_EN_AUTO_FEC; + cfg->link_fec_opt &= ICE_AQC_PHY_FEC_10G_KR_40G_KR4_EN | + ICE_AQC_PHY_FEC_25G_KR_CLAUSE74_EN; + cfg->link_fec_opt |= ICE_AQC_PHY_FEC_10G_KR_40G_KR4_REQ | + ICE_AQC_PHY_FEC_25G_KR_REQ; + break; + case ICE_FEC_RS: + /* Clear auto FEC and BASE-R bits, and AND RS ability + * bits and OR request bits. + */ + cfg->caps &= ~ICE_AQC_PHY_EN_AUTO_FEC; + cfg->link_fec_opt &= ICE_AQC_PHY_FEC_25G_RS_CLAUSE91_EN; + cfg->link_fec_opt |= ICE_AQC_PHY_FEC_25G_RS_528_REQ | + ICE_AQC_PHY_FEC_25G_RS_544_REQ; + break; + case ICE_FEC_NONE: + /* Clear auto FEC and all FEC option bits. */ + cfg->caps &= ~ICE_AQC_PHY_EN_AUTO_FEC; + cfg->link_fec_opt &= ~ICE_AQC_PHY_FEC_MASK; + break; + case ICE_FEC_AUTO: + /* AND auto FEC bit, and all caps bits. */ + cfg->caps &= ICE_AQC_PHY_CAPS_MASK; + break; + } +} + +/** * ice_get_link_status - get status of the HW network link * @pi: port information structure * @link_up: pointer to bool (true/false = linkup/linkdown) @@ -2136,6 +2267,55 @@ ice_aq_set_link_restart_an(struct ice_port_info *pi, bool ena_link, } /** + * ice_aq_set_event_mask + * @hw: pointer to the HW struct + * @port_num: port number of the physical function + * @mask: event mask to be set + * @cd: pointer to command details structure or NULL + * + * Set event mask (0x0613) + */ +enum ice_status +ice_aq_set_event_mask(struct ice_hw *hw, u8 port_num, u16 mask, + struct ice_sq_cd *cd) +{ + struct ice_aqc_set_event_mask *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.set_event_mask; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_event_mask); + + cmd->lport_num = port_num; + + cmd->event_mask = cpu_to_le16(mask); + return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); +} + +/** + * ice_aq_set_mac_loopback + * @hw: pointer to the HW struct + * @ena_lpbk: Enable or Disable loopback + * @cd: pointer to command details structure or NULL + * + * Enable/disable loopback on a given port + */ +enum ice_status +ice_aq_set_mac_loopback(struct ice_hw *hw, bool ena_lpbk, struct ice_sq_cd *cd) +{ + struct ice_aqc_set_mac_lb *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.set_mac_lb; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_mac_lb); + if (ena_lpbk) + cmd->lb_mode = ICE_AQ_MAC_LB_EN; + + return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); +} + +/** * ice_aq_set_port_id_led * @pi: pointer to the port information * @is_orig_mode: is this LED set to original mode (by the net-list) @@ -2297,7 +2477,7 @@ ice_aq_set_rss_lut(struct ice_hw *hw, u16 vsi_handle, u8 lut_type, /** * __ice_aq_get_set_rss_key - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_id: VSI FW index * @key: pointer to key info struct * @set: set true to set the key, false to get the key @@ -2332,7 +2512,7 @@ ice_status __ice_aq_get_set_rss_key(struct ice_hw *hw, u16 vsi_id, /** * ice_aq_get_rss_key - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: software VSI handle * @key: pointer to key info struct * @@ -2351,7 +2531,7 @@ ice_aq_get_rss_key(struct ice_hw *hw, u16 vsi_handle, /** * ice_aq_set_rss_key - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: software VSI handle * @keys: pointer to key info struct * @@ -2436,7 +2616,7 @@ ice_aq_add_lan_txq(struct ice_hw *hw, u8 num_qgrps, * @num_qgrps: number of groups in the list * @qg_list: the list of groups to disable * @buf_size: the total size of the qg_list buffer in bytes - * @rst_src: if called due to reset, specifies the RST source + * @rst_src: if called due to reset, specifies the reset source * @vmvf_num: the relative VM or VF number that is undergoing the reset * @cd: pointer to command details structure or NULL * @@ -2476,7 +2656,7 @@ ice_aq_dis_lan_txq(struct ice_hw *hw, u8 num_qgrps, break; case ICE_VF_RESET: cmd->cmd_type = ICE_AQC_Q_DIS_CMD_VF_RESET; - /* In this case, FW expects vmvf_num to be absolute VF id */ + /* In this case, FW expects vmvf_num to be absolute VF ID */ cmd->vmvf_and_timeout |= cpu_to_le16((vmvf_num + hw->func_caps.vf_base_id) & ICE_AQC_Q_DIS_VMVF_NUM_M); @@ -2519,7 +2699,7 @@ do_aq: ice_debug(hw, ICE_DBG_SCHED, "VM%d disable failed %d\n", vmvf_num, hw->adminq.sq_last_status); else - ice_debug(hw, ICE_DBG_SCHED, "disable Q %d failed %d\n", + ice_debug(hw, ICE_DBG_SCHED, "disable queue %d failed %d\n", le16_to_cpu(qg_list[0].q_id[0]), hw->adminq.sq_last_status); } @@ -2534,8 +2714,8 @@ do_aq: * @dest_ctx: the context to be written to * @ce_info: a description of the struct to be filled */ -static void ice_write_byte(u8 *src_ctx, u8 *dest_ctx, - const struct ice_ctx_ele *ce_info) +static void +ice_write_byte(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) { u8 src_byte, dest_byte, mask; u8 *from, *dest; @@ -2573,8 +2753,8 @@ static void ice_write_byte(u8 *src_ctx, u8 *dest_ctx, * @dest_ctx: the context to be written to * @ce_info: a description of the struct to be filled */ -static void ice_write_word(u8 *src_ctx, u8 *dest_ctx, - const struct ice_ctx_ele *ce_info) +static void +ice_write_word(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) { u16 src_word, mask; __le16 dest_word; @@ -2616,8 +2796,8 @@ static void ice_write_word(u8 *src_ctx, u8 *dest_ctx, * @dest_ctx: the context to be written to * @ce_info: a description of the struct to be filled */ -static void ice_write_dword(u8 *src_ctx, u8 *dest_ctx, - const struct ice_ctx_ele *ce_info) +static void +ice_write_dword(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) { u32 src_dword, mask; __le32 dest_dword; @@ -2667,8 +2847,8 @@ static void ice_write_dword(u8 *src_ctx, u8 *dest_ctx, * @dest_ctx: the context to be written to * @ce_info: a description of the struct to be filled */ -static void ice_write_qword(u8 *src_ctx, u8 *dest_ctx, - const struct ice_ctx_ele *ce_info) +static void +ice_write_qword(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) { u64 src_qword, mask; __le64 dest_qword; @@ -2750,24 +2930,50 @@ ice_set_ctx(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) } /** + * ice_get_lan_q_ctx - get the LAN queue context for the given VSI and TC + * @hw: pointer to the HW struct + * @vsi_handle: software VSI handle + * @tc: TC number + * @q_handle: software queue handle + */ +static struct ice_q_ctx * +ice_get_lan_q_ctx(struct ice_hw *hw, u16 vsi_handle, u8 tc, u16 q_handle) +{ + struct ice_vsi_ctx *vsi; + struct ice_q_ctx *q_ctx; + + vsi = ice_get_vsi_ctx(hw, vsi_handle); + if (!vsi) + return NULL; + if (q_handle >= vsi->num_lan_q_entries[tc]) + return NULL; + if (!vsi->lan_q_ctx[tc]) + return NULL; + q_ctx = vsi->lan_q_ctx[tc]; + return &q_ctx[q_handle]; +} + +/** * ice_ena_vsi_txq * @pi: port information structure * @vsi_handle: software VSI handle - * @tc: tc number + * @tc: TC number + * @q_handle: software queue handle * @num_qgrps: Number of added queue groups * @buf: list of queue groups to be added * @buf_size: size of buffer for indirect command * @cd: pointer to command details structure or NULL * - * This function adds one lan q + * This function adds one LAN queue */ enum ice_status -ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_qgrps, - struct ice_aqc_add_tx_qgrp *buf, u16 buf_size, +ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 q_handle, + u8 num_qgrps, struct ice_aqc_add_tx_qgrp *buf, u16 buf_size, struct ice_sq_cd *cd) { struct ice_aqc_txsched_elem_data node = { 0 }; struct ice_sched_node *parent; + struct ice_q_ctx *q_ctx; enum ice_status status; struct ice_hw *hw; @@ -2784,6 +2990,14 @@ ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_qgrps, mutex_lock(&pi->sched_lock); + q_ctx = ice_get_lan_q_ctx(hw, vsi_handle, tc, q_handle); + if (!q_ctx) { + ice_debug(hw, ICE_DBG_SCHED, "Enaq: invalid queue handle %d\n", + q_handle); + status = ICE_ERR_PARAM; + goto ena_txq_exit; + } + /* find a parent node */ parent = ice_sched_get_free_qparent(pi, vsi_handle, tc, ICE_SCHED_NODE_OWNER_LAN); @@ -2803,14 +3017,14 @@ ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_qgrps, * Bit 5-6. * - Bit 7 is reserved. * Without setting the generic section as valid in valid_sections, the - * Admin Q command will fail with error code ICE_AQ_RC_EINVAL. + * Admin queue command will fail with error code ICE_AQ_RC_EINVAL. */ buf->txqs[0].info.valid_sections = ICE_AQC_ELEM_VALID_GENERIC; - /* add the lan q */ + /* add the LAN queue */ status = ice_aq_add_lan_txq(hw, num_qgrps, buf, buf_size, cd); if (status) { - ice_debug(hw, ICE_DBG_SCHED, "enable Q %d failed %d\n", + ice_debug(hw, ICE_DBG_SCHED, "enable queue %d failed %d\n", le16_to_cpu(buf->txqs[0].txq_id), hw->adminq.sq_last_status); goto ena_txq_exit; @@ -2818,8 +3032,9 @@ ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_qgrps, node.node_teid = buf->txqs[0].q_teid; node.data.elem_type = ICE_AQC_ELEM_TYPE_LEAF; + q_ctx->q_handle = q_handle; - /* add a leaf node into schduler tree q layer */ + /* add a leaf node into schduler tree queue layer */ status = ice_sched_add_node(pi, hw->num_tx_sched_layers - 1, &node); ena_txq_exit: @@ -2830,35 +3045,42 @@ ena_txq_exit: /** * ice_dis_vsi_txq * @pi: port information structure + * @vsi_handle: software VSI handle + * @tc: TC number * @num_queues: number of queues + * @q_handles: pointer to software queue handle array * @q_ids: pointer to the q_id array * @q_teids: pointer to queue node teids - * @rst_src: if called due to reset, specifies the RST source + * @rst_src: if called due to reset, specifies the reset source * @vmvf_num: the relative VM or VF number that is undergoing the reset * @cd: pointer to command details structure or NULL * * This function removes queues and their corresponding nodes in SW DB */ enum ice_status -ice_dis_vsi_txq(struct ice_port_info *pi, u8 num_queues, u16 *q_ids, - u32 *q_teids, enum ice_disq_rst_src rst_src, u16 vmvf_num, +ice_dis_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_queues, + u16 *q_handles, u16 *q_ids, u32 *q_teids, + enum ice_disq_rst_src rst_src, u16 vmvf_num, struct ice_sq_cd *cd) { enum ice_status status = ICE_ERR_DOES_NOT_EXIST; struct ice_aqc_dis_txq_item qg_list; + struct ice_q_ctx *q_ctx; u16 i; if (!pi || pi->port_state != ICE_SCHED_PORT_STATE_READY) return ICE_ERR_CFG; - /* if queue is disabled already yet the disable queue command has to be - * sent to complete the VF reset, then call ice_aq_dis_lan_txq without - * any queue information - */ - - if (!num_queues && rst_src) - return ice_aq_dis_lan_txq(pi->hw, 0, NULL, 0, rst_src, vmvf_num, - NULL); + if (!num_queues) { + /* if queue is disabled already yet the disable queue command + * has to be sent to complete the VF reset, then call + * ice_aq_dis_lan_txq without any queue information + */ + if (rst_src) + return ice_aq_dis_lan_txq(pi->hw, 0, NULL, 0, rst_src, + vmvf_num, NULL); + return ICE_ERR_CFG; + } mutex_lock(&pi->sched_lock); @@ -2868,6 +3090,17 @@ ice_dis_vsi_txq(struct ice_port_info *pi, u8 num_queues, u16 *q_ids, node = ice_sched_find_node_by_teid(pi->root, q_teids[i]); if (!node) continue; + q_ctx = ice_get_lan_q_ctx(pi->hw, vsi_handle, tc, q_handles[i]); + if (!q_ctx) { + ice_debug(pi->hw, ICE_DBG_SCHED, "invalid queue handle%d\n", + q_handles[i]); + continue; + } + if (q_ctx->q_handle != q_handles[i]) { + ice_debug(pi->hw, ICE_DBG_SCHED, "Err:handles %d %d\n", + q_ctx->q_handle, q_handles[i]); + continue; + } qg_list.parent_teid = node->info.parent_teid; qg_list.num_qs = 1; qg_list.q_id[0] = cpu_to_le16(q_ids[i]); @@ -2878,18 +3111,19 @@ ice_dis_vsi_txq(struct ice_port_info *pi, u8 num_queues, u16 *q_ids, if (status) break; ice_free_sched_node(pi, node); + q_ctx->q_handle = ICE_INVAL_Q_HANDLE; } mutex_unlock(&pi->sched_lock); return status; } /** - * ice_cfg_vsi_qs - configure the new/exisiting VSI queues + * ice_cfg_vsi_qs - configure the new/existing VSI queues * @pi: port information structure * @vsi_handle: software VSI handle * @tc_bitmap: TC bitmap * @maxqs: max queues array per TC - * @owner: lan or rdma + * @owner: LAN or RDMA * * This function adds/updates the VSI queues per TC. */ @@ -2908,7 +3142,7 @@ ice_cfg_vsi_qs(struct ice_port_info *pi, u16 vsi_handle, u8 tc_bitmap, mutex_lock(&pi->sched_lock); - for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) { + ice_for_each_traffic_class(i) { /* configuration is possible only if TC node is present */ if (!ice_sched_get_tc_node(pi, i)) continue; @@ -2924,13 +3158,13 @@ ice_cfg_vsi_qs(struct ice_port_info *pi, u16 vsi_handle, u8 tc_bitmap, } /** - * ice_cfg_vsi_lan - configure VSI lan queues + * ice_cfg_vsi_lan - configure VSI LAN queues * @pi: port information structure * @vsi_handle: software VSI handle * @tc_bitmap: TC bitmap - * @max_lanqs: max lan queues array per TC + * @max_lanqs: max LAN queues array per TC * - * This function adds/updates the VSI lan queues per TC. + * This function adds/updates the VSI LAN queues per TC. */ enum ice_status ice_cfg_vsi_lan(struct ice_port_info *pi, u16 vsi_handle, u8 tc_bitmap, @@ -2942,7 +3176,7 @@ ice_cfg_vsi_lan(struct ice_port_info *pi, u16 vsi_handle, u8 tc_bitmap, /** * ice_replay_pre_init - replay pre initialization - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Initializes required config data for VSI, FD, ACL, and RSS before replay. */ @@ -2966,7 +3200,7 @@ static enum ice_status ice_replay_pre_init(struct ice_hw *hw) /** * ice_replay_vsi - replay VSI configuration - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: driver VSI handle * * Restore all VSI configuration after reset. It is required to call this @@ -2993,7 +3227,7 @@ enum ice_status ice_replay_vsi(struct ice_hw *hw, u16 vsi_handle) /** * ice_replay_post - post replay configuration cleanup - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Post replay cleanup. */ @@ -3012,8 +3246,9 @@ void ice_replay_post(struct ice_hw *hw) * @prev_stat: ptr to previous loaded stat value * @cur_stat: ptr to current stat value */ -void ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg, - bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat) +void +ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg, + bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat) { u64 new_data; @@ -3043,8 +3278,9 @@ void ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg, * @prev_stat: ptr to previous loaded stat value * @cur_stat: ptr to current stat value */ -void ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, - u64 *prev_stat, u64 *cur_stat) +void +ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, + u64 *prev_stat, u64 *cur_stat) { u32 new_data; @@ -3063,3 +3299,28 @@ void ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, /* to manage the potential roll-over */ *cur_stat = (new_data + BIT_ULL(32)) - *prev_stat; } + +/** + * ice_sched_query_elem - query element information from HW + * @hw: pointer to the HW struct + * @node_teid: node TEID to be queried + * @buf: buffer to element information + * + * This function queries HW element information + */ +enum ice_status +ice_sched_query_elem(struct ice_hw *hw, u32 node_teid, + struct ice_aqc_get_elem *buf) +{ + u16 buf_size, num_elem_ret = 0; + enum ice_status status; + + buf_size = sizeof(*buf); + memset(buf, 0, buf_size); + buf->generic[0].node_teid = cpu_to_le32(node_teid); + status = ice_aq_query_sched_elems(hw, 1, buf, buf_size, &num_elem_ret, + NULL); + if (status || num_elem_ret != 1) + ice_debug(hw, ICE_DBG_SCHED, "query element failed\n"); + return status; +} diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h index d7c7c2ed8823..d1f8353fe6bb 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.h +++ b/drivers/net/ethernet/intel/ice/ice_common.h @@ -9,8 +9,10 @@ #include "ice_switch.h" #include <linux/avf/virtchnl.h> -void ice_debug_cq(struct ice_hw *hw, u32 mask, void *desc, void *buf, - u16 buf_len); +enum ice_status ice_nvm_validate_checksum(struct ice_hw *hw); + +void +ice_debug_cq(struct ice_hw *hw, u32 mask, void *desc, void *buf, u16 buf_len); enum ice_status ice_init_hw(struct ice_hw *hw); void ice_deinit_hw(struct ice_hw *hw); enum ice_status ice_check_reset(struct ice_hw *hw); @@ -28,8 +30,8 @@ ice_acquire_res(struct ice_hw *hw, enum ice_aq_res_ids res, enum ice_aq_res_access_type access, u32 timeout); void ice_release_res(struct ice_hw *hw, enum ice_aq_res_ids res); enum ice_status ice_init_nvm(struct ice_hw *hw); -enum ice_status ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, - u16 *data); +enum ice_status +ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data); enum ice_status ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq, struct ice_aq_desc *desc, void *buf, u16 buf_size, @@ -84,30 +86,49 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport, enum ice_status ice_set_fc(struct ice_port_info *pi, u8 *aq_failures, bool ena_auto_link_update); - +void +ice_cfg_phy_fec(struct ice_aqc_set_phy_cfg_data *cfg, enum ice_fec_mode fec); +void +ice_copy_phy_caps_to_cfg(struct ice_aqc_get_phy_caps_data *caps, + struct ice_aqc_set_phy_cfg_data *cfg); enum ice_status ice_aq_set_link_restart_an(struct ice_port_info *pi, bool ena_link, struct ice_sq_cd *cd); enum ice_status +ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, + struct ice_link_status *link, struct ice_sq_cd *cd); +enum ice_status +ice_aq_set_event_mask(struct ice_hw *hw, u8 port_num, u16 mask, + struct ice_sq_cd *cd); +enum ice_status +ice_aq_set_mac_loopback(struct ice_hw *hw, bool ena_lpbk, struct ice_sq_cd *cd); + +enum ice_status ice_aq_set_port_id_led(struct ice_port_info *pi, bool is_orig_mode, struct ice_sq_cd *cd); enum ice_status -ice_dis_vsi_txq(struct ice_port_info *pi, u8 num_queues, u16 *q_ids, - u32 *q_teids, enum ice_disq_rst_src rst_src, u16 vmvf_num, - struct ice_sq_cd *cmd_details); +ice_dis_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_queues, + u16 *q_handle, u16 *q_ids, u32 *q_teids, + enum ice_disq_rst_src rst_src, u16 vmvf_num, + struct ice_sq_cd *cd); enum ice_status ice_cfg_vsi_lan(struct ice_port_info *pi, u16 vsi_handle, u8 tc_bitmap, u16 *max_lanqs); enum ice_status -ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u8 num_qgrps, - struct ice_aqc_add_tx_qgrp *buf, u16 buf_size, +ice_ena_vsi_txq(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 q_handle, + u8 num_qgrps, struct ice_aqc_add_tx_qgrp *buf, u16 buf_size, struct ice_sq_cd *cd); enum ice_status ice_replay_vsi(struct ice_hw *hw, u16 vsi_handle); void ice_replay_post(struct ice_hw *hw); void ice_output_fw_log(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf); -void ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg, - bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat); -void ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, - u64 *prev_stat, u64 *cur_stat); +void +ice_stat_update40(struct ice_hw *hw, u32 hireg, u32 loreg, + bool prev_stat_loaded, u64 *prev_stat, u64 *cur_stat); +void +ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, + u64 *prev_stat, u64 *cur_stat); +enum ice_status +ice_sched_query_elem(struct ice_hw *hw, u32 node_teid, + struct ice_aqc_get_elem *buf); #endif /* _ICE_COMMON_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c index 2bf5e11f559a..e91ac4df0242 100644 --- a/drivers/net/ethernet/intel/ice/ice_controlq.c +++ b/drivers/net/ethernet/intel/ice/ice_controlq.c @@ -51,7 +51,7 @@ static void ice_mailbox_init_regs(struct ice_hw *hw) /** * ice_check_sq_alive - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @cq: pointer to the specific Control queue * * Returns true if Queue is enabled else false. @@ -287,7 +287,7 @@ ice_cfg_sq_regs(struct ice_hw *hw, struct ice_ctl_q_info *cq) * @hw: pointer to the hardware structure * @cq: pointer to the specific Control queue * - * Configure base address and length registers for the receive (event q) + * Configure base address and length registers for the receive (event queue) */ static enum ice_status ice_cfg_rq_regs(struct ice_hw *hw, struct ice_ctl_q_info *cq) @@ -439,7 +439,7 @@ do { \ /* free the buffer info list */ \ if ((qi)->ring.cmd_buf) \ devm_kfree(ice_hw_to_dev(hw), (qi)->ring.cmd_buf); \ - /* free dma head */ \ + /* free DMA head */ \ devm_kfree(ice_hw_to_dev(hw), (qi)->ring.dma_head); \ } while (0) @@ -751,7 +751,7 @@ static u16 ice_clean_sq(struct ice_hw *hw, struct ice_ctl_q_info *cq) /** * ice_sq_done - check if FW has processed the Admin Send Queue (ATQ) - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @cq: pointer to the specific Control queue * * Returns true if the firmware has processed all descriptors on the @@ -767,7 +767,7 @@ static bool ice_sq_done(struct ice_hw *hw, struct ice_ctl_q_info *cq) /** * ice_sq_send_cmd - send command to Control Queue (ATQ) - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @cq: pointer to the specific Control queue * @desc: prefilled descriptor describing the command (non DMA mem) * @buf: buffer to use for indirect commands (or NULL for direct commands) @@ -962,7 +962,7 @@ void ice_fill_dflt_direct_cmd_desc(struct ice_aq_desc *desc, u16 opcode) /** * ice_clean_rq_elem - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @cq: pointer to the specific Control queue * @e: event info from the receive descriptor, includes any buffers * @pending: number of events that could be left to process diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.h b/drivers/net/ethernet/intel/ice/ice_controlq.h index 0038a4109c99..44945c2165d8 100644 --- a/drivers/net/ethernet/intel/ice/ice_controlq.h +++ b/drivers/net/ethernet/intel/ice/ice_controlq.h @@ -35,7 +35,7 @@ enum ice_ctl_q { #define ICE_CTL_Q_SQ_CMD_TIMEOUT 250 /* msecs */ struct ice_ctl_q_ring { - void *dma_head; /* Virtual address to dma head */ + void *dma_head; /* Virtual address to DMA head */ struct ice_dma_mem desc_buf; /* descriptor ring memory */ void *cmd_buf; /* command buffer memory */ @@ -79,6 +79,7 @@ struct ice_rq_event_info { /* Control Queue information */ struct ice_ctl_q_info { enum ice_ctl_q qtype; + enum ice_aq_err rq_last_status; /* last status on receive queue */ struct ice_ctl_q_ring rq; /* receive queue */ struct ice_ctl_q_ring sq; /* send queue */ u32 sq_cmd_timeout; /* send queue cmd write back timeout */ @@ -86,10 +87,9 @@ struct ice_ctl_q_info { u16 num_sq_entries; /* send queue depth */ u16 rq_buf_size; /* receive queue buffer size */ u16 sq_buf_size; /* send queue buffer size */ + enum ice_aq_err sq_last_status; /* last status on send queue */ struct mutex sq_lock; /* Send queue lock */ struct mutex rq_lock; /* Receive queue lock */ - enum ice_aq_err sq_last_status; /* last status on send queue */ - enum ice_aq_err rq_last_status; /* last status on receive queue */ }; #endif /* _ICE_CONTROLQ_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_dcb.c b/drivers/net/ethernet/intel/ice/ice_dcb.c new file mode 100644 index 000000000000..c2002ded65f6 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_dcb.c @@ -0,0 +1,1403 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2019, Intel Corporation. */ + +#include "ice_common.h" +#include "ice_sched.h" +#include "ice_dcb.h" + +/** + * ice_aq_get_lldp_mib + * @hw: pointer to the HW struct + * @bridge_type: type of bridge requested + * @mib_type: Local, Remote or both Local and Remote MIBs + * @buf: pointer to the caller-supplied buffer to store the MIB block + * @buf_size: size of the buffer (in bytes) + * @local_len: length of the returned Local LLDP MIB + * @remote_len: length of the returned Remote LLDP MIB + * @cd: pointer to command details structure or NULL + * + * Requests the complete LLDP MIB (entire packet). (0x0A00) + */ +static enum ice_status +ice_aq_get_lldp_mib(struct ice_hw *hw, u8 bridge_type, u8 mib_type, void *buf, + u16 buf_size, u16 *local_len, u16 *remote_len, + struct ice_sq_cd *cd) +{ + struct ice_aqc_lldp_get_mib *cmd; + struct ice_aq_desc desc; + enum ice_status status; + + cmd = &desc.params.lldp_get_mib; + + if (buf_size == 0 || !buf) + return ICE_ERR_PARAM; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_lldp_get_mib); + + cmd->type = mib_type & ICE_AQ_LLDP_MIB_TYPE_M; + cmd->type |= (bridge_type << ICE_AQ_LLDP_BRID_TYPE_S) & + ICE_AQ_LLDP_BRID_TYPE_M; + + desc.datalen = cpu_to_le16(buf_size); + + status = ice_aq_send_cmd(hw, &desc, buf, buf_size, cd); + if (!status) { + if (local_len) + *local_len = le16_to_cpu(cmd->local_len); + if (remote_len) + *remote_len = le16_to_cpu(cmd->remote_len); + } + + return status; +} + +/** + * ice_aq_cfg_lldp_mib_change + * @hw: pointer to the HW struct + * @ena_update: Enable or Disable event posting + * @cd: pointer to command details structure or NULL + * + * Enable or Disable posting of an event on ARQ when LLDP MIB + * associated with the interface changes (0x0A01) + */ +enum ice_status +ice_aq_cfg_lldp_mib_change(struct ice_hw *hw, bool ena_update, + struct ice_sq_cd *cd) +{ + struct ice_aqc_lldp_set_mib_change *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.lldp_set_event; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_lldp_set_mib_change); + + if (!ena_update) + cmd->command |= ICE_AQ_LLDP_MIB_UPDATE_DIS; + + return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); +} + +/** + * ice_aq_stop_lldp + * @hw: pointer to the HW struct + * @shutdown_lldp_agent: True if LLDP Agent needs to be Shutdown + * False if LLDP Agent needs to be Stopped + * @persist: True if Stop/Shutdown of LLDP Agent needs to be persistent across + * reboots + * @cd: pointer to command details structure or NULL + * + * Stop or Shutdown the embedded LLDP Agent (0x0A05) + */ +enum ice_status +ice_aq_stop_lldp(struct ice_hw *hw, bool shutdown_lldp_agent, bool persist, + struct ice_sq_cd *cd) +{ + struct ice_aqc_lldp_stop *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.lldp_stop; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_lldp_stop); + + if (shutdown_lldp_agent) + cmd->command |= ICE_AQ_LLDP_AGENT_SHUTDOWN; + + if (persist) + cmd->command |= ICE_AQ_LLDP_AGENT_PERSIST_DIS; + + return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); +} + +/** + * ice_aq_start_lldp + * @hw: pointer to the HW struct + * @persist: True if Start of LLDP Agent needs to be persistent across reboots + * @cd: pointer to command details structure or NULL + * + * Start the embedded LLDP Agent on all ports. (0x0A06) + */ +enum ice_status +ice_aq_start_lldp(struct ice_hw *hw, bool persist, struct ice_sq_cd *cd) +{ + struct ice_aqc_lldp_start *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.lldp_start; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_lldp_start); + + cmd->command = ICE_AQ_LLDP_AGENT_START; + + if (persist) + cmd->command |= ICE_AQ_LLDP_AGENT_PERSIST_ENA; + + return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); +} + +/** + * ice_aq_set_lldp_mib - Set the LLDP MIB + * @hw: pointer to the HW struct + * @mib_type: Local, Remote or both Local and Remote MIBs + * @buf: pointer to the caller-supplied buffer to store the MIB block + * @buf_size: size of the buffer (in bytes) + * @cd: pointer to command details structure or NULL + * + * Set the LLDP MIB. (0x0A08) + */ +static enum ice_status +ice_aq_set_lldp_mib(struct ice_hw *hw, u8 mib_type, void *buf, u16 buf_size, + struct ice_sq_cd *cd) +{ + struct ice_aqc_lldp_set_local_mib *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.lldp_set_mib; + + if (buf_size == 0 || !buf) + return ICE_ERR_PARAM; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_lldp_set_local_mib); + + desc.flags |= cpu_to_le16((u16)ICE_AQ_FLAG_RD); + desc.datalen = cpu_to_le16(buf_size); + + cmd->type = mib_type; + cmd->length = cpu_to_le16(buf_size); + + return ice_aq_send_cmd(hw, &desc, buf, buf_size, cd); +} + +/** + * ice_get_dcbx_status + * @hw: pointer to the HW struct + * + * Get the DCBX status from the Firmware + */ +static u8 ice_get_dcbx_status(struct ice_hw *hw) +{ + u32 reg; + + reg = rd32(hw, PRTDCB_GENS); + return (u8)((reg & PRTDCB_GENS_DCBX_STATUS_M) >> + PRTDCB_GENS_DCBX_STATUS_S); +} + +/** + * ice_parse_ieee_ets_common_tlv + * @buf: Data buffer to be parsed for ETS CFG/REC data + * @ets_cfg: Container to store parsed data + * + * Parses the common data of IEEE 802.1Qaz ETS CFG/REC TLV + */ +static void +ice_parse_ieee_ets_common_tlv(u8 *buf, struct ice_dcb_ets_cfg *ets_cfg) +{ + u8 offset = 0; + int i; + + /* Priority Assignment Table (4 octets) + * Octets:| 1 | 2 | 3 | 4 | + * ----------------------------------------- + * |pri0|pri1|pri2|pri3|pri4|pri5|pri6|pri7| + * ----------------------------------------- + * Bits:|7 4|3 0|7 4|3 0|7 4|3 0|7 4|3 0| + * ----------------------------------------- + */ + for (i = 0; i < 4; i++) { + ets_cfg->prio_table[i * 2] = + ((buf[offset] & ICE_IEEE_ETS_PRIO_1_M) >> + ICE_IEEE_ETS_PRIO_1_S); + ets_cfg->prio_table[i * 2 + 1] = + ((buf[offset] & ICE_IEEE_ETS_PRIO_0_M) >> + ICE_IEEE_ETS_PRIO_0_S); + offset++; + } + + /* TC Bandwidth Table (8 octets) + * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | + * --------------------------------- + * |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7| + * --------------------------------- + * + * TSA Assignment Table (8 octets) + * Octets:| 9 | 10| 11| 12| 13| 14| 15| 16| + * --------------------------------- + * |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7| + * --------------------------------- + */ + ice_for_each_traffic_class(i) { + ets_cfg->tcbwtable[i] = buf[offset]; + ets_cfg->tsatable[i] = buf[ICE_MAX_TRAFFIC_CLASS + offset++]; + } +} + +/** + * ice_parse_ieee_etscfg_tlv + * @tlv: IEEE 802.1Qaz ETS CFG TLV + * @dcbcfg: Local store to update ETS CFG data + * + * Parses IEEE 802.1Qaz ETS CFG TLV + */ +static void +ice_parse_ieee_etscfg_tlv(struct ice_lldp_org_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_dcb_ets_cfg *etscfg; + u8 *buf = tlv->tlvinfo; + + /* First Octet post subtype + * -------------------------- + * |will-|CBS | Re- | Max | + * |ing | |served| TCs | + * -------------------------- + * |1bit | 1bit|3 bits|3bits| + */ + etscfg = &dcbcfg->etscfg; + etscfg->willing = ((buf[0] & ICE_IEEE_ETS_WILLING_M) >> + ICE_IEEE_ETS_WILLING_S); + etscfg->cbs = ((buf[0] & ICE_IEEE_ETS_CBS_M) >> ICE_IEEE_ETS_CBS_S); + etscfg->maxtcs = ((buf[0] & ICE_IEEE_ETS_MAXTC_M) >> + ICE_IEEE_ETS_MAXTC_S); + + /* Begin parsing at Priority Assignment Table (offset 1 in buf) */ + ice_parse_ieee_ets_common_tlv(&buf[1], etscfg); +} + +/** + * ice_parse_ieee_etsrec_tlv + * @tlv: IEEE 802.1Qaz ETS REC TLV + * @dcbcfg: Local store to update ETS REC data + * + * Parses IEEE 802.1Qaz ETS REC TLV + */ +static void +ice_parse_ieee_etsrec_tlv(struct ice_lldp_org_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + u8 *buf = tlv->tlvinfo; + + /* Begin parsing at Priority Assignment Table (offset 1 in buf) */ + ice_parse_ieee_ets_common_tlv(&buf[1], &dcbcfg->etsrec); +} + +/** + * ice_parse_ieee_pfccfg_tlv + * @tlv: IEEE 802.1Qaz PFC CFG TLV + * @dcbcfg: Local store to update PFC CFG data + * + * Parses IEEE 802.1Qaz PFC CFG TLV + */ +static void +ice_parse_ieee_pfccfg_tlv(struct ice_lldp_org_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + u8 *buf = tlv->tlvinfo; + + /* ---------------------------------------- + * |will-|MBC | Re- | PFC | PFC Enable | + * |ing | |served| cap | | + * ----------------------------------------- + * |1bit | 1bit|2 bits|4bits| 1 octet | + */ + dcbcfg->pfc.willing = ((buf[0] & ICE_IEEE_PFC_WILLING_M) >> + ICE_IEEE_PFC_WILLING_S); + dcbcfg->pfc.mbc = ((buf[0] & ICE_IEEE_PFC_MBC_M) >> ICE_IEEE_PFC_MBC_S); + dcbcfg->pfc.pfccap = ((buf[0] & ICE_IEEE_PFC_CAP_M) >> + ICE_IEEE_PFC_CAP_S); + dcbcfg->pfc.pfcena = buf[1]; +} + +/** + * ice_parse_ieee_app_tlv + * @tlv: IEEE 802.1Qaz APP TLV + * @dcbcfg: Local store to update APP PRIO data + * + * Parses IEEE 802.1Qaz APP PRIO TLV + */ +static void +ice_parse_ieee_app_tlv(struct ice_lldp_org_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + u16 offset = 0; + u16 typelen; + int i = 0; + u16 len; + u8 *buf; + + typelen = ntohs(tlv->typelen); + len = ((typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S); + buf = tlv->tlvinfo; + + /* Removing sizeof(ouisubtype) and reserved byte from len. + * Remaining len div 3 is number of APP TLVs. + */ + len -= (sizeof(tlv->ouisubtype) + 1); + + /* Move offset to App Priority Table */ + offset++; + + /* Application Priority Table (3 octets) + * Octets:| 1 | 2 | 3 | + * ----------------------------------------- + * |Priority|Rsrvd| Sel | Protocol ID | + * ----------------------------------------- + * Bits:|23 21|20 19|18 16|15 0| + * ----------------------------------------- + */ + while (offset < len) { + dcbcfg->app[i].priority = ((buf[offset] & + ICE_IEEE_APP_PRIO_M) >> + ICE_IEEE_APP_PRIO_S); + dcbcfg->app[i].selector = ((buf[offset] & + ICE_IEEE_APP_SEL_M) >> + ICE_IEEE_APP_SEL_S); + dcbcfg->app[i].prot_id = (buf[offset + 1] << 0x8) | + buf[offset + 2]; + /* Move to next app */ + offset += 3; + i++; + if (i >= ICE_DCBX_MAX_APPS) + break; + } + + dcbcfg->numapps = i; +} + +/** + * ice_parse_ieee_tlv + * @tlv: IEEE 802.1Qaz TLV + * @dcbcfg: Local store to update ETS REC data + * + * Get the TLV subtype and send it to parsing function + * based on the subtype value + */ +static void +ice_parse_ieee_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) +{ + u32 ouisubtype; + u8 subtype; + + ouisubtype = ntohl(tlv->ouisubtype); + subtype = (u8)((ouisubtype & ICE_LLDP_TLV_SUBTYPE_M) >> + ICE_LLDP_TLV_SUBTYPE_S); + switch (subtype) { + case ICE_IEEE_SUBTYPE_ETS_CFG: + ice_parse_ieee_etscfg_tlv(tlv, dcbcfg); + break; + case ICE_IEEE_SUBTYPE_ETS_REC: + ice_parse_ieee_etsrec_tlv(tlv, dcbcfg); + break; + case ICE_IEEE_SUBTYPE_PFC_CFG: + ice_parse_ieee_pfccfg_tlv(tlv, dcbcfg); + break; + case ICE_IEEE_SUBTYPE_APP_PRI: + ice_parse_ieee_app_tlv(tlv, dcbcfg); + break; + default: + break; + } +} + +/** + * ice_parse_cee_pgcfg_tlv + * @tlv: CEE DCBX PG CFG TLV + * @dcbcfg: Local store to update ETS CFG data + * + * Parses CEE DCBX PG CFG TLV + */ +static void +ice_parse_cee_pgcfg_tlv(struct ice_cee_feat_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_dcb_ets_cfg *etscfg; + u8 *buf = tlv->tlvinfo; + u16 offset = 0; + int i; + + etscfg = &dcbcfg->etscfg; + + if (tlv->en_will_err & ICE_CEE_FEAT_TLV_WILLING_M) + etscfg->willing = 1; + + etscfg->cbs = 0; + /* Priority Group Table (4 octets) + * Octets:| 1 | 2 | 3 | 4 | + * ----------------------------------------- + * |pri0|pri1|pri2|pri3|pri4|pri5|pri6|pri7| + * ----------------------------------------- + * Bits:|7 4|3 0|7 4|3 0|7 4|3 0|7 4|3 0| + * ----------------------------------------- + */ + for (i = 0; i < 4; i++) { + etscfg->prio_table[i * 2] = + ((buf[offset] & ICE_CEE_PGID_PRIO_1_M) >> + ICE_CEE_PGID_PRIO_1_S); + etscfg->prio_table[i * 2 + 1] = + ((buf[offset] & ICE_CEE_PGID_PRIO_0_M) >> + ICE_CEE_PGID_PRIO_0_S); + offset++; + } + + /* PG Percentage Table (8 octets) + * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | + * --------------------------------- + * |pg0|pg1|pg2|pg3|pg4|pg5|pg6|pg7| + * --------------------------------- + */ + ice_for_each_traffic_class(i) + etscfg->tcbwtable[i] = buf[offset++]; + + /* Number of TCs supported (1 octet) */ + etscfg->maxtcs = buf[offset]; +} + +/** + * ice_parse_cee_pfccfg_tlv + * @tlv: CEE DCBX PFC CFG TLV + * @dcbcfg: Local store to update PFC CFG data + * + * Parses CEE DCBX PFC CFG TLV + */ +static void +ice_parse_cee_pfccfg_tlv(struct ice_cee_feat_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + u8 *buf = tlv->tlvinfo; + + if (tlv->en_will_err & ICE_CEE_FEAT_TLV_WILLING_M) + dcbcfg->pfc.willing = 1; + + /* ------------------------ + * | PFC Enable | PFC TCs | + * ------------------------ + * | 1 octet | 1 octet | + */ + dcbcfg->pfc.pfcena = buf[0]; + dcbcfg->pfc.pfccap = buf[1]; +} + +/** + * ice_parse_cee_app_tlv + * @tlv: CEE DCBX APP TLV + * @dcbcfg: Local store to update APP PRIO data + * + * Parses CEE DCBX APP PRIO TLV + */ +static void +ice_parse_cee_app_tlv(struct ice_cee_feat_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) +{ + u16 len, typelen, offset = 0; + struct ice_cee_app_prio *app; + u8 i; + + typelen = ntohs(tlv->hdr.typelen); + len = ((typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S); + + dcbcfg->numapps = len / sizeof(*app); + if (!dcbcfg->numapps) + return; + if (dcbcfg->numapps > ICE_DCBX_MAX_APPS) + dcbcfg->numapps = ICE_DCBX_MAX_APPS; + + for (i = 0; i < dcbcfg->numapps; i++) { + u8 up, selector; + + app = (struct ice_cee_app_prio *)(tlv->tlvinfo + offset); + for (up = 0; up < ICE_MAX_USER_PRIORITY; up++) + if (app->prio_map & BIT(up)) + break; + + dcbcfg->app[i].priority = up; + + /* Get Selector from lower 2 bits, and convert to IEEE */ + selector = (app->upper_oui_sel & ICE_CEE_APP_SELECTOR_M); + switch (selector) { + case ICE_CEE_APP_SEL_ETHTYPE: + dcbcfg->app[i].selector = ICE_APP_SEL_ETHTYPE; + break; + case ICE_CEE_APP_SEL_TCPIP: + dcbcfg->app[i].selector = ICE_APP_SEL_TCPIP; + break; + default: + /* Keep selector as it is for unknown types */ + dcbcfg->app[i].selector = selector; + } + + dcbcfg->app[i].prot_id = ntohs(app->protocol); + /* Move to next app */ + offset += sizeof(*app); + } +} + +/** + * ice_parse_cee_tlv + * @tlv: CEE DCBX TLV + * @dcbcfg: Local store to update DCBX config data + * + * Get the TLV subtype and send it to parsing function + * based on the subtype value + */ +static void +ice_parse_cee_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_cee_feat_tlv *sub_tlv; + u8 subtype, feat_tlv_count = 0; + u16 len, tlvlen, typelen; + u32 ouisubtype; + + ouisubtype = ntohl(tlv->ouisubtype); + subtype = (u8)((ouisubtype & ICE_LLDP_TLV_SUBTYPE_M) >> + ICE_LLDP_TLV_SUBTYPE_S); + /* Return if not CEE DCBX */ + if (subtype != ICE_CEE_DCBX_TYPE) + return; + + typelen = ntohs(tlv->typelen); + tlvlen = ((typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S); + len = sizeof(tlv->typelen) + sizeof(ouisubtype) + + sizeof(struct ice_cee_ctrl_tlv); + /* Return if no CEE DCBX Feature TLVs */ + if (tlvlen <= len) + return; + + sub_tlv = (struct ice_cee_feat_tlv *)((char *)tlv + len); + while (feat_tlv_count < ICE_CEE_MAX_FEAT_TYPE) { + u16 sublen; + + typelen = ntohs(sub_tlv->hdr.typelen); + sublen = ((typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S); + subtype = (u8)((typelen & ICE_LLDP_TLV_TYPE_M) >> + ICE_LLDP_TLV_TYPE_S); + switch (subtype) { + case ICE_CEE_SUBTYPE_PG_CFG: + ice_parse_cee_pgcfg_tlv(sub_tlv, dcbcfg); + break; + case ICE_CEE_SUBTYPE_PFC_CFG: + ice_parse_cee_pfccfg_tlv(sub_tlv, dcbcfg); + break; + case ICE_CEE_SUBTYPE_APP_PRI: + ice_parse_cee_app_tlv(sub_tlv, dcbcfg); + break; + default: + return; /* Invalid Sub-type return */ + } + feat_tlv_count++; + /* Move to next sub TLV */ + sub_tlv = (struct ice_cee_feat_tlv *) + ((char *)sub_tlv + sizeof(sub_tlv->hdr.typelen) + + sublen); + } +} + +/** + * ice_parse_org_tlv + * @tlv: Organization specific TLV + * @dcbcfg: Local store to update ETS REC data + * + * Currently only IEEE 802.1Qaz TLV is supported, all others + * will be returned + */ +static void +ice_parse_org_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) +{ + u32 ouisubtype; + u32 oui; + + ouisubtype = ntohl(tlv->ouisubtype); + oui = ((ouisubtype & ICE_LLDP_TLV_OUI_M) >> ICE_LLDP_TLV_OUI_S); + switch (oui) { + case ICE_IEEE_8021QAZ_OUI: + ice_parse_ieee_tlv(tlv, dcbcfg); + break; + case ICE_CEE_DCBX_OUI: + ice_parse_cee_tlv(tlv, dcbcfg); + break; + default: + break; + } +} + +/** + * ice_lldp_to_dcb_cfg + * @lldpmib: LLDPDU to be parsed + * @dcbcfg: store for LLDPDU data + * + * Parse DCB configuration from the LLDPDU + */ +static enum ice_status +ice_lldp_to_dcb_cfg(u8 *lldpmib, struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_lldp_org_tlv *tlv; + enum ice_status ret = 0; + u16 offset = 0; + u16 typelen; + u16 type; + u16 len; + + if (!lldpmib || !dcbcfg) + return ICE_ERR_PARAM; + + /* set to the start of LLDPDU */ + lldpmib += ETH_HLEN; + tlv = (struct ice_lldp_org_tlv *)lldpmib; + while (1) { + typelen = ntohs(tlv->typelen); + type = ((typelen & ICE_LLDP_TLV_TYPE_M) >> ICE_LLDP_TLV_TYPE_S); + len = ((typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S); + offset += sizeof(typelen) + len; + + /* END TLV or beyond LLDPDU size */ + if (type == ICE_TLV_TYPE_END || offset > ICE_LLDPDU_SIZE) + break; + + switch (type) { + case ICE_TLV_TYPE_ORG: + ice_parse_org_tlv(tlv, dcbcfg); + break; + default: + break; + } + + /* Move to next TLV */ + tlv = (struct ice_lldp_org_tlv *) + ((char *)tlv + sizeof(tlv->typelen) + len); + } + + return ret; +} + +/** + * ice_aq_get_dcb_cfg + * @hw: pointer to the HW struct + * @mib_type: MIB type for the query + * @bridgetype: bridge type for the query (remote) + * @dcbcfg: store for LLDPDU data + * + * Query DCB configuration from the firmware + */ +enum ice_status +ice_aq_get_dcb_cfg(struct ice_hw *hw, u8 mib_type, u8 bridgetype, + struct ice_dcbx_cfg *dcbcfg) +{ + enum ice_status ret; + u8 *lldpmib; + + /* Allocate the LLDPDU */ + lldpmib = devm_kzalloc(ice_hw_to_dev(hw), ICE_LLDPDU_SIZE, GFP_KERNEL); + if (!lldpmib) + return ICE_ERR_NO_MEMORY; + + ret = ice_aq_get_lldp_mib(hw, bridgetype, mib_type, (void *)lldpmib, + ICE_LLDPDU_SIZE, NULL, NULL, NULL); + + if (!ret) + /* Parse LLDP MIB to get DCB configuration */ + ret = ice_lldp_to_dcb_cfg(lldpmib, dcbcfg); + + devm_kfree(ice_hw_to_dev(hw), lldpmib); + + return ret; +} + +/** + * ice_aq_start_stop_dcbx - Start/Stop DCBX service in FW + * @hw: pointer to the HW struct + * @start_dcbx_agent: True if DCBX Agent needs to be started + * False if DCBX Agent needs to be stopped + * @dcbx_agent_status: FW indicates back the DCBX agent status + * True if DCBX Agent is active + * False if DCBX Agent is stopped + * @cd: pointer to command details structure or NULL + * + * Start/Stop the embedded dcbx Agent. In case that this wrapper function + * returns ICE_SUCCESS, caller will need to check if FW returns back the same + * value as stated in dcbx_agent_status, and react accordingly. (0x0A09) + */ +enum ice_status +ice_aq_start_stop_dcbx(struct ice_hw *hw, bool start_dcbx_agent, + bool *dcbx_agent_status, struct ice_sq_cd *cd) +{ + struct ice_aqc_lldp_stop_start_specific_agent *cmd; + enum ice_status status; + struct ice_aq_desc desc; + u16 opcode; + + cmd = &desc.params.lldp_agent_ctrl; + + opcode = ice_aqc_opc_lldp_stop_start_specific_agent; + + ice_fill_dflt_direct_cmd_desc(&desc, opcode); + + if (start_dcbx_agent) + cmd->command = ICE_AQC_START_STOP_AGENT_START_DCBX; + + status = ice_aq_send_cmd(hw, &desc, NULL, 0, cd); + + *dcbx_agent_status = false; + + if (!status && + cmd->command == ICE_AQC_START_STOP_AGENT_START_DCBX) + *dcbx_agent_status = true; + + return status; +} + +/** + * ice_aq_get_cee_dcb_cfg + * @hw: pointer to the HW struct + * @buff: response buffer that stores CEE operational configuration + * @cd: pointer to command details structure or NULL + * + * Get CEE DCBX mode operational configuration from firmware (0x0A07) + */ +static enum ice_status +ice_aq_get_cee_dcb_cfg(struct ice_hw *hw, + struct ice_aqc_get_cee_dcb_cfg_resp *buff, + struct ice_sq_cd *cd) +{ + struct ice_aq_desc desc; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_cee_dcb_cfg); + + return ice_aq_send_cmd(hw, &desc, (void *)buff, sizeof(*buff), cd); +} + +/** + * ice_cee_to_dcb_cfg + * @cee_cfg: pointer to CEE configuration struct + * @dcbcfg: DCB configuration struct + * + * Convert CEE configuration from firmware to DCB configuration + */ +static void +ice_cee_to_dcb_cfg(struct ice_aqc_get_cee_dcb_cfg_resp *cee_cfg, + struct ice_dcbx_cfg *dcbcfg) +{ + u32 status, tlv_status = le32_to_cpu(cee_cfg->tlv_status); + u32 ice_aqc_cee_status_mask, ice_aqc_cee_status_shift; + u16 app_prio = le16_to_cpu(cee_cfg->oper_app_prio); + u8 i, err, sync, oper, app_index, ice_app_sel_type; + u16 ice_aqc_cee_app_mask, ice_aqc_cee_app_shift; + u16 ice_app_prot_id_type; + + /* CEE PG data to ETS config */ + dcbcfg->etscfg.maxtcs = cee_cfg->oper_num_tc; + + /* Note that the FW creates the oper_prio_tc nibbles reversed + * from those in the CEE Priority Group sub-TLV. + */ + for (i = 0; i < ICE_MAX_TRAFFIC_CLASS / 2; i++) { + dcbcfg->etscfg.prio_table[i * 2] = + ((cee_cfg->oper_prio_tc[i] & ICE_CEE_PGID_PRIO_0_M) >> + ICE_CEE_PGID_PRIO_0_S); + dcbcfg->etscfg.prio_table[i * 2 + 1] = + ((cee_cfg->oper_prio_tc[i] & ICE_CEE_PGID_PRIO_1_M) >> + ICE_CEE_PGID_PRIO_1_S); + } + + ice_for_each_traffic_class(i) { + dcbcfg->etscfg.tcbwtable[i] = cee_cfg->oper_tc_bw[i]; + + if (dcbcfg->etscfg.prio_table[i] == ICE_CEE_PGID_STRICT) { + /* Map it to next empty TC */ + dcbcfg->etscfg.prio_table[i] = cee_cfg->oper_num_tc - 1; + dcbcfg->etscfg.tsatable[i] = ICE_IEEE_TSA_STRICT; + } else { + dcbcfg->etscfg.tsatable[i] = ICE_IEEE_TSA_ETS; + } + } + + /* CEE PFC data to ETS config */ + dcbcfg->pfc.pfcena = cee_cfg->oper_pfc_en; + dcbcfg->pfc.pfccap = ICE_MAX_TRAFFIC_CLASS; + + app_index = 0; + for (i = 0; i < 3; i++) { + if (i == 0) { + /* FCoE APP */ + ice_aqc_cee_status_mask = ICE_AQC_CEE_FCOE_STATUS_M; + ice_aqc_cee_status_shift = ICE_AQC_CEE_FCOE_STATUS_S; + ice_aqc_cee_app_mask = ICE_AQC_CEE_APP_FCOE_M; + ice_aqc_cee_app_shift = ICE_AQC_CEE_APP_FCOE_S; + ice_app_sel_type = ICE_APP_SEL_ETHTYPE; + ice_app_prot_id_type = ICE_APP_PROT_ID_FCOE; + } else if (i == 1) { + /* iSCSI APP */ + ice_aqc_cee_status_mask = ICE_AQC_CEE_ISCSI_STATUS_M; + ice_aqc_cee_status_shift = ICE_AQC_CEE_ISCSI_STATUS_S; + ice_aqc_cee_app_mask = ICE_AQC_CEE_APP_ISCSI_M; + ice_aqc_cee_app_shift = ICE_AQC_CEE_APP_ISCSI_S; + ice_app_sel_type = ICE_APP_SEL_TCPIP; + ice_app_prot_id_type = ICE_APP_PROT_ID_ISCSI; + } else { + /* FIP APP */ + ice_aqc_cee_status_mask = ICE_AQC_CEE_FIP_STATUS_M; + ice_aqc_cee_status_shift = ICE_AQC_CEE_FIP_STATUS_S; + ice_aqc_cee_app_mask = ICE_AQC_CEE_APP_FIP_M; + ice_aqc_cee_app_shift = ICE_AQC_CEE_APP_FIP_S; + ice_app_sel_type = ICE_APP_SEL_ETHTYPE; + ice_app_prot_id_type = ICE_APP_PROT_ID_FIP; + } + + status = (tlv_status & ice_aqc_cee_status_mask) >> + ice_aqc_cee_status_shift; + err = (status & ICE_TLV_STATUS_ERR) ? 1 : 0; + sync = (status & ICE_TLV_STATUS_SYNC) ? 1 : 0; + oper = (status & ICE_TLV_STATUS_OPER) ? 1 : 0; + /* Add FCoE/iSCSI/FIP APP if Error is False and + * Oper/Sync is True + */ + if (!err && sync && oper) { + dcbcfg->app[app_index].priority = + (app_prio & ice_aqc_cee_app_mask) >> + ice_aqc_cee_app_shift; + dcbcfg->app[app_index].selector = ice_app_sel_type; + dcbcfg->app[app_index].prot_id = ice_app_prot_id_type; + app_index++; + } + } + + dcbcfg->numapps = app_index; +} + +/** + * ice_get_ieee_dcb_cfg + * @pi: port information structure + * @dcbx_mode: mode of DCBX (IEEE or CEE) + * + * Get IEEE or CEE mode DCB configuration from the Firmware + */ +static enum ice_status +ice_get_ieee_or_cee_dcb_cfg(struct ice_port_info *pi, u8 dcbx_mode) +{ + struct ice_dcbx_cfg *dcbx_cfg = NULL; + enum ice_status ret; + + if (!pi) + return ICE_ERR_PARAM; + + if (dcbx_mode == ICE_DCBX_MODE_IEEE) + dcbx_cfg = &pi->local_dcbx_cfg; + else if (dcbx_mode == ICE_DCBX_MODE_CEE) + dcbx_cfg = &pi->desired_dcbx_cfg; + + /* Get Local DCB Config in case of ICE_DCBX_MODE_IEEE + * or get CEE DCB Desired Config in case of ICE_DCBX_MODE_CEE + */ + ret = ice_aq_get_dcb_cfg(pi->hw, ICE_AQ_LLDP_MIB_LOCAL, + ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID, dcbx_cfg); + if (ret) + goto out; + + /* Get Remote DCB Config */ + dcbx_cfg = &pi->remote_dcbx_cfg; + ret = ice_aq_get_dcb_cfg(pi->hw, ICE_AQ_LLDP_MIB_REMOTE, + ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID, dcbx_cfg); + /* Don't treat ENOENT as an error for Remote MIBs */ + if (pi->hw->adminq.sq_last_status == ICE_AQ_RC_ENOENT) + ret = 0; + +out: + return ret; +} + +/** + * ice_get_dcb_cfg + * @pi: port information structure + * + * Get DCB configuration from the Firmware + */ +enum ice_status ice_get_dcb_cfg(struct ice_port_info *pi) +{ + struct ice_aqc_get_cee_dcb_cfg_resp cee_cfg; + struct ice_dcbx_cfg *dcbx_cfg; + enum ice_status ret; + + if (!pi) + return ICE_ERR_PARAM; + + ret = ice_aq_get_cee_dcb_cfg(pi->hw, &cee_cfg, NULL); + if (!ret) { + /* CEE mode */ + dcbx_cfg = &pi->local_dcbx_cfg; + dcbx_cfg->dcbx_mode = ICE_DCBX_MODE_CEE; + dcbx_cfg->tlv_status = le32_to_cpu(cee_cfg.tlv_status); + ice_cee_to_dcb_cfg(&cee_cfg, dcbx_cfg); + ret = ice_get_ieee_or_cee_dcb_cfg(pi, ICE_DCBX_MODE_CEE); + } else if (pi->hw->adminq.sq_last_status == ICE_AQ_RC_ENOENT) { + /* CEE mode not enabled try querying IEEE data */ + dcbx_cfg = &pi->local_dcbx_cfg; + dcbx_cfg->dcbx_mode = ICE_DCBX_MODE_IEEE; + ret = ice_get_ieee_or_cee_dcb_cfg(pi, ICE_DCBX_MODE_IEEE); + } + + return ret; +} + +/** + * ice_init_dcb + * @hw: pointer to the HW struct + * + * Update DCB configuration from the Firmware + */ +enum ice_status ice_init_dcb(struct ice_hw *hw) +{ + struct ice_port_info *pi = hw->port_info; + enum ice_status ret = 0; + + if (!hw->func_caps.common_cap.dcb) + return ICE_ERR_NOT_SUPPORTED; + + pi->is_sw_lldp = true; + + /* Get DCBX status */ + pi->dcbx_status = ice_get_dcbx_status(hw); + + if (pi->dcbx_status == ICE_DCBX_STATUS_DONE || + pi->dcbx_status == ICE_DCBX_STATUS_IN_PROGRESS) { + /* Get current DCBX configuration */ + ret = ice_get_dcb_cfg(pi); + pi->is_sw_lldp = (hw->adminq.sq_last_status == ICE_AQ_RC_EPERM); + if (ret) + return ret; + } else if (pi->dcbx_status == ICE_DCBX_STATUS_DIS) { + return ICE_ERR_NOT_READY; + } + + /* Configure the LLDP MIB change event */ + ret = ice_aq_cfg_lldp_mib_change(hw, true, NULL); + if (!ret) + pi->is_sw_lldp = false; + + return ret; +} + +/** + * ice_add_ieee_ets_common_tlv + * @buf: Data buffer to be populated with ice_dcb_ets_cfg data + * @ets_cfg: Container for ice_dcb_ets_cfg data + * + * Populate the TLV buffer with ice_dcb_ets_cfg data + */ +static void +ice_add_ieee_ets_common_tlv(u8 *buf, struct ice_dcb_ets_cfg *ets_cfg) +{ + u8 priority0, priority1; + u8 offset = 0; + int i; + + /* Priority Assignment Table (4 octets) + * Octets:| 1 | 2 | 3 | 4 | + * ----------------------------------------- + * |pri0|pri1|pri2|pri3|pri4|pri5|pri6|pri7| + * ----------------------------------------- + * Bits:|7 4|3 0|7 4|3 0|7 4|3 0|7 4|3 0| + * ----------------------------------------- + */ + for (i = 0; i < ICE_MAX_TRAFFIC_CLASS / 2; i++) { + priority0 = ets_cfg->prio_table[i * 2] & 0xF; + priority1 = ets_cfg->prio_table[i * 2 + 1] & 0xF; + buf[offset] = (priority0 << ICE_IEEE_ETS_PRIO_1_S) | priority1; + offset++; + } + + /* TC Bandwidth Table (8 octets) + * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | + * --------------------------------- + * |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7| + * --------------------------------- + * + * TSA Assignment Table (8 octets) + * Octets:| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | + * --------------------------------- + * |tc0|tc1|tc2|tc3|tc4|tc5|tc6|tc7| + * --------------------------------- + */ + ice_for_each_traffic_class(i) { + buf[offset] = ets_cfg->tcbwtable[i]; + buf[ICE_MAX_TRAFFIC_CLASS + offset] = ets_cfg->tsatable[i]; + offset++; + } +} + +/** + * ice_add_ieee_ets_tlv - Prepare ETS TLV in IEEE format + * @tlv: Fill the ETS config data in IEEE format + * @dcbcfg: Local store which holds the DCB Config + * + * Prepare IEEE 802.1Qaz ETS CFG TLV + */ +static void +ice_add_ieee_ets_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_dcb_ets_cfg *etscfg; + u8 *buf = tlv->tlvinfo; + u8 maxtcwilling = 0; + u32 ouisubtype; + u16 typelen; + + typelen = ((ICE_TLV_TYPE_ORG << ICE_LLDP_TLV_TYPE_S) | + ICE_IEEE_ETS_TLV_LEN); + tlv->typelen = htons(typelen); + + ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) | + ICE_IEEE_SUBTYPE_ETS_CFG); + tlv->ouisubtype = htonl(ouisubtype); + + /* First Octet post subtype + * -------------------------- + * |will-|CBS | Re- | Max | + * |ing | |served| TCs | + * -------------------------- + * |1bit | 1bit|3 bits|3bits| + */ + etscfg = &dcbcfg->etscfg; + if (etscfg->willing) + maxtcwilling = BIT(ICE_IEEE_ETS_WILLING_S); + maxtcwilling |= etscfg->maxtcs & ICE_IEEE_ETS_MAXTC_M; + buf[0] = maxtcwilling; + + /* Begin adding at Priority Assignment Table (offset 1 in buf) */ + ice_add_ieee_ets_common_tlv(&buf[1], etscfg); +} + +/** + * ice_add_ieee_etsrec_tlv - Prepare ETS Recommended TLV in IEEE format + * @tlv: Fill ETS Recommended TLV in IEEE format + * @dcbcfg: Local store which holds the DCB Config + * + * Prepare IEEE 802.1Qaz ETS REC TLV + */ +static void +ice_add_ieee_etsrec_tlv(struct ice_lldp_org_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + struct ice_dcb_ets_cfg *etsrec; + u8 *buf = tlv->tlvinfo; + u32 ouisubtype; + u16 typelen; + + typelen = ((ICE_TLV_TYPE_ORG << ICE_LLDP_TLV_TYPE_S) | + ICE_IEEE_ETS_TLV_LEN); + tlv->typelen = htons(typelen); + + ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) | + ICE_IEEE_SUBTYPE_ETS_REC); + tlv->ouisubtype = htonl(ouisubtype); + + etsrec = &dcbcfg->etsrec; + + /* First Octet is reserved */ + /* Begin adding at Priority Assignment Table (offset 1 in buf) */ + ice_add_ieee_ets_common_tlv(&buf[1], etsrec); +} + +/** + * ice_add_ieee_pfc_tlv - Prepare PFC TLV in IEEE format + * @tlv: Fill PFC TLV in IEEE format + * @dcbcfg: Local store which holds the PFC CFG data + * + * Prepare IEEE 802.1Qaz PFC CFG TLV + */ +static void +ice_add_ieee_pfc_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg) +{ + u8 *buf = tlv->tlvinfo; + u32 ouisubtype; + u16 typelen; + + typelen = ((ICE_TLV_TYPE_ORG << ICE_LLDP_TLV_TYPE_S) | + ICE_IEEE_PFC_TLV_LEN); + tlv->typelen = htons(typelen); + + ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) | + ICE_IEEE_SUBTYPE_PFC_CFG); + tlv->ouisubtype = htonl(ouisubtype); + + /* ---------------------------------------- + * |will-|MBC | Re- | PFC | PFC Enable | + * |ing | |served| cap | | + * ----------------------------------------- + * |1bit | 1bit|2 bits|4bits| 1 octet | + */ + if (dcbcfg->pfc.willing) + buf[0] = BIT(ICE_IEEE_PFC_WILLING_S); + + if (dcbcfg->pfc.mbc) + buf[0] |= BIT(ICE_IEEE_PFC_MBC_S); + + buf[0] |= dcbcfg->pfc.pfccap & 0xF; + buf[1] = dcbcfg->pfc.pfcena; +} + +/** + * ice_add_ieee_app_pri_tlv - Prepare APP TLV in IEEE format + * @tlv: Fill APP TLV in IEEE format + * @dcbcfg: Local store which holds the APP CFG data + * + * Prepare IEEE 802.1Qaz APP CFG TLV + */ +static void +ice_add_ieee_app_pri_tlv(struct ice_lldp_org_tlv *tlv, + struct ice_dcbx_cfg *dcbcfg) +{ + u16 typelen, len, offset = 0; + u8 priority, selector, i = 0; + u8 *buf = tlv->tlvinfo; + u32 ouisubtype; + + /* No APP TLVs then just return */ + if (dcbcfg->numapps == 0) + return; + ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) | + ICE_IEEE_SUBTYPE_APP_PRI); + tlv->ouisubtype = htonl(ouisubtype); + + /* Move offset to App Priority Table */ + offset++; + /* Application Priority Table (3 octets) + * Octets:| 1 | 2 | 3 | + * ----------------------------------------- + * |Priority|Rsrvd| Sel | Protocol ID | + * ----------------------------------------- + * Bits:|23 21|20 19|18 16|15 0| + * ----------------------------------------- + */ + while (i < dcbcfg->numapps) { + priority = dcbcfg->app[i].priority & 0x7; + selector = dcbcfg->app[i].selector & 0x7; + buf[offset] = (priority << ICE_IEEE_APP_PRIO_S) | selector; + buf[offset + 1] = (dcbcfg->app[i].prot_id >> 0x8) & 0xFF; + buf[offset + 2] = dcbcfg->app[i].prot_id & 0xFF; + /* Move to next app */ + offset += 3; + i++; + if (i >= ICE_DCBX_MAX_APPS) + break; + } + /* len includes size of ouisubtype + 1 reserved + 3*numapps */ + len = sizeof(tlv->ouisubtype) + 1 + (i * 3); + typelen = ((ICE_TLV_TYPE_ORG << ICE_LLDP_TLV_TYPE_S) | (len & 0x1FF)); + tlv->typelen = htons(typelen); +} + +/** + * ice_add_dcb_tlv - Add all IEEE TLVs + * @tlv: Fill TLV data in IEEE format + * @dcbcfg: Local store which holds the DCB Config + * @tlvid: Type of IEEE TLV + * + * Add tlv information + */ +static void +ice_add_dcb_tlv(struct ice_lldp_org_tlv *tlv, struct ice_dcbx_cfg *dcbcfg, + u16 tlvid) +{ + switch (tlvid) { + case ICE_IEEE_TLV_ID_ETS_CFG: + ice_add_ieee_ets_tlv(tlv, dcbcfg); + break; + case ICE_IEEE_TLV_ID_ETS_REC: + ice_add_ieee_etsrec_tlv(tlv, dcbcfg); + break; + case ICE_IEEE_TLV_ID_PFC_CFG: + ice_add_ieee_pfc_tlv(tlv, dcbcfg); + break; + case ICE_IEEE_TLV_ID_APP_PRI: + ice_add_ieee_app_pri_tlv(tlv, dcbcfg); + break; + default: + break; + } +} + +/** + * ice_dcb_cfg_to_lldp - Convert DCB configuration to MIB format + * @lldpmib: pointer to the HW struct + * @miblen: length of LLDP MIB + * @dcbcfg: Local store which holds the DCB Config + * + * Convert the DCB configuration to MIB format + */ +static void +ice_dcb_cfg_to_lldp(u8 *lldpmib, u16 *miblen, struct ice_dcbx_cfg *dcbcfg) +{ + u16 len, offset = 0, tlvid = ICE_TLV_ID_START; + struct ice_lldp_org_tlv *tlv; + u16 typelen; + + tlv = (struct ice_lldp_org_tlv *)lldpmib; + while (1) { + ice_add_dcb_tlv(tlv, dcbcfg, tlvid++); + typelen = ntohs(tlv->typelen); + len = (typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S; + if (len) + offset += len + 2; + /* END TLV or beyond LLDPDU size */ + if (tlvid >= ICE_TLV_ID_END_OF_LLDPPDU || + offset > ICE_LLDPDU_SIZE) + break; + /* Move to next TLV */ + if (len) + tlv = (struct ice_lldp_org_tlv *) + ((char *)tlv + sizeof(tlv->typelen) + len); + } + *miblen = offset; +} + +/** + * ice_set_dcb_cfg - Set the local LLDP MIB to FW + * @pi: port information structure + * + * Set DCB configuration to the Firmware + */ +enum ice_status ice_set_dcb_cfg(struct ice_port_info *pi) +{ + u8 mib_type, *lldpmib = NULL; + struct ice_dcbx_cfg *dcbcfg; + enum ice_status ret; + struct ice_hw *hw; + u16 miblen; + + if (!pi) + return ICE_ERR_PARAM; + + hw = pi->hw; + + /* update the HW local config */ + dcbcfg = &pi->local_dcbx_cfg; + /* Allocate the LLDPDU */ + lldpmib = devm_kzalloc(ice_hw_to_dev(hw), ICE_LLDPDU_SIZE, GFP_KERNEL); + if (!lldpmib) + return ICE_ERR_NO_MEMORY; + + mib_type = SET_LOCAL_MIB_TYPE_LOCAL_MIB; + if (dcbcfg->app_mode == ICE_DCBX_APPS_NON_WILLING) + mib_type |= SET_LOCAL_MIB_TYPE_CEE_NON_WILLING; + + ice_dcb_cfg_to_lldp(lldpmib, &miblen, dcbcfg); + ret = ice_aq_set_lldp_mib(hw, mib_type, (void *)lldpmib, miblen, + NULL); + + devm_kfree(ice_hw_to_dev(hw), lldpmib); + + return ret; +} + +/** + * ice_aq_query_port_ets - query port ets configuration + * @pi: port information structure + * @buf: pointer to buffer + * @buf_size: buffer size in bytes + * @cd: pointer to command details structure or NULL + * + * query current port ets configuration + */ +static enum ice_status +ice_aq_query_port_ets(struct ice_port_info *pi, + struct ice_aqc_port_ets_elem *buf, u16 buf_size, + struct ice_sq_cd *cd) +{ + struct ice_aqc_query_port_ets *cmd; + struct ice_aq_desc desc; + enum ice_status status; + + if (!pi) + return ICE_ERR_PARAM; + cmd = &desc.params.port_ets; + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_query_port_ets); + cmd->port_teid = pi->root->info.node_teid; + + status = ice_aq_send_cmd(pi->hw, &desc, buf, buf_size, cd); + return status; +} + +/** + * ice_update_port_tc_tree_cfg - update TC tree configuration + * @pi: port information structure + * @buf: pointer to buffer + * + * update the SW DB with the new TC changes + */ +static enum ice_status +ice_update_port_tc_tree_cfg(struct ice_port_info *pi, + struct ice_aqc_port_ets_elem *buf) +{ + struct ice_sched_node *node, *tc_node; + struct ice_aqc_get_elem elem; + enum ice_status status = 0; + u32 teid1, teid2; + u8 i, j; + + if (!pi) + return ICE_ERR_PARAM; + /* suspend the missing TC nodes */ + for (i = 0; i < pi->root->num_children; i++) { + teid1 = le32_to_cpu(pi->root->children[i]->info.node_teid); + ice_for_each_traffic_class(j) { + teid2 = le32_to_cpu(buf->tc_node_teid[j]); + if (teid1 == teid2) + break; + } + if (j < ICE_MAX_TRAFFIC_CLASS) + continue; + /* TC is missing */ + pi->root->children[i]->in_use = false; + } + /* add the new TC nodes */ + ice_for_each_traffic_class(j) { + teid2 = le32_to_cpu(buf->tc_node_teid[j]); + if (teid2 == ICE_INVAL_TEID) + continue; + /* Is it already present in the tree ? */ + for (i = 0; i < pi->root->num_children; i++) { + tc_node = pi->root->children[i]; + if (!tc_node) + continue; + teid1 = le32_to_cpu(tc_node->info.node_teid); + if (teid1 == teid2) { + tc_node->tc_num = j; + tc_node->in_use = true; + break; + } + } + if (i < pi->root->num_children) + continue; + /* new TC */ + status = ice_sched_query_elem(pi->hw, teid2, &elem); + if (!status) + status = ice_sched_add_node(pi, 1, &elem.generic[0]); + if (status) + break; + /* update the TC number */ + node = ice_sched_find_node_by_teid(pi->root, teid2); + if (node) + node->tc_num = j; + } + return status; +} + +/** + * ice_query_port_ets - query port ets configuration + * @pi: port information structure + * @buf: pointer to buffer + * @buf_size: buffer size in bytes + * @cd: pointer to command details structure or NULL + * + * query current port ets configuration and update the + * SW DB with the TC changes + */ +enum ice_status +ice_query_port_ets(struct ice_port_info *pi, + struct ice_aqc_port_ets_elem *buf, u16 buf_size, + struct ice_sq_cd *cd) +{ + enum ice_status status; + + mutex_lock(&pi->sched_lock); + status = ice_aq_query_port_ets(pi, buf, buf_size, cd); + if (!status) + status = ice_update_port_tc_tree_cfg(pi, buf); + mutex_unlock(&pi->sched_lock); + return status; +} diff --git a/drivers/net/ethernet/intel/ice/ice_dcb.h b/drivers/net/ethernet/intel/ice/ice_dcb.h new file mode 100644 index 000000000000..522e1452abe2 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_dcb.h @@ -0,0 +1,183 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Intel Corporation. */ + +#ifndef _ICE_DCB_H_ +#define _ICE_DCB_H_ + +#include "ice_type.h" + +#define ICE_DCBX_STATUS_NOT_STARTED 0 +#define ICE_DCBX_STATUS_IN_PROGRESS 1 +#define ICE_DCBX_STATUS_DONE 2 +#define ICE_DCBX_STATUS_DIS 7 + +#define ICE_TLV_TYPE_END 0 +#define ICE_TLV_TYPE_ORG 127 + +#define ICE_IEEE_8021QAZ_OUI 0x0080C2 +#define ICE_IEEE_SUBTYPE_ETS_CFG 9 +#define ICE_IEEE_SUBTYPE_ETS_REC 10 +#define ICE_IEEE_SUBTYPE_PFC_CFG 11 +#define ICE_IEEE_SUBTYPE_APP_PRI 12 + +#define ICE_CEE_DCBX_OUI 0x001B21 +#define ICE_CEE_DCBX_TYPE 2 +#define ICE_CEE_SUBTYPE_PG_CFG 2 +#define ICE_CEE_SUBTYPE_PFC_CFG 3 +#define ICE_CEE_SUBTYPE_APP_PRI 4 +#define ICE_CEE_MAX_FEAT_TYPE 3 +/* Defines for LLDP TLV header */ +#define ICE_LLDP_TLV_LEN_S 0 +#define ICE_LLDP_TLV_LEN_M (0x01FF << ICE_LLDP_TLV_LEN_S) +#define ICE_LLDP_TLV_TYPE_S 9 +#define ICE_LLDP_TLV_TYPE_M (0x7F << ICE_LLDP_TLV_TYPE_S) +#define ICE_LLDP_TLV_SUBTYPE_S 0 +#define ICE_LLDP_TLV_SUBTYPE_M (0xFF << ICE_LLDP_TLV_SUBTYPE_S) +#define ICE_LLDP_TLV_OUI_S 8 +#define ICE_LLDP_TLV_OUI_M (0xFFFFFFUL << ICE_LLDP_TLV_OUI_S) + +/* Defines for IEEE ETS TLV */ +#define ICE_IEEE_ETS_MAXTC_S 0 +#define ICE_IEEE_ETS_MAXTC_M (0x7 << ICE_IEEE_ETS_MAXTC_S) +#define ICE_IEEE_ETS_CBS_S 6 +#define ICE_IEEE_ETS_CBS_M BIT(ICE_IEEE_ETS_CBS_S) +#define ICE_IEEE_ETS_WILLING_S 7 +#define ICE_IEEE_ETS_WILLING_M BIT(ICE_IEEE_ETS_WILLING_S) +#define ICE_IEEE_ETS_PRIO_0_S 0 +#define ICE_IEEE_ETS_PRIO_0_M (0x7 << ICE_IEEE_ETS_PRIO_0_S) +#define ICE_IEEE_ETS_PRIO_1_S 4 +#define ICE_IEEE_ETS_PRIO_1_M (0x7 << ICE_IEEE_ETS_PRIO_1_S) +#define ICE_CEE_PGID_PRIO_0_S 0 +#define ICE_CEE_PGID_PRIO_0_M (0xF << ICE_CEE_PGID_PRIO_0_S) +#define ICE_CEE_PGID_PRIO_1_S 4 +#define ICE_CEE_PGID_PRIO_1_M (0xF << ICE_CEE_PGID_PRIO_1_S) +#define ICE_CEE_PGID_STRICT 15 + +/* Defines for IEEE TSA types */ +#define ICE_IEEE_TSA_STRICT 0 +#define ICE_IEEE_TSA_ETS 2 + +/* Defines for IEEE PFC TLV */ +#define ICE_IEEE_PFC_CAP_S 0 +#define ICE_IEEE_PFC_CAP_M (0xF << ICE_IEEE_PFC_CAP_S) +#define ICE_IEEE_PFC_MBC_S 6 +#define ICE_IEEE_PFC_MBC_M BIT(ICE_IEEE_PFC_MBC_S) +#define ICE_IEEE_PFC_WILLING_S 7 +#define ICE_IEEE_PFC_WILLING_M BIT(ICE_IEEE_PFC_WILLING_S) + +/* Defines for IEEE APP TLV */ +#define ICE_IEEE_APP_SEL_S 0 +#define ICE_IEEE_APP_SEL_M (0x7 << ICE_IEEE_APP_SEL_S) +#define ICE_IEEE_APP_PRIO_S 5 +#define ICE_IEEE_APP_PRIO_M (0x7 << ICE_IEEE_APP_PRIO_S) + +/* TLV definitions for preparing MIB */ +#define ICE_IEEE_TLV_ID_ETS_CFG 3 +#define ICE_IEEE_TLV_ID_ETS_REC 4 +#define ICE_IEEE_TLV_ID_PFC_CFG 5 +#define ICE_IEEE_TLV_ID_APP_PRI 6 +#define ICE_TLV_ID_END_OF_LLDPPDU 7 +#define ICE_TLV_ID_START ICE_IEEE_TLV_ID_ETS_CFG + +#define ICE_IEEE_ETS_TLV_LEN 25 +#define ICE_IEEE_PFC_TLV_LEN 6 +#define ICE_IEEE_APP_TLV_LEN 11 + +/* IEEE 802.1AB LLDP Organization specific TLV */ +struct ice_lldp_org_tlv { + __be16 typelen; + __be32 ouisubtype; + u8 tlvinfo[1]; +} __packed; + +struct ice_cee_tlv_hdr { + __be16 typelen; + u8 operver; + u8 maxver; +}; + +struct ice_cee_ctrl_tlv { + struct ice_cee_tlv_hdr hdr; + __be32 seqno; + __be32 ackno; +}; + +struct ice_cee_feat_tlv { + struct ice_cee_tlv_hdr hdr; + u8 en_will_err; /* Bits: |En|Will|Err|Reserved(5)| */ +#define ICE_CEE_FEAT_TLV_ENA_M 0x80 +#define ICE_CEE_FEAT_TLV_WILLING_M 0x40 +#define ICE_CEE_FEAT_TLV_ERR_M 0x20 + u8 subtype; + u8 tlvinfo[1]; +}; + +struct ice_cee_app_prio { + __be16 protocol; + u8 upper_oui_sel; /* Bits: |Upper OUI(6)|Selector(2)| */ +#define ICE_CEE_APP_SELECTOR_M 0x03 + __be16 lower_oui; + u8 prio_map; +} __packed; + +enum ice_status +ice_aq_get_dcb_cfg(struct ice_hw *hw, u8 mib_type, u8 bridgetype, + struct ice_dcbx_cfg *dcbcfg); +enum ice_status ice_get_dcb_cfg(struct ice_port_info *pi); +enum ice_status ice_set_dcb_cfg(struct ice_port_info *pi); +enum ice_status ice_init_dcb(struct ice_hw *hw); +enum ice_status +ice_query_port_ets(struct ice_port_info *pi, + struct ice_aqc_port_ets_elem *buf, u16 buf_size, + struct ice_sq_cd *cmd_details); +#ifdef CONFIG_DCB +enum ice_status +ice_aq_stop_lldp(struct ice_hw *hw, bool shutdown_lldp_agent, bool persist, + struct ice_sq_cd *cd); +enum ice_status +ice_aq_start_lldp(struct ice_hw *hw, bool persist, struct ice_sq_cd *cd); +enum ice_status +ice_aq_start_stop_dcbx(struct ice_hw *hw, bool start_dcbx_agent, + bool *dcbx_agent_status, struct ice_sq_cd *cd); +enum ice_status +ice_aq_cfg_lldp_mib_change(struct ice_hw *hw, bool ena_update, + struct ice_sq_cd *cd); +#else /* CONFIG_DCB */ +static inline enum ice_status +ice_aq_stop_lldp(struct ice_hw __always_unused *hw, + bool __always_unused shutdown_lldp_agent, + bool __always_unused persist, + struct ice_sq_cd __always_unused *cd) +{ + return 0; +} + +static inline enum ice_status +ice_aq_start_lldp(struct ice_hw __always_unused *hw, + bool __always_unused persist, + struct ice_sq_cd __always_unused *cd) +{ + return 0; +} + +static inline enum ice_status +ice_aq_start_stop_dcbx(struct ice_hw __always_unused *hw, + bool __always_unused start_dcbx_agent, + bool *dcbx_agent_status, + struct ice_sq_cd __always_unused *cd) +{ + *dcbx_agent_status = false; + + return 0; +} + +static inline enum ice_status +ice_aq_cfg_lldp_mib_change(struct ice_hw __always_unused *hw, + bool __always_unused ena_update, + struct ice_sq_cd __always_unused *cd) +{ + return 0; +} + +#endif /* CONFIG_DCB */ +#endif /* _ICE_DCB_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c new file mode 100644 index 000000000000..fe88b127ca42 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c @@ -0,0 +1,653 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2019, Intel Corporation. */ + +#include "ice_dcb_lib.h" + +/** + * ice_dcb_get_ena_tc - return bitmap of enabled TCs + * @dcbcfg: DCB config to evaluate for enabled TCs + */ +u8 ice_dcb_get_ena_tc(struct ice_dcbx_cfg *dcbcfg) +{ + u8 i, num_tc, ena_tc = 1; + + num_tc = ice_dcb_get_num_tc(dcbcfg); + + for (i = 0; i < num_tc; i++) + ena_tc |= BIT(i); + + return ena_tc; +} + +/** + * ice_dcb_get_num_tc - Get the number of TCs from DCBX config + * @dcbcfg: config to retrieve number of TCs from + */ +u8 ice_dcb_get_num_tc(struct ice_dcbx_cfg *dcbcfg) +{ + bool tc_unused = false; + u8 num_tc = 0; + u8 ret = 0; + int i; + + /* Scan the ETS Config Priority Table to find traffic classes + * enabled and create a bitmask of enabled TCs + */ + for (i = 0; i < CEE_DCBX_MAX_PRIO; i++) + num_tc |= BIT(dcbcfg->etscfg.prio_table[i]); + + /* Scan bitmask for contiguous TCs starting with TC0 */ + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { + if (num_tc & BIT(i)) { + if (!tc_unused) { + ret++; + } else { + pr_err("Non-contiguous TCs - Disabling DCB\n"); + return 1; + } + } else { + tc_unused = true; + } + } + + /* There is always at least 1 TC */ + if (!ret) + ret = 1; + + return ret; +} + +/** + * ice_vsi_cfg_dcb_rings - Update rings to reflect DCB TC + * @vsi: VSI owner of rings being updated + */ +void ice_vsi_cfg_dcb_rings(struct ice_vsi *vsi) +{ + struct ice_ring *tx_ring, *rx_ring; + u16 qoffset, qcount; + int i, n; + + if (!test_bit(ICE_FLAG_DCB_ENA, vsi->back->flags)) { + /* Reset the TC information */ + for (i = 0; i < vsi->num_txq; i++) { + tx_ring = vsi->tx_rings[i]; + tx_ring->dcb_tc = 0; + } + for (i = 0; i < vsi->num_rxq; i++) { + rx_ring = vsi->rx_rings[i]; + rx_ring->dcb_tc = 0; + } + return; + } + + ice_for_each_traffic_class(n) { + if (!(vsi->tc_cfg.ena_tc & BIT(n))) + break; + + qoffset = vsi->tc_cfg.tc_info[n].qoffset; + qcount = vsi->tc_cfg.tc_info[n].qcount_tx; + for (i = qoffset; i < (qoffset + qcount); i++) { + tx_ring = vsi->tx_rings[i]; + rx_ring = vsi->rx_rings[i]; + tx_ring->dcb_tc = n; + rx_ring->dcb_tc = n; + } + } +} + +/** + * ice_pf_dcb_recfg - Reconfigure all VEBs and VSIs + * @pf: pointer to the PF struct + * + * Assumed caller has already disabled all VSIs before + * calling this function. Reconfiguring DCB based on + * local_dcbx_cfg. + */ +static void ice_pf_dcb_recfg(struct ice_pf *pf) +{ + struct ice_dcbx_cfg *dcbcfg = &pf->hw.port_info->local_dcbx_cfg; + u8 tc_map = 0; + int v, ret; + + /* Update each VSI */ + ice_for_each_vsi(pf, v) { + if (!pf->vsi[v]) + continue; + + if (pf->vsi[v]->type == ICE_VSI_PF) + tc_map = ice_dcb_get_ena_tc(dcbcfg); + else + tc_map = ICE_DFLT_TRAFFIC_CLASS; + + ret = ice_vsi_cfg_tc(pf->vsi[v], tc_map); + if (ret) { + dev_err(&pf->pdev->dev, + "Failed to config TC for VSI index: %d\n", + pf->vsi[v]->idx); + continue; + } + + ice_vsi_map_rings_to_vectors(pf->vsi[v]); + } +} + +/** + * ice_pf_dcb_cfg - Apply new DCB configuration + * @pf: pointer to the PF struct + * @new_cfg: DCBX config to apply + * @locked: is the RTNL held + */ +static +int ice_pf_dcb_cfg(struct ice_pf *pf, struct ice_dcbx_cfg *new_cfg, bool locked) +{ + struct ice_dcbx_cfg *old_cfg, *curr_cfg; + struct ice_aqc_port_ets_elem buf = { 0 }; + int ret = 0; + + curr_cfg = &pf->hw.port_info->local_dcbx_cfg; + + /* Enable DCB tagging only when more than one TC */ + if (ice_dcb_get_num_tc(new_cfg) > 1) { + dev_dbg(&pf->pdev->dev, "DCB tagging enabled (num TC > 1)\n"); + set_bit(ICE_FLAG_DCB_ENA, pf->flags); + } else { + dev_dbg(&pf->pdev->dev, "DCB tagging disabled (num TC = 1)\n"); + clear_bit(ICE_FLAG_DCB_ENA, pf->flags); + } + + if (!memcmp(new_cfg, curr_cfg, sizeof(*new_cfg))) { + dev_dbg(&pf->pdev->dev, "No change in DCB config required\n"); + return ret; + } + + /* Store old config in case FW config fails */ + old_cfg = devm_kzalloc(&pf->pdev->dev, sizeof(*old_cfg), GFP_KERNEL); + memcpy(old_cfg, curr_cfg, sizeof(*old_cfg)); + + /* avoid race conditions by holding the lock while disabling and + * re-enabling the VSI + */ + if (!locked) + rtnl_lock(); + ice_pf_dis_all_vsi(pf, true); + + memcpy(curr_cfg, new_cfg, sizeof(*curr_cfg)); + memcpy(&curr_cfg->etsrec, &curr_cfg->etscfg, sizeof(curr_cfg->etsrec)); + + /* Only send new config to HW if we are in SW LLDP mode. Otherwise, + * the new config came from the HW in the first place. + */ + if (pf->hw.port_info->is_sw_lldp) { + ret = ice_set_dcb_cfg(pf->hw.port_info); + if (ret) { + dev_err(&pf->pdev->dev, "Set DCB Config failed\n"); + /* Restore previous settings to local config */ + memcpy(curr_cfg, old_cfg, sizeof(*curr_cfg)); + goto out; + } + } + + ret = ice_query_port_ets(pf->hw.port_info, &buf, sizeof(buf), NULL); + if (ret) { + dev_err(&pf->pdev->dev, "Query Port ETS failed\n"); + goto out; + } + + ice_pf_dcb_recfg(pf); + +out: + ice_pf_ena_all_vsi(pf, true); + if (!locked) + rtnl_unlock(); + devm_kfree(&pf->pdev->dev, old_cfg); + return ret; +} + +/** + * ice_dcb_rebuild - rebuild DCB post reset + * @pf: physical function instance + */ +void ice_dcb_rebuild(struct ice_pf *pf) +{ + struct ice_aqc_port_ets_elem buf = { 0 }; + struct ice_dcbx_cfg *prev_cfg; + enum ice_status ret; + u8 willing; + + ret = ice_query_port_ets(pf->hw.port_info, &buf, sizeof(buf), NULL); + if (ret) { + dev_err(&pf->pdev->dev, "Query Port ETS failed\n"); + goto dcb_error; + } + + /* If DCB was not enabled previously, we are done */ + if (!test_bit(ICE_FLAG_DCB_ENA, pf->flags)) + return; + + /* Save current willing state and force FW to unwilling */ + willing = pf->hw.port_info->local_dcbx_cfg.etscfg.willing; + pf->hw.port_info->local_dcbx_cfg.etscfg.willing = 0x0; + ret = ice_set_dcb_cfg(pf->hw.port_info); + if (ret) { + dev_err(&pf->pdev->dev, "Failed to set DCB to unwilling\n"); + goto dcb_error; + } + + /* Retrieve DCB config and ensure same as current in SW */ + prev_cfg = devm_kmemdup(&pf->pdev->dev, + &pf->hw.port_info->local_dcbx_cfg, + sizeof(*prev_cfg), GFP_KERNEL); + if (!prev_cfg) { + dev_err(&pf->pdev->dev, "Failed to alloc space for DCB cfg\n"); + goto dcb_error; + } + + ice_init_dcb(&pf->hw); + if (memcmp(prev_cfg, &pf->hw.port_info->local_dcbx_cfg, + sizeof(*prev_cfg))) { + /* difference in cfg detected - disable DCB till next MIB */ + dev_err(&pf->pdev->dev, "Set local MIB not accurate\n"); + devm_kfree(&pf->pdev->dev, prev_cfg); + goto dcb_error; + } + + /* fetched config congruent to previous configuration */ + devm_kfree(&pf->pdev->dev, prev_cfg); + + /* Configuration replayed - reset willing state to previous */ + pf->hw.port_info->local_dcbx_cfg.etscfg.willing = willing; + ret = ice_set_dcb_cfg(pf->hw.port_info); + if (ret) { + dev_err(&pf->pdev->dev, "Fail restoring prev willing state\n"); + goto dcb_error; + } + dev_info(&pf->pdev->dev, "DCB restored after reset\n"); + ret = ice_query_port_ets(pf->hw.port_info, &buf, sizeof(buf), NULL); + if (ret) { + dev_err(&pf->pdev->dev, "Query Port ETS failed\n"); + goto dcb_error; + } + + return; + +dcb_error: + dev_err(&pf->pdev->dev, "Disabling DCB until new settings occur\n"); + prev_cfg = devm_kzalloc(&pf->pdev->dev, sizeof(*prev_cfg), GFP_KERNEL); + prev_cfg->etscfg.willing = true; + prev_cfg->etscfg.tcbwtable[0] = ICE_TC_MAX_BW; + prev_cfg->etscfg.tsatable[0] = ICE_IEEE_TSA_ETS; + memcpy(&prev_cfg->etsrec, &prev_cfg->etscfg, sizeof(prev_cfg->etsrec)); + ice_pf_dcb_cfg(pf, prev_cfg, false); + devm_kfree(&pf->pdev->dev, prev_cfg); +} + +/** + * ice_dcb_init_cfg - set the initial DCB config in SW + * @pf: PF to apply config to + * @locked: Is the RTNL held + */ +static int ice_dcb_init_cfg(struct ice_pf *pf, bool locked) +{ + struct ice_dcbx_cfg *newcfg; + struct ice_port_info *pi; + int ret = 0; + + pi = pf->hw.port_info; + newcfg = devm_kzalloc(&pf->pdev->dev, sizeof(*newcfg), GFP_KERNEL); + if (!newcfg) + return -ENOMEM; + + memcpy(newcfg, &pi->local_dcbx_cfg, sizeof(*newcfg)); + memset(&pi->local_dcbx_cfg, 0, sizeof(*newcfg)); + + dev_info(&pf->pdev->dev, "Configuring initial DCB values\n"); + if (ice_pf_dcb_cfg(pf, newcfg, locked)) + ret = -EINVAL; + + devm_kfree(&pf->pdev->dev, newcfg); + + return ret; +} + +/** + * ice_dcb_sw_default_config - Apply a default DCB config + * @pf: PF to apply config to + * @locked: was this function called with RTNL held + */ +static int ice_dcb_sw_dflt_cfg(struct ice_pf *pf, bool locked) +{ + struct ice_aqc_port_ets_elem buf = { 0 }; + struct ice_dcbx_cfg *dcbcfg; + struct ice_port_info *pi; + struct ice_hw *hw; + int ret; + + hw = &pf->hw; + pi = hw->port_info; + dcbcfg = devm_kzalloc(&pf->pdev->dev, sizeof(*dcbcfg), GFP_KERNEL); + + memset(dcbcfg, 0, sizeof(*dcbcfg)); + memset(&pi->local_dcbx_cfg, 0, sizeof(*dcbcfg)); + + dcbcfg->etscfg.willing = 1; + dcbcfg->etscfg.maxtcs = 8; + dcbcfg->etscfg.tcbwtable[0] = 100; + dcbcfg->etscfg.tsatable[0] = ICE_IEEE_TSA_ETS; + + memcpy(&dcbcfg->etsrec, &dcbcfg->etscfg, + sizeof(dcbcfg->etsrec)); + dcbcfg->etsrec.willing = 0; + + dcbcfg->pfc.willing = 1; + dcbcfg->pfc.pfccap = IEEE_8021QAZ_MAX_TCS; + + dcbcfg->numapps = 1; + dcbcfg->app[0].selector = ICE_APP_SEL_ETHTYPE; + dcbcfg->app[0].priority = 3; + dcbcfg->app[0].prot_id = ICE_APP_PROT_ID_FCOE; + + ret = ice_pf_dcb_cfg(pf, dcbcfg, locked); + devm_kfree(&pf->pdev->dev, dcbcfg); + if (ret) + return ret; + + return ice_query_port_ets(pi, &buf, sizeof(buf), NULL); +} + +/** + * ice_init_pf_dcb - initialize DCB for a PF + * @pf: PF to initialize DCB for + * @locked: Was function called with RTNL held + */ +int ice_init_pf_dcb(struct ice_pf *pf, bool locked) +{ + struct device *dev = &pf->pdev->dev; + struct ice_port_info *port_info; + struct ice_hw *hw = &pf->hw; + int sw_default = 0; + int err; + + port_info = hw->port_info; + + err = ice_init_dcb(hw); + if (err) { + /* FW LLDP is not active, default to SW DCBX/LLDP */ + dev_info(&pf->pdev->dev, "FW LLDP is not active\n"); + hw->port_info->dcbx_status = ICE_DCBX_STATUS_NOT_STARTED; + hw->port_info->is_sw_lldp = true; + } + + if (port_info->dcbx_status == ICE_DCBX_STATUS_DIS) + dev_info(&pf->pdev->dev, "DCBX disabled\n"); + + /* LLDP disabled in FW */ + if (port_info->is_sw_lldp) { + sw_default = 1; + dev_info(&pf->pdev->dev, "DCBx/LLDP in SW mode.\n"); + clear_bit(ICE_FLAG_ENABLE_FW_LLDP, pf->flags); + } else { + set_bit(ICE_FLAG_ENABLE_FW_LLDP, pf->flags); + } + + if (port_info->dcbx_status == ICE_DCBX_STATUS_NOT_STARTED) + dev_info(&pf->pdev->dev, "DCBX not started\n"); + + if (sw_default) { + err = ice_dcb_sw_dflt_cfg(pf, locked); + if (err) { + dev_err(&pf->pdev->dev, + "Failed to set local DCB config %d\n", err); + err = -EIO; + goto dcb_init_err; + } + + pf->dcbx_cap = DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE; + set_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); + set_bit(ICE_FLAG_DCB_ENA, pf->flags); + return 0; + } + + /* DCBX in FW and LLDP enabled in FW */ + pf->dcbx_cap = DCB_CAP_DCBX_LLD_MANAGED | DCB_CAP_DCBX_VER_IEEE; + + set_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); + + err = ice_dcb_init_cfg(pf, locked); + if (err) + goto dcb_init_err; + + dev_info(&pf->pdev->dev, "DCBX offload supported\n"); + return err; + +dcb_init_err: + dev_err(dev, "DCB init failed\n"); + return err; +} + +/** + * ice_update_dcb_stats - Update DCB stats counters + * @pf: PF whose stats needs to be updated + */ +void ice_update_dcb_stats(struct ice_pf *pf) +{ + struct ice_hw_port_stats *prev_ps, *cur_ps; + struct ice_hw *hw = &pf->hw; + u8 pf_id = hw->pf_id; + int i; + + prev_ps = &pf->stats_prev; + cur_ps = &pf->stats; + + for (i = 0; i < 8; i++) { + ice_stat_update32(hw, GLPRT_PXOFFRXC(pf_id, i), + pf->stat_prev_loaded, + &prev_ps->priority_xoff_rx[i], + &cur_ps->priority_xoff_rx[i]); + ice_stat_update32(hw, GLPRT_PXONRXC(pf_id, i), + pf->stat_prev_loaded, + &prev_ps->priority_xon_rx[i], + &cur_ps->priority_xon_rx[i]); + ice_stat_update32(hw, GLPRT_PXONTXC(pf_id, i), + pf->stat_prev_loaded, + &prev_ps->priority_xon_tx[i], + &cur_ps->priority_xon_tx[i]); + ice_stat_update32(hw, GLPRT_PXOFFTXC(pf_id, i), + pf->stat_prev_loaded, + &prev_ps->priority_xoff_tx[i], + &cur_ps->priority_xoff_tx[i]); + ice_stat_update32(hw, GLPRT_RXON2OFFCNT(pf_id, i), + pf->stat_prev_loaded, + &prev_ps->priority_xon_2_xoff[i], + &cur_ps->priority_xon_2_xoff[i]); + } +} + +/** + * ice_tx_prepare_vlan_flags_dcb - prepare VLAN tagging for DCB + * @tx_ring: ring to send buffer on + * @first: pointer to struct ice_tx_buf + */ +int +ice_tx_prepare_vlan_flags_dcb(struct ice_ring *tx_ring, + struct ice_tx_buf *first) +{ + struct sk_buff *skb = first->skb; + + if (!test_bit(ICE_FLAG_DCB_ENA, tx_ring->vsi->back->flags)) + return 0; + + /* Insert 802.1p priority into VLAN header */ + if ((first->tx_flags & (ICE_TX_FLAGS_HW_VLAN | ICE_TX_FLAGS_SW_VLAN)) || + skb->priority != TC_PRIO_CONTROL) { + first->tx_flags &= ~ICE_TX_FLAGS_VLAN_PR_M; + /* Mask the lower 3 bits to set the 802.1p priority */ + first->tx_flags |= (skb->priority & 0x7) << + ICE_TX_FLAGS_VLAN_PR_S; + if (first->tx_flags & ICE_TX_FLAGS_SW_VLAN) { + struct vlan_ethhdr *vhdr; + int rc; + + rc = skb_cow_head(skb, 0); + if (rc < 0) + return rc; + vhdr = (struct vlan_ethhdr *)skb->data; + vhdr->h_vlan_TCI = htons(first->tx_flags >> + ICE_TX_FLAGS_VLAN_S); + } else { + first->tx_flags |= ICE_TX_FLAGS_HW_VLAN; + } + } + + return 0; +} + +/** + * ice_dcb_need_recfg - Check if DCB needs reconfig + * @pf: board private structure + * @old_cfg: current DCB config + * @new_cfg: new DCB config + */ +static bool ice_dcb_need_recfg(struct ice_pf *pf, struct ice_dcbx_cfg *old_cfg, + struct ice_dcbx_cfg *new_cfg) +{ + bool need_reconfig = false; + + /* Check if ETS configuration has changed */ + if (memcmp(&new_cfg->etscfg, &old_cfg->etscfg, + sizeof(new_cfg->etscfg))) { + /* If Priority Table has changed reconfig is needed */ + if (memcmp(&new_cfg->etscfg.prio_table, + &old_cfg->etscfg.prio_table, + sizeof(new_cfg->etscfg.prio_table))) { + need_reconfig = true; + dev_dbg(&pf->pdev->dev, "ETS UP2TC changed.\n"); + } + + if (memcmp(&new_cfg->etscfg.tcbwtable, + &old_cfg->etscfg.tcbwtable, + sizeof(new_cfg->etscfg.tcbwtable))) + dev_dbg(&pf->pdev->dev, "ETS TC BW Table changed.\n"); + + if (memcmp(&new_cfg->etscfg.tsatable, + &old_cfg->etscfg.tsatable, + sizeof(new_cfg->etscfg.tsatable))) + dev_dbg(&pf->pdev->dev, "ETS TSA Table changed.\n"); + } + + /* Check if PFC configuration has changed */ + if (memcmp(&new_cfg->pfc, &old_cfg->pfc, sizeof(new_cfg->pfc))) { + need_reconfig = true; + dev_dbg(&pf->pdev->dev, "PFC config change detected.\n"); + } + + /* Check if APP Table has changed */ + if (memcmp(&new_cfg->app, &old_cfg->app, sizeof(new_cfg->app))) { + need_reconfig = true; + dev_dbg(&pf->pdev->dev, "APP Table change detected.\n"); + } + + dev_dbg(&pf->pdev->dev, "dcb need_reconfig=%d\n", need_reconfig); + return need_reconfig; +} + +/** + * ice_dcb_process_lldp_set_mib_change - Process MIB change + * @pf: ptr to ice_pf + * @event: pointer to the admin queue receive event + */ +void +ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, + struct ice_rq_event_info *event) +{ + struct ice_aqc_port_ets_elem buf = { 0 }; + struct ice_aqc_lldp_get_mib *mib; + struct ice_dcbx_cfg tmp_dcbx_cfg; + bool need_reconfig = false; + struct ice_port_info *pi; + u8 type; + int ret; + + /* Not DCB capable or capability disabled */ + if (!(test_bit(ICE_FLAG_DCB_CAPABLE, pf->flags))) + return; + + if (pf->dcbx_cap & DCB_CAP_DCBX_HOST) { + dev_dbg(&pf->pdev->dev, + "MIB Change Event in HOST mode\n"); + return; + } + + pi = pf->hw.port_info; + mib = (struct ice_aqc_lldp_get_mib *)&event->desc.params.raw; + /* Ignore if event is not for Nearest Bridge */ + type = ((mib->type >> ICE_AQ_LLDP_BRID_TYPE_S) & + ICE_AQ_LLDP_BRID_TYPE_M); + dev_dbg(&pf->pdev->dev, "LLDP event MIB bridge type 0x%x\n", type); + if (type != ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID) + return; + + /* Check MIB Type and return if event for Remote MIB update */ + type = mib->type & ICE_AQ_LLDP_MIB_TYPE_M; + dev_dbg(&pf->pdev->dev, + "LLDP event mib type %s\n", type ? "remote" : "local"); + if (type == ICE_AQ_LLDP_MIB_REMOTE) { + /* Update the remote cached instance and return */ + ret = ice_aq_get_dcb_cfg(pi->hw, ICE_AQ_LLDP_MIB_REMOTE, + ICE_AQ_LLDP_BRID_TYPE_NEAREST_BRID, + &pi->remote_dcbx_cfg); + if (ret) { + dev_err(&pf->pdev->dev, "Failed to get remote DCB config\n"); + return; + } + } + + /* store the old configuration */ + tmp_dcbx_cfg = pf->hw.port_info->local_dcbx_cfg; + + /* Reset the old DCBX configuration data */ + memset(&pi->local_dcbx_cfg, 0, sizeof(pi->local_dcbx_cfg)); + + /* Get updated DCBX data from firmware */ + ret = ice_get_dcb_cfg(pf->hw.port_info); + if (ret) { + dev_err(&pf->pdev->dev, "Failed to get DCB config\n"); + return; + } + + /* No change detected in DCBX configs */ + if (!memcmp(&tmp_dcbx_cfg, &pi->local_dcbx_cfg, sizeof(tmp_dcbx_cfg))) { + dev_dbg(&pf->pdev->dev, + "No change detected in DCBX configuration.\n"); + return; + } + + need_reconfig = ice_dcb_need_recfg(pf, &tmp_dcbx_cfg, + &pi->local_dcbx_cfg); + if (!need_reconfig) + return; + + /* Enable DCB tagging only when more than one TC */ + if (ice_dcb_get_num_tc(&pi->local_dcbx_cfg) > 1) { + dev_dbg(&pf->pdev->dev, "DCB tagging enabled (num TC > 1)\n"); + set_bit(ICE_FLAG_DCB_ENA, pf->flags); + } else { + dev_dbg(&pf->pdev->dev, "DCB tagging disabled (num TC = 1)\n"); + clear_bit(ICE_FLAG_DCB_ENA, pf->flags); + } + + rtnl_lock(); + ice_pf_dis_all_vsi(pf, true); + + ret = ice_query_port_ets(pf->hw.port_info, &buf, sizeof(buf), NULL); + if (ret) { + dev_err(&pf->pdev->dev, "Query Port ETS failed\n"); + rtnl_unlock(); + return; + } + + /* changes in configuration update VSI */ + ice_pf_dcb_recfg(pf); + + ice_pf_ena_all_vsi(pf, true); + rtnl_unlock(); +} diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h new file mode 100644 index 000000000000..819081053ff5 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h @@ -0,0 +1,62 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Intel Corporation. */ + +#ifndef _ICE_DCB_LIB_H_ +#define _ICE_DCB_LIB_H_ + +#include "ice.h" +#include "ice_lib.h" + +#ifdef CONFIG_DCB +#define ICE_TC_MAX_BW 100 /* Default Max BW percentage */ + +void ice_dcb_rebuild(struct ice_pf *pf); +u8 ice_dcb_get_ena_tc(struct ice_dcbx_cfg *dcbcfg); +u8 ice_dcb_get_num_tc(struct ice_dcbx_cfg *dcbcfg); +void ice_vsi_cfg_dcb_rings(struct ice_vsi *vsi); +int ice_init_pf_dcb(struct ice_pf *pf, bool locked); +void ice_update_dcb_stats(struct ice_pf *pf); +int +ice_tx_prepare_vlan_flags_dcb(struct ice_ring *tx_ring, + struct ice_tx_buf *first); +void +ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, + struct ice_rq_event_info *event); +static inline void +ice_set_cgd_num(struct ice_tlan_ctx *tlan_ctx, struct ice_ring *ring) +{ + tlan_ctx->cgd_num = ring->dcb_tc; +} +#else +#define ice_dcb_rebuild(pf) do {} while (0) + +static inline u8 ice_dcb_get_ena_tc(struct ice_dcbx_cfg __always_unused *dcbcfg) +{ + return ICE_DFLT_TRAFFIC_CLASS; +} + +static inline u8 ice_dcb_get_num_tc(struct ice_dcbx_cfg __always_unused *dcbcfg) +{ + return 1; +} + +static inline int +ice_init_pf_dcb(struct ice_pf *pf, bool __always_unused locked) +{ + dev_dbg(&pf->pdev->dev, "DCB not supported\n"); + return -EOPNOTSUPP; +} + +static inline int +ice_tx_prepare_vlan_flags_dcb(struct ice_ring __always_unused *tx_ring, + struct ice_tx_buf __always_unused *first) +{ + return 0; +} + +#define ice_update_dcb_stats(pf) do {} while (0) +#define ice_vsi_cfg_dcb_rings(vsi) do {} while (0) +#define ice_dcb_process_lldp_set_mib_change(pf, event) do {} while (0) +#define ice_set_cgd_num(tlan_ctx, ring) do {} while (0) +#endif /* CONFIG_DCB */ +#endif /* _ICE_DCB_LIB_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index eb8d149e317c..52083a63dee6 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -4,6 +4,8 @@ /* ethtool support for ice */ #include "ice.h" +#include "ice_lib.h" +#include "ice_dcb_lib.h" struct ice_stats { char stat_string[ETH_GSTRING_LEN]; @@ -33,26 +35,50 @@ static int ice_q_stats_len(struct net_device *netdev) #define ICE_PF_STATS_LEN ARRAY_SIZE(ice_gstrings_pf_stats) #define ICE_VSI_STATS_LEN ARRAY_SIZE(ice_gstrings_vsi_stats) -#define ICE_ALL_STATS_LEN(n) (ICE_PF_STATS_LEN + ICE_VSI_STATS_LEN + \ - ice_q_stats_len(n)) +#define ICE_PFC_STATS_LEN ( \ + (FIELD_SIZEOF(struct ice_pf, stats.priority_xoff_rx) + \ + FIELD_SIZEOF(struct ice_pf, stats.priority_xon_rx) + \ + FIELD_SIZEOF(struct ice_pf, stats.priority_xoff_tx) + \ + FIELD_SIZEOF(struct ice_pf, stats.priority_xon_tx)) \ + / sizeof(u64)) +#define ICE_ALL_STATS_LEN(n) (ICE_PF_STATS_LEN + ICE_PFC_STATS_LEN + \ + ICE_VSI_STATS_LEN + ice_q_stats_len(n)) static const struct ice_stats ice_gstrings_vsi_stats[] = { - ICE_VSI_STAT("tx_unicast", eth_stats.tx_unicast), ICE_VSI_STAT("rx_unicast", eth_stats.rx_unicast), - ICE_VSI_STAT("tx_multicast", eth_stats.tx_multicast), + ICE_VSI_STAT("tx_unicast", eth_stats.tx_unicast), ICE_VSI_STAT("rx_multicast", eth_stats.rx_multicast), - ICE_VSI_STAT("tx_broadcast", eth_stats.tx_broadcast), + ICE_VSI_STAT("tx_multicast", eth_stats.tx_multicast), ICE_VSI_STAT("rx_broadcast", eth_stats.rx_broadcast), - ICE_VSI_STAT("tx_bytes", eth_stats.tx_bytes), + ICE_VSI_STAT("tx_broadcast", eth_stats.tx_broadcast), ICE_VSI_STAT("rx_bytes", eth_stats.rx_bytes), - ICE_VSI_STAT("rx_discards", eth_stats.rx_discards), - ICE_VSI_STAT("tx_errors", eth_stats.tx_errors), - ICE_VSI_STAT("tx_linearize", tx_linearize), + ICE_VSI_STAT("tx_bytes", eth_stats.tx_bytes), + ICE_VSI_STAT("rx_dropped", eth_stats.rx_discards), ICE_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol), ICE_VSI_STAT("rx_alloc_fail", rx_buf_failed), ICE_VSI_STAT("rx_pg_alloc_fail", rx_page_failed), + ICE_VSI_STAT("tx_errors", eth_stats.tx_errors), + ICE_VSI_STAT("tx_linearize", tx_linearize), }; +enum ice_ethtool_test_id { + ICE_ETH_TEST_REG = 0, + ICE_ETH_TEST_EEPROM, + ICE_ETH_TEST_INTR, + ICE_ETH_TEST_LOOP, + ICE_ETH_TEST_LINK, +}; + +static const char ice_gstrings_test[][ETH_GSTRING_LEN] = { + "Register test (offline)", + "EEPROM test (offline)", + "Interrupt test (offline)", + "Loopback test (offline)", + "Link test (on/offline)", +}; + +#define ICE_TEST_LEN (sizeof(ice_gstrings_test) / ETH_GSTRING_LEN) + /* These PF_STATs might look like duplicates of some NETDEV_STATs, * but they aren't. This device is capable of supporting multiple * VSIs/netdevs on a single PF. The NETDEV_STATs are for individual @@ -63,45 +89,45 @@ static const struct ice_stats ice_gstrings_vsi_stats[] = { * is queried on the base PF netdev. */ static const struct ice_stats ice_gstrings_pf_stats[] = { - ICE_PF_STAT("port.tx_bytes", stats.eth.tx_bytes), - ICE_PF_STAT("port.rx_bytes", stats.eth.rx_bytes), - ICE_PF_STAT("port.tx_unicast", stats.eth.tx_unicast), - ICE_PF_STAT("port.rx_unicast", stats.eth.rx_unicast), - ICE_PF_STAT("port.tx_multicast", stats.eth.tx_multicast), - ICE_PF_STAT("port.rx_multicast", stats.eth.rx_multicast), - ICE_PF_STAT("port.tx_broadcast", stats.eth.tx_broadcast), - ICE_PF_STAT("port.rx_broadcast", stats.eth.rx_broadcast), - ICE_PF_STAT("port.tx_errors", stats.eth.tx_errors), - ICE_PF_STAT("port.tx_size_64", stats.tx_size_64), - ICE_PF_STAT("port.rx_size_64", stats.rx_size_64), - ICE_PF_STAT("port.tx_size_127", stats.tx_size_127), - ICE_PF_STAT("port.rx_size_127", stats.rx_size_127), - ICE_PF_STAT("port.tx_size_255", stats.tx_size_255), - ICE_PF_STAT("port.rx_size_255", stats.rx_size_255), - ICE_PF_STAT("port.tx_size_511", stats.tx_size_511), - ICE_PF_STAT("port.rx_size_511", stats.rx_size_511), - ICE_PF_STAT("port.tx_size_1023", stats.tx_size_1023), - ICE_PF_STAT("port.rx_size_1023", stats.rx_size_1023), - ICE_PF_STAT("port.tx_size_1522", stats.tx_size_1522), - ICE_PF_STAT("port.rx_size_1522", stats.rx_size_1522), - ICE_PF_STAT("port.tx_size_big", stats.tx_size_big), - ICE_PF_STAT("port.rx_size_big", stats.rx_size_big), - ICE_PF_STAT("port.link_xon_tx", stats.link_xon_tx), - ICE_PF_STAT("port.link_xon_rx", stats.link_xon_rx), - ICE_PF_STAT("port.link_xoff_tx", stats.link_xoff_tx), - ICE_PF_STAT("port.link_xoff_rx", stats.link_xoff_rx), - ICE_PF_STAT("port.tx_dropped_link_down", stats.tx_dropped_link_down), - ICE_PF_STAT("port.rx_undersize", stats.rx_undersize), - ICE_PF_STAT("port.rx_fragments", stats.rx_fragments), - ICE_PF_STAT("port.rx_oversize", stats.rx_oversize), - ICE_PF_STAT("port.rx_jabber", stats.rx_jabber), - ICE_PF_STAT("port.rx_csum_bad", hw_csum_rx_error), - ICE_PF_STAT("port.rx_length_errors", stats.rx_len_errors), - ICE_PF_STAT("port.rx_dropped", stats.eth.rx_discards), - ICE_PF_STAT("port.rx_crc_errors", stats.crc_errors), - ICE_PF_STAT("port.illegal_bytes", stats.illegal_bytes), - ICE_PF_STAT("port.mac_local_faults", stats.mac_local_faults), - ICE_PF_STAT("port.mac_remote_faults", stats.mac_remote_faults), + ICE_PF_STAT("rx_bytes.nic", stats.eth.rx_bytes), + ICE_PF_STAT("tx_bytes.nic", stats.eth.tx_bytes), + ICE_PF_STAT("rx_unicast.nic", stats.eth.rx_unicast), + ICE_PF_STAT("tx_unicast.nic", stats.eth.tx_unicast), + ICE_PF_STAT("rx_multicast.nic", stats.eth.rx_multicast), + ICE_PF_STAT("tx_multicast.nic", stats.eth.tx_multicast), + ICE_PF_STAT("rx_broadcast.nic", stats.eth.rx_broadcast), + ICE_PF_STAT("tx_broadcast.nic", stats.eth.tx_broadcast), + ICE_PF_STAT("tx_errors.nic", stats.eth.tx_errors), + ICE_PF_STAT("rx_size_64.nic", stats.rx_size_64), + ICE_PF_STAT("tx_size_64.nic", stats.tx_size_64), + ICE_PF_STAT("rx_size_127.nic", stats.rx_size_127), + ICE_PF_STAT("tx_size_127.nic", stats.tx_size_127), + ICE_PF_STAT("rx_size_255.nic", stats.rx_size_255), + ICE_PF_STAT("tx_size_255.nic", stats.tx_size_255), + ICE_PF_STAT("rx_size_511.nic", stats.rx_size_511), + ICE_PF_STAT("tx_size_511.nic", stats.tx_size_511), + ICE_PF_STAT("rx_size_1023.nic", stats.rx_size_1023), + ICE_PF_STAT("tx_size_1023.nic", stats.tx_size_1023), + ICE_PF_STAT("rx_size_1522.nic", stats.rx_size_1522), + ICE_PF_STAT("tx_size_1522.nic", stats.tx_size_1522), + ICE_PF_STAT("rx_size_big.nic", stats.rx_size_big), + ICE_PF_STAT("tx_size_big.nic", stats.tx_size_big), + ICE_PF_STAT("link_xon_rx.nic", stats.link_xon_rx), + ICE_PF_STAT("link_xon_tx.nic", stats.link_xon_tx), + ICE_PF_STAT("link_xoff_rx.nic", stats.link_xoff_rx), + ICE_PF_STAT("link_xoff_tx.nic", stats.link_xoff_tx), + ICE_PF_STAT("tx_dropped_link_down.nic", stats.tx_dropped_link_down), + ICE_PF_STAT("rx_undersize.nic", stats.rx_undersize), + ICE_PF_STAT("rx_fragments.nic", stats.rx_fragments), + ICE_PF_STAT("rx_oversize.nic", stats.rx_oversize), + ICE_PF_STAT("rx_jabber.nic", stats.rx_jabber), + ICE_PF_STAT("rx_csum_bad.nic", hw_csum_rx_error), + ICE_PF_STAT("rx_length_errors.nic", stats.rx_len_errors), + ICE_PF_STAT("rx_dropped.nic", stats.eth.rx_discards), + ICE_PF_STAT("rx_crc_errors.nic", stats.crc_errors), + ICE_PF_STAT("illegal_bytes.nic", stats.illegal_bytes), + ICE_PF_STAT("mac_local_faults.nic", stats.mac_local_faults), + ICE_PF_STAT("mac_remote_faults.nic", stats.mac_remote_faults), }; static const u32 ice_regs_dump_list[] = { @@ -112,6 +138,9 @@ static const u32 ice_regs_dump_list[] = { QINT_RQCTL(0), PFINT_OICR_ENA, QRX_ITR(0), + PF0INT_ITR_0(0), + PF0INT_ITR_1(0), + PF0INT_ITR_2(0), }; struct ice_priv_flag { @@ -126,6 +155,7 @@ struct ice_priv_flag { static const struct ice_priv_flag ice_gstrings_priv_flags[] = { ICE_PRIV_FLAG("link-down-on-close", ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA), + ICE_PRIV_FLAG("enable-fw-lldp", ICE_FLAG_ENABLE_FW_LLDP), }; #define ICE_PRIV_FLAG_ARRAY_SIZE ARRAY_SIZE(ice_gstrings_priv_flags) @@ -269,6 +299,571 @@ out: return ret; } +/** + * ice_active_vfs - check if there are any active VFs + * @pf: board private structure + * + * Returns true if an active VF is found, otherwise returns false + */ +static bool ice_active_vfs(struct ice_pf *pf) +{ + struct ice_vf *vf = pf->vf; + int i; + + for (i = 0; i < pf->num_alloc_vfs; i++, vf++) + if (test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) + return true; + return false; +} + +/** + * ice_link_test - perform a link test on a given net_device + * @netdev: network interface device structure + * + * This function performs one of the self-tests required by ethtool. + * Returns 0 on success, non-zero on failure. + */ +static u64 ice_link_test(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + enum ice_status status; + bool link_up = false; + + netdev_info(netdev, "link test\n"); + status = ice_get_link_status(np->vsi->port_info, &link_up); + if (status) { + netdev_err(netdev, "link query error, status = %d\n", status); + return 1; + } + + if (!link_up) + return 2; + + return 0; +} + +/** + * ice_eeprom_test - perform an EEPROM test on a given net_device + * @netdev: network interface device structure + * + * This function performs one of the self-tests required by ethtool. + * Returns 0 on success, non-zero on failure. + */ +static u64 ice_eeprom_test(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_pf *pf = np->vsi->back; + + netdev_info(netdev, "EEPROM test\n"); + return !!(ice_nvm_validate_checksum(&pf->hw)); +} + +/** + * ice_reg_pattern_test + * @hw: pointer to the HW struct + * @reg: reg to be tested + * @mask: bits to be touched + */ +static int ice_reg_pattern_test(struct ice_hw *hw, u32 reg, u32 mask) +{ + struct ice_pf *pf = (struct ice_pf *)hw->back; + static const u32 patterns[] = { + 0x5A5A5A5A, 0xA5A5A5A5, + 0x00000000, 0xFFFFFFFF + }; + u32 val, orig_val; + int i; + + orig_val = rd32(hw, reg); + for (i = 0; i < ARRAY_SIZE(patterns); ++i) { + u32 pattern = patterns[i] & mask; + + wr32(hw, reg, pattern); + val = rd32(hw, reg); + if (val == pattern) + continue; + dev_err(&pf->pdev->dev, + "%s: reg pattern test failed - reg 0x%08x pat 0x%08x val 0x%08x\n" + , __func__, reg, pattern, val); + return 1; + } + + wr32(hw, reg, orig_val); + val = rd32(hw, reg); + if (val != orig_val) { + dev_err(&pf->pdev->dev, + "%s: reg restore test failed - reg 0x%08x orig 0x%08x val 0x%08x\n" + , __func__, reg, orig_val, val); + return 1; + } + + return 0; +} + +/** + * ice_reg_test - perform a register test on a given net_device + * @netdev: network interface device structure + * + * This function performs one of the self-tests required by ethtool. + * Returns 0 on success, non-zero on failure. + */ +static u64 ice_reg_test(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_hw *hw = np->vsi->port_info->hw; + u32 int_elements = hw->func_caps.common_cap.num_msix_vectors ? + hw->func_caps.common_cap.num_msix_vectors - 1 : 1; + struct ice_diag_reg_test_info { + u32 address; + u32 mask; + u32 elem_num; + u32 elem_size; + } ice_reg_list[] = { + {GLINT_ITR(0, 0), 0x00000fff, int_elements, + GLINT_ITR(0, 1) - GLINT_ITR(0, 0)}, + {GLINT_ITR(1, 0), 0x00000fff, int_elements, + GLINT_ITR(1, 1) - GLINT_ITR(1, 0)}, + {GLINT_ITR(0, 0), 0x00000fff, int_elements, + GLINT_ITR(2, 1) - GLINT_ITR(2, 0)}, + {GLINT_CTL, 0xffff0001, 1, 0} + }; + int i; + + netdev_dbg(netdev, "Register test\n"); + for (i = 0; i < ARRAY_SIZE(ice_reg_list); ++i) { + u32 j; + + for (j = 0; j < ice_reg_list[i].elem_num; ++j) { + u32 mask = ice_reg_list[i].mask; + u32 reg = ice_reg_list[i].address + + (j * ice_reg_list[i].elem_size); + + /* bail on failure (non-zero return) */ + if (ice_reg_pattern_test(hw, reg, mask)) + return 1; + } + } + + return 0; +} + +/** + * ice_lbtest_prepare_rings - configure Tx/Rx test rings + * @vsi: pointer to the VSI structure + * + * Function configures rings of a VSI for loopback test without + * enabling interrupts or informing the kernel about new queues. + * + * Returns 0 on success, negative on failure. + */ +static int ice_lbtest_prepare_rings(struct ice_vsi *vsi) +{ + int status; + + status = ice_vsi_setup_tx_rings(vsi); + if (status) + goto err_setup_tx_ring; + + status = ice_vsi_setup_rx_rings(vsi); + if (status) + goto err_setup_rx_ring; + + status = ice_vsi_cfg(vsi); + if (status) + goto err_setup_rx_ring; + + status = ice_vsi_start_rx_rings(vsi); + if (status) + goto err_start_rx_ring; + + return status; + +err_start_rx_ring: + ice_vsi_free_rx_rings(vsi); +err_setup_rx_ring: + ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, 0); +err_setup_tx_ring: + ice_vsi_free_tx_rings(vsi); + + return status; +} + +/** + * ice_lbtest_disable_rings - disable Tx/Rx test rings after loopback test + * @vsi: pointer to the VSI structure + * + * Function stops and frees VSI rings after a loopback test. + * Returns 0 on success, negative on failure. + */ +static int ice_lbtest_disable_rings(struct ice_vsi *vsi) +{ + int status; + + status = ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, 0); + if (status) + netdev_err(vsi->netdev, "Failed to stop Tx rings, VSI %d error %d\n", + vsi->vsi_num, status); + + status = ice_vsi_stop_rx_rings(vsi); + if (status) + netdev_err(vsi->netdev, "Failed to stop Rx rings, VSI %d error %d\n", + vsi->vsi_num, status); + + ice_vsi_free_tx_rings(vsi); + ice_vsi_free_rx_rings(vsi); + + return status; +} + +/** + * ice_lbtest_create_frame - create test packet + * @pf: pointer to the PF structure + * @ret_data: allocated frame buffer + * @size: size of the packet data + * + * Function allocates a frame with a test pattern on specific offsets. + * Returns 0 on success, non-zero on failure. + */ +static int ice_lbtest_create_frame(struct ice_pf *pf, u8 **ret_data, u16 size) +{ + u8 *data; + + if (!pf) + return -EINVAL; + + data = devm_kzalloc(&pf->pdev->dev, size, GFP_KERNEL); + if (!data) + return -ENOMEM; + + /* Since the ethernet test frame should always be at least + * 64 bytes long, fill some octets in the payload with test data. + */ + memset(data, 0xFF, size); + data[32] = 0xDE; + data[42] = 0xAD; + data[44] = 0xBE; + data[46] = 0xEF; + + *ret_data = data; + + return 0; +} + +/** + * ice_lbtest_check_frame - verify received loopback frame + * @frame: pointer to the raw packet data + * + * Function verifies received test frame with a pattern. + * Returns true if frame matches the pattern, false otherwise. + */ +static bool ice_lbtest_check_frame(u8 *frame) +{ + /* Validate bytes of a frame under offsets chosen earlier */ + if (frame[32] == 0xDE && + frame[42] == 0xAD && + frame[44] == 0xBE && + frame[46] == 0xEF && + frame[48] == 0xFF) + return true; + + return false; +} + +/** + * ice_diag_send - send test frames to the test ring + * @tx_ring: pointer to the transmit ring + * @data: pointer to the raw packet data + * @size: size of the packet to send + * + * Function sends loopback packets on a test Tx ring. + */ +static int ice_diag_send(struct ice_ring *tx_ring, u8 *data, u16 size) +{ + struct ice_tx_desc *tx_desc; + struct ice_tx_buf *tx_buf; + dma_addr_t dma; + u64 td_cmd; + + tx_desc = ICE_TX_DESC(tx_ring, tx_ring->next_to_use); + tx_buf = &tx_ring->tx_buf[tx_ring->next_to_use]; + + dma = dma_map_single(tx_ring->dev, data, size, DMA_TO_DEVICE); + if (dma_mapping_error(tx_ring->dev, dma)) + return -EINVAL; + + tx_desc->buf_addr = cpu_to_le64(dma); + + /* These flags are required for a descriptor to be pushed out */ + td_cmd = (u64)(ICE_TX_DESC_CMD_EOP | ICE_TX_DESC_CMD_RS); + tx_desc->cmd_type_offset_bsz = + cpu_to_le64(ICE_TX_DESC_DTYPE_DATA | + (td_cmd << ICE_TXD_QW1_CMD_S) | + ((u64)0 << ICE_TXD_QW1_OFFSET_S) | + ((u64)size << ICE_TXD_QW1_TX_BUF_SZ_S) | + ((u64)0 << ICE_TXD_QW1_L2TAG1_S)); + + tx_buf->next_to_watch = tx_desc; + + /* Force memory write to complete before letting h/w know + * there are new descriptors to fetch. + */ + wmb(); + + tx_ring->next_to_use++; + if (tx_ring->next_to_use >= tx_ring->count) + tx_ring->next_to_use = 0; + + writel_relaxed(tx_ring->next_to_use, tx_ring->tail); + + /* Wait until the packets get transmitted to the receive queue. */ + usleep_range(1000, 2000); + dma_unmap_single(tx_ring->dev, dma, size, DMA_TO_DEVICE); + + return 0; +} + +#define ICE_LB_FRAME_SIZE 64 +/** + * ice_lbtest_receive_frames - receive and verify test frames + * @rx_ring: pointer to the receive ring + * + * Function receives loopback packets and verify their correctness. + * Returns number of received valid frames. + */ +static int ice_lbtest_receive_frames(struct ice_ring *rx_ring) +{ + struct ice_rx_buf *rx_buf; + int valid_frames, i; + u8 *received_buf; + + valid_frames = 0; + + for (i = 0; i < rx_ring->count; i++) { + union ice_32b_rx_flex_desc *rx_desc; + + rx_desc = ICE_RX_DESC(rx_ring, i); + + if (!