diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2019-09-18 12:34:53 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2019-09-18 12:34:53 -0700 |
| commit | 81160dda9a7aad13c04e78bb2cfd3c4630e3afab (patch) | |
| tree | 4bf79ffa9fc7dc5e2915ff978778c3402c491113 /drivers/net/ethernet/intel/ice | |
| parent | 8b53c76533aa4356602aea98f98a2f3b4051464c (diff) | |
| parent | 1bab8d4c488be22d57f9dd09968c90a0ddc413bf (diff) | |
| download | linux-81160dda9a7aad13c04e78bb2cfd3c4630e3afab.tar.bz2 | |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next
Pull networking updates from David Miller:
1) Support IPV6 RA Captive Portal Identifier, from Maciej Żenczykowski.
2) Use bio_vec in the networking instead of custom skb_frag_t, from
Matthew Wilcox.
3) Make use of xmit_more in r8169 driver, from Heiner Kallweit.
4) Add devmap_hash to xdp, from Toke Høiland-Jørgensen.
5) Support all variants of 5750X bnxt_en chips, from Michael Chan.
6) More RTNL avoidance work in the core and mlx5 driver, from Vlad
Buslov.
7) Add TCP syn cookies bpf helper, from Petar Penkov.
8) Add 'nettest' to selftests and use it, from David Ahern.
9) Add extack support to drop_monitor, add packet alert mode and
support for HW drops, from Ido Schimmel.
10) Add VLAN offload to stmmac, from Jose Abreu.
11) Lots of devm_platform_ioremap_resource() conversions, from
YueHaibing.
12) Add IONIC driver, from Shannon Nelson.
13) Several kTLS cleanups, from Jakub Kicinski.
* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1930 commits)
mlxsw: spectrum_buffers: Add the ability to query the CPU port's shared buffer
mlxsw: spectrum: Register CPU port with devlink
mlxsw: spectrum_buffers: Prevent changing CPU port's configuration
net: ena: fix incorrect update of intr_delay_resolution
net: ena: fix retrieval of nonadaptive interrupt moderation intervals
net: ena: fix update of interrupt moderation register
net: ena: remove all old adaptive rx interrupt moderation code from ena_com
net: ena: remove ena_restore_ethtool_params() and relevant fields
net: ena: remove old adaptive interrupt moderation code from ena_netdev
net: ena: remove code duplication in ena_com_update_nonadaptive_moderation_interval _*()
net: ena: enable the interrupt_moderation in driver_supported_features
net: ena: reimplement set/get_coalesce()
net: ena: switch to dim algorithm for rx adaptive interrupt moderation
net: ena: add intr_moder_rx_interval to struct ena_com_dev and use it
net: phy: adin: implement Energy Detect Powerdown mode via phy-tunable
ethtool: implement Energy Detect Powerdown support via phy-tunable
xen-netfront: do not assume sk_buff_head list is empty in error handling
s390/ctcm: Delete unnecessary checks before the macro call “dev_kfree_skb”
net: ena: don't wake up tx queue when down
drop_monitor: Better sanitize notified packets
...
Diffstat (limited to 'drivers/net/ethernet/intel/ice')
26 files changed, 4924 insertions, 1485 deletions
diff --git a/drivers/net/ethernet/intel/ice/Makefile b/drivers/net/ethernet/intel/ice/Makefile index 2d140ba83781..9edde960b4f2 100644 --- a/drivers/net/ethernet/intel/ice/Makefile +++ b/drivers/net/ethernet/intel/ice/Makefile @@ -15,6 +15,7 @@ ice-y := ice_main.o \ ice_sched.o \ ice_lib.o \ ice_txrx.o \ + ice_flex_pipe.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 9ee6b55553c0..45e100666049 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -8,6 +8,7 @@ #include <linux/errno.h> #include <linux/kernel.h> #include <linux/module.h> +#include <linux/firmware.h> #include <linux/netdevice.h> #include <linux/compiler.h> #include <linux/etherdevice.h> @@ -29,6 +30,7 @@ #include <linux/sctp.h> #include <linux/ipv6.h> #include <linux/if_bridge.h> +#include <linux/ctype.h> #include <linux/avf/virtchnl.h> #include <net/ipv6.h> #include "ice_devids.h" @@ -47,33 +49,16 @@ extern const char ice_drv_ver[]; #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_NUM_TX_DESC 256 +#define ICE_DFLT_NUM_RX_DESC 2048 #define ICE_DFLT_TRAFFIC_CLASS BIT(0) #define ICE_INT_NAME_STR_LEN (IFNAMSIZ + 16) -#define ICE_ETHTOOL_FWVER_LEN 32 #define ICE_AQ_LEN 64 -#define ICE_MBXQ_LEN 64 +#define ICE_MBXSQ_LEN 64 +#define ICE_MBXRQ_LEN 512 #define ICE_MIN_MSIX 2 #define ICE_NO_VSI 0xffff -#define ICE_MAX_TXQS 2048 -#define ICE_MAX_RXQS 2048 #define ICE_VSI_MAP_CONTIG 0 #define ICE_VSI_MAP_SCATTER 1 #define ICE_MAX_SCATTER_TXQS 16 @@ -86,16 +71,6 @@ extern const char ice_drv_ver[]; #define ICE_RES_MISC_VEC_ID (ICE_RES_VALID_BIT - 1) #define ICE_INVAL_Q_INDEX 0xffff #define ICE_INVAL_VFID 256 -#define ICE_MAX_VF_COUNT 256 -#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) -#define ICE_DFLT_INTR_PER_VF (ICE_DFLT_QS_PER_VF + 1) #define ICE_MAX_RESET_WAIT 20 @@ -220,6 +195,7 @@ enum ice_state { __ICE_CFG_BUSY, __ICE_SERVICE_SCHED, __ICE_SERVICE_DIS, + __ICE_OICR_INTR_DIS, /* Global OICR interrupt disabled */ __ICE_STATE_NBITS /* must be last */ }; @@ -257,9 +233,6 @@ struct ice_vsi { u16 vsi_num; /* HW (absolute) index of this VSI */ u16 idx; /* software index in pf->vsi[] */ - /* Interrupt thresholds */ - u16 work_lmt; - s16 vf_id; /* VF ID for SR-IOV VSIs */ u16 ethtype; /* Ethernet protocol for pause frame */ @@ -292,8 +265,8 @@ struct ice_vsi { /* queue information */ u8 tx_mapping_mode; /* ICE_MAP_MODE_[CONTIG|SCATTER] */ u8 rx_mapping_mode; /* ICE_MAP_MODE_[CONTIG|SCATTER] */ - u16 txq_map[ICE_MAX_TXQS]; /* index in pf->avail_txqs */ - u16 rxq_map[ICE_MAX_RXQS]; /* index in pf->avail_rxqs */ + u16 *txq_map; /* index in pf->avail_txqs */ + u16 *rxq_map; /* index in pf->avail_rxqs */ u16 alloc_txq; /* Allocated Tx queues */ u16 num_txq; /* Used Tx queues */ u16 alloc_rxq; /* Allocated Rx queues */ @@ -329,15 +302,16 @@ struct ice_q_vector { } ____cacheline_internodealigned_in_smp; enum ice_pf_flags { - ICE_FLAG_MSIX_ENA, ICE_FLAG_FLTR_SYNC, ICE_FLAG_RSS_ENA, ICE_FLAG_SRIOV_ENA, ICE_FLAG_SRIOV_CAPABLE, ICE_FLAG_DCB_CAPABLE, ICE_FLAG_DCB_ENA, + ICE_FLAG_ADV_FEATURES, ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, - ICE_FLAG_ENABLE_FW_LLDP, + ICE_FLAG_NO_MEDIA, + ICE_FLAG_FW_LLDP_AGENT, ICE_FLAG_ETHTOOL_CTXT, /* set when ethtool holds RTNL lock */ ICE_PF_FLAGS_NBITS /* must be last */ }; @@ -363,9 +337,9 @@ struct ice_pf { u16 num_vf_qps; /* num queue pairs per VF */ u16 num_vf_msix; /* num vectors per VF */ DECLARE_BITMAP(state, __ICE_STATE_NBITS); - DECLARE_BITMAP(avail_txqs, ICE_MAX_TXQS); - DECLARE_BITMAP(avail_rxqs, ICE_MAX_RXQS); DECLARE_BITMAP(flags, ICE_PF_FLAGS_NBITS); + unsigned long *avail_txqs; /* bitmap to track PF Tx queue usage */ + unsigned long *avail_rxqs; /* bitmap to track PF Rx queue usage */ unsigned long serv_tmr_period; unsigned long serv_tmr_prev; struct timer_list serv_tmr; @@ -376,11 +350,11 @@ struct ice_pf { u32 hw_csum_rx_error; u32 oicr_idx; /* Other interrupt cause MSIX vector index */ u32 num_avail_sw_msix; /* remaining MSIX SW vectors left unclaimed */ + u16 max_pf_txqs; /* Total Tx queues PF wide */ + u16 max_pf_rxqs; /* Total Rx queues PF wide */ 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 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! */ u16 num_alloc_vsi; u16 corer_count; /* Core reset count */ @@ -433,21 +407,26 @@ ice_irq_dynamic_ena(struct ice_hw *hw, 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 + * ice_netdev_to_pf - Retrieve the PF struct associated with a netdev + * @netdev: pointer to the netdev struct */ -static inline struct ice_vsi * -ice_find_vsi_by_type(struct ice_pf *pf, enum ice_vsi_type type) +static inline struct ice_pf *ice_netdev_to_pf(struct net_device *netdev) { - int i; + struct ice_netdev_priv *np = netdev_priv(netdev); - for (i = 0; i < pf->num_alloc_vsi; i++) { - struct ice_vsi *vsi = pf->vsi[i]; + return np->vsi->back; +} - if (vsi && vsi->type == type) - return vsi; - } +/** + * ice_get_main_vsi - Get the PF VSI + * @pf: PF instance + * + * returns pf->vsi[0], which by definition is the PF VSI + */ +static inline struct ice_vsi *ice_get_main_vsi(struct ice_pf *pf) +{ + if (pf->vsi) + return pf->vsi[0]; return NULL; } @@ -455,6 +434,11 @@ ice_find_vsi_by_type(struct ice_pf *pf, enum ice_vsi_type type) 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); +void ice_set_ethtool_safe_mode_ops(struct net_device *netdev); +u16 ice_get_avail_txq_count(struct ice_pf *pf); +u16 ice_get_avail_rxq_count(struct ice_pf *pf); +void ice_update_vsi_stats(struct ice_vsi *vsi); +void ice_update_pf_stats(struct ice_pf *pf); int ice_up(struct ice_vsi *vsi); int ice_down(struct ice_vsi *vsi); int ice_vsi_cfg(struct ice_vsi *vsi); diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h index 765e3c2ed045..023e3d2fee5f 100644 --- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h +++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h @@ -33,11 +33,22 @@ struct ice_aqc_get_ver { u8 api_patch; }; +/* Send driver version (indirect 0x0002) */ +struct ice_aqc_driver_ver { + u8 major_ver; + u8 minor_ver; + u8 build_ver; + u8 subbuild_ver; + u8 reserved[4]; + __le32 addr_high; + __le32 addr_low; +}; + /* Queue Shutdown (direct 0x0003) */ struct ice_aqc_q_shutdown { - __le32 driver_unloading; + u8 driver_unloading; #define ICE_AQC_DRIVER_UNLOADING BIT(0) - u8 reserved[12]; + u8 reserved[15]; }; /* Request resource ownership (direct 0x0008) @@ -91,6 +102,7 @@ struct ice_aqc_list_caps_elem { #define ICE_AQC_CAPS_SRIOV 0x0012 #define ICE_AQC_CAPS_VF 0x0013 #define ICE_AQC_CAPS_VSI 0x0017 +#define ICE_AQC_CAPS_DCB 0x0018 #define ICE_AQC_CAPS_RSS 0x0040 #define ICE_AQC_CAPS_RXQS 0x0041 #define ICE_AQC_CAPS_TXQS 0x0042 @@ -1518,6 +1530,56 @@ struct ice_aqc_get_clear_fw_log { __le32 addr_low; }; +/* Download Package (indirect 0x0C40) */ +/* Also used for Update Package (indirect 0x0C42) */ +struct ice_aqc_download_pkg { + u8 flags; +#define ICE_AQC_DOWNLOAD_PKG_LAST_BUF 0x01 + u8 reserved[3]; + __le32 reserved1; + __le32 addr_high; + __le32 addr_low; +}; + +struct ice_aqc_download_pkg_resp { + __le32 error_offset; + __le32 error_info; + __le32 addr_high; + __le32 addr_low; +}; + +/* Get Package Info List (indirect 0x0C43) */ +struct ice_aqc_get_pkg_info_list { + __le32 reserved1; + __le32 reserved2; + __le32 addr_high; + __le32 addr_low; +}; + +/* Version format for packages */ +struct ice_pkg_ver { + u8 major; + u8 minor; + u8 update; + u8 draft; +}; + +#define ICE_PKG_NAME_SIZE 32 + +struct ice_aqc_get_pkg_info { + struct ice_pkg_ver ver; + char name[ICE_PKG_NAME_SIZE]; + u8 is_in_nvm; + u8 is_active; + u8 is_active_at_boot; + u8 is_modified; +}; + +/* Get Package Info List response buffer format (0x0C43) */ +struct ice_aqc_get_pkg_info_resp { + __le32 count; + struct ice_aqc_get_pkg_info pkg_info[1]; +}; /** * struct ice_aq_desc - Admin Queue (AQ) descriptor * @flags: ICE_AQ_FLAG_* flags @@ -1546,6 +1608,7 @@ struct ice_aq_desc { u8 raw[16]; struct ice_aqc_generic generic; struct ice_aqc_get_ver get_ver; + struct ice_aqc_driver_ver driver_ver; struct ice_aqc_q_shutdown q_shutdown; struct ice_aqc_req_res res_owner; struct ice_aqc_manage_mac_read mac_read; @@ -1579,6 +1642,7 @@ 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_download_pkg download_pkg; 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; @@ -1610,12 +1674,19 @@ enum ice_aq_err { ICE_AQ_RC_EBUSY = 12, /* Device or resource busy */ ICE_AQ_RC_EEXIST = 13, /* Object already exists */ ICE_AQ_RC_ENOSPC = 16, /* No space left or allocation failure */ + ICE_AQ_RC_ENOSYS = 17, /* Function not implemented */ + ICE_AQ_RC_ENOSEC = 24, /* Missing security manifest */ + ICE_AQ_RC_EBADSIG = 25, /* Bad RSA signature */ + ICE_AQ_RC_ESVN = 26, /* SVN number prohibits this package */ + ICE_AQ_RC_EBADMAN = 27, /* Manifest hash mismatch */ + ICE_AQ_RC_EBADBUF = 28, /* Buffer hash mismatches manifest */ }; /* Admin Queue command opcodes */ enum ice_adminq_opc { /* AQ commands */ ice_aqc_opc_get_ver = 0x0001, + ice_aqc_opc_driver_ver = 0x0002, ice_aqc_opc_q_shutdown = 0x0003, /* resource ownership */ @@ -1697,6 +1768,10 @@ enum ice_adminq_opc { ice_aqc_opc_add_txqs = 0x0C30, ice_aqc_opc_dis_txqs = 0x0C31, + /* package commands */ + ice_aqc_opc_download_pkg = 0x0C40, + ice_aqc_opc_get_pkg_info_list = 0x0C43, + /* debug commands */ ice_aqc_opc_fw_logging = 0xFF09, ice_aqc_opc_fw_logging_info = 0xFF10, diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 2e0731c1e1a3..3a6b3950eb0e 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -263,21 +263,23 @@ 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) { - struct ice_link_status *hw_link_info_old, *hw_link_info; struct ice_aqc_get_link_status_data link_data = { 0 }; struct ice_aqc_get_link_status *resp; + struct ice_link_status *li_old, *li; enum ice_media_type *hw_media_type; struct ice_fc_info *hw_fc_info; bool tx_pause, rx_pause; struct ice_aq_desc desc; enum ice_status status; + struct ice_hw *hw; u16 cmd_flags; if (!pi) return ICE_ERR_PARAM; - hw_link_info_old = &pi->phy.link_info_old; + hw = pi->hw; + li_old = &pi->phy.link_info_old; hw_media_type = &pi->phy.media_type; - hw_link_info = &pi->phy.link_info; + li = &pi->phy.link_info; hw_fc_info = &pi->fc; ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_link_status); @@ -286,27 +288,27 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, resp->cmd_flags = cpu_to_le16(cmd_flags); resp->lport_num = pi->lport; - status = ice_aq_send_cmd(pi->hw, &desc, &link_data, sizeof(link_data), - cd); + status = ice_aq_send_cmd(hw, &desc, &link_data, sizeof(link_data), cd); if (status) return status; /* save off old link status information */ - *hw_link_info_old = *hw_link_info; + *li_old = *li; /* update current link status information */ - hw_link_info->link_speed = le16_to_cpu(link_data.link_speed); - hw_link_info->phy_type_low = le64_to_cpu(link_data.phy_type_low); - hw_link_info->phy_type_high = le64_to_cpu(link_data.phy_type_high); + li->link_speed = le16_to_cpu(link_data.link_speed); + li->phy_type_low = le64_to_cpu(link_data.phy_type_low); + li->phy_type_high = le64_to_cpu(link_data.phy_type_high); *hw_media_type = ice_get_media_type(pi); - hw_link_info->link_info = link_data.link_info; - 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; + li->link_info = link_data.link_info; + li->an_info = link_data.an_info; + li->ext_info = link_data.ext_info; + li->max_frame_size = le16_to_cpu(link_data.max_frame_size); + li->fec_info = link_data.cfg & ICE_AQ_FEC_MASK; + li->topo_media_conflict = link_data.topo_media_conflict; + li->pacing = link_data.cfg & (ICE_AQ_CFG_PACING_M | + ICE_AQ_CFG_PACING_TYPE_M); /* update fc info */ tx_pause = !!(link_data.an_info & ICE_AQ_LINK_PAUSE_TX); @@ -320,12 +322,24 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, else hw_fc_info->current_mode = ICE_FC_NONE; - hw_link_info->lse_ena = - !!(resp->cmd_flags & cpu_to_le16(ICE_AQ_LSE_IS_ENABLED)); + li->lse_ena = !!(resp->cmd_flags & cpu_to_le16(ICE_AQ_LSE_IS_ENABLED)); + + ice_debug(hw, ICE_DBG_LINK, "link_speed = 0x%x\n", li->link_speed); + ice_debug(hw, ICE_DBG_LINK, "phy_type_low = 0x%llx\n", + (unsigned long long)li->phy_type_low); + ice_debug(hw, ICE_DBG_LINK, "phy_type_high = 0x%llx\n", + (unsigned long long)li->phy_type_high); + ice_debug(hw, ICE_DBG_LINK, "media_type = 0x%x\n", *hw_media_type); + ice_debug(hw, ICE_DBG_LINK, "link_info = 0x%x\n", li->link_info); + ice_debug(hw, ICE_DBG_LINK, "an_info = 0x%x\n", li->an_info); + ice_debug(hw, ICE_DBG_LINK, "ext_info = 0x%x\n", li->ext_info); + ice_debug(hw, ICE_DBG_LINK, "lse_ena = 0x%x\n", li->lse_ena); + ice_debug(hw, ICE_DBG_LINK, "max_frame = 0x%x\n", li->max_frame_size); + ice_debug(hw, ICE_DBG_LINK, "pacing = 0x%x\n", li->pacing); /* save link status information */ if (link) - *link = *hw_link_info; + *link = *li; /* flag cleared so calling functions don't call AQ again */ pi->phy.get_link_info = false; @@ -715,6 +729,29 @@ static void ice_get_itr_intrl_gran(struct ice_hw *hw) } /** + * ice_get_nvm_version - get cached NVM version data + * @hw: pointer to the hardware structure + * @oem_ver: 8 bit NVM version + * @oem_build: 16 bit NVM build number + * @oem_patch: 8 NVM patch number + * @ver_hi: high 16 bits of the NVM version + * @ver_lo: low 16 bits of the NVM version + */ +void +ice_get_nvm_version(struct ice_hw *hw, u8 *oem_ver, u16 *oem_build, + u8 *oem_patch, u8 *ver_hi, u8 *ver_lo) +{ + struct ice_nvm_info *nvm = &hw->nvm; + + *oem_ver = (u8)((nvm->oem_ver & ICE_OEM_VER_MASK) >> ICE_OEM_VER_SHIFT); + *oem_patch = (u8)(nvm->oem_ver & ICE_OEM_VER_PATCH_MASK); + *oem_build = (u16)((nvm->oem_ver & ICE_OEM_VER_BUILD_MASK) >> + ICE_OEM_VER_BUILD_SHIFT); + *ver_hi = (nvm->ver & ICE_NVM_VER_HI_MASK) >> ICE_NVM_VER_HI_SHIFT; + *ver_lo = (nvm->ver & ICE_NVM_VER_LO_MASK) >> ICE_NVM_VER_LO_SHIFT; +} + +/** * ice_init_hw - main hardware initialization routine * @hw: pointer to the hardware structure */ @@ -740,7 +777,7 @@ enum ice_status ice_init_hw(struct ice_hw *hw) ice_get_itr_intrl_gran(hw); - status = ice_init_all_ctrlq(hw); + status = ice_create_all_ctrlq(hw); if (status) goto err_unroll_cqinit; @@ -845,7 +882,9 @@ enum ice_status ice_init_hw(struct ice_hw *hw) ice_init_flex_flds(hw, ICE_RXDID_FLEX_NIC); ice_init_flex_flds(hw, ICE_RXDID_FLEX_NIC_2); - + status = ice_init_hw_tbls(hw); + if (status) + goto err_unroll_fltr_mgmt_struct; return 0; err_unroll_fltr_mgmt_struct: @@ -855,7 +894,7 @@ err_unroll_sched: err_unroll_alloc: devm_kfree(ice_hw_to_dev(hw), hw->port_info); err_unroll_cqinit: - ice_shutdown_all_ctrlq(hw); + ice_destroy_all_ctrlq(hw); return status; } @@ -873,6 +912,8 @@ void ice_deinit_hw(struct ice_hw *hw) ice_sched_cleanup_all(hw); ice_sched_clear_agg(hw); + ice_free_seg(hw); + ice_free_hw_tbls(hw); if (hw->port_info) { devm_kfree(ice_hw_to_dev(hw), hw->port_info); @@ -881,7 +922,7 @@ void ice_deinit_hw(struct ice_hw *hw) /* Attempt to disable FW logging before shutting down control queues */ ice_cfg_fw_log(hw, false); - ice_shutdown_all_ctrlq(hw); + ice_destroy_all_ctrlq(hw); /* Clear VSI contexts if not already cleared */ ice_clear_all_vsi_ctx(hw); @@ -1078,6 +1119,7 @@ static const struct ice_ctx_ele ice_rlan_ctx_info[] = { ICE_CTX_STORE(ice_rlan_ctx, tphdata_ena, 1, 195), ICE_CTX_STORE(ice_rlan_ctx, tphhead_ena, 1, 196), ICE_CTX_STORE(ice_rlan_ctx, lrxqthresh, 3, 198), + ICE_CTX_STORE(ice_rlan_ctx, prefena, 1, 201), { 0 } }; @@ -1088,7 +1130,8 @@ 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 and enables the hardware to prefetch descriptors + * instead of only fetching them on demand */ enum ice_status ice_write_rxq_ctx(struct ice_hw *hw, struct ice_rlan_ctx *rlan_ctx, @@ -1096,6 +1139,11 @@ ice_write_rxq_ctx(struct ice_hw *hw, struct ice_rlan_ctx *rlan_ctx, { u8 ctx_buf[ICE_RXQ_CTX_SZ] = { 0 }; + if (!rlan_ctx) + return ICE_ERR_BAD_PTR; + + rlan_ctx->prefena = 1; + ice_set_ctx((u8 *)rlan_ctx, ctx_buf, ice_rlan_ctx_info); return ice_copy_rxq_ctx_to_hw(hw, ctx_buf, rxq_index); } @@ -1111,6 +1159,7 @@ const struct ice_ctx_ele ice_tlan_ctx_info[] = { ICE_CTX_STORE(ice_tlan_ctx, vmvf_type, 2, 78), ICE_CTX_STORE(ice_tlan_ctx, src_vsi, 10, 80), ICE_CTX_STORE(ice_tlan_ctx, tsyn_ena, 1, 90), + ICE_CTX_STORE(ice_tlan_ctx, internal_usage_flag, 1, 91), ICE_CTX_STORE(ice_tlan_ctx, alt_vlan, 1, 92), ICE_CTX_STORE(ice_tlan_ctx, cpuid, 8, 93), ICE_CTX_STORE(ice_tlan_ctx, wb_mode, 1, 101), @@ -1129,7 +1178,7 @@ const struct ice_ctx_ele ice_tlan_ctx_info[] = { ICE_CTX_STORE(ice_tlan_ctx, drop_ena, 1, 165), ICE_CTX_STORE(ice_tlan_ctx, cache_prof_idx, 2, 166), ICE_CTX_STORE(ice_tlan_ctx, pkt_shaper_prof_idx, 3, 168), - ICE_CTX_STORE(ice_tlan_ctx, int_q_state, 110, 171), + ICE_CTX_STORE(ice_tlan_ctx, int_q_state, 122, 171), { 0 } }; @@ -1185,6 +1234,12 @@ ice_debug_cq(struct ice_hw *hw, u32 __maybe_unused mask, void *desc, void *buf, /* FW Admin Queue command wrappers */ +/* Software lock/mutex that is meant to be held while the Global Config Lock + * in firmware is acquired by the software to prevent most (but not all) types + * of AQ commands from being sent to FW + */ +DEFINE_MUTEX(ice_global_cfg_lock_sw); + /** * ice_aq_send_cmd - send FW Admin Queue command to FW Admin Queue * @hw: pointer to the HW struct @@ -1199,7 +1254,38 @@ enum ice_status ice_aq_send_cmd(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf, u16 buf_size, struct ice_sq_cd *cd) { - return ice_sq_send_cmd(hw, &hw->adminq, desc, buf, buf_size, cd); + struct ice_aqc_req_res *cmd = &desc->params.res_owner; + bool lock_acquired = false; + enum ice_status status; + + /* When a package download is in process (i.e. when the firmware's + * Global Configuration Lock resource is held), only the Download + * Package, Get Version, Get Package Info List and Release Resource + * (with resource ID set to Global Config Lock) AdminQ commands are + * allowed; all others must block until the package download completes + * and the Global Config Lock is released. See also + * ice_acquire_global_cfg_lock(). + */ + switch (le16_to_cpu(desc->opcode)) { + case ice_aqc_opc_download_pkg: + case ice_aqc_opc_get_pkg_info_list: + case ice_aqc_opc_get_ver: + break; + case ice_aqc_opc_release_res: + if (le16_to_cpu(cmd->res_id) == ICE_AQC_RES_ID_GLBL_LOCK) + break; + /* fall-through */ + default: + mutex_lock(&ice_global_cfg_lock_sw); + lock_acquired = true; + break; + } + + status = ice_sq_send_cmd(hw, &hw->adminq, desc, buf, buf_size, cd); + if (lock_acquired) + mutex_unlock(&ice_global_cfg_lock_sw); + + return status; } /** @@ -1237,6 +1323,43 @@ enum ice_status ice_aq_get_fw_ver(struct ice_hw *hw, struct ice_sq_cd *cd) } /** + * ice_aq_send_driver_ver + * @hw: pointer to the HW struct + * @dv: driver's major, minor version + * @cd: pointer to command details structure or NULL + * + * Send the driver version (0x0002) to the firmware + */ +enum ice_status +ice_aq_send_driver_ver(struct ice_hw *hw, struct ice_driver_ver *dv, + struct ice_sq_cd *cd) +{ + struct ice_aqc_driver_ver *cmd; + struct ice_aq_desc desc; + u16 len; + + cmd = &desc.params.driver_ver; + + if (!dv) + return ICE_ERR_PARAM; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_driver_ver); + + desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD); + cmd->major_ver = dv->major_ver; + cmd->minor_ver = dv->minor_ver; + cmd->build_ver = dv->build_ver; + cmd->subbuild_ver = dv->subbuild_ver; + + len = 0; + while (len < sizeof(dv->driver_string) && + isascii(dv->driver_string[len]) && dv->driver_string[len]) + len++; + + return ice_aq_send_cmd(hw, &desc, dv->driver_string, len, cd); +} + +/** * ice_aq_q_shutdown * @hw: pointer to the HW struct * @unloading: is the driver unloading itself @@ -1254,7 +1377,7 @@ enum ice_status ice_aq_q_shutdown(struct ice_hw *hw, bool unloading) ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_q_shutdown); if (unloading) - cmd->driver_unloading = cpu_to_le32(ICE_AQC_DRIVER_UNLOADING); + cmd->driver_unloading = ICE_AQC_DRIVER_UNLOADING; return ice_aq_send_cmd(hw, &desc, NULL, 0, NULL); } @@ -1529,29 +1652,29 @@ 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, - "%s: valid functions = %d\n", prefix, + "%s: valid_functions (bitmap) = %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, - "%s: SR-IOV = %d\n", prefix, + "%s: sr_iov_1_1 = %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, - "%s: VFs exposed = %d\n", prefix, + "%s: num_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, - "%s: VFs allocated = %d\n", prefix, + "%s: num_allocd_vfs = %d\n", prefix, func_p->num_allocd_vfs); ice_debug(hw, ICE_DBG_INIT, - "%s: VF base_id = %d\n", prefix, + "%s: vf_base_id = %d\n", prefix, func_p->vf_base_id); } break; @@ -1559,63 +1682,75 @@ 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, - "%s: num VSI alloc to host = %d\n", + "%s: num_vsi_allocd_to_host = %d\n", prefix, dev_p->num_vsi_allocd_to_host); } else if (func_p) { 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", + "%s: guar_num_vsi (fw) = %d\n", prefix, number); ice_debug(hw, ICE_DBG_INIT, - "%s: num guaranteed VSI = %d\n", + "%s: guar_num_vsi = %d\n", prefix, func_p->guar_num_vsi); } break; + case ICE_AQC_CAPS_DCB: + caps->dcb = (number == 1); + caps->active_tc_bitmap = logical_id; + caps->maxtc = phys_id; + ice_debug(hw, ICE_DBG_INIT, + "%s: dcb = %d\n", prefix, caps->dcb); + ice_debug(hw, ICE_DBG_INIT, + "%s: active_tc_bitmap = %d\n", prefix, + caps->active_tc_bitmap); + ice_debug(hw, ICE_DBG_INIT, + "%s: maxtc = %d\n", prefix, caps->maxtc); + break; case ICE_AQC_CAPS_RSS: caps->rss_table_size = number; caps->rss_table_entry_width = logical_id; ice_debug(hw, ICE_DBG_INIT, - "%s: RSS table size = %d\n", prefix, + "%s: rss_table_size = %d\n", prefix, caps->rss_table_size); ice_debug(hw, ICE_DBG_INIT, - "%s: RSS table width = %d\n", prefix, + "%s: rss_table_entry_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, - "%s: num Rx queues = %d\n", prefix, + "%s: num_rxq = %d\n", prefix, caps->num_rxq); ice_debug(hw, ICE_DBG_INIT, - "%s: Rx first queue ID = %d\n", prefix, + "%s: rxq_first_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, - "%s: num Tx queues = %d\n", prefix, + "%s: num_txq = %d\n", prefix, caps->num_txq); ice_debug(hw, ICE_DBG_INIT, - "%s: Tx first queue ID = %d\n", prefix, + "%s: txq_first_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, - "%s: MSIX vector count = %d\n", prefix, + "%s: num_msix_vectors = %d\n", prefix, caps->num_msix_vectors); ice_debug(hw, ICE_DBG_INIT, - "%s: MSIX first vector index = %d\n", prefix, + "%s: msix_vector_first_id = %d\n", prefix, caps->msix_vector_first_id); break; case ICE_AQC_CAPS_MAX_MTU: caps->max_mtu = number; - ice_debug(hw, ICE_DBG_INIT, "%s: max MTU = %d\n", + ice_debug(hw, ICE_DBG_INIT, "%s: max_mtu = %d\n", prefix, caps->max_mtu); break; default: @@ -1712,6 +1847,75 @@ ice_discover_caps(struct ice_hw *hw, enum ice_adminq_opc opc) } /** + * ice_set_safe_mode_caps - Override dev/func capabilities when in safe mode + * @hw: pointer to the hardware structure + */ +void ice_set_safe_mode_caps(struct ice_hw *hw) +{ + struct ice_hw_func_caps *func_caps = &hw->func_caps; + struct ice_hw_dev_caps *dev_caps = &hw->dev_caps; + u32 valid_func, rxq_first_id, txq_first_id; + u32 msix_vector_first_id, max_mtu; + u32 num_func = 0; + u8 i; + + /* cache some func_caps values that should be restored after memset */ + valid_func = func_caps->common_cap.valid_functions; + txq_first_id = func_caps->common_cap.txq_first_id; + rxq_first_id = func_caps->common_cap.rxq_first_id; + msix_vector_first_id = func_caps->common_cap.msix_vector_first_id; + max_mtu = func_caps->common_cap.max_mtu; + + /* unset func capabilities */ + memset(func_caps, 0, sizeof(*func_caps)); + + /* restore cached values */ + func_caps->common_cap.valid_functions = valid_func; + func_caps->common_cap.txq_first_id = txq_first_id; + func_caps->common_cap.rxq_first_id = rxq_first_id; + func_caps->common_cap.msix_vector_first_id = msix_vector_first_id; + func_caps->common_cap.max_mtu = max_mtu; + + /* one Tx and one Rx queue in safe mode */ + func_caps->common_cap.num_rxq = 1; + func_caps->common_cap.num_txq = 1; + + /* two MSIX vectors, one for traffic and one for misc causes */ + func_caps->common_cap.num_msix_vectors = 2; + func_caps->guar_num_vsi = 1; + + /* cache some dev_caps values that should be restored after memset */ + valid_func = dev_caps->common_cap.valid_functions; + txq_first_id = dev_caps->common_cap.txq_first_id; + rxq_first_id = dev_caps->common_cap.rxq_first_id; + msix_vector_first_id = dev_caps->common_cap.msix_vector_first_id; + max_mtu = dev_caps->common_cap.max_mtu; + + /* unset dev capabilities */ + memset(dev_caps, 0, sizeof(*dev_caps)); + + /* restore cached values */ + dev_caps->common_cap.valid_functions = valid_func; + dev_caps->common_cap.txq_first_id = txq_first_id; + dev_caps->common_cap.rxq_first_id = rxq_first_id; + dev_caps->common_cap.msix_vector_first_id = msix_vector_first_id; + dev_caps->common_cap.max_mtu = max_mtu; + + /* valid_func is a bitmap. get number of functions */ +#define ICE_MAX_FUNCS 8 + for (i = 0; i < ICE_MAX_FUNCS; i++) + if (valid_func & BIT(i)) + num_func++; + + /* one Tx and one Rx queue per function in safe mode */ + dev_caps->common_cap.num_rxq = num_func; + dev_caps->common_cap.num_txq = num_func; + + /* two MSIX vectors per function */ + dev_caps->common_cap.num_msix_vectors = 2 * num_func; +} + +/** * ice_get_caps - get info about the HW * @hw: pointer to the hardware structure */ @@ -1993,6 +2197,17 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport, desc.params.set_phy.lport_num = lport; desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD); + ice_debug(hw, ICE_DBG_LINK, "phy_type_low = 0x%llx\n", + (unsigned long long)le64_to_cpu(cfg->phy_type_low)); + ice_debug(hw, ICE_DBG_LINK, "phy_type_high = 0x%llx\n", + (unsigned long long)le64_to_cpu(cfg->phy_type_high)); + ice_debug(hw, ICE_DBG_LINK, "caps = 0x%x\n", cfg->caps); + ice_debug(hw, ICE_DBG_LINK, "low_power_ctrl = 0x%x\n", + cfg->low_power_ctrl); + ice_debug(hw, ICE_DBG_LINK, "eee_cap = 0x%x\n", cfg->eee_cap); + ice_debug(hw, ICE_DBG_LINK, "eeer_value = 0x%x\n", cfg->eeer_value); + ice_debug(hw, ICE_DBG_LINK, "link_fec_opt = 0x%x\n", cfg->link_fec_opt); + return ice_aq_send_cmd(hw, &desc, cfg, sizeof(*cfg), cd); } @@ -2024,7 +2239,7 @@ enum ice_status ice_update_link_info(struct ice_port_info *pi) if (!pcaps) return ICE_ERR_NO_MEMORY; - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, + status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP, pcaps, NULL); if (!status) memcpy(li->module_type, &pcaps->module_type, @@ -2174,27 +2389,24 @@ 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 + /* Clear 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 + /* Clear 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; + /* Clear all FEC option bits. */ cfg->link_fec_opt &= ~ICE_AQC_PHY_FEC_MASK; break; case ICE_FEC_AUTO: @@ -3240,40 +3452,44 @@ void ice_replay_post(struct ice_hw *hw) /** * ice_stat_update40 - read 40 bit stat from the chip and update stat values * @hw: ptr to the hardware info - * @hireg: high 32 bit HW register to read from - * @loreg: low 32 bit HW register to read from + * @reg: offset of 64 bit HW register to read from * @prev_stat_loaded: bool to specify if previous stats are loaded * @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) +ice_stat_update40(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, + u64 *prev_stat, u64 *cur_stat) { - u64 new_data; - - new_data = rd32(hw, loreg); - new_data |= ((u64)(rd32(hw, hireg) & 0xFFFF)) << 32; + u64 new_data = rd64(hw, reg) & (BIT_ULL(40) - 1); /* device stats are not reset at PFR, they likely will not be zeroed - * when the driver starts. So save the first values read and use them as - * offsets to be subtracted from the raw values in order to report stats - * that count from zero. + * when the driver starts. Thus, save the value from the first read + * without adding to the statistic value so that we report stats which + * count up from zero. */ - if (!prev_stat_loaded) + if (!prev_stat_loaded) { *prev_stat = new_data; + return; + } + + /* Calculate the difference between the new and old values, and then + * add it to the software stat value. + */ if (new_data >= *prev_stat) - *cur_stat = new_data - *prev_stat; + *cur_stat += new_data - *prev_stat; else /* to manage the potential roll-over */ - *cur_stat = (new_data + BIT_ULL(40)) - *prev_stat; - *cur_stat &= 0xFFFFFFFFFFULL; + *cur_stat += (new_data + BIT_ULL(40)) - *prev_stat; + + /* Update the previously stored value to prepare for next read */ + *prev_stat = new_data; } /** * ice_stat_update32 - read 32 bit stat from the chip and update stat values * @hw: ptr to the hardware info - * @reg: HW register to read from + * @reg: offset of HW register to read from * @prev_stat_loaded: bool to specify if previous stats are loaded * @prev_stat: ptr to previous loaded stat value * @cur_stat: ptr to current stat value @@ -3287,17 +3503,26 @@ ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded, new_data = rd32(hw, reg); /* device stats are not reset at PFR, they likely will not be zeroed - * when the driver starts. So save the first values read and use them as - * offsets to be subtracted from the raw values in order to report stats - * that count from zero. + * when the driver starts. Thus, save the value from the first read + * without adding to the statistic value so that we report stats which + * count up from zero. */ - if (!prev_stat_loaded) + if (!prev_stat_loaded) { *prev_stat = new_data; + return; + } + + /* Calculate the difference between the new and old values, and then + * add it to the software stat value. + */ if (new_data >= *prev_stat) - *cur_stat = new_data - *prev_stat; + *cur_stat += new_data - *prev_stat; else /* to manage the potential roll-over */ - *cur_stat = (new_data + BIT_ULL(32)) - *prev_stat; + *cur_stat += (new_data + BIT_ULL(32)) - *prev_stat; + + /* Update the previously stored value to prepare for next read */ + *prev_stat = new_data; } /** diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h index d1f8353fe6bb..c3df92f57777 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.h +++ b/drivers/net/ethernet/intel/ice/ice_common.h @@ -6,6 +6,7 @@ #include "ice.h" #include "ice_type.h" +#include "ice_flex_pipe.h" #include "ice_switch.h" #include <linux/avf/virtchnl.h> @@ -17,8 +18,10 @@ 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); enum ice_status ice_reset(struct ice_hw *hw, enum ice_reset_req req); +enum ice_status ice_create_all_ctrlq(struct ice_hw *hw); enum ice_status ice_init_all_ctrlq(struct ice_hw *hw); void ice_shutdown_all_ctrlq(struct ice_hw *hw); +void ice_destroy_all_ctrlq(struct ice_hw *hw); enum ice_status ice_clean_rq_elem(struct ice_hw *hw, struct ice_ctl_q_info *cq, struct ice_rq_event_info *e, u16 *pending); @@ -39,6 +42,8 @@ ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq, void ice_clear_pxe_mode(struct ice_hw *hw); enum ice_status ice_get_caps(struct ice_hw *hw); +void ice_set_safe_mode_caps(struct ice_hw *hw); + void ice_dev_onetime_setup(struct ice_hw *hw); enum ice_status @@ -64,12 +69,18 @@ void ice_fill_dflt_direct_cmd_desc(struct ice_aq_desc *desc, u16 opcode); extern const struct ice_ctx_ele ice_tlan_ctx_info[]; enum ice_status ice_set_ctx(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info); + +extern struct mutex ice_global_cfg_lock_sw; + enum ice_status ice_aq_send_cmd(struct ice_hw *hw, struct ice_aq_desc *desc, void *buf, u16 buf_size, struct ice_sq_cd *cd); enum ice_status ice_aq_get_fw_ver(struct ice_hw *hw, struct ice_sq_cd *cd); enum ice_status +ice_aq_send_driver_ver(struct ice_hw *hw, struct ice_driver_ver *dv, + struct ice_sq_cd *cd); +enum ice_status ice_aq_get_phy_caps(struct ice_port_info *pi, bool qual_mods, u8 report_mode, struct ice_aqc_get_phy_caps_data *caps, struct ice_sq_cd *cd); @@ -123,11 +134,14 @@ 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); +ice_stat_update40(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); +void +ice_get_nvm_version(struct ice_hw *hw, u8 *oem_ver, u16 *oem_build, + u8 *oem_patch, u8 *ver_hi, u8 *ver_lo); enum ice_status ice_sched_query_elem(struct ice_hw *hw, u32 node_teid, struct ice_aqc_get_elem *buf); diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c index e91ac4df0242..2353166c654e 100644 --- a/drivers/net/ethernet/intel/ice/ice_controlq.c +++ b/drivers/net/ethernet/intel/ice/ice_controlq.c @@ -310,7 +310,7 @@ ice_cfg_rq_regs(struct ice_hw *hw, struct ice_ctl_q_info *cq) * @cq: pointer to the specific Control queue * * This is the main initialization routine for the Control Send Queue - * Prior to calling this function, drivers *MUST* set the following fields + * Prior to calling this function, the driver *MUST* set the following fields * in the cq->structure: * - cq->num_sq_entries * - cq->sq_buf_size @@ -369,7 +369,7 @@ init_ctrlq_exit: * @cq: pointer to the specific Control queue * * The main initialization routine for the Admin Receive (Event) Queue. - * Prior to calling this function, drivers *MUST* set the following fields + * Prior to calling this function, the driver *MUST* set the following fields * in the cq->structure: * - cq->num_rq_entries * - cq->rq_buf_size @@ -569,14 +569,8 @@ static enum ice_status ice_init_check_adminq(struct ice_hw *hw) return 0; init_ctrlq_free_rq: - if (cq->rq.count) { - ice_shutdown_rq(hw, cq); - mutex_destroy(&cq->rq_lock); - } - if (cq->sq.count) { - ice_shutdown_sq(hw, cq); - mutex_destroy(&cq->sq_lock); - } + ice_shutdown_rq(hw, cq); + ice_shutdown_sq(hw, cq); return status; } @@ -585,12 +579,14 @@ init_ctrlq_free_rq: * @hw: pointer to the hardware structure * @q_type: specific Control queue type * - * Prior to calling this function, drivers *MUST* set the following fields + * Prior to calling this function, the driver *MUST* set the following fields * in the cq->structure: * - cq->num_sq_entries * - cq->num_rq_entries * - cq->rq_buf_size * - cq->sq_buf_size + * + * NOTE: this function does not initialize the controlq locks */ static enum ice_status ice_init_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) { @@ -616,8 +612,6 @@ static enum ice_status ice_init_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) !cq->rq_buf_size || !cq->sq_buf_size) { return ICE_ERR_CFG; } - mutex_init(&cq->sq_lock); - mutex_init(&cq->rq_lock); /* setup SQ command write back timeout */ cq->sq_cmd_timeout = ICE_CTL_Q_SQ_CMD_TIMEOUT; @@ -625,7 +619,7 @@ static enum ice_status ice_init_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) /* allocate the ATQ */ ret_code = ice_init_sq(hw, cq); if (ret_code) - goto init_ctrlq_destroy_locks; + return ret_code; /* allocate the ARQ */ ret_code = ice_init_rq(hw, cq); @@ -637,9 +631,6 @@ static enum ice_status ice_init_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) init_ctrlq_free_sq: ice_shutdown_sq(hw, cq); -init_ctrlq_destroy_locks: - mutex_destroy(&cq->sq_lock); - mutex_destroy(&cq->rq_lock); return ret_code; } @@ -647,12 +638,14 @@ init_ctrlq_destroy_locks: * ice_init_all_ctrlq - main initialization routine for all control queues * @hw: pointer to the hardware structure * - * Prior to calling this function, drivers *MUST* set the following fields + * Prior to calling this function, the driver MUST* set the following fields * in the cq->structure for all control queues: * - cq->num_sq_entries * - cq->num_rq_entries * - cq->rq_buf_size * - cq->sq_buf_size + * + * NOTE: this function does not initialize the controlq locks. */ enum ice_status ice_init_all_ctrlq(struct ice_hw *hw) { @@ -672,9 +665,47 @@ enum ice_status ice_init_all_ctrlq(struct ice_hw *hw) } /** + * ice_init_ctrlq_locks - Initialize locks for a control queue + * @cq: pointer to the control queue + * + * Initializes the send and receive queue locks for a given control queue. + */ +static void ice_init_ctrlq_locks(struct ice_ctl_q_info *cq) +{ + mutex_init(&cq->sq_lock); + mutex_init(&cq->rq_lock); +} + +/** + * ice_create_all_ctrlq - main initialization routine for all control queues + * @hw: pointer to the hardware structure + * + * Prior to calling this function, the driver *MUST* set the following fields + * in the cq->structure for all control queues: + * - cq->num_sq_entries + * - cq->num_rq_entries + * - cq->rq_buf_size + * - cq->sq_buf_size + * + * This function creates all the control queue locks and then calls + * ice_init_all_ctrlq. It should be called once during driver load. If the + * driver needs to re-initialize control queues at run time it should call + * ice_init_all_ctrlq instead. + */ +enum ice_status ice_create_all_ctrlq(struct ice_hw *hw) +{ + ice_init_ctrlq_locks(&hw->adminq); + ice_init_ctrlq_locks(&hw->mailboxq); + + return ice_init_all_ctrlq(hw); +} + +/** * ice_shutdown_ctrlq - shutdown routine for any control queue * @hw: pointer to the hardware structure * @q_type: specific Control queue type + * + * NOTE: this function does not destroy the control queue locks. */ static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) { @@ -693,19 +724,17 @@ static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) return; } - if (cq->sq.count) { - ice_shutdown_sq(hw, cq); - mutex_destroy(&cq->sq_lock); - } - if (cq->rq.count) { - ice_shutdown_rq(hw, cq); - mutex_destroy(&cq->rq_lock); - } + ice_shutdown_sq(hw, cq); + ice_shutdown_rq(hw, cq); } /** * ice_shutdown_all_ctrlq - shutdown routine for all control queues * @hw: pointer to the hardware structure + * + * NOTE: this function does not destroy the control queue locks. The driver + * may call this at runtime to shutdown and later restart control queues, such + * as in response to a reset event. */ void ice_shutdown_all_ctrlq(struct ice_hw *hw) { @@ -716,6 +745,37 @@ void ice_shutdown_all_ctrlq(struct ice_hw *hw) } /** + * ice_destroy_ctrlq_locks - Destroy locks for a control queue + * @cq: pointer to the control queue + * + * Destroys the send and receive queue locks for a given control queue. + */ +static void +ice_destroy_ctrlq_locks(struct ice_ctl_q_info *cq) +{ + mutex_destroy(&cq->sq_lock); + mutex_destroy(&cq->rq_lock); +} + +/** + * ice_destroy_all_ctrlq - exit routine for all control queues + * @hw: pointer to the hardware structure + * + * This function shuts down all the control queues and then destroys the + * control queue locks. It should be called once during driver unload. The + * driver should call ice_shutdown_all_ctrlq if it needs to shut down and + * reinitialize control queues, such as in response to a reset event. + */ +void ice_destroy_all_ctrlq(struct ice_hw *hw) +{ + /* shut down all the control queues first */ + ice_shutdown_all_ctrlq(hw); + + ice_destroy_ctrlq_locks(&hw->adminq); + ice_destroy_ctrlq_locks(&hw->mailboxq); +} + +/** * ice_clean_sq - cleans Admin send queue (ATQ) * @hw: pointer to the hardware structure * @cq: pointer to the specific Control queue diff --git a/drivers/net/ethernet/intel/ice/ice_dcb.c b/drivers/net/ethernet/intel/ice/ice_dcb.c index c2002ded65f6..dd7efff121bd 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb.c @@ -60,7 +60,7 @@ ice_aq_get_lldp_mib(struct ice_hw *hw, u8 bridge_type, u8 mib_type, void *buf, * Enable or Disable posting of an event on ARQ when LLDP MIB * associated with the interface changes (0x0A01) */ -enum ice_status +static enum ice_status ice_aq_cfg_lldp_mib_change(struct ice_hw *hw, bool ena_update, struct ice_sq_cd *cd) { @@ -444,9 +444,15 @@ ice_parse_cee_pgcfg_tlv(struct ice_cee_feat_tlv *tlv, * |pg0|pg1|pg2|pg3|pg4|pg5|pg6|pg7| * --------------------------------- */ - ice_for_each_traffic_class(i) + ice_for_each_traffic_class(i) { etscfg->tcbwtable[i] = buf[offset++]; + if (etscfg->prio_table[i] == ICE_CEE_PGID_STRICT) + dcbcfg->etscfg.tsatable[i] = ICE_IEEE_TSA_STRICT; + else + dcbcfg->etscfg.tsatable[i] = ICE_IEEE_TSA_ETS; + } + /* Number of TCs supported (1 octet) */ etscfg->maxtcs = buf[offset]; } @@ -937,10 +943,11 @@ enum ice_status ice_get_dcb_cfg(struct ice_port_info *pi) /** * ice_init_dcb * @hw: pointer to the HW struct + * @enable_mib_change: enable MIB change event * * Update DCB configuration from the Firmware */ -enum ice_status ice_init_dcb(struct ice_hw *hw) +enum ice_status ice_init_dcb(struct ice_hw *hw, bool enable_mib_change) { struct ice_port_info *pi = hw->port_info; enum ice_status ret = 0; @@ -954,7 +961,8 @@ enum ice_status ice_init_dcb(struct ice_hw *hw) pi->dcbx_status = ice_get_dcbx_status(hw); if (pi->dcbx_status == ICE_DCBX_STATUS_DONE || - pi->dcbx_status == ICE_DCBX_STATUS_IN_PROGRESS) { + pi->dcbx_status == ICE_DCBX_STATUS_IN_PROGRESS || + pi->dcbx_status == ICE_DCBX_STATUS_NOT_STARTED) { /* Get current DCBX configuration */ ret = ice_get_dcb_cfg(pi); pi->is_sw_lldp = (hw->adminq.sq_last_status == ICE_AQ_RC_EPERM); @@ -965,9 +973,39 @@ enum ice_status ice_init_dcb(struct ice_hw *hw) } /* Configure the LLDP MIB change event */ - ret = ice_aq_cfg_lldp_mib_change(hw, true, NULL); + if (enable_mib_change) { + ret = ice_aq_cfg_lldp_mib_change(hw, true, NULL); + if (!ret) + pi->is_sw_lldp = false; + } + + return ret; +} + +/** + * ice_cfg_lldp_mib_change + * @hw: pointer to the HW struct + * @ena_mib: enable/disable MIB change event + * + * Configure (disable/enable) MIB + */ +enum ice_status ice_cfg_lldp_mib_change(struct ice_hw *hw, bool ena_mib) +{ + struct ice_port_info *pi = hw->port_info; + enum ice_status ret; + + if (!hw->func_caps.common_cap.dcb) + return ICE_ERR_NOT_SUPPORTED; + + /* Get DCBX status */ + pi->dcbx_status = ice_get_dcbx_status(hw); + + if (pi->dcbx_status == ICE_DCBX_STATUS_DIS) + return ICE_ERR_NOT_READY; + + ret = ice_aq_cfg_lldp_mib_change(hw, ena_mib, NULL); if (!ret) - pi->is_sw_lldp = false; + pi->is_sw_lldp = !ena_mib; return ret; } diff --git a/drivers/net/ethernet/intel/ice/ice_dcb.h b/drivers/net/ethernet/intel/ice/ice_dcb.h index 522e1452abe2..ee138f9bdc7c 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb.h +++ b/drivers/net/ethernet/intel/ice/ice_dcb.h @@ -125,7 +125,7 @@ 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_init_dcb(struct ice_hw *hw, bool enable_mib_change); enum ice_status ice_query_port_ets(struct ice_port_info *pi, struct ice_aqc_port_ets_elem *buf, u16 buf_size, @@ -139,9 +139,7 @@ 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); +enum ice_status ice_cfg_lldp_mib_change(struct ice_hw *hw, bool ena_mib); #else /* CONFIG_DCB */ static inline enum ice_status ice_aq_stop_lldp(struct ice_hw __always_unused *hw, @@ -172,9 +170,8 @@ ice_aq_start_stop_dcbx(struct ice_hw __always_unused *hw, } 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) +ice_cfg_lldp_mib_change(struct ice_hw __always_unused *hw, + bool __always_unused ena_mib) { return 0; } diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c index fe88b127ca42..dd47869c4ad4 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c @@ -4,6 +4,48 @@ #include "ice_dcb_lib.h" /** + * ice_vsi_cfg_netdev_tc - Setup the netdev TC configuration + * @vsi: the VSI being configured + * @ena_tc: TC map to be enabled + */ +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_dcb_get_ena_tc - return bitmap of enabled TCs * @dcbcfg: DCB config to evaluate for enabled TCs */ @@ -204,15 +246,86 @@ out: } /** + * ice_cfg_etsrec_defaults - Set default ETS recommended DCB config + * @pi: port information structure + */ +static void ice_cfg_etsrec_defaults(struct ice_port_info *pi) +{ + struct ice_dcbx_cfg *dcbcfg = &pi->local_dcbx_cfg; + u8 i; + + /* Ensure ETS recommended DCB configuration is not already set */ + if (dcbcfg->etsrec.maxtcs) + return; + + /* In CEE mode, set the default to 1 TC */ + dcbcfg->etsrec.maxtcs = 1; + for (i = 0; i < ICE_MAX_TRAFFIC_CLASS; i++) { + dcbcfg->etsrec.tcbwtable[i] = i ? 0 : 100; + dcbcfg->etsrec.tsatable[i] = i ? ICE_IEEE_TSA_STRICT : + ICE_IEEE_TSA_ETS; + } +} + +/** + * 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_rebuild - rebuild DCB post reset * @pf: physical function instance */ void ice_dcb_rebuild(struct ice_pf *pf) { + struct ice_dcbx_cfg *local_dcbx_cfg, *desired_dcbx_cfg, *prev_cfg; 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) { @@ -224,9 +337,15 @@ void ice_dcb_rebuild(struct ice_pf *pf) if (!test_bit(ICE_FLAG_DCB_ENA, pf->flags)) return; + local_dcbx_cfg = &pf->hw.port_info->local_dcbx_cfg; + desired_dcbx_cfg = &pf->hw.port_info->desired_dcbx_cfg; + /* 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; + local_dcbx_cfg->etscfg.willing = 0x0; + local_dcbx_cfg->pfc.willing = 0x0; + local_dcbx_cfg->app_mode = ICE_DCBX_APPS_NON_WILLING; + + ice_cfg_etsrec_defaults(pf->hw.port_info); ret = ice_set_dcb_cfg(pf->hw.port_info); if (ret) { dev_err(&pf->pdev->dev, "Failed to set DCB to unwilling\n"); @@ -234,31 +353,37 @@ void ice_dcb_rebuild(struct ice_pf *pf) } /* Retrieve DCB config and ensure same as current in SW */ - prev_cfg = devm_kmemdup(&pf->pdev->dev, - &pf->hw.port_info->local_dcbx_cfg, + prev_cfg = devm_kmemdup(&pf->pdev->dev, 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))) { + ice_init_dcb(&pf->hw, true); + if (pf->hw.port_info->dcbx_status == ICE_DCBX_STATUS_DIS) + pf->hw.port_info->is_sw_lldp = true; + else + pf->hw.port_info->is_sw_lldp = false; + + if (ice_dcb_need_recfg(pf, prev_cfg, local_dcbx_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; + /* Set the local desired config */ + if (local_dcbx_cfg->dcbx_mode == ICE_DCBX_MODE_CEE) + memcpy(local_dcbx_cfg, desired_dcbx_cfg, + sizeof(*local_dcbx_cfg)); + + ice_cfg_etsrec_defaults(pf->hw.port_info); ret = ice_set_dcb_cfg(pf->hw.port_info); if (ret) { - dev_err(&pf->pdev->dev, "Fail restoring prev willing state\n"); + dev_err(&pf->pdev->dev, "Failed to set desired config\n"); goto dcb_error; } dev_info(&pf->pdev->dev, "DCB restored after reset\n"); @@ -330,7 +455,7 @@ static int ice_dcb_sw_dflt_cfg(struct ice_pf *pf, bool locked) memset(&pi->local_dcbx_cfg, 0, sizeof(*dcbcfg)); dcbcfg->etscfg.willing = 1; - dcbcfg->etscfg.maxtcs = 8; + dcbcfg->etscfg.maxtcs = hw->func_caps.common_cap.maxtc; dcbcfg->etscfg.tcbwtable[0] = 100; dcbcfg->etscfg.tsatable[0] = ICE_IEEE_TSA_ETS; @@ -339,7 +464,7 @@ static int ice_dcb_sw_dflt_cfg(struct ice_pf *pf, bool locked) dcbcfg->etsrec.willing = 0; dcbcfg->pfc.willing = 1; - dcbcfg->pfc.pfccap = IEEE_8021QAZ_MAX_TCS; + dcbcfg->pfc.pfccap = hw->func_caps.common_cap.maxtc; dcbcfg->numapps = 1; dcbcfg->app[0].selector = ICE_APP_SEL_ETHTYPE; @@ -364,35 +489,24 @@ 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); + err = ice_init_dcb(hw, false); + if (err && !port_info->is_sw_lldp) { + dev_err(&pf->pdev->dev, "Error initializing DCB %d\n", err); + goto dcb_init_err; } - if (port_info->dcbx_status == ICE_DCBX_STATUS_NOT_STARTED) - dev_info(&pf->pdev->dev, "DCBX not started\n"); - - if (sw_default) { + dev_info(&pf->pdev->dev, + "DCB is enabled in the hardware, max number of TCs supported on this port are %d\n", + pf->hw.func_caps.common_cap.maxtc); + if (err) { + /* FW LLDP is disabled, activate SW DCBX/LLDP mode */ + dev_info(&pf->pdev->dev, + "FW LLDP is disabled, DCBx/LLDP in SW mode.