(rx_desc->wb.status_error0 & + cpu_to_le16(ICE_TX_DESC_CMD_EOP | ICE_TX_DESC_CMD_RS))) + continue; + + rx_buf = &rx_ring->rx_buf[i]; + received_buf = page_address(rx_buf->page); + + if (ice_lbtest_check_frame(received_buf)) + valid_frames++; + } + + return valid_frames; +} + +/** + * ice_loopback_test - perform a loopback test on a given net_device + * @netdev: network interface device structure + * + * This function performs one of the self-tests required by ethtool. + * Returns 0 on success, non-zero on failure. + */ +static u64 ice_loopback_test(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *orig_vsi = np->vsi, *test_vsi; + struct ice_pf *pf = orig_vsi->back; + struct ice_ring *tx_ring, *rx_ring; + u8 broadcast[ETH_ALEN], ret = 0; + int num_frames, valid_frames; + LIST_HEAD(tmp_list); + u8 *tx_frame; + int i; + + netdev_info(netdev, "loopback test\n"); + + test_vsi = ice_lb_vsi_setup(pf, pf->hw.port_info); + if (!test_vsi) { + netdev_err(netdev, "Failed to create a VSI for the loopback test"); + return 1; + } + + test_vsi->netdev = netdev; + tx_ring = test_vsi->tx_rings[0]; + rx_ring = test_vsi->rx_rings[0]; + + if (ice_lbtest_prepare_rings(test_vsi)) { + ret = 2; + goto lbtest_vsi_close; + } + + if (ice_alloc_rx_bufs(rx_ring, rx_ring->count)) { + ret = 3; + goto lbtest_rings_dis; + } + + /* Enable MAC loopback in firmware */ + if (ice_aq_set_mac_loopback(&pf->hw, true, NULL)) { + ret = 4; + goto lbtest_mac_dis; + } + + /* Test VSI needs to receive broadcast packets */ + eth_broadcast_addr(broadcast); + if (ice_add_mac_to_list(test_vsi, &tmp_list, broadcast)) { + ret = 5; + goto lbtest_mac_dis; + } + + if (ice_add_mac(&pf->hw, &tmp_list)) { + ret = 6; + goto free_mac_list; + } + + if (ice_lbtest_create_frame(pf, &tx_frame, ICE_LB_FRAME_SIZE)) { + ret = 7; + goto remove_mac_filters; + } + + num_frames = min_t(int, tx_ring->count, 32); + for (i = 0; i < num_frames; i++) { + if (ice_diag_send(tx_ring, tx_frame, ICE_LB_FRAME_SIZE)) { + ret = 8; + goto lbtest_free_frame; + } + } + + valid_frames = ice_lbtest_receive_frames(rx_ring); + if (!valid_frames) + ret = 9; + else if (valid_frames != num_frames) + ret = 10; + +lbtest_free_frame: + devm_kfree(&pf->pdev->dev, tx_frame); +remove_mac_filters: + if (ice_remove_mac(&pf->hw, &tmp_list)) + netdev_err(netdev, "Could not remove MAC filter for the test VSI"); +free_mac_list: + ice_free_fltr_list(&pf->pdev->dev, &tmp_list); +lbtest_mac_dis: + /* Disable MAC loopback after the test is completed. */ + if (ice_aq_set_mac_loopback(&pf->hw, false, NULL)) + netdev_err(netdev, "Could not disable MAC loopback\n"); +lbtest_rings_dis: + if (ice_lbtest_disable_rings(test_vsi)) + netdev_err(netdev, "Could not disable test rings\n"); +lbtest_vsi_close: + test_vsi->netdev = NULL; + if (ice_vsi_release(test_vsi)) + netdev_err(netdev, "Failed to remove the test VSI"); + + return ret; +} + +/** + * ice_intr_test - perform an interrupt test on a given net_device + * @netdev: network interface device structure + * + * This function performs one of the self-tests required by ethtool. + * Returns 0 on success, non-zero on failure. + */ +static u64 ice_intr_test(struct net_device *netdev) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_pf *pf = np->vsi->back; + u16 swic_old = pf->sw_int_count; + + netdev_info(netdev, "interrupt test\n"); + + wr32(&pf->hw, GLINT_DYN_CTL(pf->oicr_idx), + GLINT_DYN_CTL_SW_ITR_INDX_M | + GLINT_DYN_CTL_INTENA_MSK_M | + GLINT_DYN_CTL_SWINT_TRIG_M); + + usleep_range(1000, 2000); + return (swic_old == pf->sw_int_count); +} + +/** + * ice_self_test - handler function for performing a self-test by ethtool + * @netdev: network interface device structure + * @eth_test: ethtool_test structure + * @data: required by ethtool.self_test + * + * This function is called after invoking 'ethtool -t devname' command where + * devname is the name of the network device on which ethtool should operate. + * It performs a set of self-tests to check if a device works properly. + */ +static void +ice_self_test(struct net_device *netdev, struct ethtool_test *eth_test, + u64 *data) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + bool if_running = netif_running(netdev); + struct ice_pf *pf = np->vsi->back; + + if (eth_test->flags == ETH_TEST_FL_OFFLINE) { + netdev_info(netdev, "offline testing starting\n"); + + set_bit(__ICE_TESTING, pf->state); + + if (ice_active_vfs(pf)) { + dev_warn(&pf->pdev->dev, + "Please take active VFs and Netqueues offline and restart the adapter before running NIC diagnostics\n"); + data[ICE_ETH_TEST_REG] = 1; + data[ICE_ETH_TEST_EEPROM] = 1; + data[ICE_ETH_TEST_INTR] = 1; + data[ICE_ETH_TEST_LOOP] = 1; + data[ICE_ETH_TEST_LINK] = 1; + eth_test->flags |= ETH_TEST_FL_FAILED; + clear_bit(__ICE_TESTING, pf->state); + goto skip_ol_tests; + } + /* If the device is online then take it offline */ + if (if_running) + /* indicate we're in test mode */ + ice_stop(netdev); + + data[ICE_ETH_TEST_LINK] = ice_link_test(netdev); + data[ICE_ETH_TEST_EEPROM] = ice_eeprom_test(netdev); + data[ICE_ETH_TEST_INTR] = ice_intr_test(netdev); + data[ICE_ETH_TEST_LOOP] = ice_loopback_test(netdev); + data[ICE_ETH_TEST_REG] = ice_reg_test(netdev); + + if (data[ICE_ETH_TEST_LINK] || + data[ICE_ETH_TEST_EEPROM] || + data[ICE_ETH_TEST_LOOP] || + data[ICE_ETH_TEST_INTR] || + data[ICE_ETH_TEST_REG]) + eth_test->flags |= ETH_TEST_FL_FAILED; + + clear_bit(__ICE_TESTING, pf->state); + + if (if_running) { + int status = ice_open(netdev); + + if (status) { + dev_err(&pf->pdev->dev, + "Could not open device %s, err %d", + pf->int_name, status); + } + } + } else { + /* Online tests */ + netdev_info(netdev, "online testing starting\n"); + + data[ICE_ETH_TEST_LINK] = ice_link_test(netdev); + if (data[ICE_ETH_TEST_LINK]) + eth_test->flags |= ETH_TEST_FL_FAILED; + + /* Offline only tests, not run in online; pass by default */ + data[ICE_ETH_TEST_REG] = 0; + data[ICE_ETH_TEST_EEPROM] = 0; + data[ICE_ETH_TEST_INTR] = 0; + data[ICE_ETH_TEST_LOOP] = 0; + } + +skip_ol_tests: + netdev_info(netdev, "testing finished\n"); +} + static void ice_get_strings(struct net_device *netdev, u32 stringset, u8 *data) { struct ice_netdev_priv *np = netdev_priv(netdev); @@ -286,17 +881,17 @@ static void ice_get_strings(struct net_device *netdev, u32 stringset, u8 *data) ice_for_each_alloc_txq(vsi, i) { snprintf(p, ETH_GSTRING_LEN, - "tx-queue-%u.tx_packets", i); + "tx_queue_%u_packets", i); p += ETH_GSTRING_LEN; - snprintf(p, ETH_GSTRING_LEN, "tx-queue-%u.tx_bytes", i); + snprintf(p, ETH_GSTRING_LEN, "tx_queue_%u_bytes", i); p += ETH_GSTRING_LEN; } ice_for_each_alloc_rxq(vsi, i) { snprintf(p, ETH_GSTRING_LEN, - "rx-queue-%u.rx_packets", i); + "rx_queue_%u_packets", i); p += ETH_GSTRING_LEN; - snprintf(p, ETH_GSTRING_LEN, "rx-queue-%u.rx_bytes", i); + snprintf(p, ETH_GSTRING_LEN, "rx_queue_%u_bytes", i); p += ETH_GSTRING_LEN; } @@ -309,6 +904,25 @@ static void ice_get_strings(struct net_device *netdev, u32 stringset, u8 *data) p += ETH_GSTRING_LEN; } + for (i = 0; i < ICE_MAX_USER_PRIORITY; i++) { + snprintf(p, ETH_GSTRING_LEN, + "tx_priority_%u_xon.nic", i); + p += ETH_GSTRING_LEN; + snprintf(p, ETH_GSTRING_LEN, + "tx_priority_%u_xoff.nic", i); + p += ETH_GSTRING_LEN; + } + for (i = 0; i < ICE_MAX_USER_PRIORITY; i++) { + snprintf(p, ETH_GSTRING_LEN, + "rx_priority_%u_xon.nic", i); + p += ETH_GSTRING_LEN; + snprintf(p, ETH_GSTRING_LEN, + "rx_priority_%u_xoff.nic", i); + p += ETH_GSTRING_LEN; + } + break; + case ETH_SS_TEST: + memcpy(data, ice_gstrings_test, ICE_TEST_LEN * ETH_GSTRING_LEN); break; case ETH_SS_PRIV_FLAGS: for (i = 0; i < ICE_PRIV_FLAG_ARRAY_SIZE; i++) { @@ -346,6 +960,185 @@ ice_set_phys_id(struct net_device *netdev, enum ethtool_phys_id_state state) } /** + * ice_set_fec_cfg - Set link FEC options + * @netdev: network interface device structure + * @req_fec: FEC mode to configure + */ +static int ice_set_fec_cfg(struct net_device *netdev, enum ice_fec_mode req_fec) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_aqc_set_phy_cfg_data config = { 0 }; + struct ice_aqc_get_phy_caps_data *caps; + struct ice_vsi *vsi = np->vsi; + u8 sw_cfg_caps, sw_cfg_fec; + struct ice_port_info *pi; + enum ice_status status; + int err = 0; + + pi = vsi->port_info; + if (!pi) + return -EOPNOTSUPP; + + /* Changing the FEC parameters is not supported if not the PF VSI */ + if (vsi->type != ICE_VSI_PF) { + netdev_info(netdev, "Changing FEC parameters only supported for PF VSI\n"); + return -EOPNOTSUPP; + } + + /* Get last SW configuration */ + caps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*caps), GFP_KERNEL); + if (!caps) + return -ENOMEM; + + status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, + caps, NULL); + if (status) { + err = -EAGAIN; + goto done; + } + + /* Copy SW configuration returned from PHY caps to PHY config */ + ice_copy_phy_caps_to_cfg(caps, &config); + sw_cfg_caps = caps->caps; + sw_cfg_fec = caps->link_fec_options; + + /* Get toloplogy caps, then copy PHY FEC topoloy caps to PHY config */ + memset(caps, 0, sizeof(*caps)); + + status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP, + caps, NULL); + if (status) { + err = -EAGAIN; + goto done; + } + + config.caps |= (caps->caps & ICE_AQC_PHY_EN_AUTO_FEC); + config.link_fec_opt = caps->link_fec_options; + + ice_cfg_phy_fec(&config, req_fec); + + /* If FEC mode has changed, then set PHY configuration and enable AN. */ + if ((config.caps & ICE_AQ_PHY_ENA_AUTO_FEC) != + (sw_cfg_caps & ICE_AQC_PHY_EN_AUTO_FEC) || + config.link_fec_opt != sw_cfg_fec) { + if (caps->caps & ICE_AQC_PHY_AN_MODE) + config.caps |= ICE_AQ_PHY_ENA_AUTO_LINK_UPDT; + + status = ice_aq_set_phy_cfg(pi->hw, pi->lport, &config, NULL); + + if (status) + err = -EAGAIN; + } + +done: + devm_kfree(&vsi->back->pdev->dev, caps); + return err; +} + +/** + * ice_set_fecparam - Set FEC link options + * @netdev: network interface device structure + * @fecparam: Ethtool structure to retrieve FEC parameters + */ +static int +ice_set_fecparam(struct net_device *netdev, struct ethtool_fecparam *fecparam) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + enum ice_fec_mode fec; + + switch (fecparam->fec) { + case ETHTOOL_FEC_AUTO: + fec = ICE_FEC_AUTO; + break; + case ETHTOOL_FEC_RS: + fec = ICE_FEC_RS; + break; + case ETHTOOL_FEC_BASER: + fec = ICE_FEC_BASER; + break; + case ETHTOOL_FEC_OFF: + case ETHTOOL_FEC_NONE: + fec = ICE_FEC_NONE; + break; + default: + dev_warn(&vsi->back->pdev->dev, "Unsupported FEC mode: %d\n", + fecparam->fec); + return -EINVAL; + } + + return ice_set_fec_cfg(netdev, fec); +} + +/** + * ice_get_fecparam - Get link FEC options + * @netdev: network interface device structure + * @fecparam: Ethtool structure to retrieve FEC parameters + */ +static int +ice_get_fecparam(struct net_device *netdev, struct ethtool_fecparam *fecparam) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_aqc_get_phy_caps_data *caps; + struct ice_link_status *link_info; + struct ice_vsi *vsi = np->vsi; + struct ice_port_info *pi; + enum ice_status status; + int err = 0; + + pi = vsi->port_info; + + if (!pi) + return -EOPNOTSUPP; + link_info = &pi->phy.link_info; + + /* Set FEC mode based on negotiated link info */ + switch (link_info->fec_info) { + case ICE_AQ_LINK_25G_KR_FEC_EN: + fecparam->active_fec = ETHTOOL_FEC_BASER; + break; + case ICE_AQ_LINK_25G_RS_528_FEC_EN: + /* fall through */ + case ICE_AQ_LINK_25G_RS_544_FEC_EN: + fecparam->active_fec = ETHTOOL_FEC_RS; + break; + default: + fecparam->active_fec = ETHTOOL_FEC_OFF; + break; + } + + caps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*caps), GFP_KERNEL); + if (!caps) + return -ENOMEM; + + status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP, + caps, NULL); + if (status) { + err = -EAGAIN; + goto done; + } + + /* Set supported/configured FEC modes based on PHY capability */ + if (caps->caps & ICE_AQC_PHY_EN_AUTO_FEC) + fecparam->fec |= ETHTOOL_FEC_AUTO; + if (caps->link_fec_options & ICE_AQC_PHY_FEC_10G_KR_40G_KR4_EN || + caps->link_fec_options & ICE_AQC_PHY_FEC_10G_KR_40G_KR4_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_KR_CLAUSE74_EN || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_KR_REQ) + fecparam->fec |= ETHTOOL_FEC_BASER; + if (caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_528_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_544_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_CLAUSE91_EN) + fecparam->fec |= ETHTOOL_FEC_RS; + if (caps->link_fec_options == 0) + fecparam->fec |= ETHTOOL_FEC_OFF; + +done: + devm_kfree(&vsi->back->pdev->dev, caps); + return err; +} + +/** * ice_get_priv_flags - report device private flags * @netdev: network interface device structure * @@ -382,13 +1175,19 @@ static u32 ice_get_priv_flags(struct net_device *netdev) static int ice_set_priv_flags(struct net_device *netdev, u32 flags) { struct ice_netdev_priv *np = netdev_priv(netdev); + DECLARE_BITMAP(change_flags, ICE_PF_FLAGS_NBITS); + DECLARE_BITMAP(orig_flags, ICE_PF_FLAGS_NBITS); struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; + int ret = 0; u32 i; if (flags > BIT(ICE_PRIV_FLAG_ARRAY_SIZE)) return -EINVAL; + set_bit(ICE_FLAG_ETHTOOL_CTXT, pf->flags); + + bitmap_copy(orig_flags, pf->flags, ICE_PF_FLAGS_NBITS); for (i = 0; i < ICE_PRIV_FLAG_ARRAY_SIZE; i++) { const struct ice_priv_flag *priv_flag; @@ -400,7 +1199,84 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags) clear_bit(priv_flag->bitno, pf->flags); } - return 0; + bitmap_xor(change_flags, pf->flags, orig_flags, ICE_PF_FLAGS_NBITS); + + if (test_bit(ICE_FLAG_ENABLE_FW_LLDP, change_flags)) { + if (!test_bit(ICE_FLAG_ENABLE_FW_LLDP, pf->flags)) { + enum ice_status status; + + /* Disable FW LLDP engine */ + status = ice_aq_cfg_lldp_mib_change(&pf->hw, false, + NULL); + /* If unregistering for LLDP events fails, this is + * not an error state, as there shouldn't be any + * events to respond to. + */ + if (status) + dev_info(&pf->pdev->dev, + "Failed to unreg for LLDP events\n"); + + /* The AQ call to stop the FW LLDP agent will generate + * an error if the agent is already stopped. + */ + status = ice_aq_stop_lldp(&pf->hw, true, true, NULL); + if (status) + dev_warn(&pf->pdev->dev, + "Fail to stop LLDP agent\n"); + /* Use case for having the FW LLDP agent stopped + * will likely not need DCB, so failure to init is + * not a concern of ethtool + */ + status = ice_init_pf_dcb(pf, true); + if (status) + dev_warn(&pf->pdev->dev, "Fail to init DCB\n"); + + /* Forward LLDP packets to default VSI so that they + * are passed up the stack + */ + ice_cfg_sw_lldp(vsi, false, true); + } else { + enum ice_status status; + bool dcbx_agent_status; + + /* AQ command to start FW LLDP agent will return an + * error if the agent is already started + */ + status = ice_aq_start_lldp(&pf->hw, true, NULL); + if (status) + dev_warn(&pf->pdev->dev, + "Fail to start LLDP Agent\n"); + + /* AQ command to start FW DCBX agent will fail if + * the agent is already started + */ + status = ice_aq_start_stop_dcbx(&pf->hw, true, + &dcbx_agent_status, + NULL); + if (status) + dev_dbg(&pf->pdev->dev, + "Failed to start FW DCBX\n"); + + dev_info(&pf->pdev->dev, "FW DCBX agent is %s\n", + dcbx_agent_status ? "ACTIVE" : "DISABLED"); + + /* Failure to configure MIB change or init DCB is not + * relevant to ethtool. Print notification that + * registration/init failed but do not return error + * state to ethtool + */ + status = ice_init_pf_dcb(pf, true); + if (status) + dev_dbg(&pf->pdev->dev, "Fail to init DCB\n"); + + /* Remove rule to direct LLDP packets to default VSI. + * The FW LLDP engine will now be consuming them. + */ + ice_cfg_sw_lldp(vsi, false, false); + } + } + clear_bit(ICE_FLAG_ETHTOOL_CTXT, pf->flags); + return ret; } static int ice_get_sset_count(struct net_device *netdev, int sset) @@ -426,6 +1302,8 @@ static int ice_get_sset_count(struct net_device *netdev, int sset) * not safe. */ return ICE_ALL_STATS_LEN(netdev); + case ETH_SS_TEST: + return ICE_TEST_LEN; case ETH_SS_PRIV_FLAGS: return ICE_PRIV_FLAG_ARRAY_SIZE; default: @@ -486,6 +1364,16 @@ ice_get_ethtool_stats(struct net_device *netdev, data[i++] = (ice_gstrings_pf_stats[j].sizeof_stat == sizeof(u64)) ? *(u64 *)p : *(u32 *)p; } + + for (j = 0; j < ICE_MAX_USER_PRIORITY; j++) { + data[i++] = pf->stats.priority_xon_tx[j]; + data[i++] = pf->stats.priority_xoff_tx[j]; + } + + for (j = 0; j < ICE_MAX_USER_PRIORITY; j++) { + data[i++] = pf->stats.priority_xon_rx[j]; + data[i++] = pf->stats.priority_xoff_rx[j]; + } } /** @@ -515,7 +1403,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_100M_SGMII) { ethtool_link_ksettings_add_link_mode(ks, supported, 100baseT_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100MB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100MB) ethtool_link_ksettings_add_link_mode(ks, advertising, 100baseT_Full); } @@ -523,14 +1412,16 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_1G_SGMII) { ethtool_link_ksettings_add_link_mode(ks, supported, 1000baseT_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_1000MB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_1000MB) ethtool_link_ksettings_add_link_mode(ks, advertising, 1000baseT_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_1000BASE_KX) { ethtool_link_ksettings_add_link_mode(ks, supported, 1000baseKX_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_1000MB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_1000MB) ethtool_link_ksettings_add_link_mode(ks, advertising, 1000baseKX_Full); } @@ -538,14 +1429,16 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_1000BASE_LX) { ethtool_link_ksettings_add_link_mode(ks, supported, 1000baseX_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_1000MB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_1000MB) ethtool_link_ksettings_add_link_mode(ks, advertising, 1000baseX_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_2500BASE_T) { ethtool_link_ksettings_add_link_mode(ks, supported, 2500baseT_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_2500MB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_2500MB) ethtool_link_ksettings_add_link_mode(ks, advertising, 2500baseT_Full); } @@ -553,7 +1446,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_2500BASE_KX) { ethtool_link_ksettings_add_link_mode(ks, supported, 2500baseX_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_2500MB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_2500MB) ethtool_link_ksettings_add_link_mode(ks, advertising, 2500baseX_Full); } @@ -561,7 +1455,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_5GBASE_KR) { ethtool_link_ksettings_add_link_mode(ks, supported, 5000baseT_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_5GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_5GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 5000baseT_Full); } @@ -571,28 +1466,32 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_10G_SFI_C2C) { ethtool_link_ksettings_add_link_mode(ks, supported, 10000baseT_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 10000baseT_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_10GBASE_KR_CR1) { ethtool_link_ksettings_add_link_mode(ks, supported, 10000baseKR_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 10000baseKR_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_10GBASE_SR) { ethtool_link_ksettings_add_link_mode(ks, supported, 10000baseSR_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 10000baseSR_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_10GBASE_LR) { ethtool_link_ksettings_add_link_mode(ks, supported, 10000baseLR_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_10GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 10000baseLR_Full); } @@ -604,7 +1503,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_25G_AUI_C2C) { ethtool_link_ksettings_add_link_mode(ks, supported, 25000baseCR_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_25GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_25GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 25000baseCR_Full); } @@ -612,7 +1512,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_25GBASE_LR) { ethtool_link_ksettings_add_link_mode(ks, supported, 25000baseSR_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_25GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_25GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 25000baseSR_Full); } @@ -621,14 +1522,16 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_25GBASE_KR1) { ethtool_link_ksettings_add_link_mode(ks, supported, 25000baseKR_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_25GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_25GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 25000baseKR_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_40GBASE_KR4) { ethtool_link_ksettings_add_link_mode(ks, supported, 40000baseKR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 40000baseKR4_Full); } @@ -637,21 +1540,24 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_40G_XLAUI) { ethtool_link_ksettings_add_link_mode(ks, supported, 40000baseCR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 40000baseCR4_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_40GBASE_SR4) { ethtool_link_ksettings_add_link_mode(ks, supported, 40000baseSR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 40000baseSR4_Full); } if (phy_types_low & ICE_PHY_TYPE_LOW_40GBASE_LR4) { ethtool_link_ksettings_add_link_mode(ks, supported, 40000baseLR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_40GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 40000baseLR4_Full); } @@ -666,7 +1572,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_50G_AUI1) { ethtool_link_ksettings_add_link_mode(ks, supported, 50000baseCR2_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_50GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_50GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 50000baseCR2_Full); } @@ -674,7 +1581,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_50GBASE_KR_PAM4) { ethtool_link_ksettings_add_link_mode(ks, supported, 50000baseKR2_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_50GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_50GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 50000baseKR2_Full); } @@ -684,7 +1592,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_50GBASE_LR) { ethtool_link_ksettings_add_link_mode(ks, supported, 50000baseSR2_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_50GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_50GB) ethtool_link_ksettings_add_link_mode(ks, advertising, 50000baseSR2_Full); } @@ -701,7 +1610,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_high & ICE_PHY_TYPE_HIGH_100G_AUI2) { ethtool_link_ksettings_add_link_mode(ks, supported, 100000baseCR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) need_add_adv_mode = true; } if (need_add_adv_mode) { @@ -713,7 +1623,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_100GBASE_SR2) { ethtool_link_ksettings_add_link_mode(ks, supported, 100000baseSR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) need_add_adv_mode = true; } if (need_add_adv_mode) { @@ -725,7 +1636,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_low & ICE_PHY_TYPE_LOW_100GBASE_DR) { ethtool_link_ksettings_add_link_mode(ks, supported, 100000baseLR4_ER4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) need_add_adv_mode = true; } if (need_add_adv_mode) { @@ -738,7 +1650,8 @@ ice_phy_type_to_ethtool(struct net_device *netdev, phy_types_high & ICE_PHY_TYPE_HIGH_100GBASE_KR2_PAM4) { ethtool_link_ksettings_add_link_mode(ks, supported, 100000baseKR4_Full); - if (hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) + if (!hw_link_info->req_speeds || + hw_link_info->req_speeds & ICE_AQ_LINK_SPEED_100GB) need_add_adv_mode = true; } if (need_add_adv_mode) @@ -811,7 +1724,7 @@ ice_get_settings_link_up(struct ethtool_link_ksettings *ks, link_info = &vsi->port_info->phy.link_info; - /* Initialize supported and advertised settings based on phy settings */ + /* Initialize supported and advertised settings based on PHY settings */ switch (link_info->phy_type_low) { case ICE_PHY_TYPE_LOW_100BASE_TX: ethtool_link_ksettings_add_link_mode(ks, supported, Autoneg); @@ -921,6 +1834,7 @@ ice_get_settings_link_up(struct ethtool_link_ksettings *ks, 25000baseCR_Full); break; case ICE_PHY_TYPE_LOW_25G_AUI_AOC_ACC: + case ICE_PHY_TYPE_LOW_25G_AUI_C2C: ethtool_link_ksettings_add_link_mode(ks, supported, 25000baseCR_Full); break; @@ -1137,10 +2051,10 @@ ice_get_settings_link_up(struct ethtool_link_ksettings *ks, */ static void ice_get_settings_link_down(struct ethtool_link_ksettings *ks, - struct net_device __always_unused *netdev) + struct net_device *netdev) { /* link is down and the driver needs to fall back on - * supported phy types to figure out what info to display + * supported PHY types to figure out what info to display */ ice_phy_type_to_ethtool(netdev, ks); @@ -1156,10 +2070,12 @@ ice_get_settings_link_down(struct ethtool_link_ksettings *ks, * * Reports speed/duplex settings based on media_type */ -static int ice_get_link_ksettings(struct net_device *netdev, - struct ethtool_link_ksettings *ks) +static int +ice_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *ks) { struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_aqc_get_phy_caps_data *caps; struct ice_link_status *hw_link_info; struct ice_vsi *vsi = np->vsi; @@ -1230,6 +2146,40 @@ static int ice_get_link_ksettings(struct net_device *netdev, break; } + caps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*caps), GFP_KERNEL); + if (!caps) + goto done; + + if (ice_aq_get_phy_caps(vsi->port_info, false, ICE_AQC_REPORT_TOPO_CAP, + caps, NULL)) + netdev_info(netdev, "Get phy capability failed.\n"); + + /* Set supported FEC modes based on PHY capability */ + ethtool_link_ksettings_add_link_mode(ks, supported, FEC_NONE); + + if (caps->link_fec_options & ICE_AQC_PHY_FEC_10G_KR_40G_KR4_EN || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_KR_CLAUSE74_EN) + ethtool_link_ksettings_add_link_mode(ks, supported, FEC_BASER); + if (caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_CLAUSE91_EN) + ethtool_link_ksettings_add_link_mode(ks, supported, FEC_RS); + + if (ice_aq_get_phy_caps(vsi->port_info, false, ICE_AQC_REPORT_SW_CFG, + caps, NULL)) + netdev_info(netdev, "Get phy capability failed.\n"); + + /* Set advertised FEC modes based on PHY capability */ + ethtool_link_ksettings_add_link_mode(ks, advertising, FEC_NONE); + + if (caps->link_fec_options & ICE_AQC_PHY_FEC_10G_KR_40G_KR4_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_KR_REQ) + ethtool_link_ksettings_add_link_mode(ks, advertising, + FEC_BASER); + if (caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_528_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_544_REQ) + ethtool_link_ksettings_add_link_mode(ks, advertising, FEC_RS); + +done: + devm_kfree(&vsi->back->pdev->dev, caps); return 0; } @@ -1349,7 +2299,7 @@ ice_setup_autoneg(struct ice_port_info *p, struct ethtool_link_ksettings *ks, } else { /* If autoneg is currently enabled */ if (p->phy.link_info.an_info & ICE_AQ_AN_COMPLETED) { - /* If autoneg is supported 10GBASE_T is the only phy + /* If autoneg is supported 10GBASE_T is the only PHY * that can disable it, so otherwise return error */ if (ethtool_link_ksettings_test_link_mode(ks, @@ -1399,14 +2349,13 @@ ice_set_link_ksettings(struct net_device *netdev, if (!p) return -EOPNOTSUPP; - /* Check if this is lan vsi */ - for (idx = 0 ; idx < pf->num_alloc_vsi ; idx++) { + /* Check if this is LAN VSI */ + ice_for_each_vsi(pf, idx) if (pf->vsi[idx]->type == ICE_VSI_PF) { if (np->vsi != pf->vsi[idx]) return -EOPNOTSUPP; break; } - } if (p->phy.media_type != ICE_MEDIA_BASET && p->phy.media_type != ICE_MEDIA_FIBER && @@ -1464,7 +2413,7 @@ ice_set_link_ksettings(struct net_device *netdev, if (!abilities) return -ENOMEM; - /* Get the current phy config */ + /* Get the current PHY config */ status = ice_aq_get_phy_caps(p, false, ICE_AQC_REPORT_SW_CFG, abilities, NULL); if (status) { @@ -1559,15 +2508,16 @@ done: } /** - * ice_get_rxnfc - command to get RX flow classification rules + * ice_get_rxnfc - command to get Rx flow classification rules * @netdev: network interface device structure * @cmd: ethtool rxnfc command * @rule_locs: buffer to rturn Rx flow classification rules * * Returns Success if the command is supported. */ -static int ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, - u32 __always_unused *rule_locs) +static int +ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, + u32 __always_unused *rule_locs) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; @@ -1821,18 +2771,21 @@ ice_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) struct ice_port_info *pi = np->vsi->port_info; struct ice_aqc_get_phy_caps_data *pcaps; struct ice_vsi *vsi = np->vsi; + struct ice_dcbx_cfg *dcbx_cfg; enum ice_status status; /* Initialize pause params */ pause->rx_pause = 0; pause->tx_pause = 0; + dcbx_cfg = &pi->local_dcbx_cfg; + pcaps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*pcaps), GFP_KERNEL); if (!pcaps) return; - /* Get current phy config */ + /* Get current PHY config */ status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps, NULL); if (status) @@ -1841,6 +2794,10 @@ ice_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) pause->autoneg = ((pcaps->caps & ICE_AQC_PHY_AN_MODE) ? AUTONEG_ENABLE : AUTONEG_DISABLE); + if (dcbx_cfg->pfc.pfcena) + /* PFC enabled so report LFC as off */ + goto out; + if (pcaps->caps & ICE_AQC_PHY_EN_TX_LINK_PAUSE) pause->tx_pause = 1; if (pcaps->caps & ICE_AQC_PHY_EN_RX_LINK_PAUSE) @@ -1861,6 +2818,7 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_link_status *hw_link_info; struct ice_pf *pf = np->vsi->back; + struct ice_dcbx_cfg *dcbx_cfg; struct ice_vsi *vsi = np->vsi; struct ice_hw *hw = &pf->hw; struct ice_port_info *pi; @@ -1871,6 +2829,7 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) pi = vsi->port_info; hw_link_info = &pi->phy.link_info; + dcbx_cfg = &pi->local_dcbx_cfg; link_up = hw_link_info->link_info & ICE_AQ_LINK_UP; /* Changing the port's flow control is not supported if this isn't the @@ -1893,6 +2852,10 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) netdev_info(netdev, "Autoneg did not complete so changing settings may not result in an actual change.\n"); } + if (dcbx_cfg->pfc.pfcena) { + netdev_info(netdev, "Priority flow control enabled. Cannot set link flow control.\n"); + return -EOPNOTSUPP; + } if (pause->rx_pause && pause->tx_pause) pi->fc.req_mode = ICE_FC_FULL; else if (pause->rx_pause && !pause->tx_pause) @@ -2021,11 +2984,12 @@ out: * @key: hash key * @hfunc: hash function * - * Returns -EINVAL if the table specifies an invalid queue id, otherwise + * Returns -EINVAL if the table specifies an invalid queue ID, otherwise * returns 0 after programming the table. */ -static int ice_set_rxfh(struct net_device *netdev, const u32 *indir, - const u8 *key, const u8 hfunc) +static int +ice_set_rxfh(struct net_device *netdev, const u32 *indir, const u8 *key, + const u8 hfunc) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; @@ -2087,7 +3051,7 @@ enum ice_container_type { /** * ice_get_rc_coalesce - get ITR values for specific ring container * @ec: ethtool structure to fill with driver's coalesce settings - * @c_type: container type, RX or TX + * @c_type: container type, Rx or Tx * @rc: ring container that the ITR values will come from * * Query the device for ice_ring_container specific ITR values. This is @@ -2100,12 +3064,18 @@ static int ice_get_rc_coalesce(struct ethtool_coalesce *ec, enum ice_container_type c_type, struct ice_ring_container *rc) { - struct ice_pf *pf = rc->ring->vsi->back; + struct ice_pf *pf; + + if (!rc->ring) + return -EINVAL; + + pf = rc->ring->vsi->back; switch (c_type) { case ICE_RX_CONTAINER: ec->use_adaptive_rx_coalesce = ITR_IS_DYNAMIC(rc->itr_setting); ec->rx_coalesce_usecs = rc->itr_setting & ~ICE_ITR_DYNAMIC; + ec->rx_coalesce_usecs_high = rc->ring->q_vector->intrl; break; case ICE_TX_CONTAINER: ec->use_adaptive_tx_coalesce = ITR_IS_DYNAMIC(rc->itr_setting); @@ -2120,49 +3090,60 @@ ice_get_rc_coalesce(struct ethtool_coalesce *ec, enum ice_container_type c_type, } /** + * ice_get_q_coalesce - get a queue's ITR/INTRL (coalesce) settings + * @vsi: VSI associated to the queue for getting ITR/INTRL (coalesce) settings + * @ec: coalesce settings to program the device with + * @q_num: update ITR/INTRL (coalesce) settings for this queue number/index + * + * Return 0 on success, and negative under the following conditions: + * 1. Getting Tx or Rx ITR/INTRL (coalesce) settings failed. + * 2. The q_num passed in is not a valid number/index for Tx and Rx rings. + */ +static int +ice_get_q_coalesce(struct ice_vsi *vsi, struct ethtool_coalesce *ec, int q_num) +{ + if (q_num < vsi->num_rxq && q_num < vsi->num_txq) { + if (ice_get_rc_coalesce(ec, ICE_RX_CONTAINER, + &vsi->rx_rings[q_num]->q_vector->rx)) + return -EINVAL; + if (ice_get_rc_coalesce(ec, ICE_TX_CONTAINER, + &vsi->tx_rings[q_num]->q_vector->tx)) + return -EINVAL; + } else if (q_num < vsi->num_rxq) { + if (ice_get_rc_coalesce(ec, ICE_RX_CONTAINER, + &vsi->rx_rings[q_num]->q_vector->rx)) + return -EINVAL; + } else if (q_num < vsi->num_txq) { + if (ice_get_rc_coalesce(ec, ICE_TX_CONTAINER, + &vsi->tx_rings[q_num]->q_vector->tx)) + return -EINVAL; + } else { + return -EINVAL; + } + + return 0; +} + +/** * __ice_get_coalesce - get ITR/INTRL values for the device * @netdev: pointer to the netdev associated with this query * @ec: ethtool structure to fill with driver's coalesce settings * @q_num: queue number to get the coalesce settings for + * + * If the caller passes in a negative q_num then we return coalesce settings + * based on queue number 0, else use the actual q_num passed in. */ static int __ice_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, int q_num) { struct ice_netdev_priv *np = netdev_priv(netdev); - int tx = -EINVAL, rx = -EINVAL; struct ice_vsi *vsi = np->vsi; - if (q_num < 0) { - rx = ice_get_rc_coalesce(ec, ICE_RX_CONTAINER, - &vsi->rx_rings[0]->q_vector->rx); - tx = ice_get_rc_coalesce(ec, ICE_TX_CONTAINER, - &vsi->tx_rings[0]->q_vector->tx); - - goto update_coalesced_frames; - } + if (q_num < 0) + q_num = 0; - if (q_num < vsi->num_rxq && q_num < vsi->num_txq) { - rx = ice_get_rc_coalesce(ec, ICE_RX_CONTAINER, - &vsi->rx_rings[q_num]->q_vector->rx); - tx = ice_get_rc_coalesce(ec, ICE_TX_CONTAINER, - &vsi->tx_rings[q_num]->q_vector->tx); - } else if (q_num < vsi->num_rxq) { - rx = ice_get_rc_coalesce(ec, ICE_RX_CONTAINER, - &vsi->rx_rings[q_num]->q_vector->rx); - } else if (q_num < vsi->num_txq) { - tx = ice_get_rc_coalesce(ec, ICE_TX_CONTAINER, - &vsi->tx_rings[q_num]->q_vector->tx); - } else { - /* q_num is invalid for both Rx and Tx queues */ - return -EINVAL; - } - -update_coalesced_frames: - /* either q_num is invalid for both Rx and Tx queues or setting coalesce - * failed completely - */ - if (tx && rx) + if (ice_get_q_coalesce(vsi, ec, q_num)) return -EINVAL; if (q_num < vsi->num_txq) @@ -2180,15 +3161,16 @@ ice_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec) return __ice_get_coalesce(netdev, ec, -1); } -static int ice_get_per_q_coalesce(struct net_device *netdev, u32 q_num, - struct ethtool_coalesce *ec) +static int +ice_get_per_q_coalesce(struct net_device *netdev, u32 q_num, + struct ethtool_coalesce *ec) { return __ice_get_coalesce(netdev, ec, q_num); } /** * ice_set_rc_coalesce - set ITR values for specific ring container - * @c_type: container type, RX or TX + * @c_type: container type, Rx or Tx * @ec: ethtool structure from user to update ITR settings * @rc: ring container that the ITR values will come from * @vsi: VSI associated to the ring container @@ -2213,6 +3195,22 @@ ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, switch (c_type) { case ICE_RX_CONTAINER: + if (ec->rx_coalesce_usecs_high > ICE_MAX_INTRL || + (ec->rx_coalesce_usecs_high && + ec->rx_coalesce_usecs_high < pf->hw.intrl_gran)) { + netdev_info(vsi->netdev, + "Invalid value, rx-usecs-high valid values are 0 (disabled), %d-%d\n", + pf->hw.intrl_gran, ICE_MAX_INTRL); + return -EINVAL; + } + + if (ec->rx_coalesce_usecs_high != rc->ring->q_vector->intrl) { + rc->ring->q_vector->intrl = ec->rx_coalesce_usecs_high; + wr32(&pf->hw, GLINT_RATE(rc->ring->q_vector->reg_idx), + ice_intrl_usec_to_reg(ec->rx_coalesce_usecs_high, + pf->hw.intrl_gran)); + } + if (ec->rx_coalesce_usecs != itr_setting && ec->use_adaptive_rx_coalesce) { netdev_info(vsi->netdev, @@ -2235,6 +3233,12 @@ ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, } break; case ICE_TX_CONTAINER: + if (ec->tx_coalesce_usecs_high) { + netdev_info(vsi->netdev, + "setting tx-usecs-high is not supported\n"); + return -EINVAL; + } + if (ec->tx_coalesce_usecs != itr_setting && ec->use_adaptive_tx_coalesce) { netdev_info(vsi->netdev, @@ -2264,54 +3268,77 @@ ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, return 0; } +/** + * ice_set_q_coalesce - set a queue's ITR/INTRL (coalesce) settings + * @vsi: VSI associated to the queue that need updating + * @ec: coalesce settings to program the device with + * @q_num: update ITR/INTRL (coalesce) settings for this queue number/index + * + * Return 0 on success, and negative under the following conditions: + * 1. Setting Tx or Rx ITR/INTRL (coalesce) settings failed. + * 2. The q_num passed in is not a valid number/index for Tx and Rx rings. + */ +static int +ice_set_q_coalesce(struct ice_vsi *vsi, struct ethtool_coalesce *ec, int q_num) +{ + if (q_num < vsi->num_rxq && q_num < vsi->num_txq) { + if (ice_set_rc_coalesce(ICE_RX_CONTAINER, ec, + &vsi->rx_rings[q_num]->q_vector->rx, + vsi)) + return -EINVAL; + + if (ice_set_rc_coalesce(ICE_TX_CONTAINER, ec, + &vsi->tx_rings[q_num]->q_vector->tx, + vsi)) + return -EINVAL; + } else if (q_num < vsi->num_rxq) { + if (ice_set_rc_coalesce(ICE_RX_CONTAINER, ec, + &vsi->rx_rings[q_num]->q_vector->rx, + vsi)) + return -EINVAL; + } else if (q_num < vsi->num_txq) { + if (ice_set_rc_coalesce(ICE_TX_CONTAINER, ec, + &vsi->tx_rings[q_num]->q_vector->tx, + vsi)) + return -EINVAL; + } else { + return -EINVAL; + } + + return 0; +} + +/** + * __ice_set_coalesce - set ITR/INTRL values for the device + * @netdev: pointer to the netdev associated with this query + * @ec: ethtool structure to fill with driver's coalesce settings + * @q_num: queue number to get the coalesce settings for + * + * If the caller passes in a negative q_num then we set the coalesce settings + * for all Tx/Rx queues, else use the actual q_num passed in. + */ static int __ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, int q_num) { struct ice_netdev_priv *np = netdev_priv(netdev); - int rx = -EINVAL, tx = -EINVAL; struct ice_vsi *vsi = np->vsi; if (q_num < 0) { int i; ice_for_each_q_vector(vsi, i) { - struct ice_q_vector *q_vector = vsi->q_vectors[i]; - - if (ice_set_rc_coalesce(ICE_RX_CONTAINER, ec, - &q_vector->rx, vsi) || - ice_set_rc_coalesce(ICE_TX_CONTAINER, ec, - &q_vector->tx, vsi)) + if (ice_set_q_coalesce(vsi, ec, i)) return -EINVAL; } - goto set_work_lmt; } - if (q_num < vsi->num_rxq && q_num < vsi->num_txq) { - rx = ice_set_rc_coalesce(ICE_RX_CONTAINER, ec, - &vsi->rx_rings[q_num]->q_vector->rx, - vsi); - tx = ice_set_rc_coalesce(ICE_TX_CONTAINER, ec, - &vsi->tx_rings[q_num]->q_vector->tx, - vsi); - } else if (q_num < vsi->num_rxq) { - rx = ice_set_rc_coalesce(ICE_RX_CONTAINER, ec, - &vsi->rx_rings[q_num]->q_vector->rx, - vsi); - } else if (q_num < vsi->num_txq) { - tx = ice_set_rc_coalesce(ICE_TX_CONTAINER, ec, - &vsi->tx_rings[q_num]->q_vector->tx, - vsi); - } - - /* either q_num is invalid for both Rx and Tx queues or setting coalesce - * failed completely - */ - if (rx && tx) + if (ice_set_q_coalesce(vsi, ec, q_num)) return -EINVAL; set_work_lmt: + if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq) vsi->work_lmt = max(ec->tx_max_coalesced_frames_irq, ec->rx_max_coalesced_frames_irq); @@ -2325,8 +3352,9 @@ ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec) return __ice_set_coalesce(netdev, ec, -1); } -static int ice_set_per_q_coalesce(struct net_device *netdev, u32 q_num, - struct ethtool_coalesce *ec) +static int +ice_set_per_q_coalesce(struct net_device *netdev, u32 q_num, + struct ethtool_coalesce *ec) { return __ice_set_coalesce(netdev, ec, q_num); } @@ -2339,6 +3367,7 @@ static const struct ethtool_ops ice_ethtool_ops = { .get_regs = ice_get_regs, .get_msglevel = ice_get_msglevel, .set_msglevel = ice_set_msglevel, + .self_test = ice_self_test, .get_link = ethtool_op_get_link, .get_eeprom_len = ice_get_eeprom_len, .get_eeprom = ice_get_eeprom, @@ -2363,6 +3392,8 @@ static const struct ethtool_ops ice_ethtool_ops = { .get_ts_info = ethtool_op_get_ts_info, .get_per_queue_coalesce = ice_get_per_q_coalesce, .set_per_queue_coalesce = ice_set_per_q_coalesce, + .get_fecparam = ice_get_fecparam, + .set_fecparam = ice_set_fecparam, }; /** diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h index 6bf5cc064270..6c5ce05742b1 100644 --- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h +++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h @@ -6,6 +6,9 @@ #ifndef _ICE_HW_AUTOGEN_H_ #define _ICE_HW_AUTOGEN_H_ +#define PF0INT_ITR_0(_i) (0x03000004 + ((_i) * 4096)) +#define PF0INT_ITR_1(_i) (0x03000008 + ((_i) * 4096)) +#define PF0INT_ITR_2(_i) (0x0300000C + ((_i) * 4096)) #define QTX_COMM_DBELL(_DBQM) (0x002C0000 + ((_DBQM) * 4)) #define QTX_COMM_HEAD(_DBQM) (0x000E0000 + ((_DBQM) * 4)) #define QTX_COMM_HEAD_HEAD_S 0 @@ -49,6 +52,9 @@ #define PF_MBX_ATQLEN_ATQLEN_M ICE_M(0x3FF, 0) #define PF_MBX_ATQLEN_ATQENABLE_M BIT(31) #define PF_MBX_ATQT 0x0022E300 +#define PRTDCB_GENS 0x00083020 +#define PRTDCB_GENS_DCBX_STATUS_S 0 +#define PRTDCB_GENS_DCBX_STATUS_M ICE_M(0x7, 0) #define GLFLXP_RXDID_FLAGS(_i, _j) (0x0045D000 + ((_i) * 4 + (_j) * 256)) #define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_S 0 #define GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_M ICE_M(0x3F, 0) @@ -106,6 +112,16 @@ #define VPGEN_VFRTRIG_VFSWR_M BIT(0) #define PFHMC_ERRORDATA 0x00520500 #define PFHMC_ERRORINFO 0x00520400 +#define GLINT_CTL 0x0016CC54 +#define GLINT_CTL_DIS_AUTOMASK_M BIT(0) +#define GLINT_CTL_ITR_GRAN_200_S 16 +#define GLINT_CTL_ITR_GRAN_200_M ICE_M(0xF, 16) +#define GLINT_CTL_ITR_GRAN_100_S 20 +#define GLINT_CTL_ITR_GRAN_100_M ICE_M(0xF, 20) +#define GLINT_CTL_ITR_GRAN_50_S 24 +#define GLINT_CTL_ITR_GRAN_50_M ICE_M(0xF, 24) +#define GLINT_CTL_ITR_GRAN_25_S 28 +#define GLINT_CTL_ITR_GRAN_25_M ICE_M(0xF, 28) #define GLINT_DYN_CTL(_INT) (0x00160000 + ((_INT) * 4)) #define GLINT_DYN_CTL_INTENA_M BIT(0) #define GLINT_DYN_CTL_CLEARPBA_M BIT(1) @@ -142,6 +158,7 @@ #define PFINT_OICR_HMC_ERR_M BIT(26) #define PFINT_OICR_PE_CRITERR_M BIT(28) #define PFINT_OICR_VFLR_M BIT(29) +#define PFINT_OICR_SWINT_M BIT(31) #define PFINT_OICR_CTL 0x0016CA80 #define PFINT_OICR_CTL_MSIX_INDX_M ICE_M(0x7FF, 0) #define PFINT_OICR_CTL_ITR_INDX_S 11 @@ -150,11 +167,15 @@ #define PFINT_OICR_ENA 0x0016C900 #define QINT_RQCTL(_QRX) (0x00150000 + ((_QRX) * 4)) #define QINT_RQCTL_MSIX_INDX_S 0 +#define QINT_RQCTL_MSIX_INDX_M ICE_M(0x7FF, 0) #define QINT_RQCTL_ITR_INDX_S 11 +#define QINT_RQCTL_ITR_INDX_M ICE_M(0x3, 11) #define QINT_RQCTL_CAUSE_ENA_M BIT(30) #define QINT_TQCTL(_DBQM) (0x00140000 + ((_DBQM) * 4)) #define QINT_TQCTL_MSIX_INDX_S 0 +#define QINT_TQCTL_MSIX_INDX_M ICE_M(0x7FF, 0) #define QINT_TQCTL_ITR_INDX_S 11 +#define QINT_TQCTL_ITR_INDX_M ICE_M(0x3, 11) #define QINT_TQCTL_CAUSE_ENA_M BIT(30) #define VPINT_ALLOC(_VF) (0x001D1000 + ((_VF) * 4)) #define VPINT_ALLOC_FIRST_S 0 @@ -168,6 +189,8 @@ #define VPINT_ALLOC_PCI_LAST_S 12 #define VPINT_ALLOC_PCI_LAST_M ICE_M(0x7FF, 12) #define VPINT_ALLOC_PCI_VALID_M BIT(31) +#define VPINT_MBX_CTL(_VSI) (0x0016A000 + ((_VSI) * 4)) +#define VPINT_MBX_CTL_CAUSE_ENA_M BIT(30) #define GLLAN_RCTL_0 0x002941F8 #define QRX_CONTEXT(_i, _QRX) (0x00280000 + ((_i) * 8192 + (_QRX) * 4)) #define QRX_CTRL(_QRX) (0x00120000 + ((_QRX) * 4)) @@ -306,11 +329,16 @@ #define GLPRT_PTC64L(_i) (0x00380B80 + ((_i) * 8)) #define GLPRT_PTC9522H(_i) (0x00380D04 + ((_i) * 8)) #define GLPRT_PTC9522L(_i) (0x00380D00 + ((_i) * 8)) +#define GLPRT_PXOFFRXC(_i, _j) (0x00380500 + ((_i) * 8 + (_j) * 64)) +#define GLPRT_PXOFFTXC(_i, _j) (0x00380F40 + ((_i) * 8 + (_j) * 64)) +#define GLPRT_PXONRXC(_i, _j) (0x00380300 + ((_i) * 8 + (_j) * 64)) +#define GLPRT_PXONTXC(_i, _j) (0x00380D40 + ((_i) * 8 + (_j) * 64)) #define GLPRT_RFC(_i) (0x00380AC0 + ((_i) * 8)) #define GLPRT_RJC(_i) (0x00380B00 + ((_i) * 8)) #define GLPRT_RLEC(_i) (0x00380140 + ((_i) * 8)) #define GLPRT_ROC(_i) (0x00380240 + ((_i) * 8)) #define GLPRT_RUC(_i) (0x00380200 + ((_i) * 8)) +#define GLPRT_RXON2OFFCNT(_i, _j) (0x00380700 + ((_i) * 8 + (_j) * 64)) #define GLPRT_TDOLD(_i) (0x00381280 + ((_i) * 8)) #define GLPRT_UPRCH(_i) (0x00381304 + ((_i) * 8)) #define GLPRT_UPRCL(_i) (0x00381300 + ((_i) * 8)) diff --git a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h index ef4c79b5aa32..510a8c900e61 100644 --- a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h +++ b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h @@ -20,7 +20,7 @@ union ice_32byte_rx_desc { } lo_dword; union { __le32 rss; /* RSS Hash */ - __le32 fd_id; /* Flow Director filter id */ + __le32 fd_id; /* Flow Director filter ID */ } hi_dword; } qword0; struct { @@ -99,7 +99,7 @@ enum ice_rx_ptype_payload_layer { ICE_RX_PTYPE_PAYLOAD_LAYER_PAY4 = 3, }; -/* RX Flex Descriptor +/* Rx Flex Descriptor * This descriptor is used instead of the legacy version descriptor when * ice_rlan_ctx.adv_desc is set */ @@ -113,7 +113,7 @@ union ice_32b_rx_flex_desc { } read; struct { /* Qword 0 */ - u8 rxdid; /* descriptor builder profile id */ + u8 rxdid; /* descriptor builder profile ID */ u8 mir_id_umb_cast; /* mirror=[5:0], umb=[7:6] */ __le16 ptype_flex_flags0; /* ptype=[9:0], ff0=[15:10] */ __le16 pkt_len; /* [15:14] are reserved */ @@ -149,7 +149,7 @@ union ice_32b_rx_flex_desc { /* Rx Flex Descriptor NIC Profile * This descriptor corresponds to RxDID 2 which contains - * metadata fields for RSS, flow id and timestamp info + * metadata fields for RSS, flow ID and timestamp info */ struct ice_32b_rx_flex_desc_nic { /* Qword 0 */ @@ -208,23 +208,23 @@ enum ice_flex_rx_mdid { ICE_RX_MDID_HASH_HIGH, }; -/* Rx Flag64 packet flag bits */ -enum ice_rx_flg64_bits { - ICE_RXFLG_PKT_DSI = 0, - ICE_RXFLG_EVLAN_x8100 = 15, - ICE_RXFLG_EVLAN_x9100, - ICE_RXFLG_VLAN_x8100, - ICE_RXFLG_TNL_MAC = 22, - ICE_RXFLG_TNL_VLAN, - ICE_RXFLG_PKT_FRG, - ICE_RXFLG_FIN = 32, - ICE_RXFLG_SYN, - ICE_RXFLG_RST, - ICE_RXFLG_TNL0 = 38, - ICE_RXFLG_TNL1, - ICE_RXFLG_TNL2, - ICE_RXFLG_UDP_GRE, - ICE_RXFLG_RSVD = 63 +/* Rx/Tx Flag64 packet flag bits */ +enum ice_flg64_bits { + ICE_FLG_PKT_DSI = 0, + ICE_FLG_EVLAN_x8100 = 15, + ICE_FLG_EVLAN_x9100, + ICE_FLG_VLAN_x8100, + ICE_FLG_TNL_MAC = 22, + ICE_FLG_TNL_VLAN, + ICE_FLG_PKT_FRG, + ICE_FLG_FIN = 32, + ICE_FLG_SYN, + ICE_FLG_RST, + ICE_FLG_TNL0 = 38, + ICE_FLG_TNL1, + ICE_FLG_TNL2, + ICE_FLG_UDP_GRE, + ICE_FLG_RSVD = 63 }; /* for ice_32byte_rx_flex_desc.ptype_flexi_flags0 member */ @@ -322,7 +322,7 @@ enum ice_rlan_ctx_rx_hsplit_1 { ICE_RLAN_RX_HSPLIT_1_SPLIT_ALWAYS = 2, }; -/* TX Descriptor */ +/* Tx Descriptor */ struct ice_tx_desc { __le64 buf_addr; /* Address of descriptor's data buf */ __le64 cmd_type_offset_bsz; @@ -342,12 +342,12 @@ enum ice_tx_desc_cmd_bits { ICE_TX_DESC_CMD_EOP = 0x0001, ICE_TX_DESC_CMD_RS = 0x0002, ICE_TX_DESC_CMD_IL2TAG1 = 0x0008, - ICE_TX_DESC_CMD_IIPT_IPV6 = 0x0020, /* 2 BITS */ - ICE_TX_DESC_CMD_IIPT_IPV4 = 0x0040, /* 2 BITS */ - ICE_TX_DESC_CMD_IIPT_IPV4_CSUM = 0x0060, /* 2 BITS */ - ICE_TX_DESC_CMD_L4T_EOFT_TCP = 0x0100, /* 2 BITS */ - ICE_TX_DESC_CMD_L4T_EOFT_SCTP = 0x0200, /* 2 BITS */ - ICE_TX_DESC_CMD_L4T_EOFT_UDP = 0x0300, /* 2 BITS */ + ICE_TX_DESC_CMD_IIPT_IPV6 = 0x0020, + ICE_TX_DESC_CMD_IIPT_IPV4 = 0x0040, + ICE_TX_DESC_CMD_IIPT_IPV4_CSUM = 0x0060, + ICE_TX_DESC_CMD_L4T_EOFT_TCP = 0x0100, + ICE_TX_DESC_CMD_L4T_EOFT_SCTP = 0x0200, + ICE_TX_DESC_CMD_L4T_EOFT_UDP = 0x0300, }; #define ICE_TXD_QW1_OFFSET_S 16 diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index fa61203bee26..a19f5920733b 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -3,6 +3,7 @@ #include "ice.h" #include "ice_lib.h" +#include "ice_dcb_lib.h" /** * ice_setup_rx_ctx - Configure a receive ring context @@ -73,7 +74,7 @@ static int ice_setup_rx_ctx(struct ice_ring *ring) regval |= (rxdid << QRXFLXP_CNTXT_RXDID_IDX_S) & QRXFLXP_CNTXT_RXDID_IDX_M; - /* increasing context priority to pick up profile id; + /* increasing context priority to pick up profile ID; * default is 0x01; setting to 0x03 to ensure profile * is programming if prev context is of same priority */ @@ -124,6 +125,8 @@ ice_setup_tx_ctx(struct ice_ring *ring, struct ice_tlan_ctx *tlan_ctx, u16 pf_q) /* Transmit Queue Length */ tlan_ctx->qlen = ring->count; + ice_set_cgd_num(tlan_ctx, ring); + /* PF number */ tlan_ctx->pf_num = hw->pf_id; @@ -134,11 +137,13 @@ ice_setup_tx_ctx(struct ice_ring *ring, struct ice_tlan_ctx *tlan_ctx, u16 pf_q) * for PF or EMP this field should be set to zero */ switch (vsi->type) { + case ICE_VSI_LB: + /* fall through */ case ICE_VSI_PF: tlan_ctx->vmvf_type = ICE_TLAN_CTX_VMVF_TYPE_PF; break; case ICE_VSI_VF: - /* Firmware expects vmvf_num to be absolute VF id */ + /* Firmware expects vmvf_num to be absolute VF ID */ tlan_ctx->vmvf_num = hw->func_caps.vf_base_id + vsi->vf_id; tlan_ctx->vmvf_type = ICE_TLAN_CTX_VMVF_TYPE_VF; break; @@ -175,17 +180,14 @@ static int ice_pf_rxq_wait(struct ice_pf *pf, int pf_q, bool ena) int i; for (i = 0; i < ICE_Q_WAIT_MAX_RETRY; i++) { - u32 rx_reg = rd32(&pf->hw, QRX_CTRL(pf_q)); - - if (ena == !!(rx_reg & QRX_CTRL_QENA_STAT_M)) - break; + if (ena == !!(rd32(&pf->hw, QRX_CTRL(pf_q)) & + QRX_CTRL_QENA_STAT_M)) + return 0; usleep_range(20, 40); } - if (i >= ICE_Q_WAIT_MAX_RETRY) - return -ETIMEDOUT; - return 0; + return -ETIMEDOUT; } /** @@ -197,19 +199,13 @@ static int ice_vsi_ctrl_rx_rings(struct ice_vsi *vsi, bool ena) { struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; - int i, j, ret = 0; + int i, ret = 0; for (i = 0; i < vsi->num_rxq; i++) { int pf_q = vsi->rxq_map[i]; u32 rx_reg; - for (j = 0; j < ICE_Q_WAIT_MAX_RETRY; j++) { - rx_reg = rd32(hw, QRX_CTRL(pf_q)); - if (((rx_reg >> QRX_CTRL_QENA_REQ_S) & 1) == - ((rx_reg >> QRX_CTRL_QENA_STAT_S) & 1)) - break; - usleep_range(1000, 2000); - } + rx_reg = rd32(hw, QRX_CTRL(pf_q)); /* Skip if the queue is already in the requested state */ if (ena == !!(rx_reg & QRX_CTRL_QENA_STAT_M)) @@ -238,12 +234,11 @@ static int ice_vsi_ctrl_rx_rings(struct ice_vsi *vsi, bool ena) /** * ice_vsi_alloc_arrays - Allocate queue and vector pointer arrays for the VSI * @vsi: VSI pointer - * @alloc_qvectors: a bool to specify if q_vectors need to be allocated. * * On error: returns error code (negative) * On success: returns 0 */ -static int ice_vsi_alloc_arrays(struct ice_vsi *vsi, bool alloc_qvectors) +static int ice_vsi_alloc_arrays(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; @@ -258,15 +253,15 @@ static int ice_vsi_alloc_arrays(struct ice_vsi *vsi, bool alloc_qvectors) if (!vsi->rx_rings) goto err_rxrings; - if (alloc_qvectors) { - /* allocate memory for q_vector pointers */ - vsi->q_vectors = devm_kcalloc(&pf->pdev->dev, - vsi->num_q_vectors, - sizeof(*vsi->q_vectors), - GFP_KERNEL); - if (!vsi->q_vectors) - goto err_vectors; - } + /* There is no need to allocate q_vectors for a loopback VSI. */ + if (vsi->type == ICE_VSI_LB) + return 0; + + /* allocate memory for q_vector pointers */ + vsi->q_vectors = devm_kcalloc(&pf->pdev->dev, vsi->num_q_vectors, + sizeof(*vsi->q_vectors), GFP_KERNEL); + if (!vsi->q_vectors) + goto err_vectors; return 0; @@ -279,37 +274,68 @@ err_txrings: } /** - * ice_vsi_set_num_qs - Set num queues, descriptors and vectors for a VSI + * ice_vsi_set_num_desc - Set number of descriptors for queues on this VSI + * @vsi: the VSI being configured + */ +static void ice_vsi_set_num_desc(struct ice_vsi *vsi) +{ + switch (vsi->type) { + case ICE_VSI_PF: + /* fall through */ + case ICE_VSI_LB: + vsi->num_rx_desc = ICE_DFLT_NUM_RX_DESC; + vsi->num_tx_desc = ICE_DFLT_NUM_TX_DESC; + break; + default: + dev_dbg(&vsi->back->pdev->dev, + "Not setting number of Tx/Rx descriptors for VSI type %d\n", + vsi->type); + break; + } +} + +/** + * ice_vsi_set_num_qs - Set number of queues, descriptors and vectors for a VSI * @vsi: the VSI being configured + * @vf_id: ID of the VF being configured * * Return 0 on success and a negative value on error */ -static void ice_vsi_set_num_qs(struct ice_vsi *vsi) +static void ice_vsi_set_num_qs(struct ice_vsi *vsi, u16 vf_id) { struct ice_pf *pf = vsi->back; + struct ice_vf *vf = NULL; + + if (vsi->type == ICE_VSI_VF) + vsi->vf_id = vf_id; switch (vsi->type) { case ICE_VSI_PF: vsi->alloc_txq = pf->num_lan_tx; vsi->alloc_rxq = pf->num_lan_rx; - vsi->num_desc = ALIGN(ICE_DFLT_NUM_DESC, ICE_REQ_DESC_MULTIPLE); vsi->num_q_vectors = max_t(int, pf->num_lan_rx, pf->num_lan_tx); break; case ICE_VSI_VF: - vsi->alloc_txq = pf->num_vf_qps; - vsi->alloc_rxq = pf->num_vf_qps; + vf = &pf->vf[vsi->vf_id]; + vsi->alloc_txq = vf->num_vf_qs; + vsi->alloc_rxq = vf->num_vf_qs; /* pf->num_vf_msix includes (VF miscellaneous vector + * data queue interrupts). Since vsi->num_q_vectors is number - * of queues vectors, subtract 1 from the original vector - * count + * of queues vectors, subtract 1 (ICE_NONQ_VECS_VF) from the + * original vector count */ - vsi->num_q_vectors = pf->num_vf_msix - 1; + vsi->num_q_vectors = pf->num_vf_msix - ICE_NONQ_VECS_VF; + break; + case ICE_VSI_LB: + vsi->alloc_txq = 1; + vsi->alloc_rxq = 1; break; default: - dev_warn(&vsi->back->pdev->dev, "Unknown VSI type %d\n", - vsi->type); + dev_warn(&pf->pdev->dev, "Unknown VSI type %d\n", vsi->type); break; } + + ice_vsi_set_num_desc(vsi); } /** @@ -370,16 +396,15 @@ void ice_vsi_delete(struct ice_vsi *vsi) } /** - * ice_vsi_free_arrays - clean up VSI resources + * ice_vsi_free_arrays - De-allocate queue and vector pointer arrays for the VSI * @vsi: pointer to VSI being cleared - * @free_qvectors: bool to specify if q_vectors should be deallocated */ -static void ice_vsi_free_arrays(struct ice_vsi *vsi, bool free_qvectors) +static void ice_vsi_free_arrays(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; /* free the ring and vector containers */ - if (free_qvectors && vsi->q_vectors) { + if (vsi->q_vectors) { devm_kfree(&pf->pdev->dev, vsi->q_vectors); vsi->q_vectors = NULL; } @@ -427,7 +452,7 @@ int ice_vsi_clear(struct ice_vsi *vsi) if (vsi->idx < pf->next_vsi) pf->next_vsi = vsi->idx; - ice_vsi_free_arrays(vsi, true); + ice_vsi_free_arrays(vsi); mutex_unlock(&pf->sw_mutex); devm_kfree(&pf->pdev->dev, vsi); @@ -455,10 +480,12 @@ static irqreturn_t ice_msix_clean_rings(int __always_unused irq, void *data) * ice_vsi_alloc - Allocates the next available struct VSI in the PF * @pf: board private structure * @type: type of VSI + * @vf_id: ID of the VF being configured * * returns a pointer to a VSI on success, NULL on failure. */ -static struct ice_vsi *ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type) +static struct ice_vsi * +ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type, u16 vf_id) { struct ice_vsi *vsi = NULL; @@ -484,18 +511,25 @@ static struct ice_vsi *ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type) vsi->idx = pf->next_vsi; vsi->work_lmt = ICE_DFLT_IRQ_WORK; - ice_vsi_set_num_qs(vsi); + if (type == ICE_VSI_VF) + ice_vsi_set_num_qs(vsi, vf_id); + else + ice_vsi_set_num_qs(vsi, ICE_INVAL_VFID); switch (vsi->type) { case ICE_VSI_PF: - if (ice_vsi_alloc_arrays(vsi, true)) + if (ice_vsi_alloc_arrays(vsi)) goto err_rings; /* Setup default MSIX irq handler for VSI */ vsi->irq_handler = ice_msix_clean_rings; break; case ICE_VSI_VF: - if (ice_vsi_alloc_arrays(vsi, true)) + if (ice_vsi_alloc_arrays(vsi)) + goto err_rings; + break; + case ICE_VSI_LB: + if (ice_vsi_alloc_arrays(vsi)) goto err_rings; break; default: @@ -547,7 +581,7 @@ static int __ice_vsi_get_qs_contig(struct ice_qs_cfg *qs_cfg) /** * __ice_vsi_get_qs_sc - Assign a scattered queues from PF to VSI - * @qs_cfg: gathered variables needed for PF->VSI queues assignment + * @qs_cfg: gathered variables needed for pf->vsi queues assignment * * Return 0 on success and -ENOMEM in case of no left space in PF queue bitmap */ @@ -579,11 +613,10 @@ err_scatter: /** * __ice_vsi_get_qs - helper function for assigning queues from PF to VSI - * @qs_cfg: gathered variables needed for PF->VSI queues assignment + * @qs_cfg: gathered variables needed for pf->vsi queues assignment * - * This is an internal function for assigning queues from the PF to VSI and - * initially tries to find contiguous space. If it is not successful to find - * contiguous space, then it tries with the scatter approach. + * This function first tries to find contiguous space. If it is not successful, + * it tries with the scatter approach. * * Return 0 on success and -ENOMEM in case of no left space in PF queue bitmap */ @@ -715,6 +748,8 @@ static void ice_vsi_set_rss_params(struct ice_vsi *vsi) BIT(cap->rss_table_entry_width)); vsi->rss_lut_type = ICE_AQC_GSET_RSS_LUT_TABLE_TYPE_VSI; break; + case ICE_VSI_LB: + break; default: dev_warn(&pf->pdev->dev, "Unknown VSI type %d\n", vsi->type); @@ -827,7 +862,7 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt) /* find the (rounded up) power-of-2 of qcount */ pow = order_base_2(qcount_rx); - for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) { + ice_for_each_traffic_class(i) { if (!