\n"); + clear_bit(ICE_FLAG_FW_LLDP_AGENT, pf->flags); err = ice_dcb_sw_dflt_cfg(pf, locked); if (err) { dev_err(&pf->pdev->dev, @@ -402,21 +516,18 @@ int ice_init_pf_dcb(struct ice_pf *pf, bool locked) } 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; } + set_bit(ICE_FLAG_FW_LLDP_AGENT, pf->flags); + /* 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: @@ -432,30 +543,31 @@ 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; + u8 port; int i; + port = hw->port_info->lport; prev_ps = &pf->stats_prev; cur_ps = &pf->stats; for (i = 0; i < 8; i++) { - ice_stat_update32(hw, GLPRT_PXOFFRXC(pf_id, i), + ice_stat_update32(hw, GLPRT_PXOFFRXC(port, 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), + ice_stat_update32(hw, GLPRT_PXONRXC(port, 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), + ice_stat_update32(hw, GLPRT_PXONTXC(port, 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), + ice_stat_update32(hw, GLPRT_PXOFFTXC(port, 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), + ice_stat_update32(hw, GLPRT_RXON2OFFCNT(port, i), pf->stat_prev_loaded, &prev_ps->priority_xon_2_xoff[i], &cur_ps->priority_xon_2_xoff[i]); @@ -502,55 +614,6 @@ ice_tx_prepare_vlan_flags_dcb(struct ice_ring *tx_ring, } /** - * 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 diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h index 819081053ff5..661a6f7bca64 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h @@ -22,6 +22,7 @@ ice_tx_prepare_vlan_flags_dcb(struct ice_ring *tx_ring, void ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, struct ice_rq_event_info *event); +void ice_vsi_cfg_netdev_tc(struct ice_vsi *vsi, u8 ena_tc); static inline void ice_set_cgd_num(struct ice_tlan_ctx *tlan_ctx, struct ice_ring *ring) { @@ -58,5 +59,6 @@ ice_tx_prepare_vlan_flags_dcb(struct ice_ring __always_unused *tx_ring, #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) +#define ice_vsi_cfg_netdev_tc(vsi, ena_tc) 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 52083a63dee6..7e23034df955 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -155,36 +155,11 @@ 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), + ICE_PRIV_FLAG("fw-lldp-agent", ICE_FLAG_FW_LLDP_AGENT), }; #define ICE_PRIV_FLAG_ARRAY_SIZE ARRAY_SIZE(ice_gstrings_priv_flags) -/** - * ice_nvm_version_str - format the NVM version strings - * @hw: ptr to the hardware info - */ -static char *ice_nvm_version_str(struct ice_hw *hw) -{ - static char buf[ICE_ETHTOOL_FWVER_LEN]; - u8 ver, patch; - u32 full_ver; - u16 build; - - full_ver = hw->nvm.oem_ver; - ver = (u8)((full_ver & ICE_OEM_VER_MASK) >> ICE_OEM_VER_SHIFT); - build = (u16)((full_ver & ICE_OEM_VER_BUILD_MASK) >> - ICE_OEM_VER_BUILD_SHIFT); - patch = (u8)(full_ver & ICE_OEM_VER_PATCH_MASK); - - snprintf(buf, sizeof(buf), "%x.%02x 0x%x %d.%d.%d", - (hw->nvm.ver & ICE_NVM_VER_HI_MASK) >> ICE_NVM_VER_HI_SHIFT, - (hw->nvm.ver & ICE_NVM_VER_LO_MASK) >> ICE_NVM_VER_LO_SHIFT, - hw->nvm.eetrack, ver, build, patch); - - return buf; -} - static void ice_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { @@ -1201,13 +1176,13 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags) 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)) { + if (test_bit(ICE_FLAG_FW_LLDP_AGENT, change_flags)) { + if (!test_bit(ICE_FLAG_FW_LLDP_AGENT, pf->flags)) { enum ice_status status; /* Disable FW LLDP engine */ - status = ice_aq_cfg_lldp_mib_change(&pf->hw, false, - NULL); + status = ice_cfg_lldp_mib_change(&pf->hw, false); + /* If unregistering for LLDP events fails, this is * not an error state, as there shouldn't be any * events to respond to. @@ -1273,6 +1248,12 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags) * The FW LLDP engine will now be consuming them. */ ice_cfg_sw_lldp(vsi, false, false); + + /* Register for MIB change events */ + status = ice_cfg_lldp_mib_change(&pf->hw, true); + if (status) + dev_dbg(&pf->pdev->dev, + "Fail to enable MIB change events\n"); } } clear_bit(ICE_FLAG_ETHTOOL_CTXT, pf->flags); @@ -1319,14 +1300,17 @@ ice_get_ethtool_stats(struct net_device *netdev, struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; struct ice_ring *ring; - unsigned int j = 0; + unsigned int j; int i = 0; char *p; + ice_update_pf_stats(pf); + ice_update_vsi_stats(vsi); + for (j = 0; j < ICE_VSI_STATS_LEN; j++) { p = (char *)vsi + ice_gstrings_vsi_stats[j].stat_offset; data[i++] = (ice_gstrings_vsi_stats[j].sizeof_stat == - sizeof(u64)) ? *(u64 *)p : *(u32 *)p; + sizeof(u64)) ? *(u64 *)p : *(u32 *)p; } /* populate per queue stats */ @@ -1716,6 +1700,7 @@ ice_get_settings_link_up(struct ethtool_link_ksettings *ks, struct net_device *netdev) { struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_port_info *pi = np->vsi->port_info; struct ethtool_link_ksettings cap_ksettings; struct ice_link_status *link_info; struct ice_vsi *vsi = np->vsi; @@ -2040,6 +2025,33 @@ ice_get_settings_link_up(struct ethtool_link_ksettings *ks, break; } ks->base.duplex = DUPLEX_FULL; + + if (link_info->an_info & ICE_AQ_AN_COMPLETED) + ethtool_link_ksettings_add_link_mode(ks, lp_advertising, + Autoneg); + + /* Set flow control negotiated Rx/Tx pause */ + switch (pi->fc.current_mode) { + case ICE_FC_FULL: + ethtool_link_ksettings_add_link_mode(ks, lp_advertising, Pause); + break; + case ICE_FC_TX_PAUSE: + ethtool_link_ksettings_add_link_mode(ks, lp_advertising, Pause); + ethtool_link_ksettings_add_link_mode(ks, lp_advertising, + Asym_Pause); + break; + case ICE_FC_RX_PAUSE: + ethtool_link_ksettings_add_link_mode(ks, lp_advertising, + Asym_Pause); + break; + case ICE_FC_PFC: + /* fall through */ + default: + ethtool_link_ksettings_del_link_mode(ks, lp_advertising, Pause); + ethtool_link_ksettings_del_link_mode(ks, lp_advertising, + Asym_Pause); + break; + } } /** @@ -2078,9 +2090,12 @@ ice_get_link_ksettings(struct net_device *netdev, struct ice_aqc_get_phy_caps_data *caps; struct ice_link_status *hw_link_info; struct ice_vsi *vsi = np->vsi; + enum ice_status status; + int err = 0; ethtool_link_ksettings_zero_link_mode(ks, supported); ethtool_link_ksettings_zero_link_mode(ks, advertising); + ethtool_link_ksettings_zero_link_mode(ks, lp_advertising); hw_link_info = &vsi->port_info->phy.link_info; /* set speed and duplex */ @@ -2125,48 +2140,36 @@ ice_get_link_ksettings(struct net_device *netdev, /* flow control is symmetric and always supported */ ethtool_link_ksettings_add_link_mode(ks, supported, Pause); - switch (vsi->port_info->fc.req_mode) { - case ICE_FC_FULL: + caps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*caps), GFP_KERNEL); + if (!caps) + return -ENOMEM; + + status = ice_aq_get_phy_caps(vsi->port_info, false, + ICE_AQC_REPORT_SW_CFG, caps, NULL); + if (status) { + err = -EIO; + goto done; + } + + /* Set the advertised flow control based on the PHY capability */ + if ((caps->caps & ICE_AQC_PHY_EN_TX_LINK_PAUSE) && + (caps->caps & ICE_AQC_PHY_EN_RX_LINK_PAUSE)) { ethtool_link_ksettings_add_link_mode(ks, advertising, Pause); - break; - case ICE_FC_TX_PAUSE: ethtool_link_ksettings_add_link_mode(ks, advertising, Asym_Pause); - break; - case ICE_FC_RX_PAUSE: + } else if (caps->caps & ICE_AQC_PHY_EN_TX_LINK_PAUSE) { + ethtool_link_ksettings_add_link_mode(ks, advertising, + Asym_Pause); + } else if (caps->caps & ICE_AQC_PHY_EN_RX_LINK_PAUSE) { ethtool_link_ksettings_add_link_mode(ks, advertising, Pause); ethtool_link_ksettings_add_link_mode(ks, advertising, Asym_Pause); - break; - case ICE_FC_PFC: - default: + } else { ethtool_link_ksettings_del_link_mode(ks, advertising, Pause); ethtool_link_ksettings_del_link_mode(ks, advertising, Asym_Pause); - 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); @@ -2178,9 +2181,25 @@ ice_get_link_ksettings(struct net_device *netdev, caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_544_REQ) ethtool_link_ksettings_add_link_mode(ks, advertising, FEC_RS); + status = ice_aq_get_phy_caps(vsi->port_info, false, + ICE_AQC_REPORT_TOPO_CAP, caps, NULL); + if (status) { + err = -EIO; + goto done; + } + + /* 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); + done: devm_kfree(&vsi->back->pdev->dev, caps); - return 0; + return err; } /** @@ -2763,6 +2782,11 @@ static int ice_nway_reset(struct net_device *netdev) * ice_get_pauseparam - Get Flow Control status * @netdev: network interface device structure * @pause: ethernet pause (flow control) parameters + * + * Get requested flow control status from PHY capability. + * If autoneg is true, then ethtool will send the ETHTOOL_GSET ioctl which + * is handled by ice_get_link_ksettings. ice_get_link_ksettings will report + * the negotiated Rx/Tx pause via lp_advertising. */ static void ice_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) @@ -2816,6 +2840,7 @@ static int ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) { struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_aqc_get_phy_caps_data *pcaps; struct ice_link_status *hw_link_info; struct ice_pf *pf = np->vsi->back; struct ice_dcbx_cfg *dcbx_cfg; @@ -2826,6 +2851,7 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) u8 aq_failures; bool link_up; int err = 0; + u32 is_an; pi = vsi->port_info; hw_link_info = &pi->phy.link_info; @@ -2840,7 +2866,30 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) return -EOPNOTSUPP; } - if (pause->autoneg != (hw_link_info->an_info & ICE_AQ_AN_COMPLETED)) { + /* Get pause param reports configured and negotiated flow control pause + * when ETHTOOL_GLINKSETTINGS is defined. Since ETHTOOL_GLINKSETTINGS is + * defined get pause param pause->autoneg reports SW configured setting, + * so compare pause->autoneg with SW configured to prevent the user from + * using set pause param to chance autoneg. + */ + pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL); + if (!pcaps) + return -ENOMEM; + + /* Get current PHY config */ + status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps, + NULL); + if (status) { + kfree(pcaps); + return -EIO; + } + + is_an = ((pcaps->caps & ICE_AQC_PHY_AN_MODE) ? + AUTONEG_ENABLE : AUTONEG_DISABLE); + + kfree(pcaps); + + if (pause->autoneg != is_an) { netdev_info(netdev, "To change autoneg please use: ethtool -s <dev> autoneg <on|off>\n"); return -EOPNOTSUPP; } @@ -3146,12 +3195,6 @@ __ice_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, if (ice_get_q_coalesce(vsi, ec, q_num)) return -EINVAL; - if (q_num < vsi->num_txq) - ec->tx_max_coalesced_frames_irq = vsi->work_lmt; - - if (q_num < vsi->num_rxq) - ec->rx_max_coalesced_frames_irq = vsi->work_lmt; - return 0; } @@ -3185,25 +3228,25 @@ static int ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, struct ice_ring_container *rc, struct ice_vsi *vsi) { + const char *c_type_str = (c_type == ICE_RX_CONTAINER) ? "rx" : "tx"; + u32 use_adaptive_coalesce, coalesce_usecs; struct ice_pf *pf = vsi->back; u16 itr_setting; if (!rc->ring) return -EINVAL; - itr_setting = rc->itr_setting & ~ICE_ITR_DYNAMIC; - 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); + "Invalid value, %s-usecs-high valid values are 0 (disabled), %d-%d\n", + c_type_str, 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), @@ -3211,60 +3254,60 @@ ice_set_rc_coalesce(enum ice_container_type c_type, struct ethtool_coalesce *ec, pf->hw.intrl_gran)); } - if (ec->rx_coalesce_usecs != itr_setting && - ec->use_adaptive_rx_coalesce) { - netdev_info(vsi->netdev, - "Rx interrupt throttling cannot be changed if adaptive-rx is enabled\n"); - return -EINVAL; - } + use_adaptive_coalesce = ec->use_adaptive_rx_coalesce; + coalesce_usecs = ec->rx_coalesce_usecs; - if (ec->rx_coalesce_usecs > ICE_ITR_MAX) { - netdev_info(vsi->netdev, - "Invalid value, rx-usecs range is 0-%d\n", - ICE_ITR_MAX); - return -EINVAL; - } - - if (ec->use_adaptive_rx_coalesce) { - rc->itr_setting |= ICE_ITR_DYNAMIC; - } else { - rc->itr_setting = ITR_REG_ALIGN(ec->rx_coalesce_usecs); - rc->target_itr = ITR_TO_REG(rc->itr_setting); - } break; case ICE_TX_CONTAINER: if (ec->tx_coalesce_usecs_high) { netdev_info(vsi->netdev, - "setting tx-usecs-high is not supported\n"); + "setting %s-usecs-high is not supported\n", + c_type_str); return -EINVAL; } - if (ec->tx_coalesce_usecs != itr_setting && - ec->use_adaptive_tx_coalesce) { - netdev_info(vsi->netdev, - "Tx interrupt throttling cannot be changed if adaptive-tx is enabled\n"); - return -EINVAL; - } + use_adaptive_coalesce = ec->use_adaptive_tx_coalesce; + coalesce_usecs = ec->tx_coalesce_usecs; - if (ec->tx_coalesce_usecs > ICE_ITR_MAX) { - netdev_info(vsi->netdev, - "Invalid value, tx-usecs range is 0-%d\n", - ICE_ITR_MAX); - return -EINVAL; - } - - if (ec->use_adaptive_tx_coalesce) { - rc->itr_setting |= ICE_ITR_DYNAMIC; - } else { - rc->itr_setting = ITR_REG_ALIGN(ec->tx_coalesce_usecs); - rc->target_itr = ITR_TO_REG(rc->itr_setting); - } break; default: dev_dbg(&pf->pdev->dev, "Invalid container type %d\n", c_type); return -EINVAL; } + itr_setting = rc->itr_setting & ~ICE_ITR_DYNAMIC; + if (coalesce_usecs != itr_setting && use_adaptive_coalesce) { + netdev_info(vsi->netdev, + "%s interrupt throttling cannot be changed if adaptive-%s is enabled\n", + c_type_str, c_type_str); + return -EINVAL; + } + + if (coalesce_usecs > ICE_ITR_MAX) { + netdev_info(vsi->netdev, + "Invalid value, %s-usecs range is 0-%d\n", + c_type_str, ICE_ITR_MAX); + return -EINVAL; + } + + /* hardware only supports an ITR granularity of 2us */ + if (coalesce_usecs % 2 != 0) { + netdev_info(vsi->netdev, + "Invalid value, %s-usecs must be even\n", + c_type_str); + return -EINVAL; + } + + if (use_adaptive_coalesce) { + rc->itr_setting |= ICE_ITR_DYNAMIC; + } else { + /* store user facing value how it was set */ + rc->itr_setting = coalesce_usecs; + /* set to static and convert to value HW understands */ + rc->target_itr = + ITR_TO_REG(ITR_REG_ALIGN(rc->itr_setting)); + } + return 0; } @@ -3331,17 +3374,13 @@ __ice_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec, if (ice_set_q_coalesce(vsi, ec, i)) return -EINVAL; } - goto set_work_lmt; + goto set_complete; } 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); +set_complete: return 0; } @@ -3396,6 +3435,33 @@ static const struct ethtool_ops ice_ethtool_ops = { .set_fecparam = ice_set_fecparam, }; +static const struct ethtool_ops ice_ethtool_safe_mode_ops = { + .get_link_ksettings = ice_get_link_ksettings, + .set_link_ksettings = ice_set_link_ksettings, + .get_drvinfo = ice_get_drvinfo, + .get_regs_len = ice_get_regs_len, + .get_regs = ice_get_regs, + .get_msglevel = ice_get_msglevel, + .set_msglevel = ice_set_msglevel, + .get_eeprom_len = ice_get_eeprom_len, + .get_eeprom = ice_get_eeprom, + .get_strings = ice_get_strings, + .get_ethtool_stats = ice_get_ethtool_stats, + .get_sset_count = ice_get_sset_count, + .get_ringparam = ice_get_ringparam, + .set_ringparam = ice_set_ringparam, + .nway_reset = ice_nway_reset, +}; + +/** + * ice_set_ethtool_safe_mode_ops - setup safe mode ethtool ops + * @netdev: network interface device structure + */ +void ice_set_ethtool_safe_mode_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &ice_ethtool_safe_mode_ops; +} + /** * ice_set_ethtool_ops - setup netdev ethtool ops * @netdev: network interface device structure diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c new file mode 100644 index 000000000000..cbd53b586c36 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -0,0 +1,1549 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2019, Intel Corporation. */ + +#include "ice_common.h" +#include "ice_flex_pipe.h" + +/** + * ice_pkg_val_buf + * @buf: pointer to the ice buffer + * + * This helper function validates a buffer's header. + */ +static struct ice_buf_hdr *ice_pkg_val_buf(struct ice_buf *buf) +{ + struct ice_buf_hdr *hdr; + u16 section_count; + u16 data_end; + + hdr = (struct ice_buf_hdr *)buf->buf; + /* verify data */ + section_count = le16_to_cpu(hdr->section_count); + if (section_count < ICE_MIN_S_COUNT || section_count > ICE_MAX_S_COUNT) + return NULL; + + data_end = le16_to_cpu(hdr->data_end); + if (data_end < ICE_MIN_S_DATA_END || data_end > ICE_MAX_S_DATA_END) + return NULL; + + return hdr; +} + +/** + * ice_find_buf_table + * @ice_seg: pointer to the ice segment + * + * Returns the address of the buffer table within the ice segment. + */ +static struct ice_buf_table *ice_find_buf_table(struct ice_seg *ice_seg) +{ + struct ice_nvm_table *nvms; + + nvms = (struct ice_nvm_table *) + (ice_seg->device_table + + le32_to_cpu(ice_seg->device_table_count)); + + return (__force struct ice_buf_table *) + (nvms->vers + le32_to_cpu(nvms->table_count)); +} + +/** + * ice_pkg_enum_buf + * @ice_seg: pointer to the ice segment (or NULL on subsequent calls) + * @state: pointer to the enum state + * + * This function will enumerate all the buffers in the ice segment. The first + * call is made with the ice_seg parameter non-NULL; on subsequent calls, + * ice_seg is set to NULL which continues the enumeration. When the function + * returns a NULL pointer, then the end of the buffers has been reached, or an + * unexpected value has been detected (for example an invalid section count or + * an invalid buffer end value). + */ +static struct ice_buf_hdr * +ice_pkg_enum_buf(struct ice_seg *ice_seg, struct ice_pkg_enum *state) +{ + if (ice_seg) { + state->buf_table = ice_find_buf_table(ice_seg); + if (!state->buf_table) + return NULL; + + state->buf_idx = 0; + return ice_pkg_val_buf(state->buf_table->buf_array); + } + + if (++state->buf_idx < le32_to_cpu(state->buf_table->buf_count)) + return ice_pkg_val_buf(state->buf_table->buf_array + + state->buf_idx); + else + return NULL; +} + +/** + * ice_pkg_advance_sect + * @ice_seg: pointer to the ice segment (or NULL on subsequent calls) + * @state: pointer to the enum state + * + * This helper function will advance the section within the ice segment, + * also advancing the buffer if needed. + */ +static bool +ice_pkg_advance_sect(struct ice_seg *ice_seg, struct ice_pkg_enum *state) +{ + if (!ice_seg && !state->buf) + return false; + + if (!ice_seg && state->buf) + if (++state->sect_idx < le16_to_cpu(state->buf->section_count)) + return true; + + state->buf = ice_pkg_enum_buf(ice_seg, state); + if (!state->buf) + return false; + + /* start of new buffer, reset section index */ + state->sect_idx = 0; + return true; +} + +/** + * ice_pkg_enum_section + * @ice_seg: pointer to the ice segment (or NULL on subsequent calls) + * @state: pointer to the enum state + * @sect_type: section type to enumerate + * + * This function will enumerate all the sections of a particular type in the + * ice segment. The first call is made with the ice_seg parameter non-NULL; + * on subsequent calls, ice_seg is set to NULL which continues the enumeration. + * When the function returns a NULL pointer, then the end of the matching + * sections has been reached. + */ +static void * +ice_pkg_enum_section(struct ice_seg *ice_seg, struct ice_pkg_enum *state, + u32 sect_type) +{ + u16 offset, size; + + if (ice_seg) + state->type = sect_type; + + if (!ice_pkg_advance_sect(ice_seg, state)) + return NULL; + + /* scan for next matching section */ + while (state->buf->section_entry[state->sect_idx].type != + cpu_to_le32(state->type)) + if (!ice_pkg_advance_sect(NULL, state)) + return NULL; + + /* validate section */ + offset = le16_to_cpu(state->buf->section_entry[state->sect_idx].offset); + if (offset < ICE_MIN_S_OFF || offset > ICE_MAX_S_OFF) + return NULL; + + size = le16_to_cpu(state->buf->section_entry[state->sect_idx].size); + if (size < ICE_MIN_S_SZ || size > ICE_MAX_S_SZ) + return NULL; + + /* make sure the section fits in the buffer */ + if (offset + size > ICE_PKG_BUF_SIZE) + return NULL; + + state->sect_type = + le32_to_cpu(state->buf->section_entry[state->sect_idx].type); + + /* calc pointer to this section */ + state->sect = ((u8 *)state->buf) + + le16_to_cpu(state->buf->section_entry[state->sect_idx].offset); + + return state->sect; +} + +/** + * ice_acquire_global_cfg_lock + * @hw: pointer to the HW structure + * @access: access type (read or write) + * + * This function will request ownership of the global config lock for reading + * or writing of the package. When attempting to obtain write access, the + * caller must check for the following two return values: + * + * ICE_SUCCESS - Means the caller has acquired the global config lock + * and can perform writing of the package. + * ICE_ERR_AQ_NO_WORK - Indicates another driver has already written the + * package or has found that no update was necessary; in + * this case, the caller can just skip performing any + * update of the package. + */ +static enum ice_status +ice_acquire_global_cfg_lock(struct ice_hw *hw, + enum ice_aq_res_access_type access) +{ + enum ice_status status; + + status = ice_acquire_res(hw, ICE_GLOBAL_CFG_LOCK_RES_ID, access, + ICE_GLOBAL_CFG_LOCK_TIMEOUT); + + if (!status) + mutex_lock(&ice_global_cfg_lock_sw); + else if (status == ICE_ERR_AQ_NO_WORK) + ice_debug(hw, ICE_DBG_PKG, + "Global config lock: No work to do\n"); + + return status; +} + +/** + * ice_release_global_cfg_lock + * @hw: pointer to the HW structure + * + * This function will release the global config lock. + */ +static void ice_release_global_cfg_lock(struct ice_hw *hw) +{ + mutex_unlock(&ice_global_cfg_lock_sw); + ice_release_res(hw, ICE_GLOBAL_CFG_LOCK_RES_ID); +} + +/** + * ice_aq_download_pkg + * @hw: pointer to the hardware structure + * @pkg_buf: the package buffer to transfer + * @buf_size: the size of the package buffer + * @last_buf: last buffer indicator + * @error_offset: returns error offset + * @error_info: returns error information + * @cd: pointer to command details structure or NULL + * + * Download Package (0x0C40) + */ +static enum ice_status +ice_aq_download_pkg(struct ice_hw *hw, struct ice_buf_hdr *pkg_buf, + u16 buf_size, bool last_buf, u32 *error_offset, + u32 *error_info, struct ice_sq_cd *cd) +{ + struct ice_aqc_download_pkg *cmd; + struct ice_aq_desc desc; + enum ice_status status; + + if (error_offset) + *error_offset = 0; + if (error_info) + *error_info = 0; + + cmd = &desc.params.download_pkg; + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_download_pkg); + desc.flags |= cpu_to_le16(ICE_AQ_FLAG_RD); + + if (last_buf) + cmd->flags |= ICE_AQC_DOWNLOAD_PKG_LAST_BUF; + + status = ice_aq_send_cmd(hw, &desc, pkg_buf, buf_size, cd); + if (status == ICE_ERR_AQ_ERROR) { + /* Read error from buffer only when the FW returned an error */ + struct ice_aqc_download_pkg_resp *resp; + + resp = (struct ice_aqc_download_pkg_resp *)pkg_buf; + if (error_offset) + *error_offset = le32_to_cpu(resp->error_offset); + if (error_info) + *error_info = le32_to_cpu(resp->error_info); + } + + return status; +} + +/** + * ice_find_seg_in_pkg + * @hw: pointer to the hardware structure + * @seg_type: the segment type to search for (i.e., SEGMENT_TYPE_CPK) + * @pkg_hdr: pointer to the package header to be searched + * + * This function searches a package file for a particular segment type. On + * success it returns a pointer to the segment header, otherwise it will + * return NULL. + */ +static struct ice_generic_seg_hdr * +ice_find_seg_in_pkg(struct ice_hw *hw, u32 seg_type, + struct ice_pkg_hdr *pkg_hdr) +{ + u32 i; + + ice_debug(hw, ICE_DBG_PKG, "Package format version: %d.%d.%d.%d\n", + pkg_hdr->format_ver.major, pkg_hdr->format_ver.minor, + pkg_hdr->format_ver.update, pkg_hdr->format_ver.draft); + + /* Search all package segments for the requested segment type */ + for (i = 0; i < le32_to_cpu(pkg_hdr->seg_count); i++) { + struct ice_generic_seg_hdr *seg; + + seg = (struct ice_generic_seg_hdr *) + ((u8 *)pkg_hdr + le32_to_cpu(pkg_hdr->seg_offset[i])); + + if (le32_to_cpu(seg->seg_type) == seg_type) + return seg; + } + + return NULL; +} + +/** + * ice_dwnld_cfg_bufs + * @hw: pointer to the hardware structure + * @bufs: pointer to an array of buffers + * @count: the number of buffers in the array + * + * Obtains global config lock and downloads the package configuration buffers + * to the firmware. Metadata buffers are skipped, and the first metadata buffer + * found indicates that the rest of the buffers are all metadata buffers. + */ +static enum ice_status +ice_dwnld_cfg_bufs(struct ice_hw *hw, struct ice_buf *bufs, u32 count) +{ + enum ice_status status; + struct ice_buf_hdr *bh; + u32 offset, info, i; + + if (!bufs || !count) + return ICE_ERR_PARAM; + + /* If the first buffer's first section has its metadata bit set + * then there are no buffers to be downloaded, and the operation is + * considered a success. + */ + bh = (struct ice_buf_hdr *)bufs; + if (le32_to_cpu(bh->section_entry[0].type) & ICE_METADATA_BUF) + return 0; + + /* reset pkg_dwnld_status in case this function is called in the + * reset/rebuild flow + */ + hw->pkg_dwnld_status = ICE_AQ_RC_OK; + + status = ice_acquire_global_cfg_lock(hw, ICE_RES_WRITE); + if (status) { + if (status == ICE_ERR_AQ_NO_WORK) + hw->pkg_dwnld_status = ICE_AQ_RC_EEXIST; + else + hw->pkg_dwnld_status = hw->adminq.sq_last_status; + return status; + } + + for (i = 0; i < count; i++) { + bool last = ((i + 1) == count); + + if (!last) { + /* check next buffer for metadata flag */ + bh = (struct ice_buf_hdr *)(bufs + i + 1); + + /* A set metadata flag in the next buffer will signal + * that the current buffer will be the last buffer + * downloaded + */ + if (le16_to_cpu(bh->section_count)) + if (le32_to_cpu(bh->section_entry[0].type) & + ICE_METADATA_BUF) + last = true; + } + + bh = (struct ice_buf_hdr *)(bufs + i); + + status = ice_aq_download_pkg(hw, bh, ICE_PKG_BUF_SIZE, last, + &offset, &info, NULL); + + /* Save AQ status from download package */ + hw->pkg_dwnld_status = hw->adminq.sq_last_status; + if (status) { + ice_debug(hw, ICE_DBG_PKG, + "Pkg download failed: err %d off %d inf %d\n", + status, offset, info); + + break; + } + + if (last) + break; + } + + ice_release_global_cfg_lock(hw); + + return status; +} + +/** + * ice_aq_get_pkg_info_list + * @hw: pointer to the hardware structure + * @pkg_info: the buffer which will receive the information list + * @buf_size: the size of the pkg_info information buffer + * @cd: pointer to command details structure or NULL + * + * Get Package Info List (0x0C43) + */ +static enum ice_status +ice_aq_get_pkg_info_list(struct ice_hw *hw, + struct ice_aqc_get_pkg_info_resp *pkg_info, + u16 buf_size, struct ice_sq_cd *cd) +{ + struct ice_aq_desc desc; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_get_pkg_info_list); + + return ice_aq_send_cmd(hw, &desc, pkg_info, buf_size, cd); +} + +/** + * ice_download_pkg + * @hw: pointer to the hardware structure + * @ice_seg: pointer to the segment of the package to be downloaded + * + * Handles the download of a complete package. + */ +static enum ice_status +ice_download_pkg(struct ice_hw *hw, struct ice_seg *ice_seg) +{ + struct ice_buf_table *ice_buf_tbl; + + ice_debug(hw, ICE_DBG_PKG, "Segment version: %d.%d.%d.%d\n", + ice_seg->hdr.seg_ver.major, ice_seg->hdr.seg_ver.minor, + ice_seg->hdr.seg_ver.update, ice_seg->hdr.seg_ver.draft); + + ice_debug(hw, ICE_DBG_PKG, "Seg: type 0x%X, size %d, name %s\n", + le32_to_cpu(ice_seg->hdr.seg_type), + le32_to_cpu(ice_seg->hdr.seg_size), ice_seg->hdr.seg_name); + + ice_buf_tbl = ice_find_buf_table(ice_seg); + + ice_debug(hw, ICE_DBG_PKG, "Seg buf count: %d\n", + le32_to_cpu(ice_buf_tbl->buf_count)); + + return ice_dwnld_cfg_bufs(hw, ice_buf_tbl->buf_array, + le32_to_cpu(ice_buf_tbl->buf_count)); +} + +/** + * ice_init_pkg_info + * @hw: pointer to the hardware structure + * @pkg_hdr: pointer to the driver's package hdr + * + * Saves off the package details into the HW structure. + */ +static enum ice_status +ice_init_pkg_info(struct ice_hw *hw, struct ice_pkg_hdr *pkg_hdr) +{ + struct ice_global_metadata_seg *meta_seg; + struct ice_generic_seg_hdr *seg_hdr; + + if (!pkg_hdr) + return ICE_ERR_PARAM; + + meta_seg = (struct ice_global_metadata_seg *) + ice_find_seg_in_pkg(hw, SEGMENT_TYPE_METADATA, pkg_hdr); + if (meta_seg) { + hw->pkg_ver = meta_seg->pkg_ver; + memcpy(hw->pkg_name, meta_seg->pkg_name, sizeof(hw->pkg_name)); + + ice_debug(hw, ICE_DBG_PKG, "Pkg: %d.%d.%d.%d, %s\n", + meta_seg->pkg_ver.major, meta_seg->pkg_ver.minor, + meta_seg->pkg_ver.update, meta_seg->pkg_ver.draft, + meta_seg->pkg_name); + } else { + ice_debug(hw, ICE_DBG_INIT, + "Did not find metadata segment in driver package\n"); + return ICE_ERR_CFG; + } + + seg_hdr = ice_find_seg_in_pkg(hw, SEGMENT_TYPE_ICE, pkg_hdr); + if (seg_hdr) { + hw->ice_pkg_ver = seg_hdr->seg_ver; + memcpy(hw->ice_pkg_name, seg_hdr->seg_name, + sizeof(hw->ice_pkg_name)); + + ice_debug(hw, ICE_DBG_PKG, "Ice Pkg: %d.%d.%d.%d, %s\n", + seg_hdr->seg_ver.major, seg_hdr->seg_ver.minor, + seg_hdr->seg_ver.update, seg_hdr->seg_ver.draft, + seg_hdr->seg_name); + } else { + ice_debug(hw, ICE_DBG_INIT, + "Did not find ice segment in driver package\n"); + return ICE_ERR_CFG; + } + + return 0; +} + +/** + * ice_get_pkg_info + * @hw: pointer to the hardware structure + * + * Store details of the package currently loaded in HW into the HW structure. + */ +static enum ice_status ice_get_pkg_info(struct ice_hw *hw) +{ + struct ice_aqc_get_pkg_info_resp *pkg_info; + enum ice_status status; + u16 size; + u32 i; + + size = sizeof(*pkg_info) + (sizeof(pkg_info->pkg_info[0]) * + (ICE_PKG_CNT - 1)); + pkg_info = kzalloc(size, GFP_KERNEL); + if (!pkg_info) + return ICE_ERR_NO_MEMORY; + + status = ice_aq_get_pkg_info_list(hw, pkg_info, size, NULL); + if (status) + goto init_pkg_free_alloc; + + for (i = 0; i < le32_to_cpu(pkg_info->count); i++) { +#define ICE_PKG_FLAG_COUNT 4 + char flags[ICE_PKG_FLAG_COUNT + 1] = { 0 }; + u8 place = 0; + + if (pkg_info->pkg_info[i].is_active) { + flags[place++] = 'A'; + hw->active_pkg_ver = pkg_info->pkg_info[i].ver; + memcpy(hw->active_pkg_name, + pkg_info->pkg_info[i].name, + sizeof(hw->active_pkg_name)); + hw->active_pkg_in_nvm = pkg_info->pkg_info[i].is_in_nvm; + } + if (pkg_info->pkg_info[i].is_active_at_boot) + flags[place++] = 'B'; + if (pkg_info->pkg_info[i].is_modified) + flags[place++] = 'M'; + if (pkg_info->pkg_info[i].is_in_nvm) + flags[place++] = 'N'; + + ice_debug(hw, ICE_DBG_PKG, "Pkg[%d]: %d.%d.%d.%d,%s,%s\n", + i, pkg_info->pkg_info[i].ver.major, + pkg_info->pkg_info[i].ver.minor, + pkg_info->pkg_info[i].ver.update, + pkg_info->pkg_info[i].ver.draft, + pkg_info->pkg_info[i].name, flags); + } + +init_pkg_free_alloc: + kfree(pkg_info); + + return status; +} + +/** + * ice_verify_pkg - verify package + * @pkg: pointer to the package buffer + * @len: size of the package buffer + * + * Verifies various attributes of the package file, including length, format + * version, and the requirement of at least one segment. + */ +static enum ice_status ice_verify_pkg(struct ice_pkg_hdr *pkg, u32 len) +{ + u32 seg_count; + u32 i; + + if (len < sizeof(*pkg)) + return ICE_ERR_BUF_TOO_SHORT; + + if (pkg->format_ver.major != ICE_PKG_FMT_VER_MAJ || + pkg->format_ver.minor != ICE_PKG_FMT_VER_MNR || + pkg->format_ver.update != ICE_PKG_FMT_VER_UPD || + pkg->format_ver.draft != ICE_PKG_FMT_VER_DFT) + return ICE_ERR_CFG; + + /* pkg must have at least one segment */ + seg_count = le32_to_cpu(pkg->seg_count); + if (seg_count < 1) + return ICE_ERR_CFG; + + /* make sure segment array fits in package length */ + if (len < sizeof(*pkg) + ((seg_count - 1) * sizeof(pkg->seg_offset))) + return ICE_ERR_BUF_TOO_SHORT; + + /* all segments must fit within length */ + for (i = 0; i < seg_count; i++) { + u32 off = le32_to_cpu(pkg->seg_offset[i]); + struct ice_generic_seg_hdr *seg; + + /* segment header must fit */ + if (len < off + sizeof(*seg)) + return ICE_ERR_BUF_TOO_SHORT; + + seg = (struct ice_generic_seg_hdr *)((u8 *)pkg + off); + + /* segment body must fit */ + if (len < off + le32_to_cpu(seg->seg_size)) + return ICE_ERR_BUF_TOO_SHORT; + } + + return 0; +} + +/** + * ice_free_seg - free package segment pointer + * @hw: pointer to the hardware structure + * + * Frees the package segment pointer in the proper manner, depending on if the + * segment was allocated or just the passed in pointer was stored. + */ +void ice_free_seg(struct ice_hw *hw) +{ + if (hw->pkg_copy) { + devm_kfree(ice_hw_to_dev(hw), hw->pkg_copy); + hw->pkg_copy = NULL; + hw->pkg_size = 0; + } + hw->seg = NULL; +} + +/** + * ice_init_pkg_regs - initialize additional package registers + * @hw: pointer to the hardware structure + */ +static void ice_init_pkg_regs(struct ice_hw *hw) +{ +#define ICE_SW_BLK_INP_MASK_L 0xFFFFFFFF +#define ICE_SW_BLK_INP_MASK_H 0x0000FFFF +#define ICE_SW_BLK_IDX 0 + + /* setup Switch block input mask, which is 48-bits in two parts */ + wr32(hw, GL_PREEXT_L2_PMASK0(ICE_SW_BLK_IDX), ICE_SW_BLK_INP_MASK_L); + wr32(hw, GL_PREEXT_L2_PMASK1(ICE_SW_BLK_IDX), ICE_SW_BLK_INP_MASK_H); +} + +/** + * ice_chk_pkg_version - check package version for compatibility with driver + * @pkg_ver: pointer to a version structure to check + * + * Check to make sure that the package about to be downloaded is compatible with + * the driver. To be compatible, the major and minor components of the package + * version must match our ICE_PKG_SUPP_VER_MAJ and ICE_PKG_SUPP_VER_MNR + * definitions. + */ +static enum ice_status ice_chk_pkg_version(struct ice_pkg_ver *pkg_ver) +{ + if (pkg_ver->major != ICE_PKG_SUPP_VER_MAJ || + pkg_ver->minor != ICE_PKG_SUPP_VER_MNR) + return ICE_ERR_NOT_SUPPORTED; + + return 0; +} + +/** + * ice_init_pkg - initialize/download package + * @hw: pointer to the hardware structure + * @buf: pointer to the package buffer + * @len: size of the package buffer + * + * This function initializes a package. The package contains HW tables + * required to do packet processing. First, the function extracts package + * information such as version. Then it finds the ice configuration segment + * within the package; this function then saves a copy of the segment pointer + * within the supplied package buffer. Next, the function will cache any hints + * from the package, followed by downloading the package itself. Note, that if + * a previous PF driver has already downloaded the package successfully, then + * the current driver will not have to download the package again. + * + * The local package contents will be used to query default behavior and to + * update specific sections of the HW's version of the package (e.g. to update + * the parse graph to understand new protocols). + * + * This function stores a pointer to the package buffer memory, and it is + * expected that the supplied buffer will not be freed immediately. If the + * package buffer needs to be freed, such as when read from a file, use + * ice_copy_and_init_pkg() instead of directly calling ice_init_pkg() in this + * case. + */ +enum ice_status ice_init_pkg(struct ice_hw *hw, u8 *buf, u32 len) +{ + struct ice_pkg_hdr *pkg; + enum ice_status status; + struct ice_seg *seg; + + if (!buf || !len) + return ICE_ERR_PARAM; + + pkg = (struct ice_pkg_hdr *)buf; + status = ice_verify_pkg(pkg, len); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "failed to verify pkg (err: %d)\n", + status); + return status; + } + + /* initialize package info */ + status = ice_init_pkg_info(hw, pkg); + if (status) + return status; + + /* before downloading the package, check package version for + * compatibility with driver + */ + status = ice_chk_pkg_version(&hw->pkg_ver); + if (status) + return status; + + /* find segment in given package */ + seg = (struct ice_seg *)ice_find_seg_in_pkg(hw, SEGMENT_TYPE_ICE, pkg); + if (!seg) { + ice_debug(hw, ICE_DBG_INIT, "no ice segment in package.\n"); + return ICE_ERR_CFG; + } + + /* download package */ + status = ice_download_pkg(hw, seg); + if (status == ICE_ERR_AQ_NO_WORK) { + ice_debug(hw, ICE_DBG_INIT, + "package previously loaded - no work.\n"); + status = 0; + } + + /* Get information on the package currently loaded in HW, then make sure + * the driver is compatible with this version. + */ + if (!status) { + status = ice_get_pkg_info(hw); + if (!status) + status = ice_chk_pkg_version(&hw->active_pkg_ver); + } + + if (!status) { + hw->seg = seg; + /* on successful package download update other required + * registers to support the package and fill HW tables + * with package content. + */ + ice_init_pkg_regs(hw); + ice_fill_blk_tbls(hw); + } else { + ice_debug(hw, ICE_DBG_INIT, "package load failed, %d\n", + status); + } + + return status; +} + +/** + * ice_copy_and_init_pkg - initialize/download a copy of the package + * @hw: pointer to the hardware structure + * @buf: pointer to the package buffer + * @len: size of the package buffer + * + * This function copies the package buffer, and then calls ice_init_pkg() to + * initialize the copied package contents. + * + * The copying is necessary if the package buffer supplied is constant, or if + * the memory may disappear shortly after calling this function. + * + * If the package buffer resides in the data segment and can be modified, the + * caller is free to use ice_init_pkg() instead of ice_copy_and_init_pkg(). + * + * However, if the package buffer needs to be copied first, such as when being + * read from a file, the caller should use ice_copy_and_init_pkg(). + * + * This function will first copy the package buffer, before calling + * ice_init_pkg(). The caller is free to immediately destroy the original + * package buffer, as the new copy will be managed by this function and + * related routines. + */ +enum ice_status ice_copy_and_init_pkg(struct ice_hw *hw, const u8 *buf, u32 len) +{ + enum ice_status status; + u8 *buf_copy; + + if (!buf || !len) + return ICE_ERR_PARAM; + + buf_copy = devm_kmemdup(ice_hw_to_dev(hw), buf, len, GFP_KERNEL); + + status = ice_init_pkg(hw, buf_copy, len); + if (status) { + /* Free the copy, since we failed to initialize the package */ + devm_kfree(ice_hw_to_dev(hw), buf_copy); + } else { + /* Track the copied pkg so we can free it later */ + hw->pkg_copy = buf_copy; + hw->pkg_size = len; + } + + return status; +} + +/* PTG Management */ + +/** + * ice_ptg_find_ptype - Search for packet type group using packet type (ptype) + * @hw: pointer to the hardware structure + * @blk: HW block + * @ptype: the ptype to search for + * @ptg: pointer to variable that receives the PTG + * + * This function will search the PTGs for a particular ptype, returning the + * PTG ID that contains it through the PTG parameter, with the value of + * ICE_DEFAULT_PTG (0) meaning it is part the default PTG. + */ +static enum ice_status +ice_ptg_find_ptype(struct ice_hw *hw, enum ice_block blk, u16 ptype, u8 *ptg) +{ + if (ptype >= ICE_XLT1_CNT || !ptg) + return ICE_ERR_PARAM; + + *ptg = hw->blk[blk].xlt1.ptypes[ptype].ptg; + return 0; +} + +/** + * ice_ptg_alloc_val - Allocates a new packet type group ID by value + * @hw: pointer to the hardware structure + * @blk: HW block + * @ptg: the PTG to allocate + * + * This function allocates a given packet type group ID specified by the PTG + * parameter. + */ +static void ice_ptg_alloc_val(struct ice_hw *hw, enum ice_block blk, u8 ptg) +{ + hw->blk[blk].xlt1.ptg_tbl[ptg].in_use = true; +} + +/** + * ice_ptg_remove_ptype - Removes ptype from a particular packet type group + * @hw: pointer to the hardware structure + * @blk: HW block + * @ptype: the ptype to remove + * @ptg: the PTG to remove the ptype from + * + * This function will remove the ptype from the specific PTG, and move it to + * the default PTG (ICE_DEFAULT_PTG). + */ +static enum ice_status +ice_ptg_remove_ptype(struct ice_hw *hw, enum ice_block blk, u16 ptype, u8 ptg) +{ + struct ice_ptg_ptype **ch; + struct ice_ptg_ptype *p; + + if (ptype > ICE_XLT1_CNT - 1) + return ICE_ERR_PARAM; + + if (!hw->blk[blk].xlt1.ptg_tbl[ptg].in_use) + return ICE_ERR_DOES_NOT_EXIST; + + /* Should not happen if .in_use is set, bad config */ + if (!hw->blk[blk].xlt1.ptg_tbl[ptg].first_ptype) + return ICE_ERR_CFG; + + /* find the ptype within this PTG, and bypass the link over it */ + p = hw->blk[blk].xlt1.ptg_tbl[ptg].first_ptype; + ch = &hw->blk[blk].xlt1.ptg_tbl[ptg].first_ptype; + while (p) { + if (ptype == (p - hw->blk[blk].xlt1.ptypes)) { + *ch = p->next_ptype; + break; + } + + ch = &p->next_ptype; + p = p->next_ptype; + } + + hw->blk[blk].xlt1.ptypes[ptype].ptg = ICE_DEFAULT_PTG; + hw->blk[blk].xlt1.ptypes[ptype].next_ptype = NULL; + + return 0; +} + +/** + * ice_ptg_add_mv_ptype - Adds/moves ptype to a particular packet type group + * @hw: pointer to the hardware structure + * @blk: HW block + * @ptype: the ptype to add or move + * @ptg: the PTG to add or move the ptype to + * + * This function will either add or move a ptype to a particular PTG depending + * on if the ptype is already part of another group. Note that using a + * a destination PTG ID of ICE_DEFAULT_PTG (0) will move the ptype to the + * default PTG. + */ +static enum ice_status +ice_ptg_add_mv_ptype(struct ice_hw *hw, enum ice_block blk, u16 ptype, u8 ptg) +{ + enum ice_status status; + u8 original_ptg; + + if (ptype > ICE_XLT1_CNT - 1) + return ICE_ERR_PARAM; + + if (!hw->blk[blk].xlt1.ptg_tbl[ptg].in_use && ptg != ICE_DEFAULT_PTG) + return ICE_ERR_DOES_NOT_EXIST; + + status = ice_ptg_find_ptype(hw, blk, ptype, &original_ptg); + if (status) + return status; + + /* Is ptype already in the correct PTG? */ + if (original_ptg == ptg) + return 0; + + /* Remove from original PTG and move back to the default PTG */ + if (original_ptg != ICE_DEFAULT_PTG) + ice_ptg_remove_ptype(hw, blk, ptype, original_ptg); + + /* Moving to default PTG? Then we're done with this request */ + if (ptg == ICE_DEFAULT_PTG) + return 0; + + /* Add ptype to PTG at beginning of list */ + hw->blk[blk].xlt1.ptypes[ptype].next_ptype = + hw->blk[blk].xlt1.ptg_tbl[ptg].first_ptype; + hw->blk[blk].xlt1.ptg_tbl[ptg].first_ptype = + &hw->blk[blk].xlt1.ptypes[ptype]; + + hw->blk[blk].xlt1.ptypes[ptype].ptg = ptg; + hw->blk[blk].xlt1.t[ptype] = ptg; + + return 0; +} + +/* Block / table size info */ +struct ice_blk_size_details { + u16 xlt1; /* # XLT1 entries */ + u16 xlt2; /* # XLT2 entries */ + u16 prof_tcam; /* # profile ID TCAM entries */ + u16 prof_id; /* # profile IDs */ + u8 prof_cdid_bits; /* # CDID one-hot bits used in key */ + u16 prof_redir; /* # profile redirection entries */ + u16 es; /* # extraction sequence entries */ + u16 fvw; /* # field vector words */ + u8 overwrite; /* overwrite existing entries allowed */ + u8 reverse; /* reverse FV order */ +}; + +static const struct ice_blk_size_details blk_sizes[ICE_BLK_COUNT] = { + /** + * Table Definitions + * XLT1 - Number of entries in XLT1 table + * XLT2 - Number of entries in XLT2 table + * TCAM - Number of entries Profile ID TCAM table + * CDID - Control Domain ID of the hardware block + * PRED - Number of entries in the Profile Redirection Table + * FV - Number of entries in the Field Vector + * FVW - Width (in WORDs) of the Field Vector + * OVR - Overwrite existing table entries + * REV - Reverse FV + */ + /* XLT1 , XLT2 ,TCAM, PID,CDID,PRED, FV, FVW */ + /* Overwrite , Reverse FV */ + /* SW */ { ICE_XLT1_CNT, ICE_XLT2_CNT, 512, 256, 0, 256, 256, 48, + false, false }, + /* ACL */ { ICE_XLT1_CNT, ICE_XLT2_CNT, 512, 128, 0, 128, 128, 32, + false, false }, + /* FD */ { ICE_XLT1_CNT, ICE_XLT2_CNT, 512, 128, 0, 128, 128, 24, + false, true }, + /* RSS */ { ICE_XLT1_CNT, ICE_XLT2_CNT, 512, 128, 0, 128, 128, 24, + true, true }, + /* PE */ { ICE_XLT1_CNT, ICE_XLT2_CNT, 64, 32, 0, 32, 32, 24, + false, false }, +}; + +enum ice_sid_all { + ICE_SID_XLT1_OFF = 0, + ICE_SID_XLT2_OFF, + ICE_SID_PR_OFF, + ICE_SID_PR_REDIR_OFF, + ICE_SID_ES_OFF, + ICE_SID_OFF_COUNT, +}; + +/* VSIG Management */ + +/** + * ice_vsig_find_vsi - find a VSIG that contains a specified VSI + * @hw: pointer to the hardware structure + * @blk: HW block + * @vsi: VSI of interest + * @vsig: pointer to receive the VSI group + * + * This function will lookup the VSI entry in the XLT2 list and return + * the VSI group its associated with. + */ +static enum ice_status +ice_vsig_find_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 *vsig) +{ + if (!vsig || vsi >= ICE_MAX_VSI) + return ICE_ERR_PARAM; + + /* As long as there's a default or valid VSIG associated with the input + * VSI, the functions returns a success. Any handling of VSIG will be + * done by the following add, update or remove functions. + */ + *vsig = hw->blk[blk].xlt2.vsis[vsi].vsig; + + return 0; +} + +/** + * ice_vsig_alloc_val - allocate a new VSIG by value + * @hw: pointer to the hardware structure + * @blk: HW block + * @vsig: the VSIG to allocate + * + * This function will allocate a given VSIG specified by the VSIG parameter. + */ +static u16 ice_vsig_alloc_val(struct ice_hw *hw, enum ice_block blk, u16 vsig) +{ + u16 idx = vsig & ICE_VSIG_IDX_M; + + if (!hw->blk[blk].xlt2.vsig_tbl[idx].in_use) { + INIT_LIST_HEAD(&hw->blk[blk].xlt2.vsig_tbl[idx].prop_lst); + hw->blk[blk].xlt2.vsig_tbl[idx].in_use = true; + } + + return ICE_VSIG_VALUE(idx, hw->pf_id); +} + +/** + * ice_vsig_remove_vsi - remove VSI from VSIG + * @hw: pointer to the hardware structure + * @blk: HW block + * @vsi: VSI to remove + * @vsig: VSI group to remove from + * + * The function will remove the input VSI from its VSI group and move it + * to the DEFAULT_VSIG. + */ +static enum ice_status +ice_vsig_remove_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 vsig) +{ + struct ice_vsig_vsi **vsi_head, *vsi_cur, *vsi_tgt; + u16 idx; + + idx = vsig & ICE_VSIG_IDX_M; + + if (vsi >= ICE_MAX_VSI || idx >= ICE_MAX_VSIGS) + return ICE_ERR_PARAM; + + if (!hw->blk[blk].xlt2.vsig_tbl[idx].in_use) + return ICE_ERR_DOES_NOT_EXIST; + + /* entry already in default VSIG, don't have to remove */ + if (idx == ICE_DEFAULT_VSIG) + return 0; + + vsi_head = &hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi; + if (!(*vsi_head)) + return ICE_ERR_CFG; + + vsi_tgt = &hw->blk[blk].xlt2.vsis[vsi]; + vsi_cur = (*vsi_head); + + /* iterate the VSI list, skip over the entry to be removed */ + while (vsi_cur) { + if (vsi_tgt == vsi_cur) { + (*vsi_head) = vsi_cur->next_vsi; + break; + } + vsi_head = &vsi_cur->next_vsi; + vsi_cur = vsi_cur->next_vsi; + } + + /* verify if VSI was removed from group list */ + if (!vsi_cur) + return ICE_ERR_DOES_NOT_EXIST; + + vsi_cur->vsig = ICE_DEFAULT_VSIG; + vsi_cur->changed = 1; + vsi_cur->next_vsi = NULL; + + return 0; +} + +/** + * ice_vsig_add_mv_vsi - add or move a VSI to a VSI group + * @hw: pointer to the hardware structure + * @blk: HW block + * @vsi: VSI to move + * @vsig: destination VSI group + * + * This function will move or add the input VSI to the target VSIG. + * The function will find the original VSIG the VSI belongs to and + * move the entry to the DEFAULT_VSIG, update the original VSIG and + * then move entry to the new VSIG. + */ +static enum ice_status +ice_vsig_add_mv_vsi(struct ice_hw *hw, enum ice_block blk, u16 vsi, u16 vsig) +{ + struct ice_vsig_vsi *tmp; + enum ice_status status; + u16 orig_vsig, idx; + + idx = vsig & ICE_VSIG_IDX_M; + + if (vsi >= ICE_MAX_VSI || idx >= ICE_MAX_VSIGS) + return ICE_ERR_PARAM; + + /* if VSIG not in use and VSIG is not default type this VSIG + * doesn't exist. + */ + if (!hw->blk[blk].xlt2.vsig_tbl[idx].in_use && + vsig != ICE_DEFAULT_VSIG) + return ICE_ERR_DOES_NOT_EXIST; + + status = ice_vsig_find_vsi(hw, blk, vsi, &orig_vsig); + if (status) + return status; + + /* no update required if vsigs match */ + if (orig_vsig == vsig) + return 0; + + if (orig_vsig != ICE_DEFAULT_VSIG) { + /* remove entry from orig_vsig and add to default VSIG */ + status = ice_vsig_remove_vsi(hw, blk, vsi, orig_vsig); + if (status) + return status; + } + + if (idx == ICE_DEFAULT_VSIG) + return 0; + + /* Create VSI entry and add VSIG and prop_mask values */ + hw->blk[blk].xlt2.vsis[vsi].vsig = vsig; + hw->blk[blk].xlt2.vsis[vsi].changed = 1; + + /* Add new entry to the head of the VSIG list */ + tmp = hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi; + hw->blk[blk].xlt2.vsig_tbl[idx].first_vsi = + &hw->blk[blk].xlt2.vsis[vsi]; + hw->blk[blk].xlt2.vsis[vsi].next_vsi = tmp; + hw->blk[blk].xlt2.t[vsi] = vsig; + + return 0; +} + +/* Block / table section IDs */ +static const u32 ice_blk_sids[ICE_BLK_COUNT][ICE_SID_OFF_COUNT] = { + /* SWITCH */ + { ICE_SID_XLT1_SW, + ICE_SID_XLT2_SW, + ICE_SID_PROFID_TCAM_SW, + ICE_SID_PROFID_REDIR_SW, + ICE_SID_FLD_VEC_SW + }, + + /* ACL */ + { ICE_SID_XLT1_ACL, + ICE_SID_XLT2_ACL, + ICE_SID_PROFID_TCAM_ACL, + ICE_SID_PROFID_REDIR_ACL, + ICE_SID_FLD_VEC_ACL + }, + + /* FD */ + { ICE_SID_XLT1_FD, + ICE_SID_XLT2_FD, + ICE_SID_PROFID_TCAM_FD, + ICE_SID_PROFID_REDIR_FD, + ICE_SID_FLD_VEC_FD + }, + + /* RSS */ + { ICE_SID_XLT1_RSS, + ICE_SID_XLT2_RSS, + ICE_SID_PROFID_TCAM_RSS, + ICE_SID_PROFID_REDIR_RSS, + ICE_SID_FLD_VEC_RSS + }, + + /* PE */ + { ICE_SID_XLT1_PE, + ICE_SID_XLT2_PE, + ICE_SID_PROFID_TCAM_PE, + ICE_SID_PROFID_REDIR_PE, + ICE_SID_FLD_VEC_PE + } +}; + +/** + * ice_init_sw_xlt1_db - init software XLT1 database from HW tables + * @hw: pointer to the hardware structure + * @blk: the HW block to initialize + */ +static void ice_init_sw_xlt1_db(struct ice_hw *hw, enum ice_block blk) +{ + u16 pt; + + for (pt = 0; pt < hw->blk[blk].xlt1.count; pt++) { + u8 ptg; + + ptg = hw->blk[blk].xlt1.t[pt]; + if (ptg != ICE_DEFAULT_PTG) { + ice_ptg_alloc_val(hw, blk, ptg); + ice_ptg_add_mv_ptype(hw, blk, pt, ptg); + } + } +} + +/** + * ice_init_sw_xlt2_db - init software XLT2 database from HW tables + * @hw: pointer to the hardware structure + * @blk: the HW block to initialize + */ +static void ice_init_sw_xlt2_db(struct ice_hw *hw, enum ice_block blk) +{ + u16 vsi; + + for (vsi = 0; vsi < hw->blk[blk].xlt2.count; vsi++) { + u16 vsig; + + vsig = hw->blk[blk].xlt2.t[vsi]; + if (vsig) { + ice_vsig_alloc_val(hw, blk, vsig); + ice_vsig_add_mv_vsi(hw, blk, vsi, vsig); + /* no changes at this time, since this has been + * initialized from the original package + */ + hw->blk[blk].xlt2.vsis[vsi].changed = 0; + } + } +} + +/** + * ice_init_sw_db - init software database from HW tables + * @hw: pointer to the hardware structure + */ +static void ice_init_sw_db(struct ice_hw *hw) +{ + u16 i; + + for (i = 0; i < ICE_BLK_COUNT; i++) { + ice_init_sw_xlt1_db(hw, (enum ice_block)i); + ice_init_sw_xlt2_db(hw, (enum ice_block)i); + } +} + +/** + * ice_fill_tbl - Reads content of a single table type into database + * @hw: pointer to the hardware structure + * @block_id: Block ID of the table to copy + * @sid: Section ID of the table to copy + * + * Will attempt to read the entire content of a given table of a single block + * into the driver database. We assume that the buffer will always + * be as large or larger than the data contained in the package. If + * this condition is not met, there is most likely an error in the package + * contents. + */ +static void ice_fill_tbl(struct ice_hw *hw, enum ice_block block_id, u32 sid) +{ + u32 dst_len, sect_len, offset = 0; + struct ice_prof_redir_section *pr; + struct ice_prof_id_section *pid; + struct ice_xlt1_section *xlt1; + struct ice_xlt2_section *xlt2; + struct ice_sw_fv_section *es; + struct