(vsi->tc_cfg.ena_tc & BIT(i))) { /* TC is not enabled */ vsi->tc_cfg.tc_info[i].qoffset = 0; @@ -852,7 +887,18 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt) tx_count += tx_numq_tc; ctxt->info.tc_mapping[i] = cpu_to_le16(qmap); } - vsi->num_rxq = offset; + + /* if offset is non-zero, means it is calculated correctly based on + * enabled TCs for a given VSI otherwise qcount_rx will always + * be correct and non-zero because it is based off - VSI's + * allocated Rx queues which is at least 1 (hence qcount_tx will be + * at least 1) + */ + if (offset) + vsi->num_rxq = offset; + else + vsi->num_rxq = qcount_rx; + vsi->num_txq = tx_count; if (vsi->type == ICE_VSI_VF && vsi->num_txq != vsi->num_rxq) { @@ -881,6 +927,9 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt) static void ice_set_rss_vsi_ctx(struct ice_vsi_ctx *ctxt, struct ice_vsi *vsi) { u8 lut_type, hash_type; + struct ice_pf *pf; + + pf = vsi->back; switch (vsi->type) { case ICE_VSI_PF: @@ -893,9 +942,11 @@ static void ice_set_rss_vsi_ctx(struct ice_vsi_ctx *ctxt, struct ice_vsi *vsi) lut_type = ICE_AQ_VSI_Q_OPT_RSS_LUT_VSI; hash_type = ICE_AQ_VSI_Q_OPT_RSS_TPLZ; break; + case ICE_VSI_LB: + dev_dbg(&pf->pdev->dev, "Unsupported VSI type %d\n", vsi->type); + return; default: - dev_warn(&vsi->back->pdev->dev, "Unknown VSI type %d\n", - vsi->type); + dev_warn(&pf->pdev->dev, "Unknown VSI type %d\n", vsi->type); return; } @@ -923,7 +974,10 @@ static int ice_vsi_init(struct ice_vsi *vsi) if (!ctxt) return -ENOMEM; + ctxt->info = vsi->info; switch (vsi->type) { + case ICE_VSI_LB: + /* fall through */ case ICE_VSI_PF: ctxt->flags = ICE_AQ_VSI_TYPE_PF; break; @@ -948,6 +1002,14 @@ static int ice_vsi_init(struct ice_vsi *vsi) ctxt->info.sw_id = vsi->port_info->sw_id; ice_vsi_setup_q_map(vsi, ctxt); + /* Enable MAC Antispoof with new VSI being initialized or updated */ + if (vsi->type == ICE_VSI_VF && pf->vf[vsi->vf_id].spoofchk) { + ctxt->info.valid_sections |= + cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID); + ctxt->info.sec_flags |= + ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF; + } + ret = ice_add_vsi(hw, vsi->idx, ctxt, NULL); if (ret) { dev_err(&pf->pdev->dev, @@ -973,10 +1035,11 @@ static int ice_vsi_init(struct ice_vsi *vsi) static void ice_free_q_vector(struct ice_vsi *vsi, int v_idx) { struct ice_q_vector *q_vector; + struct ice_pf *pf = vsi->back; struct ice_ring *ring; if (!vsi->q_vectors[v_idx]) { - dev_dbg(&vsi->back->pdev->dev, "Queue vector at index %d not found\n", + dev_dbg(&pf->pdev->dev, "Queue vector at index %d not found\n", v_idx); return; } @@ -991,7 +1054,7 @@ static void ice_free_q_vector(struct ice_vsi *vsi, int v_idx) if (vsi->netdev) netif_napi_del(&q_vector->napi); - devm_kfree(&vsi->back->pdev->dev, q_vector); + devm_kfree(&pf->pdev->dev, q_vector); vsi->q_vectors[v_idx] = NULL; } @@ -1003,7 +1066,7 @@ void ice_vsi_free_q_vectors(struct ice_vsi *vsi) { int v_idx; - for (v_idx = 0; v_idx < vsi->num_q_vectors; v_idx++) + ice_for_each_q_vector(vsi, v_idx) ice_free_q_vector(vsi, v_idx); } @@ -1105,62 +1168,32 @@ err_out: static int ice_vsi_setup_vector_base(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; - int num_q_vectors = 0; + u16 num_q_vectors; - if (vsi->sw_base_vector || vsi->hw_base_vector) { - dev_dbg(&pf->pdev->dev, "VSI %d has non-zero HW base vector %d or SW base vector %d\n", - vsi->vsi_num, vsi->hw_base_vector, vsi->sw_base_vector); + /* SRIOV doesn't grab irq_tracker entries for each VSI */ + if (vsi->type == ICE_VSI_VF) + return 0; + + if (vsi->base_vector) { + dev_dbg(&pf->pdev->dev, "VSI %d has non-zero base vector %d\n", + vsi->vsi_num, vsi->base_vector); return -EEXIST; } if (!test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) return -ENOENT; - switch (vsi->type) { - case ICE_VSI_PF: - num_q_vectors = vsi->num_q_vectors; - /* reserve slots from OS requested IRQs */ - vsi->sw_base_vector = ice_get_res(pf, pf->sw_irq_tracker, - num_q_vectors, vsi->idx); - if (vsi->sw_base_vector < 0) { - dev_err(&pf->pdev->dev, - "Failed to get tracking for %d SW vectors for VSI %d, err=%d\n", - num_q_vectors, vsi->vsi_num, - vsi->sw_base_vector); - return -ENOENT; - } - pf->num_avail_sw_msix -= num_q_vectors; - - /* reserve slots from HW interrupts */ - vsi->hw_base_vector = ice_get_res(pf, pf->hw_irq_tracker, - num_q_vectors, vsi->idx); - break; - case ICE_VSI_VF: - /* take VF misc vector and data vectors into account */ - num_q_vectors = pf->num_vf_msix; - /* For VF VSI, reserve slots only from HW interrupts */ - vsi->hw_base_vector = ice_get_res(pf, pf->hw_irq_tracker, - num_q_vectors, vsi->idx); - break; - default: - dev_warn(&vsi->back->pdev->dev, "Unknown VSI type %d\n", - vsi->type); - break; - } - - if (vsi->hw_base_vector < 0) { + num_q_vectors = vsi->num_q_vectors; + /* reserve slots from OS requested IRQs */ + vsi->base_vector = ice_get_res(pf, pf->irq_tracker, num_q_vectors, + vsi->idx); + if (vsi->base_vector < 0) { dev_err(&pf->pdev->dev, - "Failed to get tracking for %d HW vectors for VSI %d, err=%d\n", - num_q_vectors, vsi->vsi_num, vsi->hw_base_vector); - if (vsi->type != ICE_VSI_VF) { - ice_free_res(vsi->back->sw_irq_tracker, - vsi->sw_base_vector, vsi->idx); - pf->num_avail_sw_msix += num_q_vectors; - } + "Failed to get tracking for %d vectors for VSI %d, err=%d\n", + num_q_vectors, vsi->vsi_num, vsi->base_vector); return -ENOENT; } - - pf->num_avail_hw_msix -= num_q_vectors; + pf->num_avail_sw_msix -= num_q_vectors; return 0; } @@ -1215,7 +1248,7 @@ static int ice_vsi_alloc_rings(struct ice_vsi *vsi) ring->ring_active = false; ring->vsi = vsi; ring->dev = &pf->pdev->dev; - ring->count = vsi->num_desc; + ring->count = vsi->num_tx_desc; vsi->tx_rings[i] = ring; } @@ -1234,7 +1267,7 @@ static int ice_vsi_alloc_rings(struct ice_vsi *vsi) ring->vsi = vsi; ring->netdev = vsi->netdev; ring->dev = &pf->pdev->dev; - ring->count = vsi->num_desc; + ring->count = vsi->num_rx_desc; vsi->rx_rings[i] = ring; } @@ -1253,7 +1286,11 @@ err_out: * through the MSI-X enabling code. On a constrained vector budget, we map Tx * and Rx rings to the vector as "efficiently" as possible. */ +#ifdef CONFIG_DCB +void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi) +#else static void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi) +#endif /* CONFIG_DCB */ { int q_vectors = vsi->num_q_vectors; int tx_rings_rem, rx_rings_rem; @@ -1339,7 +1376,6 @@ int ice_vsi_manage_rss_lut(struct ice_vsi *vsi, bool ena) */ static int ice_vsi_cfg_rss_lut_key(struct ice_vsi *vsi) { - u8 seed[ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE]; struct ice_aqc_get_set_rss_keys *key; struct ice_pf *pf = vsi->back; enum ice_status status; @@ -1361,31 +1397,30 @@ static int ice_vsi_cfg_rss_lut_key(struct ice_vsi *vsi) vsi->rss_table_size); if (status) { - dev_err(&vsi->back->pdev->dev, + dev_err(&pf->pdev->dev, "set_rss_lut failed, error %d\n", status); err = -EIO; goto ice_vsi_cfg_rss_exit; } - key = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*key), GFP_KERNEL); + key = devm_kzalloc(&pf->pdev->dev, sizeof(*key), GFP_KERNEL); if (!key) { err = -ENOMEM; goto ice_vsi_cfg_rss_exit; } if (vsi->rss_hkey_user) - memcpy(seed, vsi->rss_hkey_user, - ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE); + memcpy(key, + (struct ice_aqc_get_set_rss_keys *)vsi->rss_hkey_user, + ICE_GET_SET_RSS_KEY_EXTEND_KEY_SIZE); else - netdev_rss_key_fill((void *)seed, - ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE); - memcpy(&key->standard_rss_key, seed, - ICE_AQC_GET_SET_RSS_KEY_DATA_RSS_KEY_SIZE); + netdev_rss_key_fill((void *)key, + ICE_GET_SET_RSS_KEY_EXTEND_KEY_SIZE); status = ice_aq_set_rss_key(&pf->hw, vsi->idx, key); if (status) { - dev_err(&vsi->back->pdev->dev, "set_rss_key failed, error %d\n", + dev_err(&pf->pdev->dev, "set_rss_key failed, error %d\n", status); err = -EIO; } @@ -1397,12 +1432,12 @@ ice_vsi_cfg_rss_exit: } /** - * ice_add_mac_to_list - Add a mac address filter entry to the list + * ice_add_mac_to_list - Add a MAC address filter entry to the list * @vsi: the VSI to be forwarded to * @add_list: pointer to the list which contains MAC filter entries * @macaddr: the MAC address to be added. * - * Adds mac address filter entry to the temp list + * Adds MAC address filter entry to the temp list * * Returns 0 on success or ENOMEM on failure. */ @@ -1504,7 +1539,7 @@ void ice_free_fltr_list(struct device *dev, struct list_head *h) /** * ice_vsi_add_vlan - Add VSI membership for given VLAN * @vsi: the VSI being configured - * @vid: VLAN id to be added + * @vid: VLAN ID to be added */ int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid) { @@ -1542,7 +1577,7 @@ int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid) /** * ice_vsi_kill_vlan - Remove VSI membership for a given VLAN * @vsi: the VSI being configured - * @vid: VLAN id to be removed + * @vid: VLAN ID to be removed * * Returns 0 on success and negative on failure */ @@ -1551,7 +1586,8 @@ int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid) struct ice_fltr_list_entry *list; struct ice_pf *pf = vsi->back; LIST_HEAD(tmp_add_list); - int status = 0; + enum ice_status status; + int err = 0; list = devm_kzalloc(&pf->pdev->dev, sizeof(*list), GFP_KERNEL); if (!list) @@ -1567,14 +1603,20 @@ int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid) INIT_LIST_HEAD(&list->list_entry); list_add(&list->list_entry, &tmp_add_list); - if (ice_remove_vlan(&pf->hw, &tmp_add_list)) { - dev_err(&pf->pdev->dev, "Error removing VLAN %d on vsi %i\n", - vid, vsi->vsi_num); - status = -EIO; + status = ice_remove_vlan(&pf->hw, &tmp_add_list); + if (status == ICE_ERR_DOES_NOT_EXIST) { + dev_dbg(&pf->pdev->dev, + "Failed to remove VLAN %d on VSI %i, it does not exist, status: %d\n", + vid, vsi->vsi_num, status); + } else if (status) { + dev_err(&pf->pdev->dev, + "Error removing VLAN %d on vsi %i error: %d\n", + vid, vsi->vsi_num, status); + err = -EIO; } ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); - return status; + return err; } /** @@ -1586,7 +1628,6 @@ int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid) */ int ice_vsi_cfg_rxqs(struct ice_vsi *vsi) { - int err = 0; u16 i; if (vsi->type == ICE_VSI_VF) @@ -1601,14 +1642,19 @@ int ice_vsi_cfg_rxqs(struct ice_vsi *vsi) vsi->rx_buf_len = ICE_RXBUF_2048; setup_rings: /* set up individual rings */ - for (i = 0; i < vsi->num_rxq && !err; i++) - err = ice_setup_rx_ctx(vsi->rx_rings[i]); + for (i = 0; i < vsi->num_rxq; i++) { + int err; - if (err) { - dev_err(&vsi->back->pdev->dev, "ice_setup_rx_ctx failed\n"); - return -EIO; + err = ice_setup_rx_ctx(vsi->rx_rings[i]); + if (err) { + dev_err(&vsi->back->pdev->dev, + "ice_setup_rx_ctx failed for RxQ %d, err %d\n", + i, err); + return err; + } } - return err; + + return 0; } /** @@ -1640,7 +1686,7 @@ ice_vsi_cfg_txqs(struct ice_vsi *vsi, struct ice_ring **rings, int offset) num_q_grps = 1; /* set up and configure the Tx queues for each enabled TC */ - for (tc = 0; tc < ICE_MAX_TRAFFIC_CLASS; tc++) { + ice_for_each_traffic_class(tc) { if (!(vsi->tc_cfg.ena_tc & BIT(tc))) break; @@ -1660,10 +1706,10 @@ ice_vsi_cfg_txqs(struct ice_vsi *vsi, struct ice_ring **rings, int offset) rings[q_idx]->tail = pf->hw.hw_addr + QTX_COMM_DBELL(pf_q); status = ice_ena_vsi_txq(vsi->port_info, vsi->idx, tc, - num_q_grps, qg_buf, buf_len, - NULL); + i, num_q_grps, qg_buf, + buf_len, NULL); if (status) { - dev_err(&vsi->back->pdev->dev, + dev_err(&pf->pdev->dev, "Failed to set LAN Tx queue context, error: %d\n", status); err = -ENODEV; @@ -1707,7 +1753,7 @@ int ice_vsi_cfg_lan_txqs(struct ice_vsi *vsi) * This function converts a decimal interrupt rate limit in usecs to the format * expected by firmware. */ -static u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran) +u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran) { u32 val = intrl / gran; @@ -1717,17 +1763,49 @@ static u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran) } /** + * ice_cfg_itr_gran - set the ITR granularity to 2 usecs if not already set + * @hw: board specific structure + */ +static void ice_cfg_itr_gran(struct ice_hw *hw) +{ + u32 regval = rd32(hw, GLINT_CTL); + + /* no need to update global register if ITR gran is already set */ + if (!(regval & GLINT_CTL_DIS_AUTOMASK_M) && + (((regval & GLINT_CTL_ITR_GRAN_200_M) >> + GLINT_CTL_ITR_GRAN_200_S) == ICE_ITR_GRAN_US) && + (((regval & GLINT_CTL_ITR_GRAN_100_M) >> + GLINT_CTL_ITR_GRAN_100_S) == ICE_ITR_GRAN_US) && + (((regval & GLINT_CTL_ITR_GRAN_50_M) >> + GLINT_CTL_ITR_GRAN_50_S) == ICE_ITR_GRAN_US) && + (((regval & GLINT_CTL_ITR_GRAN_25_M) >> + GLINT_CTL_ITR_GRAN_25_S) == ICE_ITR_GRAN_US)) + return; + + regval = ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_200_S) & + GLINT_CTL_ITR_GRAN_200_M) | + ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_100_S) & + GLINT_CTL_ITR_GRAN_100_M) | + ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_50_S) & + GLINT_CTL_ITR_GRAN_50_M) | + ((ICE_ITR_GRAN_US << GLINT_CTL_ITR_GRAN_25_S) & + GLINT_CTL_ITR_GRAN_25_M); + wr32(hw, GLINT_CTL, regval); +} + +/** * ice_cfg_itr - configure the initial interrupt throttle values * @hw: pointer to the HW structure * @q_vector: interrupt vector that's being configured - * @vector: HW vector index to apply the interrupt throttling to * * Configure interrupt throttling values for the ring containers that are * associated with the interrupt vector passed in. */ static void -ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector, u16 vector) +ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector) { + ice_cfg_itr_gran(hw); + if (q_vector->num_ring_rx) { struct ice_ring_container *rc = &q_vector->rx; @@ -1738,8 +1816,7 @@ ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector, u16 vector) rc->target_itr = ITR_TO_REG(rc->itr_setting); rc->next_update = jiffies + 1; rc->current_itr = rc->target_itr; - rc->latency_range = ICE_LOW_LATENCY; - wr32(hw, GLINT_ITR(rc->itr_idx, vector), + wr32(hw, GLINT_ITR(rc->itr_idx, q_vector->reg_idx), ITR_REG_ALIGN(rc->current_itr) >> ICE_ITR_GRAN_S); } @@ -1753,30 +1830,94 @@ ice_cfg_itr(struct ice_hw *hw, struct ice_q_vector *q_vector, u16 vector) rc->target_itr = ITR_TO_REG(rc->itr_setting); rc->next_update = jiffies + 1; rc->current_itr = rc->target_itr; - rc->latency_range = ICE_LOW_LATENCY; - wr32(hw, GLINT_ITR(rc->itr_idx, vector), + wr32(hw, GLINT_ITR(rc->itr_idx, q_vector->reg_idx), ITR_REG_ALIGN(rc->current_itr) >> ICE_ITR_GRAN_S); } } /** + * ice_cfg_txq_interrupt - configure interrupt on Tx queue + * @vsi: the VSI being configured + * @txq: Tx queue being mapped to MSI-X vector + * @msix_idx: MSI-X vector index within the function + * @itr_idx: ITR index of the interrupt cause + * + * Configure interrupt on Tx queue by associating Tx queue to MSI-X vector + * within the function space. + */ +#ifdef CONFIG_PCI_IOV +void +ice_cfg_txq_interrupt(struct ice_vsi *vsi, u16 txq, u16 msix_idx, u16 itr_idx) +#else +static void +ice_cfg_txq_interrupt(struct ice_vsi *vsi, u16 txq, u16 msix_idx, u16 itr_idx) +#endif /* CONFIG_PCI_IOV */ +{ + struct ice_pf *pf = vsi->back; + struct ice_hw *hw = &pf->hw; + u32 val; + + itr_idx = (itr_idx << QINT_TQCTL_ITR_INDX_S) & QINT_TQCTL_ITR_INDX_M; + + val = QINT_TQCTL_CAUSE_ENA_M | itr_idx | + ((msix_idx << QINT_TQCTL_MSIX_INDX_S) & QINT_TQCTL_MSIX_INDX_M); + + wr32(hw, QINT_TQCTL(vsi->txq_map[txq]), val); +} + +/** + * ice_cfg_rxq_interrupt - configure interrupt on Rx queue + * @vsi: the VSI being configured + * @rxq: Rx queue being mapped to MSI-X vector + * @msix_idx: MSI-X vector index within the function + * @itr_idx: ITR index of the interrupt cause + * + * Configure interrupt on Rx queue by associating Rx queue to MSI-X vector + * within the function space. + */ +#ifdef CONFIG_PCI_IOV +void +ice_cfg_rxq_interrupt(struct ice_vsi *vsi, u16 rxq, u16 msix_idx, u16 itr_idx) +#else +static void +ice_cfg_rxq_interrupt(struct ice_vsi *vsi, u16 rxq, u16 msix_idx, u16 itr_idx) +#endif /* CONFIG_PCI_IOV */ +{ + struct ice_pf *pf = vsi->back; + struct ice_hw *hw = &pf->hw; + u32 val; + + itr_idx = (itr_idx << QINT_RQCTL_ITR_INDX_S) & QINT_RQCTL_ITR_INDX_M; + + val = QINT_RQCTL_CAUSE_ENA_M | itr_idx | + ((msix_idx << QINT_RQCTL_MSIX_INDX_S) & QINT_RQCTL_MSIX_INDX_M); + + wr32(hw, QINT_RQCTL(vsi->rxq_map[rxq]), val); + + ice_flush(hw); +} + +/** * ice_vsi_cfg_msix - MSIX mode Interrupt Config in the HW * @vsi: the VSI being configured + * + * This configures MSIX mode interrupts for the PF VSI, and should not be used + * for the VF VSI. */ void ice_vsi_cfg_msix(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; - u16 vector = vsi->hw_base_vector; struct ice_hw *hw = &pf->hw; u32 txq = 0, rxq = 0; int i, q; - for (i = 0; i < vsi->num_q_vectors; i++, vector++) { + for (i = 0; i < vsi->num_q_vectors; i++) { struct ice_q_vector *q_vector = vsi->q_vectors[i]; + u16 reg_idx = q_vector->reg_idx; - ice_cfg_itr(hw, q_vector, vector); + ice_cfg_itr(hw, q_vector); - wr32(hw, GLINT_RATE(vector), + wr32(hw, GLINT_RATE(reg_idx), ice_intrl_usec_to_reg(q_vector->intrl, hw->intrl_gran)); /* Both Transmit Queue Interrupt Cause Control register @@ -1791,39 +1932,17 @@ void ice_vsi_cfg_msix(struct ice_vsi *vsi) * tracked for this PF. */ for (q = 0; q < q_vector->num_ring_tx; q++) { - int itr_idx = q_vector->tx.itr_idx; - u32 val; - - if (vsi->type == ICE_VSI_VF) - val = QINT_TQCTL_CAUSE_ENA_M | - (itr_idx << QINT_TQCTL_ITR_INDX_S) | - ((i + 1) << QINT_TQCTL_MSIX_INDX_S); - else - val = QINT_TQCTL_CAUSE_ENA_M | - (itr_idx << QINT_TQCTL_ITR_INDX_S) | - (vector << QINT_TQCTL_MSIX_INDX_S); - wr32(hw, QINT_TQCTL(vsi->txq_map[txq]), val); + ice_cfg_txq_interrupt(vsi, txq, reg_idx, + q_vector->tx.itr_idx); txq++; } for (q = 0; q < q_vector->num_ring_rx; q++) { - int itr_idx = q_vector->rx.itr_idx; - u32 val; - - if (vsi->type == ICE_VSI_VF) - val = QINT_RQCTL_CAUSE_ENA_M | - (itr_idx << QINT_RQCTL_ITR_INDX_S) | - ((i + 1) << QINT_RQCTL_MSIX_INDX_S); - else - val = QINT_RQCTL_CAUSE_ENA_M | - (itr_idx << QINT_RQCTL_ITR_INDX_S) | - (vector << QINT_RQCTL_MSIX_INDX_S); - wr32(hw, QINT_RQCTL(vsi->rxq_map[rxq]), val); + ice_cfg_rxq_interrupt(vsi, rxq, reg_idx, + q_vector->rx.itr_idx); rxq++; } } - - ice_flush(hw); } /** @@ -1848,6 +1967,10 @@ int ice_vsi_manage_vlan_insertion(struct ice_vsi *vsi) */ ctxt->info.vlan_flags = ICE_AQ_VSI_VLAN_MODE_ALL; + /* Preserve existing VLAN strip setting */ + ctxt->info.vlan_flags |= (vsi->info.vlan_flags & + ICE_AQ_VSI_VLAN_EMOD_M); + ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID); status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); @@ -1934,10 +2057,23 @@ int ice_vsi_stop_rx_rings(struct ice_vsi *vsi) } /** + * ice_trigger_sw_intr - trigger a software interrupt + * @hw: pointer to the HW structure + * @q_vector: interrupt vector to trigger the software interrupt for + */ +void ice_trigger_sw_intr(struct ice_hw *hw, struct ice_q_vector *q_vector) +{ + wr32(hw, GLINT_DYN_CTL(q_vector->reg_idx), + (ICE_ITR_NONE << GLINT_DYN_CTL_ITR_INDX_S) | + GLINT_DYN_CTL_SWINT_TRIG_M | + GLINT_DYN_CTL_INTENA_M); +} + +/** * ice_vsi_stop_tx_rings - Disable Tx rings * @vsi: the VSI being configured * @rst_src: reset source - * @rel_vmvf_num: Relative id of VF/VM + * @rel_vmvf_num: Relative ID of VF/VM * @rings: Tx ring array to be stopped * @offset: offset within vsi->txq_map */ @@ -1947,10 +2083,10 @@ ice_vsi_stop_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, { struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; + int tc, q_idx = 0, err = 0; + u16 *q_ids, *q_handles, i; enum ice_status status; u32 *q_teids, val; - u16 *q_ids, i; - int err = 0; if (vsi->num_txq > ICE_LAN_TXQ_MAX_QDIS) return -EINVAL; @@ -1967,50 +2103,71 @@ ice_vsi_stop_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, goto err_alloc_q_ids; } - /* set up the Tx queue list to be disabled */ - ice_for_each_txq(vsi, i) { - u16 v_idx; + q_handles = devm_kcalloc(&pf->pdev->dev, vsi->num_txq, + sizeof(*q_handles), GFP_KERNEL); + if (!q_handles) { + err = -ENOMEM; + goto err_alloc_q_handles; + } - if (!rings || !rings[i] || !rings[i]->q_vector) { - err = -EINVAL; - goto err_out; - } + /* set up the Tx queue list to be disabled for each enabled TC */ + ice_for_each_traffic_class(tc) { + if (!(vsi->tc_cfg.ena_tc & BIT(tc))) + break; + + for (i = 0; i < vsi->tc_cfg.tc_info[tc].qcount_tx; i++) { + struct ice_q_vector *q_vector; - q_ids[i] = vsi->txq_map[i + offset]; - q_teids[i] = rings[i]->txq_teid; + if (!rings || !rings[q_idx]) { + err = -EINVAL; + goto err_out; + } - /* clear cause_ena bit for disabled queues */ - val = rd32(hw, QINT_TQCTL(rings[i]->reg_idx)); - val &= ~QINT_TQCTL_CAUSE_ENA_M; - wr32(hw, QINT_TQCTL(rings[i]->reg_idx), val); + q_ids[i] = vsi->txq_map[q_idx + offset]; + q_teids[i] = rings[q_idx]->txq_teid; + q_handles[i] = i; - /* software is expected to wait for 100 ns */ - ndelay(100); + /* clear cause_ena bit for disabled queues */ + val = rd32(hw, QINT_TQCTL(rings[i]->reg_idx)); + val &= ~QINT_TQCTL_CAUSE_ENA_M; + wr32(hw, QINT_TQCTL(rings[i]->reg_idx), val); + + /* software is expected to wait for 100 ns */ + ndelay(100); + + /* trigger a software interrupt for the vector + * associated to the queue to schedule NAPI handler + */ + q_vector = rings[i]->q_vector; + if (q_vector) + ice_trigger_sw_intr(hw, q_vector); - /* trigger a software interrupt for the vector associated to - * the queue to schedule NAPI handler + q_idx++; + } + status = ice_dis_vsi_txq(vsi->port_info, vsi->idx, tc, + vsi->num_txq, q_handles, q_ids, + q_teids, rst_src, rel_vmvf_num, NULL); + + /* if the disable queue command was exercised during an active + * reset flow, ICE_ERR_RESET_ONGOING is returned. This is not + * an error as the reset operation disables queues at the + * hardware level anyway. */ - v_idx = rings[i]->q_vector->v_idx; - wr32(hw, GLINT_DYN_CTL(vsi->hw_base_vector + v_idx), - GLINT_DYN_CTL_SWINT_TRIG_M | GLINT_DYN_CTL_INTENA_MSK_M); - } - status = ice_dis_vsi_txq(vsi->port_info, vsi->num_txq, q_ids, q_teids, - rst_src, rel_vmvf_num, NULL); - /* if the disable queue command was exercised during an active reset - * flow, ICE_ERR_RESET_ONGOING is returned. This is not an error as - * the reset operation disables queues at the hardware level anyway. - */ - if (status == ICE_ERR_RESET_ONGOING) { - dev_info(&pf->pdev->dev, - "Reset in progress. LAN Tx queues already disabled\n"); - } else if (status) { - dev_err(&pf->pdev->dev, - "Failed to disable LAN Tx queues, error: %d\n", - status); - err = -ENODEV; + if (status == ICE_ERR_RESET_ONGOING) { + dev_dbg(&pf->pdev->dev, + "Reset in progress. LAN Tx queues already disabled\n"); + } else if (status) { + dev_err(&pf->pdev->dev, + "Failed to disable LAN Tx queues, error: %d\n", + status); + err = -ENODEV; + } } err_out: + devm_kfree(&pf->pdev->dev, q_handles); + +err_alloc_q_handles: devm_kfree(&pf->pdev->dev, q_ids); err_alloc_q_ids: @@ -2023,10 +2180,11 @@ err_alloc_q_ids: * ice_vsi_stop_lan_tx_rings - Disable LAN Tx rings * @vsi: the VSI being configured * @rst_src: reset source - * @rel_vmvf_num: Relative id of VF/VM + * @rel_vmvf_num: Relative ID of VF/VM */ -int ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, - enum ice_disq_rst_src rst_src, u16 rel_vmvf_num) +int +ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, + u16 rel_vmvf_num) { return ice_vsi_stop_tx_rings(vsi, rst_src, rel_vmvf_num, vsi->tx_rings, 0); @@ -2036,19 +2194,22 @@ int ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, * ice_cfg_vlan_pruning - enable or disable VLAN pruning on the VSI * @vsi: VSI to enable or disable VLAN pruning on * @ena: set to true to enable VLAN pruning and false to disable it + * @vlan_promisc: enable valid security flags if not in VLAN promiscuous mode * * returns 0 if VSI is updated, negative otherwise */ -int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena) +int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc) { struct ice_vsi_ctx *ctxt; struct device *dev; + struct ice_pf *pf; int status; if (!vsi) return -EINVAL; - dev = &vsi->back->pdev->dev; + pf = vsi->back; + dev = &pf->pdev->dev; ctxt = devm_kzalloc(dev, sizeof(*ctxt), GFP_KERNEL); if (!ctxt) return -ENOMEM; @@ -2067,14 +2228,16 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena) ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; } - ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID | - ICE_AQ_VSI_PROP_SW_VALID); + if (!vlan_promisc) + ctxt->info.valid_sections = + cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID | + ICE_AQ_VSI_PROP_SW_VALID); - status = ice_update_vsi(&vsi->back->hw, vsi->idx, ctxt, NULL); + status = ice_update_vsi(&pf->hw, vsi->idx, ctxt, NULL); if (status) { netdev_err(vsi->netdev, "%sabling VLAN pruning on VSI handle: %d, VSI HW ID: %d failed, err = %d, aq_err = %d\n", ena ? "En" : "Dis", vsi->idx, vsi->vsi_num, status, - vsi->back->hw.adminq.sq_last_status); + pf->hw.adminq.sq_last_status); goto err_out; } @@ -2089,12 +2252,153 @@ err_out: return -EIO; } +static void ice_vsi_set_tc_cfg(struct ice_vsi *vsi) +{ + struct ice_dcbx_cfg *cfg = &vsi->port_info->local_dcbx_cfg; + + vsi->tc_cfg.ena_tc = ice_dcb_get_ena_tc(cfg); + vsi->tc_cfg.numtc = ice_dcb_get_num_tc(cfg); +} + +/** + * ice_vsi_set_q_vectors_reg_idx - set the HW register index for all q_vectors + * @vsi: VSI to set the q_vectors register index on + */ +static int +ice_vsi_set_q_vectors_reg_idx(struct ice_vsi *vsi) +{ + u16 i; + + if (!vsi || !vsi->q_vectors) + return -EINVAL; + + ice_for_each_q_vector(vsi, i) { + struct ice_q_vector *q_vector = vsi->q_vectors[i]; + + if (!q_vector) { + dev_err(&vsi->back->pdev->dev, + "Failed to set reg_idx on q_vector %d VSI %d\n", + i, vsi->vsi_num); + goto clear_reg_idx; + } + + if (vsi->type == ICE_VSI_VF) { + struct ice_vf *vf = &vsi->back->vf[vsi->vf_id]; + + q_vector->reg_idx = ice_calc_vf_reg_idx(vf, q_vector); + } else { + q_vector->reg_idx = + q_vector->v_idx + vsi->base_vector; + } + } + + return 0; + +clear_reg_idx: + ice_for_each_q_vector(vsi, i) { + struct ice_q_vector *q_vector = vsi->q_vectors[i]; + + if (q_vector) + q_vector->reg_idx = 0; + } + + return -EINVAL; +} + +/** + * ice_vsi_add_rem_eth_mac - Program VSI ethertype based filter with rule + * @vsi: the VSI being configured + * @add_rule: boolean value to add or remove ethertype filter rule + */ +static void +ice_vsi_add_rem_eth_mac(struct ice_vsi *vsi, bool add_rule) +{ + struct ice_fltr_list_entry *list; + struct ice_pf *pf = vsi->back; + LIST_HEAD(tmp_add_list); + enum ice_status status; + + list = devm_kzalloc(&pf->pdev->dev, sizeof(*list), GFP_KERNEL); + if (!list) + return; + + list->fltr_info.lkup_type = ICE_SW_LKUP_ETHERTYPE; + list->fltr_info.fltr_act = ICE_DROP_PACKET; + list->fltr_info.flag = ICE_FLTR_TX; + list->fltr_info.src_id = ICE_SRC_ID_VSI; + list->fltr_info.vsi_handle = vsi->idx; + list->fltr_info.l_data.ethertype_mac.ethertype = vsi->ethtype; + + INIT_LIST_HEAD(&list->list_entry); + list_add(&list->list_entry, &tmp_add_list); + + if (add_rule) + status = ice_add_eth_mac(&pf->hw, &tmp_add_list); + else + status = ice_remove_eth_mac(&pf->hw, &tmp_add_list); + + if (status) + dev_err(&pf->pdev->dev, + "Failure Adding or Removing Ethertype on VSI %i error: %d\n", + vsi->vsi_num, status); + + ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); +} + +/** + * ice_cfg_sw_lldp - Config switch rules for LLDP packet handling + * @vsi: the VSI being configured + * @tx: bool to determine Tx or Rx rule + * @create: bool to determine create or remove Rule + */ +void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create) +{ + struct ice_fltr_list_entry *list; + struct ice_pf *pf = vsi->back; + LIST_HEAD(tmp_add_list); + enum ice_status status; + + list = devm_kzalloc(&pf->pdev->dev, sizeof(*list), GFP_KERNEL); + if (!list) + return; + + list->fltr_info.lkup_type = ICE_SW_LKUP_ETHERTYPE; + list->fltr_info.vsi_handle = vsi->idx; + list->fltr_info.l_data.ethertype_mac.ethertype = ETH_P_LLDP; + + if (tx) { + list->fltr_info.fltr_act = ICE_DROP_PACKET; + list->fltr_info.flag = ICE_FLTR_TX; + list->fltr_info.src_id = ICE_SRC_ID_VSI; + } else { + list->fltr_info.fltr_act = ICE_FWD_TO_VSI; + list->fltr_info.flag = ICE_FLTR_RX; + list->fltr_info.src_id = ICE_SRC_ID_LPORT; + } + + INIT_LIST_HEAD(&list->list_entry); + list_add(&list->list_entry, &tmp_add_list); + + if (create) + status = ice_add_eth_mac(&pf->hw, &tmp_add_list); + else + status = ice_remove_eth_mac(&pf->hw, &tmp_add_list); + + if (status) + dev_err(&pf->pdev->dev, + "Fail %s %s LLDP rule on VSI %i error: %d\n", + create ? "adding" : "removing", tx ? "TX" : "RX", + vsi->vsi_num, status); + + ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); +} + /** * ice_vsi_setup - Set up a VSI by a given type * @pf: board private structure * @pi: pointer to the port_info instance * @type: VSI type - * @vf_id: defines VF id to which this VSI connects. This field is meant to be + * @vf_id: defines VF ID to which this VSI connects. This field is meant to be * used only for ICE_VSI_VF VSI type. For other VSI types, should * fill-in ICE_INVAL_VFID as input. * @@ -2109,10 +2413,15 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; struct device *dev = &pf->pdev->dev; + enum ice_status status; struct ice_vsi *vsi; int ret, i; - vsi = ice_vsi_alloc(pf, type); + if (type == ICE_VSI_VF) + vsi = ice_vsi_alloc(pf, type, vf_id); + else + vsi = ice_vsi_alloc(pf, type, ICE_INVAL_VFID); + if (!vsi) { dev_err(dev, "could not allocate VSI\n"); return NULL; @@ -2120,6 +2429,9 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, vsi->port_info = pi; vsi->vsw = pf->first_sw; + if (vsi->type == ICE_VSI_PF) + vsi->ethtype = ETH_P_PAUSE; + if (vsi->type == ICE_VSI_VF) vsi->vf_id = vf_id; @@ -2132,7 +2444,7 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, /* set RSS capabilities */ ice_vsi_set_rss_params(vsi); - /* set tc configuration */ + /* set TC configuration */ ice_vsi_set_tc_cfg(vsi); /* create the VSI */ @@ -2150,6 +2462,10 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, if (ret) goto unroll_alloc_q_vector; + ret = ice_vsi_set_q_vectors_reg_idx(vsi); + if (ret) + goto unroll_vector_base; + ret = ice_vsi_alloc_rings(vsi); if (ret) goto unroll_vector_base; @@ -2177,19 +2493,24 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, if (ret) goto unroll_alloc_q_vector; - /* Setup Vector base only during VF init phase or when VF asks - * for more vectors than assigned number. In all other cases, - * assign hw_base_vector to the value given earlier. - */ - if (test_bit(ICE_VF_STATE_CFG_INTR, pf->vf[vf_id].vf_states)) { - ret = ice_vsi_setup_vector_base(vsi); - if (ret) - goto unroll_vector_base; - } else { - vsi->hw_base_vector = pf->vf[vf_id].first_vector_idx; - } + ret = ice_vsi_set_q_vectors_reg_idx(vsi); + if (ret) + goto unroll_vector_base; + pf->q_left_tx -= vsi->alloc_txq; pf->q_left_rx -= vsi->alloc_rxq; + + /* Do not exit if configuring RSS had an issue, at least + * receive traffic on first queue. Hence no need to capture + * return value + */ + if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) + ice_vsi_cfg_rss_lut_key(vsi); + break; + case ICE_VSI_LB: + ret = ice_vsi_alloc_rings(vsi); + if (ret) + goto unroll_vsi_init; break; default: /* clean up the resources and exit */ @@ -2200,22 +2521,42 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, for (i = 0; i < vsi->tc_cfg.numtc; i++) max_txqs[i] = pf->num_lan_tx; - ret = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, - max_txqs); - if (ret) { - dev_info(&pf->pdev->dev, "Failed VSI lan queue config\n"); + status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, + max_txqs); + if (status) { + dev_err(&pf->pdev->dev, + "VSI %d failed lan queue config, error %d\n", + vsi->vsi_num, status); goto unroll_vector_base; } + /* Add switch rule to drop all Tx Flow Control Frames, of look up + * type ETHERTYPE from VSIs, and restrict malicious VF from sending + * out PAUSE or PFC frames. If enabled, FW can still send FC frames. + * The rule is added once for PF VSI in order to create appropriate + * recipe, since VSI/VSI list is ignored with drop action... + * Also add rules to handle LLDP Tx and Rx packets. Tx LLDP packets + * need to be dropped so that VFs cannot send LLDP packets to reconfig + * DCB settings in the HW. Also, if the FW DCBX engine is not running + * then Rx LLDP packets need to be redirected up the stack. + */ + if (vsi->type == ICE_VSI_PF) { + ice_vsi_add_rem_eth_mac(vsi, true); + + /* Tx LLDP packets */ + ice_cfg_sw_lldp(vsi, true, true); + + /* Rx LLDP packets */ + if (!test_bit(ICE_FLAG_ENABLE_FW_LLDP, pf->flags)) + ice_cfg_sw_lldp(vsi, false, true); + } + return vsi; unroll_vector_base: /* reclaim SW interrupts back to the common pool */ - ice_free_res(vsi->back->sw_irq_tracker, vsi->sw_base_vector, vsi->idx); + ice_free_res(pf->irq_tracker, vsi->base_vector, vsi->idx); pf->num_avail_sw_msix += vsi->num_q_vectors; - /* reclaim HW interrupt back to the common pool */ - ice_free_res(vsi->back->hw_irq_tracker, vsi->hw_base_vector, vsi->idx); - pf->num_avail_hw_msix += vsi->num_q_vectors; unroll_alloc_q_vector: ice_vsi_free_q_vectors(vsi); unroll_vsi_init: @@ -2236,17 +2577,17 @@ unroll_get_qs: static void ice_vsi_release_msix(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; - u16 vector = vsi->hw_base_vector; struct ice_hw *hw = &pf->hw; u32 txq = 0; u32 rxq = 0; int i, q; - for (i = 0; i < vsi->num_q_vectors; i++, vector++) { + for (i = 0; i < vsi->num_q_vectors; i++) { struct ice_q_vector *q_vector = vsi->q_vectors[i]; + u16 reg_idx = q_vector->reg_idx; - wr32(hw, GLINT_ITR(ICE_IDX_ITR0, vector), 0); - wr32(hw, GLINT_ITR(ICE_IDX_ITR1, vector), 0); + wr32(hw, GLINT_ITR(ICE_IDX_ITR0, reg_idx), 0); + wr32(hw, GLINT_ITR(ICE_IDX_ITR1, reg_idx), 0); for (q = 0; q < q_vector->num_ring_tx; q++) { wr32(hw, QINT_TQCTL(vsi->txq_map[txq]), 0); txq++; @@ -2268,7 +2609,7 @@ static void ice_vsi_release_msix(struct ice_vsi *vsi) void ice_vsi_free_irq(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; - int base = vsi->sw_base_vector; + int base = vsi->base_vector; if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { int i; @@ -2281,7 +2622,7 @@ void ice_vsi_free_irq(struct ice_vsi *vsi) return; vsi->irqs_ready = false; - for (i = 0; i < vsi->num_q_vectors; i++) { + ice_for_each_q_vector(vsi, i) { u16 vector = i + base; int irq_num; @@ -2364,11 +2705,11 @@ int ice_free_res(struct ice_res_tracker *res, u16 index, u16 id) int count = 0; int i; - if (!res || index >= res->num_entries) + if (!res || index >= res->end) return -EINVAL; id |= ICE_RES_VALID_BIT; - for (i = index; i < res->num_entries && res->list[i] == id; i++) { + for (i = index; i < res->end && res->list[i] == id; i++) { res->list[i] = 0; count++; } @@ -2386,10 +2727,9 @@ int ice_free_res(struct ice_res_tracker *res, u16 index, u16 id) */ static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) { - int start = res->search_hint; - int end = start; + int start = 0, end = 0; - if ((start + needed) > res->num_entries) + if (needed > res->end) return -ENOMEM; id |= ICE_RES_VALID_BIT; @@ -2398,7 +2738,7 @@ static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) /* skip already allocated entries */ if (res->list[end++] & ICE_RES_VALID_BIT) { start = end; - if ((start + needed) > res->num_entries) + if ((start + needed) > res->end) break; } @@ -2409,13 +2749,9 @@ static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) while (i != end) res->list[i++] = id; - if (end == res->num_entries) - end = 0; - - res->search_hint = end; return start; } - } while (1); + } while (end < res->end); return -ENOMEM; } @@ -2427,16 +2763,11 @@ static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) * @needed: size of the block needed * @id: identifier to track owner * - * Returns the base item index of the block, or -ENOMEM for error - * The search_hint trick and lack of advanced fit-finding only works - * because we're highly likely to have all the same sized requests. - * Linear search time and any fragmentation should be minimal. + * Returns the base item index of the block, or negative for error */ int ice_get_res(struct ice_pf *pf, struct ice_res_tracker *res, u16 needed, u16 id) { - int ret; - if (!res || !pf) return -EINVAL; @@ -2447,16 +2778,7 @@ ice_get_res(struct ice_pf *pf, struct ice_res_tracker *res, u16 needed, u16 id) return -EINVAL; } - /* search based on search_hint */ - ret = ice_search_res(res, needed, id); - - if (ret < 0) { - /* previous search failed. Reset search hint and try again */ - res->search_hint = 0; - ret = ice_search_res(res, needed, id); - } - - return ret; + return ice_search_res(res, needed, id); } /** @@ -2465,7 +2787,7 @@ ice_get_res(struct ice_pf *pf, struct ice_res_tracker *res, u16 needed, u16 id) */ void ice_vsi_dis_irq(struct ice_vsi *vsi) { - int base = vsi->sw_base_vector; + int base = vsi->base_vector; struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; u32 val; @@ -2500,17 +2822,32 @@ void ice_vsi_dis_irq(struct ice_vsi *vsi) /* disable each interrupt */ if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - for (i = vsi->hw_base_vector; - i < (vsi->num_q_vectors + vsi->hw_base_vector); i++) - wr32(hw, GLINT_DYN_CTL(i), 0); + ice_for_each_q_vector(vsi, i) + wr32(hw, GLINT_DYN_CTL(vsi->q_vectors[i]->reg_idx), 0); ice_flush(hw); - for (i = 0; i < vsi->num_q_vectors; i++) + + ice_for_each_q_vector(vsi, i) synchronize_irq(pf->msix_entries[i + base].vector); } } /** + * ice_napi_del - Remove NAPI handler for the VSI + * @vsi: VSI for which NAPI handler is to be removed + */ +void ice_napi_del(struct ice_vsi *vsi) +{ + int v_idx; + + if (!vsi->netdev) + return; + + ice_for_each_q_vector(vsi, v_idx) + netif_napi_del(&vsi->q_vectors[v_idx]->napi); +} + +/** * ice_vsi_release - Delete a VSI and free its resources * @vsi: the VSI being removed * @@ -2518,59 +2855,61 @@ void ice_vsi_dis_irq(struct ice_vsi *vsi) */ int ice_vsi_release(struct ice_vsi *vsi) { - struct ice_vf *vf = NULL; struct ice_pf *pf; if (!vsi->back) return -ENODEV; pf = vsi->back; - if (vsi->type == ICE_VSI_VF) - vf = &pf->vf[vsi->vf_id]; - /* do not unregister and free netdevs while driver is in the reset - * recovery pending state. Since reset/rebuild happens through PF - * service task workqueue, its not a good idea to unregister netdev - * that is associated to the PF that is running the work queue items - * currently. This is done to avoid check_flush_dependency() warning - * on this wq + /* do not unregister while driver is in the reset recovery pending + * state. Since reset/rebuild happens through PF service task workqueue, + * it's not a good idea to unregister netdev that is associated to the + * PF that is running the work queue items currently. This is done to + * avoid check_flush_dependency() warning on this wq */ - if (vsi->netdev && !ice_is_reset_in_progress(pf->state)) { - ice_napi_del(vsi); + if (vsi->netdev && !ice_is_reset_in_progress(pf->state)) unregister_netdev(vsi->netdev); - free_netdev(vsi->netdev); - vsi->netdev = NULL; - } if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) ice_rss_clean(vsi); /* Disable VSI and free resources */ - ice_vsi_dis_irq(vsi); + if (vsi->type != ICE_VSI_LB) + ice_vsi_dis_irq(vsi); ice_vsi_close(vsi); - /* reclaim interrupt vectors back to PF */ + /* SR-IOV determines needed MSIX resources all at once instead of per + * VSI since when VFs are spawned we know how many VFs there are and how + * many interrupts each VF needs. SR-IOV MSIX resources are also + * cleared in the same manner. + */ if (vsi->type != ICE_VSI_VF) { /* reclaim SW interrupts back to the common pool */ - ice_free_res(vsi->back->sw_irq_tracker, vsi->sw_base_vector, - vsi->idx); + ice_free_res(pf->irq_tracker, vsi->base_vector, vsi->idx); pf->num_avail_sw_msix += vsi->num_q_vectors; - /* reclaim HW interrupts back to the common pool */ - ice_free_res(vsi->back->hw_irq_tracker, vsi->hw_base_vector, - vsi->idx); - pf->num_avail_hw_msix += vsi->num_q_vectors; - } else if (test_bit(ICE_VF_STATE_CFG_INTR, vf->vf_states)) { - /* Reclaim VF resources back only while freeing all VFs or - * vector reassignment is requested + } + + if (vsi->type == ICE_VSI_PF) { + ice_vsi_add_rem_eth_mac(vsi, false); + ice_cfg_sw_lldp(vsi, true, false); + /* The Rx rule will only exist to remove if the LLDP FW + * engine is currently stopped */ - ice_free_res(vsi->back->hw_irq_tracker, vf->first_vector_idx, - vsi->idx); - pf->num_avail_hw_msix += pf->num_vf_msix; + if (!test_bit(ICE_FLAG_ENABLE_FW_LLDP, pf->flags)) + ice_cfg_sw_lldp(vsi, false, false); } ice_remove_vsi_fltr(&pf->hw, vsi->idx); ice_rm_vsi_lan_cfg(vsi->port_info, vsi->idx); ice_vsi_delete(vsi); ice_vsi_free_q_vectors(vsi); + + /* make sure unregister_netdev() was called by checking __ICE_DOWN */ + if (vsi->netdev && test_bit(__ICE_DOWN, vsi->state)) { + free_netdev(vsi->netdev); + vsi->netdev = NULL; + } + ice_vsi_clear_rings(vsi); ice_vsi_put_qs(vsi); @@ -2596,6 +2935,8 @@ int ice_vsi_release(struct ice_vsi *vsi) int ice_vsi_rebuild(struct ice_vsi *vsi) { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; + struct ice_vf *vf = NULL; + enum ice_status status; struct ice_pf *pf; int ret, i; @@ -2603,16 +2944,31 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) return -EINVAL; pf = vsi->back; + if (vsi->type == ICE_VSI_VF) + vf = &pf->vf[vsi->vf_id]; + ice_rm_vsi_lan_cfg(vsi->port_info, vsi->idx); ice_vsi_free_q_vectors(vsi); - ice_free_res(vsi->back->sw_irq_tracker, vsi->sw_base_vector, vsi->idx); - ice_free_res(vsi->back->hw_irq_tracker, vsi->hw_base_vector, vsi->idx); - vsi->sw_base_vector = 0; - vsi->hw_base_vector = 0; + + /* SR-IOV determines needed MSIX resources all at once instead of per + * VSI since when VFs are spawned we know how many VFs there are and how + * many interrupts each VF needs. SR-IOV MSIX resources are also + * cleared in the same manner. + */ + if (vsi->type != ICE_VSI_VF) { + /* reclaim SW interrupts back to the common pool */ + ice_free_res(pf->irq_tracker, vsi->base_vector, vsi->idx); + pf->num_avail_sw_msix += vsi->num_q_vectors; + vsi->base_vector = 0; + } + ice_vsi_clear_rings(vsi); - ice_vsi_free_arrays(vsi, false); - ice_dev_onetime_setup(&vsi->back->hw); - ice_vsi_set_num_qs(vsi); + ice_vsi_free_arrays(vsi); + ice_dev_onetime_setup(&pf->hw); + if (vsi->type == ICE_VSI_VF) + ice_vsi_set_num_qs(vsi, vf->vf_id); + else + ice_vsi_set_num_qs(vsi, ICE_INVAL_VFID); ice_vsi_set_tc_cfg(vsi); /* Initialize VSI struct elements and create VSI in FW */ @@ -2620,7 +2976,7 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret < 0) goto err_vsi; - ret = ice_vsi_alloc_arrays(vsi, false); + ret = ice_vsi_alloc_arrays(vsi); if (ret < 0) goto err_vsi; @@ -2630,7 +2986,7 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret) goto err_rings; - ret = ice_vsi_setup_vector_base(vsi); + ret = ice_vsi_set_q_vectors_reg_idx(vsi); if (ret) goto err_vectors; @@ -2643,7 +2999,7 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) * receive traffic on first queue. Hence no need to capture * return value */ - if (test_bit(ICE_FLAG_RSS_ENA, vsi->back->flags)) + if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) ice_vsi_cfg_rss_lut_key(vsi); break; case ICE_VSI_VF: @@ -2655,12 +3011,16 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret) goto err_vectors; + ret = ice_vsi_set_q_vectors_reg_idx(vsi); + if (ret) + goto err_vectors; + ret = ice_vsi_alloc_rings(vsi); if (ret) goto err_vectors; - vsi->back->q_left_tx -= vsi->alloc_txq; - vsi->back->q_left_rx -= vsi->alloc_rxq; + pf->q_left_tx -= vsi->alloc_txq; + pf->q_left_rx -= vsi->alloc_rxq; break; default: break; @@ -2670,11 +3030,12 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) for (i = 0; i < vsi->tc_cfg.numtc; i++) max_txqs[i] = pf->num_lan_tx; - ret = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, - max_txqs); - if (ret) { - dev_info(&vsi->back->pdev->dev, - "Failed VSI lan queue config\n"); + status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, + max_txqs); + if (status) { + dev_err(&pf->pdev->dev, + "VSI %d failed lan queue config, error %d\n", + vsi->vsi_num, status); goto err_vectors; } return 0; @@ -2690,13 +3051,13 @@ err_rings: } err_vsi: ice_vsi_clear(vsi); - set_bit(__ICE_RESET_FAILED, vsi->back->state); + set_bit(__ICE_RESET_FAILED, pf->state); return ret; } /** * ice_is_reset_in_progress - check for a reset in progress - * @state: pf state field + * @state: PF state field */ bool ice_is_reset_in_progress(unsigned long *state) { @@ -2705,3 +3066,125 @@ bool ice_is_reset_in_progress(unsigned long *state) test_bit(__ICE_CORER_REQ, state) || test_bit(__ICE_GLOBR_REQ, state); } + +#ifdef CONFIG_DCB +/** + * ice_vsi_update_q_map - update our copy of the VSI info with new queue map + * @vsi: VSI being configured + * @ctx: the context buffer returned from AQ VSI update command + */ +static void ice_vsi_update_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctx) +{ + vsi->info.mapping_flags = ctx->info.mapping_flags; + memcpy(&vsi->info.q_mapping, &ctx->info.q_mapping, + sizeof(vsi->info.q_mapping)); + memcpy(&vsi->info.tc_mapping, ctx->info.tc_mapping, + sizeof(vsi->info.tc_mapping)); +} + +/** + * ice_vsi_cfg_netdev_tc - Setup the netdev TC configuration + * @vsi: the VSI being configured + * @ena_tc: TC map to be enabled + */ +static void ice_vsi_cfg_netdev_tc(struct ice_vsi *vsi, u8 ena_tc) +{ + struct net_device *netdev = vsi->netdev; + struct ice_pf *pf = vsi->back; + struct ice_dcbx_cfg *dcbcfg; + u8 netdev_tc; + int i; + + if (!netdev) + return; + + if (!ena_tc) { + netdev_reset_tc(netdev); + return; + } + + if (netdev_set_num_tc(netdev, vsi->tc_cfg.numtc)) + return; + + dcbcfg = &pf->hw.port_info->local_dcbx_cfg; + + ice_for_each_traffic_class(i) + if (vsi->tc_cfg.ena_tc & BIT(i)) + netdev_set_tc_queue(netdev, + vsi->tc_cfg.tc_info[i].netdev_tc, + vsi->tc_cfg.tc_info[i].qcount_tx, + vsi->tc_cfg.tc_info[i].qoffset); + + for (i = 0; i < ICE_MAX_USER_PRIORITY; i++) { + u8 ets_tc = dcbcfg->etscfg.prio_table[i]; + + /* Get the mapped netdev TC# for the UP */ + netdev_tc = vsi->tc_cfg.tc_info[ets_tc].netdev_tc; + netdev_set_prio_tc_map(netdev, i, netdev_tc); + } +} + +/** + * ice_vsi_cfg_tc - Configure VSI Tx Sched for given TC map + * @vsi: VSI to be configured + * @ena_tc: TC bitmap + * + * VSI queues expected to be quiesced before calling this function + */ +int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc) +{ + u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; + struct ice_vsi_ctx *ctx; + struct ice_pf *pf = vsi->back; + enum ice_status status; + int i, ret = 0; + u8 num_tc = 0; + + ice_for_each_traffic_class(i) { + /* build bitmap of enabled TCs */ + if (ena_tc & BIT(i)) + num_tc++; + /* populate max_txqs per TC */ + max_txqs[i] = pf->num_lan_tx; + } + + vsi->tc_cfg.ena_tc = ena_tc; + vsi->tc_cfg.numtc = num_tc; + + ctx = devm_kzalloc(&pf->pdev->dev, sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->vf_num = 0; + ctx->info = vsi->info; + + ice_vsi_setup_q_map(vsi, ctx); + + /* must to indicate which section of VSI context are being modified */ + ctx->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_RXQ_MAP_VALID); + status = ice_update_vsi(&pf->hw, vsi->idx, ctx, NULL); + if (status) { + dev_info(&pf->pdev->dev, "Failed VSI Update\n"); + ret = -EIO; + goto out; + } + + status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, + max_txqs); + + if (status) { + dev_err(&pf->pdev->dev, + "VSI %d failed TC config, error %d\n", + vsi->vsi_num, status); + ret = -EIO; + goto out; + } + ice_vsi_update_q_map(vsi, ctx); + vsi->info.valid_sections = 0; + + ice_vsi_cfg_netdev_tc(vsi, ena_tc); +out: + devm_kfree(&pf->pdev->dev, ctx); + return ret; +} +#endif /* CONFIG_DCB */ diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h index 7988a53729a9..6e43ef03bfc3 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_lib.h @@ -19,6 +19,14 @@ int ice_vsi_cfg_lan_txqs(struct ice_vsi *vsi); void ice_vsi_cfg_msix(struct ice_vsi *vsi); +#ifdef CONFIG_PCI_IOV +void +ice_cfg_txq_interrupt(struct ice_vsi *vsi, u16 txq, u16 msix_idx, u16 itr_idx); + +void +ice_cfg_rxq_interrupt(struct ice_vsi *vsi, u16 rxq, u16 msix_idx, u16 itr_idx); +#endif /* CONFIG_PCI_IOV */ + int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid); int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid); @@ -35,16 +43,24 @@ int ice_vsi_stop_lan_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, u16 rel_vmvf_num); -int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena); +int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc); + +void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create); void ice_vsi_delete(struct ice_vsi *vsi); int ice_vsi_clear(struct ice_vsi *vsi); +#ifdef CONFIG_DCB +int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc); +#endif /* CONFIG_DCB */ + struct ice_vsi * ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, enum ice_vsi_type type, u16 vf_id); +void ice_napi_del(struct ice_vsi *vsi); + int ice_vsi_release(struct ice_vsi *vsi); void ice_vsi_close(struct ice_vsi *vsi); @@ -60,8 +76,14 @@ bool ice_is_reset_in_progress(unsigned long *state); void ice_vsi_free_q_vectors(struct ice_vsi *vsi); +void ice_trigger_sw_intr(struct ice_hw *hw, struct ice_q_vector *q_vector); + void ice_vsi_put_qs(struct ice_vsi *vsi); +#ifdef CONFIG_DCB +void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi); +#endif /* CONFIG_DCB */ + void ice_vsi_dis_irq(struct ice_vsi *vsi); void ice_vsi_free_irq(struct ice_vsi *vsi); @@ -70,8 +92,7 @@ void ice_vsi_free_rx_rings(struct ice_vsi *vsi); void ice_vsi_free_tx_rings(struct ice_vsi *vsi); -int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc); - int ice_vsi_manage_rss_lut(struct ice_vsi *vsi, bool ena); +u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran); #endif /* !_ICE_LIB_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 47cc3f905b7f..28ec0d57941d 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -7,8 +7,9 @@ #include "ice.h" #include "ice_lib.h" +#include "ice_dcb_lib.h" -#define DRV_VERSION "0.7.2-k" +#define DRV_VERSION "0.7.4-k" #define DRV_SUMMARY "Intel(R) Ethernet Connection E800 Series Linux Driver" const char ice_drv_ver[] = DRV_VERSION; static const char ice_driver_string[] = DRV_SUMMARY; @@ -30,7 +31,6 @@ MODULE_PARM_DESC(debug, "netif level (0=none,...,16=all)"); static struct workqueue_struct *ice_wq; static const struct net_device_ops ice_netdev_ops; -static void ice_pf_dis_all_vsi(struct ice_pf *pf); static void ice_rebuild(struct ice_pf *pf); static void ice_vsi_release_all(struct ice_pf *pf); @@ -61,9 +61,10 @@ static u32 ice_get_tx_pending(struct ice_ring *ring) static void ice_check_for_hang_subtask(struct ice_pf *pf) { struct ice_vsi *vsi = NULL; + struct ice_hw *hw; unsigned int i; - u32 v, v_idx; int packets; + u32 v; ice_for_each_vsi(pf, v) if (pf->vsi[v] && pf->vsi[v]->type == ICE_VSI_PF) { @@ -77,12 +78,12 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) if (!(vsi->netdev && netif_carrier_ok(vsi->netdev))) return; + hw = &vsi->back->hw; + for (i = 0; i < vsi->num_txq; i++) { struct ice_ring *tx_ring = vsi->tx_rings[i]; if (tx_ring && tx_ring->desc) { - int itr = ICE_ITR_NONE; - /* If packet counter has not changed the queue is * likely stalled, so force an interrupt for this * queue. @@ -93,12 +94,7 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) packets = tx_ring->stats.pkts & INT_MAX; if (tx_ring->tx_stats.prev_pkt == packets) { /* Trigger sw interrupt to revive the queue */ - v_idx = tx_ring->q_vector->v_idx; - wr32(&vsi->back->hw, - GLINT_DYN_CTL(vsi->hw_base_vector + v_idx), - (itr << GLINT_DYN_CTL_ITR_INDX_S) | - GLINT_DYN_CTL_SWINT_TRIG_M | - GLINT_DYN_CTL_INTENA_MSK_M); + ice_trigger_sw_intr(hw, tx_ring->q_vector); continue; } @@ -113,14 +109,75 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) } /** - * ice_add_mac_to_sync_list - creates list of mac addresses to be synced + * ice_init_mac_fltr - Set initial MAC filters + * @pf: board private structure + * + * Set initial set of MAC filters for PF VSI; configure filters for permanent + * address and broadcast address. If an error is encountered, netdevice will be + * unregistered. + */ +static int ice_init_mac_fltr(struct ice_pf *pf) +{ + LIST_HEAD(tmp_add_list); + u8 broadcast[ETH_ALEN]; + struct ice_vsi *vsi; + int status; + + vsi = ice_find_vsi_by_type(pf, ICE_VSI_PF); + if (!vsi) + return -EINVAL; + + /* To add a MAC filter, first add the MAC to a list and then + * pass the list to ice_add_mac. + */ + + /* Add a unicast MAC filter so the VSI can get its packets */ + status = ice_add_mac_to_list(vsi, &tmp_add_list, + vsi->port_info->mac.perm_addr); + if (status) + goto unregister; + + /* VSI needs to receive broadcast traffic, so add the broadcast + * MAC address to the list as well. + */ + eth_broadcast_addr(broadcast); + status = ice_add_mac_to_list(vsi, &tmp_add_list, broadcast); + if (status) + goto free_mac_list; + + /* Program MAC filters for entries in tmp_add_list */ + status = ice_add_mac(&pf->hw, &tmp_add_list); + if (status) + status = -ENOMEM; + +free_mac_list: + ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); + +unregister: + /* We aren't useful with no MAC filters, so unregister if we + * had an error + */ + if (status && vsi->netdev->reg_state == NETREG_REGISTERED) { + dev_err(&pf->pdev->dev, + "Could not add MAC filters error %d. Unregistering device\n", + status); + unregister_netdev(vsi->netdev); + free_netdev(vsi->netdev); + vsi->netdev = NULL; + } + + return status; +} + +/** + * ice_add_mac_to_sync_list - creates list of MAC addresses to be synced * @netdev: the net device on which the sync is happening - * @addr: mac address to sync + * @addr: MAC address to sync * * This is a callback function which is called by the in kernel device sync * functions (like __dev_uc_sync, __dev_mc_sync, etc). This function only * populates the tmp_sync_list, which is later used by ice_add_mac to add the - * mac filters from the hardware. + * MAC filters from the hardware. */ static int ice_add_mac_to_sync_list(struct net_device *netdev, const u8 *addr) { @@ -134,14 +191,14 @@ static int ice_add_mac_to_sync_list(struct net_device *netdev, const u8 *addr) } /** - * ice_add_mac_to_unsync_list - creates list of mac addresses to be unsynced + * ice_add_mac_to_unsync_list - creates list of MAC addresses to be unsynced * @netdev: the net device on which the unsync is happening - * @addr: mac address to unsync + * @addr: MAC address to unsync * * This is a callback function which is called by the in kernel device unsync * functions (like __dev_uc_unsync, __dev_mc_unsync, etc). This function only * populates the tmp_unsync_list, which is later used by ice_remove_mac to - * delete the mac filters from the hardware. + * delete the MAC filters from the hardware. */ static int ice_add_mac_to_unsync_list(struct net_device *netdev, const u8 *addr) { @@ -168,6 +225,39 @@ static bool ice_vsi_fltr_changed(struct ice_vsi *vsi) } /** + * ice_cfg_promisc - Enable or disable promiscuous mode for a given PF + * @vsi: the VSI being configured + * @promisc_m: mask of promiscuous config bits + * @set_promisc: enable or disable promisc flag request + * + */ +static int ice_cfg_promisc(struct ice_vsi *vsi, u8 promisc_m, bool set_promisc) +{ + struct ice_hw *hw = &vsi->back->hw; + enum ice_status status = 0; + + if (vsi->type != ICE_VSI_PF) + return 0; + + if (vsi->vlan_ena) { + status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m, + set_promisc); + } else { + if (set_promisc) + status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m, + 0); + else + status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m, + 0); + } + + if (status) + return -EIO; + + return 0; +} + +/** * ice_vsi_sync_fltr - Update the VSI filter list to the HW * @vsi: ptr to the VSI * @@ -182,6 +272,7 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) struct ice_hw *hw = &pf->hw; enum ice_status status = 0; u32 changed_flags = 0; + u8 promisc_m; int err = 0; if (!vsi->netdev) @@ -211,7 +302,7 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) netif_addr_unlock_bh(netdev); } - /* Remove mac addresses in the unsync list */ + /* Remove MAC addresses in the unsync list */ status = ice_remove_mac(hw, &vsi->tmp_unsync_list); ice_free_fltr_list(dev, &vsi->tmp_unsync_list); if (status) { @@ -223,12 +314,16 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) } } - /* Add mac addresses in the sync list */ + /* Add MAC addresses in the sync list */ status = ice_add_mac(hw, &vsi->tmp_sync_list); ice_free_fltr_list(dev, &vsi->tmp_sync_list); - if (status) { + /* If filter is added successfully or already exists, do not go into + * 'if' condition and report it as error. Instead continue processing + * rest of the function. + */ + if (status && status != ICE_ERR_ALREADY_EXISTS) { netdev_err(netdev, "Failed to add MAC filters\n"); - /* If there is no more space for new umac filters, vsi + /* If there is no more space for new umac filters, VSI * should go into promiscuous mode. There should be some * space reserved for promiscuous filters. */ @@ -245,49 +340,56 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) } } /* check for changes in promiscuous modes */ - if (changed_flags & IFF_ALLMULTI) - netdev_warn(netdev, "Unsupported configuration\n"); + if (changed_flags & IFF_ALLMULTI) { + if (vsi->current_netdev_flags & IFF_ALLMULTI) { + if (vsi->vlan_ena) + promisc_m = ICE_MCAST_VLAN_PROMISC_BITS; + else + promisc_m = ICE_MCAST_PROMISC_BITS; + + err = ice_cfg_promisc(vsi, promisc_m, true); + if (err) { + netdev_err(netdev, "Error setting Multicast promiscuous mode on VSI %i\n", + vsi->vsi_num); + vsi->current_netdev_flags &= ~IFF_ALLMULTI; + goto out_promisc; + } + } else if (!