ice_pkg_enum state; + u8 *src, *dst; + void *sect; + + /* if the HW segment pointer is null then the first iteration of + * ice_pkg_enum_section() will fail. In this case the HW tables will + * not be filled and return success. + */ + if (!hw->seg) { + ice_debug(hw, ICE_DBG_PKG, "hw->seg is NULL, tables are not filled\n"); + return; + } + + memset(&state, 0, sizeof(state)); + + sect = ice_pkg_enum_section(hw->seg, &state, sid); + + while (sect) { + switch (sid) { + case ICE_SID_XLT1_SW: + case ICE_SID_XLT1_FD: + case ICE_SID_XLT1_RSS: + case ICE_SID_XLT1_ACL: + case ICE_SID_XLT1_PE: + xlt1 = (struct ice_xlt1_section *)sect; + src = xlt1->value; + sect_len = le16_to_cpu(xlt1->count) * + sizeof(*hw->blk[block_id].xlt1.t); + dst = hw->blk[block_id].xlt1.t; + dst_len = hw->blk[block_id].xlt1.count * + sizeof(*hw->blk[block_id].xlt1.t); + break; + case ICE_SID_XLT2_SW: + case ICE_SID_XLT2_FD: + case ICE_SID_XLT2_RSS: + case ICE_SID_XLT2_ACL: + case ICE_SID_XLT2_PE: + xlt2 = (struct ice_xlt2_section *)sect; + src = (__force u8 *)xlt2->value; + sect_len = le16_to_cpu(xlt2->count) * + sizeof(*hw->blk[block_id].xlt2.t); + dst = (u8 *)hw->blk[block_id].xlt2.t; + dst_len = hw->blk[block_id].xlt2.count * + sizeof(*hw->blk[block_id].xlt2.t); + break; + case ICE_SID_PROFID_TCAM_SW: + case ICE_SID_PROFID_TCAM_FD: + case ICE_SID_PROFID_TCAM_RSS: + case ICE_SID_PROFID_TCAM_ACL: + case ICE_SID_PROFID_TCAM_PE: + pid = (struct ice_prof_id_section *)sect; + src = (u8 *)pid->entry; + sect_len = le16_to_cpu(pid->count) * + sizeof(*hw->blk[block_id].prof.t); + dst = (u8 *)hw->blk[block_id].prof.t; + dst_len = hw->blk[block_id].prof.count * + sizeof(*hw->blk[block_id].prof.t); + break; + case ICE_SID_PROFID_REDIR_SW: + case ICE_SID_PROFID_REDIR_FD: + case ICE_SID_PROFID_REDIR_RSS: + case ICE_SID_PROFID_REDIR_ACL: + case ICE_SID_PROFID_REDIR_PE: + pr = (struct ice_prof_redir_section *)sect; + src = pr->redir_value; + sect_len = le16_to_cpu(pr->count) * + sizeof(*hw->blk[block_id].prof_redir.t); + dst = hw->blk[block_id].prof_redir.t; + dst_len = hw->blk[block_id].prof_redir.count * + sizeof(*hw->blk[block_id].prof_redir.t); + break; + case ICE_SID_FLD_VEC_SW: + case ICE_SID_FLD_VEC_FD: + case ICE_SID_FLD_VEC_RSS: + case ICE_SID_FLD_VEC_ACL: + case ICE_SID_FLD_VEC_PE: + es = (struct ice_sw_fv_section *)sect; + src = (u8 *)es->fv; + sect_len = (u32)(le16_to_cpu(es->count) * + hw->blk[block_id].es.fvw) * + sizeof(*hw->blk[block_id].es.t); + dst = (u8 *)hw->blk[block_id].es.t; + dst_len = (u32)(hw->blk[block_id].es.count * + hw->blk[block_id].es.fvw) * + sizeof(*hw->blk[block_id].es.t); + break; + default: + return; + } + + /* if the section offset exceeds destination length, terminate + * table fill. + */ + if (offset > dst_len) + return; + + /* if the sum of section size and offset exceed destination size + * then we are out of bounds of the HW table size for that PF. + * Changing section length to fill the remaining table space + * of that PF. + */ + if ((offset + sect_len) > dst_len) + sect_len = dst_len - offset; + + memcpy(dst + offset, src, sect_len); + offset += sect_len; + sect = ice_pkg_enum_section(NULL, &state, sid); + } +} + +/** + * ice_fill_blk_tbls - Read package context for tables + * @hw: pointer to the hardware structure + * + * Reads the current package contents and populates the driver + * database with the data iteratively for all advanced feature + * blocks. Assume that the HW tables have been allocated. + */ +void ice_fill_blk_tbls(struct ice_hw *hw) +{ + u8 i; + + for (i = 0; i < ICE_BLK_COUNT; i++) { + enum ice_block blk_id = (enum ice_block)i; + + ice_fill_tbl(hw, blk_id, hw->blk[blk_id].xlt1.sid); + ice_fill_tbl(hw, blk_id, hw->blk[blk_id].xlt2.sid); + ice_fill_tbl(hw, blk_id, hw->blk[blk_id].prof.sid); + ice_fill_tbl(hw, blk_id, hw->blk[blk_id].prof_redir.sid); + ice_fill_tbl(hw, blk_id, hw->blk[blk_id].es.sid); + } + + ice_init_sw_db(hw); +} + +/** + * ice_free_hw_tbls - free hardware table memory + * @hw: pointer to the hardware structure + */ +void ice_free_hw_tbls(struct ice_hw *hw) +{ + u8 i; + + for (i = 0; i < ICE_BLK_COUNT; i++) { + hw->blk[i].is_list_init = false; + + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.ptypes); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.ptg_tbl); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt1.t); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt2.t); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt2.vsig_tbl); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].xlt2.vsis); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].prof.t); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].prof_redir.t); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].es.t); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].es.ref_count); + devm_kfree(ice_hw_to_dev(hw), hw->blk[i].es.written); + } + + memset(hw->blk, 0, sizeof(hw->blk)); +} + +/** + * ice_clear_hw_tbls - clear HW tables and flow profiles + * @hw: pointer to the hardware structure + */ +void ice_clear_hw_tbls(struct ice_hw *hw) +{ + u8 i; + + for (i = 0; i < ICE_BLK_COUNT; i++) { + struct ice_prof_redir *prof_redir = &hw->blk[i].prof_redir; + struct ice_prof_tcam *prof = &hw->blk[i].prof; + struct ice_xlt1 *xlt1 = &hw->blk[i].xlt1; + struct ice_xlt2 *xlt2 = &hw->blk[i].xlt2; + struct ice_es *es = &hw->blk[i].es; + + memset(xlt1->ptypes, 0, xlt1->count * sizeof(*xlt1->ptypes)); + memset(xlt1->ptg_tbl, 0, + ICE_MAX_PTGS * sizeof(*xlt1->ptg_tbl)); + memset(xlt1->t, 0, xlt1->count * sizeof(*xlt1->t)); + + memset(xlt2->vsis, 0, xlt2->count * sizeof(*xlt2->vsis)); + memset(xlt2->vsig_tbl, 0, + xlt2->count * sizeof(*xlt2->vsig_tbl)); + memset(xlt2->t, 0, xlt2->count * sizeof(*xlt2->t)); + + memset(prof->t, 0, prof->count * sizeof(*prof->t)); + memset(prof_redir->t, 0, + prof_redir->count * sizeof(*prof_redir->t)); + + memset(es->t, 0, es->count * sizeof(*es->t)); + memset(es->ref_count, 0, es->count * sizeof(*es->ref_count)); + memset(es->written, 0, es->count * sizeof(*es->written)); + } +} + +/** + * ice_init_hw_tbls - init hardware table memory + * @hw: pointer to the hardware structure + */ +enum ice_status ice_init_hw_tbls(struct ice_hw *hw) +{ + u8 i; + + for (i = 0; i < ICE_BLK_COUNT; i++) { + struct ice_prof_redir *prof_redir = &hw->blk[i].prof_redir; + struct ice_prof_tcam *prof = &hw->blk[i].prof; + struct ice_xlt1 *xlt1 = &hw->blk[i].xlt1; + struct ice_xlt2 *xlt2 = &hw->blk[i].xlt2; + struct ice_es *es = &hw->blk[i].es; + u16 j; + + if (hw->blk[i].is_list_init) + continue; + + hw->blk[i].is_list_init = true; + + hw->blk[i].overwrite = blk_sizes[i].overwrite; + es->reverse = blk_sizes[i].reverse; + + xlt1->sid = ice_blk_sids[i][ICE_SID_XLT1_OFF]; + xlt1->count = blk_sizes[i].xlt1; + + xlt1->ptypes = devm_kcalloc(ice_hw_to_dev(hw), xlt1->count, + sizeof(*xlt1->ptypes), GFP_KERNEL); + + if (!xlt1->ptypes) + goto err; + + xlt1->ptg_tbl = devm_kcalloc(ice_hw_to_dev(hw), ICE_MAX_PTGS, + sizeof(*xlt1->ptg_tbl), + GFP_KERNEL); + + if (!xlt1->ptg_tbl) + goto err; + + xlt1->t = devm_kcalloc(ice_hw_to_dev(hw), xlt1->count, + sizeof(*xlt1->t), GFP_KERNEL); + if (!xlt1->t) + goto err; + + xlt2->sid = ice_blk_sids[i][ICE_SID_XLT2_OFF]; + xlt2->count = blk_sizes[i].xlt2; + + xlt2->vsis = devm_kcalloc(ice_hw_to_dev(hw), xlt2->count, + sizeof(*xlt2->vsis), GFP_KERNEL); + + if (!xlt2->vsis) + goto err; + + xlt2->vsig_tbl = devm_kcalloc(ice_hw_to_dev(hw), xlt2->count, + sizeof(*xlt2->vsig_tbl), + GFP_KERNEL); + if (!xlt2->vsig_tbl) + goto err; + + for (j = 0; j < xlt2->count; j++) + INIT_LIST_HEAD(&xlt2->vsig_tbl[j].prop_lst); + + xlt2->t = devm_kcalloc(ice_hw_to_dev(hw), xlt2->count, + sizeof(*xlt2->t), GFP_KERNEL); + if (!xlt2->t) + goto err; + + prof->sid = ice_blk_sids[i][ICE_SID_PR_OFF]; + prof->count = blk_sizes[i].prof_tcam; + prof->max_prof_id = blk_sizes[i].prof_id; + prof->cdid_bits = blk_sizes[i].prof_cdid_bits; + prof->t = devm_kcalloc(ice_hw_to_dev(hw), prof->count, + sizeof(*prof->t), GFP_KERNEL); + + if (!prof->t) + goto err; + + prof_redir->sid = ice_blk_sids[i][ICE_SID_PR_REDIR_OFF]; + prof_redir->count = blk_sizes[i].prof_redir; + prof_redir->t = devm_kcalloc(ice_hw_to_dev(hw), + prof_redir->count, + sizeof(*prof_redir->t), + GFP_KERNEL); + + if (!prof_redir->t) + goto err; + + es->sid = ice_blk_sids[i][ICE_SID_ES_OFF]; + es->count = blk_sizes[i].es; + es->fvw = blk_sizes[i].fvw; + es->t = devm_kcalloc(ice_hw_to_dev(hw), + (u32)(es->count * es->fvw), + sizeof(*es->t), GFP_KERNEL); + if (!es->t) + goto err; + + es->ref_count = devm_kcalloc(ice_hw_to_dev(hw), es->count, + sizeof(*es->ref_count), + GFP_KERNEL); + + es->written = devm_kcalloc(ice_hw_to_dev(hw), es->count, + sizeof(*es->written), GFP_KERNEL); + if (!es->ref_count) + goto err; + } + return 0; + +err: + ice_free_hw_tbls(hw); + return ICE_ERR_NO_MEMORY; +} diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h new file mode 100644 index 000000000000..37eb282742d1 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h @@ -0,0 +1,29 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Intel Corporation. */ + +#ifndef _ICE_FLEX_PIPE_H_ +#define _ICE_FLEX_PIPE_H_ + +#include "ice_type.h" + +/* Package minimal version supported */ +#define ICE_PKG_SUPP_VER_MAJ 1 +#define ICE_PKG_SUPP_VER_MNR 3 + +/* Package format version */ +#define ICE_PKG_FMT_VER_MAJ 1 +#define ICE_PKG_FMT_VER_MNR 0 +#define ICE_PKG_FMT_VER_UPD 0 +#define ICE_PKG_FMT_VER_DFT 0 + +#define ICE_PKG_CNT 4 + +enum ice_status ice_init_pkg(struct ice_hw *hw, u8 *buff, u32 len); +enum ice_status +ice_copy_and_init_pkg(struct ice_hw *hw, const u8 *buf, u32 len); +enum ice_status ice_init_hw_tbls(struct ice_hw *hw); +void ice_free_seg(struct ice_hw *hw); +void ice_fill_blk_tbls(struct ice_hw *hw); +void ice_clear_hw_tbls(struct ice_hw *hw); +void ice_free_hw_tbls(struct ice_hw *hw); +#endif /* _ICE_FLEX_PIPE_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h new file mode 100644 index 000000000000..5d5a7eaffa30 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h @@ -0,0 +1,374 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Intel Corporation. */ + +#ifndef _ICE_FLEX_TYPE_H_ +#define _ICE_FLEX_TYPE_H_ +/* Extraction Sequence (Field Vector) Table */ +struct ice_fv_word { + u8 prot_id; + u16 off; /* Offset within the protocol header */ + u8 resvrd; +} __packed; + +#define ICE_MAX_FV_WORDS 48 +struct ice_fv { + struct ice_fv_word ew[ICE_MAX_FV_WORDS]; +}; + +/* Package and segment headers and tables */ +struct ice_pkg_hdr { + struct ice_pkg_ver format_ver; + __le32 seg_count; + __le32 seg_offset[1]; +}; + +/* generic segment */ +struct ice_generic_seg_hdr { +#define SEGMENT_TYPE_METADATA 0x00000001 +#define SEGMENT_TYPE_ICE 0x00000010 + __le32 seg_type; + struct ice_pkg_ver seg_ver; + __le32 seg_size; + char seg_name[ICE_PKG_NAME_SIZE]; +}; + +/* ice specific segment */ + +union ice_device_id { + struct { + __le16 device_id; + __le16 vendor_id; + } dev_vend_id; + __le32 id; +}; + +struct ice_device_id_entry { + union ice_device_id device; + union ice_device_id sub_device; +}; + +struct ice_seg { + struct ice_generic_seg_hdr hdr; + __le32 device_table_count; + struct ice_device_id_entry device_table[1]; +}; + +struct ice_nvm_table { + __le32 table_count; + __le32 vers[1]; +}; + +struct ice_buf { +#define ICE_PKG_BUF_SIZE 4096 + u8 buf[ICE_PKG_BUF_SIZE]; +}; + +struct ice_buf_table { + __le32 buf_count; + struct ice_buf buf_array[1]; +}; + +/* global metadata specific segment */ +struct ice_global_metadata_seg { + struct ice_generic_seg_hdr hdr; + struct ice_pkg_ver pkg_ver; + __le32 track_id; + char pkg_name[ICE_PKG_NAME_SIZE]; +}; + +#define ICE_MIN_S_OFF 12 +#define ICE_MAX_S_OFF 4095 +#define ICE_MIN_S_SZ 1 +#define ICE_MAX_S_SZ 4084 + +/* section information */ +struct ice_section_entry { + __le32 type; + __le16 offset; + __le16 size; +}; + +#define ICE_MIN_S_COUNT 1 +#define ICE_MAX_S_COUNT 511 +#define ICE_MIN_S_DATA_END 12 +#define ICE_MAX_S_DATA_END 4096 + +#define ICE_METADATA_BUF 0x80000000 + +struct ice_buf_hdr { + __le16 section_count; + __le16 data_end; + struct ice_section_entry section_entry[1]; +}; + +#define ICE_MAX_ENTRIES_IN_BUF(hd_sz, ent_sz) ((ICE_PKG_BUF_SIZE - \ + sizeof(struct ice_buf_hdr) - (hd_sz)) / (ent_sz)) + +/* ice package section IDs */ +#define ICE_SID_XLT1_SW 12 +#define ICE_SID_XLT2_SW 13 +#define ICE_SID_PROFID_TCAM_SW 14 +#define ICE_SID_PROFID_REDIR_SW 15 +#define ICE_SID_FLD_VEC_SW 16 + +#define ICE_SID_XLT1_ACL 22 +#define ICE_SID_XLT2_ACL 23 +#define ICE_SID_PROFID_TCAM_ACL 24 +#define ICE_SID_PROFID_REDIR_ACL 25 +#define ICE_SID_FLD_VEC_ACL 26 + +#define ICE_SID_XLT1_FD 32 +#define ICE_SID_XLT2_FD 33 +#define ICE_SID_PROFID_TCAM_FD 34 +#define ICE_SID_PROFID_REDIR_FD 35 +#define ICE_SID_FLD_VEC_FD 36 + +#define ICE_SID_XLT1_RSS 42 +#define ICE_SID_XLT2_RSS 43 +#define ICE_SID_PROFID_TCAM_RSS 44 +#define ICE_SID_PROFID_REDIR_RSS 45 +#define ICE_SID_FLD_VEC_RSS 46 + +#define ICE_SID_RXPARSER_BOOST_TCAM 56 + +#define ICE_SID_XLT1_PE 82 +#define ICE_SID_XLT2_PE 83 +#define ICE_SID_PROFID_TCAM_PE 84 +#define ICE_SID_PROFID_REDIR_PE 85 +#define ICE_SID_FLD_VEC_PE 86 + +/* Label Metadata section IDs */ +#define ICE_SID_LBL_FIRST 0x80000010 +#define ICE_SID_LBL_RXPARSER_TMEM 0x80000018 +/* The following define MUST be updated to reflect the last label section ID */ +#define ICE_SID_LBL_LAST 0x80000038 + +enum ice_block { + ICE_BLK_SW = 0, + ICE_BLK_ACL, + ICE_BLK_FD, + ICE_BLK_RSS, + ICE_BLK_PE, + ICE_BLK_COUNT +}; + +/* package labels */ +struct ice_label { + __le16 value; +#define ICE_PKG_LABEL_SIZE 64 + char name[ICE_PKG_LABEL_SIZE]; +}; + +struct ice_label_section { + __le16 count; + struct ice_label label[1]; +}; + +#define ICE_MAX_LABELS_IN_BUF ICE_MAX_ENTRIES_IN_BUF( \ + sizeof(struct ice_label_section) - sizeof(struct ice_label), \ + sizeof(struct ice_label)) + +struct ice_sw_fv_section { + __le16 count; + __le16 base_offset; + struct ice_fv fv[1]; +}; + +/* The BOOST TCAM stores the match packet header in reverse order, meaning + * the fields are reversed; in addition, this means that the normally big endian + * fields of the packet are now little endian. + */ +struct ice_boost_key_value { +#define ICE_BOOST_REMAINING_HV_KEY 15 + u8 remaining_hv_key[ICE_BOOST_REMAINING_HV_KEY]; + __le16 hv_dst_port_key; + __le16 hv_src_port_key; + u8 tcam_search_key; +} __packed; + +struct ice_boost_key { + struct ice_boost_key_value key; + struct ice_boost_key_value key2; +}; + +/* package Boost TCAM entry */ +struct ice_boost_tcam_entry { + __le16 addr; + __le16 reserved; + /* break up the 40 bytes of key into different fields */ + struct ice_boost_key key; + u8 boost_hit_index_group; + /* The following contains bitfields which are not on byte boundaries. + * These fields are currently unused by driver software. + */ +#define ICE_BOOST_BIT_FIELDS 43 + u8 bit_fields[ICE_BOOST_BIT_FIELDS]; +}; + +struct ice_boost_tcam_section { + __le16 count; + __le16 reserved; + struct ice_boost_tcam_entry tcam[1]; +}; + +#define ICE_MAX_BST_TCAMS_IN_BUF ICE_MAX_ENTRIES_IN_BUF( \ + sizeof(struct ice_boost_tcam_section) - \ + sizeof(struct ice_boost_tcam_entry), \ + sizeof(struct ice_boost_tcam_entry)) + +struct ice_xlt1_section { + __le16 count; + __le16 offset; + u8 value[1]; +} __packed; + +struct ice_xlt2_section { + __le16 count; + __le16 offset; + __le16 value[1]; +}; + +struct ice_prof_redir_section { + __le16 count; + __le16 offset; + u8 redir_value[1]; +}; + +struct ice_pkg_enum { + struct ice_buf_table *buf_table; + u32 buf_idx; + + u32 type; + struct ice_buf_hdr *buf; + u32 sect_idx; + void *sect; + u32 sect_type; + + u32 entry_idx; + void *(*handler)(u32 sect_type, void *section, u32 index, u32 *offset); +}; + +struct ice_es { + u32 sid; + u16 count; + u16 fvw; + u16 *ref_count; + struct list_head prof_map; + struct ice_fv_word *t; + struct mutex prof_map_lock; /* protect access to profiles list */ + u8 *written; + u8 reverse; /* set to true to reverse FV order */ +}; + +/* PTYPE Group management */ + +/* Note: XLT1 table takes 13-bit as input, and results in an 8-bit packet type + * group (PTG) ID as output. + * + * Note: PTG 0 is the default packet type group and it is assumed that all PTYPE + * are a part of this group until moved to a new PTG. + */ +#define ICE_DEFAULT_PTG 0 + +struct ice_ptg_entry { + struct ice_ptg_ptype *first_ptype; + u8 in_use; +}; + +struct ice_ptg_ptype { + struct ice_ptg_ptype *next_ptype; + u8 ptg; +}; + +struct ice_vsig_entry { + struct list_head prop_lst; + struct ice_vsig_vsi *first_vsi; + u8 in_use; +}; + +struct ice_vsig_vsi { + struct ice_vsig_vsi *next_vsi; + u32 prop_mask; + u16 changed; + u16 vsig; +}; + +#define ICE_XLT1_CNT 1024 +#define ICE_MAX_PTGS 256 + +/* XLT1 Table */ +struct ice_xlt1 { + struct ice_ptg_entry *ptg_tbl; + struct ice_ptg_ptype *ptypes; + u8 *t; + u32 sid; + u16 count; +}; + +#define ICE_XLT2_CNT 768 +#define ICE_MAX_VSIGS 768 + +/* VSIG bit layout: + * [0:12]: incremental VSIG index 1 to ICE_MAX_VSIGS + * [13:15]: PF number of device + */ +#define ICE_VSIG_IDX_M (0x1FFF) +#define ICE_PF_NUM_S 13 +#define ICE_PF_NUM_M (0x07 << ICE_PF_NUM_S) +#define ICE_VSIG_VALUE(vsig, pf_id) \ + (u16)((((u16)(vsig)) & ICE_VSIG_IDX_M) | \ + (((u16)(pf_id) << ICE_PF_NUM_S) & ICE_PF_NUM_M)) +#define ICE_DEFAULT_VSIG 0 + +/* XLT2 Table */ +struct ice_xlt2 { + struct ice_vsig_entry *vsig_tbl; + struct ice_vsig_vsi *vsis; + u16 *t; + u32 sid; + u16 count; +}; + +/* Keys are made up of two values, each one-half the size of the key. + * For TCAM, the entire key is 80 bits wide (or 2, 40-bit wide values) + */ +#define ICE_TCAM_KEY_VAL_SZ 5 +#define ICE_TCAM_KEY_SZ (2 * ICE_TCAM_KEY_VAL_SZ) + +struct ice_prof_tcam_entry { + __le16 addr; + u8 key[ICE_TCAM_KEY_SZ]; + u8 prof_id; +} __packed; + +struct ice_prof_id_section { + __le16 count; + struct ice_prof_tcam_entry entry[1]; +} __packed; + +struct ice_prof_tcam { + u32 sid; + u16 count; + u16 max_prof_id; + struct ice_prof_tcam_entry *t; + u8 cdid_bits; /* # CDID bits to use in key, 0, 2, 4, or 8 */ +}; + +struct ice_prof_redir { + u8 *t; + u32 sid; + u16 count; +}; + +/* Tables per block */ +struct ice_blk_info { + struct ice_xlt1 xlt1; + struct ice_xlt2 xlt2; + struct ice_prof_tcam prof; + struct ice_prof_redir prof_redir; + struct ice_es es; + u8 overwrite; /* set to true to allow overwrite of table entries */ + u8 is_list_init; +}; + +#endif /* _ICE_FLEX_TYPE_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h index 6c5ce05742b1..152fbd556e9b 100644 --- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h +++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h @@ -55,6 +55,8 @@ #define PRTDCB_GENS 0x00083020 #define PRTDCB_GENS_DCBX_STATUS_S 0 #define PRTDCB_GENS_DCBX_STATUS_M ICE_M(0x7, 0) +#define GL_PREEXT_L2_PMASK0(_i) (0x0020F0FC + ((_i) * 4)) +#define GL_PREEXT_L2_PMASK1(_i) (0x0020F108 + ((_i) * 4)) #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) @@ -127,8 +129,11 @@ #define GLINT_DYN_CTL_CLEARPBA_M BIT(1) #define GLINT_DYN_CTL_SWINT_TRIG_M BIT(2) #define GLINT_DYN_CTL_ITR_INDX_S 3 +#define GLINT_DYN_CTL_ITR_INDX_M ICE_M(0x3, 3) #define GLINT_DYN_CTL_INTERVAL_S 5 +#define GLINT_DYN_CTL_INTERVAL_M ICE_M(0xFFF, 5) #define GLINT_DYN_CTL_SW_ITR_INDX_M ICE_M(0x3, 25) +#define GLINT_DYN_CTL_WB_ON_ITR_M BIT(30) #define GLINT_DYN_CTL_INTENA_MSK_M BIT(31) #define GLINT_ITR(_i, _INT) (0x00154000 + ((_i) * 8192 + (_INT) * 4)) #define GLINT_RATE(_INT) (0x0015A000 + ((_INT) * 4)) @@ -281,14 +286,10 @@ #define GL_PWR_MODE_CTL 0x000B820C #define GL_PWR_MODE_CTL_CAR_MAX_BW_S 30 #define GL_PWR_MODE_CTL_CAR_MAX_BW_M ICE_M(0x3, 30) -#define GLPRT_BPRCH(_i) (0x00381384 + ((_i) * 8)) #define GLPRT_BPRCL(_i) (0x00381380 + ((_i) * 8)) -#define GLPRT_BPTCH(_i) (0x00381244 + ((_i) * 8)) #define GLPRT_BPTCL(_i) (0x00381240 + ((_i) * 8)) #define GLPRT_CRCERRS(_i) (0x00380100 + ((_i) * 8)) -#define GLPRT_GORCH(_i) (0x00380004 + ((_i) * 8)) #define GLPRT_GORCL(_i) (0x00380000 + ((_i) * 8)) -#define GLPRT_GOTCH(_i) (0x00380B44 + ((_i) * 8)) #define GLPRT_GOTCL(_i) (0x00380B40 + ((_i) * 8)) #define GLPRT_ILLERRC(_i) (0x003801C0 + ((_i) * 8)) #define GLPRT_LXOFFRXC(_i) (0x003802C0 + ((_i) * 8)) @@ -296,38 +297,22 @@ #define GLPRT_LXONRXC(_i) (0x00380280 + ((_i) * 8)) #define GLPRT_LXONTXC(_i) (0x00381140 + ((_i) * 8)) #define GLPRT_MLFC(_i) (0x00380040 + ((_i) * 8)) -#define GLPRT_MPRCH(_i) (0x00381344 + ((_i) * 8)) #define GLPRT_MPRCL(_i) (0x00381340 + ((_i) * 8)) -#define GLPRT_MPTCH(_i) (0x00381204 + ((_i) * 8)) #define GLPRT_MPTCL(_i) (0x00381200 + ((_i) * 8)) #define GLPRT_MRFC(_i) (0x00380080 + ((_i) * 8)) -#define GLPRT_PRC1023H(_i) (0x00380A04 + ((_i) * 8)) #define GLPRT_PRC1023L(_i) (0x00380A00 + ((_i) * 8)) -#define GLPRT_PRC127H(_i) (0x00380944 + ((_i) * 8)) #define GLPRT_PRC127L(_i) (0x00380940 + ((_i) * 8)) -#define GLPRT_PRC1522H(_i) (0x00380A44 + ((_i) * 8)) #define GLPRT_PRC1522L(_i) (0x00380A40 + ((_i) * 8)) -#define GLPRT_PRC255H(_i) (0x00380984 + ((_i) * 8)) #define GLPRT_PRC255L(_i) (0x00380980 + ((_i) * 8)) -#define GLPRT_PRC511H(_i) (0x003809C4 + ((_i) * 8)) #define GLPRT_PRC511L(_i) (0x003809C0 + ((_i) * 8)) -#define GLPRT_PRC64H(_i) (0x00380904 + ((_i) * 8)) #define GLPRT_PRC64L(_i) (0x00380900 + ((_i) * 8)) -#define GLPRT_PRC9522H(_i) (0x00380A84 + ((_i) * 8)) #define GLPRT_PRC9522L(_i) (0x00380A80 + ((_i) * 8)) -#define GLPRT_PTC1023H(_i) (0x00380C84 + ((_i) * 8)) #define GLPRT_PTC1023L(_i) (0x00380C80 + ((_i) * 8)) -#define GLPRT_PTC127H(_i) (0x00380BC4 + ((_i) * 8)) #define GLPRT_PTC127L(_i) (0x00380BC0 + ((_i) * 8)) -#define GLPRT_PTC1522H(_i) (0x00380CC4 + ((_i) * 8)) #define GLPRT_PTC1522L(_i) (0x00380CC0 + ((_i) * 8)) -#define GLPRT_PTC255H(_i) (0x00380C04 + ((_i) * 8)) #define GLPRT_PTC255L(_i) (0x00380C00 + ((_i) * 8)) -#define GLPRT_PTC511H(_i) (0x00380C44 + ((_i) * 8)) #define GLPRT_PTC511L(_i) (0x00380C40 + ((_i) * 8)) -#define GLPRT_PTC64H(_i) (0x00380B84 + ((_i) * 8)) #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)) @@ -340,32 +325,23 @@ #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)) -#define GLPRT_UPTCH(_i) (0x003811C4 + ((_i) * 8)) #define GLPRT_UPTCL(_i) (0x003811C0 + ((_i) * 8)) -#define GLV_BPRCH(_i) (0x003B6004 + ((_i) * 8)) #define GLV_BPRCL(_i) (0x003B6000 + ((_i) * 8)) -#define GLV_BPTCH(_i) (0x0030E004 + ((_i) * 8)) #define GLV_BPTCL(_i) (0x0030E000 + ((_i) * 8)) -#define GLV_GORCH(_i) (0x003B0004 + ((_i) * 8)) #define GLV_GORCL(_i) (0x003B0000 + ((_i) * 8)) -#define GLV_GOTCH(_i) (0x00300004 + ((_i) * 8)) #define GLV_GOTCL(_i) (0x00300000 + ((_i) * 8)) -#define GLV_MPRCH(_i) (0x003B4004 + ((_i) * 8)) #define GLV_MPRCL(_i) (0x003B4000 + ((_i) * 8)) -#define GLV_MPTCH(_i) (0x0030C004 + ((_i) * 8)) #define GLV_MPTCL(_i) (0x0030C000 + ((_i) * 8)) #define GLV_RDPC(_i) (0x00294C04 + ((_i) * 4)) #define GLV_TEPC(_VSI) (0x00312000 + ((_VSI) * 4)) -#define GLV_UPRCH(_i) (0x003B2004 + ((_i) * 8)) #define GLV_UPRCL(_i) (0x003B2000 + ((_i) * 8)) -#define GLV_UPTCH(_i) (0x0030A004 + ((_i) * 8)) #define GLV_UPTCL(_i) (0x0030A000 + ((_i) * 8)) #define PF_VT_PFALLOC_HIF 0x0009DD80 #define VSIQF_HKEY_MAX_INDEX 12 #define VSIQF_HLUT_MAX_INDEX 15 #define VFINT_DYN_CTLN(_i) (0x00003800 + ((_i) * 4)) #define VFINT_DYN_CTLN_CLEARPBA_M BIT(1) +#define PRTRPB_RDPC 0x000AC260 #endif /* _ICE_HW_AUTOGEN_H_ */ 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 510a8c900e61..2aac8f13daeb 100644 --- a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h +++ b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h @@ -290,6 +290,7 @@ struct ice_rlan_ctx { u8 tphdata_ena; u8 tphhead_ena; u16 lrxqthresh; /* bigger than needed, see above for reason */ + u8 prefena; /* NOTE: normally must be set to 1 at init */ }; struct ice_ctx_ele { @@ -427,6 +428,7 @@ struct ice_tlan_ctx { #define ICE_TLAN_CTX_VMVF_TYPE_PF 2 u16 src_vsi; u8 tsyn_ena; + u8 internal_usage_flag; u8 alt_vlan; u16 cpuid; /* bigger than needed, see above for reason */ u8 wb_mode; diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index a19f5920733b..cc755382df25 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -191,41 +191,58 @@ static int ice_pf_rxq_wait(struct ice_pf *pf, int pf_q, bool ena) } /** - * ice_vsi_ctrl_rx_rings - Start or stop a VSI's Rx rings + * ice_vsi_ctrl_rx_ring - Start or stop a VSI's Rx ring * @vsi: the VSI being configured * @ena: start or stop the Rx rings + * @rxq_idx: Rx queue index */ -static int ice_vsi_ctrl_rx_rings(struct ice_vsi *vsi, bool ena) +#ifndef CONFIG_PCI_IOV +static +#endif /* !CONFIG_PCI_IOV */ +int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx) { + int pf_q = vsi->rxq_map[rxq_idx]; struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; - int i, ret = 0; + int ret = 0; + u32 rx_reg; - for (i = 0; i < vsi->num_rxq; i++) { - int pf_q = vsi->rxq_map[i]; - u32 rx_reg; + rx_reg = rd32(hw, QRX_CTRL(pf_q)); - 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)) + return 0; - /* Skip if the queue is already in the requested state */ - if (ena == !!(rx_reg & QRX_CTRL_QENA_STAT_M)) - continue; + /* turn on/off the queue */ + if (ena) + rx_reg |= QRX_CTRL_QENA_REQ_M; + else + rx_reg &= ~QRX_CTRL_QENA_REQ_M; + wr32(hw, QRX_CTRL(pf_q), rx_reg); - /* turn on/off the queue */ - if (ena) - rx_reg |= QRX_CTRL_QENA_REQ_M; - else - rx_reg &= ~QRX_CTRL_QENA_REQ_M; - wr32(hw, QRX_CTRL(pf_q), rx_reg); - - /* wait for the change to finish */ - ret = ice_pf_rxq_wait(pf, pf_q, ena); - if (ret) { - dev_err(&pf->pdev->dev, - "VSI idx %d Rx ring %d %sable timeout\n", - vsi->idx, pf_q, (ena ? "en" : "dis")); + /* wait for the change to finish */ + ret = ice_pf_rxq_wait(pf, pf_q, ena); + if (ret) + dev_err(&pf->pdev->dev, + "VSI idx %d Rx ring %d %sable timeout\n", + vsi->idx, pf_q, (ena ? "en" : "dis")); + + return ret; +} + +/** + * ice_vsi_ctrl_rx_rings - Start or stop a VSI's Rx rings + * @vsi: the VSI being configured + * @ena: start or stop the Rx rings + */ +static int ice_vsi_ctrl_rx_rings(struct ice_vsi *vsi, bool ena) +{ + int i, ret = 0; + + for (i = 0; i < vsi->num_rxq; i++) { + ret = ice_vsi_ctrl_rx_ring(vsi, ena, i); + if (ret) break; - } } return ret; @@ -246,12 +263,24 @@ static int ice_vsi_alloc_arrays(struct ice_vsi *vsi) vsi->tx_rings = devm_kcalloc(&pf->pdev->dev, vsi->alloc_txq, sizeof(*vsi->tx_rings), GFP_KERNEL); if (!vsi->tx_rings) - goto err_txrings; + return -ENOMEM; vsi->rx_rings = devm_kcalloc(&pf->pdev->dev, vsi->alloc_rxq, sizeof(*vsi->rx_rings), GFP_KERNEL); if (!vsi->rx_rings) - goto err_rxrings; + goto err_rings; + + vsi->txq_map = devm_kcalloc(&pf->pdev->dev, vsi->alloc_txq, + sizeof(*vsi->txq_map), GFP_KERNEL); + + if (!vsi->txq_map) + goto err_txq_map; + + vsi->rxq_map = devm_kcalloc(&pf->pdev->dev, vsi->alloc_rxq, + sizeof(*vsi->rxq_map), GFP_KERNEL); + if (!vsi->rxq_map) + goto err_rxq_map; + /* There is no need to allocate q_vectors for a loopback VSI. */ if (vsi->type == ICE_VSI_LB) @@ -266,10 +295,13 @@ static int ice_vsi_alloc_arrays(struct ice_vsi *vsi) return 0; err_vectors: + devm_kfree(&pf->pdev->dev, vsi->rxq_map); +err_rxq_map: + devm_kfree(&pf->pdev->dev, vsi->txq_map); +err_txq_map: devm_kfree(&pf->pdev->dev, vsi->rx_rings); -err_rxrings: +err_rings: devm_kfree(&pf->pdev->dev, vsi->tx_rings); -err_txrings: return -ENOMEM; } @@ -311,9 +343,21 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, u16 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_q_vectors = max_t(int, pf->num_lan_rx, pf->num_lan_tx); + vsi->alloc_txq = min_t(int, ice_get_avail_txq_count(pf), + num_online_cpus()); + + pf->num_lan_tx = vsi->alloc_txq; + + /* only 1 Rx queue unless RSS is enabled */ + if (!test_bit(ICE_FLAG_RSS_ENA, pf->flags)) + vsi->alloc_rxq = 1; + else + vsi->alloc_rxq = min_t(int, ice_get_avail_rxq_count(pf), + num_online_cpus()); + + pf->num_lan_rx = vsi->alloc_rxq; + + vsi->num_q_vectors = max_t(int, vsi->alloc_rxq, vsi->alloc_txq); break; case ICE_VSI_VF: vf = &pf->vf[vsi->vf_id]; @@ -416,6 +460,14 @@ static void ice_vsi_free_arrays(struct ice_vsi *vsi) devm_kfree(&pf->pdev->dev, vsi->rx_rings); vsi->rx_rings = NULL; } + if (vsi->txq_map) { + devm_kfree(&pf->pdev->dev, vsi->txq_map); + vsi->txq_map = NULL; + } + if (vsi->rxq_map) { + devm_kfree(&pf->pdev->dev, vsi->rxq_map); + vsi->rxq_map = NULL; + } } /** @@ -508,8 +560,8 @@ ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type type, u16 vf_id) vsi->type = type; vsi->back = pf; set_bit(__ICE_DOWN, vsi->state); + vsi->idx = pf->next_vsi; - vsi->work_lmt = ICE_DFLT_IRQ_WORK; if (type == ICE_VSI_VF) ice_vsi_set_num_qs(vsi, vf_id); @@ -647,7 +699,7 @@ static int ice_vsi_get_qs(struct ice_vsi *vsi) struct ice_qs_cfg tx_qs_cfg = { .qs_mutex = &pf->avail_q_mutex, .pf_map = pf->avail_txqs, - .pf_map_size = ICE_MAX_TXQS, + .pf_map_size = pf->max_pf_txqs, .q_count = vsi->alloc_txq, .scatter_count = ICE_MAX_SCATTER_TXQS, .vsi_map = vsi->txq_map, @@ -657,7 +709,7 @@ static int ice_vsi_get_qs(struct ice_vsi *vsi) struct ice_qs_cfg rx_qs_cfg = { .qs_mutex = &pf->avail_q_mutex, .pf_map = pf->avail_rxqs, - .pf_map_size = ICE_MAX_RXQS, + .pf_map_size = pf->max_pf_rxqs, .q_count = vsi->alloc_rxq, .scatter_count = ICE_MAX_SCATTER_RXQS, .vsi_map = vsi->rxq_map, @@ -701,6 +753,17 @@ void ice_vsi_put_qs(struct ice_vsi *vsi) } /** + * ice_is_safe_mode + * @pf: pointer to the PF struct + * + * returns true if driver is in safe mode, false otherwise + */ +bool ice_is_safe_mode(struct ice_pf *pf) +{ + return !test_bit(ICE_FLAG_ADV_FEATURES, pf->flags); +} + +/** * ice_rss_clean - Delete RSS related VSI structures that hold user inputs * @vsi: the VSI being removed */ @@ -1010,6 +1073,13 @@ static int ice_vsi_init(struct ice_vsi *vsi) ICE_AQ_VSI_SEC_FLAG_ENA_MAC_ANTI_SPOOF; } + /* Allow control frames out of main VSI */ + if (vsi->type == ICE_VSI_PF) { + ctxt->info.sec_flags |= ICE_AQ_VSI_SEC_FLAG_ALLOW_DEST_OVRD; + ctxt->info.valid_sections |= + cpu_to_le16(ICE_AQ_VSI_PROP_SECURITY_VALID); + } + ret = ice_add_vsi(hw, vsi->idx, ctxt, NULL); if (ret) { dev_err(&pf->pdev->dev, @@ -1129,12 +1199,7 @@ static int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi) return -EEXIST; } - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - num_q_vectors = vsi->num_q_vectors; - } else { - err = -EINVAL; - goto err_out; - } + num_q_vectors = vsi->num_q_vectors; for (v_idx = 0; v_idx < num_q_vectors; v_idx++) { err = ice_vsi_alloc_q_vector(vsi, v_idx); @@ -1180,9 +1245,6 @@ static int ice_vsi_setup_vector_base(struct ice_vsi *vsi) return -EEXIST; } - if (!test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - return -ENOENT; - 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, @@ -1477,40 +1539,32 @@ void ice_update_eth_stats(struct ice_vsi *vsi) prev_es = &vsi->eth_stats_prev; cur_es = &vsi->eth_stats; - ice_stat_update40(hw, GLV_GORCH(vsi_num), GLV_GORCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->rx_bytes, - &cur_es->rx_bytes); + ice_stat_update40(hw, GLV_GORCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->rx_bytes, &cur_es->rx_bytes); - ice_stat_update40(hw, GLV_UPRCH(vsi_num), GLV_UPRCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->rx_unicast, - &cur_es->rx_unicast); + ice_stat_update40(hw, GLV_UPRCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->rx_unicast, &cur_es->rx_unicast); - ice_stat_update40(hw, GLV_MPRCH(vsi_num), GLV_MPRCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->rx_multicast, - &cur_es->rx_multicast); + ice_stat_update40(hw, GLV_MPRCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->rx_multicast, &cur_es->rx_multicast); - ice_stat_update40(hw, GLV_BPRCH(vsi_num), GLV_BPRCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->rx_broadcast, - &cur_es->rx_broadcast); + ice_stat_update40(hw, GLV_BPRCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->rx_broadcast, &cur_es->rx_broadcast); ice_stat_update32(hw, GLV_RDPC(vsi_num), vsi->stat_offsets_loaded, &prev_es->rx_discards, &cur_es->rx_discards); - ice_stat_update40(hw, GLV_GOTCH(vsi_num), GLV_GOTCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->tx_bytes, - &cur_es->tx_bytes); + ice_stat_update40(hw, GLV_GOTCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->tx_bytes, &cur_es->tx_bytes); - ice_stat_update40(hw, GLV_UPTCH(vsi_num), GLV_UPTCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->tx_unicast, - &cur_es->tx_unicast); + ice_stat_update40(hw, GLV_UPTCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->tx_unicast, &cur_es->tx_unicast); - ice_stat_update40(hw, GLV_MPTCH(vsi_num), GLV_MPTCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->tx_multicast, - &cur_es->tx_multicast); + ice_stat_update40(hw, GLV_MPTCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->tx_multicast, &cur_es->tx_multicast); - ice_stat_update40(hw, GLV_BPTCH(vsi_num), GLV_BPTCL(vsi_num), - vsi->stat_offsets_loaded, &prev_es->tx_broadcast, - &cur_es->tx_broadcast); + ice_stat_update40(hw, GLV_BPTCL(vsi_num), vsi->stat_offsets_loaded, + &prev_es->tx_broadcast, &cur_es->tx_broadcast); ice_stat_update32(hw, GLV_TEPC(vsi_num), vsi->stat_offsets_loaded, &prev_es->tx_errors, &cur_es->tx_errors); @@ -1658,6 +1712,62 @@ setup_rings: } /** + * ice_vsi_cfg_txq - Configure single Tx queue + * @vsi: the VSI that queue belongs to + * @ring: Tx ring to be configured + * @tc_q_idx: queue index within given TC + * @qg_buf: queue group buffer + * @tc: TC that Tx ring belongs to + */ +static int +ice_vsi_cfg_txq(struct ice_vsi *vsi, struct ice_ring *ring, u16 tc_q_idx, + struct ice_aqc_add_tx_qgrp *qg_buf, u8 tc) +{ + struct ice_tlan_ctx tlan_ctx = { 0 }; + struct ice_aqc_add_txqs_perq *txq; + struct ice_pf *pf = vsi->back; + u8 buf_len = sizeof(*qg_buf); + enum ice_status status; + u16 pf_q; + + pf_q = ring->reg_idx; + ice_setup_tx_ctx(ring, &tlan_ctx, pf_q); + /* copy context contents into the qg_buf */ + qg_buf->txqs[0].txq_id = cpu_to_le16(pf_q); + ice_set_ctx((u8 *)&tlan_ctx, qg_buf->txqs[0].txq_ctx, + ice_tlan_ctx_info); + + /* init queue specific tail reg. It is referred as + * transmit comm scheduler queue doorbell. + */ + ring->tail = pf->hw.hw_addr + QTX_COMM_DBELL(pf_q); + + /* Add unique software queue handle of the Tx queue per + * TC into the VSI Tx ring + */ + ring->q_handle = tc_q_idx; + + status = ice_ena_vsi_txq(vsi->port_info, vsi->idx, tc, ring->q_handle, + 1, qg_buf, buf_len, NULL); + if (status) { + dev_err(&pf->pdev->dev, + "Failed to set LAN Tx queue context, error: %d\n", + status); + return -ENODEV; + } + + /* Add Tx Queue TEID into the VSI Tx ring from the + * response. This will complete configuring and + * enabling the queue. + */ + txq = &qg_buf->txqs[0]; + if (pf_q == le16_to_cpu(txq->txq_id)) + ring->txq_teid = le32_to_cpu(txq->q_teid); + + return 0; +} + +/** * ice_vsi_cfg_txqs - Configure the VSI for Tx * @vsi: the VSI being configured * @rings: Tx ring array to be configured @@ -1670,20 +1780,16 @@ static int ice_vsi_cfg_txqs(struct ice_vsi *vsi, struct ice_ring **rings, int offset) { struct ice_aqc_add_tx_qgrp *qg_buf; - struct ice_aqc_add_txqs_perq *txq; struct ice_pf *pf = vsi->back; - u8 num_q_grps, q_idx = 0; - enum ice_status status; - u16 buf_len, i, pf_q; - int err = 0, tc; + u16 q_idx = 0, i; + int err = 0; + u8 tc; - buf_len = sizeof(*qg_buf); - qg_buf = devm_kzalloc(&pf->pdev->dev, buf_len, GFP_KERNEL); + qg_buf = devm_kzalloc(&pf->pdev->dev, sizeof(*qg_buf), GFP_KERNEL); if (!qg_buf) return -ENOMEM; qg_buf->num_txqs = 1; - num_q_grps = 1; /* set up and configure the Tx queues for each enabled TC */ ice_for_each_traffic_class(tc) { @@ -1691,39 +1797,10 @@ ice_vsi_cfg_txqs(struct ice_vsi *vsi, struct ice_ring **rings, int offset) break; for (i = 0; i < vsi->tc_cfg.tc_info[tc].qcount_tx; i++) { - struct ice_tlan_ctx tlan_ctx = { 0 }; - - pf_q = vsi->txq_map[q_idx + offset]; - ice_setup_tx_ctx(rings[q_idx], &tlan_ctx, pf_q); - /* copy context contents into the qg_buf */ - qg_buf->txqs[0].txq_id = cpu_to_le16(pf_q); - ice_set_ctx((u8 *)&tlan_ctx, qg_buf->txqs[0].txq_ctx, - ice_tlan_ctx_info); - - /* init queue specific tail reg. It is referred as - * transmit comm scheduler queue doorbell. - */ - rings[q_idx]->tail = - pf->hw.hw_addr + QTX_COMM_DBELL(pf_q); - status = ice_ena_vsi_txq(vsi->port_info, vsi->idx, tc, - i, num_q_grps, qg_buf, - buf_len, NULL); - if (status) { - dev_err(&pf->pdev->dev, - "Failed to set LAN Tx queue context, error: %d\n", - status); - err = -ENODEV; + err = ice_vsi_cfg_txq(vsi, rings[q_idx], i + offset, + qg_buf, tc); + if (err) goto err_cfg_txqs; - } - - /* Add Tx Queue TEID into the VSI Tx ring from the - * response. This will complete configuring and - * enabling the queue. - */ - txq = &qg_buf->txqs[0]; - if (pf_q == le16_to_cpu(txq->txq_id)) - rings[q_idx]->txq_teid = - le32_to_cpu(txq->q_teid); q_idx++; } @@ -2070,45 +2147,112 @@ void ice_trigger_sw_intr(struct ice_hw *hw, struct ice_q_vector *q_vector) } /** - * ice_vsi_stop_tx_rings - Disable Tx rings + * ice_vsi_stop_tx_ring - Disable single Tx ring * @vsi: the VSI being configured * @rst_src: reset source * @rel_vmvf_num: Relative ID of VF/VM - * @rings: Tx ring array to be stopped - * @offset: offset within vsi->txq_map + * @ring: Tx ring to be stopped + * @txq_meta: Meta data of Tx ring to be stopped */ -static int -ice_vsi_stop_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, - u16 rel_vmvf_num, struct ice_ring **rings, int offset) +#ifndef CONFIG_PCI_IOV +static +#endif /* !CONFIG_PCI_IOV */ +int +ice_vsi_stop_tx_ring(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, + u16 rel_vmvf_num, struct ice_ring *ring, + struct ice_txq_meta *txq_meta) { struct ice_pf *pf = vsi->back; + struct ice_q_vector *q_vector; 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; + u32 val; - if (vsi->num_txq > ICE_LAN_TXQ_MAX_QDIS) - return -EINVAL; + /* clear cause_ena bit for disabled queues */ + val = rd32(hw, QINT_TQCTL(ring->reg_idx)); + val &= ~QINT_TQCTL_CAUSE_ENA_M; + wr32(hw, QINT_TQCTL(ring->reg_idx), val); - q_teids = devm_kcalloc(&pf->pdev->dev, vsi->num_txq, sizeof(*q_teids), - GFP_KERNEL); - if (!q_teids) - return -ENOMEM; + /* software is expected to wait for 100 ns */ + ndelay(100); - q_ids = devm_kcalloc(&pf->pdev->dev, vsi->num_txq, sizeof(*q_ids), - GFP_KERNEL); - if (!q_ids) { - err = -ENOMEM; - goto err_alloc_q_ids; + /* trigger a software interrupt for the vector + * associated to the queue to schedule NAPI handler + */ + q_vector = ring->q_vector; + if (q_vector) + ice_trigger_sw_intr(hw, q_vector); + + status = ice_dis_vsi_txq(vsi->port_info, txq_meta->vsi_idx, + txq_meta->tc, 1, &txq_meta->q_handle, + &txq_meta->q_id, &txq_meta->q_teid, 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_dbg(&vsi->back->pdev->dev, + "Reset in progress. LAN Tx queues already disabled\n"); + } else if (status == ICE_ERR_DOES_NOT_EXIST) { + dev_dbg(&vsi->back->pdev->dev, + "LAN Tx queues do not exist, nothing to disable\n"); + } else if (status) { + dev_err(&vsi->back->pdev->dev, + "Failed to disable LAN Tx queues, error: %d\n", status); + return -ENODEV; } - 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; - } + return 0; +} + +/** + * ice_fill_txq_meta - Prepare the Tx queue's meta data + * @vsi: VSI that ring belongs to + * @ring: ring that txq_meta will be based on + * @txq_meta: a helper struct that wraps Tx queue's information + * + * Set up a helper struct that will contain all the necessary fields that + * are needed for stopping Tx queue + */ +#ifndef CONFIG_PCI_IOV +static +#endif /* !CONFIG_PCI_IOV */ +void +ice_fill_txq_meta(struct ice_vsi *vsi, struct ice_ring *ring, + struct ice_txq_meta *txq_meta) +{ + u8 tc = 0; + +#ifdef CONFIG_DCB + tc = ring->dcb_tc; +#endif /* CONFIG_DCB */ + txq_meta->q_id = ring->reg_idx; + txq_meta->q_teid = ring->txq_teid; + txq_meta->q_handle = ring->q_handle; + txq_meta->vsi_idx = vsi->idx; + txq_meta->tc = tc; +} + +/** + * 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 + * @rings: Tx ring array to be stopped + */ +static int +ice_vsi_stop_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, + u16 rel_vmvf_num, struct ice_ring **rings) +{ + u16 i, q_idx = 0; + int status; + u8 tc; + + if (vsi->num_txq > ICE_LAN_TXQ_MAX_QDIS) + return -EINVAL; /* set up the Tx queue list to be disabled for each enabled TC */ ice_for_each_traffic_class(tc) { @@ -2116,64 +2260,24 @@ ice_vsi_stop_tx_rings(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, break; for (i = 0; i < vsi->tc_cfg.tc_info[tc].qcount_tx; i++) { - struct ice_q_vector *q_vector; + struct ice_txq_meta txq_meta = { }; - if (!rings || !rings[q_idx]) { - err = -EINVAL; - goto err_out; - } - - q_ids[i] = vsi->txq_map[q_idx + offset]; - q_teids[i] = rings[q_idx]->txq_teid; - q_handles[i] = i; + if (!rings || !rings[q_idx]) + return -EINVAL; - /* 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); + ice_fill_txq_meta(vsi, rings[q_idx], &txq_meta); + status = ice_vsi_stop_tx_ring(vsi, rst_src, + rel_vmvf_num, + rings[q_idx], &txq_meta); - /* 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); + if (status) + return status; 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. - */ - 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: - devm_kfree(&pf->pdev->dev, q_teids); - - return err; + return 0; } /** @@ -2186,8 +2290,7 @@ 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); + return ice_vsi_stop_tx_rings(vsi, rst_src, rel_vmvf_num, vsi->tx_rings); } /** @@ -2497,9 +2600,6 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, 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 @@ -2519,7 +2619,7 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, /* configure VSI nodes based on number of queues and TC's */ for (i = 0; i < vsi->tc_cfg.numtc; i++) - max_txqs[i] = pf->num_lan_tx; + max_txqs[i] = vsi->alloc_txq; status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); @@ -2540,15 +2640,17 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, * 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); + if (!ice_is_safe_mode(pf)) { + if (vsi->type == ICE_VSI_PF) { + ice_vsi_add_rem_eth_mac(vsi, true); - /* Tx LLDP packets */ - ice_cfg_sw_lldp(vsi, true, 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); + /* Rx LLDP packets */ + if (!test_bit(ICE_FLAG_FW_LLDP_AGENT, pf->flags)) + ice_cfg_sw_lldp(vsi, false, true); + } } return vsi; @@ -2563,8 +2665,6 @@ unroll_vsi_init: ice_vsi_delete(vsi); unroll_get_qs: ice_vsi_put_qs(vsi); - pf->q_left_tx += vsi->alloc_txq; - pf->q_left_rx += vsi->alloc_rxq; ice_vsi_clear(vsi); return NULL; @@ -2610,39 +2710,36 @@ void ice_vsi_free_irq(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; int base = vsi->base_vector; + int i; - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - int i; - - if (!vsi->q_vectors || !vsi->irqs_ready) - return; + if (!vsi->q_vectors || !vsi->irqs_ready) + return; - ice_vsi_release_msix(vsi); - if (vsi->type == ICE_VSI_VF) - return; + ice_vsi_release_msix(vsi); + if (vsi->type == ICE_VSI_VF) + return; - vsi->irqs_ready = false; - ice_for_each_q_vector(vsi, i) { - u16 vector = i + base; - int irq_num; + vsi->irqs_ready = false; + ice_for_each_q_vector(vsi, i) { + u16 vector = i + base; + int irq_num; - irq_num = pf->msix_entries[vector].vector; + irq_num = pf->msix_entries[vector].vector; - /* free only the irqs that were actually requested */ - if (!vsi->q_vectors[i] || - !(vsi->q_vectors[i]->num_ring_tx || - vsi->q_vectors[i]->num_ring_rx)) - continue; + /* free only the irqs that were actually requested */ + if (!vsi->q_vectors[i] || + !(vsi->q_vectors[i]->num_ring_tx || + vsi->q_vectors[i]->num_ring_rx)) + continue; - /* clear the affinity notifier in the IRQ descriptor */ - irq_set_affinity_notifier(irq_num, NULL); + /* clear the affinity notifier in the IRQ descriptor */ + irq_set_affinity_notifier(irq_num, NULL); - /* clear the affinity_mask in the IRQ descriptor */ - irq_set_affinity_hint(irq_num, NULL); - synchronize_irq(irq_num); - devm_free_irq(&pf->pdev->dev, irq_num, - vsi->q_vectors[i]); - } + /* clear the affinity_mask in the IRQ descriptor */ + irq_set_affinity_hint(irq_num, NULL); + synchronize_irq(irq_num); + devm_free_irq(&pf->pdev->dev, irq_num, + vsi->q_vectors[i]); } } @@ -2821,15 +2918,20 @@ void ice_vsi_dis_irq(struct ice_vsi *vsi) } /* disable each interrupt */ - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - ice_for_each_q_vector(vsi, i) - wr32(hw, GLINT_DYN_CTL(vsi->q_vectors[i]->reg_idx), 0); + ice_for_each_q_vector(vsi, i) { + if (!vsi->q_vectors[i]) + continue; + wr32(hw, GLINT_DYN_CTL(vsi->q_vectors[i]->reg_idx), 0); + } - ice_flush(hw); + ice_flush(hw); - ice_for_each_q_vector(vsi, i) - synchronize_irq(pf->msix_entries[i + base].vector); - } + /* don't call synchronize_irq() for VF's from the host */ + if (vsi->type == ICE_VSI_VF) + return; + + ice_for_each_q_vector(vsi, i) + synchronize_irq(pf->msix_entries[i + base].vector); } /** @@ -2889,14 +2991,16 @@ int ice_vsi_release(struct ice_vsi *vsi) pf->num_avail_sw_msix += vsi->num_q_vectors; } - 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 - */ - if (!test_bit(ICE_FLAG_ENABLE_FW_LLDP, pf->flags)) - ice_cfg_sw_lldp(vsi, false, false); + if (!ice_is_safe_mode(pf)) { + 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 + */ + if (!test_bit(ICE_FLAG_FW_LLDP_AGENT, pf->flags)) + ice_cfg_sw_lldp(vsi, false, false); + } } ice_remove_vsi_fltr(&pf->hw, vsi->idx); @@ -2913,8 +3017,6 @@ int ice_vsi_release(struct ice_vsi *vsi) ice_vsi_clear_rings(vsi); ice_vsi_put_qs(vsi); - pf->q_left_tx += vsi->alloc_txq; - pf->q_left_rx += vsi->alloc_rxq; /* retain SW VSI data structure since it is needed to unregister and * free VSI netdev when PF is not in reset recovery pending state,\ @@ -2962,6 +3064,7 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) vsi->base_vector = 0; } + ice_vsi_put_qs(vsi); ice_vsi_clear_rings(vsi); ice_vsi_free_arrays(vsi); ice_dev_onetime_setup(&pf->hw); @@ -2969,6 +3072,12 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) ice_vsi_set_num_qs(vsi, vf->vf_id); else ice_vsi_set_num_qs(vsi, ICE_INVAL_VFID); + + ret = ice_vsi_alloc_arrays(vsi); + if (ret < 0) + goto err_vsi; + + ice_vsi_get_qs(vsi); ice_vsi_set_tc_cfg(vsi); /* Initialize VSI struct elements and create VSI in FW */ @@ -2976,9 +3085,6 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret < 0) goto err_vsi; - ret = ice_vsi_alloc_arrays(vsi); - if (ret < 0) - goto err_vsi; switch (vsi->type) { case ICE_VSI_PF: @@ -2986,6 +3092,10 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret) goto err_rings; + ret = ice_vsi_setup_vector_base(vsi); + if (ret) + goto err_vectors; + ret = ice_vsi_set_q_vectors_reg_idx(vsi); if (ret) goto err_vectors; @@ -3007,10 +3117,6 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret) goto err_rings; - ret = ice_vsi_setup_vector_base(vsi); - if (ret) - goto err_vectors; - ret = ice_vsi_set_q_vectors_reg_idx(vsi); if (ret) goto err_vectors; @@ -3019,8 +3125,6 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) if (ret) goto err_vectors; - pf->q_left_tx -= vsi->alloc_txq; - pf->q_left_rx -= vsi->alloc_rxq; break; default: break; @@ -3028,7 +3132,7 @@ int ice_vsi_rebuild(struct ice_vsi *vsi) /* configure VSI nodes based on number of queues and TC's */ for (i = 0; i < vsi->tc_cfg.numtc; i++) - max_txqs[i] = pf->num_lan_tx; + max_txqs[i] = vsi->alloc_txq; status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); @@ -3083,48 +3187,6 @@ static void ice_vsi_update_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctx) } /** - * 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 @@ -3145,7 +3207,7 @@ int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc) if (ena_tc & BIT(i)) num_tc++; /* populate max_txqs per TC */ - max_txqs[i] = pf->num_lan_tx; + max_txqs[i] = vsi->alloc_txq; } vsi->tc_cfg.ena_tc = ena_tc; @@ -3188,3 +3250,52 @@ out: return ret; } #endif /* CONFIG_DCB */ + +/** + * ice_nvm_version_str - format the NVM version strings + * @hw: ptr to the hardware info + */ +char *ice_nvm_version_str(struct ice_hw *hw) +{ + u8 oem_ver, oem_patch, ver_hi, ver_lo; + static char buf[ICE_NVM_VER_LEN]; + u16 oem_build; + + ice_get_nvm_version(hw, &oem_ver, &oem_build, &oem_patch, &ver_hi, + &ver_lo); + + snprintf(buf, sizeof(buf), "%x.%02x 0x%x %d.%d.