(vsi->current_netdev_flags & IFF_ALLMULTI)) { + if (vsi->vlan_ena) + promisc_m = ICE_MCAST_VLAN_PROMISC_BITS; + else + promisc_m = ICE_MCAST_PROMISC_BITS; + + err = ice_cfg_promisc(vsi, promisc_m, false); + if (err) { + netdev_err(netdev, "Error clearing Multicast promiscuous mode on VSI %i\n", + vsi->vsi_num); + vsi->current_netdev_flags |= IFF_ALLMULTI; + goto out_promisc; + } + } + } if (((changed_flags & IFF_PROMISC) || promisc_forced_on) || test_bit(ICE_VSI_FLAG_PROMISC_CHANGED, vsi->flags)) { clear_bit(ICE_VSI_FLAG_PROMISC_CHANGED, vsi->flags); if (vsi->current_netdev_flags & IFF_PROMISC) { - /* Apply TX filter rule to get traffic from VMs */ - status = ice_cfg_dflt_vsi(hw, vsi->idx, true, - ICE_FLTR_TX); - if (status) { - netdev_err(netdev, "Error setting default VSI %i tx rule\n", - vsi->vsi_num); - vsi->current_netdev_flags &= ~IFF_PROMISC; - err = -EIO; - goto out_promisc; - } - /* Apply RX filter rule to get traffic from wire */ + /* Apply Rx filter rule to get traffic from wire */ status = ice_cfg_dflt_vsi(hw, vsi->idx, true, ICE_FLTR_RX); if (status) { - netdev_err(netdev, "Error setting default VSI %i rx rule\n", + netdev_err(netdev, "Error setting default VSI %i Rx rule\n", vsi->vsi_num); vsi->current_netdev_flags &= ~IFF_PROMISC; err = -EIO; goto out_promisc; } } else { - /* Clear TX filter rule to stop traffic from VMs */ - status = ice_cfg_dflt_vsi(hw, vsi->idx, false, - ICE_FLTR_TX); - if (status) { - netdev_err(netdev, "Error clearing default VSI %i tx rule\n", - vsi->vsi_num); - vsi->current_netdev_flags |= IFF_PROMISC; - err = -EIO; - goto out_promisc; - } - /* Clear RX filter to remove traffic from wire */ + /* Clear Rx filter to remove traffic from wire */ status = ice_cfg_dflt_vsi(hw, vsi->idx, false, ICE_FLTR_RX); if (status) { - netdev_err(netdev, "Error clearing default VSI %i rx rule\n", + netdev_err(netdev, "Error clearing default VSI %i Rx rule\n", vsi->vsi_num); vsi->current_netdev_flags |= IFF_PROMISC; err = -EIO; @@ -322,7 +424,7 @@ static void ice_sync_fltr_subtask(struct ice_pf *pf) clear_bit(ICE_FLAG_FLTR_SYNC, pf->flags); - for (v = 0; v < pf->num_alloc_vsi; v++) + ice_for_each_vsi(pf, v) if (pf->vsi[v] && ice_vsi_fltr_changed(pf->vsi[v]) && ice_vsi_sync_fltr(pf->vsi[v])) { /* come back and try again later */ @@ -332,6 +434,51 @@ static void ice_sync_fltr_subtask(struct ice_pf *pf) } /** + * ice_dis_vsi - pause a VSI + * @vsi: the VSI being paused + * @locked: is the rtnl_lock already held + */ +static void ice_dis_vsi(struct ice_vsi *vsi, bool locked) +{ + if (test_bit(__ICE_DOWN, vsi->state)) + return; + + set_bit(__ICE_NEEDS_RESTART, vsi->state); + + if (vsi->type == ICE_VSI_PF && vsi->netdev) { + if (netif_running(vsi->netdev)) { + if (!locked) { + rtnl_lock(); + vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); + rtnl_unlock(); + } else { + vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); + } + } else { + ice_vsi_close(vsi); + } + } +} + +/** + * ice_pf_dis_all_vsi - Pause all VSIs on a PF + * @pf: the PF + * @locked: is the rtnl_lock already held + */ +#ifdef CONFIG_DCB +void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked) +#else +static void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked) +#endif /* CONFIG_DCB */ +{ + int v; + + ice_for_each_vsi(pf, v) + if (pf->vsi[v]) + ice_dis_vsi(pf->vsi[v], locked); +} + +/** * ice_prepare_for_reset - prep for the core to reset * @pf: board private structure * @@ -342,12 +489,16 @@ ice_prepare_for_reset(struct ice_pf *pf) { struct ice_hw *hw = &pf->hw; + /* already prepared for reset */ + if (test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) + return; + /* Notify VFs of impending reset */ if (ice_check_sq_alive(hw, &hw->mailboxq)) ice_vc_notify_reset(pf); /* disable the VSIs and their queues that are not already DOWN */ - ice_pf_dis_all_vsi(pf); + ice_pf_dis_all_vsi(pf, false); if (hw->port_info) ice_sched_clear_port(hw->port_info); @@ -394,6 +545,7 @@ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type) ice_rebuild(pf); clear_bit(__ICE_PREPARED_FOR_RESET, pf->state); clear_bit(__ICE_PFR_REQ, pf->state); + ice_reset_all_vfs(pf, true); } } @@ -416,10 +568,15 @@ static void ice_reset_subtask(struct ice_pf *pf) * for the reset now), poll for reset done, rebuild and return. */ if (test_bit(__ICE_RESET_OICR_RECV, pf->state)) { - clear_bit(__ICE_GLOBR_RECV, pf->state); - clear_bit(__ICE_CORER_RECV, pf->state); - if (!test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) - ice_prepare_for_reset(pf); + /* Perform the largest reset requested */ + if (test_and_clear_bit(__ICE_CORER_RECV, pf->state)) + reset_type = ICE_RESET_CORER; + if (test_and_clear_bit(__ICE_GLOBR_RECV, pf->state)) + reset_type = ICE_RESET_GLOBR; + /* return if no valid reset type requested */ + if (reset_type == ICE_RESET_INVAL) + return; + ice_prepare_for_reset(pf); /* make sure we are ready to rebuild */ if (ice_check_reset(&pf->hw)) { @@ -429,13 +586,14 @@ static void ice_reset_subtask(struct ice_pf *pf) pf->hw.reset_ongoing = false; ice_rebuild(pf); /* clear bit to resume normal operations, but - * ICE_NEEDS_RESTART bit is set incase rebuild failed + * ICE_NEEDS_RESTART bit is set in case rebuild failed */ clear_bit(__ICE_RESET_OICR_RECV, pf->state); clear_bit(__ICE_PREPARED_FOR_RESET, pf->state); clear_bit(__ICE_PFR_REQ, pf->state); clear_bit(__ICE_CORER_REQ, pf->state); clear_bit(__ICE_GLOBR_REQ, pf->state); + ice_reset_all_vfs(pf, true); } return; @@ -466,9 +624,16 @@ static void ice_reset_subtask(struct ice_pf *pf) */ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) { + struct ice_aqc_get_phy_caps_data *caps; + enum ice_status status; + const char *fec_req; const char *speed; + const char *fec; const char *fc; + if (!vsi) + return; + if (vsi->current_isup == isup) return; @@ -480,6 +645,12 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) } switch (vsi->port_info->phy.link_info.link_speed) { + case ICE_AQ_LINK_SPEED_100GB: + speed = "100 G"; + break; + case ICE_AQ_LINK_SPEED_50GB: + speed = "50 G"; + break; case ICE_AQ_LINK_SPEED_40GB: speed = "40 G"; break; @@ -511,39 +682,82 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) switch (vsi->port_info->fc.current_mode) { case ICE_FC_FULL: - fc = "RX/TX"; + fc = "Rx/Tx"; break; case ICE_FC_TX_PAUSE: - fc = "TX"; + fc = "Tx"; break; case ICE_FC_RX_PAUSE: - fc = "RX"; + fc = "Rx"; + break; + case ICE_FC_NONE: + fc = "None"; break; default: fc = "Unknown"; break; } - netdev_info(vsi->netdev, "NIC Link is up %sbps, Flow Control: %s\n", - speed, fc); + /* Get FEC mode based on negotiated link info */ + switch (vsi->port_info->phy.link_info.fec_info) { + case ICE_AQ_LINK_25G_RS_528_FEC_EN: + /* fall through */ + case ICE_AQ_LINK_25G_RS_544_FEC_EN: + fec = "RS-FEC"; + break; + case ICE_AQ_LINK_25G_KR_FEC_EN: + fec = "FC-FEC/BASE-R"; + break; + default: + fec = "NONE"; + break; + } + + /* Get FEC mode requested based on PHY caps last SW configuration */ + caps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*caps), GFP_KERNEL); + if (!caps) { + fec_req = "Unknown"; + goto done; + } + + status = ice_aq_get_phy_caps(vsi->port_info, false, + ICE_AQC_REPORT_SW_CFG, caps, NULL); + if (status) + netdev_info(vsi->netdev, "Get phy capability failed.\n"); + + if (caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_528_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_544_REQ) + fec_req = "RS-FEC"; + else if (caps->link_fec_options & ICE_AQC_PHY_FEC_10G_KR_40G_KR4_REQ || + caps->link_fec_options & ICE_AQC_PHY_FEC_25G_KR_REQ) + fec_req = "FC-FEC/BASE-R"; + else + fec_req = "NONE"; + + devm_kfree(&vsi->back->pdev->dev, caps); + +done: + netdev_info(vsi->netdev, "NIC Link is up %sbps, Requested FEC: %s, FEC: %s, Flow Control: %s\n", + speed, fec_req, fec, fc); } /** - * ice_vsi_link_event - update the vsi's netdev - * @vsi: the vsi on which the link event occurred - * @link_up: whether or not the vsi needs to be set up or down + * ice_vsi_link_event - update the VSI's netdev + * @vsi: the VSI on which the link event occurred + * @link_up: whether or not the VSI needs to be set up or down */ static void ice_vsi_link_event(struct ice_vsi *vsi, bool link_up) { - if (!vsi || test_bit(__ICE_DOWN, vsi->state)) + if (!vsi) + return; + + if (test_bit(__ICE_DOWN, vsi->state) || !vsi->netdev) return; if (vsi->type == ICE_VSI_PF) { - if (!vsi->netdev) { - dev_dbg(&vsi->back->pdev->dev, - "vsi->netdev is not initialized!\n"); + if (link_up == netif_carrier_ok(vsi->netdev)) return; - } + if (link_up) { netif_carrier_on(vsi->netdev); netif_tx_wake_all_queues(vsi->netdev); @@ -556,63 +770,53 @@ static void ice_vsi_link_event(struct ice_vsi *vsi, bool link_up) /** * ice_link_event - process the link event - * @pf: pf that the link event is associated with + * @pf: PF that the link event is associated with * @pi: port_info for the port that the link event is associated with + * @link_up: true if the physical link is up and false if it is down + * @link_speed: current link speed received from the link event * - * Returns -EIO if ice_get_link_status() fails - * Returns 0 on success + * Returns 0 on success and negative on failure */ static int -ice_link_event(struct ice_pf *pf, struct ice_port_info *pi) +ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, + u16 link_speed) { - u8 new_link_speed, old_link_speed; struct ice_phy_info *phy_info; - bool new_link_same_as_old; - bool new_link, old_link; - u8 lport; - u16 v; + struct ice_vsi *vsi; + u16 old_link_speed; + bool old_link; + int result; phy_info = &pi->phy; phy_info->link_info_old = phy_info->link_info; - /* Force ice_get_link_status() to update link info */ - phy_info->get_link_info = true; - old_link = (phy_info->link_info_old.link_info & ICE_AQ_LINK_UP); + old_link = !!(phy_info->link_info_old.link_info & ICE_AQ_LINK_UP); old_link_speed = phy_info->link_info_old.link_speed; - lport = pi->lport; - if (ice_get_link_status(pi, &new_link)) { + /* update the link info structures and re-enable link events, + * don't bail on failure due to other book keeping needed + */ + result = ice_update_link_info(pi); + if (result) dev_dbg(&pf->pdev->dev, - "Could not get link status for port %d\n", lport); - return -EIO; - } - - new_link_speed = phy_info->link_info.link_speed; + "Failed to update link status and re-enable link events for port %d\n", + pi->lport); - new_link_same_as_old = (new_link == old_link && - new_link_speed == old_link_speed); + /* if the old link up/down and speed is the same as the new */ + if (link_up == old_link && link_speed == old_link_speed) + return result; - ice_for_each_vsi(pf, v) { - struct ice_vsi *vsi = pf->vsi[v]; - - if (!vsi || !vsi->port_info) - continue; - - if (new_link_same_as_old && - (test_bit(__ICE_DOWN, vsi->state) || - new_link == netif_carrier_ok(vsi->netdev))) - continue; + vsi = ice_find_vsi_by_type(pf, ICE_VSI_PF); + if (!vsi || !vsi->port_info) + return -EINVAL; - if (vsi->port_info->lport == lport) { - ice_print_link_msg(vsi, new_link); - ice_vsi_link_event(vsi, new_link); - } - } + ice_vsi_link_event(vsi, link_up); + ice_print_link_msg(vsi, link_up); - if (!new_link_same_as_old && pf->num_alloc_vfs) + if (pf->num_alloc_vfs) ice_vc_notify_link_state(pf); - return 0; + return result; } /** @@ -635,19 +839,73 @@ static void ice_watchdog_subtask(struct ice_pf *pf) pf->serv_tmr_prev = jiffies; - if (ice_link_event(pf, pf->hw.port_info)) - dev_dbg(&pf->pdev->dev, "ice_link_event failed\n"); - /* Update the stats for active netdevs so the network stack * can look at updated numbers whenever it cares to */ ice_update_pf_stats(pf); - for (i = 0; i < pf->num_alloc_vsi; i++) + ice_for_each_vsi(pf, i) if (pf->vsi[i] && pf->vsi[i]->netdev) ice_update_vsi_stats(pf->vsi[i]); } /** + * ice_init_link_events - enable/initialize link events + * @pi: pointer to the port_info instance + * + * Returns -EIO on failure, 0 on success + */ +static int ice_init_link_events(struct ice_port_info *pi) +{ + u16 mask; + + mask = ~((u16)(ICE_AQ_LINK_EVENT_UPDOWN | ICE_AQ_LINK_EVENT_MEDIA_NA | + ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL)); + + if (ice_aq_set_event_mask(pi->hw, pi->lport, mask, NULL)) { + dev_dbg(ice_hw_to_dev(pi->hw), + "Failed to set link event mask for port %d\n", + pi->lport); + return -EIO; + } + + if (ice_aq_get_link_info(pi, true, NULL, NULL)) { + dev_dbg(ice_hw_to_dev(pi->hw), + "Failed to enable link events for port %d\n", + pi->lport); + return -EIO; + } + + return 0; +} + +/** + * ice_handle_link_event - handle link event via ARQ + * @pf: PF that the link event is associated with + * @event: event structure containing link status info + */ +static int +ice_handle_link_event(struct ice_pf *pf, struct ice_rq_event_info *event) +{ + struct ice_aqc_get_link_status_data *link_data; + struct ice_port_info *port_info; + int status; + + link_data = (struct ice_aqc_get_link_status_data *)event->msg_buf; + port_info = pf->hw.port_info; + if (!port_info) + return -EINVAL; + + status = ice_link_event(pf, port_info, + !!(link_data->link_info & ICE_AQ_LINK_UP), + le16_to_cpu(link_data->link_speed)); + if (status) + dev_dbg(&pf->pdev->dev, + "Could not process link event, error %d\n", status); + + return status; +} + +/** * __ice_clean_ctrlq - helper function to clean controlq rings * @pf: ptr to struct ice_pf * @q_type: specific Control queue type @@ -750,12 +1008,20 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type) opcode = le16_to_cpu(event.desc.opcode); switch (opcode) { + case ice_aqc_opc_get_link_status: + if (ice_handle_link_event(pf, &event)) + dev_err(&pf->pdev->dev, + "Could not handle link event\n"); + break; case ice_mbx_opc_send_msg_to_pf: ice_vc_process_vf_msg(pf, &event); break; case ice_aqc_opc_fw_logging: ice_output_fw_log(hw, &event.desc, event.msg_buf); break; + case ice_aqc_opc_lldp_set_mib_change: + ice_dcb_process_lldp_set_mib_change(pf, &event); + break; default: dev_dbg(&pf->pdev->dev, "%s Receive Queue unknown event 0x%04x ignored\n", @@ -877,6 +1143,18 @@ static void ice_service_task_stop(struct ice_pf *pf) } /** + * ice_service_task_restart - restart service task and schedule works + * @pf: board private structure + * + * This function is needed for suspend and resume works (e.g WoL scenario) + */ +static void ice_service_task_restart(struct ice_pf *pf) +{ + clear_bit(__ICE_SERVICE_DIS, pf->state); + ice_service_task_schedule(pf); +} + +/** * ice_service_timer - timer callback to schedule service task * @t: pointer to timer_list */ @@ -901,7 +1179,7 @@ static void ice_handle_mdd_event(struct ice_pf *pf) u32 reg; int i; - if (!test_bit(__ICE_MDD_EVENT_PENDING, pf->state)) + if (!test_and_clear_bit(__ICE_MDD_EVENT_PENDING, pf->state)) return; /* find what triggered the MDD event */ @@ -989,14 +1267,16 @@ static void ice_handle_mdd_event(struct ice_pf *pf) } } - /* see if one of the VFs needs to be reset */ - for (i = 0; i < pf->num_alloc_vfs && mdd_detected; i++) { + /* check to see if one of the VFs caused the MDD */ + for (i = 0; i < pf->num_alloc_vfs; i++) { struct ice_vf *vf = &pf->vf[i]; + bool vf_mdd_detected = false; + reg = rd32(hw, VP_MDET_TX_PQM(i)); if (reg & VP_MDET_TX_PQM_VALID_M) { wr32(hw, VP_MDET_TX_PQM(i), 0xFFFF); - vf->num_mdd_events++; + vf_mdd_detected = true; dev_info(&pf->pdev->dev, "TX driver issue detected on VF %d\n", i); } @@ -1004,7 +1284,7 @@ static void ice_handle_mdd_event(struct ice_pf *pf) reg = rd32(hw, VP_MDET_TX_TCLAN(i)); if (reg & VP_MDET_TX_TCLAN_VALID_M) { wr32(hw, VP_MDET_TX_TCLAN(i), 0xFFFF); - vf->num_mdd_events++; + vf_mdd_detected = true; dev_info(&pf->pdev->dev, "TX driver issue detected on VF %d\n", i); } @@ -1012,7 +1292,7 @@ static void ice_handle_mdd_event(struct ice_pf *pf) reg = rd32(hw, VP_MDET_TX_TDPU(i)); if (reg & VP_MDET_TX_TDPU_VALID_M) { wr32(hw, VP_MDET_TX_TDPU(i), 0xFFFF); - vf->num_mdd_events++; + vf_mdd_detected = true; dev_info(&pf->pdev->dev, "TX driver issue detected on VF %d\n", i); } @@ -1020,26 +1300,18 @@ static void ice_handle_mdd_event(struct ice_pf *pf) reg = rd32(hw, VP_MDET_RX(i)); if (reg & VP_MDET_RX_VALID_M) { wr32(hw, VP_MDET_RX(i), 0xFFFF); - vf->num_mdd_events++; + vf_mdd_detected = true; dev_info(&pf->pdev->dev, "RX driver issue detected on VF %d\n", i); } - if (vf->num_mdd_events > ICE_DFLT_NUM_MDD_EVENTS_ALLOWED) { - dev_info(&pf->pdev->dev, - "Too many MDD events on VF %d, disabled\n", i); - dev_info(&pf->pdev->dev, - "Use PF Control I/F to re-enable the VF\n"); - set_bit(ICE_VF_STATE_DIS, vf->vf_states); + if (vf_mdd_detected) { + vf->num_mdd_events++; + if (vf->num_mdd_events > 1) + dev_info(&pf->pdev->dev, "VF %d has had %llu MDD events since last boot\n", + i, vf->num_mdd_events); } } - - /* re-enable MDD interrupt cause */ - clear_bit(__ICE_MDD_EVENT_PENDING, pf->state); - reg = rd32(hw, PFINT_OICR_ENA); - reg |= PFINT_OICR_MAL_DETECT_M; - wr32(hw, PFINT_OICR_ENA, reg); - ice_flush(hw); } /** @@ -1089,7 +1361,7 @@ static void ice_service_task(struct work_struct *work) /** * ice_set_ctrlq_len - helper function to set controlq length - * @hw: pointer to the hw instance + * @hw: pointer to the HW instance */ static void ice_set_ctrlq_len(struct ice_hw *hw) { @@ -1111,8 +1383,9 @@ static void ice_set_ctrlq_len(struct ice_hw *hw) * This is a callback function used by the irq_set_affinity_notifier function * so that we may register to receive changes to the irq affinity masks. */ -static void ice_irq_affinity_notify(struct irq_affinity_notify *notify, - const cpumask_t *mask) +static void +ice_irq_affinity_notify(struct irq_affinity_notify *notify, + const cpumask_t *mask) { struct ice_q_vector *q_vector = container_of(notify, struct ice_q_vector, affinity_notify); @@ -1142,7 +1415,7 @@ static int ice_vsi_ena_irq(struct ice_vsi *vsi) if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { int i; - for (i = 0; i < vsi->num_q_vectors; i++) + ice_for_each_q_vector(vsi, i) ice_irq_dynamic_ena(hw, vsi, vsi->q_vectors[i]); } @@ -1159,7 +1432,7 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) { int q_vectors = vsi->num_q_vectors; struct ice_pf *pf = vsi->back; - int base = vsi->sw_base_vector; + int base = vsi->base_vector; int rx_int_idx = 0; int tx_int_idx = 0; int vector, err; @@ -1184,10 +1457,9 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) /* skip this unused q_vector */ continue; } - err = devm_request_irq(&pf->pdev->dev, - pf->msix_entries[base + vector].vector, - vsi->irq_handler, 0, q_vector->name, - q_vector); + err = devm_request_irq(&pf->pdev->dev, irq_num, + vsi->irq_handler, 0, + q_vector->name, q_vector); if (err) { netdev_err(vsi->netdev, "MSIX request_irq failed, error: %d\n", err); @@ -1241,7 +1513,7 @@ static void ice_ena_misc_vector(struct ice_pf *pf) wr32(hw, PFINT_OICR_ENA, val); /* SW_ITR_IDX = 0, but don't change INTENA */ - wr32(hw, GLINT_DYN_CTL(pf->hw_oicr_idx), + wr32(hw, GLINT_DYN_CTL(pf->oicr_idx), GLINT_DYN_CTL_SW_ITR_INDX_M | GLINT_DYN_CTL_INTENA_MSK_M); } @@ -1263,6 +1535,11 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) oicr = rd32(hw, PFINT_OICR); ena_mask = rd32(hw, PFINT_OICR_ENA); + if (oicr & PFINT_OICR_SWINT_M) { + ena_mask &= ~PFINT_OICR_SWINT_M; + pf->sw_int_count++; + } + if (oicr & PFINT_OICR_MAL_DETECT_M) { ena_mask &= ~PFINT_OICR_MAL_DETECT_M; set_bit(__ICE_MDD_EVENT_PENDING, pf->state); @@ -1328,7 +1605,7 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) rd32(hw, PFHMC_ERRORDATA)); } - /* Report and mask off any remaining unexpected interrupts */ + /* Report any remaining unexpected interrupts */ oicr &= ena_mask; if (oicr) { dev_dbg(&pf->pdev->dev, "unhandled interrupt oicr=0x%08x\n", @@ -1342,12 +1619,9 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) set_bit(__ICE_PFR_REQ, pf->state); ice_service_task_schedule(pf); } - ena_mask &= ~oicr; } ret = IRQ_HANDLED; - /* re-enable interrupt causes that are not handled during this pass */ - wr32(hw, PFINT_OICR_ENA, ena_mask); if (!test_bit(__ICE_DOWN, pf->state)) { ice_service_task_schedule(pf); ice_irq_dynamic_ena(hw, NULL, NULL); @@ -1392,37 +1666,35 @@ static void ice_free_irq_msix_misc(struct ice_pf *pf) ice_flush(hw); if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags) && pf->msix_entries) { - synchronize_irq(pf->msix_entries[pf->sw_oicr_idx].vector); + synchronize_irq(pf->msix_entries[pf->oicr_idx].vector); devm_free_irq(&pf->pdev->dev, - pf->msix_entries[pf->sw_oicr_idx].vector, pf); + pf->msix_entries[pf->oicr_idx].vector, pf); } pf->num_avail_sw_msix += 1; - ice_free_res(pf->sw_irq_tracker, pf->sw_oicr_idx, ICE_RES_MISC_VEC_ID); - pf->num_avail_hw_msix += 1; - ice_free_res(pf->hw_irq_tracker, pf->hw_oicr_idx, ICE_RES_MISC_VEC_ID); + ice_free_res(pf->irq_tracker, pf->oicr_idx, ICE_RES_MISC_VEC_ID); } /** * ice_ena_ctrlq_interrupts - enable control queue interrupts * @hw: pointer to HW structure - * @v_idx: HW vector index to associate the control queue interrupts with + * @reg_idx: HW vector index to associate the control queue interrupts with */ -static void ice_ena_ctrlq_interrupts(struct ice_hw *hw, u16 v_idx) +static void ice_ena_ctrlq_interrupts(struct ice_hw *hw, u16 reg_idx) { u32 val; - val = ((v_idx & PFINT_OICR_CTL_MSIX_INDX_M) | + val = ((reg_idx & PFINT_OICR_CTL_MSIX_INDX_M) | PFINT_OICR_CTL_CAUSE_ENA_M); wr32(hw, PFINT_OICR_CTL, val); /* enable Admin queue Interrupt causes */ - val = ((v_idx & PFINT_FW_CTL_MSIX_INDX_M) | + val = ((reg_idx & PFINT_FW_CTL_MSIX_INDX_M) | PFINT_FW_CTL_CAUSE_ENA_M); wr32(hw, PFINT_FW_CTL, val); /* enable Mailbox queue Interrupt causes */ - val = ((v_idx & PFINT_MBX_CTL_MSIX_INDX_M) | + val = ((reg_idx & PFINT_MBX_CTL_MSIX_INDX_M) | PFINT_MBX_CTL_CAUSE_ENA_M); wr32(hw, PFINT_MBX_CTL, val); @@ -1454,43 +1726,31 @@ static int ice_req_irq_msix_misc(struct ice_pf *pf) if (ice_is_reset_in_progress(pf->state)) goto skip_req_irq; - /* reserve one vector in sw_irq_tracker for misc interrupts */ - oicr_idx = ice_get_res(pf, pf->sw_irq_tracker, 1, ICE_RES_MISC_VEC_ID); + /* reserve one vector in irq_tracker for misc interrupts */ + oicr_idx = ice_get_res(pf, pf->irq_tracker, 1, ICE_RES_MISC_VEC_ID); if (oicr_idx < 0) return oicr_idx; pf->num_avail_sw_msix -= 1; - pf->sw_oicr_idx = oicr_idx; - - /* reserve one vector in hw_irq_tracker for misc interrupts */ - oicr_idx = ice_get_res(pf, pf->hw_irq_tracker, 1, ICE_RES_MISC_VEC_ID); - if (oicr_idx < 0) { - ice_free_res(pf->sw_irq_tracker, 1, ICE_RES_MISC_VEC_ID); - pf->num_avail_sw_msix += 1; - return oicr_idx; - } - pf->num_avail_hw_msix -= 1; - pf->hw_oicr_idx = oicr_idx; + pf->oicr_idx = oicr_idx; err = devm_request_irq(&pf->pdev->dev, - pf->msix_entries[pf->sw_oicr_idx].vector, + pf->msix_entries[pf->oicr_idx].vector, ice_misc_intr, 0, pf->int_name, pf); if (err) { dev_err(&pf->pdev->dev, "devm_request_irq for %s failed: %d\n", pf->int_name, err); - ice_free_res(pf->sw_irq_tracker, 1, ICE_RES_MISC_VEC_ID); + ice_free_res(pf->irq_tracker, 1, ICE_RES_MISC_VEC_ID); pf->num_avail_sw_msix += 1; - ice_free_res(pf->hw_irq_tracker, 1, ICE_RES_MISC_VEC_ID); - pf->num_avail_hw_msix += 1; return err; } skip_req_irq: ice_ena_misc_vector(pf); - ice_ena_ctrlq_interrupts(hw, pf->hw_oicr_idx); - wr32(hw, GLINT_ITR(ICE_RX_ITR, pf->hw_oicr_idx), + ice_ena_ctrlq_interrupts(hw, pf->oicr_idx); + wr32(hw, GLINT_ITR(ICE_RX_ITR, pf->oicr_idx), ITR_REG_ALIGN(ICE_ITR_8K) >> ICE_ITR_GRAN_S); ice_flush(hw); @@ -1500,21 +1760,6 @@ skip_req_irq: } /** - * ice_napi_del - Remove NAPI handler for the VSI - * @vsi: VSI for which NAPI handler is to be removed - */ -void ice_napi_del(struct ice_vsi *vsi) -{ - int v_idx; - - if (!vsi->netdev) - return; - - for (v_idx = 0; v_idx < vsi->num_q_vectors; v_idx++) - netif_napi_del(&vsi->q_vectors[v_idx]->napi); -} - -/** * ice_napi_add - register NAPI handler for the VSI * @vsi: VSI for which NAPI handler is to be registered * @@ -1529,7 +1774,7 @@ static void ice_napi_add(struct ice_vsi *vsi) if (!vsi->netdev) return; - for (v_idx = 0; v_idx < vsi->num_q_vectors; v_idx++) + ice_for_each_q_vector(vsi, v_idx) netif_napi_add(vsi->netdev, &vsi->q_vectors[v_idx]->napi, ice_napi_poll, NAPI_POLL_WEIGHT); } @@ -1639,8 +1884,8 @@ void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size) * @pf: board private structure * @pi: pointer to the port_info instance * - * Returns pointer to the successfully allocated VSI sw struct on success, - * otherwise returns NULL on failure. + * Returns pointer to the successfully allocated VSI software struct + * on success, otherwise returns NULL on failure. */ static struct ice_vsi * ice_pf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) @@ -1649,18 +1894,34 @@ ice_pf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) } /** - * ice_vlan_rx_add_vid - Add a vlan id filter to HW offload + * ice_lb_vsi_setup - Set up a loopback VSI + * @pf: board private structure + * @pi: pointer to the port_info instance + * + * Returns pointer to the successfully allocated VSI software struct + * on success, otherwise returns NULL on failure. + */ +struct ice_vsi * +ice_lb_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) +{ + return ice_vsi_setup(pf, pi, ICE_VSI_LB, ICE_INVAL_VFID); +} + +/** + * ice_vlan_rx_add_vid - Add a VLAN ID filter to HW offload * @netdev: network interface to be adjusted * @proto: unused protocol - * @vid: vlan id to be added + * @vid: VLAN ID to be added * - * net_device_ops implementation for adding vlan ids + * net_device_ops implementation for adding VLAN IDs */ -static int ice_vlan_rx_add_vid(struct net_device *netdev, - __always_unused __be16 proto, u16 vid) +static int +ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto, + u16 vid) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; + int ret; if (vid >= VLAN_N_VID) { netdev_err(netdev, "VLAN id requested %d is out of range %d\n", @@ -1673,33 +1934,39 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev, /* Enable VLAN pruning when VLAN 0 is added */ if (unlikely(!vid)) { - int ret = ice_cfg_vlan_pruning(vsi, true); - + ret = ice_cfg_vlan_pruning(vsi, true, false); if (ret) return ret; } - /* Add all VLAN ids including 0 to the switch filter. VLAN id 0 is + /* Add all VLAN IDs including 0 to the switch filter. VLAN ID 0 is * needed to continue allowing all untagged packets since VLAN prune * list is applied to all packets by the switch */ - return ice_vsi_add_vlan(vsi, vid); + ret = ice_vsi_add_vlan(vsi, vid); + if (!ret) { + vsi->vlan_ena = true; + set_bit(ICE_VSI_FLAG_VLAN_FLTR_CHANGED, vsi->flags); + } + + return ret; } /** - * ice_vlan_rx_kill_vid - Remove a vlan id filter from HW offload + * ice_vlan_rx_kill_vid - Remove a VLAN ID filter from HW offload * @netdev: network interface to be adjusted * @proto: unused protocol - * @vid: vlan id to be removed + * @vid: VLAN ID to be removed * - * net_device_ops implementation for removing vlan ids + * net_device_ops implementation for removing VLAN IDs */ -static int ice_vlan_rx_kill_vid(struct net_device *netdev, - __always_unused __be16 proto, u16 vid) +static int +ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto, + u16 vid) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - int status; + int ret; if (vsi->info.pvid) return -EINVAL; @@ -1707,15 +1974,17 @@ static int ice_vlan_rx_kill_vid(struct net_device *netdev, /* Make sure ice_vsi_kill_vlan is successful before updating VLAN * information */ - status = ice_vsi_kill_vlan(vsi, vid); - if (status) - return status; + ret = ice_vsi_kill_vlan(vsi, vid); + if (ret) + return ret; /* Disable VLAN pruning when VLAN 0 is removed */ if (unlikely(!vid)) - status = ice_cfg_vlan_pruning(vsi, false); + ret = ice_cfg_vlan_pruning(vsi, false, false); - return status; + vsi->vlan_ena = false; + set_bit(ICE_VSI_FLAG_VLAN_FLTR_CHANGED, vsi->flags); + return ret; } /** @@ -1726,8 +1995,6 @@ static int ice_vlan_rx_kill_vid(struct net_device *netdev, */ static int ice_setup_pf_sw(struct ice_pf *pf) { - LIST_HEAD(tmp_add_list); - u8 broadcast[ETH_ALEN]; struct ice_vsi *vsi; int status = 0; @@ -1752,38 +2019,12 @@ static int ice_setup_pf_sw(struct ice_pf *pf) */ ice_napi_add(vsi); - /* To add a MAC filter, first add the MAC to a list and then - * pass the list to ice_add_mac. - */ - - /* Add a unicast MAC filter so the VSI can get its packets */ - status = ice_add_mac_to_list(vsi, &tmp_add_list, - vsi->port_info->mac.perm_addr); + status = ice_init_mac_fltr(pf); if (status) goto unroll_napi_add; - /* VSI needs to receive broadcast traffic, so add the broadcast - * MAC address to the list as well. - */ - eth_broadcast_addr(broadcast); - status = ice_add_mac_to_list(vsi, &tmp_add_list, broadcast); - if (status) - goto free_mac_list; - - /* program MAC filters for entries in tmp_add_list */ - status = ice_add_mac(&pf->hw, &tmp_add_list); - if (status) { - dev_err(&pf->pdev->dev, "Could not add MAC filters\n"); - status = -ENOMEM; - goto free_mac_list; - } - - ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); return status; -free_mac_list: - ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); - unroll_napi_add: if (vsi) { ice_napi_del(vsi); @@ -1975,14 +2216,9 @@ static void ice_clear_interrupt_scheme(struct ice_pf *pf) if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) ice_dis_msix(pf); - if (pf->sw_irq_tracker) { - devm_kfree(&pf->pdev->dev, pf->sw_irq_tracker); - pf->sw_irq_tracker = NULL; - } - - if (pf->hw_irq_tracker) { - devm_kfree(&pf->pdev->dev, pf->hw_irq_tracker); - pf->hw_irq_tracker = NULL; + if (pf->irq_tracker) { + devm_kfree(&pf->pdev->dev, pf->irq_tracker); + pf->irq_tracker = NULL; } } @@ -1992,7 +2228,7 @@ static void ice_clear_interrupt_scheme(struct ice_pf *pf) */ static int ice_init_interrupt_scheme(struct ice_pf *pf) { - int vectors = 0, hw_vectors = 0; + int vectors; if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) vectors = ice_ena_msix_range(pf); @@ -2003,53 +2239,23 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf) return vectors; /* set up vector assignment tracking */ - pf->sw_irq_tracker = - devm_kzalloc(&pf->pdev->dev, sizeof(*pf->sw_irq_tracker) + + pf->irq_tracker = + devm_kzalloc(&pf->pdev->dev, sizeof(*pf->irq_tracker) + (sizeof(u16) * vectors), GFP_KERNEL); - if (!pf->sw_irq_tracker) { + if (!pf->irq_tracker) { ice_dis_msix(pf); return -ENOMEM; } /* populate SW interrupts pool with number of OS granted IRQs. */ pf->num_avail_sw_msix = vectors; - pf->sw_irq_tracker->num_entries = vectors; - - /* set up HW vector assignment tracking */ - hw_vectors = pf->hw.func_caps.common_cap.num_msix_vectors; - pf->hw_irq_tracker = - devm_kzalloc(&pf->pdev->dev, sizeof(*pf->hw_irq_tracker) + - (sizeof(u16) * hw_vectors), GFP_KERNEL); - if (!pf->hw_irq_tracker) { - ice_clear_interrupt_scheme(pf); - return -ENOMEM; - } - - /* populate HW interrupts pool with number of HW supported irqs. */ - pf->num_avail_hw_msix = hw_vectors; - pf->hw_irq_tracker->num_entries = hw_vectors; + pf->irq_tracker->num_entries = vectors; + pf->irq_tracker->end = pf->irq_tracker->num_entries; return 0; } /** - * ice_verify_itr_gran - verify driver's assumption of ITR granularity - * @pf: pointer to the PF structure - * - * There is no error returned here because the driver will be able to handle a - * different ITR granularity, but interrupt moderation will not be accurate if - * the driver's assumptions are not verified. This assumption is made so we can - * use constants in the hot path instead of accessing structure members. - */ -static void ice_verify_itr_gran(struct ice_pf *pf) -{ - if (pf->hw.itr_gran != (ICE_ITR_GRAN_S << 1)) - dev_warn(&pf->pdev->dev, - "%d ITR granularity assumption is invalid, actual ITR granularity is %d. Interrupt moderation will be inaccurate!\n", - (ICE_ITR_GRAN_S << 1), pf->hw.itr_gran); -} - -/** * ice_verify_cacheline_size - verify driver's assumption of 64 Byte cache lines * @pf: pointer to the PF structure * @@ -2072,9 +2278,10 @@ static void ice_verify_cacheline_size(struct ice_pf *pf) * * Returns 0 on success, negative on failure */ -static int ice_probe(struct pci_dev *pdev, - const struct pci_device_id __always_unused *ent) +static int +ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) { + struct device *dev = &pdev->dev; struct ice_pf *pf; struct ice_hw *hw; int err; @@ -2086,20 +2293,20 @@ static int ice_probe(struct pci_dev *pdev, err = pcim_iomap_regions(pdev, BIT(ICE_BAR0), pci_name(pdev)); if (err) { - dev_err(&pdev->dev, "BAR0 I/O map error %d\n", err); + dev_err(dev, "BAR0 I/O map error %d\n", err); return err; } - pf = devm_kzalloc(&pdev->dev, sizeof(*pf), GFP_KERNEL); + pf = devm_kzalloc(dev, sizeof(*pf), GFP_KERNEL); if (!pf) return -ENOMEM; - /* set up for high or low dma */ - err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); + /* set up for high or low DMA */ + err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)); if (err) - err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32)); + err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32)); if (err) { - dev_err(&pdev->dev, "DMA configuration failed: 0x%x\n", err); + dev_err(dev, "DMA configuration failed: 0x%x\n", err); return err; } @@ -2133,17 +2340,26 @@ static int ice_probe(struct pci_dev *pdev, err = ice_init_hw(hw); if (err) { - dev_err(&pdev->dev, "ice_init_hw failed: %d\n", err); + dev_err(dev, "ice_init_hw failed: %d\n", err); err = -EIO; goto err_exit_unroll; } - dev_info(&pdev->dev, "firmware %d.%d.%05d api %d.%d\n", + dev_info(dev, "firmware %d.%d.%05d api %d.%d\n", hw->fw_maj_ver, hw->fw_min_ver, hw->fw_build, hw->api_maj_ver, hw->api_min_ver); ice_init_pf(pf); + err = ice_init_pf_dcb(pf, false); + if (err) { + clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); + clear_bit(ICE_FLAG_DCB_ENA, pf->flags); + + /* do not fail overall init if DCB init fails */ + err = 0; + } + ice_determine_q_usage(pf); pf->num_alloc_vsi = hw->func_caps.guar_num_vsi; @@ -2152,8 +2368,8 @@ static int ice_probe(struct pci_dev *pdev, goto err_init_pf_unroll; } - pf->vsi = devm_kcalloc(&pdev->dev, pf->num_alloc_vsi, - sizeof(*pf->vsi), GFP_KERNEL); + pf->vsi = devm_kcalloc(dev, pf->num_alloc_vsi, sizeof(*pf->vsi), + GFP_KERNEL); if (!pf->vsi) { err = -ENOMEM; goto err_init_pf_unroll; @@ -2161,8 +2377,7 @@ static int ice_probe(struct pci_dev *pdev, err = ice_init_interrupt_scheme(pf); if (err) { - dev_err(&pdev->dev, - "ice_init_interrupt_scheme failed: %d\n", err); + dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err); err = -EIO; goto err_init_interrupt_unroll; } @@ -2178,15 +2393,13 @@ static int ice_probe(struct pci_dev *pdev, if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { err = ice_req_irq_msix_misc(pf); if (err) { - dev_err(&pdev->dev, - "setup of misc vector failed: %d\n", err); + dev_err(dev, "setup of misc vector failed: %d\n", err); goto err_init_interrupt_unroll; } } /* create switch struct for the switch element created by FW on boot */ - pf->first_sw = devm_kzalloc(&pdev->dev, sizeof(*pf->first_sw), - GFP_KERNEL); + pf->first_sw = devm_kzalloc(dev, sizeof(*pf->first_sw), GFP_KERNEL); if (!pf->first_sw) { err = -ENOMEM; goto err_msix_misc_unroll; @@ -2204,8 +2417,7 @@ static int ice_probe(struct pci_dev *pdev, err = ice_setup_pf_sw(pf); if (err) { - dev_err(&pdev->dev, - "probe failed due to setup pf switch:%d\n", err); + dev_err(dev, "probe failed due to setup PF switch:%d\n", err); goto err_alloc_sw_unroll; } @@ -2214,8 +2426,13 @@ static int ice_probe(struct pci_dev *pdev, /* since everything is good, start the service timer */ mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period)); + err = ice_init_link_events(pf->hw.port_info); + if (err) { + dev_err(dev, "ice_init_link_events failed: %d\n", err); + goto err_alloc_sw_unroll; + } + ice_verify_cacheline_size(pf); - ice_verify_itr_gran(pf); return 0; @@ -2227,7 +2444,7 @@ err_msix_misc_unroll: ice_free_irq_msix_misc(pf); err_init_interrupt_unroll: ice_clear_interrupt_scheme(pf); - devm_kfree(&pdev->dev, pf->vsi); + devm_kfree(dev, pf->vsi); err_init_pf_unroll: ice_deinit_pf(pf); ice_deinit_hw(hw); @@ -2272,6 +2489,136 @@ static void ice_remove(struct pci_dev *pdev) pci_disable_pcie_error_reporting(pdev); } +/** + * ice_pci_err_detected - warning that PCI error has been detected + * @pdev: PCI device information struct + * @err: the type of PCI error + * + * Called to warn that something happened on the PCI bus and the error handling + * is in progress. Allows the driver to gracefully prepare/handle PCI errors. + */ +static pci_ers_result_t +ice_pci_err_detected(struct pci_dev *pdev, enum pci_channel_state err) +{ + struct ice_pf *pf = pci_get_drvdata(pdev); + + if (!pf) { + dev_err(&pdev->dev, "%s: unrecoverable device error %d\n", + __func__, err); + return PCI_ERS_RESULT_DISCONNECT; + } + + if (!test_bit(__ICE_SUSPENDED, pf->state)) { + ice_service_task_stop(pf); + + if (!test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) { + set_bit(__ICE_PFR_REQ, pf->state); + ice_prepare_for_reset(pf); + } + } + + return PCI_ERS_RESULT_NEED_RESET; +} + +/** + * ice_pci_err_slot_reset - a PCI slot reset has just happened + * @pdev: PCI device information struct + * + * Called to determine if the driver can recover from the PCI slot reset by + * using a register read to determine if the device is recoverable. + */ +static pci_ers_result_t ice_pci_err_slot_reset(struct pci_dev *pdev) +{ + struct ice_pf *pf = pci_get_drvdata(pdev); + pci_ers_result_t result; + int err; + u32 reg; + + err = pci_enable_device_mem(pdev); + if (err) { + dev_err(&pdev->dev, + "Cannot re-enable PCI device after reset, error %d\n", + err); + result = PCI_ERS_RESULT_DISCONNECT; + } else { + pci_set_master(pdev); + pci_restore_state(pdev); + pci_save_state(pdev); + pci_wake_from_d3(pdev, false); + + /* Check for life */ + reg = rd32(&pf->hw, GLGEN_RTRIG); + if (!reg) + result = PCI_ERS_RESULT_RECOVERED; + else + result = PCI_ERS_RESULT_DISCONNECT; + } + + err = pci_cleanup_aer_uncorrect_error_status(pdev); + if (err) + dev_dbg(&pdev->dev, + "pci_cleanup_aer_uncorrect_error_status failed, error %d\n", + err); + /* non-fatal, continue */ + + return result; +} + +/** + * ice_pci_err_resume - restart operations after PCI error recovery + * @pdev: PCI device information struct + * + * Called to allow the driver to bring things back up after PCI error and/or + * reset recovery have finished + */ +static void ice_pci_err_resume(struct pci_dev *pdev) +{ + struct ice_pf *pf = pci_get_drvdata(pdev); + + if (!pf) { + dev_err(&pdev->dev, + "%s failed, device is unrecoverable\n", __func__); + return; + } + + if (test_bit(__ICE_SUSPENDED, pf->state)) { + dev_dbg(&pdev->dev, "%s failed to resume normal operations!\n", + __func__); + return; + } + + ice_do_reset(pf, ICE_RESET_PFR); + ice_service_task_restart(pf); + mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period)); +} + +/** + * ice_pci_err_reset_prepare - prepare device driver for PCI reset + * @pdev: PCI device information struct + */ +static void ice_pci_err_reset_prepare(struct pci_dev *pdev) +{ + struct ice_pf *pf = pci_get_drvdata(pdev); + + if (!test_bit(__ICE_SUSPENDED, pf->state)) { + ice_service_task_stop(pf); + + if (!test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) { + set_bit(__ICE_PFR_REQ, pf->state); + ice_prepare_for_reset(pf); + } + } +} + +/** + * ice_pci_err_reset_done - PCI reset done, device driver reset can begin + * @pdev: PCI device information struct + */ +static void ice_pci_err_reset_done(struct pci_dev *pdev) +{ + ice_pci_err_resume(pdev); +} + /* ice_pci_tbl - PCI Device ID Table * * Wildcard entries (PCI_ANY_ID) should come last @@ -2289,12 +2636,21 @@ static const struct pci_device_id ice_pci_tbl[] = { }; MODULE_DEVICE_TABLE(pci, ice_pci_tbl); +static const struct pci_error_handlers ice_pci_err_handler = { + .error_detected = ice_pci_err_detected, + .slot_reset = ice_pci_err_slot_reset, + .reset_prepare = ice_pci_err_reset_prepare, + .reset_done = ice_pci_err_reset_done, + .resume = ice_pci_err_resume +}; + static struct pci_driver ice_driver = { .name = KBUILD_MODNAME, .id_table = ice_pci_tbl, .probe = ice_probe, .remove = ice_remove, .sriov_configure = ice_sriov_configure, + .err_handler = &ice_pci_err_handler }; /** @@ -2318,7 +2674,7 @@ static int __init ice_module_init(void) status = pci_register_driver(&ice_driver); if (status) { - pr_err("failed to register pci driver, err %d\n", status); + pr_err("failed to register PCI driver, err %d\n", status); destroy_workqueue(ice_wq); } @@ -2341,7 +2697,7 @@ static void __exit ice_module_exit(void) module_exit(ice_module_exit); /** - * ice_set_mac_address - NDO callback to set mac address + * ice_set_mac_address - NDO callback to set MAC address * @netdev: network interface device structure * @pi: pointer to an address structure * @@ -2378,14 +2734,14 @@ static int ice_set_mac_address(struct net_device *netdev, void *pi) return -EBUSY; } - /* When we change the mac address we also have to change the mac address - * based filter rules that were created previously for the old mac + /* When we change the MAC address we also have to change the MAC address + * based filter rules that were created previously for the old MAC * address. So first, we remove the old filter rule using ice_remove_mac * and then create a new filter rule using ice_add_mac. Note that for - * both these operations, we first need to form a "list" of mac - * addresses (even though in this case, we have only 1 mac address to be + * both these operations, we first need to form a "list" of MAC + * addresses (even though in this case, we have only 1 MAC address to be * added/removed) and this done using ice_add_mac_to_list. Depending on - * the ensuing operation this "list" of mac addresses is either to be + * the ensuing operation this "list" of MAC addresses is either to be * added or removed from the filter. */ err = ice_add_mac_to_list(vsi, &r_mac_list, netdev->dev_addr); @@ -2418,21 +2774,21 @@ free_lists: ice_free_fltr_list(&pf->pdev->dev, &a_mac_list); if (err) { - netdev_err(netdev, "can't set mac %pM. filter update failed\n", + netdev_err(netdev, "can't set MAC %pM. filter update failed\n", mac); return err; } - /* change the netdev's mac address */ + /* change the netdev's MAC address */ memcpy(netdev->dev_addr, mac, netdev->addr_len); - netdev_dbg(vsi->netdev, "updated mac address to %pM\n", + netdev_dbg(vsi->netdev, "updated MAC address to %pM\n", netdev->dev_addr); - /* write new mac address to the firmware */ + /* write new MAC address to the firmware */ flags = ICE_AQC_MAN_MAC_UPDATE_LAA_WOL; status = ice_aq_manage_mac_write(hw, mac, flags, NULL); if (status) { - netdev_err(netdev, "can't set mac %pM. write to firmware failed.\n", + netdev_err(netdev, "can't set MAC %pM. write to firmware failed.\n", mac); } return 0; @@ -2470,7 +2826,7 @@ static void ice_set_rx_mode(struct net_device *netdev) * @tb: pointer to array of nladdr (unused) * @dev: the net device pointer * @addr: the MAC address entry being added - * @vid: VLAN id + * @vid: VLAN ID * @flags: instructions from stack about fdb operation * @extack: netlink extended ack */ @@ -2510,11 +2866,12 @@ ice_fdb_add(struct ndmsg *ndm, struct nlattr __always_unused *tb[], * @tb: pointer to array of nladdr (unused) * @dev: the net device pointer * @addr: the MAC address entry being added - * @vid: VLAN id + * @vid: VLAN ID */ -static int ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[], - struct net_device *dev, const unsigned char *addr, - __always_unused u16 vid) +static int +ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[], + struct net_device *dev, const unsigned char *addr, + __always_unused u16 vid) { int err; @@ -2538,13 +2895,16 @@ static int ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[], * @netdev: ptr to the netdev being adjusted * @features: the feature set that the stack is suggesting */ -static int ice_set_features(struct net_device *netdev, - netdev_features_t features) +static int +ice_set_features(struct net_device *netdev, netdev_features_t features) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; int ret = 0; + /* Multiple features can be changed in one call so keep features in + * separate if/else statements to guarantee each feature is checked + */ if (features & NETIF_F_RXHASH && !(netdev->features & NETIF_F_RXHASH)) ret = ice_vsi_manage_rss_lut(vsi, true); else if (!(features & NETIF_F_RXHASH) && @@ -2557,19 +2917,27 @@ static int ice_set_features(struct net_device *netdev, else if (!(features & NETIF_F_HW_VLAN_CTAG_RX) && (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)) ret = ice_vsi_manage_vlan_stripping(vsi, false); - else if ((features & NETIF_F_HW_VLAN_CTAG_TX) && - !(netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) + + if ((features & NETIF_F_HW_VLAN_CTAG_TX) && + !(netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) ret = ice_vsi_manage_vlan_insertion(vsi); else if (!(features & NETIF_F_HW_VLAN_CTAG_TX) && (netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) ret = ice_vsi_manage_vlan_insertion(vsi); + if ((features & NETIF_F_HW_VLAN_CTAG_FILTER) && + !(netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) + ret = ice_cfg_vlan_pruning(vsi, true, false); + else if (!(features & NETIF_F_HW_VLAN_CTAG_FILTER) && + (netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) + ret = ice_cfg_vlan_pruning(vsi, false, false); + return ret; } /** - * ice_vsi_vlan_setup - Setup vlan offload properties on a VSI - * @vsi: VSI to setup vlan properties for + * ice_vsi_vlan_setup - Setup VLAN offload properties on a VSI + * @vsi: VSI to setup VLAN properties for */ static int ice_vsi_vlan_setup(struct ice_vsi *vsi) { @@ -2589,7 +2957,7 @@ static int ice_vsi_vlan_setup(struct ice_vsi *vsi) * * Return 0 on success and negative value on error */ -static int ice_vsi_cfg(struct ice_vsi *vsi) +int ice_vsi_cfg(struct ice_vsi *vsi) { int err; @@ -2601,6 +2969,7 @@ static int ice_vsi_cfg(struct ice_vsi *vsi) if (err) return err; } + ice_vsi_cfg_dcb_rings(vsi); err = ice_vsi_cfg_lan_txqs(vsi); if (!err) @@ -2620,7 +2989,7 @@ static void ice_napi_enable_all(struct ice_vsi *vsi) if (!vsi->netdev) return; - for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) { + ice_for_each_q_vector(vsi, q_idx) { struct ice_q_vector *q_vector = vsi->q_vectors[q_idx]; if (q_vector->rx.ring || q_vector->tx.ring) @@ -2666,7 +3035,7 @@ static int ice_up_complete(struct ice_vsi *vsi) ice_service_task_schedule(pf); - return err; + return 0; } /** @@ -2693,8 +3062,8 @@ int ice_up(struct ice_vsi *vsi) * This function fetches stats from the ring considering the atomic operations * that needs to be performed to read u64 values in 32 bit machine. */ -static void ice_fetch_u64_stats_per_ring(struct ice_ring *ring, u64 *pkts, - u64 *bytes) +static void +ice_fetch_u64_stats_per_ring(struct ice_ring *ring, u64 *pkts, u64 *bytes) { unsigned int start; *pkts = 0; @@ -2911,6 +3280,8 @@ static void ice_update_pf_stats(struct ice_pf *pf) ice_stat_update32(hw, GLPRT_LXOFFTXC(pf_id), pf->stat_prev_loaded, &prev_ps->link_xoff_tx, &cur_ps->link_xoff_tx); + ice_update_dcb_stats(pf); + ice_stat_update32(hw, GLPRT_CRCERRS(pf_id), pf->stat_prev_loaded, &prev_ps->crc_errors, &cur_ps->crc_errors); @@ -2992,7 +3363,7 @@ static void ice_napi_disable_all(struct ice_vsi *vsi) if (!vsi->netdev) return; - for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) { + ice_for_each_q_vector(vsi, q_idx) { struct ice_q_vector *q_vector = vsi->q_vectors[q_idx]; if (q_vector->rx.ring || q_vector->tx.ring) @@ -3141,7 +3512,7 @@ int ice_down(struct ice_vsi *vsi) * * Return 0 on success, negative on failure */ -static int ice_vsi_setup_tx_rings(struct ice_vsi *vsi) +int ice_vsi_setup_tx_rings(struct ice_vsi *vsi) { int i, err = 0; @@ -3167,7 +3538,7 @@ static int ice_vsi_setup_tx_rings(struct ice_vsi *vsi) * * Return 0 on success, negative on failure */ -static int ice_vsi_setup_rx_rings(struct ice_vsi *vsi) +int ice_vsi_setup_rx_rings(struct ice_vsi *vsi) { int i, err = 0; @@ -3276,7 +3647,7 @@ static void ice_vsi_release_all(struct ice_pf *pf) if (!pf->vsi) return; - for (i = 0; i < pf->num_alloc_vsi; i++) { + ice_for_each_vsi(pf, i) { if (!pf->vsi[i]) continue; @@ -3289,47 +3660,31 @@ static void ice_vsi_release_all(struct ice_pf *pf) } /** - * ice_dis_vsi - pause a VSI - * @vsi: the VSI being paused + * ice_ena_vsi - resume a VSI + * @vsi: the VSI being resume * @locked: is the rtnl_lock already held */ -static void ice_dis_vsi(struct ice_vsi *vsi, bool locked) +static int ice_ena_vsi(struct ice_vsi *vsi, bool locked) { - if (test_bit(__ICE_DOWN, vsi->state)) - return; + int err = 0; - set_bit(__ICE_NEEDS_RESTART, vsi->state); + if (!test_bit(__ICE_NEEDS_RESTART, vsi->state)) + return err; + + clear_bit(__ICE_NEEDS_RESTART, vsi->state); + + if (vsi->netdev && vsi->type == ICE_VSI_PF) { + struct net_device *netd = vsi->netdev; - if (vsi->type == ICE_VSI_PF && vsi->netdev) { if (netif_running(vsi->netdev)) { - if (!locked) { + if (locked) { + err = netd->netdev_ops->ndo_open(netd); + } else { rtnl_lock(); - vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); + err = netd->netdev_ops->ndo_open(netd); rtnl_unlock(); - } else { - vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); } } else { - ice_vsi_close(vsi); - } - } -} - -/** - * ice_ena_vsi - resume a VSI - * @vsi: the VSI being resume - */ -static int ice_ena_vsi(struct ice_vsi *vsi) -{ - int err = 0; - - if (test_and_clear_bit(__ICE_NEEDS_RESTART, vsi->state) && - vsi->netdev) { - if (netif_running(vsi->netdev)) { - rtnl_lock(); - err = vsi->netdev->netdev_ops->ndo_open(vsi->netdev); - rtnl_unlock(); - } else { err = ice_vsi_open(vsi); } } @@ -3338,36 +3693,28 @@ static int ice_ena_vsi(struct ice_vsi *vsi) } /** - * ice_pf_dis_all_vsi - Pause all VSIs on a PF - * @pf: the PF - */ -static void ice_pf_dis_all_vsi(struct ice_pf *pf) -{ - int v; - - ice_for_each_vsi(pf, v) - if (pf->vsi[v]) - ice_dis_vsi(pf->vsi[v], false); -} - -/** * ice_pf_ena_all_vsi - Resume all VSIs on a PF * @pf: the PF + * @locked: is the rtnl_lock already held */ -static int ice_pf_ena_all_vsi(struct ice_pf *pf) +#ifdef CONFIG_DCB +int ice_pf_ena_all_vsi(struct ice_pf *pf, bool locked) +#else +static int ice_pf_ena_all_vsi(struct ice_pf *pf, bool locked) +#endif /* CONFIG_DCB */ { int v; ice_for_each_vsi(pf, v) if (pf->vsi[v]) - if (ice_ena_vsi(pf->vsi[v])) + if (ice_ena_vsi(pf->vsi[v], locked)) return -EIO; return 0; } /** - * ice_vsi_rebuild_all - rebuild all VSIs in pf + * ice_vsi_rebuild_all - rebuild all VSIs in PF * @pf: the PF */ static int ice_vsi_rebuild_all(struct ice_pf *pf) @@ -3375,16 +3722,12 @@ static int ice_vsi_rebuild_all(struct ice_pf *pf) int i; /* loop through pf->vsi array and reinit the VSI if found */ - for (i = 0; i < pf->num_alloc_vsi; i++) { + ice_for_each_vsi(pf, i) { int err; if (!pf->vsi[i]) continue; - /* VF VSI rebuild isn't supported yet */ - if (pf->vsi[i]->type == ICE_VSI_VF) - continue; - err = ice_vsi_rebuild(pf->vsi[i]); if (err) { dev_err(&pf->pdev->dev, @@ -3412,7 +3755,7 @@ static int ice_vsi_replay_all(struct ice_pf *pf) int i; /* loop through pf->vsi array and replay the VSI if found */ - for (i = 0; i < pf->num_alloc_vsi; i++) { + ice_for_each_vsi(pf, i) { if (!pf->vsi[i]) continue; @@ -3441,7 +3784,7 @@ static int ice_vsi_replay_all(struct ice_pf *pf) /** * ice_rebuild - rebuild after reset - * @pf: pf to rebuild + * @pf: PF to rebuild */ static void ice_rebuild(struct ice_pf *pf) { @@ -3453,7 +3796,7 @@ static void ice_rebuild(struct ice_pf *pf) if (test_bit(__ICE_DOWN, pf->state)) goto clear_recovery; - dev_dbg(dev, "rebuilding pf\n"); + dev_dbg(dev, "rebuilding PF\n"); ret = ice_init_all_ctrlq(hw); if (ret) { @@ -3479,11 +3822,7 @@ static void ice_rebuild(struct ice_pf *pf) if (err) goto err_sched_init_port; - /* reset search_hint of irq_trackers to 0 since interrupts are - * reclaimed and could be allocated from beginning during VSI rebuild - */ - pf->sw_irq_tracker->search_hint = 0; - pf->hw_irq_tracker->search_hint = 0; + ice_dcb_rebuild(pf); err = ice_vsi_rebuild_all(pf); if (err) { @@ -3512,7 +3851,7 @@ static void ice_rebuild(struct ice_pf *pf) } /* restart the VSIs that were rebuilt and running before the reset */ - err = ice_pf_ena_all_vsi(pf); + err = ice_pf_ena_all_vsi(pf, false); if (err) { dev_err(&pf->pdev->dev, "error enabling VSIs\n"); /* no need to disable VSIs in tear down path in ice_rebuild() @@ -3521,9 +3860,7 @@ static void ice_rebuild(struct ice_pf *pf) goto err_vsi_rebuild; } - ice_reset_all_vfs(pf, true); - - for (i = 0; i < pf->num_alloc_vsi; i++) { + ice_for_each_vsi(pf, i) { bool link_up; if (!pf->vsi[i] || pf->vsi[i]->type != ICE_VSI_PF) @@ -3570,16 +3907,16 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) u8 count = 0; if (new_mtu == netdev->mtu) { - netdev_warn(netdev, "mtu is already %u\n", netdev->mtu); + netdev_warn(netdev, "MTU is already %u\n", netdev->mtu); return 0; } if (new_mtu < netdev->min_mtu) { - netdev_err(netdev, "new mtu invalid. min_mtu is %d\n", + netdev_err(netdev, "new MTU invalid. min_mtu is %d\n", netdev->min_mtu); return -EINVAL; } else if (new_mtu > netdev->max_mtu) { - netdev_err(netdev, "new mtu invalid. max_mtu is %d\n", + netdev_err(netdev, "new MTU invalid. max_mtu is %d\n", netdev->min_mtu); return -EINVAL; } @@ -3595,7 +3932,7 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) } while (count < 100); if (count == 100) { - netdev_err(netdev, "can't change mtu. Device is busy\n"); + netdev_err(netdev, "can't change MTU. Device is busy\n"); return -EBUSY; } @@ -3607,18 +3944,18 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) err = ice_down(vsi); if (err) { - netdev_err(netdev, "change mtu if_up err %d\n", err); + netdev_err(netdev, "change MTU if_up err %d\n", err); return err; } err = ice_up(vsi); if (err) { - netdev_err(netdev, "change mtu if_up err %d\n", err); + netdev_err(netdev, "change MTU if_up err %d\n", err); return err; } } - netdev_dbg(netdev, "changed mtu to %d\n", new_mtu); + netdev_info(netdev, "changed MTU to %d\n", new_mtu); return 0; } @@ -3710,7 +4047,7 @@ int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size) /** * ice_bridge_getlink - Get the hardware bridge mode * @skb: skb buff - * @pid: process id + * @pid: process ID * @seq: RTNL message seq * @dev: the netdev being configured * @filter_mask: filter mask passed in @@ -3909,8 +4246,7 @@ static void ice_tx_timeout(struct net_device *netdev) /* Read interrupt register */ if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) val = rd32(hw, - GLINT_DYN_CTL(tx_ring->q_vector->v_idx + - tx_ring->vsi->hw_base_vector)); + GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %d, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n", vsi->vsi_num, hung_queue, tx_ring->next_to_clean, @@ -3955,7 +4291,7 @@ static void ice_tx_timeout(struct net_device *netdev) * * Returns 0 on success, negative value on failure */ -static int ice_open(struct net_device *netdev) +int ice_open(struct net_device *netdev) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; @@ -3992,7 +4328,7 @@ static int ice_open(struct net_device *netdev) * * Returns success only - not allowed to fail */ -static int ice_stop(struct net_device *netdev) +int ice_stop(struct net_device *netdev) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; diff --git a/drivers/net/ethernet/intel/ice/ice_nvm.c b/drivers/net/ethernet/intel/ice/ice_nvm.c index 413fdbbcc4d0..bcb431f1bd92 100644 --- a/drivers/net/ethernet/intel/ice/ice_nvm.c +++ b/drivers/net/ethernet/intel/ice/ice_nvm.c @@ -5,7 +5,7 @@ /** * ice_aq_read_nvm - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @module_typeid: module pointer location in words from the NVM beginning * @offset: byte offset from the module beginning * @length: length of the section to be read (in bytes from the offset) @@ -119,7 +119,7 @@ ice_read_sr_word_aq(struct ice_hw *hw, u16 offset, u16 *data) status = ice_read_sr_aq(hw, offset, 1, data, true); if (!status) - *data = le16_to_cpu(*(__le16 *)data); + *data = le16_to_cpu(*(__force __le16 *)data); return status; } @@ -174,7 +174,7 @@ ice_read_sr_buf_aq(struct ice_hw *hw, u16 offset, u16 *words, u16 *data) } while (words_read < *words); for (i = 0; i < *words; i++) - data[i] = le16_to_cpu(((__le16 *)data)[i]); + data[i] = le16_to_cpu(((__force __le16 *)data)[i]); read_nvm_buf_aq_exit: *words = words_read; @@ -235,7 +235,7 @@ ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data) /** * ice_init_nvm - initializes NVM setting - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * This function reads and populates NVM settings such as Shadow RAM size, * max_timeout, and blank_nvm_mode @@ -248,7 +248,7 @@ enum ice_status ice_init_nvm(struct ice_hw *hw) u32 fla, gens_stat; u8 sr_size; - /* The SR size is stored regardless of the nvm programming mode + /* The SR size is stored regardless of the NVM programming mode * as the blank mode may be used in the factory line. */ gens_stat = rd32(hw, GLNVM_GENS); @@ -316,3 +316,34 @@ ice_read_sr_buf(struct ice_hw *hw, u16 offset, u16 *words, u16 *data) return status; } + +/** + * ice_nvm_validate_checksum + * @hw: pointer to the HW struct + * + * Verify NVM PFA checksum validity (0x0706) + */ +enum ice_status ice_nvm_validate_checksum(struct ice_hw *hw) +{ + struct ice_aqc_nvm_checksum *cmd; + struct ice_aq_desc desc; + enum ice_status status; + + status = ice_acquire_nvm(hw, ICE_RES_READ); + if (status) + return status; + + cmd = &desc.params.nvm_checksum; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_nvm_checksum); + cmd->flags = ICE_AQC_NVM_CHECKSUM_VERIFY; + + status = ice_aq_send_cmd(hw, &desc, NULL, 0, NULL); + ice_release_nvm(hw); + + if (!status) + if (le16_to_cpu(cmd->checksum) != ICE_AQC_NVM_CHECKSUM_CORRECT) + status = ICE_ERR_NVM_CHECKSUM; + + return status; +} diff --git a/drivers/net/ethernet/intel/ice/ice_sched.c b/drivers/net/ethernet/intel/ice/ice_sched.c index 56049739a250..2a232504379d 100644 --- a/drivers/net/ethernet/intel/ice/ice_sched.c +++ b/drivers/net/ethernet/intel/ice/ice_sched.c @@ -43,9 +43,9 @@ ice_sched_add_root_node(struct