%d", ver_hi, ver_lo, + hw->nvm.eetrack, oem_ver, oem_build, oem_patch); + + return buf; +} + +/** + * ice_vsi_cfg_mac_fltr - Add or remove a MAC address filter for a VSI + * @vsi: the VSI being configured MAC filter + * @macaddr: the MAC address to be added. + * @set: Add or delete a MAC filter + * + * Adds or removes MAC address filter entry for VF VSI + */ +enum ice_status +ice_vsi_cfg_mac_fltr(struct ice_vsi *vsi, const u8 *macaddr, bool set) +{ + LIST_HEAD(tmp_add_list); + enum ice_status status; + + /* Update MAC filter list to be added or removed for a VSI */ + if (ice_add_mac_to_list(vsi, &tmp_add_list, macaddr)) { + status = ICE_ERR_NO_MEMORY; + goto cfg_mac_fltr_exit; + } + + if (set) + status = ice_add_mac(&vsi->back->hw, &tmp_add_list); + else + status = ice_remove_mac(&vsi->back->hw, &tmp_add_list); + +cfg_mac_fltr_exit: + ice_free_fltr_list(&vsi->back->pdev->dev, &tmp_add_list); + return status; +} diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h index 6e43ef03bfc3..47bc033fff20 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_lib.h @@ -6,8 +6,22 @@ #include "ice.h" -int ice_add_mac_to_list(struct ice_vsi *vsi, struct list_head *add_list, - const u8 *macaddr); +struct ice_txq_meta { + /* Tx-scheduler element identifier */ + u32 q_teid; + /* Entry in VSI's txq_map bitmap */ + u16 q_id; + /* Relative index of Tx queue within TC */ + u16 q_handle; + /* VSI index that Tx queue belongs to */ + u16 vsi_idx; + /* TC number that Tx queue belongs to */ + u8 tc; +}; + +int +ice_add_mac_to_list(struct ice_vsi *vsi, struct list_head *add_list, + const u8 *macaddr); void ice_free_fltr_list(struct device *dev, struct list_head *h); @@ -25,6 +39,16 @@ 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); + +int +ice_vsi_stop_tx_ring(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, + u16 rel_vmvf_num, struct ice_ring *ring, + struct ice_txq_meta *txq_meta); + +void ice_fill_txq_meta(struct ice_vsi *vsi, struct ice_ring *ring, + struct ice_txq_meta *txq_meta); + +int ice_vsi_ctrl_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx); #endif /* CONFIG_PCI_IOV */ int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid); @@ -95,4 +119,11 @@ void ice_vsi_free_tx_rings(struct ice_vsi *vsi); int ice_vsi_manage_rss_lut(struct ice_vsi *vsi, bool ena); u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran); + +char *ice_nvm_version_str(struct ice_hw *hw); + +enum ice_status +ice_vsi_cfg_mac_fltr(struct ice_vsi *vsi, const u8 *macaddr, bool set); + +bool ice_is_safe_mode(struct ice_pf *pf); #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 63db08d9bafa..214cd6eca405 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -9,16 +9,27 @@ #include "ice_lib.h" #include "ice_dcb_lib.h" -#define DRV_VERSION "0.7.4-k" +#define DRV_VERSION_MAJOR 0 +#define DRV_VERSION_MINOR 8 +#define DRV_VERSION_BUILD 1 + +#define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ + __stringify(DRV_VERSION_MINOR) "." \ + __stringify(DRV_VERSION_BUILD) "-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; static const char ice_copyright[] = "Copyright (c) 2018, Intel Corporation."; +/* DDP Package file located in firmware search paths (e.g. /lib/firmware/) */ +#define ICE_DDP_PKG_PATH "intel/ice/ddp/" +#define ICE_DDP_PKG_FILE ICE_DDP_PKG_PATH "ice.pkg" + MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>"); MODULE_DESCRIPTION(DRV_SUMMARY); MODULE_LICENSE("GPL v2"); MODULE_VERSION(DRV_VERSION); +MODULE_FIRMWARE(ICE_DDP_PKG_FILE); static int debug = -1; module_param(debug, int, 0644); @@ -29,24 +40,23 @@ MODULE_PARM_DESC(debug, "netif level (0=none,...,16=all)"); #endif /* !CONFIG_DYNAMIC_DEBUG */ static struct workqueue_struct *ice_wq; +static const struct net_device_ops ice_netdev_safe_mode_ops; static const struct net_device_ops ice_netdev_ops; -static void ice_rebuild(struct ice_pf *pf); +static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type); static void ice_vsi_release_all(struct ice_pf *pf); -static void ice_update_vsi_stats(struct ice_vsi *vsi); -static void ice_update_pf_stats(struct ice_pf *pf); /** * ice_get_tx_pending - returns number of Tx descriptors not processed * @ring: the ring of descriptors */ -static u32 ice_get_tx_pending(struct ice_ring *ring) +static u16 ice_get_tx_pending(struct ice_ring *ring) { - u32 head, tail; + u16 head, tail; head = ring->next_to_clean; - tail = readl(ring->tail); + tail = ring->next_to_use; if (head != tail) return (head < tail) ? @@ -118,12 +128,11 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) */ static int ice_init_mac_fltr(struct ice_pf *pf) { - LIST_HEAD(tmp_add_list); + enum ice_status status; u8 broadcast[ETH_ALEN]; struct ice_vsi *vsi; - int status; - vsi = ice_find_vsi_by_type(pf, ICE_VSI_PF); + vsi = ice_get_main_vsi(pf); if (!vsi) return -EINVAL; @@ -132,8 +141,7 @@ static int ice_init_mac_fltr(struct ice_pf *pf) */ /* 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_vsi_cfg_mac_fltr(vsi, vsi->port_info->mac.perm_addr, true); if (status) goto unregister; @@ -141,18 +149,11 @@ static int ice_init_mac_fltr(struct ice_pf *pf) * 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); + status = ice_vsi_cfg_mac_fltr(vsi, broadcast, true); if (status) - status = -ENOMEM; - -free_mac_list: - ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); + goto unregister; + return 0; unregister: /* We aren't useful with no MAC filters, so unregister if we * had an error @@ -166,7 +167,7 @@ unregister: vsi->netdev = NULL; } - return status; + return -EIO; } /** @@ -447,13 +448,13 @@ static void ice_dis_vsi(struct ice_vsi *vsi, bool locked) if (vsi->type == ICE_VSI_PF && vsi->netdev) { if (netif_running(vsi->netdev)) { - if (!locked) { + if (!locked) rtnl_lock(); - vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); + + ice_stop(vsi->netdev); + + if (!locked) rtnl_unlock(); - } else { - vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); - } } else { ice_vsi_close(vsi); } @@ -488,6 +489,7 @@ static void ice_prepare_for_reset(struct ice_pf *pf) { struct ice_hw *hw = &pf->hw; + int i; /* already prepared for reset */ if (test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) @@ -497,6 +499,12 @@ ice_prepare_for_reset(struct ice_pf *pf) if (ice_check_sq_alive(hw, &hw->mailboxq)) ice_vc_notify_reset(pf); + /* Disable VFs until reset is completed */ + for (i = 0; i < pf->num_alloc_vfs; i++) + ice_set_vf_state_qs_dis(&pf->vf[i]); + + /* clear SW filtering DB */ + ice_clear_hw_tbls(hw); /* disable the VSIs and their queues that are not already DOWN */ ice_pf_dis_all_vsi(pf, false); @@ -542,7 +550,7 @@ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type) */ if (reset_type == ICE_RESET_PFR) { pf->pfr_count++; - ice_rebuild(pf); + ice_rebuild(pf, reset_type); clear_bit(__ICE_PREPARED_FOR_RESET, pf->state); clear_bit(__ICE_PFR_REQ, pf->state); ice_reset_all_vfs(pf, true); @@ -573,6 +581,8 @@ static void ice_reset_subtask(struct ice_pf *pf) reset_type = ICE_RESET_CORER; if (test_and_clear_bit(__ICE_GLOBR_RECV, pf->state)) reset_type = ICE_RESET_GLOBR; + if (test_and_clear_bit(__ICE_EMPR_RECV, pf->state)) + reset_type = ICE_RESET_EMPR; /* return if no valid reset type requested */ if (reset_type == ICE_RESET_INVAL) return; @@ -584,7 +594,7 @@ static void ice_reset_subtask(struct ice_pf *pf) } else { /* done with reset. start rebuild */ pf->hw.reset_ongoing = false; - ice_rebuild(pf); + ice_rebuild(pf, reset_type); /* clear bit to resume normal operations, but * ICE_NEEDS_RESTART bit is set in case rebuild failed */ @@ -618,6 +628,22 @@ static void ice_reset_subtask(struct ice_pf *pf) } /** + * ice_print_topo_conflict - print topology conflict message + * @vsi: the VSI whose topology status is being checked + */ +static void ice_print_topo_conflict(struct ice_vsi *vsi) +{ + switch (vsi->port_info->phy.link_info.topo_media_conflict) { + case ICE_AQ_LINK_TOPO_CONFLICT: + case ICE_AQ_LINK_MEDIA_CONFLICT: + netdev_info(vsi->netdev, "Possible mis-configuration of the Ethernet port detected, please use the Intel(R) Ethernet Port Configuration Tool application to address the issue.\n"); + break; + default: + break; + } +} + +/** * ice_print_link_msg - print link up or down message * @vsi: the VSI whose link status is being queried * @isup: boolean for if the link is now up or down @@ -630,6 +656,7 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) const char *speed; const char *fec; const char *fc; + const char *an; if (!vsi) return; @@ -713,6 +740,12 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) break; } + /* check if autoneg completed, might be false due to not supported */ + if (vsi->port_info->phy.link_info.an_info & ICE_AQ_AN_COMPLETED) + an = "True"; + else + an = "False"; + /* Get FEC mode requested based on PHY caps last SW configuration */ caps = devm_kzalloc(&vsi->back->pdev->dev, sizeof(*caps), GFP_KERNEL); if (!caps) { @@ -737,8 +770,9 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) 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); + netdev_info(vsi->netdev, "NIC Link is up %sbps, Requested FEC: %s, FEC: %s, Autoneg: %s, Flow Control: %s\n", + speed, fec_req, fec, an, fc); + ice_print_topo_conflict(vsi); } /** @@ -806,10 +840,24 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, if (link_up == old_link && link_speed == old_link_speed) return result; - vsi = ice_find_vsi_by_type(pf, ICE_VSI_PF); + vsi = ice_get_main_vsi(pf); if (!vsi || !vsi->port_info) return -EINVAL; + /* turn off PHY if media was removed */ + if (!test_bit(ICE_FLAG_NO_MEDIA, pf->flags) && + !(pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE)) { + set_bit(ICE_FLAG_NO_MEDIA, pf->flags); + + result = ice_aq_set_link_restart_an(pi, false, NULL); + if (result) { + dev_dbg(&pf->pdev->dev, + "Failed to set link down, VSI %d error %d\n", + vsi->vsi_num, result); + return result; + } + } + ice_vsi_link_event(vsi, link_up); ice_print_link_msg(vsi, link_up); @@ -1307,14 +1355,134 @@ static void ice_handle_mdd_event(struct ice_pf *pf) 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", + if (vf->num_mdd_events && + vf->num_mdd_events <= ICE_MDD_EVENTS_THRESHOLD) + dev_info(&pf->pdev->dev, + "VF %d has had %llu MDD events since last boot, Admin might need to reload AVF driver with this number of events\n", i, vf->num_mdd_events); } } } /** + * ice_force_phys_link_state - Force the physical link state + * @vsi: VSI to force the physical link state to up/down + * @link_up: true/false indicates to set the physical link to up/down + * + * Force the physical link state by getting the current PHY capabilities from + * hardware and setting the PHY config based on the determined capabilities. If + * link changes a link event will be triggered because both the Enable Automatic + * Link Update and LESM Enable bits are set when setting the PHY capabilities. + * + * Returns 0 on success, negative on failure + */ +static int ice_force_phys_link_state(struct ice_vsi *vsi, bool link_up) +{ + struct ice_aqc_get_phy_caps_data *pcaps; + struct ice_aqc_set_phy_cfg_data *cfg; + struct ice_port_info *pi; + struct device *dev; + int retcode; + + if (!vsi || !vsi->port_info || !vsi->back) + return -EINVAL; + if (vsi->type != ICE_VSI_PF) + return 0; + + dev = &vsi->back->pdev->dev; + + pi = vsi->port_info; + + pcaps = devm_kzalloc(dev, sizeof(*pcaps), GFP_KERNEL); + if (!pcaps) + return -ENOMEM; + + retcode = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps, + NULL); + if (retcode) { + dev_err(dev, + "Failed to get phy capabilities, VSI %d error %d\n", + vsi->vsi_num, retcode); + retcode = -EIO; + goto out; + } + + /* No change in link */ + if (link_up == !!(pcaps->caps & ICE_AQC_PHY_EN_LINK) && + link_up == !!(pi->phy.link_info.link_info & ICE_AQ_LINK_UP)) + goto out; + + cfg = devm_kzalloc(dev, sizeof(*cfg), GFP_KERNEL); + if (!cfg) { + retcode = -ENOMEM; + goto out; + } + + cfg->phy_type_low = pcaps->phy_type_low; + cfg->phy_type_high = pcaps->phy_type_high; + cfg->caps = pcaps->caps | ICE_AQ_PHY_ENA_AUTO_LINK_UPDT; + cfg->low_power_ctrl = pcaps->low_power_ctrl; + cfg->eee_cap = pcaps->eee_cap; + cfg->eeer_value = pcaps->eeer_value; + cfg->link_fec_opt = pcaps->link_fec_options; + if (link_up) + cfg->caps |= ICE_AQ_PHY_ENA_LINK; + else + cfg->caps &= ~ICE_AQ_PHY_ENA_LINK; + + retcode = ice_aq_set_phy_cfg(&vsi->back->hw, pi->lport, cfg, NULL); + if (retcode) { + dev_err(dev, "Failed to set phy config, VSI %d error %d\n", + vsi->vsi_num, retcode); + retcode = -EIO; + } + + devm_kfree(dev, cfg); +out: + devm_kfree(dev, pcaps); + return retcode; +} + +/** + * ice_check_media_subtask - Check for media; bring link up if detected. + * @pf: pointer to PF struct + */ +static void ice_check_media_subtask(struct ice_pf *pf) +{ + struct ice_port_info *pi; + struct ice_vsi *vsi; + int err; + + vsi = ice_get_main_vsi(pf); + if (!vsi) + return; + + /* No need to check for media if it's already present or the interface + * is down + */ + if (!test_bit(ICE_FLAG_NO_MEDIA, pf->flags) || + test_bit(__ICE_DOWN, vsi->state)) + return; + + /* Refresh link info and check if media is present */ + pi = vsi->port_info; + err = ice_update_link_info(pi); + if (err) + return; + + if (pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) { + err = ice_force_phys_link_state(vsi, true); + if (err) + return; + clear_bit(ICE_FLAG_NO_MEDIA, pf->flags); + + /* A Link Status Event will be generated; the event handler + * will complete bringing the interface up + */ + } +} + +/** * ice_service_task - manage and run subtasks * @work: pointer to work_struct contained by the PF struct */ @@ -1336,12 +1504,19 @@ static void ice_service_task(struct work_struct *work) return; } + ice_clean_adminq_subtask(pf); + ice_check_media_subtask(pf); ice_check_for_hang_subtask(pf); ice_sync_fltr_subtask(pf); ice_handle_mdd_event(pf); - ice_process_vflr_event(pf); ice_watchdog_subtask(pf); - ice_clean_adminq_subtask(pf); + + if (ice_is_safe_mode(pf)) { + ice_service_task_complete(pf); + return; + } + + ice_process_vflr_event(pf); ice_clean_mailboxq_subtask(pf); /* Clear __ICE_SERVICE_SCHED flag to allow scheduling next event */ @@ -1369,8 +1544,8 @@ static void ice_set_ctrlq_len(struct ice_hw *hw) hw->adminq.num_sq_entries = ICE_AQ_LEN; hw->adminq.rq_buf_size = ICE_AQ_MAX_BUF_LEN; hw->adminq.sq_buf_size = ICE_AQ_MAX_BUF_LEN; - hw->mailboxq.num_rq_entries = ICE_MBXQ_LEN; - hw->mailboxq.num_sq_entries = ICE_MBXQ_LEN; + hw->mailboxq.num_rq_entries = ICE_MBXRQ_LEN; + hw->mailboxq.num_sq_entries = ICE_MBXSQ_LEN; hw->mailboxq.rq_buf_size = ICE_MBXQ_MAX_BUF_LEN; hw->mailboxq.sq_buf_size = ICE_MBXQ_MAX_BUF_LEN; } @@ -1409,15 +1584,11 @@ static void ice_irq_affinity_release(struct kref __always_unused *ref) {} */ static int ice_vsi_ena_irq(struct ice_vsi *vsi) { - struct ice_pf *pf = vsi->back; - struct ice_hw *hw = &pf->hw; + struct ice_hw *hw = &vsi->back->hw; + int i; - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - int i; - - ice_for_each_q_vector(vsi, i) - ice_irq_dynamic_ena(hw, vsi, vsi->q_vectors[i]); - } + ice_for_each_q_vector(vsi, i) + ice_irq_dynamic_ena(hw, vsi, vsi->q_vectors[i]); ice_flush(hw); return 0; @@ -1665,7 +1836,7 @@ static void ice_free_irq_msix_misc(struct ice_pf *pf) wr32(hw, PFINT_OICR_ENA, 0); ice_flush(hw); - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags) && pf->msix_entries) { + if (pf->msix_entries) { synchronize_irq(pf->msix_entries[pf->oicr_idx].vector); devm_free_irq(&pf->pdev->dev, pf->msix_entries[pf->oicr_idx].vector, pf); @@ -1780,30 +1951,41 @@ static void ice_napi_add(struct ice_vsi *vsi) } /** - * ice_cfg_netdev - Allocate, configure and register a netdev - * @vsi: the VSI associated with the new netdev - * - * Returns 0 on success, negative value on failure + * ice_set_ops - set netdev and ethtools ops for the given netdev + * @netdev: netdev instance */ -static int ice_cfg_netdev(struct ice_vsi *vsi) +static void ice_set_ops(struct net_device *netdev) { + struct ice_pf *pf = ice_netdev_to_pf(netdev); + + if (ice_is_safe_mode(pf)) { + netdev->netdev_ops = &ice_netdev_safe_mode_ops; + ice_set_ethtool_safe_mode_ops(netdev); + return; + } + + netdev->netdev_ops = &ice_netdev_ops; + ice_set_ethtool_ops(netdev); +} + +/** + * ice_set_netdev_features - set features for the given netdev + * @netdev: netdev instance + */ +static void ice_set_netdev_features(struct net_device *netdev) +{ + struct ice_pf *pf = ice_netdev_to_pf(netdev); netdev_features_t csumo_features; netdev_features_t vlano_features; netdev_features_t dflt_features; netdev_features_t tso_features; - struct ice_netdev_priv *np; - struct net_device *netdev; - u8 mac_addr[ETH_ALEN]; - int err; - - netdev = alloc_etherdev_mqs(sizeof(*np), vsi->alloc_txq, - vsi->alloc_rxq); - if (!netdev) - return -ENOMEM; - vsi->netdev = netdev; - np = netdev_priv(netdev); - np->vsi = vsi; + if (ice_is_safe_mode(pf)) { + /* safe mode */ + netdev->features = NETIF_F_SG | NETIF_F_HIGHDMA; + netdev->hw_features = netdev->features; + return; + } dflt_features = NETIF_F_SG | NETIF_F_HIGHDMA | @@ -1831,25 +2013,50 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) tso_features; netdev->vlan_features |= dflt_features | csumo_features | tso_features; +} + +/** + * ice_cfg_netdev - Allocate, configure and register a netdev + * @vsi: the VSI associated with the new netdev + * + * Returns 0 on success, negative value on failure + */ +static int ice_cfg_netdev(struct ice_vsi *vsi) +{ + struct ice_pf *pf = vsi->back; + struct ice_netdev_priv *np; + struct net_device *netdev; + u8 mac_addr[ETH_ALEN]; + int err; + + netdev = alloc_etherdev_mqs(sizeof(*np), vsi->alloc_txq, + vsi->alloc_rxq); + if (!netdev) + return -ENOMEM; + + vsi->netdev = netdev; + np = netdev_priv(netdev); + np->vsi = vsi; + + ice_set_netdev_features(netdev); + + ice_set_ops(netdev); if (vsi->type == ICE_VSI_PF) { - SET_NETDEV_DEV(netdev, &vsi->back->pdev->dev); + SET_NETDEV_DEV(netdev, &pf->pdev->dev); ether_addr_copy(mac_addr, vsi->port_info->mac.perm_addr); - ether_addr_copy(netdev->dev_addr, mac_addr); ether_addr_copy(netdev->perm_addr, mac_addr); } netdev->priv_flags |= IFF_UNICAST_FLT; - /* assign netdev_ops */ - netdev->netdev_ops = &ice_netdev_ops; + /* Setup netdev TC information */ + ice_vsi_cfg_netdev_tc(vsi, vsi->tc_cfg.ena_tc); /* setup watchdog timeout value to be 5 second */ netdev->watchdog_timeo = 5 * HZ; - ice_set_ethtool_ops(netdev); - netdev->min_mtu = ETH_MIN_MTU; netdev->max_mtu = ICE_MAX_MTU; @@ -2041,36 +2248,48 @@ unroll_vsi_setup: ice_vsi_free_q_vectors(vsi); ice_vsi_delete(vsi); ice_vsi_put_qs(vsi); - pf->q_left_tx += vsi->alloc_txq; - pf->q_left_rx += vsi->alloc_rxq; ice_vsi_clear(vsi); } return status; } /** - * ice_determine_q_usage - Calculate queue distribution - * @pf: board private structure - * - * Return -ENOMEM if we don't get enough queues for all ports + * ice_get_avail_q_count - Get count of queues in use + * @pf_qmap: bitmap to get queue use count from + * @lock: pointer to a mutex that protects access to pf_qmap + * @size: size of the bitmap */ -static void ice_determine_q_usage(struct ice_pf *pf) +static u16 +ice_get_avail_q_count(unsigned long *pf_qmap, struct mutex *lock, u16 size) { - u16 q_left_tx, q_left_rx; + u16 count = 0, bit; - q_left_tx = pf->hw.func_caps.common_cap.num_txq; - q_left_rx = pf->hw.func_caps.common_cap.num_rxq; + mutex_lock(lock); + for_each_clear_bit(bit, pf_qmap, size) + count++; + mutex_unlock(lock); - pf->num_lan_tx = min_t(int, q_left_tx, num_online_cpus()); + return count; +} - /* only 1 Rx queue unless RSS is enabled */ - if (!test_bit(ICE_FLAG_RSS_ENA, pf->flags)) - pf->num_lan_rx = 1; - else - pf->num_lan_rx = min_t(int, q_left_rx, num_online_cpus()); +/** + * ice_get_avail_txq_count - Get count of Tx queues in use + * @pf: pointer to an ice_pf instance + */ +u16 ice_get_avail_txq_count(struct ice_pf *pf) +{ + return ice_get_avail_q_count(pf->avail_txqs, &pf->avail_q_mutex, + pf->max_pf_txqs); +} - pf->q_left_tx = q_left_tx - pf->num_lan_tx; - pf->q_left_rx = q_left_rx - pf->num_lan_rx; +/** + * ice_get_avail_rxq_count - Get count of Rx queues in use + * @pf: pointer to an ice_pf instance + */ +u16 ice_get_avail_rxq_count(struct ice_pf *pf) +{ + return ice_get_avail_q_count(pf->avail_rxqs, &pf->avail_q_mutex, + pf->max_pf_rxqs); } /** @@ -2082,43 +2301,74 @@ static void ice_deinit_pf(struct ice_pf *pf) ice_service_task_stop(pf); mutex_destroy(&pf->sw_mutex); mutex_destroy(&pf->avail_q_mutex); + + if (pf->avail_txqs) { + bitmap_free(pf->avail_txqs); + pf->avail_txqs = NULL; + } + + if (pf->avail_rxqs) { + bitmap_free(pf->avail_rxqs); + pf->avail_rxqs = NULL; + } } /** - * ice_init_pf - Initialize general software structures (struct ice_pf) - * @pf: board private structure to initialize + * ice_set_pf_caps - set PFs capability flags + * @pf: pointer to the PF instance */ -static void ice_init_pf(struct ice_pf *pf) +static void ice_set_pf_caps(struct ice_pf *pf) { - bitmap_zero(pf->flags, ICE_PF_FLAGS_NBITS); - set_bit(ICE_FLAG_MSIX_ENA, pf->flags); -#ifdef CONFIG_PCI_IOV - if (pf->hw.func_caps.common_cap.sr_iov_1_1) { - struct ice_hw *hw = &pf->hw; + struct ice_hw_func_caps *func_caps = &pf->hw.func_caps; + clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); + if (func_caps->common_cap.dcb) + set_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); +#ifdef CONFIG_PCI_IOV + clear_bit(ICE_FLAG_SRIOV_CAPABLE, pf->flags); + if (func_caps->common_cap.sr_iov_1_1) { set_bit(ICE_FLAG_SRIOV_CAPABLE, pf->flags); - pf->num_vfs_supported = min_t(int, hw->func_caps.num_allocd_vfs, + pf->num_vfs_supported = min_t(int, func_caps->num_allocd_vfs, ICE_MAX_VF_COUNT); } #endif /* CONFIG_PCI_IOV */ + clear_bit(ICE_FLAG_RSS_ENA, pf->flags); + if (func_caps->common_cap.rss_table_size) + set_bit(ICE_FLAG_RSS_ENA, pf->flags); - mutex_init(&pf->sw_mutex); - mutex_init(&pf->avail_q_mutex); + pf->max_pf_txqs = func_caps->common_cap.num_txq; + pf->max_pf_rxqs = func_caps->common_cap.num_rxq; +} - /* Clear avail_[t|r]x_qs bitmaps (set all to avail) */ - mutex_lock(&pf->avail_q_mutex); - bitmap_zero(pf->avail_txqs, ICE_MAX_TXQS); - bitmap_zero(pf->avail_rxqs, ICE_MAX_RXQS); - mutex_unlock(&pf->avail_q_mutex); +/** + * ice_init_pf - Initialize general software structures (struct ice_pf) + * @pf: board private structure to initialize + */ +static int ice_init_pf(struct ice_pf *pf) +{ + ice_set_pf_caps(pf); - if (pf->hw.func_caps.common_cap.rss_table_size) - set_bit(ICE_FLAG_RSS_ENA, pf->flags); + mutex_init(&pf->sw_mutex); /* setup service timer and periodic service task */ timer_setup(&pf->serv_tmr, ice_service_timer, 0); pf->serv_tmr_period = HZ; INIT_WORK(&pf->serv_task, ice_service_task); clear_bit(__ICE_SERVICE_SCHED, pf->state); + + mutex_init(&pf->avail_q_mutex); + pf->avail_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); + if (!pf->avail_txqs) + return -ENOMEM; + + pf->avail_rxqs = bitmap_zalloc(pf->max_pf_rxqs, GFP_KERNEL); + if (!pf->avail_rxqs) { + devm_kfree(&pf->pdev->dev, pf->avail_txqs); + pf->avail_txqs = NULL; + return -ENOMEM; + } + + return 0; } /** @@ -2137,13 +2387,18 @@ static int ice_ena_msix_range(struct ice_pf *pf) /* reserve one vector for miscellaneous handler */ needed = 1; + if (v_left < needed) + goto no_hw_vecs_left_err; v_budget += needed; v_left -= needed; /* reserve vectors for LAN traffic */ - pf->num_lan_msix = min_t(int, num_online_cpus(), v_left); - v_budget += pf->num_lan_msix; - v_left -= pf->num_lan_msix; + needed = min_t(int, num_online_cpus(), v_left); + if (v_left < needed) + goto no_hw_vecs_left_err; + pf->num_lan_msix = needed; + v_budget += needed; + v_left -= needed; pf->msix_entries = devm_kcalloc(&pf->pdev->dev, v_budget, sizeof(*pf->msix_entries), GFP_KERNEL); @@ -2168,18 +2423,18 @@ static int ice_ena_msix_range(struct ice_pf *pf) if (v_actual < v_budget) { dev_warn(&pf->pdev->dev, - "not enough vectors. requested = %d, obtained = %d\n", + "not enough OS MSI-X vectors. requested = %d, obtained = %d\n", v_budget, v_actual); - if (v_actual >= (pf->num_lan_msix + 1)) { - pf->num_avail_sw_msix = v_actual - - (pf->num_lan_msix + 1); - } else if (v_actual >= 2) { - pf->num_lan_msix = 1; - pf->num_avail_sw_msix = v_actual - 2; - } else { +/* 2 vectors for LAN (traffic + OICR) */ +#define ICE_MIN_LAN_VECS 2 + + if (v_actual < ICE_MIN_LAN_VECS) { + /* error if we can't get minimum vectors */ pci_disable_msix(pf->pdev); err = -ERANGE; goto msix_err; + } else { + pf->num_lan_msix = ICE_MIN_LAN_VECS; } } @@ -2189,9 +2444,13 @@ msix_err: devm_kfree(&pf->pdev->dev, pf->msix_entries); goto exit_err; +no_hw_vecs_left_err: + dev_err(&pf->pdev->dev, + "not enough device MSI-X vectors. requested = %d, available = %d\n", + needed, v_left); + err = -ERANGE; exit_err: pf->num_lan_msix = 0; - clear_bit(ICE_FLAG_MSIX_ENA, pf->flags); return err; } @@ -2204,7 +2463,6 @@ static void ice_dis_msix(struct ice_pf *pf) pci_disable_msix(pf->pdev); devm_kfree(&pf->pdev->dev, pf->msix_entries); pf->msix_entries = NULL; - clear_bit(ICE_FLAG_MSIX_ENA, pf->flags); } /** @@ -2213,8 +2471,7 @@ static void ice_dis_msix(struct ice_pf *pf) */ static void ice_clear_interrupt_scheme(struct ice_pf *pf) { - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - ice_dis_msix(pf); + ice_dis_msix(pf); if (pf->irq_tracker) { devm_kfree(&pf->pdev->dev, pf->irq_tracker); @@ -2230,10 +2487,7 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf) { int vectors; - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - vectors = ice_ena_msix_range(pf); - else - return -ENODEV; + vectors = ice_ena_msix_range(pf); if (vectors < 0) return vectors; @@ -2256,6 +2510,163 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf) } /** + * ice_log_pkg_init - log result of DDP package load + * @hw: pointer to hardware info + * @status: status of package load + */ +static void +ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) +{ + struct ice_pf *pf = (struct ice_pf *)hw->back; + struct device *dev = &pf->pdev->dev; + + switch (*status) { + case ICE_SUCCESS: + /* The package download AdminQ command returned success because + * this download succeeded or ICE_ERR_AQ_NO_WORK since there is + * already a package loaded on the device. + */ + if (hw->pkg_ver.major == hw->active_pkg_ver.major && + hw->pkg_ver.minor == hw->active_pkg_ver.minor && + hw->pkg_ver.update == hw->active_pkg_ver.update && + hw->pkg_ver.draft == hw->active_pkg_ver.draft && + !memcmp(hw->pkg_name, hw->active_pkg_name, + sizeof(hw->pkg_name))) { + if (hw->pkg_dwnld_status == ICE_AQ_RC_EEXIST) + dev_info(dev, + "DDP package already present on device: %s version %d.%d.%d.%d\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + hw->active_pkg_ver.update, + hw->active_pkg_ver.draft); + else + dev_info(dev, + "The DDP package was successfully loaded: %s version %d.%d.%d.%d\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + hw->active_pkg_ver.update, + hw->active_pkg_ver.draft); + } else if (hw->active_pkg_ver.major != ICE_PKG_SUPP_VER_MAJ || + hw->active_pkg_ver.minor != ICE_PKG_SUPP_VER_MNR) { + dev_err(dev, + "The device has a DDP package that is not supported by the driver. The device has package '%s' version %d.%d.x.x. The driver requires version %d.%d.x.x. Entering Safe Mode.\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + ICE_PKG_SUPP_VER_MAJ, ICE_PKG_SUPP_VER_MNR); + *status = ICE_ERR_NOT_SUPPORTED; + } else if (hw->active_pkg_ver.major == ICE_PKG_SUPP_VER_MAJ && + hw->active_pkg_ver.minor == ICE_PKG_SUPP_VER_MNR) { + dev_info(dev, + "The driver could not load the DDP package file because a compatible DDP package is already present on the device. The device has package '%s' version %d.