ice_port_info *pi, /** * ice_sched_find_node_by_teid - Find the Tx scheduler node in SW DB * @start_node: pointer to the starting ice_sched_node struct in a sub-tree - * @teid: node teid to search + * @teid: node TEID to search * - * This function searches for a node matching the teid in the scheduling tree + * This function searches for a node matching the TEID in the scheduling tree * from the SW DB. The search is recursive and is restricted by the number of * layers it has searched through; stopping at the max supported layer. * @@ -66,7 +66,7 @@ ice_sched_find_node_by_teid(struct ice_sched_node *start_node, u32 teid) start_node->info.data.elem_type == ICE_AQC_ELEM_TYPE_LEAF) return NULL; - /* Check if teid matches to any of the children nodes */ + /* Check if TEID matches to any of the children nodes */ for (i = 0; i < start_node->num_children; i++) if (ICE_TXSCHED_GET_NODE_TEID(start_node->children[i]) == teid) return start_node->children[i]; @@ -86,7 +86,7 @@ ice_sched_find_node_by_teid(struct ice_sched_node *start_node, u32 teid) /** * ice_aqc_send_sched_elem_cmd - send scheduling elements cmd - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @cmd_opc: cmd opcode * @elems_req: number of elements to request * @buf: pointer to buffer @@ -118,7 +118,7 @@ ice_aqc_send_sched_elem_cmd(struct ice_hw *hw, enum ice_adminq_opc cmd_opc, /** * ice_aq_query_sched_elems - query scheduler elements - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @elems_req: number of elements to query * @buf: pointer to buffer * @buf_size: buffer size in bytes @@ -127,7 +127,7 @@ ice_aqc_send_sched_elem_cmd(struct ice_hw *hw, enum ice_adminq_opc cmd_opc, * * Query scheduling elements (0x0404) */ -static enum ice_status +enum ice_status ice_aq_query_sched_elems(struct ice_hw *hw, u16 elems_req, struct ice_aqc_get_elem *buf, u16 buf_size, u16 *elems_ret, struct ice_sq_cd *cd) @@ -138,31 +138,6 @@ ice_aq_query_sched_elems(struct ice_hw *hw, u16 elems_req, } /** - * ice_sched_query_elem - query element information from hw - * @hw: pointer to the hw struct - * @node_teid: node teid to be queried - * @buf: buffer to element information - * - * This function queries HW element information - */ -static enum ice_status -ice_sched_query_elem(struct ice_hw *hw, u32 node_teid, - struct ice_aqc_get_elem *buf) -{ - u16 buf_size, num_elem_ret = 0; - enum ice_status status; - - buf_size = sizeof(*buf); - memset(buf, 0, buf_size); - buf->generic[0].node_teid = cpu_to_le32(node_teid); - status = ice_aq_query_sched_elems(hw, 1, buf, buf_size, &num_elem_ret, - NULL); - if (status || num_elem_ret != 1) - ice_debug(hw, ICE_DBG_SCHED, "query element failed\n"); - return status; -} - -/** * ice_sched_add_node - Insert the Tx scheduler node in SW DB * @pi: port information structure * @layer: Scheduler layer of the node @@ -226,7 +201,7 @@ ice_sched_add_node(struct ice_port_info *pi, u8 layer, /** * ice_aq_delete_sched_elems - delete scheduler elements - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @grps_req: number of groups to delete * @buf: pointer to buffer * @buf_size: buffer size in bytes @@ -246,13 +221,13 @@ ice_aq_delete_sched_elems(struct ice_hw *hw, u16 grps_req, } /** - * ice_sched_remove_elems - remove nodes from hw - * @hw: pointer to the hw struct + * ice_sched_remove_elems - remove nodes from HW + * @hw: pointer to the HW struct * @parent: pointer to the parent node * @num_nodes: number of nodes * @node_teids: array of node teids to be deleted * - * This function remove nodes from hw + * This function remove nodes from HW */ static enum ice_status ice_sched_remove_elems(struct ice_hw *hw, struct ice_sched_node *parent, @@ -276,7 +251,8 @@ ice_sched_remove_elems(struct ice_hw *hw, struct ice_sched_node *parent, status = ice_aq_delete_sched_elems(hw, 1, buf, buf_size, &num_groups_removed, NULL); if (status || num_groups_removed != 1) - ice_debug(hw, ICE_DBG_SCHED, "remove elements failed\n"); + ice_debug(hw, ICE_DBG_SCHED, "remove node failed FW error %d\n", + hw->adminq.sq_last_status); devm_kfree(ice_hw_to_dev(hw), buf); return status; @@ -284,7 +260,7 @@ ice_sched_remove_elems(struct ice_hw *hw, struct ice_sched_node *parent, /** * ice_sched_get_first_node - get the first node of the given layer - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @parent: pointer the base node of the subtree * @layer: layer number * @@ -360,12 +336,8 @@ void ice_free_sched_node(struct ice_port_info *pi, struct ice_sched_node *node) node->info.data.elem_type != ICE_AQC_ELEM_TYPE_ROOT_PORT && node->info.data.elem_type != ICE_AQC_ELEM_TYPE_LEAF) { u32 teid = le32_to_cpu(node->info.node_teid); - enum ice_status status; - status = ice_sched_remove_elems(hw, node->parent, 1, &teid); - if (status) - ice_debug(hw, ICE_DBG_SCHED, - "remove element failed %d\n", status); + ice_sched_remove_elems(hw, node->parent, 1, &teid); } parent = node->parent; /* root has no parent */ @@ -409,7 +381,7 @@ err_exit: /** * ice_aq_get_dflt_topo - gets default scheduler topology - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @lport: logical port number * @buf: pointer to buffer * @buf_size: buffer size in bytes @@ -439,7 +411,7 @@ ice_aq_get_dflt_topo(struct ice_hw *hw, u8 lport, /** * ice_aq_add_sched_elems - adds scheduling element - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @grps_req: the number of groups that are requested to be added * @buf: pointer to buffer * @buf_size: buffer size in bytes @@ -460,7 +432,7 @@ ice_aq_add_sched_elems(struct ice_hw *hw, u16 grps_req, /** * ice_aq_suspend_sched_elems - suspend scheduler elements - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @elems_req: number of elements to suspend * @buf: pointer to buffer * @buf_size: buffer size in bytes @@ -481,7 +453,7 @@ ice_aq_suspend_sched_elems(struct ice_hw *hw, u16 elems_req, /** * ice_aq_resume_sched_elems - resume scheduler elements - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @elems_req: number of elements to resume * @buf: pointer to buffer * @buf_size: buffer size in bytes @@ -502,7 +474,7 @@ ice_aq_resume_sched_elems(struct ice_hw *hw, u16 elems_req, /** * ice_aq_query_sched_res - query scheduler resource - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @buf_size: buffer size in bytes * @buf: pointer to buffer * @cd: pointer to command details structure or NULL @@ -521,13 +493,13 @@ ice_aq_query_sched_res(struct ice_hw *hw, u16 buf_size, } /** - * ice_sched_suspend_resume_elems - suspend or resume hw nodes - * @hw: pointer to the hw struct + * ice_sched_suspend_resume_elems - suspend or resume HW nodes + * @hw: pointer to the HW struct * @num_nodes: number of nodes * @node_teids: array of node teids to be suspended or resumed * @suspend: true means suspend / false means resume * - * This function suspends or resumes hw nodes + * This function suspends or resumes HW nodes */ static enum ice_status ice_sched_suspend_resume_elems(struct ice_hw *hw, u8 num_nodes, u32 *node_teids, @@ -561,10 +533,54 @@ ice_sched_suspend_resume_elems(struct ice_hw *hw, u8 num_nodes, u32 *node_teids, } /** - * ice_sched_clear_agg - clears the agg related information + * ice_alloc_lan_q_ctx - allocate LAN queue contexts for the given VSI and TC + * @hw: pointer to the HW struct + * @vsi_handle: VSI handle + * @tc: TC number + * @new_numqs: number of queues + */ +static enum ice_status +ice_alloc_lan_q_ctx(struct ice_hw *hw, u16 vsi_handle, u8 tc, u16 new_numqs) +{ + struct ice_vsi_ctx *vsi_ctx; + struct ice_q_ctx *q_ctx; + + vsi_ctx = ice_get_vsi_ctx(hw, vsi_handle); + if (!vsi_ctx) + return ICE_ERR_PARAM; + /* allocate LAN queue contexts */ + if (!vsi_ctx->lan_q_ctx[tc]) { + vsi_ctx->lan_q_ctx[tc] = devm_kcalloc(ice_hw_to_dev(hw), + new_numqs, + sizeof(*q_ctx), + GFP_KERNEL); + if (!vsi_ctx->lan_q_ctx[tc]) + return ICE_ERR_NO_MEMORY; + vsi_ctx->num_lan_q_entries[tc] = new_numqs; + return 0; + } + /* num queues are increased, update the queue contexts */ + if (new_numqs > vsi_ctx->num_lan_q_entries[tc]) { + u16 prev_num = vsi_ctx->num_lan_q_entries[tc]; + + q_ctx = devm_kcalloc(ice_hw_to_dev(hw), new_numqs, + sizeof(*q_ctx), GFP_KERNEL); + if (!q_ctx) + return ICE_ERR_NO_MEMORY; + memcpy(q_ctx, vsi_ctx->lan_q_ctx[tc], + prev_num * sizeof(*q_ctx)); + devm_kfree(ice_hw_to_dev(hw), vsi_ctx->lan_q_ctx[tc]); + vsi_ctx->lan_q_ctx[tc] = q_ctx; + vsi_ctx->num_lan_q_entries[tc] = new_numqs; + } + return 0; +} + +/** + * ice_sched_clear_agg - clears the aggregator related information * @hw: pointer to the hardware structure * - * This function removes agg list and free up agg related memory + * This function removes aggregator list and free up aggregator related memory * previously allocated. */ void ice_sched_clear_agg(struct ice_hw *hw) @@ -622,7 +638,7 @@ void ice_sched_clear_port(struct ice_port_info *pi) /** * ice_sched_cleanup_all - cleanup scheduler elements from SW DB for all ports - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Cleanup scheduling elements from SW DB for all the ports */ @@ -646,16 +662,16 @@ void ice_sched_cleanup_all(struct ice_hw *hw) } /** - * ice_sched_add_elems - add nodes to hw and SW DB + * ice_sched_add_elems - add nodes to HW and SW DB * @pi: port information structure * @tc_node: pointer to the branch node * @parent: pointer to the parent node * @layer: layer number to add nodes * @num_nodes: number of nodes * @num_nodes_added: pointer to num nodes added - * @first_node_teid: if new nodes are added then return the teid of first node + * @first_node_teid: if new nodes are added then return the TEID of first node * - * This function add nodes to hw as well as to SW DB for a given layer + * This function add nodes to HW as well as to SW DB for a given layer */ static enum ice_status ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node, @@ -667,10 +683,10 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node, u16 i, num_groups_added = 0; enum ice_status status = 0; struct ice_hw *hw = pi->hw; - u16 buf_size; + size_t buf_size; u32 teid; - buf_size = sizeof(*buf) + sizeof(*buf->generic) * (num_nodes - 1); + buf_size = struct_size(buf, generic, num_nodes - 1); buf = devm_kzalloc(ice_hw_to_dev(hw), buf_size, GFP_KERNEL); if (!buf) return ICE_ERR_NO_MEMORY; @@ -697,7 +713,8 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node, status = ice_aq_add_sched_elems(hw, 1, buf, buf_size, &num_groups_added, NULL); if (status || num_groups_added != 1) { - ice_debug(hw, ICE_DBG_SCHED, "add elements failed\n"); + ice_debug(hw, ICE_DBG_SCHED, "add node failed FW Error %d\n", + hw->adminq.sq_last_status); devm_kfree(ice_hw_to_dev(hw), buf); return ICE_ERR_CFG; } @@ -748,7 +765,7 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node, * @parent: pointer to parent node * @layer: layer number to add nodes * @num_nodes: number of nodes to be added - * @first_node_teid: pointer to the first node teid + * @first_node_teid: pointer to the first node TEID * @num_nodes_added: pointer to number of nodes added * * This function add nodes to a given layer. @@ -800,7 +817,7 @@ ice_sched_add_nodes_to_layer(struct ice_port_info *pi, *num_nodes_added += num_added; } - /* Don't modify the first node teid memory if the first node was + /* Don't modify the first node TEID memory if the first node was * added already in the above call. Instead send some temp * memory for all other recursive calls. */ @@ -832,7 +849,7 @@ ice_sched_add_nodes_to_layer(struct ice_port_info *pi, /** * ice_sched_get_qgrp_layer - get the current queue group layer number - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * This function returns the current queue group layer number */ @@ -844,7 +861,7 @@ static u8 ice_sched_get_qgrp_layer(struct ice_hw *hw) /** * ice_sched_get_vsi_layer - get the current VSI layer number - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * This function returns the current VSI layer number */ @@ -855,7 +872,7 @@ static u8 ice_sched_get_vsi_layer(struct ice_hw *hw) * 7 4 * 5 or less sw_entry_point_layer */ - /* calculate the vsi layer based on number of layers. */ + /* calculate the VSI layer based on number of layers. */ if (hw->num_tx_sched_layers > ICE_VSI_LAYER_OFFSET + 1) { u8 layer = hw->num_tx_sched_layers - ICE_VSI_LAYER_OFFSET; @@ -973,7 +990,7 @@ enum ice_status ice_sched_init_port(struct ice_port_info *pi) goto err_init_port; } - /* If the last node is a leaf node then the index of the Q group + /* If the last node is a leaf node then the index of the queue group * layer is two less than the number of elements. */ if (num_elems > 2 && buf[0].generic[num_elems - 1].data.elem_type == @@ -1082,7 +1099,7 @@ sched_query_out: /** * ice_sched_find_node_in_subtree - Find node in part of base node subtree - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @base: pointer to the base node * @node: pointer to the node to search * @@ -1114,13 +1131,13 @@ ice_sched_find_node_in_subtree(struct ice_hw *hw, struct ice_sched_node *base, } /** - * ice_sched_get_free_qparent - Get a free lan or rdma q group node + * ice_sched_get_free_qparent - Get a free LAN or RDMA queue group node * @pi: port information structure * @vsi_handle: software VSI handle * @tc: branch number - * @owner: lan or rdma + * @owner: LAN or RDMA * - * This function retrieves a free lan or rdma q group node + * This function retrieves a free LAN or RDMA queue group node */ struct ice_sched_node * ice_sched_get_free_qparent(struct ice_port_info *pi, u16 vsi_handle, u8 tc, @@ -1138,11 +1155,11 @@ ice_sched_get_free_qparent(struct ice_port_info *pi, u16 vsi_handle, u8 tc, if (!vsi_ctx) return NULL; vsi_node = vsi_ctx->sched.vsi_node[tc]; - /* validate invalid VSI id */ + /* validate invalid VSI ID */ if (!vsi_node) goto lan_q_exit; - /* get the first q group node from VSI sub-tree */ + /* get the first queue group node from VSI sub-tree */ qgrp_node = ice_sched_get_first_node(pi->hw, vsi_node, qgrp_layer); while (qgrp_node) { /* make sure the qgroup node is part of the VSI subtree */ @@ -1158,12 +1175,12 @@ lan_q_exit: } /** - * ice_sched_get_vsi_node - Get a VSI node based on VSI id - * @hw: pointer to the hw struct + * ice_sched_get_vsi_node - Get a VSI node based on VSI ID + * @hw: pointer to the HW struct * @tc_node: pointer to the TC node * @vsi_handle: software VSI handle * - * This function retrieves a VSI node for a given VSI id from a given + * This function retrieves a VSI node for a given VSI ID from a given * TC branch */ static struct ice_sched_node * @@ -1188,7 +1205,7 @@ ice_sched_get_vsi_node(struct ice_hw *hw, struct ice_sched_node *tc_node, /** * ice_sched_calc_vsi_child_nodes - calculate number of VSI child nodes - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @num_qs: number of queues * @num_nodes: num nodes array * @@ -1204,7 +1221,7 @@ ice_sched_calc_vsi_child_nodes(struct ice_hw *hw, u16 num_qs, u16 *num_nodes) qgl = ice_sched_get_qgrp_layer(hw); vsil = ice_sched_get_vsi_layer(hw); - /* calculate num nodes from q group to VSI layer */ + /* calculate num nodes from queue group to VSI layer */ for (i = qgl; i > vsil; i--) { /* round to the next integer if there is a remainder */ num = DIV_ROUND_UP(num, hw->max_children[i]); @@ -1220,10 +1237,10 @@ ice_sched_calc_vsi_child_nodes(struct ice_hw *hw, u16 num_qs, u16 *num_nodes) * @vsi_handle: software VSI handle * @tc_node: pointer to the TC node * @num_nodes: pointer to the num nodes that needs to be added per layer - * @owner: node owner (lan or rdma) + * @owner: node owner (LAN or RDMA) * * This function adds the VSI child nodes to tree. It gets called for - * lan and rdma separately. + * LAN and RDMA separately. */ static enum ice_status ice_sched_add_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle, @@ -1271,44 +1288,8 @@ ice_sched_add_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle, } /** - * ice_sched_rm_vsi_child_nodes - remove VSI child nodes from the tree - * @pi: port information structure - * @vsi_node: pointer to the VSI node - * @num_nodes: pointer to the num nodes that needs to be removed per layer - * @owner: node owner (lan or rdma) - * - * This function removes the VSI child nodes from the tree. It gets called for - * lan and rdma separately. - */ -static void -ice_sched_rm_vsi_child_nodes(struct ice_port_info *pi, - struct ice_sched_node *vsi_node, u16 *num_nodes, - u8 owner) -{ - struct ice_sched_node *node, *next; - u8 i, qgl, vsil; - u16 num; - - qgl = ice_sched_get_qgrp_layer(pi->hw); - vsil = ice_sched_get_vsi_layer(pi->hw); - - for (i = qgl; i > vsil; i--) { - num = num_nodes[i]; - node = ice_sched_get_first_node(pi->hw, vsi_node, i); - while (node && num) { - next = node->sibling; - if (node->owner == owner && !node->num_children) { - ice_free_sched_node(pi, node); - num--; - } - node = next; - } - } -} - -/** * ice_sched_calc_vsi_support_nodes - calculate number of VSI support nodes - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @tc_node: pointer to TC node * @num_nodes: pointer to num nodes array * @@ -1427,7 +1408,7 @@ ice_sched_add_vsi_to_topo(struct ice_port_info *pi, u16 vsi_handle, u8 tc) /* calculate number of supported nodes needed for this VSI */ ice_sched_calc_vsi_support_nodes(hw, tc_node, num_nodes); - /* add vsi supported nodes to tc subtree */ + /* add VSI supported nodes to TC subtree */ return ice_sched_add_vsi_support_nodes(pi, vsi_handle, tc_node, num_nodes); } @@ -1446,7 +1427,6 @@ static enum ice_status ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 new_numqs, u8 owner) { - u16 prev_num_nodes[ICE_AQC_TOPO_MAX_LEVEL_NUM] = { 0 }; u16 new_num_nodes[ICE_AQC_TOPO_MAX_LEVEL_NUM] = { 0 }; struct ice_sched_node *vsi_node; struct ice_sched_node *tc_node; @@ -1454,7 +1434,6 @@ ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle, enum ice_status status = 0; struct ice_hw *hw = pi->hw; u16 prev_numqs; - u8 i; tc_node = ice_sched_get_tc_node(pi, tc); if (!tc_node) @@ -1468,41 +1447,30 @@ ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle, if (!vsi_ctx) return ICE_ERR_PARAM; - if (owner == ICE_SCHED_NODE_OWNER_LAN) - prev_numqs = vsi_ctx->sched.max_lanq[tc]; - else - return ICE_ERR_PARAM; - - /* num queues are not changed */ - if (prev_numqs == new_numqs) + prev_numqs = vsi_ctx->sched.max_lanq[tc]; + /* num queues are not changed or less than the previous number */ + if (new_numqs <= prev_numqs) + return status; + status = ice_alloc_lan_q_ctx(hw, vsi_handle, tc, new_numqs); + if (status) return status; - - /* calculate number of nodes based on prev/new number of qs */ - if (prev_numqs) - ice_sched_calc_vsi_child_nodes(hw, prev_numqs, prev_num_nodes); if (new_numqs) ice_sched_calc_vsi_child_nodes(hw, new_numqs, new_num_nodes); - - if (prev_numqs > new_numqs) { - for (i = 0; i < ICE_AQC_TOPO_MAX_LEVEL_NUM; i++) - new_num_nodes[i] = prev_num_nodes[i] - new_num_nodes[i]; - - ice_sched_rm_vsi_child_nodes(pi, vsi_node, new_num_nodes, - owner); - } else { - for (i = 0; i < ICE_AQC_TOPO_MAX_LEVEL_NUM; i++) - new_num_nodes[i] -= prev_num_nodes[i]; - - status = ice_sched_add_vsi_child_nodes(pi, vsi_handle, tc_node, - new_num_nodes, owner); - if (status) - return status; - } - + /* Keep the max number of queue configuration all the time. Update the + * tree only if number of queues > previous number of queues. This may + * leave some extra nodes in the tree if number of queues < previous + * number but that wouldn't harm anything. Removing those extra nodes + * may complicate the code if those nodes are part of SRL or + * individually rate limited. + */ + status = ice_sched_add_vsi_child_nodes(pi, vsi_handle, tc_node, + new_num_nodes, owner); + if (status) + return status; vsi_ctx->sched.max_lanq[tc] = new_numqs; - return status; + return 0; } /** @@ -1511,7 +1479,7 @@ ice_sched_update_vsi_child_nodes(struct ice_port_info *pi, u16 vsi_handle, * @vsi_handle: software VSI handle * @tc: TC number * @maxqs: max number of queues - * @owner: lan or rdma + * @owner: LAN or RDMA * @enable: TC enabled or disabled * * This function adds/updates VSI nodes based on the number of queues. If TC is @@ -1527,6 +1495,7 @@ ice_sched_cfg_vsi(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 maxqs, enum ice_status status = 0; struct ice_hw *hw = pi->hw; + ice_debug(pi->hw, ICE_DBG_SCHED, "add/config VSI %d\n", vsi_handle); tc_node = ice_sched_get_tc_node(pi, tc); if (!tc_node) return ICE_ERR_PARAM; @@ -1535,7 +1504,7 @@ ice_sched_cfg_vsi(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 maxqs, return ICE_ERR_PARAM; vsi_node = ice_sched_get_vsi_node(hw, tc_node, vsi_handle); - /* suspend the VSI if tc is not enabled */ + /* suspend the VSI if TC is not enabled */ if (!enable) { if (vsi_node && vsi_node->in_use) { u32 teid = le32_to_cpu(vsi_node->info.node_teid); @@ -1586,7 +1555,7 @@ ice_sched_cfg_vsi(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 maxqs, } /** - * ice_sched_rm_agg_vsi_entry - remove agg related VSI info entry + * ice_sched_rm_agg_vsi_entry - remove aggregator related VSI info entry * @pi: port information structure * @vsi_handle: software VSI handle * @@ -1646,8 +1615,9 @@ ice_sched_rm_vsi_cfg(struct ice_port_info *pi, u16 vsi_handle, u8 owner) { enum ice_status status = ICE_ERR_PARAM; struct ice_vsi_ctx *vsi_ctx; - u8 i, j = 0; + u8 i; + ice_debug(pi->hw, ICE_DBG_SCHED, "removing VSI %d\n", vsi_handle); if (!ice_is_vsi_valid(pi->hw, vsi_handle)) return status; mutex_lock(&pi->sched_lock); @@ -1655,8 +1625,9 @@ ice_sched_rm_vsi_cfg(struct ice_port_info *pi, u16 vsi_handle, u8 owner) if (!vsi_ctx) goto exit_sched_rm_vsi_cfg; - for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) { + ice_for_each_traffic_class(i) { struct ice_sched_node *vsi_node, *tc_node; + u8 j = 0; tc_node = ice_sched_get_tc_node(pi, i); if (!tc_node) @@ -1689,7 +1660,7 @@ ice_sched_rm_vsi_cfg(struct ice_port_info *pi, u16 vsi_handle, u8 owner) ice_free_sched_node(pi, vsi_node); vsi_ctx->sched.vsi_node[i] = NULL; - /* clean up agg related vsi info if any */ + /* clean up aggregator related VSI info if any */ ice_sched_rm_agg_vsi_info(pi, vsi_handle); } if (owner == ICE_SCHED_NODE_OWNER_LAN) diff --git a/drivers/net/ethernet/intel/ice/ice_sched.h b/drivers/net/ethernet/intel/ice/ice_sched.h index bee8221ad146..3902a8ad3025 100644 --- a/drivers/net/ethernet/intel/ice/ice_sched.h +++ b/drivers/net/ethernet/intel/ice/ice_sched.h @@ -24,6 +24,10 @@ struct ice_sched_agg_info { }; /* FW AQ command calls */ +enum ice_status +ice_aq_query_sched_elems(struct ice_hw *hw, u16 elems_req, + struct ice_aqc_get_elem *buf, u16 buf_size, + u16 *elems_ret, struct ice_sq_cd *cd); enum ice_status ice_sched_init_port(struct ice_port_info *pi); enum ice_status ice_sched_query_res_alloc(struct ice_hw *hw); void ice_sched_clear_port(struct ice_port_info *pi); diff --git a/drivers/net/ethernet/intel/ice/ice_status.h b/drivers/net/ethernet/intel/ice/ice_status.h index 683f48824a29..c01597885629 100644 --- a/drivers/net/ethernet/intel/ice/ice_status.h +++ b/drivers/net/ethernet/intel/ice/ice_status.h @@ -12,6 +12,7 @@ enum ice_status { ICE_ERR_PARAM = -1, ICE_ERR_NOT_IMPL = -2, ICE_ERR_NOT_READY = -3, + ICE_ERR_NOT_SUPPORTED = -4, ICE_ERR_BAD_PTR = -5, ICE_ERR_INVAL_SIZE = -6, ICE_ERR_DEVICE_NOT_SUPPORTED = -8, @@ -25,6 +26,7 @@ enum ice_status { ICE_ERR_IN_USE = -16, ICE_ERR_MAX_LIMIT = -17, ICE_ERR_RESET_ONGOING = -18, + ICE_ERR_NVM_CHECKSUM = -51, ICE_ERR_BUF_TOO_SHORT = -52, ICE_ERR_NVM_BLANK_MODE = -53, ICE_ERR_AQ_ERROR = -100, diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c index 09d1c314b68f..8271fd651725 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.c +++ b/drivers/net/ethernet/intel/ice/ice_switch.c @@ -19,7 +19,7 @@ * byte 6 = 0x2: to identify it as locally administered SA MAC * byte 12 = 0x81 & byte 13 = 0x00: * In case of VLAN filter first two bytes defines ether type (0x8100) - * and remaining two bytes are placeholder for programming a given VLAN id + * and remaining two bytes are placeholder for programming a given VLAN ID * In case of Ether type filter it is treated as header without VLAN tag * and byte 12 and 13 is used to program a given Ether type instead */ @@ -51,7 +51,7 @@ static const u8 dummy_eth_header[DUMMY_ETH_HDR_LEN] = { 0x2, 0, 0, 0, 0, 0, /** * ice_aq_alloc_free_res - command to allocate/free resources - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @num_entries: number of resource entries in buffer * @buf: Indirect buffer to hold data parameters and response * @buf_size: size of buffer for indirect commands @@ -87,7 +87,7 @@ ice_aq_alloc_free_res(struct ice_hw *hw, u16 num_entries, /** * ice_init_def_sw_recp - initialize the recipe book keeping tables - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Allocate memory for the entire recipe table and initialize the structures/ * entries corresponding to basic recipes. @@ -163,7 +163,7 @@ ice_aq_get_sw_cfg(struct ice_hw *hw, struct ice_aqc_get_sw_cfg_resp *buf, /** * ice_aq_add_vsi - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_ctx: pointer to a VSI context struct * @cd: pointer to command details structure or NULL * @@ -206,7 +206,7 @@ ice_aq_add_vsi(struct ice_hw *hw, struct ice_vsi_ctx *vsi_ctx, /** * ice_aq_free_vsi - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_ctx: pointer to a VSI context struct * @keep_vsi_alloc: keep VSI allocation as part of this PF's resources * @cd: pointer to command details structure or NULL @@ -242,7 +242,7 @@ ice_aq_free_vsi(struct ice_hw *hw, struct ice_vsi_ctx *vsi_ctx, /** * ice_aq_update_vsi - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_ctx: pointer to a VSI context struct * @cd: pointer to command details structure or NULL * @@ -279,7 +279,7 @@ ice_aq_update_vsi(struct ice_hw *hw, struct ice_vsi_ctx *vsi_ctx, /** * ice_is_vsi_valid - check whether the VSI is valid or not - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: VSI handle * * check whether the VSI is valid or not @@ -290,11 +290,11 @@ bool ice_is_vsi_valid(struct ice_hw *hw, u16 vsi_handle) } /** - * ice_get_hw_vsi_num - return the hw VSI number - * @hw: pointer to the hw struct + * ice_get_hw_vsi_num - return the HW VSI number + * @hw: pointer to the HW struct * @vsi_handle: VSI handle * - * return the hw VSI number + * return the HW VSI number * Caution: call this function only if VSI is valid (ice_is_vsi_valid) */ u16 ice_get_hw_vsi_num(struct ice_hw *hw, u16 vsi_handle) @@ -304,7 +304,7 @@ u16 ice_get_hw_vsi_num(struct ice_hw *hw, u16 vsi_handle) /** * ice_get_vsi_ctx - return the VSI context entry for a given VSI handle - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: VSI handle * * return the VSI context entry for a given VSI handle @@ -316,21 +316,42 @@ struct ice_vsi_ctx *ice_get_vsi_ctx(struct ice_hw *hw, u16 vsi_handle) /** * ice_save_vsi_ctx - save the VSI context for a given VSI handle - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: VSI handle * @vsi: VSI context pointer * * save the VSI context entry for a given VSI handle */ -static void ice_save_vsi_ctx(struct ice_hw *hw, u16 vsi_handle, - struct ice_vsi_ctx *vsi) +static void +ice_save_vsi_ctx(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi) { hw->vsi_ctx[vsi_handle] = vsi; } /** + * ice_clear_vsi_q_ctx - clear VSI queue contexts for all TCs + * @hw: pointer to the HW struct + * @vsi_handle: VSI handle + */ +static void ice_clear_vsi_q_ctx(struct ice_hw *hw, u16 vsi_handle) +{ + struct ice_vsi_ctx *vsi; + u8 i; + + vsi = ice_get_vsi_ctx(hw, vsi_handle); + if (!vsi) + return; + ice_for_each_traffic_class(i) { + if (vsi->lan_q_ctx[i]) { + devm_kfree(ice_hw_to_dev(hw), vsi->lan_q_ctx[i]); + vsi->lan_q_ctx[i] = NULL; + } + } +} + +/** * ice_clear_vsi_ctx - clear the VSI context entry - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: VSI handle * * clear the VSI context entry @@ -341,6 +362,7 @@ static void ice_clear_vsi_ctx(struct ice_hw *hw, u16 vsi_handle) vsi = ice_get_vsi_ctx(hw, vsi_handle); if (vsi) { + ice_clear_vsi_q_ctx(hw, vsi_handle); devm_kfree(ice_hw_to_dev(hw), vsi); hw->vsi_ctx[vsi_handle] = NULL; } @@ -348,7 +370,7 @@ static void ice_clear_vsi_ctx(struct ice_hw *hw, u16 vsi_handle) /** * ice_clear_all_vsi_ctx - clear all the VSI context entries - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct */ void ice_clear_all_vsi_ctx(struct ice_hw *hw) { @@ -360,7 +382,7 @@ void ice_clear_all_vsi_ctx(struct ice_hw *hw) /** * ice_add_vsi - add VSI context to the hardware and VSI handle list - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: unique VSI handle provided by drivers * @vsi_ctx: pointer to a VSI context struct * @cd: pointer to command details structure or NULL @@ -383,7 +405,7 @@ ice_add_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx, return status; tmp_vsi_ctx = ice_get_vsi_ctx(hw, vsi_handle); if (!tmp_vsi_ctx) { - /* Create a new vsi context */ + /* Create a new VSI context */ tmp_vsi_ctx = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*tmp_vsi_ctx), GFP_KERNEL); if (!tmp_vsi_ctx) { @@ -398,12 +420,12 @@ ice_add_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx, tmp_vsi_ctx->vsi_num = vsi_ctx->vsi_num; } - return status; + return 0; } /** * ice_free_vsi- free VSI context from hardware and VSI handle list - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: unique VSI handle * @vsi_ctx: pointer to a VSI context struct * @keep_vsi_alloc: keep VSI allocation as part of this PF's resources @@ -428,7 +450,7 @@ ice_free_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx, /** * ice_update_vsi - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle: unique VSI handle * @vsi_ctx: pointer to a VSI context struct * @cd: pointer to command details structure or NULL @@ -447,8 +469,8 @@ ice_update_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx, /** * ice_aq_alloc_free_vsi_list - * @hw: pointer to the hw struct - * @vsi_list_id: VSI list id returned or used for lookup + * @hw: pointer to the HW struct + * @vsi_list_id: VSI list ID returned or used for lookup * @lkup_type: switch rule filter lookup type * @opc: switch rules population command type - pass in the command opcode * @@ -504,7 +526,7 @@ ice_aq_alloc_free_vsi_list_exit: /** * ice_aq_sw_rules - add/update/remove switch rules - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @rule_list: pointer to switch rule population list * @rule_list_sz: total size of the rule list in bytes * @num_rules: number of switch rules in the rule_list @@ -643,21 +665,43 @@ static void ice_fill_sw_info(struct ice_hw *hw, struct ice_fltr_info *fi) fi->fltr_act == ICE_FWD_TO_VSI_LIST || fi->fltr_act == ICE_FWD_TO_Q || fi->fltr_act == ICE_FWD_TO_QGRP)) { - fi->lb_en = true; - /* Do not set lan_en to TRUE if + /* Setting LB for prune actions will result in replicated + * packets to the internal switch that will be dropped. + */ + if (fi->lkup_type != ICE_SW_LKUP_VLAN) + fi->lb_en = true; + + /* Set lan_en to TRUE if * 1. The switch is a VEB AND * 2 - * 2.1 The lookup is MAC with unicast addr for MAC, OR - * 2.2 The lookup is MAC_VLAN with unicast addr for MAC + * 2.1 The lookup is a directional lookup like ethertype, + * promiscuous, ethertype-MAC, promiscuous-VLAN + * and default-port OR + * 2.2 The lookup is VLAN, OR + * 2.3 The lookup is MAC with mcast or bcast addr for MAC, OR + * 2.4 The lookup is MAC_VLAN with mcast or bcast addr for MAC. + * + * OR + * + * The switch is a VEPA. * - * In all other cases, the LAN enable has to be set to true. + * In all other cases, the LAN enable has to be set to false. */ - if (!(hw->evb_veb && - ((fi->lkup_type == ICE_SW_LKUP_MAC && - is_unicast_ether_addr(fi->l_data.mac.mac_addr)) || - (fi->lkup_type == ICE_SW_LKUP_MAC_VLAN && - is_unicast_ether_addr(fi->l_data.mac_vlan.mac_addr))))) + if (hw->evb_veb) { + if (fi->lkup_type == ICE_SW_LKUP_ETHERTYPE || + fi->lkup_type == ICE_SW_LKUP_PROMISC || + fi->lkup_type == ICE_SW_LKUP_ETHERTYPE_MAC || + fi->lkup_type == ICE_SW_LKUP_PROMISC_VLAN || + fi->lkup_type == ICE_SW_LKUP_DFLT || + fi->lkup_type == ICE_SW_LKUP_VLAN || + (fi->lkup_type == ICE_SW_LKUP_MAC && + !is_unicast_ether_addr(fi->l_data.mac.mac_addr)) || + (fi->lkup_type == ICE_SW_LKUP_MAC_VLAN && + !is_unicast_ether_addr(fi->l_data.mac.mac_addr))) + fi->lan_en = true; + } else { fi->lan_en = true; + } } } @@ -755,7 +799,7 @@ ice_fill_sw_rule(struct ice_hw *hw, struct ice_fltr_info *f_info, daddr = f_info->l_data.ethertype_mac.mac_addr; /* fall-through */ case ICE_SW_LKUP_ETHERTYPE: - off = (__be16 *)(eth_hdr + ICE_ETH_ETHTYPE_OFFSET); + off = (__force __be16 *)(eth_hdr + ICE_ETH_ETHTYPE_OFFSET); *off = cpu_to_be16(f_info->l_data.ethertype_mac.ethertype); break; case ICE_SW_LKUP_MAC_VLAN: @@ -785,7 +829,7 @@ ice_fill_sw_rule(struct ice_hw *hw, struct ice_fltr_info *f_info, ether_addr_copy(eth_hdr + ICE_ETH_DA_OFFSET, daddr); if (!(vlan_id > ICE_MAX_VLAN_ID)) { - off = (__be16 *)(eth_hdr + ICE_ETH_VLAN_TCI_OFFSET); + off = (__force __be16 *)(eth_hdr + ICE_ETH_VLAN_TCI_OFFSET); *off = cpu_to_be16(vlan_id); } @@ -799,7 +843,7 @@ ice_fill_sw_rule(struct ice_hw *hw, struct ice_fltr_info *f_info, * @hw: pointer to the hardware structure * @m_ent: the management entry for which sw marker needs to be added * @sw_marker: sw marker to tag the Rx descriptor with - * @l_id: large action resource id + * @l_id: large action resource ID * * Create a large action to hold software marker and update the switch rule * entry pointed by m_ent with newly created large action @@ -811,8 +855,8 @@ ice_add_marker_act(struct ice_hw *hw, struct ice_fltr_mgmt_list_entry *m_ent, struct ice_aqc_sw_rules_elem *lg_act, *rx_tx; /* For software marker we need 3 large actions * 1. FWD action: FWD TO VSI or VSI LIST - * 2. GENERIC VALUE action to hold the profile id - * 3. GENERIC VALUE action to hold the software marker id + * 2. GENERIC VALUE action to hold the profile ID + * 3. GENERIC VALUE action to hold the software marker ID */ const u16 num_lg_acts = 3; enum ice_status status; @@ -875,13 +919,13 @@ ice_add_marker_act(struct ice_hw *hw, struct ice_fltr_mgmt_list_entry *m_ent, ice_fill_sw_rule(hw, &m_ent->fltr_info, rx_tx, ice_aqc_opc_update_sw_rules); - /* Update the action to point to the large action id */ + /* Update the action to point to the large action ID */ rx_tx->pdata.lkup_tx_rx.act = cpu_to_le32(ICE_SINGLE_ACT_PTR | ((l_id << ICE_SINGLE_ACT_PTR_VAL_S) & ICE_SINGLE_ACT_PTR_VAL_M)); - /* Use the filter rule id of the previously created rule with single + /* Use the filter rule ID of the previously created rule with single * act. Once the update happens, hardware will treat this as large * action */ @@ -904,10 +948,10 @@ ice_add_marker_act(struct ice_hw *hw, struct ice_fltr_mgmt_list_entry *m_ent, * @hw: pointer to the hardware structure * @vsi_handle_arr: array of VSI handles to set in the VSI mapping * @num_vsi: number of VSI handles in the array - * @vsi_list_id: VSI list id generated as part of allocate resource + * @vsi_list_id: VSI list ID generated as part of allocate resource * - * Helper function to create a new entry of VSI list id to VSI mapping - * using the given VSI list id + * Helper function to create a new entry of VSI list ID to VSI mapping + * using the given VSI list ID */ static struct ice_vsi_list_map_info * ice_create_vsi_list_map(struct ice_hw *hw, u16 *vsi_handle_arr, u16 num_vsi, @@ -935,13 +979,13 @@ ice_create_vsi_list_map(struct ice_hw *hw, u16 *vsi_handle_arr, u16 num_vsi, * @hw: pointer to the hardware structure * @vsi_handle_arr: array of VSI handles to form a VSI list * @num_vsi: number of VSI handles in the array - * @vsi_list_id: VSI list id generated as part of allocate resource + * @vsi_list_id: VSI list ID generated as part of allocate resource * @remove: Boolean value to indicate if this is a remove action * @opc: switch rules population command type - pass in the command opcode * @lkup_type: lookup type of the filter * * Call AQ command to add a new switch rule or update existing switch rule - * using the given VSI list id + * using the given VSI list ID */ static enum ice_status ice_update_vsi_list_rule(struct ice_hw *hw, u16 *vsi_handle_arr, u16 num_vsi, @@ -998,7 +1042,7 @@ exit: /** * ice_create_vsi_list_rule - Creates and populates a VSI list rule - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * @vsi_handle_arr: array of VSI handles to form a VSI list * @num_vsi: number of VSI handles in the array * @vsi_list_id: stores the ID of the VSI list to be created @@ -1092,7 +1136,7 @@ ice_create_pkt_fwd_rule_exit: * @f_info: filter information for switch rule * * Call AQ command to update a previously created switch rule with a - * VSI list id + * VSI list ID */ static enum ice_status ice_update_pkt_fwd_rule(struct ice_hw *hw, struct ice_fltr_info *f_info) @@ -1119,7 +1163,7 @@ ice_update_pkt_fwd_rule(struct ice_hw *hw, struct ice_fltr_info *f_info) /** * ice_update_sw_rule_bridge_mode - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Updates unicast switch filter rules based on VEB/VEPA mode */ @@ -1174,7 +1218,7 @@ enum ice_status ice_update_sw_rule_bridge_mode(struct ice_hw *hw) * Allocate a new VSI list and add two VSIs * to this list using switch rule command * Update the previously created switch rule with the - * newly created VSI list id + * newly created VSI list ID * if a VSI list was previously created * Add the new VSI to the previously created VSI list set * using the update switch rule command @@ -1255,7 +1299,7 @@ ice_add_update_vsi_list(struct ice_hw *hw, return 0; /* Update the previously created VSI list set with - * the new VSI id passed in + * the new VSI ID passed in */ vsi_list_id = cur_fltr->fwd_id.vsi_list_id; opcode = ice_aqc_opc_update_sw_rules; @@ -1263,7 +1307,7 @@ ice_add_update_vsi_list(struct ice_hw *hw, status = ice_update_vsi_list_rule(hw, &vsi_handle, 1, vsi_list_id, false, opcode, new_fltr->lkup_type); - /* update VSI list mapping info with new VSI id */ + /* update VSI list mapping info with new VSI ID */ if (!status) set_bit(vsi_handle, m_entry->vsi_list_info->vsi_map); } @@ -1305,7 +1349,7 @@ ice_find_rule_entry(struct ice_hw *hw, u8 recp_id, struct ice_fltr_info *f_info) * @hw: pointer to the hardware structure * @recp_id: lookup type for which VSI lists needs to be searched * @vsi_handle: VSI handle to be found in VSI list - * @vsi_list_id: VSI list id found containing vsi_handle + * @vsi_list_id: VSI list ID found containing vsi_handle * * Helper function to search a VSI list with single entry containing given VSI * handle element. This can be extended further to search VSI list with more @@ -1336,7 +1380,7 @@ ice_find_vsi_list_entry(struct ice_hw *hw, u8 recp_id, u16 vsi_handle, /** * ice_add_rule_internal - add rule for a given lookup type * @hw: pointer to the hardware structure - * @recp_id: lookup type (recipe id) for which rule has to be added + * @recp_id: lookup type (recipe ID) for which rule has to be added * @f_entry: structure containing MAC forwarding information * * Adds or updates the rule lists for a given recipe @@ -1381,7 +1425,7 @@ ice_add_rule_internal(struct ice_hw *hw, u8 recp_id, /** * ice_remove_vsi_list_rule * @hw: pointer to the hardware structure - * @vsi_list_id: VSI list id generated as part of allocate resource + * @vsi_list_id: VSI list ID generated as part of allocate resource * @lkup_type: switch rule filter lookup type * * The VSI list should be emptied before this function is called to remove the @@ -1506,7 +1550,7 @@ ice_rem_update_vsi_list(struct ice_hw *hw, u16 vsi_handle, /** * ice_remove_rule_internal - Remove a filter rule of a given type * @hw: pointer to the hardware structure - * @recp_id: recipe id for which the rule needs to removed + * @recp_id: recipe ID for which the rule needs to removed * @f_entry: rule entry containing filter information */ static enum ice_status @@ -1556,7 +1600,7 @@ ice_remove_rule_internal(struct ice_hw *hw, u8 recp_id, status = ice_rem_update_vsi_list(hw, vsi_handle, list_elem); if (status) goto exit; - /* if vsi count goes to zero after updating the vsi list */ + /* if VSI count goes to zero after updating the VSI list */ if (list_elem->vsi_count == 0) remove_rule = true; } @@ -1634,7 +1678,7 @@ ice_add_mac(struct ice_hw *hw, struct list_head *m_list) return ICE_ERR_PARAM; hw_vsi_id = ice_get_hw_vsi_num(hw, vsi_handle); m_list_itr->fltr_info.fwd_id.hw_vsi_id = hw_vsi_id; - /* update the src in case it is vsi num */ + /* update the src in case it is VSI num */ if (m_list_itr->fltr_info.src_id != ICE_SRC_ID_VSI) return ICE_ERR_PARAM; m_list_itr->fltr_info.src = hw_vsi_id; @@ -1710,7 +1754,7 @@ ice_add_mac(struct ice_hw *hw, struct list_head *m_list) ((u8 *)r_iter + (elem_sent * s_rule_size)); } - /* Fill up rule id based on the value returned from FW */ + /* Fill up rule ID based on the value returned from FW */ r_iter = s_rule; list_for_each_entry(m_list_itr, m_list, list_entry) { struct ice_fltr_info *f_info = &m_list_itr->fltr_info; @@ -1770,7 +1814,7 @@ ice_add_vlan_internal(struct ice_hw *hw, struct ice_fltr_list_entry *f_entry) ice_get_hw_vsi_num(hw, f_entry->fltr_info.vsi_handle); new_fltr = &f_entry->fltr_info; - /* VLAN id should only be 12 bits */ + /* VLAN ID should only be 12 bits */ if (new_fltr->l_data.vlan.vlan_id > ICE_MAX_VLAN_ID) return ICE_ERR_PARAM; @@ -1828,7 +1872,7 @@ ice_add_vlan_internal(struct ice_hw *hw, struct ice_fltr_list_entry *f_entry) } } } else if (v_list_itr->vsi_list_info->ref_cnt == 1) { - /* Update existing VSI list to add new VSI id only if it used + /* Update existing VSI list to add new VSI ID only if it used * by one VLAN rule. */ cur_fltr = &v_list_itr->fltr_info; @@ -1838,7 +1882,7 @@ ice_add_vlan_internal(struct ice_hw *hw, struct ice_fltr_list_entry *f_entry) /* If VLAN rule exists and VSI list being used by this rule is * referenced by more than 1 VLAN rule. Then create a new VSI * list appending previous VSI with new VSI and update existing - * VLAN rule to point to new VSI list id + * VLAN rule to point to new VSI list ID */ struct ice_fltr_info tmp_fltr; u16 vsi_handle_arr[2]; @@ -1926,6 +1970,68 @@ ice_add_vlan(struct ice_hw *hw, struct list_head *v_list) } /** + * ice_add_eth_mac - Add ethertype and MAC based filter rule + * @hw: pointer to the hardware structure + * @em_list: list of ether type MAC filter, MAC is optional + * + * This function requires the caller to populate the entries in + * the filter list with the necessary fields (including flags to + * indicate Tx or Rx rules). + */ +enum ice_status +ice_add_eth_mac(struct ice_hw *hw, struct list_head *em_list) +{ + struct ice_fltr_list_entry *em_list_itr; + + if (!em_list || !hw) + return ICE_ERR_PARAM; + + list_for_each_entry(em_list_itr, em_list, list_entry) { + enum ice_sw_lkup_type l_type = + em_list_itr->fltr_info.lkup_type; + + if (l_type != ICE_SW_LKUP_ETHERTYPE_MAC && + l_type != ICE_SW_LKUP_ETHERTYPE) + return ICE_ERR_PARAM; + + em_list_itr->status = ice_add_rule_internal(hw, l_type, + em_list_itr); + if (em_list_itr->status) + return em_list_itr->status; + } + return 0; +} + +/** + * ice_remove_eth_mac - Remove an ethertype (or MAC) based filter rule + * @hw: pointer to the hardware structure + * @em_list: list of ethertype or ethertype MAC entries + */ +enum ice_status +ice_remove_eth_mac(struct ice_hw *hw, struct list_head *em_list) +{ + struct ice_fltr_list_entry *em_list_itr, *tmp; + + if (!em_list || !hw) + return ICE_ERR_PARAM; + + list_for_each_entry_safe(em_list_itr, tmp, em_list, list_entry) { + enum ice_sw_lkup_type l_type = + em_list_itr->fltr_info.lkup_type; + + if (l_type != ICE_SW_LKUP_ETHERTYPE_MAC && + l_type != ICE_SW_LKUP_ETHERTYPE) + return ICE_ERR_PARAM; + + em_list_itr->status = ice_remove_rule_internal(hw, l_type, + em_list_itr); + if (em_list_itr->status) + return em_list_itr->status; + } + return 0; +} + +/** * ice_rem_sw_rule_info * @hw: pointer to the hardware structure * @rule_head: pointer to the switch list structure that we want to delete @@ -2170,7 +2276,7 @@ ice_add_to_vsi_fltr_list(struct ice_hw *hw, u16 vsi_handle, struct ice_fltr_mgmt_list_entry *fm_entry; enum ice_status status = 0; - /* check to make sure VSI id is valid and within boundary */ + /* check to make sure VSI ID is valid and within boundary */ if (!ice_is_vsi_valid(hw, vsi_handle)) return ICE_ERR_PARAM; @@ -2190,6 +2296,291 @@ ice_add_to_vsi_fltr_list(struct ice_hw *hw, u16 vsi_handle, } /** + * ice_determine_promisc_mask + * @fi: filter info to parse + * + * Helper function to determine which ICE_PROMISC_ mask corresponds + * to given filter into. + */ +static u8 ice_determine_promisc_mask(struct ice_fltr_info *fi) +{ + u16 vid = fi->l_data.mac_vlan.vlan_id; + u8 *macaddr = fi->l_data.mac.mac_addr; + bool is_tx_fltr = false; + u8 promisc_mask = 0; + + if (fi->flag == ICE_FLTR_TX) + is_tx_fltr = true; + + if (is_broadcast_ether_addr(macaddr)) + promisc_mask |= is_tx_fltr ? + ICE_PROMISC_BCAST_TX : ICE_PROMISC_BCAST_RX; + else if (is_multicast_ether_addr(macaddr)) + promisc_mask |= is_tx_fltr ? + ICE_PROMISC_MCAST_TX : ICE_PROMISC_MCAST_RX; + else if (is_unicast_ether_addr(macaddr)) + promisc_mask |= is_tx_fltr ? + ICE_PROMISC_UCAST_TX : ICE_PROMISC_UCAST_RX; + if (vid) + promisc_mask |= is_tx_fltr ? + ICE_PROMISC_VLAN_TX : ICE_PROMISC_VLAN_RX; + + return promisc_mask; +} + +/** + * ice_remove_promisc - Remove promisc based filter rules + * @hw: pointer to the hardware structure + * @recp_id: recipe ID for which the rule needs to removed + * @v_list: list of promisc entries + */ +static enum ice_status +ice_remove_promisc(struct ice_hw *hw, u8 recp_id, + struct list_head *v_list) +{ + struct ice_fltr_list_entry *v_list_itr, *tmp; + + list_for_each_entry_safe(v_list_itr, tmp, v_list, list_entry) { + v_list_itr->status = + ice_remove_rule_internal(hw, recp_id, v_list_itr); + if (v_list_itr->status) + return v_list_itr->status; + } + return 0; +} + +/** + * ice_clear_vsi_promisc - clear specified promiscuous mode(s) for given VSI + * @hw: pointer to the hardware structure + * @vsi_handle: VSI handle to clear mode + * @promisc_mask: mask of promiscuous config bits to clear + * @vid: VLAN ID to clear VLAN promiscuous + */ +enum ice_status +ice_clear_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, + u16 vid) +{ + struct ice_switch_info *sw = hw->switch_info; + struct ice_fltr_list_entry *fm_entry, *tmp; + struct list_head remove_list_head; + struct ice_fltr_mgmt_list_entry *itr; + struct list_head *rule_head; + struct mutex *rule_lock; /* Lock to protect filter rule list */ + enum ice_status status = 0; + u8 recipe_id; + + if (!ice_is_vsi_valid(hw, vsi_handle)) + return ICE_ERR_PARAM; + + if (vid) + recipe_id = ICE_SW_LKUP_PROMISC_VLAN; + else + recipe_id = ICE_SW_LKUP_PROMISC; + + rule_head = &sw->recp_list[recipe_id].filt_rules; + rule_lock = &sw->recp_list[recipe_id].filt_rule_lock; + + INIT_LIST_HEAD(&remove_list_head); + + mutex_lock(rule_lock); + list_for_each_entry(itr, rule_head, list_entry) { + u8 fltr_promisc_mask = 0; + + if (!ice_vsi_uses_fltr(itr, vsi_handle)) + continue; + + fltr_promisc_mask |= + ice_determine_promisc_mask(&itr->fltr_info); + + /* Skip if filter is not completely specified by given mask */ + if (fltr_promisc_mask & ~promisc_mask) + continue; + + status = ice_add_entry_to_vsi_fltr_list(hw, vsi_handle, + &remove_list_head, + &itr->fltr_info); + if (status) { + mutex_unlock(rule_lock); + goto free_fltr_list; + } + } + mutex_unlock(rule_lock); + + status = ice_remove_promisc(hw, recipe_id, &remove_list_head); + +free_fltr_list: + list_for_each_entry_safe(fm_entry, tmp, &remove_list_head, list_entry) { + list_del(&fm_entry->list_entry); + devm_kfree(ice_hw_to_dev(hw), fm_entry); + } + + return status; +} + +/** + * ice_set_vsi_promisc - set given VSI to given promiscuous mode(s) + * @hw: pointer to the hardware structure + * @vsi_handle: VSI handle to configure + * @promisc_mask: mask of promiscuous config bits + * @vid: VLAN ID to set VLAN promiscuous + */ +enum ice_status +ice_set_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, u16 vid) +{ + enum { UCAST_FLTR = 1, MCAST_FLTR, BCAST_FLTR }; + struct ice_fltr_list_entry f_list_entry; + struct ice_fltr_info new_fltr; + enum ice_status status = 0; + bool is_tx_fltr; + u16 hw_vsi_id; + int pkt_type; + u8 recipe_id; + + if (!ice_is_vsi_valid(hw, vsi_handle)) + return ICE_ERR_PARAM; + hw_vsi_id = ice_get_hw_vsi_num(hw, vsi_handle); + + memset(&new_fltr, 0, sizeof(new_fltr)); + + if (promisc_mask & (ICE_PROMISC_VLAN_RX | ICE_PROMISC_VLAN_TX)) { + new_fltr.lkup_type = ICE_SW_LKUP_PROMISC_VLAN; + new_fltr.l_data.mac_vlan.vlan_id = vid; + recipe_id = ICE_SW_LKUP_PROMISC_VLAN; + } else { + new_fltr.lkup_type = ICE_SW_LKUP_PROMISC; + recipe_id = ICE_SW_LKUP_PROMISC; + } + + /* Separate filters must be set for each direction/packet type + * combination, so we will loop over the mask value, store the + * individual type, and clear it out in the input mask as it + * is found. + */ + while (promisc_mask) { + u8 *mac_addr; + + pkt_type = 0; + is_tx_fltr = false; + + if (promisc_mask & ICE_PROMISC_UCAST_RX) { + promisc_mask &= ~ICE_PROMISC_UCAST_RX; + pkt_type = UCAST_FLTR; + } else if (promisc_mask & ICE_PROMISC_UCAST_TX) { + promisc_mask &= ~ICE_PROMISC_UCAST_TX; + pkt_type = UCAST_FLTR; + is_tx_fltr = true; + } else if (promisc_mask & ICE_PROMISC_MCAST_RX) { + promisc_mask &= ~ICE_PROMISC_MCAST_RX; + pkt_type = MCAST_FLTR; + } else if (promisc_mask & ICE_PROMISC_MCAST_TX) { + promisc_mask &= ~ICE_PROMISC_MCAST_TX; + pkt_type = MCAST_FLTR; + is_tx_fltr = true; + } else if (promisc_mask & ICE_PROMISC_BCAST_RX) { + promisc_mask &= ~ICE_PROMISC_BCAST_RX; + pkt_type = BCAST_FLTR; + } else if (promisc_mask & ICE_PROMISC_BCAST_TX) { + promisc_mask &= ~ICE_PROMISC_BCAST_TX; + pkt_type = BCAST_FLTR; + is_tx_fltr = true; + } + + /* Check for VLAN promiscuous flag */ + if (promisc_mask & ICE_PROMISC_VLAN_RX) { + promisc_mask &= ~ICE_PROMISC_VLAN_RX; + } else if (promisc_mask & ICE_PROMISC_VLAN_TX) { + promisc_mask &= ~ICE_PROMISC_VLAN_TX; + is_tx_fltr = true; + } + + /* Set filter DA based on packet type */ + mac_addr = new_fltr.l_data.mac.mac_addr; + if (pkt_type == BCAST_FLTR) { + eth_broadcast_addr(mac_addr); + } else if (pkt_type == MCAST_FLTR || + pkt_type == UCAST_FLTR) { + /* Use the dummy ether header DA */ + ether_addr_copy(mac_addr, dummy_eth_header); + if (pkt_type == MCAST_FLTR) + mac_addr[0] |= 0x1; /* Set multicast bit */ + } + + /* Need to reset this to zero for all iterations */ + new_fltr.flag = 0; + if (is_tx_fltr) { + new_fltr.flag |= ICE_FLTR_TX; + new_fltr.src = hw_vsi_id; + } else { + new_fltr.flag |= ICE_FLTR_RX; + new_fltr.src = hw->port_info->lport; + } + + new_fltr.fltr_act = ICE_FWD_TO_VSI; + new_fltr.vsi_handle = vsi_handle; + new_fltr.fwd_id.hw_vsi_id = hw_vsi_id; + f_list_entry.fltr_info = new_fltr; + + status = ice_add_rule_internal(hw, recipe_id, &f_list_entry); + if (status) + goto set_promisc_exit; + } + +set_promisc_exit: + return status; +} + +/** + * ice_set_vlan_vsi_promisc + * @hw: pointer to the hardware structure + * @vsi_handle: VSI handle to configure + * @promisc_mask: mask of promiscuous config bits + * @rm_vlan_promisc: Clear VLANs VSI promisc mode + * + * Configure VSI with all associated VLANs to given promiscuous mode(s) + */ +enum ice_status +ice_set_vlan_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, + bool rm_vlan_promisc) +{ + struct ice_switch_info *sw = hw->switch_info; + struct ice_fltr_list_entry *list_itr, *tmp; + struct list_head vsi_list_head; + struct list_head *vlan_head; + struct mutex *vlan_lock; /* Lock to protect filter rule list */ + enum ice_status status; + u16 vlan_id; + + INIT_LIST_HEAD(&vsi_list_head); + vlan_lock = &sw->recp_list[ICE_SW_LKUP_VLAN].filt_rule_lock; + vlan_head = &sw->recp_list[ICE_SW_LKUP_VLAN].filt_rules; + mutex_lock(vlan_lock); + status = ice_add_to_vsi_fltr_list(hw, vsi_handle, vlan_head, + &vsi_list_head); + mutex_unlock(vlan_lock); + if (status) + goto free_fltr_list; + + list_for_each_entry(list_itr, &vsi_list_head, list_entry) { + vlan_id = list_itr->fltr_info.l_data.vlan.vlan_id; + if (rm_vlan_promisc) + status = ice_clear_vsi_promisc(hw, vsi_handle, + promisc_mask, vlan_id); + else + status = ice_set_vsi_promisc(hw, vsi_handle, + promisc_mask, vlan_id); + if (status) + break; + } + +free_fltr_list: + list_for_each_entry_safe(list_itr, tmp, &vsi_list_head, list_entry) { + list_del(&list_itr->list_entry); + devm_kfree(ice_hw_to_dev(hw), list_itr); + } + return status; +} + +/** * ice_remove_vsi_lkup_fltr - Remove lookup type filters for a VSI * @hw: pointer to the hardware structure * @vsi_handle: VSI handle to remove filters from @@ -2224,12 +2615,14 @@ ice_remove_vsi_lkup_fltr(struct ice_hw *hw, u16 vsi_handle, case ICE_SW_LKUP_VLAN: ice_remove_vlan(hw, &remove_list_head); break; + case ICE_SW_LKUP_PROMISC: + case ICE_SW_LKUP_PROMISC_VLAN: + ice_remove_promisc(hw, lkup, &remove_list_head); + break; case ICE_SW_LKUP_MAC_VLAN: case ICE_SW_LKUP_ETHERTYPE: case ICE_SW_LKUP_ETHERTYPE_MAC: - case ICE_SW_LKUP_PROMISC: case ICE_SW_LKUP_DFLT: - case ICE_SW_LKUP_PROMISC_VLAN: case ICE_SW_LKUP_LAST: default: ice_debug(hw, ICE_DBG_SW, "Unsupported lookup type %d\n", lkup); @@ -2263,7 +2656,7 @@ void ice_remove_vsi_fltr(struct ice_hw *hw, u16 vsi_handle) * ice_replay_vsi_fltr - Replay filters for requested VSI * @hw: pointer to the hardware structure * @vsi_handle: driver VSI handle - * @recp_id: Recipe id for which rules need to be replayed + * @recp_id: Recipe ID for which rules need to be replayed * @list_head: list for which filters need to be replayed * * Replays the filter of recipe recp_id for a VSI represented via vsi_handle. @@ -2287,7 +2680,7 @@ ice_replay_vsi_fltr(struct ice_hw *hw, u16 vsi_handle, u8 recp_id, f_entry.fltr_info = itr->fltr_info; if (itr->vsi_count < 2 && recp_id != ICE_SW_LKUP_VLAN && itr->fltr_info.vsi_handle == vsi_handle) { - /* update the src in case it is vsi num */ + /* update the src in case it is VSI num */ if (f_entry.fltr_info.src_id == ICE_SRC_ID_VSI) f_entry.fltr_info.src = hw_vsi_id; status = ice_add_rule_internal(hw, recp_id, &f_entry); @@ -2302,7 +2695,7 @@ ice_replay_vsi_fltr(struct ice_hw *hw, u16 vsi_handle, u8 recp_id, clear_bit(vsi_handle, itr->vsi_list_info->vsi_map); f_entry.fltr_info.vsi_handle = vsi_handle; f_entry.fltr_info.fltr_act = ICE_FWD_TO_VSI; - /* update the src in case it is vsi num */ + /* update the src in case it is VSI num */ if (f_entry.fltr_info.src_id == ICE_SRC_ID_VSI) f_entry.fltr_info.src = hw_vsi_id; if (recp_id == ICE_SW_LKUP_VLAN) @@ -2342,7 +2735,7 @@ enum ice_status ice_replay_vsi_all_fltr(struct ice_hw *hw, u16 vsi_handle) /** * ice_rm_all_sw_replay_rule_info - deletes filter replay rules - * @hw: pointer to the hw struct + * @hw: pointer to the HW struct * * Deletes the filter replay rules. */ diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h index d5ef0bd58bf9..cb123fbe30be 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.h +++ b/drivers/net/ethernet/intel/ice/ice_switch.h @@ -8,7 +8,16 @@ #define ICE_SW_CFG_MAX_BUF_LEN 2048 #define ICE_DFLT_VSI_INVAL 0xff +#define ICE_FLTR_RX BIT(0) +#define ICE_FLTR_TX BIT(1) +#define ICE_FLTR_TX_RX (ICE_FLTR_RX | ICE_FLTR_TX) #define ICE_VSI_INVAL_ID 0xffff +#define ICE_INVAL_Q_HANDLE 0xFFFF + +/* VSI queue context structure */ +struct ice_q_ctx { + u16 q_handle; +}; /* VSI context structure for add/get/update/free operations */ struct ice_vsi_ctx { @@ -20,6 +29,8 @@ struct ice_vsi_ctx { struct ice_sched_vsi_info sched; u8 alloc_from_pool; u8 vf_num; + u16 num_lan_q_entries[ICE_MAX_TRAFFIC_CLASS]; + struct ice_q_ctx *lan_q_ctx[ICE_MAX_TRAFFIC_CLASS]; }; enum ice_sw_fwd_act_type { @@ -44,7 +55,7 @@ enum ice_sw_lkup_type { ICE_SW_LKUP_LAST }; -/* type of filter src id */ +/* type of filter src ID */ enum ice_src_id { ICE_SRC_ID_UNKNOWN = 0, ICE_SRC_ID_VSI, @@ -60,9 +71,6 @@ struct ice_fltr_info { /* rule ID returned by firmware once filter rule is created */ u16 fltr_rule_id; u16 flag; -#define ICE_FLTR_RX BIT(0) -#define ICE_FLTR_TX BIT(1) -#define ICE_FLTR_TX_RX (ICE_FLTR_RX | ICE_FLTR_TX) /* Source VSI for LOOKUP_TX or source port for LOOKUP_RX */ u16 src; @@ -95,8 +103,8 @@ struct ice_fltr_info { /* Depending on filter action */ union { - /* queue id in case of ICE_FWD_TO_Q and starting - * queue id in case of ICE_FWD_TO_QGRP. + /* queue ID in case of ICE_FWD_TO_Q and starting + * queue ID in case of ICE_FWD_TO_QGRP. */ u16 q_id:11; u16 hw_vsi_id:10; @@ -143,7 +151,7 @@ struct ice_sw_recipe { DECLARE_BITMAP(r_bitmap, ICE_MAX_NUM_RECIPES); }; -/* Bookkeeping structure to hold bitmap of VSIs corresponding to VSI list id */ +/* Bookkeeping structure to hold bitmap of VSIs corresponding to VSI list ID */ struct ice_vsi_list_map_info { struct list_head list_entry; DECLARE_BITMAP(vsi_map, ICE_MAX_VSI); @@ -165,7 +173,7 @@ struct ice_fltr_list_entry { * used for VLAN membership. */ struct ice_fltr_mgmt_list_entry { - /* back pointer to VSI list id to VSI list mapping */ + /* back pointer to VSI list ID to VSI list mapping */ struct ice_vsi_list_map_info *vsi_list_info; u16 vsi_count; #define ICE_INVAL_LG_ACT_INDEX 0xffff @@ -178,6 +186,17 @@ struct ice_fltr_mgmt_list_entry { u8 counter_index; }; +enum ice_promisc_flags { + ICE_PROMISC_UCAST_RX = 0x1, + ICE_PROMISC_UCAST_TX = 0x2, + ICE_PROMISC_MCAST_RX = 0x4, + ICE_PROMISC_MCAST_TX = 0x8, + ICE_PROMISC_BCAST_RX = 0x10, + ICE_PROMISC_BCAST_TX = 0x20, + ICE_PROMISC_VLAN_RX = 0x40, + ICE_PROMISC_VLAN_TX = 0x80, +}; + /* VSI related commands */ enum ice_status ice_add_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx, @@ -198,11 +217,27 @@ enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw); enum ice_status ice_update_sw_rule_bridge_mode(struct ice_hw *hw); enum ice_status ice_add_mac(struct ice_hw *hw, struct list_head *m_lst); enum ice_status ice_remove_mac(struct ice_hw *hw, struct list_head *m_lst); +enum ice_status +ice_add_eth_mac(struct ice_hw *hw, struct list_head *em_list); +enum ice_status +ice_remove_eth_mac(struct ice_hw *hw, struct list_head *em_list); void ice_remove_vsi_fltr(struct ice_hw *hw, u16 vsi_handle); -enum ice_status ice_add_vlan(struct ice_hw *hw, struct list_head *m_list); +enum ice_status +ice_add_vlan(struct ice_hw *hw, struct list_head *m_list); enum ice_status ice_remove_vlan(struct ice_hw *hw, struct list_head *v_list); + +/* Promisc/defport setup for VSIs */ enum ice_status ice_cfg_dflt_vsi(struct ice_hw *hw, u16 vsi_handle, bool set, u8 direction); +enum ice_status +ice_set_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, + u16 vid); +enum ice_status +ice_clear_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, + u16 vid); +enum ice_status +ice_set_vlan_vsi_promisc(struct ice_hw *hw, u16 vsi_handle, u8 promisc_mask, + bool rm_vlan_promisc); enum ice_status ice_init_def_sw_recp(struct ice_hw *hw); u16 ice_get_hw_vsi_num(struct ice_hw *hw, u16 vsi_handle); diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 1af21bbe180e..3c83230434b6 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -6,6 +6,7 @@ #include <linux/prefetch.h> #include <linux/mm.h> #include "ice.h" +#include "ice_dcb_lib.h" #define ICE_RX_HDR_SIZE 256 @@ -54,7 +55,7 @@ void ice_clean_tx_ring(struct ice_ring *tx_ring) if (!tx_ring->tx_buf) return; - /* Free all the Tx ring sk_bufss */ + /* Free all the Tx ring sk_buffs */ for (i = 0; i < tx_ring->count; i++) ice_unmap_and_free_tx_buf(tx_ring, &tx_ring->tx_buf[i]); @@ -100,8 +101,8 @@ void ice_free_tx_ring(struct ice_ring *tx_ring) * * Returns true if there's any budget left (e.g. the clean is finished) */ -static bool ice_clean_tx_irq(struct ice_vsi *vsi, struct ice_ring *tx_ring, - int napi_budget) +static bool +ice_clean_tx_irq(struct ice_vsi *vsi, struct ice_ring *tx_ring, int napi_budget) { unsigned int total_bytes = 0, total_pkts = 0; unsigned int budget = vsi->work_lmt; @@ -236,9 +237,9 @@ int ice_setup_tx_ring(struct ice_ring *tx_ring) if (!tx_ring->tx_buf) return -ENOMEM; - /* round up to nearest 4K */ + /* round up to nearest page */ tx_ring->size = ALIGN(tx_ring->count * sizeof(struct ice_tx_desc), - 4096); + PAGE_SIZE); tx_ring->desc = dmam_alloc_coherent(dev, tx_ring->size, &tx_ring->dma, GFP_KERNEL); if (!tx_ring->desc) { @@ -282,8 +283,17 @@ void ice_clean_rx_ring(struct ice_ring *rx_ring) if (!rx_buf->page) continue; - dma_unmap_page(dev, rx_buf->dma, PAGE_SIZE, DMA_FROM_DEVICE); - __free_pages(rx_buf->page, 0); + /* Invalidate cache lines that may have been written to by + * device so that we avoid corrupting memory. + */ + dma_sync_single_range_for_cpu(dev, rx_buf->dma, + rx_buf->page_offset, + ICE_RXBUF_2048, DMA_FROM_DEVICE); + + /* free resources associated with mapping */ + dma_unmap_page_attrs(dev, rx_buf->dma, PAGE_SIZE, + DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); + __page_frag_cache_drain(rx_buf->page, rx_buf->pagecnt_bias); rx_buf->page = NULL; rx_buf->page_offset = 0; @@ -339,9 +349,9 @@ int ice_setup_rx_ring(struct ice_ring *rx_ring) if (!rx_ring->rx_buf) return -ENOMEM; - /* round up to nearest 4K */ - rx_ring->size = rx_ring->count * sizeof(union ice_32byte_rx_desc); - rx_ring->size = ALIGN(rx_ring->size, 4096); + /* round up to nearest page */ + rx_ring->size = ALIGN(rx_ring->count * sizeof(union ice_32byte_rx_desc), + PAGE_SIZE); rx_ring->desc = dmam_alloc_coherent(dev, rx_ring->size, &rx_ring->dma, GFP_KERNEL); if (!rx_ring->desc) { @@ -389,8 +399,8 @@ static void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val) * Returns true if the page was successfully allocated or * reused. */ -static bool ice_alloc_mapped_page(struct ice_ring *rx_ring, - struct ice_rx_buf *bi) +static bool +ice_alloc_mapped_page(struct ice_ring *rx_ring, struct ice_rx_buf *bi) { struct page *page = bi->page; dma_addr_t dma; @@ -409,7 +419,8 @@ static bool ice_alloc_mapped_page(struct ice_ring *rx_ring, } /* map page for use */ - dma = dma_map_page(rx_ring->dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE); + dma = dma_map_page_attrs(rx_ring->dev, page, 0, PAGE_SIZE, + DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); /* if mapping failed free memory back to system since * there isn't much point in holding memory we can't use @@ -423,6 +434,8 @@ static bool ice_alloc_mapped_page(struct ice_ring *rx_ring, bi->dma = dma; bi->page = page; bi->page_offset = 0; + page_ref_add(page, USHRT_MAX - 1); + bi->pagecnt_bias = USHRT_MAX; return true; } @@ -444,7 +457,7 @@ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count) if (!rx_ring->netdev || !cleaned_count) return false; - /* get the RX descriptor and buffer based on next_to_use */ + /* get the Rx descriptor and buffer based on next_to_use */ rx_desc = ICE_RX_DESC(rx_ring, ntu); bi = &rx_ring->rx_buf[ntu]; @@ -452,6 +465,12 @@ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count) if (!ice_alloc_mapped_page(rx_ring, bi)) goto no_bufs; + /* sync the buffer for use by the device */ + dma_sync_single_range_for_device(rx_ring->dev, bi->dma, + bi->page_offset, + ICE_RXBUF_2048, + DMA_FROM_DEVICE); + /* Refresh the desc even if buffer_addrs didn't change * because each write-back erases this info. */ @@ -497,61 +516,43 @@ static bool ice_page_is_reserved(struct page *page) } /** - * ice_add_rx_frag - Add contents of Rx buffer to sk_buff - * @rx_buf: buffer containing page to add - * @rx_desc: descriptor containing length of buffer written by hardware - * @skb: sk_buf to place the data into - * - * This function will add the data contained in rx_buf->page to the skb. - * This is done either through a direct copy if the data in the buffer is - * less than the skb header size, otherwise it will just attach the page as - * a frag to the skb. + * ice_rx_buf_adjust_pg_offset - Prepare Rx buffer for reuse + * @rx_buf: Rx buffer to adjust + * @size: Size of adjustment * - * The function will then update the page offset if necessary and return - * true if the buffer can be reused by the adapter. + * Update the offset within page so that Rx buf will be ready to be reused. + * For systems with PAGE_SIZE < 8192 this function will flip the page offset + * so the second half of page assigned to Rx buffer will be used, otherwise + * the offset is moved by the @size bytes */ -static bool ice_add_rx_frag(struct ice_rx_buf *rx_buf, - union ice_32b_rx_flex_desc *rx_desc, - struct sk_buff *skb) +static void +ice_rx_buf_adjust_pg_offset(struct ice_rx_buf *rx_buf, unsigned int size) { #if (PAGE_SIZE < 8192) - unsigned int truesize = ICE_RXBUF_2048; + /* flip page offset to other buffer */ + rx_buf->page_offset ^= size; #else - unsigned int last_offset = PAGE_SIZE - ICE_RXBUF_2048; - unsigned int truesize; -#endif /* PAGE_SIZE < 8192) */ - - struct page *page; - unsigned int size; - - size = le16_to_cpu(rx_desc->wb.pkt_len) & - ICE_RX_FLX_DESC_PKT_LEN_M; - - page = rx_buf->page; + /* move offset up to the next cache line */ + rx_buf->page_offset += size; +#endif +} +/** + * ice_can_reuse_rx_page - Determine if page can be reused for another Rx + * @rx_buf: buffer containing the page + * + * If page is reusable, we have a green light for calling ice_reuse_rx_page, + * which will assign the current buffer to the buffer that next_to_alloc is + * pointing to; otherwise, the DMA mapping needs to be destroyed and + * page freed + */ +static bool ice_can_reuse_rx_page(struct ice_rx_buf *rx_buf) +{ #if (PAGE_SIZE >= 8192) - truesize = ALIGN(size, L1_CACHE_BYTES); -#endif /* PAGE_SIZE >= 8192) */ - - /* will the data fit in the skb we allocated? if so, just - * copy it as it is pretty small anyway - */ - if (size <= ICE_RX_HDR_SIZE && !skb_is_nonlinear(skb)) { - unsigned char *va = page_address(page) + rx_buf->page_offset; - - memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long))); - - /* page is not reserved, we can reuse buffer as-is */ - if (likely(!ice_page_is_reserved(page))) - return true; - - /* this page cannot be reused so discard it */ - __free_pages(page, 0); - return false; - } - - skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, - rx_buf->page_offset, size, truesize); + unsigned int last_offset = PAGE_SIZE - ICE_RXBUF_2048; +#endif + unsigned int pagecnt_bias = rx_buf->pagecnt_bias; + struct page *page = rx_buf->page; /* avoid re-using remote pages */ if (unlikely(ice_page_is_reserved(page))) @@ -559,36 +560,61 @@ static bool ice_add_rx_frag(struct ice_rx_buf *rx_buf, #if (PAGE_SIZE < 8192) /* if we are only owner of page we can reuse it */ - if (unlikely(page_count(page) != 1)) + if (unlikely((page_count(page) - pagecnt_bias) > 1)) return false; - - /* flip page offset to other buffer */ - rx_buf->page_offset ^= truesize; #else - /* move offset up to the next cache line */ - rx_buf->page_offset += truesize; - if (rx_buf->page_offset > last_offset) return false; #endif /* PAGE_SIZE < 8192) */ - /* Even if we own the page, we are not allowed to use atomic_set() - * This would break get_page_unless_zero() users. + /* If we have drained the page fragment pool we need to update + * the pagecnt_bias and page count so that we fully restock the + * number of references the driver holds. */ - get_page(rx_buf->page); + if (unlikely(pagecnt_bias == 1)) { + page_ref_add(page, USHRT_MAX - 1); + rx_buf->pagecnt_bias = USHRT_MAX; + } return true; } /** + * ice_add_rx_frag - Add contents of Rx buffer to sk_buff as a frag + * @rx_buf: buffer containing page to add + * @skb: sk_buff to place the data into + * @size: packet length from rx_desc + * + * This function will add the data contained in rx_buf->page to the skb. + * It will just attach the page as a frag to the skb. + * The function will then update the page offset. + */ +static void +ice_add_rx_frag(struct ice_rx_buf *rx_buf, struct sk_buff *skb, + unsigned int size) +{ +#if (PAGE_SIZE >= 8192) + unsigned int truesize = SKB_DATA_ALIGN(size); +#else + unsigned int truesize = ICE_RXBUF_2048; +#endif + + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page, + rx_buf->page_offset, size, truesize); + + /* page is being used so we must update the page offset */ + ice_rx_buf_adjust_pg_offset(rx_buf, truesize); +} + +/** * ice_reuse_rx_page - page flip buffer and store it back on the ring * @rx_ring: Rx descriptor ring to store buffers on * @old_buf: donor buffer to have page reused * * Synchronizes page for reuse by the adapter */ -static void ice_reuse_rx_page(struct ice_ring *rx_ring, - struct ice_rx_buf *old_buf) +static void +ice_reuse_rx_page(struct ice_ring *rx_ring, struct ice_rx_buf *old_buf) { u16 nta = rx_ring->next_to_alloc; struct ice_rx_buf *new_buf; @@ -599,121 +625,132 @@ static void ice_reuse_rx_page(struct ice_ring *rx_ring, nta++; rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - /* transfer page from old buffer to new buffer */ - *new_buf = *old_buf; + /* Transfer page from old buffer to new buffer. + * Move each member individually to avoid possible store + * forwarding stalls and unnecessary copy of skb. + */ + new_buf->dma = old_buf->dma; + new_buf->page = old_buf->page; + new_buf->page_offset = old_buf->page_offset; + new_buf->pagecnt_bias = old_buf->pagecnt_bias; } /** - * ice_fetch_rx_buf - Allocate skb and populate it + * ice_get_rx_buf - Fetch Rx buffer and synchronize data for use * @rx_ring: Rx descriptor ring to transact packets on - * @rx_desc: descriptor containing info written by hardware + * @skb: skb to be used + * @size: size of buffer to add to skb * - * This function allocates an skb on the fly, and populates it with the page - * data from the current receive descriptor, taking care to set up the skb - * correctly, as well as handling calling the page recycle function if - * necessary. + * This function will pull an Rx buffer from the ring and synchronize it + * for use by the CPU. */ -static struct sk_buff *ice_fetch_rx_buf(struct ice_ring *rx_ring, - union ice_32b_rx_flex_desc *rx_desc) +static struct ice_rx_buf * +ice_get_rx_buf(struct ice_ring *rx_ring, struct sk_buff **skb, + const unsigned int size) { struct ice_rx_buf *rx_buf; - struct sk_buff *skb; - struct page *page; rx_buf = &rx_ring->rx_buf[rx_ring->next_to_clean]; - page = rx_buf->page; - prefetchw(page); + prefetchw(rx_buf->page); + *skb = rx_buf->skb; - skb = rx_buf->skb; + /* we are reusing so sync this buffer for CPU use */ + dma_sync_single_range_for_cpu(rx_ring->dev, rx_buf->dma, + rx_buf->page_offset, size, + DMA_FROM_DEVICE); - if (likely(!skb)) { - u8 *page_addr = page_address(page) + rx_buf->page_offset; + /* We have pulled a buffer for use, so decrement pagecnt_bias */ + rx_buf->pagecnt_bias--; - /* prefetch first cache line of first page */ - prefetch(page_addr); + return rx_buf; +} + +/** + * ice_construct_skb - Allocate skb and populate it + * @rx_ring: Rx descriptor ring to transact packets on + * @rx_buf: Rx buffer to pull data from + * @size: the length of the packet + * + * This function allocates an skb. It then populates it with the page + * data from the current receive descriptor, taking care to set up the + * skb correctly. + */ +static struct sk_buff * +ice_construct_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, + unsigned int size) +{ + void *va = page_address(rx_buf->page) + rx_buf->page_offset; + unsigned int headlen; + struct sk_buff *skb; + + /* prefetch first cache line of first page */ + prefetch(va); #if L1_CACHE_BYTES < 128 - prefetch((void *)(page_addr + L1_CACHE_BYTES)); + prefetch((u8 *)va + L1_CACHE_BYTES); #endif /* L1_CACHE_BYTES */ - /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, - ICE_RX_HDR_SIZE, - GFP_ATOMIC | __GFP_NOWARN); - if (unlikely(!skb)) { - rx_ring->rx_stats.alloc_buf_failed++; - return NULL; - } - - /* we will be copying header into skb->data in - * pskb_may_pull so it is in our interest to prefetch - * it now to avoid a possible cache miss - */ - prefetchw(skb->data); + /* allocate a skb to store the frags */ + skb = __napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE, + GFP_ATOMIC | __GFP_NOWARN); + if (unlikely(!skb)) + return NULL; - skb_record_rx_queue(skb, rx_ring->q_index); - } else { - /* we are reusing so sync this buffer for CPU use */ - dma_sync_single_range_for_cpu(rx_ring->dev, rx_buf->dma, - rx_buf->page_offset, - ICE_RXBUF_2048, - DMA_FROM_DEVICE); + skb_record_rx_queue(skb, rx_ring->q_index); + /* Determine available headroom for copy */ + headlen = size; + if (headlen > ICE_RX_HDR_SIZE) + headlen = eth_get_headlen(skb->dev, va, ICE_RX_HDR_SIZE); - rx_buf->skb = NULL; - } + /* align pull length to size of long to optimize memcpy performance */ + memcpy(__skb_put(skb, headlen), va, ALIGN(headlen, sizeof(long))); - /* pull page into skb */ - if (ice_add_rx_frag(rx_buf, rx_desc, skb)) { - /* hand second half of page back to the ring */ - ice_reuse_rx_page(rx_ring, rx_buf); - rx_ring->rx_stats.page_reuse_count++; + /* if we exhaust the linear part then add what is left as a frag */ + size -= headlen; + if (size) { +#if (PAGE_SIZE >= 8192) + unsigned int truesize = SKB_DATA_ALIGN(size); +#else + unsigned int truesize = ICE_RXBUF_2048; +#endif + skb_add_rx_frag(skb, 0, rx_buf->page, + rx_buf->page_offset + headlen, size, truesize); + /* buffer is used by skb, update page_offset */ + ice_rx_buf_adjust_pg_offset(rx_buf, truesize); } else { - /* we are not reusing the buffer so unmap it */ - dma_unmap_page(rx_ring->dev, rx_buf->dma, PAGE_SIZE, - DMA_FROM_DEVICE); + /* buffer is unused, reset bias back to rx_buf; data was copied + * onto skb's linear part so there's no need for adjusting + * page offset and we can reuse this buffer as-is + */ + rx_buf->pagecnt_bias++; } - /* clear contents of buffer_info */ - rx_buf->page = NULL; - return skb; } /** - * ice_pull_tail - ice specific version of skb_pull_tail - * @skb: pointer to current skb being adjusted + * ice_put_rx_buf - Clean up used buffer and either recycle or free + * @rx_ring: Rx descriptor ring to transact packets on + * @rx_buf: Rx buffer to pull data from * - * This function is an ice specific version of __pskb_pull_tail. The - * main difference between this version and the original function is that - * this function can make several assumptions about the state of things - * that allow for significant optimizations versus the standard function. - * As a result we can do things like drop a frag and maintain an accurate - * truesize for the skb. + * This function will clean up the contents of the rx_buf. It will + * either recycle the buffer or unmap it and free the associated resources. */ -static void ice_pull_tail(struct sk_buff *skb) +static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) { - struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0]; - unsigned int pull_len; - unsigned char *va; - - /* it is valid to use page_address instead of kmap since we are - * working with pages allocated out of the lomem pool per - * alloc_page(GFP_ATOMIC) - */ - va = skb_frag_address(frag); - - /* we need the header to contain the greater of either ETH_HLEN or - * 60 bytes if the skb->len is less than 60 for skb_pad. - */ - pull_len = eth_get_headlen(va, ICE_RX_HDR_SIZE); - - /* align pull length to size of long to optimize memcpy performance */ - skb_copy_to_linear_data(skb, va, ALIGN(pull_len, sizeof(long))); + /* hand second half of page back to the ring */ + if (ice_can_reuse_rx_page(rx_buf)) { + ice_reuse_rx_page(rx_ring, rx_buf); + rx_ring->rx_stats.page_reuse_count++; + } else { + /* we are not reusing the buffer so unmap it */ + dma_unmap_page_attrs(rx_ring->dev, rx_buf->dma, PAGE_SIZE, + DMA_FROM_DEVICE, ICE_RX_DMA_ATTR); + __page_frag_cache_drain(rx_buf->page, rx_buf->pagecnt_bias); + } - /* update all of the pointers */ - skb_frag_size_sub(frag, pull_len); - frag->page_offset += pull_len; - skb->data_len -= pull_len; - skb->tail += pull_len; + /* clear contents of buffer_info */ + rx_buf->page = NULL; + rx_buf->skb = NULL; } /** @@ -730,10 +767,6 @@ static void ice_pull_tail(struct sk_buff *skb) */ static bool ice_cleanup_headers(struct sk_buff *skb) { - /* place header in linear portion of buffer */ - if (skb_is_nonlinear(skb)) - ice_pull_tail(skb); - /* if eth_skb_pad returns an error the skb was freed */ if (eth_skb_pad(skb)) return true; @@ -751,8 +784,8 @@ static bool ice_cleanup_headers(struct sk_buff *skb) * The status_error_len doesn't need to be shifted because it begins * at offset zero. */ -static bool ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc, - const u16 stat_err_bits) +static bool +ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc, const u16 stat_err_bits) { return !!(rx_desc->wb.status_error0 & cpu_to_le16(stat_err_bits)); @@ -769,9 +802,9 @@ static bool ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc, * sk_buff in the next buffer to be chained and return true indicating * that this is in fact a non-EOP buffer. */ -static bool ice_is_non_eop(struct ice_ring *rx_ring, - union ice_32b_rx_flex_desc *rx_desc, - struct sk_buff *skb) +static bool +ice_is_non_eop(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, + struct sk_buff *skb) { u32 ntc = rx_ring->next_to_clean + 1; @@ -838,8 +871,9 @@ ice_rx_hash(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, * * skb->protocol must be set before this function is called */ -static void ice_rx_csum(struct ice_vsi *vsi, struct sk_buff *skb, - union ice_32b_rx_flex_desc *rx_desc, u8 ptype) +static void +ice_rx_csum(struct ice_vsi *vsi, struct sk_buff *skb, + union ice_32b_rx_flex_desc *rx_desc, u8 ptype) { struct ice_rx_ptype_decoded decoded; u32 rx_error, rx_status; @@ -909,9 +943,10 @@ checksum_fail: * order to populate the hash, checksum, VLAN, protocol, and * other fields within the skb. */ -static void ice_process_skb_fields(struct ice_ring *rx_ring, - union ice_32b_rx_flex_desc *rx_desc, - struct sk_buff *skb, u8 ptype) +static void +ice_process_skb_fields(struct ice_ring *rx_ring, + union ice_32b_rx_flex_desc *rx_desc, + struct sk_buff *skb, u8 ptype) { ice_rx_hash(rx_ring, rx_desc, skb, ptype); @@ -925,18 +960,17 @@ static void ice_process_skb_fields(struct ice_ring *rx_ring, * ice_receive_skb - Send a completed packet up the stack * @rx_ring: Rx ring in play * @skb: packet to send up - * @vlan_tag: vlan tag for packet + * @vlan_tag: VLAN tag for packet * * This function sends the completed packet (via. skb) up the stack using - * gro receive functions (with/without vlan tag) + * gro receive functions (with/without VLAN tag) */ -static void ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb, - u16 vlan_tag) +static void +ice_receive_skb(struct ice_ring *rx_ring, struct sk_buff *skb, u16 vlan_tag) { if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) && - (vlan_tag & VLAN_VID_MASK)) { + (vlan_tag & VLAN_VID_MASK)) __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag); - } napi_gro_receive(&rx_ring->q_vector->napi, skb); } @@ -958,10 +992,12 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) u16 cleaned_count = ICE_DESC_UNUSED(rx_ring); bool failure = false; - /* start the loop to process RX packets bounded by 'budget' */ + /* start the loop to process Rx packets bounded by 'budget' */ while (likely(total_rx_pkts < (unsigned int)budget)) { union ice_32b_rx_flex_desc *rx_desc; + struct ice_rx_buf *rx_buf; struct sk_buff *skb; + unsigned int size; u16 stat_err_bits; u16 vlan_tag = 0; u8 rx_ptype; @@ -973,7 +1009,7 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) cleaned_count = 0; } - /* get the RX desc from RX ring based on 'next_to_clean' */ + /* get the Rx desc from Rx ring based on 'next_to_clean' */ rx_desc = ICE_RX_DESC(rx_ring, rx_ring->next_to_clean); /* status_error_len will always be zero for unused descriptors @@ -991,11 +1027,24 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) */ dma_rmb(); + size = le16_to_cpu(rx_desc->wb.pkt_len) & + ICE_RX_FLX_DESC_PKT_LEN_M; + + rx_buf = ice_get_rx_buf(rx_ring, &skb, size); /* allocate (if needed) and populate skb */ - skb = ice_fetch_rx_buf(rx_ring, rx_desc); - if (!skb) + if (skb) + ice_add_rx_frag(rx_buf, skb, size); + else + skb = ice_construct_skb(rx_ring, rx_buf, size); + + /* exit if we failed to retrieve a buffer */ + if (!skb) { + rx_ring->rx_stats.alloc_buf_failed++; + rx_buf->pagecnt_bias++; break; + } + ice_put_rx_buf(rx_ring, rx_buf); cleaned_count++; /* skip if it is NOP desc */ @@ -1049,17 +1098,247 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) } /** + * ice_adjust_itr_by_size_and_speed - Adjust ITR based on current traffic + * @port_info: port_info structure containing the current link speed + * @avg_pkt_size: average size of Tx or Rx packets based on clean routine + * @itr: ITR value to update + * + * Calculate how big of an increment should be applied to the ITR value passed + * in based on wmem_default, SKB overhead, Ethernet overhead, and the current + * link speed. + * + * The following is a calculation derived from: + * wmem_default / (size + overhead) = desired_pkts_per_int + * rate / bits_per_byte / (size + Ethernet overhead) = pkt_rate + * (desired_pkt_rate / pkt_rate) * usecs_per_sec = ITR value + * + * Assuming wmem_default is 212992 and overhead is 640 bytes per + * packet, (256 skb, 64 headroom, 320 shared info), we can reduce the + * formula down to: + * + * wmem_default * bits_per_byte * usecs_per_sec pkt_size + 24 + * ITR = -------------------------------------------- * -------------- + * rate pkt_size + 640 + */ +static unsigned int +ice_adjust_itr_by_size_and_speed(struct ice_port_info *port_info, + unsigned int avg_pkt_size, + unsigned int itr) +{ + switch (port_info->phy.link_info.link_speed) { + case ICE_AQ_LINK_SPEED_100GB: + itr += DIV_ROUND_UP(17 * (avg_pkt_size + 24), + avg_pkt_size + 640); + break; + case ICE_AQ_LINK_SPEED_50GB: + itr += DIV_ROUND_UP(34 * (avg_pkt_size + 24), + avg_pkt_size + 640); + break; + case ICE_AQ_LINK_SPEED_40GB: + itr += DIV_ROUND_UP(43 * (avg_pkt_size + 24), + avg_pkt_size + 640); + break; + case ICE_AQ_LINK_SPEED_25GB: + itr += DIV_ROUND_UP(68 * (avg_pkt_size + 24), + avg_pkt_size + 640); + break; + case ICE_AQ_LINK_SPEED_20GB: + itr += DIV_ROUND_UP(85 * (avg_pkt_size + 24), + avg_pkt_size + 640); + break; + case ICE_AQ_LINK_SPEED_10GB: + /* fall through */ + default: + itr += DIV_ROUND_UP(170 * (avg_pkt_size + 24), + avg_pkt_size + 640); + break; + } + + if ((itr & ICE_ITR_MASK) > ICE_ITR_ADAPTIVE_MAX_USECS) { + itr &= ICE_ITR_ADAPTIVE_LATENCY; + itr += ICE_ITR_ADAPTIVE_MAX_USECS; + } + + return itr; +} + +/** + * ice_update_itr - update the adaptive ITR value based on statistics + * @q_vector: structure containing interrupt and ring information + * @rc: structure containing ring performance data + * + * Stores a new ITR value based on packets and byte + * counts during the last interrupt. The advantage of per interrupt + * computation is faster updates and more accurate ITR for the current + * traffic pattern. Constants in this function were computed + * based on theoretical maximum wire speed and thresholds were set based + * on testing data as well as attempting to minimize response time + * while increasing bulk throughput. + */ +static void +ice_update_itr(struct ice_q_vector *q_vector, struct ice_ring_container *rc) +{ + unsigned long next_update = jiffies; + unsigned int packets, bytes, itr; + bool container_is_rx; + + if (!rc->ring || !ITR_IS_DYNAMIC(rc->itr_setting)) + return; + + /* If itr_countdown is set it means we programmed an ITR within + * the last 4 interrupt cycles. This has a side effect of us + * potentially firing an early interrupt. In order to work around + * this we need to throw out any data received for a few + * interrupts following the update. + */ + if (q_vector->itr_countdown) { + itr = rc->target_itr; + goto clear_counts; + } + + container_is_rx = (&q_vector->rx == rc); + /* For Rx we want to push the delay up and default to low latency. + * for Tx we want to pull the delay down and default to high latency. + */ + itr = container_is_rx ? + ICE_ITR_ADAPTIVE_MIN_USECS | ICE_ITR_ADAPTIVE_LATENCY : + ICE_ITR_ADAPTIVE_MAX_USECS | ICE_ITR_ADAPTIVE_LATENCY; + + /* If we didn't update within up to 1 - 2 jiffies we can assume + * that either packets are coming in so slow there hasn't been + * any work, or that there is so much work that NAPI is dealing + * with interrupt moderation and we don't need to do anything. + */ + if (time_after(next_update, rc->next_update)) + goto clear_counts; + + packets = rc->total_pkts; + bytes = rc->total_bytes; + + if (container_is_rx) { + /* If Rx there are 1 to 4 packets and bytes are less than + * 9000 assume insufficient data to use bulk rate limiting + * approach unless Tx is already in bulk rate limiting. We + * are likely latency driven. + */ + if (packets && packets < 4 && bytes < 9000 && + (q_vector->tx.target_itr & ICE_ITR_ADAPTIVE_LATENCY)) { + itr = ICE_ITR_ADAPTIVE_LATENCY; + goto adjust_by_size_and_speed; + } + } else if (packets < 4) { + /* If we have Tx and Rx ITR maxed and Tx ITR is running in + * bulk mode and we are receiving 4 or fewer packets just + * reset the ITR_ADAPTIVE_LATENCY bit for latency mode so + * that the Rx can relax. + */ + if (rc->target_itr == ICE_ITR_ADAPTIVE_MAX_USECS && + (q_vector->rx.target_itr & ICE_ITR_MASK) == + ICE_ITR_ADAPTIVE_MAX_USECS) + goto clear_counts; + } else if (packets > 32) { + /* If we have processed over 32 packets in a single interrupt + * for Tx assume we need to switch over to "bulk" mode. + */ + rc->target_itr &= ~ICE_ITR_ADAPTIVE_LATENCY; + } + + /* We have no packets to actually measure against. This means + * either one of the other queues on this vector is active or + * we are a Tx queue doing TSO with too high of an interrupt rate. + * + * Between 4 and 56 we can assume that our current interrupt delay + * is only slightly too low. As such we should increase it by a small + * fixed amount. + */ + if (packets < 56) { + itr = rc->target_itr + ICE_ITR_ADAPTIVE_MIN_INC; + if ((itr & ICE_ITR_MASK) > ICE_ITR_ADAPTIVE_MAX_USECS) { + itr &= ICE_ITR_ADAPTIVE_LATENCY; + itr += ICE_ITR_ADAPTIVE_MAX_USECS; + } + goto clear_counts; + } + + if (packets <= 256) { + itr = min(q_vector->tx.current_itr, q_vector->rx.current_itr); + itr &= ICE_ITR_MASK; + + /* Between 56 and 112 is our "goldilocks" zone where we are + * working out "just right". Just report that our current + * ITR is good for us. + */ + if (packets <= 112) + goto clear_counts; + + /* If packet count is 128 or greater we are likely looking + * at a slight overrun of the delay we want. Try halving + * our delay to see if that will cut the number of packets + * in half per interrupt. + */ + itr >>= 1; + itr &= ICE_ITR_MASK; + if (itr < ICE_ITR_ADAPTIVE_MIN_USECS) + itr = ICE_ITR_ADAPTIVE_MIN_USECS; + + goto clear_counts; + } + + /* The paths below assume we are dealing with a bulk ITR since + * number of packets is greater than 256. We are just going to have + * to compute a value and try to bring the count under control, + * though for smaller packet sizes there isn't much we can do as + * NAPI polling will likely be kicking in sooner rather than later. + */ + itr = ICE_ITR_ADAPTIVE_BULK; + +adjust_by_size_and_speed: + + /* based on checks above packets cannot be 0 so division is safe */ + itr = ice_adjust_itr_by_size_and_speed(q_vector->vsi->port_info, + bytes / packets, itr); + +clear_counts: + /* write back value */ + rc->target_itr = itr; + + /* next update should occur within next jiffy */ + rc->next_update = next_update + 1; + + rc->total_bytes = 0; + rc->total_pkts = 0; +} + +/** * ice_buildreg_itr - build value for writing to the GLINT_DYN_CTL register * @itr_idx: interrupt throttling index - * @reg_itr: interrupt throttling value adjusted based on ITR granularity + * @itr: interrupt throttling value in usecs */ -static u32 ice_buildreg_itr(int itr_idx, u16 reg_itr) +static u32 ice_buildreg_itr(u16 itr_idx, u16 itr) { + /* The ITR value is reported in microseconds, and the register value is + * recorded in 2 microsecond units. For this reason we only need to + * shift by the GLINT_DYN_CTL_INTERVAL_S - ICE_ITR_GRAN_S to apply this + * granularity as a shift instead of division. The mask makes sure the + * ITR value is never odd so we don't accidentally write into the field + * prior to the ITR field. + */ + itr &= ICE_ITR_MASK; + return GLINT_DYN_CTL_INTENA_M | GLINT_DYN_CTL_CLEARPBA_M | (itr_idx << GLINT_DYN_CTL_ITR_INDX_S) | - (reg_itr << GLINT_DYN_CTL_INTERVAL_S); + (itr << (GLINT_DYN_CTL_INTERVAL_S - ICE_ITR_GRAN_S)); } +/* The act of updating the ITR will cause it to immediately trigger. In order + * to prevent this from throwing off adaptive update statistics we defer the + * update so that it can only happen so often. So after either Tx or Rx are + * updated we make the adaptive scheme wait until either the ITR completely + * expires via the next_update expiration or we have been through at least + * 3 interrupts. + */ +#define ITR_COUNTDOWN_START 3 + /** * ice_update_ena_itr - Update ITR and re-enable MSIX interrupt * @vsi: the VSI associated with the q_vector @@ -1068,10 +1347,14 @@ static u32 ice_buildreg_itr(int itr_idx, u16 reg_itr) static void ice_update_ena_itr(struct ice_vsi *vsi, struct ice_q_vector *q_vector) { - struct ice_hw *hw = &vsi->back->hw; - struct ice_ring_container *rc; + struct ice_ring_container *tx = &q_vector->tx; + struct ice_ring_container *rx = &q_vector->rx; u32 itr_val; + /* This will do nothing if dynamic updates are not enabled */ + ice_update_itr(q_vector, tx); + ice_update_itr(q_vector, rx); + /* This block of logic allows us to get away with only updating * one ITR value with each interrupt. The idea is to perform a * pseudo-lazy update with the following criteria. @@ -1080,35 +1363,36 @@ ice_update_ena_itr(struct ice_vsi *vsi, struct ice_q_vector *q_vector) * 2. If we must reduce an ITR that is given highest priority. * 3. We then give priority to increasing ITR based on amount. */ - if (q_vector->rx.target_itr < q_vector->rx.current_itr) { - rc = &q_vector->rx; + if (rx->target_itr < rx->current_itr) { /* Rx ITR needs to be reduced, this is highest priority */ - itr_val = ice_buildreg_itr(rc->itr_idx, rc->target_itr); - rc->current_itr = rc->target_itr; - } else if ((q_vector->tx.target_itr < q_vector->tx.current_itr) || - ((q_vector->rx.target_itr - q_vector->rx.current_itr) < - (q_vector->tx.target_itr - q_vector->tx.current_itr))) { - rc = &q_vector->tx; + itr_val = ice_buildreg_itr(rx->itr_idx, rx->target_itr); + rx->current_itr = rx->target_itr; + q_vector->itr_countdown = ITR_COUNTDOWN_START; + } else if ((tx->target_itr < tx->current_itr) || + ((rx->target_itr - rx->current_itr) < + (tx->target_itr - tx->current_itr))) { /* Tx ITR needs to be reduced, this is second priority * Tx ITR needs to be increased more than Rx, fourth priority */ - itr_val = ice_buildreg_itr(rc->itr_idx, rc->target_itr); - rc->current_itr = rc->target_itr; - } else if (q_vector->rx.current_itr != q_vector->rx.target_itr) { - rc = &q_vector->rx; + itr_val = ice_buildreg_itr(tx->itr_idx, tx->target_itr); + tx->current_itr = tx->target_itr; + q_vector->itr_countdown = ITR_COUNTDOWN_START; + } else if (rx->current_itr != rx->target_itr) { /* Rx ITR needs to be increased, third priority */ - itr_val = ice_buildreg_itr(rc->itr_idx, rc->target_itr); - rc->current_itr = rc->target_itr; + itr_val = ice_buildreg_itr(rx->itr_idx, rx->target_itr); + rx->current_itr = rx->target_itr; + q_vector->itr_countdown = ITR_COUNTDOWN_START; } else { /* Still have to re-enable the interrupts */ itr_val = ice_buildreg_itr(ICE_ITR_NONE, 0); + if (q_vector->itr_countdown) + q_vector->itr_countdown--; } - if (!test_bit(__ICE_DOWN, vsi->state)) { - int vector = vsi->hw_base_vector + q_vector->v_idx; - - wr32(hw, GLINT_DYN_CTL(vector), itr_val); - } + if (!test_bit(__ICE_DOWN, vsi->state)) + wr32(&vsi->back->hw, + GLINT_DYN_CTL(q_vector->reg_idx), + itr_val); } /** @@ -1354,14 +1638,14 @@ ice_tx_map(struct ice_ring *tx_ring, struct ice_tx_buf *first, ice_maybe_stop_tx(tx_ring, DESC_NEEDED); /* notify HW of packet */ - if (netif_xmit_stopped(txring_txq(tx_ring)) || !skb->xmit_more) { + if (netif_xmit_stopped(txring_txq(tx_ring)) || !netdev_xmit_more()) { writel(i, tx_ring->tail); } return; dma_error: - /* clear dma mappings for failed tx_buf map */ + /* clear DMA mappings for failed tx_buf map */ for (;;) { tx_buf = &tx_ring->tx_buf[i]; ice_unmap_and_free_tx_buf(tx_ring, tx_buf); @@ -1475,7 +1759,7 @@ int ice_tx_csum(struct ice_tx_buf *first, struct ice_tx_offload_params *off) } /** - * ice_tx_prepare_vlan_flags - prepare generic TX VLAN tagging flags for HW + * ice_tx_prepare_vlan_flags - prepare generic Tx VLAN tagging flags for HW * @tx_ring: ring to send buffer on * @first: pointer to struct ice_tx_buf * @@ -1501,7 +1785,7 @@ ice_tx_prepare_vlan_flags(struct ice_ring *tx_ring, struct ice_tx_buf *first) * to the encapsulated ethertype. */ skb->protocol = vlan_get_protocol(skb); - goto out; + return 0; } /* if we have a HW VLAN tag being added, default to the HW one */ @@ -1523,8 +1807,7 @@ ice_tx_prepare_vlan_flags(struct ice_ring *tx_ring, struct ice_tx_buf *first) first->tx_flags |= ICE_TX_FLAGS_SW_VLAN; } -out: - return 0; + return ice_tx_prepare_vlan_flags_dcb(tx_ring, first); } /** @@ -1561,6 +1844,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) if (err < 0) return err; + /* cppcheck-suppress unreadVariable */ ip.hdr = skb_network_header(skb); l4.hdr = skb_transport_header(skb); @@ -1590,10 +1874,10 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) cd_mss = skb_shinfo(skb)->gso_size; /* record cdesc_qw1 with TSO parameters */ - off->cd_qw1 |= ICE_TX_DESC_DTYPE_CTX | - (ICE_TX_CTX_DESC_TSO << ICE_TXD_CTX_QW1_CMD_S) | - (cd_tso_len << ICE_TXD_CTX_QW1_TSO_LEN_S) | - (cd_mss << ICE_TXD_CTX_QW1_MSS_S); + off->cd_qw1 |= (u64)(ICE_TX_DESC_DTYPE_CTX | + (ICE_TX_CTX_DESC_TSO << ICE_TXD_CTX_QW1_CMD_S) | + (cd_tso_len << ICE_TXD_CTX_QW1_TSO_LEN_S) | + (cd_mss << ICE_TXD_CTX_QW1_MSS_S)); first->tx_flags |= ICE_TX_FLAGS_TSO; return 1; } diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index fc358ea81816..ec76aba347b9 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -45,27 +45,32 @@ #define ICE_TX_FLAGS_HW_VLAN BIT(1) #define ICE_TX_FLAGS_SW_VLAN BIT(2) #define ICE_TX_FLAGS_VLAN_M 0xffff0000 +#define ICE_TX_FLAGS_VLAN_PR_M 0xe0000000 +#define ICE_TX_FLAGS_VLAN_PR_S 29 #define ICE_TX_FLAGS_VLAN_S 16 +#define ICE_RX_DMA_ATTR \ + (DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING) + struct ice_tx_buf { struct ice_tx_desc *next_to_watch; struct sk_buff *skb; unsigned int bytecount; unsigned short gso_segs; u32 tx_flags; - DEFINE_DMA_UNMAP_ADDR(dma); DEFINE_DMA_UNMAP_LEN(len); + DEFINE_DMA_UNMAP_ADDR(dma); }; struct ice_tx_offload_params { - u8 header_len; + u64 cd_qw1; + struct ice_ring *tx_ring; u32 td_cmd; u32 td_offset; u32 td_l2tag1; - u16 cd_l2tag2; u32 cd_tunnel_params; - u64 cd_qw1; - struct ice_ring *tx_ring; + u16 cd_l2tag2; + u8 header_len; }; struct ice_rx_buf { @@ -73,6 +78,7 @@ struct ice_rx_buf { dma_addr_t dma; struct page *page; unsigned int page_offset; + u16 pagecnt_bias; }; struct ice_q_stats { @@ -124,11 +130,19 @@ enum ice_rx_dtype { #define ICE_ITR_DYNAMIC 0x8000 /* used as flag for itr_setting */ #define ITR_IS_DYNAMIC(setting) (!!((setting) & ICE_ITR_DYNAMIC)) #define ITR_TO_REG(setting) ((setting) & ~ICE_ITR_DYNAMIC) -#define ICE_ITR_GRAN_S 1 /* Assume ITR granularity is 2us */ +#define ICE_ITR_GRAN_S 1 /* ITR granularity is always 2us */ +#define ICE_ITR_GRAN_US BIT(ICE_ITR_GRAN_S) #define ICE_ITR_MASK 0x1FFE /* ITR register value alignment mask */ #define ITR_REG_ALIGN(setting) __ALIGN_MASK(setting, ~ICE_ITR_MASK) +#define ICE_ITR_ADAPTIVE_MIN_INC 0x0002 +#define ICE_ITR_ADAPTIVE_MIN_USECS 0x0002 +#define ICE_ITR_ADAPTIVE_MAX_USECS 0x00FA +#define ICE_ITR_ADAPTIVE_LATENCY 0x8000 +#define ICE_ITR_ADAPTIVE_BULK 0x0000 + #define ICE_DFLT_INTRL 0 +#define ICE_MAX_INTRL 236 /* Legacy or Advanced Mode Queue */ #define ICE_TX_ADVANCED 0 @@ -136,6 +150,7 @@ enum ice_rx_dtype { /* descriptor ring, associated with a VSI */ struct ice_ring { + /* CL1 - 1st cacheline starts here */ struct ice_ring *next; /* pointer to next ring in q_vector */ void *desc; /* Descriptor ring memory */ struct device *dev; /* Used for DMA mapping */ @@ -147,8 +162,11 @@ struct ice_ring { struct ice_tx_buf *tx_buf; struct ice_rx_buf *rx_buf; }; + /* CL2 - 2nd cacheline starts here */ u16 q_index; /* Queue number of ring */ - u32 txq_teid; /* Added Tx queue TEID */ + u16 q_handle; /* Queue handle per TC */ + + u8 ring_active:1; /* is ring online or not */ u16 count; /* Number of descriptors */ u16 reg_idx; /* HW register index of the ring */ @@ -156,8 +174,7 @@ struct ice_ring { /* used in interrupt processing */ u16 next_to_use; u16 next_to_clean; - - u8 ring_active; /* is ring online or not */ + u16 next_to_alloc; /* stats structs */ struct ice_q_stats stats; @@ -167,27 +184,26 @@ struct ice_ring { struct ice_rxq_stats rx_stats; }; - unsigned int size; /* length of descriptor ring in bytes */ - dma_addr_t dma; /* physical address of ring */ struct rcu_head rcu; /* to avoid race on free */ - u16 next_to_alloc; + /* CLX - the below items are only accessed infrequently and should be + * in their own cache line if possible + */ + dma_addr_t dma; /* physical address of ring */ + unsigned int size; /* length of descriptor ring in bytes */ + u32 txq_teid; /* Added Tx queue TEID */ + u16 rx_buf_len; +#ifdef CONFIG_DCB + u8 dcb_tc; /* Traffic class of ring */ +#endif /* CONFIG_DCB */ } ____cacheline_internodealigned_in_smp; -enum ice_latency_range { - ICE_LOWEST_LATENCY = 0, - ICE_LOW_LATENCY = 1, - ICE_BULK_LATENCY = 2, - ICE_ULTRA_LATENCY = 3, -}; - struct ice_ring_container { /* head of linked-list of rings */ struct ice_ring *ring; unsigned long next_update; /* jiffies value of next queue update */ unsigned int total_bytes; /* total bytes processed this int */ unsigned int total_pkts; /* total packets processed this int */ - enum ice_latency_range latency_range; - int itr_idx; /* index in the interrupt vector */ + u16 itr_idx; /* index in the interrupt vector */ u16 target_itr; /* value in usecs divided by the hw->itr_gran */ u16 current_itr; /* value in usecs divided by the hw->itr_gran */ /* high bit set means dynamic ITR, rest is used to store user diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 17086d5b5c33..24bbef8bbe69 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -23,7 +23,9 @@ static inline bool ice_is_tc_ena(u8 bitmap, u8 tc) /* debug masks - set these bits in hw->debug_mask to control output */ #define ICE_DBG_INIT BIT_ULL(1) +#define ICE_DBG_FW_LOG BIT_ULL(3) #define ICE_DBG_LINK BIT_ULL(4) +#define ICE_DBG_PHY BIT_ULL(5) #define ICE_DBG_QCTX BIT_ULL(6) #define ICE_DBG_NVM BIT_ULL(7) #define ICE_DBG_LAN BIT_ULL(8) @@ -60,6 +62,13 @@ enum ice_fc_mode { ICE_FC_DFLT }; +enum ice_fec_mode { + ICE_FEC_NONE = 0, + ICE_FEC_RS, + ICE_FEC_BASER, + ICE_FEC_AUTO +}; + enum ice_set_fc_aq_failures { ICE_SET_FC_AQ_FAIL_NONE = 0, ICE_SET_FC_AQ_FAIL_GET, @@ -85,12 +94,14 @@ enum ice_media_type { enum ice_vsi_type { ICE_VSI_PF = 0, ICE_VSI_VF, + ICE_VSI_LB = 6, }; struct ice_link_status { /* Refer to ice_aq_phy_type for bits definition */ u64 phy_type_low; u64 phy_type_high; + u8 topo_media_conflict; u16 max_frame_size; u16 link_speed; u16 req_speeds; @@ -98,6 +109,7 @@ struct ice_link_status { u8 link_info; u8 an_info; u8 ext_info; + u8 fec_info; u8 pacing; /* Refer to #define from module_type[ICE_MODULE_TYPE_TOTAL_BYTE] of * ice_aqc_get_phy_caps structure @@ -106,7 +118,7 @@ struct ice_link_status { }; /* Different reset sources for which a disable queue AQ call has to be made in - * order to clean the TX scheduler as a part of the reset + * order to clean the Tx scheduler as a part of the reset */ enum ice_disq_rst_src { ICE_NO_RESET = 0, @@ -128,11 +140,11 @@ struct ice_phy_info { struct ice_hw_common_caps { u32 valid_functions; - /* TX/RX queues */ - u16 num_rxq; /* Number/Total RX queues */ - u16 rxq_first_id; /* First queue ID for RX queues */ - u16 num_txq; /* Number/Total TX queues */ - u16 txq_first_id; /* First queue ID for TX queues */ + /* Tx/Rx queues */ + u16 num_rxq; /* Number/Total Rx queues */ + u16 rxq_first_id; /* First queue ID for Rx queues */ + u16 num_txq; /* Number/Total Tx queues */ + u16 txq_first_id; /* First queue ID for Tx queues */ /* MSI-X vectors */ u16 num_msix_vectors; @@ -147,6 +159,8 @@ struct ice_hw_common_caps { /* RSS related capabilities */ u16 rss_table_size; /* 512 for PFs and 64 for VFs */ u8 rss_table_entry_width; /* RSS Entry width in bits */ + + u8 dcb; }; /* Function specific capabilities */ @@ -209,12 +223,17 @@ struct ice_nvm_info { #define ICE_MAX_TRAFFIC_CLASS 8 #define ICE_TXSCHED_MAX_BRANCHES ICE_MAX_TRAFFIC_CLASS +#define ice_for_each_traffic_class(_i) \ + for ((_i) = 0; (_i) < ICE_MAX_TRAFFIC_CLASS; (_i)++) + +#define ICE_INVAL_TEID 0xFFFFFFFF + struct ice_sched_node { struct ice_sched_node *parent; struct ice_sched_node *sibling; /* next sibling in the same layer */ struct ice_sched_node **children; struct ice_aqc_txsched_elem_data info; - u32 agg_id; /* aggregator group id */ + u32 agg_id; /* aggregator group ID */ u16 vsi_handle; u8 in_use; /* suspended or in use */ u8 tx_sched_layer; /* Logical Layer (1-9) */ @@ -241,13 +260,12 @@ enum ice_agg_type { #define ICE_SCHED_DFLT_RL_PROF_ID 0 #define ICE_SCHED_DFLT_BW_WT 1 -/* vsi type list entry to locate corresponding vsi/ag nodes */ +/* VSI type list entry to locate corresponding VSI/ag nodes */ struct ice_sched_vsi_info { struct ice_sched_node *vsi_node[ICE_MAX_TRAFFIC_CLASS]; struct ice_sched_node *ag_node[ICE_MAX_TRAFFIC_CLASS]; struct list_head list_entry; u16 max_lanq[ICE_MAX_TRAFFIC_CLASS]; - u16 vsi_id; }; /* driver defines the policy */ @@ -257,15 +275,70 @@ struct ice_sched_tx_policy { u8 rdma_ena; }; +/* CEE or IEEE 802.1Qaz ETS Configuration data */ +struct ice_dcb_ets_cfg { + u8 willing; + u8 cbs; + u8 maxtcs; + u8 prio_table[ICE_MAX_TRAFFIC_CLASS]; + u8 tcbwtable[ICE_MAX_TRAFFIC_CLASS]; + u8 tsatable[ICE_MAX_TRAFFIC_CLASS]; +}; + +/* CEE or IEEE 802.1Qaz PFC Configuration data */ +struct ice_dcb_pfc_cfg { + u8 willing; + u8 mbc; + u8 pfccap; + u8 pfcena; +}; + +/* CEE or IEEE 802.1Qaz Application Priority data */ +struct ice_dcb_app_priority_table { + u16 prot_id; + u8 priority; + u8 selector; +}; + +#define ICE_MAX_USER_PRIORITY 8 +#define ICE_DCBX_MAX_APPS 32 +#define ICE_LLDPDU_SIZE 1500 +#define ICE_TLV_STATUS_OPER 0x1 +#define ICE_TLV_STATUS_SYNC 0x2 +#define ICE_TLV_STATUS_ERR 0x4 +#define ICE_APP_PROT_ID_FCOE 0x8906 +#define ICE_APP_PROT_ID_ISCSI 0x0cbc +#define ICE_APP_PROT_ID_FIP 0x8914 +#define ICE_APP_SEL_ETHTYPE 0x1 +#define ICE_APP_SEL_TCPIP 0x2 +#define ICE_CEE_APP_SEL_ETHTYPE 0x0 +#define ICE_CEE_APP_SEL_TCPIP 0x1 + +struct ice_dcbx_cfg { + u32 numapps; + u32 tlv_status; /* CEE mode TLV status */ + struct ice_dcb_ets_cfg etscfg; + struct ice_dcb_ets_cfg etsrec; + struct ice_dcb_pfc_cfg pfc; + struct ice_dcb_app_priority_table app[ICE_DCBX_MAX_APPS]; + u8 dcbx_mode; +#define ICE_DCBX_MODE_CEE 0x1 +#define ICE_DCBX_MODE_IEEE 0x2 + u8 app_mode; +#define ICE_DCBX_APPS_NON_WILLING 0x1 +}; + struct ice_port_info { struct ice_sched_node *root; /* Root Node per Port */ - struct ice_hw *hw; /* back pointer to hw instance */ + struct ice_hw *hw; /* back pointer to HW instance */ u32 last_node_teid; /* scheduler last node info */ u16 sw_id; /* Initial switch ID belongs to port */ u16 pf_vf_num; u8 port_state; #define ICE_SCHED_PORT_STATE_INIT 0x0 #define ICE_SCHED_PORT_STATE_READY 0x1 + u8 lport; +#define ICE_LPORT_MASK 0xff u16 dflt_tx_vsi_rule_id; u16 dflt_tx_vsi_num; u16 dflt_rx_vsi_rule_id; @@ -274,9 +347,14 @@ struct ice_port_info { struct ice_mac_info mac; struct ice_phy_info phy; struct mutex sched_lock; /* protect access to TXSched tree */ - u8 lport; -#define ICE_LPORT_MASK 0xff - u8 is_vf; + struct ice_dcbx_cfg local_dcbx_cfg; /* Oper/Local Cfg */ + /* DCBX info */ + struct ice_dcbx_cfg remote_dcbx_cfg; /* Peer Cfg */ + struct ice_dcbx_cfg desired_dcbx_cfg; /* CEE Desired Cfg */ + /* LLDP/DCBX Status */ + u8 dcbx_status:3; /* see ICE_DCBX_STATUS_DIS */ + u8 is_sw_lldp:1; + u8 is_vf:1; }; struct ice_switch_info { @@ -320,7 +398,7 @@ struct ice_hw { u8 pf_id; /* device profile info */ - /* TX Scheduler values */ + /* Tx Scheduler values */ u16 num_tx_sched_layers; u16 num_tx_sched_phys_layers; u8 flattened_layers; @@ -331,7 +409,7 @@ struct ice_hw { struct ice_vsi_ctx *vsi_ctx[ICE_MAX_VSI]; u8 evb_veb; /* true for VEB, false for VEPA */ - u8 reset_ongoing; /* true if hw is in reset, false otherwise */ + u8 reset_ongoing; /* true if HW is in reset, false otherwise */ struct ice_bus_info bus; struct ice_nvm_info nvm; struct ice_hw_dev_caps dev_caps; /* device capabilities */ @@ -356,7 +434,7 @@ struct ice_hw { struct ice_fw_log_cfg fw_log; /* Device max aggregate bandwidths corresponding to the GL_PWR_MODE_CTL - * register. Used for determining the itr/intrl granularity during + * register. Used for determining the ITR/intrl granularity during * initialization. */ #define ICE_MAX_AGG_BW_200G 0x0 @@ -410,6 +488,11 @@ struct ice_hw_port_stats { u64 link_xoff_rx; /* lxoffrxc */ u64 link_xon_tx; /* lxontxc */ u64 link_xoff_tx; /* lxofftxc */ + u64 priority_xon_rx[8]; /* pxonrxc[8] */ + u64 priority_xoff_rx[8]; /* pxoffrxc[8] */ + u64 priority_xon_tx[8]; /* pxontxc[8] */ + u64 priority_xoff_tx[8]; /* pxofftxc[8] */ + u64 priority_xon_2_xoff[8]; /* pxon2offc[8] */ u64 rx_size_64; /* prc64 */ u64 rx_size_127; /* prc127 */ u64 rx_size_255; /* prc255 */ diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 57155b4a59dc..5d24b539648f 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -5,6 +5,37 @@ #include "ice_lib.h" /** + * ice_err_to_virt err - translate errors for VF return code + * @ice_err: error return code + */ +static enum virtchnl_status_code ice_err_to_virt_err(enum ice_status ice_err) +{ + switch (ice_err) { + case ICE_SUCCESS: + return VIRTCHNL_STATUS_SUCCESS; + case ICE_ERR_BAD_PTR: + case ICE_ERR_INVAL_SIZE: + case ICE_ERR_DEVICE_NOT_SUPPORTED: + case ICE_ERR_PARAM: + case ICE_ERR_CFG: + return VIRTCHNL_STATUS_ERR_PARAM; + case ICE_ERR_NO_MEMORY: + return VIRTCHNL_STATUS_ERR_NO_MEMORY; + case ICE_ERR_NOT_READY: + case ICE_ERR_RESET_FAILED: + case ICE_ERR_FW_API_VER: + case ICE_ERR_AQ_ERROR: + case ICE_ERR_AQ_TIMEOUT: + case ICE_ERR_AQ_FULL: + case ICE_ERR_AQ_NO_WORK: + case ICE_ERR_AQ_EMPTY: + return VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; + default: + return VIRTCHNL_STATUS_ERR_NOT_SUPPORTED; + } +} + +/** * ice_vc_vf_broadcast - Broadcast a message to all VFs on PF * @pf: pointer to the PF structure * @v_opcode: operation code @@ -14,7 +45,7 @@ */ static void ice_vc_vf_broadcast(struct ice_pf *pf, enum virtchnl_ops v_opcode, - enum ice_status v_retval, u8 *msg, u16 msglen) + enum virtchnl_status_code v_retval, u8 *msg, u16 msglen) { struct ice_hw *hw = &pf->hw; struct ice_vf *vf = pf->vf; @@ -72,7 +103,7 @@ ice_set_pfe_link_forced(struct ice_vf *vf, struct virtchnl_pf_event *pfe, u16 link_speed; if (link_up) - link_speed = ICE_AQ_LINK_SPEED_40GB; + link_speed = ICE_AQ_LINK_SPEED_100GB; else link_speed = ICE_AQ_LINK_SPEED_UNKNOWN; @@ -104,37 +135,26 @@ static void ice_vc_notify_vf_link_state(struct ice_vf *vf) ice_set_pfe_link(vf, &pfe, ls->link_speed, ls->link_info & ICE_AQ_LINK_UP); - ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, 0, (u8 *)&pfe, + ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, + VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe, sizeof(pfe), NULL); } /** - * ice_get_vf_vector - get VF interrupt vector register offset - * @vf_msix: number of MSIx vector per VF on a PF - * @vf_id: VF identifier - * @i: index of MSIx vector - */ -static u32 ice_get_vf_vector(int vf_msix, int vf_id, int i) -{ - return ((i == 0) ? VFINT_DYN_CTLN(vf_id) : - VFINT_DYN_CTLN(((vf_msix - 1) * (vf_id)) + (i - 1))); -} - -/** * ice_free_vf_res - Free a VF's resources * @vf: pointer to the VF info */ static void ice_free_vf_res(struct ice_vf *vf) { struct ice_pf *pf = vf->pf; - int i, pf_vf_msix; + int i, last_vector_idx; /* First, disable VF's configuration API to prevent OS from * accessing the VF's VSI after it's freed or invalidated. */ clear_bit(ICE_VF_STATE_INIT, vf->vf_states); - /* free vsi & disconnect it from the parent uplink */ + /* free VSI and disconnect it from the parent uplink */ if (vf->lan_vsi_idx) { ice_vsi_release(pf->vsi[vf->lan_vsi_idx]); vf->lan_vsi_idx = 0; @@ -142,13 +162,10 @@ static void ice_free_vf_res(struct ice_vf *vf) vf->num_mac = 0; } - pf_vf_msix = pf->num_vf_msix; + last_vector_idx = vf->first_vector_idx + pf->num_vf_msix - 1; /* Disable interrupts so that VF starts in a known state */ - for (i = 0; i < pf_vf_msix; i++) { - u32 reg_idx; - - reg_idx = ice_get_vf_vector(pf_vf_msix, vf->vf_id, i); - wr32(&pf->hw, reg_idx, VFINT_DYN_CTLN_CLEARPBA_M); + for (i = vf->first_vector_idx; i <= last_vector_idx; i++) { + wr32(&pf->hw, GLINT_DYN_CTL(i), GLINT_DYN_CTL_CLEARPBA_M); ice_flush(&pf->hw); } /* reset some of the state variables keeping track of the resources */ @@ -173,8 +190,7 @@ static void ice_dis_vf_mappings(struct ice_vf *vf) wr32(hw, VPINT_ALLOC(vf->vf_id), 0); wr32(hw, VPINT_ALLOC_PCI(vf->vf_id), 0); - first = vf->first_vector_idx + - hw->func_caps.common_cap.msix_vector_first_id; + first = vf->first_vector_idx; last = first + pf->num_vf_msix - 1; for (v = first; v <= last; v++) { u32 reg; @@ -200,6 +216,42 @@ static void ice_dis_vf_mappings(struct ice_vf *vf) } /** + * ice_sriov_free_msix_res - Reset/free any used MSIX resources + * @pf: pointer to the PF structure + * + * If MSIX entries from the pf->irq_tracker were needed then we need to + * reset the irq_tracker->end and give back the entries we needed to + * num_avail_sw_msix. + * + * If no MSIX entries were taken from the pf->irq_tracker then just clear + * the pf->sriov_base_vector. + * + * Returns 0 on success, and -EINVAL on error. + */ +static int ice_sriov_free_msix_res(struct ice_pf *pf) +{ + struct ice_res_tracker *res; + + if (!pf) + return -EINVAL; + + res = pf->irq_tracker; + if (!res) + return -EINVAL; + + /* give back irq_tracker resources used */ + if (pf->sriov_base_vector < res->num_entries) { + res->end = res->num_entries; + pf->num_avail_sw_msix += + res->num_entries - pf->sriov_base_vector; + } + + pf->sriov_base_vector = 0; + + return 0; +} + +/** * ice_free_vfs - Free all VFs * @pf: pointer to the PF structure */ @@ -214,15 +266,6 @@ void ice_free_vfs(struct ice_pf *pf) while (test_and_set_bit(__ICE_VF_DIS, pf->state)) usleep_range(1000, 2000); - /* Disable IOV before freeing resources. This lets any VF drivers - * running in the host get themselves cleaned up before we yank - * the carpet out from underneath their feet. - */ - if (!pci_vfs_assigned(pf->pdev)) - pci_disable_sriov(pf->pdev); - else - dev_warn(&pf->pdev->dev, "VFs are assigned - not disabling SR-IOV\n"); - /* Avoid wait time by stopping all VFs at the same time */ for (i = 0; i < pf->num_alloc_vfs; i++) { struct ice_vsi *vsi; @@ -238,6 +281,15 @@ void ice_free_vfs(struct ice_pf *pf) clear_bit(ICE_VF_STATE_ENA, pf->vf[i].vf_states); } + /* Disable IOV before freeing resources. This lets any VF drivers + * running in the host get themselves cleaned up before we yank + * the carpet out from underneath their feet. + */ + if (!pci_vfs_assigned(pf->pdev)) + pci_disable_sriov(pf->pdev); + else + dev_warn(&pf->pdev->dev, "VFs are assigned - not disabling SR-IOV\n"); + tmp = pf->num_alloc_vfs; pf->num_vf_qps = 0; pf->num_alloc_vfs = 0; @@ -256,6 +308,10 @@ void ice_free_vfs(struct ice_pf *pf) } } + if (ice_sriov_free_msix_res(pf)) + dev_err(&pf->pdev->dev, + "Failed to free MSIX resources used by SR-IOV\n"); + devm_kfree(&pf->pdev->dev, pf->vf); pf->vf = NULL; @@ -343,11 +399,41 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr) } /** - * ice_vsi_set_pvid - Set port VLAN id for the VSI - * @vsi: the VSI being changed - * @vid: the VLAN id to set as a PVID + * ice_vsi_set_pvid_fill_ctxt - Set VSI ctxt for add PVID + * @ctxt: the VSI ctxt to fill + * @vid: the VLAN ID to set as a PVID */ -static int ice_vsi_set_pvid(struct ice_vsi *vsi, u16 vid) +static void ice_vsi_set_pvid_fill_ctxt(struct ice_vsi_ctx *ctxt, u16 vid) +{ + ctxt->info.vlan_flags = (ICE_AQ_VSI_VLAN_MODE_UNTAGGED | + ICE_AQ_VSI_PVLAN_INSERT_PVID | + ICE_AQ_VSI_VLAN_EMOD_STR); + ctxt->info.pvid = cpu_to_le16(vid); + ctxt->info.sw_flags2 |= ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; + ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID | + ICE_AQ_VSI_PROP_SW_VALID); +} + +/** + * ice_vsi_kill_pvid_fill_ctxt - Set VSI ctx for remove PVID + * @ctxt: the VSI ctxt to fill + */ +static void ice_vsi_kill_pvid_fill_ctxt(struct ice_vsi_ctx *ctxt) +{ + ctxt->info.vlan_flags = ICE_AQ_VSI_VLAN_EMOD_NOTHING; + ctxt->info.vlan_flags |= ICE_AQ_VSI_VLAN_MODE_ALL; + ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; + ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID | + ICE_AQ_VSI_PROP_SW_VALID); +} + +/** + * ice_vsi_manage_pvid - Enable or disable port VLAN for VSI + * @vsi: the VSI to update + * @vid: the VLAN ID to set as a PVID + * @enable: true for enable PVID false for disable + */ +static int ice_vsi_manage_pvid(struct ice_vsi *vsi, u16 vid, bool enable) { struct device *dev = &vsi->back->pdev->dev; struct ice_hw *hw = &vsi->back->hw; @@ -359,50 +445,31 @@ static int ice_vsi_set_pvid(struct ice_vsi *vsi, u16 vid) if (!ctxt) return -ENOMEM; - ctxt->info.vlan_flags = (ICE_AQ_VSI_VLAN_MODE_UNTAGGED | - ICE_AQ_VSI_PVLAN_INSERT_PVID | - ICE_AQ_VSI_VLAN_EMOD_STR); - ctxt->info.pvid = cpu_to_le16(vid); - ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_VLAN_VALID); + ctxt->info = vsi->info; + if (enable) + ice_vsi_set_pvid_fill_ctxt(ctxt, vid); + else + ice_vsi_kill_pvid_fill_ctxt(ctxt); status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_info(dev, "update VSI for VLAN insert failed, err %d aq_err %d\n", + dev_info(dev, "update VSI for port VLAN failed, err %d aq_err %d\n", status, hw->adminq.sq_last_status); ret = -EIO; goto out; } - vsi->info.pvid = ctxt->info.pvid; - vsi->info.vlan_flags = ctxt->info.vlan_flags; + vsi->info = ctxt->info; out: devm_kfree(dev, ctxt); return ret; } /** - * ice_vsi_kill_pvid - Remove port VLAN id from the VSI - * @vsi: the VSI being changed - */ -static int ice_vsi_kill_pvid(struct ice_vsi *vsi) -{ - struct ice_pf *pf = vsi->back; - - if (ice_vsi_manage_vlan_stripping(vsi, false)) { - dev_err(&pf->pdev->dev, "Error removing Port VLAN on VSI %i\n", - vsi->vsi_num); - return -ENODEV; - } - - vsi->info.pvid = 0; - return 0; -} - -/** * ice_vf_vsi_setup - Set up a VF VSI * @pf: board private structure * @pi: pointer to the port_info instance - * @vf_id: defines VF id to which this VSI connects. + * @vf_id: defines VF ID to which this VSI connects. * * Returns pointer to the successfully allocated VSI struct on success, * otherwise returns NULL on failure. @@ -414,6 +481,22 @@ ice_vf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, u16 vf_id) } /** + * ice_calc_vf_first_vector_idx - Calculate absolute MSIX vector index in HW + * @pf: pointer to PF structure + * @vf: pointer to VF that the first MSIX vector index is being calculated for + * + * This returns the first MSIX vector index in HW that is used by this VF and + * this will always be the OICR index in the AVF driver so any functionality + * using vf->first_vector_idx for queue configuration will have to increment by + * 1 to avoid meddling with the OICR index. + */ +static int ice_calc_vf_first_vector_idx(struct ice_pf *pf, struct ice_vf *vf) +{ + return pf->hw.func_caps.common_cap.msix_vector_first_id + + pf->sriov_base_vector + vf->vf_id * pf->num_vf_msix; +} + +/** * ice_alloc_vsi_res - Setup VF VSI and its resources * @vf: pointer to the VF structure * @@ -427,8 +510,10 @@ static int ice_alloc_vsi_res(struct ice_vf *vf) struct ice_vsi *vsi; int status = 0; - vsi = ice_vf_vsi_setup(pf, pf->hw.port_info, vf->vf_id); + /* first vector index is the VFs OICR index */ + vf->first_vector_idx = ice_calc_vf_first_vector_idx(pf, vf); + vsi = ice_vf_vsi_setup(pf, pf->hw.port_info, vf->vf_id); if (!vsi) { dev_err(&pf->pdev->dev, "Failed to create VF VSI\n"); return -ENOMEM; @@ -437,17 +522,11 @@ static int ice_alloc_vsi_res(struct ice_vf *vf) vf->lan_vsi_idx = vsi->idx; vf->lan_vsi_num = vsi->vsi_num; - /* first vector index is the VFs OICR index */ - vf->first_vector_idx = vsi->hw_base_vector; - /* Since hw_base_vector holds the vector where data queue interrupts - * starts, increment by 1 since VFs allocated vectors include OICR intr - * as well. - */ - vsi->hw_base_vector += 1; - /* Check if port VLAN exist before, and restore it accordingly */ - if (vf->port_vlan_id) - ice_vsi_set_pvid(vsi, vf->port_vlan_id); + if (vf->port_vlan_id) { + ice_vsi_manage_pvid(vsi, vf->port_vlan_id, true); + ice_vsi_add_vlan(vsi, vf->port_vlan_id & ICE_VLAN_M); + } eth_broadcast_addr(broadcast); @@ -468,7 +547,7 @@ static int ice_alloc_vsi_res(struct ice_vf *vf) /* Clear this bit after VF initialization since we shouldn't reclaim * and reassign interrupts for synchronous or asynchronous VFR events. - * We dont want to reconfigure interrupts since AVF driver doesn't + * We don't want to reconfigure interrupts since AVF driver doesn't * expect vector assignment to be changed unless there is a request for * more vectors. */ @@ -484,6 +563,8 @@ ice_alloc_vsi_res_exit: */ static int ice_alloc_vf_res(struct ice_vf *vf) { + struct ice_pf *pf = vf->pf; + int tx_rx_queue_left; int status; /* setup VF VSI and necessary resources */ @@ -491,6 +572,15 @@ static int ice_alloc_vf_res(struct ice_vf *vf) if (status) goto ice_alloc_vf_res_exit; + /* Update number of VF queues, in case VF had requested for queue + * changes + */ + tx_rx_queue_left = min_t(int, pf->q_left_tx, pf->q_left_rx); + tx_rx_queue_left += ICE_DFLT_QS_PER_VF; + if (vf->num_req_qs && vf->num_req_qs <= tx_rx_queue_left && + vf->num_req_qs != vf->num_vf_qs) + vf->num_vf_qs = vf->num_req_qs; + if (vf->trusted) set_bit(ICE_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps); else @@ -524,8 +614,7 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) hw = &pf->hw; vsi = pf->vsi[vf->lan_vsi_idx]; - first = vf->first_vector_idx + - hw->func_caps.common_cap.msix_vector_first_id; + first = vf->first_vector_idx; last = (first + pf->num_vf_msix) - 1; abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; @@ -548,6 +637,10 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) wr32(hw, GLINT_VECT2FUNC(v), reg); } + /* Map mailbox interrupt. We put an explicit 0 here to remind us that + * VF admin queue interrupts will go to VF MSI-X vector 0. + */ + wr32(hw, VPINT_MBX_CTL(abs_vf_id), VPINT_MBX_CTL_CAUSE_ENA_M | 0); /* set regardless of mapping mode */ wr32(hw, VPLAN_TXQ_MAPENA(vf->vf_id), VPLAN_TXQ_MAPENA_TX_ENA_M); @@ -627,6 +720,97 @@ ice_determine_res(struct ice_pf *pf, u16 avail_res, u16 max_res, u16 min_res) } /** + * ice_calc_vf_reg_idx - Calculate the VF's register index in the PF space + * @vf: VF to calculate the register index for + * @q_vector: a q_vector associated to the VF + */ +int ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector) +{ + struct ice_pf *pf; + + if (!vf || !q_vector) + return -EINVAL; + + pf = vf->pf; + + /* always add one to account for the OICR being the first MSIX */ + return pf->sriov_base_vector + pf->num_vf_msix * vf->vf_id + + q_vector->v_idx + 1; +} + +/** + * ice_get_max_valid_res_idx - Get the max valid resource index + * @res: pointer to the resource to find the max valid index for + * + * Start from the end of the ice_res_tracker and return right when we find the + * first res->list entry with the ICE_RES_VALID_BIT set. This function is only + * valid for SR-IOV because it is the only consumer that manipulates the + * res->end and this is always called when res->end is set to res->num_entries. + */ +static int ice_get_max_valid_res_idx(struct ice_res_tracker *res) +{ + int i; + + if (!res) + return -EINVAL; + + for (i = res->num_entries - 1; i >= 0; i--) + if (res->list[i] & ICE_RES_VALID_BIT) + return i; + + return 0; +} + +/** + * ice_sriov_set_msix_res - Set any used MSIX resources + * @pf: pointer to PF structure + * @num_msix_needed: number of MSIX vectors needed for all SR-IOV VFs + * + * This function allows SR-IOV resources to be taken from the end of the PF's + * allowed HW MSIX vectors so in many cases the irq_tracker will not + * be needed. In these cases we just set the pf->sriov_base_vector and return + * success. + * + * If SR-IOV needs to use any pf->irq_tracker entries it updates the + * irq_tracker->end based on the first entry needed for SR-IOV. This makes it + * so any calls to ice_get_res() using the irq_tracker will not try to use + * resources at or beyond the newly set value. + * + * Return 0 on success, and -EINVAL when there are not enough MSIX vectors in + * in the PF's space available for SR-IOV. + */ +static int ice_sriov_set_msix_res(struct ice_pf *pf, u16 num_msix_needed) +{ + int max_valid_res_idx = ice_get_max_valid_res_idx(pf->irq_tracker); + u16 pf_total_msix_vectors = + pf->hw.func_caps.common_cap.num_msix_vectors; + struct ice_res_tracker *res = pf->irq_tracker; + int sriov_base_vector; + + if (max_valid_res_idx < 0) + return max_valid_res_idx; + + sriov_base_vector = pf_total_msix_vectors - num_msix_needed; + + /* make sure we only grab irq_tracker entries from the list end and + * that we have enough available MSIX vectors + */ + if (sriov_base_vector <= max_valid_res_idx) + return -EINVAL; + + pf->sriov_base_vector = sriov_base_vector; + + /* dip into irq_tracker entries and update used resources */ + if (num_msix_needed > (pf_total_msix_vectors - res->num_entries)) { + pf->num_avail_sw_msix -= + res->num_entries - pf->sriov_base_vector; + res->end = pf->sriov_base_vector; + } + + return 0; +} + +/** * ice_check_avail_res - check if vectors and queues are available * @pf: pointer to the PF structure * @@ -636,11 +820,16 @@ ice_determine_res(struct ice_pf *pf, u16 avail_res, u16 max_res, u16 min_res) */ static int ice_check_avail_res(struct ice_pf *pf) { - u16 num_msix, num_txq, num_rxq; + int max_valid_res_idx = ice_get_max_valid_res_idx(pf->irq_tracker); + u16 num_msix, num_txq, num_rxq, num_avail_msix; - if (!pf->num_alloc_vfs) + if (!pf->num_alloc_vfs || max_valid_res_idx < 0) return -EINVAL; + /* add 1 to max_valid_res_idx to account for it being 0-based */ + num_avail_msix = pf->hw.func_caps.common_cap.num_msix_vectors - + (max_valid_res_idx + 1); + /* Grab from HW interrupts common pool * Note: By the time the user decides it needs more vectors in a VF * its already too late since one must decide this prior to creating the @@ -657,11 +846,11 @@ static int ice_check_avail_res(struct ice_pf *pf) * grab default interrupt vectors (5 as supported by AVF driver). */ if (pf->num_alloc_vfs <= 16) { - num_msix = ice_determine_res(pf, pf->num_avail_hw_msix, + num_msix = ice_determine_res(pf, num_avail_msix, ICE_MAX_INTR_PER_VF, ICE_MIN_INTR_PER_VF); } else if (pf->num_alloc_vfs <= ICE_MAX_VF_COUNT) { - num_msix = ice_determine_res(pf, pf->num_avail_hw_msix, + num_msix = ice_determine_res(pf, num_avail_msix, ICE_DFLT_INTR_PER_VF, ICE_MIN_INTR_PER_VF); } else { @@ -690,6 +879,9 @@ static int ice_check_avail_res(struct ice_pf *pf) if (!num_txq || !num_rxq) return -EIO; + if (ice_sriov_set_msix_res(pf, num_msix * pf->num_alloc_vfs)) + return -EINVAL; + /* since AVF driver works with only queue pairs which means, it expects * to have equal number of Rx and Tx queues, so take the minimum of * available Tx or Rx queues @@ -750,6 +942,47 @@ static void ice_cleanup_and_realloc_vf(struct ice_vf *vf) } /** + * ice_vf_set_vsi_promisc - set given VF VSI to given promiscuous mode(s) + * @vf: pointer to the VF info + * @vsi: the VSI being configured + * @promisc_m: mask of promiscuous config bits + * @rm_promisc: promisc flag request from the VF to remove or add filter + * + * This function configures VF VSI promiscuous mode, based on the VF requests, + * for Unicast, Multicast and VLAN + */ +static enum ice_status +ice_vf_set_vsi_promisc(struct ice_vf *vf, struct ice_vsi *vsi, u8 promisc_m, + bool rm_promisc) +{ + struct ice_pf *pf = vf->pf; + enum ice_status status = 0; + struct ice_hw *hw; + + hw = &pf->hw; + if (vf->num_vlan) { + status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m, + rm_promisc); + } else if (vf->port_vlan_id) { + if (rm_promisc) + status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m, + vf->port_vlan_id); + else + status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m, + vf->port_vlan_id); + } else { + if (rm_promisc) + status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m, + 0); + else + status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m, + 0); + } + + return status; +} + +/** * ice_reset_all_vfs - reset all allocated VFs in one go * @pf: pointer to the PF structure * @is_vflr: true if VFLR was issued, false if not @@ -764,6 +997,7 @@ static void ice_cleanup_and_realloc_vf(struct ice_vf *vf) bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) { struct ice_hw *hw = &pf->hw; + struct ice_vf *vf; int v, i; /* If we don't have any VFs, then there is nothing to reset */ @@ -778,12 +1012,17 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) for (v = 0; v < pf->num_alloc_vfs; v++) ice_trigger_vf_reset(&pf->vf[v], is_vflr); - /* Call Disable LAN Tx queue AQ call with VFR bit set and 0 - * queues to inform Firmware about VF reset. - */ - for (v = 0; v < pf->num_alloc_vfs; v++) - ice_dis_vsi_txq(pf->vsi[0]->port_info, 0, NULL, NULL, - ICE_VF_RESET, v, NULL); + for (v = 0; v < pf->num_alloc_vfs; v++) { + struct ice_vsi *vsi; + + vf = &pf->vf[v]; + vsi = pf->vsi[vf->lan_vsi_idx]; + if (test_bit(ICE_VF_STATE_ENA, vf->vf_states)) { + ice_vsi_stop_lan_tx_rings(vsi, ICE_VF_RESET, vf->vf_id); + ice_vsi_stop_rx_rings(vsi); + clear_bit(ICE_VF_STATE_ENA, vf->vf_states); + } + } /* HW requires some time to make sure it can flush the FIFO for a VF * when it resets it. Poll the VPGEN_VFRSTAT register for each VF in @@ -796,9 +1035,9 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) /* Check each VF in sequence */ while (v < pf->num_alloc_vfs) { - struct ice_vf *vf = &pf->vf[v]; u32 reg; + vf = &pf->vf[v]; reg = rd32(hw, VPGEN_VFRSTAT(vf->vf_id)); if (!(reg & VPGEN_VFRSTAT_VFRD_M)) break; @@ -818,8 +1057,22 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) usleep_range(10000, 20000); /* free VF resources to begin resetting the VSI state */ - for (v = 0; v < pf->num_alloc_vfs; v++) - ice_free_vf_res(&pf->vf[v]); + for (v = 0; v < pf->num_alloc_vfs; v++) { + vf = &pf->vf[v]; + + ice_free_vf_res(vf); + + /* Free VF queues as well, and reallocate later. + * If a given VF has different number of queues + * configured, the request for update will come + * via mailbox communication. + */ + vf->num_vf_qs = 0; + } + + if (ice_sriov_free_msix_res(pf)) + dev_err(&pf->pdev->dev, + "Failed to free MSIX resources used by SR-IOV\n"); if (ice_check_avail_res(pf)) { dev_err(&pf->pdev->dev, @@ -828,8 +1081,15 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) } /* Finish the reset on each VF */ - for (v = 0; v < pf->num_alloc_vfs; v++) - ice_cleanup_and_realloc_vf(&pf->vf[v]); + for (v = 0; v < pf->num_alloc_vfs; v++) { + vf = &pf->vf[v]; + + vf->num_vf_qs = pf->num_vf_qps; + dev_dbg(&pf->pdev->dev, + "VF-id %d has %d queues configured\n", + vf->vf_id, vf->num_vf_qs); + ice_cleanup_and_realloc_vf(vf); + } ice_flush(hw); clear_bit(__ICE_VF_DIS, pf->state); @@ -847,9 +1107,10 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) { struct ice_pf *pf = vf->pf; - struct ice_hw *hw = &pf->hw; struct ice_vsi *vsi; + struct ice_hw *hw; bool rsd = false; + u8 promisc_m; u32 reg; int i; @@ -871,10 +1132,11 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) /* Call Disable LAN Tx queue AQ call even when queues are not * enabled. This is needed for successful completiom of VFR */ - ice_dis_vsi_txq(vsi->port_info, 0, NULL, NULL, ICE_VF_RESET, - vf->vf_id, NULL); + ice_dis_vsi_txq(vsi->port_info, vsi->idx, 0, 0, NULL, NULL, + NULL, ICE_VF_RESET, vf->vf_id, NULL); } + hw = &pf->hw; /* poll VPGEN_VFRSTAT reg to make sure * that reset is complete */ @@ -900,6 +1162,21 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) usleep_range(10000, 20000); + /* disable promiscuous modes in case they were enabled + * ignore any error if disabling process failed + */ + if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) || + test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) { + if (vf->port_vlan_id || vf->num_vlan) + promisc_m = ICE_UCAST_VLAN_PROMISC_BITS; + else + promisc_m = ICE_UCAST_PROMISC_BITS; + + vsi = pf->vsi[vf->lan_vsi_idx]; + if (ice_vf_set_vsi_promisc(vf, vsi, promisc_m, true)) + dev_err(&pf->pdev->dev, "disabling promiscuous mode failed\n"); + } + /* free VF resources to begin resetting the VSI state */ ice_free_vf_res(vf); @@ -938,7 +1215,7 @@ void ice_vc_notify_reset(struct ice_pf *pf) pfe.event = VIRTCHNL_EVENT_RESET_IMPENDING; pfe.severity = PF_EVENT_SEVERITY_CERTAIN_DOOM; - ice_vc_vf_broadcast(pf, VIRTCHNL_OP_EVENT, ICE_SUCCESS, + ice_vc_vf_broadcast(pf, VIRTCHNL_OP_EVENT, VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe, sizeof(struct virtchnl_pf_event)); } @@ -961,8 +1238,9 @@ static void ice_vc_notify_vf_reset(struct ice_vf *vf) pfe.event = VIRTCHNL_EVENT_RESET_IMPENDING; pfe.severity = PF_EVENT_SEVERITY_CERTAIN_DOOM; - ice_aq_send_msg_to_vf(&vf->pf->hw, vf->vf_id, VIRTCHNL_OP_EVENT, 0, - (u8 *)&pfe, sizeof(pfe), NULL); + ice_aq_send_msg_to_vf(&vf->pf->hw, vf->vf_id, VIRTCHNL_OP_EVENT, + VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe, sizeof(pfe), + NULL); } /** @@ -977,7 +1255,7 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) int i, ret; /* Disable global interrupt 0 so we don't try to handle the VFLR. */ - wr32(hw, GLINT_DYN_CTL(pf->hw_oicr_idx), + wr32(hw, GLINT_DYN_CTL(pf->oicr_idx), ICE_ITR_NONE << GLINT_DYN_CTL_ITR_INDX_S); ice_flush(hw); @@ -992,7 +1270,7 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) GFP_KERNEL); if (!vfs) { ret = -ENOMEM; - goto err_unroll_sriov; + goto err_pci_disable_sriov; } pf->vf = vfs; @@ -1012,12 +1290,19 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) pf->num_alloc_vfs = num_alloc_vfs; /* VF resources get allocated during reset */ - if (!ice_reset_all_vfs(pf, false)) + if (!ice_reset_all_vfs(pf, true)) { + ret = -EIO; goto err_unroll_sriov; + } goto err_unroll_intr; err_unroll_sriov: + pf->vf = NULL; + devm_kfree(&pf->pdev->dev, vfs); + vfs = NULL; + pf->num_alloc_vfs = 0; +err_pci_disable_sriov: pci_disable_sriov(pf->pdev); err_unroll_intr: /* rearm interrupts here */ @@ -1026,8 +1311,8 @@ err_unroll_intr: } /** - * ice_pf_state_is_nominal - checks the pf for nominal state - * @pf: pointer to pf to check + * ice_pf_state_is_nominal - checks the PF for nominal state + * @pf: pointer to PF to check * * Check the PF's state for a collection of bits that would indicate * the PF is in a state that would inhibit normal operation for @@ -1131,21 +1416,10 @@ void ice_process_vflr_event(struct ice_pf *pf) int vf_id; u32 reg; - if (!test_bit(__ICE_VFLR_EVENT_PENDING, pf->state) || + if (!test_and_clear_bit(__ICE_VFLR_EVENT_PENDING, pf->state) || !pf->num_alloc_vfs) return; - /* Re-enable the VFLR interrupt cause here, before looking for which - * VF got reset. Otherwise, if another VF gets a reset while the - * first one is being processed, that interrupt will be lost, and - * that VF will be stuck in reset forever. - */ - reg = rd32(hw, PFINT_OICR_ENA); - reg |= PFINT_OICR_VFLR_M; - wr32(hw, PFINT_OICR_ENA, reg); - ice_flush(hw); - - clear_bit(__ICE_VFLR_EVENT_PENDING, pf->state); for (vf_id = 0; vf_id < pf->num_alloc_vfs; vf_id++) { struct ice_vf *vf = &pf->vf[vf_id]; u32 reg_idx, bit_idx; @@ -1182,8 +1456,9 @@ static void ice_vc_dis_vf(struct ice_vf *vf) * * send msg to VF */ -static int ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode, - enum ice_status v_retval, u8 *msg, u16 msglen) +static int +ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode, + enum virtchnl_status_code v_retval, u8 *msg, u16 msglen) { enum ice_status aq_ret; struct ice_pf *pf; @@ -1243,8 +1518,8 @@ static int ice_vc_get_ver_msg(struct ice_vf *vf, u8 *msg) if (VF_IS_V10(&vf->vf_ver)) info.minor = VIRTCHNL_VERSION_MINOR_NO_VF_CAPS; - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_VERSION, ICE_SUCCESS, - (u8 *)&info, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_VERSION, + VIRTCHNL_STATUS_SUCCESS, (u8 *)&info, sizeof(struct virtchnl_version_info)); } @@ -1257,15 +1532,15 @@ static int ice_vc_get_ver_msg(struct ice_vf *vf, u8 *msg) */ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_vf_resource *vfres = NULL; - enum ice_status aq_ret = 0; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; int len = 0; int ret; if (!test_bit(ICE_VF_STATE_INIT, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto err; } @@ -1273,7 +1548,7 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg) vfres = devm_kzalloc(&pf->pdev->dev, len, GFP_KERNEL); if (!vfres) { - aq_ret = ICE_ERR_NO_MEMORY; + v_ret = VIRTCHNL_STATUS_ERR_NO_MEMORY; len = 0; goto err; } @@ -1286,6 +1561,11 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg) vfres->vf_cap_flags = VIRTCHNL_VF_OFFLOAD_L2; vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto err; + } + if (!vsi->info.pvid) vfres->vf_cap_flags |= VIRTCHNL_VF_OFFLOAD_VLAN; @@ -1336,7 +1616,7 @@ static int ice_vc_get_vf_res_msg(struct ice_vf *vf, u8 *msg) err: /* send the response back to the VF */ - ret = ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_VF_RESOURCES, aq_ret, + ret = ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_VF_RESOURCES, v_ret, (u8 *)vfres, len); devm_kfree(&pf->pdev->dev, vfres); @@ -1359,16 +1639,16 @@ static void ice_vc_reset_vf_msg(struct ice_vf *vf) /** * ice_find_vsi_from_id - * @pf: the pf structure to search for the VSI - * @id: id of the VSI it is searching for + * @pf: the PF structure to search for the VSI + * @id: ID of the VSI it is searching for * - * searches for the VSI with the given id + * searches for the VSI with the given ID */ static struct ice_vsi *ice_find_vsi_from_id(struct ice_pf *pf, u16 id) { int i; - for (i = 0; i < pf->num_alloc_vsi; i++) + ice_for_each_vsi(pf, i) if (pf->vsi[i] && pf->vsi[i]->vsi_num == id) return pf->vsi[i]; @@ -1378,9 +1658,9 @@ static struct ice_vsi *ice_find_vsi_from_id(struct ice_pf *pf, u16 id) /** * ice_vc_isvalid_vsi_id * @vf: pointer to the VF info - * @vsi_id: VF relative VSI id + * @vsi_id: VF relative VSI ID * - * check for the valid VSI id + * check for the valid VSI ID */ static bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id) { @@ -1395,10 +1675,10 @@ static bool ice_vc_isvalid_vsi_id(struct ice_vf *vf, u16 vsi_id) /** * ice_vc_isvalid_q_id * @vf: pointer to the VF info - * @vsi_id: VSI id - * @qid: VSI relative queue id + * @vsi_id: VSI ID + * @qid: VSI relative queue ID * - * check for the valid queue id + * check for the valid queue ID */ static bool ice_vc_isvalid_q_id(struct ice_vf *vf, u16 vsi_id, u8 qid) { @@ -1416,42 +1696,42 @@ static bool ice_vc_isvalid_q_id(struct ice_vf *vf, u16 vsi_id, u8 qid) */ static int ice_vc_config_rss_key(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_rss_key *vrk = (struct virtchnl_rss_key *)msg; + struct ice_pf *pf = vf->pf; struct ice_vsi *vsi = NULL; - enum ice_status aq_ret; - int ret; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, vrk->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, vrk->vsi_id); + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (vrk->key_len != ICE_VSIQF_HKEY_ARRAY_SIZE) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - ret = ice_set_rss(vsi, vrk->key, NULL, 0); - aq_ret = ret ? ICE_ERR_PARAM : ICE_SUCCESS; + if (ice_set_rss(vsi, vrk->key, NULL, 0)) + v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; error_param: - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_KEY, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_KEY, v_ret, NULL, 0); } @@ -1465,40 +1745,40 @@ error_param: static int ice_vc_config_rss_lut(struct ice_vf *vf, u8 *msg) { struct virtchnl_rss_lut *vrl = (struct virtchnl_rss_lut *)msg; + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; + struct ice_pf *pf = vf->pf; struct ice_vsi *vsi = NULL; - enum ice_status aq_ret; - int ret; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, vrl->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, vrl->vsi_id); + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (vrl->lut_entries != ICE_VSIQF_HLUT_ARRAY_SIZE) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - ret = ice_set_rss(vsi, NULL, vrl->lut, ICE_VSIQF_HLUT_ARRAY_SIZE); - aq_ret = ret ? ICE_ERR_PARAM : ICE_SUCCESS; + if (ice_set_rss(vsi, NULL, vrl->lut, ICE_VSIQF_HLUT_ARRAY_SIZE)) + v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; error_param: - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_LUT, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_RSS_LUT, v_ret, NULL, 0); } @@ -1511,25 +1791,26 @@ error_param: */ static int ice_vc_get_stats_msg(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_queue_select *vqs = (struct virtchnl_queue_select *)msg; - enum ice_status aq_ret = 0; + struct ice_pf *pf = vf->pf; struct ice_eth_stats stats; struct ice_vsi *vsi; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, vqs->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, vqs->vsi_id); + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1540,7 +1821,7 @@ static int ice_vc_get_stats_msg(struct ice_vf *vf, u8 *msg) error_param: /* send the response to the VF */ - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_STATS, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_GET_STATS, v_ret, (u8 *)&stats, sizeof(stats)); } @@ -1553,29 +1834,30 @@ error_param: */ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_queue_select *vqs = (struct virtchnl_queue_select *)msg; - enum ice_status aq_ret = 0; + struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, vqs->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!vqs->rx_queues && !vqs->tx_queues) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, vqs->vsi_id); + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1584,15 +1866,15 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) * programmed using ice_vsi_cfg_txqs */ if (ice_vsi_start_rx_rings(vsi)) - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; /* Set flag to indicate that queues are enabled */ - if (!aq_ret) + if (v_ret == VIRTCHNL_STATUS_SUCCESS) set_bit(ICE_VF_STATE_ENA, vf->vf_states); error_param: /* send the response to the VF */ - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_QUEUES, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_QUEUES, v_ret, NULL, 0); } @@ -1606,30 +1888,31 @@ error_param: */ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_queue_select *vqs = (struct virtchnl_queue_select *)msg; - enum ice_status aq_ret = 0; + struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) && !test_bit(ICE_VF_STATE_ENA, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, vqs->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!vqs->rx_queues && !vqs->tx_queues) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, vqs->vsi_id); + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1637,23 +1920,23 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) dev_err(&vsi->back->pdev->dev, "Failed to stop tx rings on VSI %d\n", vsi->vsi_num); - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; } if (ice_vsi_stop_rx_rings(vsi)) { dev_err(&vsi->back->pdev->dev, "Failed to stop rx rings on VSI %d\n", vsi->vsi_num); - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; } /* Clear enabled queues flag */ - if (!aq_ret) + if (v_ret == VIRTCHNL_STATUS_SUCCESS) clear_bit(ICE_VF_STATE_ENA, vf->vf_states); error_param: /* send the response to the VF */ - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_QUEUES, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_QUEUES, v_ret, NULL, 0); } @@ -1666,74 +1949,97 @@ error_param: */ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) { - struct virtchnl_irq_map_info *irqmap_info = - (struct virtchnl_irq_map_info *)msg; + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; + struct virtchnl_irq_map_info *irqmap_info; u16 vsi_id, vsi_q_id, vector_id; struct virtchnl_vector_map *map; - struct ice_vsi *vsi = NULL; struct ice_pf *pf = vf->pf; - enum ice_status aq_ret = 0; + u16 num_q_vectors_mapped; + struct ice_vsi *vsi; unsigned long qmap; int i; - if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + irqmap_info = (struct virtchnl_irq_map_info *)msg; + num_q_vectors_mapped = irqmap_info->num_vectors; + + vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - for (i = 0; i < irqmap_info->num_vectors; i++) { + /* Check to make sure number of VF vectors mapped is not greater than + * number of VF vectors originally allocated, and check that + * there is actually at least a single VF queue vector mapped + */ + if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) || + pf->num_vf_msix < num_q_vectors_mapped || + !irqmap_info->num_vectors) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + for (i = 0; i < num_q_vectors_mapped; i++) { + struct ice_q_vector *q_vector; + map = &irqmap_info->vecmap[i]; vector_id = map->vector_id; vsi_id = map->vsi_id; /* validate msg params */ if (!(vector_id < pf->hw.func_caps.common_cap - .num_msix_vectors) || !ice_vc_isvalid_vsi_id(vf, vsi_id)) { - aq_ret = ICE_ERR_PARAM; + .num_msix_vectors) || !ice_vc_isvalid_vsi_id(vf, vsi_id) || + (!vector_id && (map->rxq_map || map->txq_map))) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, vsi_id); - if (!vsi) { - aq_ret = ICE_ERR_PARAM; + /* No need to map VF miscellaneous or rogue vector */ + if (!vector_id) + continue; + + /* Subtract non queue vector from vector_id passed by VF + * to get actual number of VSI queue vector array index + */ + q_vector = vsi->q_vectors[vector_id - ICE_NONQ_VECS_VF]; + if (!q_vector) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } /* lookout for the invalid queue index */ qmap = map->rxq_map; + q_vector->num_ring_rx = 0; for_each_set_bit(vsi_q_id, &qmap, ICE_MAX_BASE_QS_PER_VF) { - struct ice_q_vector *q_vector; - if (!ice_vc_isvalid_q_id(vf, vsi_id, vsi_q_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - q_vector = vsi->q_vectors[i]; q_vector->num_ring_rx++; q_vector->rx.itr_idx = map->rxitr_idx; vsi->rx_rings[vsi_q_id]->q_vector = q_vector; + ice_cfg_rxq_interrupt(vsi, vsi_q_id, vector_id, + q_vector->rx.itr_idx); } qmap = map->txq_map; + q_vector->num_ring_tx = 0; for_each_set_bit(vsi_q_id, &qmap, ICE_MAX_BASE_QS_PER_VF) { - struct ice_q_vector *q_vector; - if (!ice_vc_isvalid_q_id(vf, vsi_id, vsi_q_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - q_vector = vsi->q_vectors[i]; q_vector->num_ring_tx++; q_vector->tx.itr_idx = map->txitr_idx; vsi->tx_rings[vsi_q_id]->q_vector = q_vector; + ice_cfg_txq_interrupt(vsi, vsi_q_id, vector_id, + q_vector->tx.itr_idx); } } - if (vsi) - ice_vsi_cfg_msix(vsi); error_param: /* send the response to the VF */ - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_IRQ_MAP, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_IRQ_MAP, v_ret, NULL, 0); } @@ -1746,26 +2052,33 @@ error_param: */ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_vsi_queue_config_info *qci = (struct virtchnl_vsi_queue_config_info *)msg; struct virtchnl_queue_pair_info *qpi; - enum ice_status aq_ret = 0; + struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; int i; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, qci->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - vsi = ice_find_vsi_from_id(vf->pf, qci->vsi_id); - if (!vsi) { - aq_ret = ICE_ERR_PARAM; + vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) + goto error_param; + + if (qci->num_queue_pairs > ICE_MAX_BASE_QS_PER_VF) { + dev_err(&pf->pdev->dev, + "VF-%d requesting more than supported number of queues: %d\n", + vf->vf_id, qci->num_queue_pairs); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1775,7 +2088,7 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) qpi->rxq.vsi_id != qci->vsi_id || qpi->rxq.queue_id != qpi->txq.queue_id || !ice_vc_isvalid_q_id(vf, qci->vsi_id, qpi->txq.queue_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } /* copy Tx queue info from VF into VSI */ @@ -1785,13 +2098,13 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) vsi->rx_rings[i]->dma = qpi->rxq.dma_ring_addr; vsi->rx_rings[i]->count = qpi->rxq.ring_len; if (qpi->rxq.databuffer_size > ((16 * 1024) - 128)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } vsi->rx_buf_len = qpi->rxq.databuffer_size; if (qpi->rxq.max_pkt_size >= (16 * 1024) || qpi->rxq.max_pkt_size < 64) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } vsi->max_frame = qpi->rxq.max_pkt_size; @@ -1802,15 +2115,16 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) */ vsi->num_txq = qci->num_queue_pairs; vsi->num_rxq = qci->num_queue_pairs; + /* All queues of VF VSI are in TC 0 */ + vsi->tc_cfg.tc_info[0].qcount_tx = qci->num_queue_pairs; + vsi->tc_cfg.tc_info[0].qcount_rx = qci->num_queue_pairs; - if (!ice_vsi_cfg_lan_txqs(vsi) && !ice_vsi_cfg_rxqs(vsi)) - aq_ret = 0; - else - aq_ret = ICE_ERR_PARAM; + if (ice_vsi_cfg_lan_txqs(vsi) || ice_vsi_cfg_rxqs(vsi)) + v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; error_param: /* send the response to the VF */ - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_VSI_QUEUES, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_VSI_QUEUES, v_ret, NULL, 0); } @@ -1845,18 +2159,18 @@ static bool ice_can_vf_change_mac(struct ice_vf *vf) * ice_vc_handle_mac_addr_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer - * @set: true if mac filters are being set, false otherwise + * @set: true if MAC filters are being set, false otherwise * * add guest MAC address filter */ static int ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_ether_addr_list *al = (struct virtchnl_ether_addr_list *)msg; struct ice_pf *pf = vf->pf; enum virtchnl_ops vc_op; - enum ice_status ret; LIST_HEAD(mac_list); struct ice_vsi *vsi; int mac_count = 0; @@ -1869,19 +2183,27 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) || !ice_vc_isvalid_vsi_id(vf, al->vsi_id)) { - ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto handle_mac_exit; } if (set && !ice_is_vf_trusted(vf) && (vf->num_mac + al->num_elements) > ICE_MAX_MACADDR_PER_VF) { dev_err(&pf->pdev->dev, - "Can't add more MAC addresses, because VF is not trusted, switch the VF to trusted mode in order to add more functionalities\n"); - ret = ICE_ERR_PARAM; + "Can't add more MAC addresses, because VF-%d is not trusted, switch the VF to trusted mode in order to add more functionalities\n", + vf->vf_id); + /* There is no need to let VF know about not being trusted + * to add more MAC addr, so we can just return success message. + */ + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto handle_mac_exit; } vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto handle_mac_exit; + } for (i = 0; i < al->num_elements; i++) { u8 *maddr = al->list[i].addr; @@ -1893,40 +2215,39 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) * already added. Just continue. */ dev_info(&pf->pdev->dev, - "mac %pM already set for VF %d\n", + "MAC %pM already set for VF %d\n", maddr, vf->vf_id); continue; } else { - /* VF can't remove dflt_lan_addr/bcast mac */ + /* VF can't remove dflt_lan_addr/bcast MAC */ dev_err(&pf->pdev->dev, - "can't remove mac %pM for VF %d\n", + "VF can't remove default MAC address or MAC %pM programmed by PF for VF %d\n", maddr, vf->vf_id); - ret = ICE_ERR_PARAM; - goto handle_mac_exit; + continue; } } /* check for the invalid cases and bail if necessary */ if (is_zero_ether_addr(maddr)) { dev_err(&pf->pdev->dev, - "invalid mac %pM provided for VF %d\n", + "invalid MAC %pM provided for VF %d\n", maddr, vf->vf_id); - ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto handle_mac_exit; } if (is_unicast_ether_addr(maddr) && !ice_can_vf_change_mac(vf)) { dev_err(&pf->pdev->dev, - "can't change unicast mac for untrusted VF %d\n", + "can't change unicast MAC for untrusted VF %d\n", vf->vf_id); - ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto handle_mac_exit; } - /* get here if maddr is multicast or if VF can change mac */ + /* get here if maddr is multicast or if VF can change MAC */ if (ice_add_mac_to_list(vsi, &mac_list, al->list[i].addr)) { - ret = ICE_ERR_NO_MEMORY; + v_ret = VIRTCHNL_STATUS_ERR_NO_MEMORY; goto handle_mac_exit; } mac_count++; @@ -1934,14 +2255,14 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) /* program the updated filter list */ if (set) - ret = ice_add_mac(&pf->hw, &mac_list); + v_ret = ice_err_to_virt_err(ice_add_mac(&pf->hw, &mac_list)); else - ret = ice_remove_mac(&pf->hw, &mac_list); + v_ret = ice_err_to_virt_err(ice_remove_mac(&pf->hw, &mac_list)); - if (ret) { + if (v_ret) { dev_err(&pf->pdev->dev, - "can't update mac filters for VF %d, error %d\n", - vf->vf_id, ret); + "can't update MAC filters for VF %d, error %d\n", + vf->vf_id, v_ret); } else { if (set) vf->num_mac += mac_count; @@ -1952,7 +2273,7 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) handle_mac_exit: ice_free_fltr_list(&pf->pdev->dev, &mac_list); /* send the response to the VF */ - return ice_vc_send_msg_to_vf(vf, vc_op, ret, NULL, 0); + return ice_vc_send_msg_to_vf(vf, vc_op, v_ret, NULL, 0); } /** @@ -1987,39 +2308,42 @@ static int ice_vc_del_mac_addr_msg(struct ice_vf *vf, u8 *msg) * VFs get a default number of queues but can use this message to request a * different number. If the request is successful, PF will reset the VF and * return 0. If unsuccessful, PF will send message informing VF of number of - * available queue pairs via virtchnl message response to vf. + * available queue pairs via virtchnl message response to VF. */ static int ice_vc_request_qs_msg(struct ice_vf *vf, u8 *msg) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_vf_res_request *vfres = (struct virtchnl_vf_res_request *)msg; int req_queues = vfres->num_queue_pairs; - enum ice_status aq_ret = 0; struct ice_pf *pf = vf->pf; + int max_allowed_vf_queues; int tx_rx_queue_left; int cur_queues; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - cur_queues = pf->num_vf_qps; + cur_queues = vf->num_vf_qs; tx_rx_queue_left = min_t(int, pf->q_left_tx, pf->q_left_rx); + max_allowed_vf_queues = tx_rx_queue_left + cur_queues; if (req_queues <= 0) { dev_err(&pf->pdev->dev, "VF %d tried to request %d queues. Ignoring.\n", vf->vf_id, req_queues); - } else if (req_queues > ICE_MAX_QS_PER_VF) { + } else if (req_queues > ICE_MAX_BASE_QS_PER_VF) { dev_err(&pf->pdev->dev, "VF %d tried to request more than %d queues.\n", - vf->vf_id, ICE_MAX_QS_PER_VF); - vfres->num_queue_pairs = ICE_MAX_QS_PER_VF; + vf->vf_id, ICE_MAX_BASE_QS_PER_VF); + vfres->num_queue_pairs = ICE_MAX_BASE_QS_PER_VF; } else if (req_queues - cur_queues > tx_rx_queue_left) { dev_warn(&pf->pdev->dev, "VF %d requested %d more queues, but only %d left.\n", vf->vf_id, req_queues - cur_queues, tx_rx_queue_left); - vfres->num_queue_pairs = tx_rx_queue_left + cur_queues; + vfres->num_queue_pairs = min_t(int, max_allowed_vf_queues, + ICE_MAX_BASE_QS_PER_VF); } else { /* request is successful, then reset VF */ vf->num_req_qs = req_queues; @@ -2033,18 +2357,18 @@ static int ice_vc_request_qs_msg(struct ice_vf *vf, u8 *msg) error_param: /* send the response to the VF */ return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_REQUEST_QUEUES, - aq_ret, (u8 *)vfres, sizeof(*vfres)); + v_ret, (u8 *)vfres, sizeof(*vfres)); } /** * ice_set_vf_port_vlan * @netdev: network interface device structure * @vf_id: VF identifier - * @vlan_id: VLAN id being set + * @vlan_id: VLAN ID being set * @qos: priority setting * @vlan_proto: VLAN protocol * - * program VF Port VLAN id and/or qos + * program VF Port VLAN ID and/or QoS */ int ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, @@ -2087,17 +2411,18 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, return ret; } - /* If pvid, then remove all filters on the old VLAN */ + /* If PVID, then remove all filters on the old VLAN */ if (vsi->info.pvid) ice_vsi_kill_vlan(vsi, (le16_to_cpu(vsi->info.pvid) & VLAN_VID_MASK)); if (vlan_id || qos) { - ret = ice_vsi_set_pvid(vsi, vlanprio); + ret = ice_vsi_manage_pvid(vsi, vlanprio, true); if (ret) goto error_set_pvid; } else { - ice_vsi_kill_pvid(vsi); + ice_vsi_manage_pvid(vsi, 0, false); + vsi->info.pvid = 0; } if (vlan_id) { @@ -2125,52 +2450,60 @@ error_set_pvid: * @msg: pointer to the msg buffer * @add_v: Add VLAN if true, otherwise delete VLAN * - * Process virtchnl op to add or remove programmed guest VLAN id + * Process virtchnl op to add or remove programmed guest VLAN ID */ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) { + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct virtchnl_vlan_filter_list *vfl = (struct virtchnl_vlan_filter_list *)msg; - enum ice_status aq_ret = 0; struct ice_pf *pf = vf->pf; + bool vlan_promisc = false; struct ice_vsi *vsi; + struct ice_hw *hw; + int status = 0; + u8 promisc_m; int i; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (!ice_vc_isvalid_vsi_id(vf, vfl->vsi_id)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (add_v && !ice_is_vf_trusted(vf) && vf->num_vlan >= ICE_MAX_VLAN_PER_VF) { dev_info(&pf->pdev->dev, - "VF is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n"); - aq_ret = ICE_ERR_PARAM; + "VF-%d is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n", + vf->vf_id); + /* There is no need to let VF know about being not trusted, + * so we can just return success message here + */ goto error_param; } for (i = 0; i < vfl->num_elements; i++) { if (vfl->vlan_id[i] > ICE_MAX_VLANID) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; dev_err(&pf->pdev->dev, "invalid VF VLAN id %d\n", vfl->vlan_id[i]); goto error_param; } } - vsi = ice_find_vsi_from_id(vf->pf, vfl->vsi_id); + hw = &pf->hw; + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } if (vsi->info.pvid) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -2178,38 +2511,94 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) dev_err(&pf->pdev->dev, "%sable VLAN stripping failed for VSI %i\n", add_v ? "en" : "dis", vsi->vsi_num); - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } + if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) || + test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) + vlan_promisc = true; + if (add_v) { for (i = 0; i < vfl->num_elements; i++) { u16 vid = vfl->vlan_id[i]; - if (!ice_vsi_add_vlan(vsi, vid)) { - vf->num_vlan++; + if (!ice_is_vf_trusted(vf) && + vf->num_vlan >= ICE_MAX_VLAN_PER_VF) { + dev_info(&pf->pdev->dev, + "VF-%d is not trusted, switch the VF to trusted mode, in order to add more VLAN addresses\n", + vf->vf_id); + /* There is no need to let VF know about being + * not trusted, so we can just return success + * message here as well. + */ + goto error_param; + } + + if (ice_vsi_add_vlan(vsi, vid)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } - /* Enable VLAN pruning when VLAN 0 is added */ - if (unlikely(!vid)) - if (ice_cfg_vlan_pruning(vsi, true)) - aq_ret = ICE_ERR_PARAM; + vf->num_vlan++; + /* Enable VLAN pruning when VLAN is added */ + if (!vlan_promisc) { + status = ice_cfg_vlan_pruning(vsi, true, false); + if (status) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + dev_err(&pf->pdev->dev, + "Enable VLAN pruning on VLAN ID: %d failed error-%d\n", + vid, status); + goto error_param; + } } else { - aq_ret = ICE_ERR_PARAM; + /* Enable Ucast/Mcast VLAN promiscuous mode */ + promisc_m = ICE_PROMISC_VLAN_TX | + ICE_PROMISC_VLAN_RX; + + status = ice_set_vsi_promisc(hw, vsi->idx, + promisc_m, vid); + if (status) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + dev_err(&pf->pdev->dev, + "Enable Unicast/multicast promiscuous mode on VLAN ID:%d failed error-%d\n", + vid, status); + } } } } else { - for (i = 0; i < vfl->num_elements; i++) { + /* In case of non_trusted VF, number of VLAN elements passed + * to PF for removal might be greater than number of VLANs + * filter programmed for that VF - So, use actual number of + * VLANS added earlier with add VLAN opcode. In order to avoid + * removing VLAN that doesn't exist, which result to sending + * erroneous failed message back to the VF + */ + int num_vf_vlan; + + num_vf_vlan = vf->num_vlan; + for (i = 0; i < vfl->num_elements && i < num_vf_vlan; i++) { u16 vid = vfl->vlan_id[i]; /* Make sure ice_vsi_kill_vlan is successful before * updating VLAN information */ - if (!ice_vsi_kill_vlan(vsi, vid)) { - vf->num_vlan--; + if (ice_vsi_kill_vlan(vsi, vid)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + vf->num_vlan--; + /* Disable VLAN pruning when removing VLAN */ + ice_cfg_vlan_pruning(vsi, false, false); - /* Disable VLAN pruning when removing VLAN 0 */ - if (unlikely(!vid)) - ice_cfg_vlan_pruning(vsi, false); + /* Disable Unicast/Multicast VLAN promiscuous mode */ + if (vlan_promisc) { + promisc_m = ICE_PROMISC_VLAN_TX | + ICE_PROMISC_VLAN_RX; + + ice_clear_vsi_promisc(hw, vsi->idx, + promisc_m, vid); } } } @@ -2217,10 +2606,10 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) error_param: /* send the response to the VF */ if (add_v) - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ADD_VLAN, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ADD_VLAN, v_ret, NULL, 0); else - return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DEL_VLAN, aq_ret, + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DEL_VLAN, v_ret, NULL, 0); } @@ -2229,7 +2618,7 @@ error_param: * @vf: pointer to the VF info * @msg: pointer to the msg buffer * - * Add and program guest VLAN id + * Add and program guest VLAN ID */ static int ice_vc_add_vlan_msg(struct ice_vf *vf, u8 *msg) { @@ -2241,7 +2630,7 @@ static int ice_vc_add_vlan_msg(struct ice_vf *vf, u8 *msg) * @vf: pointer to the VF info * @msg: pointer to the msg buffer * - * remove programmed guest VLAN id + * remove programmed guest VLAN ID */ static int ice_vc_remove_vlan_msg(struct ice_vf *vf, u8 *msg) { @@ -2256,22 +2645,22 @@ static int ice_vc_remove_vlan_msg(struct ice_vf *vf, u8 *msg) */ static int ice_vc_ena_vlan_stripping(struct ice_vf *vf) { - enum ice_status aq_ret = 0; + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } vsi = pf->vsi[vf->lan_vsi_idx]; if (ice_vsi_manage_vlan_stripping(vsi, true)) - aq_ret = ICE_ERR_AQ_ERROR; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; error_param: return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ENABLE_VLAN_STRIPPING, - aq_ret, NULL, 0); + v_ret, NULL, 0); } /** @@ -2282,22 +2671,27 @@ error_param: */ static int ice_vc_dis_vlan_stripping(struct ice_vf *vf) { - enum ice_status aq_ret = 0; + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { - aq_ret = ICE_ERR_PARAM; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + if (ice_vsi_manage_vlan_stripping(vsi, false)) - aq_ret = ICE_ERR_AQ_ERROR; + v_ret = VIRTCHNL_STATUS_ERR_PARAM; error_param: return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_DISABLE_VLAN_STRIPPING, - aq_ret, NULL, 0); + v_ret, NULL, 0); } /** @@ -2333,7 +2727,7 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event) /* Perform basic checks on the msg */ err = virtchnl_vc_validate_vf_msg(&vf->vf_ver, v_opcode, msg, msglen); if (err) { - if (err == VIRTCHNL_ERR_PARAM) + if (err == VIRTCHNL_STATUS_ERR_PARAM) err = -EPERM; else err = -EINVAL; @@ -2355,7 +2749,8 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event) error_handler: if (err) { - ice_vc_send_msg_to_vf(vf, v_opcode, ICE_ERR_PARAM, NULL, 0); + ice_vc_send_msg_to_vf(vf, v_opcode, VIRTCHNL_STATUS_ERR_PARAM, + NULL, 0); dev_err(&pf->pdev->dev, "Invalid message from VF %d, opcode %d, len %d, error %d\n", vf_id, v_opcode, msglen, err); return; @@ -2418,7 +2813,8 @@ error_handler: default: dev_err(&pf->pdev->dev, "Unsupported opcode %d from VF %d\n", v_opcode, vf_id); - err = ice_vc_send_msg_to_vf(vf, v_opcode, ICE_ERR_NOT_IMPL, + err = ice_vc_send_msg_to_vf(vf, v_opcode, + VIRTCHNL_STATUS_ERR_NOT_SUPPORTED, NULL, 0); break; } @@ -2427,7 +2823,7 @@ error_handler: * as it is busy with pending work. */ dev_info(&pf->pdev->dev, - "PF failed to honor VF %d, opcode %d\n, error %d\n", + "PF failed to honor VF %d, opcode %d, error %d\n", vf_id, v_opcode, err); } } @@ -2440,8 +2836,8 @@ error_handler: * * return VF configuration */ -int ice_get_vf_cfg(struct net_device *netdev, int vf_id, - struct ifla_vf_info *ivi) +int +ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; @@ -2550,9 +2946,9 @@ out: * ice_set_vf_mac * @netdev: network interface device structure * @vf_id: VF identifier - * @mac: mac address + * @mac: MAC address * - * program VF mac address + * program VF MAC address */ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) { @@ -2579,7 +2975,7 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) return -EINVAL; } - /* copy mac into dflt_lan_addr and trigger a VF reset. The reset + /* copy MAC into dflt_lan_addr and trigger a VF reset. The reset * flow will use the updated dflt_lan_addr and add a MAC filter * using ice_add_mac. Also set pf_set_mac to indicate that the PF has * set the MAC address for this VF. @@ -2587,7 +2983,7 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) ether_addr_copy(vf->dflt_lan_addr.addr, mac); vf->pf_set_mac = true; netdev_info(netdev, - "mac on VF %d set to %pM\n. VF driver will be reinitialized\n", + "MAC on VF %d set to %pM. VF driver will be reinitialized\n", vf_id, mac); ice_vc_dis_vf(vf); @@ -2690,7 +3086,8 @@ int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state) ice_set_pfe_link(vf, &pfe, ls->link_speed, vf->link_up); /* Notify the VF of its new link state */ - ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, 0, (u8 *)&pfe, + ice_aq_send_msg_to_vf(hw, vf->vf_id, VIRTCHNL_OP_EVENT, + VIRTCHNL_STATUS_SUCCESS, (u8 *)&pfe, sizeof(pfe), NULL); return 0; diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h index 01470a8ee03a..c3ca522c245a 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h @@ -48,37 +48,43 @@ enum ice_virtchnl_cap { struct ice_vf { struct ice_pf *pf; - s16 vf_id; /* VF id in the PF space */ - u32 driver_caps; /* reported by VF driver */ + s16 vf_id; /* VF ID in the PF space */ + u16 lan_vsi_idx; /* index into PF struct */ int first_vector_idx; /* first vector index of this VF */ - struct ice_sw *vf_sw_id; /* switch id the VF VSIs connect to */ + struct ice_sw *vf_sw_id; /* switch ID the VF VSIs connect to */ struct virtchnl_version_info vf_ver; + u32 driver_caps; /* reported by VF driver */ struct virtchnl_ether_addr dflt_lan_addr; u16 port_vlan_id; - u8 pf_set_mac; /* VF MAC address set by VMM admin */ - u8 trusted; - u16 lan_vsi_idx; /* index into PF struct */ + u8 pf_set_mac:1; /* VF MAC address set by VMM admin */ + u8 trusted:1; + u8 spoofchk:1; + u8 link_forced:1; + u8 link_up:1; /* only valid if VF link is forced */ + /* VSI indices - actual VSI pointers are maintained in the PF structure + * When assigned, these will be non-zero, because VSI 0 is always + * the main LAN VSI for the PF. + */ u16 lan_vsi_num; /* ID as used by firmware */ - u64 num_mdd_events; /* number of mdd events detected */ + unsigned int tx_rate; /* Tx bandwidth limit in Mbps */ + DECLARE_BITMAP(vf_states, ICE_VF_STATES_NBITS); /* VF runtime states */ + + u64 num_mdd_events; /* number of MDD events detected */ u64 num_inval_msgs; /* number of continuous invalid msgs */ u64 num_valid_msgs; /* number of valid msgs detected */ - unsigned long vf_caps; /* vf's adv. capabilities */ - DECLARE_BITMAP(vf_states, ICE_VF_STATES_NBITS); /* VF runtime states */ - unsigned int tx_rate; /* Tx bandwidth limit in Mbps */ - u8 link_forced; - u8 link_up; /* only valid if VF link is forced */ - u8 spoofchk; + unsigned long vf_caps; /* VF's adv. capabilities */ + u8 num_req_qs; /* num of queue pairs requested by VF */ u16 num_mac; u16 num_vlan; - u8 num_req_qs; /* num of queue pairs requested by VF */ + u16 num_vf_qs; /* num of queue configured per VF */ }; #ifdef CONFIG_PCI_IOV void ice_process_vflr_event(struct ice_pf *pf); int ice_sriov_configure(struct pci_dev *pdev, int num_vfs); int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac); -int ice_get_vf_cfg(struct net_device *netdev, int vf_id, - struct ifla_vf_info *ivi); +int +ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi); void ice_free_vfs(struct ice_pf *pf); void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event); @@ -86,17 +92,17 @@ void ice_vc_notify_link_state(struct ice_pf *pf); void ice_vc_notify_reset(struct ice_pf *pf); bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr); -int ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, - u16 vlan_id, u8 qos, __be16 vlan_proto); - -int ice_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate, - int max_tx_rate); +int +ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, + __be16 vlan_proto); int ice_set_vf_trust(struct net_device *netdev, int vf_id, bool trusted); int ice_set_vf_link_state(struct net_device *netdev, int vf_id, int link_state); int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena); + +int ice_calc_vf_reg_idx(struct ice_vf *vf, struct ice_q_vector *q_vector); #else /* CONFIG_PCI_IOV */ #define ice_process_vflr_event(pf) do {} while (0) #define ice_free_vfs(pf) do {} while (0) @@ -163,11 +169,10 @@ ice_set_vf_link_state(struct net_device __always_unused *netdev, } static inline int -ice_set_vf_bw(struct net_device __always_unused *netdev, - int __always_unused vf_id, int __always_unused min_tx_rate, - int __always_unused max_tx_rate) +ice_calc_vf_reg_idx(struct ice_vf __always_unused *vf, + struct ice_q_vector __always_unused *q_vector) { - return -EOPNOTSUPP; + return 0; } #endif /* CONFIG_PCI_IOV */ #endif /* _ICE_VIRTCHNL_PF_H_ */ |