%d.%d.%d. The package file found by the driver: '%s' version %d.%d.%d.%d.\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + hw->active_pkg_ver.update, + hw->active_pkg_ver.draft, + hw->pkg_name, + hw->pkg_ver.major, + hw->pkg_ver.minor, + hw->pkg_ver.update, + hw->pkg_ver.draft); + } else { + dev_err(dev, + "An unknown error occurred when loading the DDP package, please reboot the system. If the problem persists, update the NVM. Entering Safe Mode.\n"); + *status = ICE_ERR_NOT_SUPPORTED; + } + break; + case ICE_ERR_BUF_TOO_SHORT: + /* fall-through */ + case ICE_ERR_CFG: + dev_err(dev, + "The DDP package file is invalid. Entering Safe Mode.\n"); + break; + case ICE_ERR_NOT_SUPPORTED: + /* Package File version not supported */ + if (hw->pkg_ver.major > ICE_PKG_SUPP_VER_MAJ || + (hw->pkg_ver.major == ICE_PKG_SUPP_VER_MAJ && + hw->pkg_ver.minor > ICE_PKG_SUPP_VER_MNR)) + dev_err(dev, + "The DDP package file version is higher than the driver supports. Please use an updated driver. Entering Safe Mode.\n"); + else if (hw->pkg_ver.major < ICE_PKG_SUPP_VER_MAJ || + (hw->pkg_ver.major == ICE_PKG_SUPP_VER_MAJ && + hw->pkg_ver.minor < ICE_PKG_SUPP_VER_MNR)) + dev_err(dev, + "The DDP package file version is lower than the driver supports. The driver requires version %d.%d.x.x. Please use an updated DDP Package file. Entering Safe Mode.\n", + ICE_PKG_SUPP_VER_MAJ, ICE_PKG_SUPP_VER_MNR); + break; + case ICE_ERR_AQ_ERROR: + switch (hw->adminq.sq_last_status) { + case ICE_AQ_RC_ENOSEC: + case ICE_AQ_RC_EBADSIG: + dev_err(dev, + "The DDP package could not be loaded because its signature is not valid. Please use a valid DDP Package. Entering Safe Mode.\n"); + return; + case ICE_AQ_RC_ESVN: + dev_err(dev, + "The DDP Package could not be loaded because its security revision is too low. Please use an updated DDP Package. Entering Safe Mode.\n"); + return; + case ICE_AQ_RC_EBADMAN: + case ICE_AQ_RC_EBADBUF: + dev_err(dev, + "An error occurred on the device while loading the DDP package. The device will be reset.\n"); + return; + default: + break; + } + /* fall-through */ + default: + dev_err(dev, + "An unknown error (%d) occurred when loading the DDP package. Entering Safe Mode.\n", + *status); + break; + } +} + +/** + * ice_load_pkg - load/reload the DDP Package file + * @firmware: firmware structure when firmware requested or NULL for reload + * @pf: pointer to the PF instance + * + * Called on probe and post CORER/GLOBR rebuild to load DDP Package and + * initialize HW tables. + */ +static void +ice_load_pkg(const struct firmware *firmware, struct ice_pf *pf) +{ + enum ice_status status = ICE_ERR_PARAM; + struct device *dev = &pf->pdev->dev; + struct ice_hw *hw = &pf->hw; + + /* Load DDP Package */ + if (firmware && !hw->pkg_copy) { + status = ice_copy_and_init_pkg(hw, firmware->data, + firmware->size); + ice_log_pkg_init(hw, &status); + } else if (!firmware && hw->pkg_copy) { + /* Reload package during rebuild after CORER/GLOBR reset */ + status = ice_init_pkg(hw, hw->pkg_copy, hw->pkg_size); + ice_log_pkg_init(hw, &status); + } else { + dev_err(dev, + "The DDP package file failed to load. Entering Safe Mode.\n"); + } + + if (status) { + /* Safe Mode */ + clear_bit(ICE_FLAG_ADV_FEATURES, pf->flags); + return; + } + + /* Successful download package is the precondition for advanced + * features, hence setting the ICE_FLAG_ADV_FEATURES flag + */ + set_bit(ICE_FLAG_ADV_FEATURES, pf->flags); +} + +/** * ice_verify_cacheline_size - verify driver's assumption of 64 Byte cache lines * @pf: pointer to the PF structure * @@ -2272,6 +2683,105 @@ static void ice_verify_cacheline_size(struct ice_pf *pf) } /** + * ice_send_version - update firmware with driver version + * @pf: PF struct + * + * Returns ICE_SUCCESS on success, else error code + */ +static enum ice_status ice_send_version(struct ice_pf *pf) +{ + struct ice_driver_ver dv; + + dv.major_ver = DRV_VERSION_MAJOR; + dv.minor_ver = DRV_VERSION_MINOR; + dv.build_ver = DRV_VERSION_BUILD; + dv.subbuild_ver = 0; + strscpy((char *)dv.driver_string, DRV_VERSION, + sizeof(dv.driver_string)); + return ice_aq_send_driver_ver(&pf->hw, &dv, NULL); +} + +/** + * ice_get_opt_fw_name - return optional firmware file name or NULL + * @pf: pointer to the PF instance + */ +static char *ice_get_opt_fw_name(struct ice_pf *pf) +{ + /* Optional firmware name same as default with additional dash + * followed by a EUI-64 identifier (PCIe Device Serial Number) + */ + struct pci_dev *pdev = pf->pdev; + char *opt_fw_filename = NULL; + u32 dword; + u8 dsn[8]; + int pos; + + /* Determine the name of the optional file using the DSN (two + * dwords following the start of the DSN Capability). + */ + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_DSN); + if (pos) { + opt_fw_filename = kzalloc(NAME_MAX, GFP_KERNEL); + if (!opt_fw_filename) + return NULL; + + pci_read_config_dword(pdev, pos + 4, &dword); + put_unaligned_le32(dword, &dsn[0]); + pci_read_config_dword(pdev, pos + 8, &dword); + put_unaligned_le32(dword, &dsn[4]); + snprintf(opt_fw_filename, NAME_MAX, + "%sice-%02x%02x%02x%02x%02x%02x%02x%02x.pkg", + ICE_DDP_PKG_PATH, + dsn[7], dsn[6], dsn[5], dsn[4], + dsn[3], dsn[2], dsn[1], dsn[0]); + } + + return opt_fw_filename; +} + +/** + * ice_request_fw - Device initialization routine + * @pf: pointer to the PF instance + */ +static void ice_request_fw(struct ice_pf *pf) +{ + char *opt_fw_filename = ice_get_opt_fw_name(pf); + const struct firmware *firmware = NULL; + struct device *dev = &pf->pdev->dev; + int err = 0; + + /* optional device-specific DDP (if present) overrides the default DDP + * package file. kernel logs a debug message if the file doesn't exist, + * and warning messages for other errors. + */ + if (opt_fw_filename) { + err = firmware_request_nowarn(&firmware, opt_fw_filename, dev); + if (err) { + kfree(opt_fw_filename); + goto dflt_pkg_load; + } + + /* request for firmware was successful. Download to device */ + ice_load_pkg(firmware, pf); + kfree(opt_fw_filename); + release_firmware(firmware); + return; + } + +dflt_pkg_load: + err = request_firmware(&firmware, ICE_DDP_PKG_FILE, dev); + if (err) { + dev_err(dev, + "The DDP package file was not found or could not be read. Entering Safe Mode\n"); + return; + } + + /* request for firmware was successful. Download to device */ + ice_load_pkg(firmware, pf); + release_firmware(firmware); +} + +/** * ice_probe - Device initialization routine * @pdev: PCI device information struct * @ent: entry in ice_pci_tbl @@ -2345,22 +2855,33 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) goto err_exit_unroll; } - 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); + dev_info(dev, "firmware %d.%d.%d api %d.%d.%d nvm %s build 0x%08x\n", + hw->fw_maj_ver, hw->fw_min_ver, hw->fw_patch, + hw->api_maj_ver, hw->api_min_ver, hw->api_patch, + ice_nvm_version_str(hw), hw->fw_build); - ice_init_pf(pf); + ice_request_fw(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; + /* if ice_request_fw fails, ICE_FLAG_ADV_FEATURES bit won't be + * set in pf->state, which will cause ice_is_safe_mode to return + * true + */ + if (ice_is_safe_mode(pf)) { + dev_err(dev, + "Package download failed. Advanced features disabled - Device now in Safe Mode\n"); + /* we already got function/device capabilities but these don't + * reflect what the driver needs to do in safe mode. Instead of + * adding conditional logic everywhere to ignore these + * device/function capabilities, override them. + */ + ice_set_safe_mode_caps(hw); } - ice_determine_q_usage(pf); + err = ice_init_pf(pf); + if (err) { + dev_err(dev, "ice_init_pf failed: %d\n", err); + goto err_init_pf_unroll; + } pf->num_alloc_vsi = hw->func_caps.guar_num_vsi; if (!pf->num_alloc_vsi) { @@ -2390,12 +2911,10 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) * the misc functionality and queue processing is combined in * the same vector and that gets setup at open. */ - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - err = ice_req_irq_msix_misc(pf); - if (err) { - dev_err(dev, "setup of misc vector failed: %d\n", err); - goto err_init_interrupt_unroll; - } + err = ice_req_irq_msix_misc(pf); + if (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 */ @@ -2423,6 +2942,15 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) clear_bit(__ICE_SERVICE_DIS, pf->state); + /* tell the firmware we are up */ + err = ice_send_version(pf); + if (err) { + dev_err(dev, + "probe failed sending driver version %s. error: %d\n", + ice_drv_ver, err); + goto err_alloc_sw_unroll; + } + /* since everything is good, start the service timer */ mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period)); @@ -2434,6 +2962,20 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) ice_verify_cacheline_size(pf); + /* If no DDP driven features have to be setup, return here */ + if (ice_is_safe_mode(pf)) + return 0; + + /* initialize DDP driven features */ + + /* Note: DCB init failure is non-fatal to load */ + if (ice_init_pf_dcb(pf, false)) { + clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); + clear_bit(ICE_FLAG_DCB_ENA, pf->flags); + } else { + ice_cfg_lldp_mib_change(&pf->hw, true); + } + return 0; err_alloc_sw_unroll: @@ -2483,9 +3025,14 @@ static void ice_remove(struct pci_dev *pdev) continue; ice_vsi_free_q_vectors(pf->vsi[i]); } - ice_clear_interrupt_scheme(pf); ice_deinit_pf(pf); ice_deinit_hw(&pf->hw); + ice_clear_interrupt_scheme(pf); + /* Issue a PFR as part of the prescribed driver unload flow. Do not + * do it via ice_schedule_reset() since there is no need to rebuild + * and the service task is already stopped. + */ + ice_reset(&pf->hw, ICE_RESET_PFR); pci_disable_pcie_error_reporting(pdev); } @@ -2711,10 +3258,8 @@ static int ice_set_mac_address(struct net_device *netdev, void *pi) struct ice_hw *hw = &pf->hw; struct sockaddr *addr = pi; enum ice_status status; - LIST_HEAD(a_mac_list); - LIST_HEAD(r_mac_list); u8 flags = 0; - int err; + int err = 0; u8 *mac; mac = (u8 *)addr->sa_data; @@ -2737,42 +3282,23 @@ static int ice_set_mac_address(struct net_device *netdev, void *pi) /* 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 - * 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 - * added or removed from the filter. + * and then create a new filter rule using ice_add_mac via + * ice_vsi_cfg_mac_fltr function call for both add and/or remove + * filters. */ - err = ice_add_mac_to_list(vsi, &r_mac_list, netdev->dev_addr); - if (err) { - err = -EADDRNOTAVAIL; - goto free_lists; - } - - status = ice_remove_mac(hw, &r_mac_list); + status = ice_vsi_cfg_mac_fltr(vsi, netdev->dev_addr, false); if (status) { err = -EADDRNOTAVAIL; - goto free_lists; + goto err_update_filters; } - err = ice_add_mac_to_list(vsi, &a_mac_list, mac); - if (err) { - err = -EADDRNOTAVAIL; - goto free_lists; - } - - status = ice_add_mac(hw, &a_mac_list); + status = ice_vsi_cfg_mac_fltr(vsi, mac, true); if (status) { err = -EADDRNOTAVAIL; - goto free_lists; + goto err_update_filters; } -free_lists: - /* free list entries */ - ice_free_fltr_list(&pf->pdev->dev, &r_mac_list); - ice_free_fltr_list(&pf->pdev->dev, &a_mac_list); - +err_update_filters: if (err) { netdev_err(netdev, "can't set MAC %pM. filter update failed\n", mac); @@ -2788,8 +3314,8 @@ free_lists: 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", - mac); + netdev_err(netdev, "can't set MAC %pM. write to firmware failed error %d\n", + mac, status); } return 0; } @@ -2902,6 +3428,13 @@ ice_set_features(struct net_device *netdev, netdev_features_t features) struct ice_vsi *vsi = np->vsi; int ret = 0; + /* Don't set any netdev advanced features with device in Safe Mode */ + if (ice_is_safe_mode(vsi->back)) { + dev_err(&vsi->back->pdev->dev, + "Device is in Safe Mode - not enabling advanced netdev features\n"); + return ret; + } + /* Multiple features can be changed in one call so keep features in * separate if/else statements to guarantee each feature is checked */ @@ -3008,10 +3541,7 @@ static int ice_up_complete(struct ice_vsi *vsi) struct ice_pf *pf = vsi->back; int err; - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - ice_vsi_cfg_msix(vsi); - else - return -ENOTSUPP; + ice_vsi_cfg_msix(vsi); /* Enable only Rx rings, Tx rings were enabled by the FW when the * Tx queue group list was configured and the context bits were @@ -3132,7 +3662,7 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) * ice_update_vsi_stats - Update VSI stats counters * @vsi: the VSI to be updated */ -static void ice_update_vsi_stats(struct ice_vsi *vsi) +void ice_update_vsi_stats(struct ice_vsi *vsi) { struct rtnl_link_stats64 *cur_ns = &vsi->net_stats; struct ice_eth_stats *cur_es = &vsi->eth_stats; @@ -3159,6 +3689,8 @@ static void ice_update_vsi_stats(struct ice_vsi *vsi) cur_ns->rx_errors = pf->stats.crc_errors + pf->stats.illegal_bytes; cur_ns->rx_length_errors = pf->stats.rx_len_errors; + /* record drops from the port level */ + cur_ns->rx_missed_errors = pf->stats.eth.rx_discards; } } @@ -3166,149 +3698,139 @@ static void ice_update_vsi_stats(struct ice_vsi *vsi) * ice_update_pf_stats - Update PF port stats counters * @pf: PF whose stats needs to be updated */ -static void ice_update_pf_stats(struct ice_pf *pf) +void ice_update_pf_stats(struct ice_pf *pf) { struct ice_hw_port_stats *prev_ps, *cur_ps; struct ice_hw *hw = &pf->hw; - u8 pf_id; + u8 port; + port = hw->port_info->lport; prev_ps = &pf->stats_prev; cur_ps = &pf->stats; - pf_id = hw->pf_id; - ice_stat_update40(hw, GLPRT_GORCH(pf_id), GLPRT_GORCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.rx_bytes, + ice_stat_update40(hw, GLPRT_GORCL(port), pf->stat_prev_loaded, + &prev_ps->eth.rx_bytes, &cur_ps->eth.rx_bytes); - ice_stat_update40(hw, GLPRT_UPRCH(pf_id), GLPRT_UPRCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.rx_unicast, + ice_stat_update40(hw, GLPRT_UPRCL(port), pf->stat_prev_loaded, + &prev_ps->eth.rx_unicast, &cur_ps->eth.rx_unicast); - ice_stat_update40(hw, GLPRT_MPRCH(pf_id), GLPRT_MPRCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.rx_multicast, + ice_stat_update40(hw, GLPRT_MPRCL(port), pf->stat_prev_loaded, + &prev_ps->eth.rx_multicast, &cur_ps->eth.rx_multicast); - ice_stat_update40(hw, GLPRT_BPRCH(pf_id), GLPRT_BPRCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.rx_broadcast, + ice_stat_update40(hw, GLPRT_BPRCL(port), pf->stat_prev_loaded, + &prev_ps->eth.rx_broadcast, &cur_ps->eth.rx_broadcast); - ice_stat_update40(hw, GLPRT_GOTCH(pf_id), GLPRT_GOTCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.tx_bytes, + ice_stat_update32(hw, PRTRPB_RDPC, pf->stat_prev_loaded, + &prev_ps->eth.rx_discards, + &cur_ps->eth.rx_discards); + + ice_stat_update40(hw, GLPRT_GOTCL(port), pf->stat_prev_loaded, + &prev_ps->eth.tx_bytes, &cur_ps->eth.tx_bytes); - ice_stat_update40(hw, GLPRT_UPTCH(pf_id), GLPRT_UPTCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.tx_unicast, + ice_stat_update40(hw, GLPRT_UPTCL(port), pf->stat_prev_loaded, + &prev_ps->eth.tx_unicast, &cur_ps->eth.tx_unicast); - ice_stat_update40(hw, GLPRT_MPTCH(pf_id), GLPRT_MPTCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.tx_multicast, + ice_stat_update40(hw, GLPRT_MPTCL(port), pf->stat_prev_loaded, + &prev_ps->eth.tx_multicast, &cur_ps->eth.tx_multicast); - ice_stat_update40(hw, GLPRT_BPTCH(pf_id), GLPRT_BPTCL(pf_id), - pf->stat_prev_loaded, &prev_ps->eth.tx_broadcast, + ice_stat_update40(hw, GLPRT_BPTCL(port), pf->stat_prev_loaded, + &prev_ps->eth.tx_broadcast, &cur_ps->eth.tx_broadcast); - ice_stat_update32(hw, GLPRT_TDOLD(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_TDOLD(port), pf->stat_prev_loaded, &prev_ps->tx_dropped_link_down, &cur_ps->tx_dropped_link_down); - ice_stat_update40(hw, GLPRT_PRC64H(pf_id), GLPRT_PRC64L(pf_id), - pf->stat_prev_loaded, &prev_ps->rx_size_64, - &cur_ps->rx_size_64); + ice_stat_update40(hw, GLPRT_PRC64L(port), pf->stat_prev_loaded, + &prev_ps->rx_size_64, &cur_ps->rx_size_64); - ice_stat_update40(hw, GLPRT_PRC127H(pf_id), GLPRT_PRC127L(pf_id), - pf->stat_prev_loaded, &prev_ps->rx_size_127, - &cur_ps->rx_size_127); + ice_stat_update40(hw, GLPRT_PRC127L(port), pf->stat_prev_loaded, + &prev_ps->rx_size_127, &cur_ps->rx_size_127); - ice_stat_update40(hw, GLPRT_PRC255H(pf_id), GLPRT_PRC255L(pf_id), - pf->stat_prev_loaded, &prev_ps->rx_size_255, - &cur_ps->rx_size_255); + ice_stat_update40(hw, GLPRT_PRC255L(port), pf->stat_prev_loaded, + &prev_ps->rx_size_255, &cur_ps->rx_size_255); - ice_stat_update40(hw, GLPRT_PRC511H(pf_id), GLPRT_PRC511L(pf_id), - pf->stat_prev_loaded, &prev_ps->rx_size_511, - &cur_ps->rx_size_511); + ice_stat_update40(hw, GLPRT_PRC511L(port), pf->stat_prev_loaded, + &prev_ps->rx_size_511, &cur_ps->rx_size_511); - ice_stat_update40(hw, GLPRT_PRC1023H(pf_id), - GLPRT_PRC1023L(pf_id), pf->stat_prev_loaded, + ice_stat_update40(hw, GLPRT_PRC1023L(port), pf->stat_prev_loaded, &prev_ps->rx_size_1023, &cur_ps->rx_size_1023); - ice_stat_update40(hw, GLPRT_PRC1522H(pf_id), - GLPRT_PRC1522L(pf_id), pf->stat_prev_loaded, + ice_stat_update40(hw, GLPRT_PRC1522L(port), pf->stat_prev_loaded, &prev_ps->rx_size_1522, &cur_ps->rx_size_1522); - ice_stat_update40(hw, GLPRT_PRC9522H(pf_id), - GLPRT_PRC9522L(pf_id), pf->stat_prev_loaded, + ice_stat_update40(hw, GLPRT_PRC9522L(port), pf->stat_prev_loaded, &prev_ps->rx_size_big, &cur_ps->rx_size_big); - ice_stat_update40(hw, GLPRT_PTC64H(pf_id), GLPRT_PTC64L(pf_id), - pf->stat_prev_loaded, &prev_ps->tx_size_64, - &cur_ps->tx_size_64); + ice_stat_update40(hw, GLPRT_PTC64L(port), pf->stat_prev_loaded, + &prev_ps->tx_size_64, &cur_ps->tx_size_64); - ice_stat_update40(hw, GLPRT_PTC127H(pf_id), GLPRT_PTC127L(pf_id), - pf->stat_prev_loaded, &prev_ps->tx_size_127, - &cur_ps->tx_size_127); + ice_stat_update40(hw, GLPRT_PTC127L(port), pf->stat_prev_loaded, + &prev_ps->tx_size_127, &cur_ps->tx_size_127); - ice_stat_update40(hw, GLPRT_PTC255H(pf_id), GLPRT_PTC255L(pf_id), - pf->stat_prev_loaded, &prev_ps->tx_size_255, - &cur_ps->tx_size_255); + ice_stat_update40(hw, GLPRT_PTC255L(port), pf->stat_prev_loaded, + &prev_ps->tx_size_255, &cur_ps->tx_size_255); - ice_stat_update40(hw, GLPRT_PTC511H(pf_id), GLPRT_PTC511L(pf_id), - pf->stat_prev_loaded, &prev_ps->tx_size_511, - &cur_ps->tx_size_511); + ice_stat_update40(hw, GLPRT_PTC511L(port), pf->stat_prev_loaded, + &prev_ps->tx_size_511, &cur_ps->tx_size_511); - ice_stat_update40(hw, GLPRT_PTC1023H(pf_id), - GLPRT_PTC1023L(pf_id), pf->stat_prev_loaded, + ice_stat_update40(hw, GLPRT_PTC1023L(port), pf->stat_prev_loaded, &prev_ps->tx_size_1023, &cur_ps->tx_size_1023); - ice_stat_update40(hw, GLPRT_PTC1522H(pf_id), - GLPRT_PTC1522L(pf_id), pf->stat_prev_loaded, + ice_stat_update40(hw, GLPRT_PTC1522L(port), pf->stat_prev_loaded, &prev_ps->tx_size_1522, &cur_ps->tx_size_1522); - ice_stat_update40(hw, GLPRT_PTC9522H(pf_id), - GLPRT_PTC9522L(pf_id), pf->stat_prev_loaded, + ice_stat_update40(hw, GLPRT_PTC9522L(port), pf->stat_prev_loaded, &prev_ps->tx_size_big, &cur_ps->tx_size_big); - ice_stat_update32(hw, GLPRT_LXONRXC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_LXONRXC(port), pf->stat_prev_loaded, &prev_ps->link_xon_rx, &cur_ps->link_xon_rx); - ice_stat_update32(hw, GLPRT_LXOFFRXC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_LXOFFRXC(port), pf->stat_prev_loaded, &prev_ps->link_xoff_rx, &cur_ps->link_xoff_rx); - ice_stat_update32(hw, GLPRT_LXONTXC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_LXONTXC(port), pf->stat_prev_loaded, &prev_ps->link_xon_tx, &cur_ps->link_xon_tx); - ice_stat_update32(hw, GLPRT_LXOFFTXC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_LXOFFTXC(port), 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, + ice_stat_update32(hw, GLPRT_CRCERRS(port), pf->stat_prev_loaded, &prev_ps->crc_errors, &cur_ps->crc_errors); - ice_stat_update32(hw, GLPRT_ILLERRC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_ILLERRC(port), pf->stat_prev_loaded, &prev_ps->illegal_bytes, &cur_ps->illegal_bytes); - ice_stat_update32(hw, GLPRT_MLFC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_MLFC(port), pf->stat_prev_loaded, &prev_ps->mac_local_faults, &cur_ps->mac_local_faults); - ice_stat_update32(hw, GLPRT_MRFC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_MRFC(port), pf->stat_prev_loaded, &prev_ps->mac_remote_faults, &cur_ps->mac_remote_faults); - ice_stat_update32(hw, GLPRT_RLEC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_RLEC(port), pf->stat_prev_loaded, &prev_ps->rx_len_errors, &cur_ps->rx_len_errors); - ice_stat_update32(hw, GLPRT_RUC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_RUC(port), pf->stat_prev_loaded, &prev_ps->rx_undersize, &cur_ps->rx_undersize); - ice_stat_update32(hw, GLPRT_RFC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_RFC(port), pf->stat_prev_loaded, &prev_ps->rx_fragments, &cur_ps->rx_fragments); - ice_stat_update32(hw, GLPRT_ROC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_ROC(port), pf->stat_prev_loaded, &prev_ps->rx_oversize, &cur_ps->rx_oversize); - ice_stat_update32(hw, GLPRT_RJC(pf_id), pf->stat_prev_loaded, + ice_stat_update32(hw, GLPRT_RJC(port), pf->stat_prev_loaded, &prev_ps->rx_jabber, &cur_ps->rx_jabber); pf->stat_prev_loaded = true; @@ -3328,12 +3850,16 @@ void ice_get_stats64(struct net_device *netdev, struct rtnl_link_stats64 *stats) vsi_stats = &vsi->net_stats; - if (test_bit(__ICE_DOWN, vsi->state) || !vsi->num_txq || !vsi->num_rxq) + if (!vsi->num_txq || !vsi->num_rxq) return; + /* netdev packet/byte stats come from ring counter. These are obtained * by summing up ring counters (done by ice_update_vsi_ring_stats). + * But, only call the update routine and read the registers if VSI is + * not down. */ - ice_update_vsi_ring_stats(vsi); + if (!test_bit(__ICE_DOWN, vsi->state)) + ice_update_vsi_ring_stats(vsi); stats->tx_packets = vsi_stats->tx_packets; stats->tx_bytes = vsi_stats->tx_bytes; stats->rx_packets = vsi_stats->rx_packets; @@ -3372,85 +3898,6 @@ static void ice_napi_disable_all(struct ice_vsi *vsi) } /** - * ice_force_phys_link_state - Force the physical link state - * @vsi: VSI to force the physical link state to up/down - * @link_up: true/false indicates to set the physical link to up/down - * - * Force the physical link state by getting the current PHY capabilities from - * hardware and setting the PHY config based on the determined capabilities. If - * link changes a link event will be triggered because both the Enable Automatic - * Link Update and LESM Enable bits are set when setting the PHY capabilities. - * - * Returns 0 on success, negative on failure - */ -static int ice_force_phys_link_state(struct ice_vsi *vsi, bool link_up) -{ - struct ice_aqc_get_phy_caps_data *pcaps; - struct ice_aqc_set_phy_cfg_data *cfg; - struct ice_port_info *pi; - struct device *dev; - int retcode; - - if (!vsi || !vsi->port_info || !vsi->back) - return -EINVAL; - if (vsi->type != ICE_VSI_PF) - return 0; - - dev = &vsi->back->pdev->dev; - - pi = vsi->port_info; - - pcaps = devm_kzalloc(dev, sizeof(*pcaps), GFP_KERNEL); - if (!pcaps) - return -ENOMEM; - - retcode = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_SW_CFG, pcaps, - NULL); - if (retcode) { - dev_err(dev, - "Failed to get phy capabilities, VSI %d error %d\n", - vsi->vsi_num, retcode); - retcode = -EIO; - goto out; - } - - /* No change in link */ - if (link_up == !!(pcaps->caps & ICE_AQC_PHY_EN_LINK) && - link_up == !!(pi->phy.link_info.link_info & ICE_AQ_LINK_UP)) - goto out; - - cfg = devm_kzalloc(dev, sizeof(*cfg), GFP_KERNEL); - if (!cfg) { - retcode = -ENOMEM; - goto out; - } - - cfg->phy_type_low = pcaps->phy_type_low; - cfg->phy_type_high = pcaps->phy_type_high; - cfg->caps = pcaps->caps | ICE_AQ_PHY_ENA_AUTO_LINK_UPDT; - cfg->low_power_ctrl = pcaps->low_power_ctrl; - cfg->eee_cap = pcaps->eee_cap; - cfg->eeer_value = pcaps->eeer_value; - cfg->link_fec_opt = pcaps->link_fec_options; - if (link_up) - cfg->caps |= ICE_AQ_PHY_ENA_LINK; - else - cfg->caps &= ~ICE_AQ_PHY_ENA_LINK; - - retcode = ice_aq_set_phy_cfg(&vsi->back->hw, pi->lport, cfg, NULL); - if (retcode) { - dev_err(dev, "Failed to set phy config, VSI %d error %d\n", - vsi->vsi_num, retcode); - retcode = -EIO; - } - - devm_kfree(dev, cfg); -out: - devm_kfree(dev, pcaps); - return retcode; -} - -/** * ice_down - Shutdown the connection * @vsi: The VSI being stopped */ @@ -3559,24 +4006,6 @@ int ice_vsi_setup_rx_rings(struct ice_vsi *vsi) } /** - * ice_vsi_req_irq - Request IRQ from the OS - * @vsi: The VSI IRQ is being requested for - * @basename: name for the vector - * - * Return 0 on success and a negative value on error - */ -static int ice_vsi_req_irq(struct ice_vsi *vsi, char *basename) -{ - struct ice_pf *pf = vsi->back; - int err = -EINVAL; - - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - err = ice_vsi_req_irq_msix(vsi, basename); - - return err; -} - -/** * ice_vsi_open - Called when a network interface is made active * @vsi: the VSI to open * @@ -3605,7 +4034,7 @@ static int ice_vsi_open(struct ice_vsi *vsi) snprintf(int_name, sizeof(int_name) - 1, "%s-%s", dev_driver_string(&pf->pdev->dev), vsi->netdev->name); - err = ice_vsi_req_irq(vsi, int_name); + err = ice_vsi_req_irq_msix(vsi, int_name); if (err) goto err_setup_rx; @@ -3669,23 +4098,19 @@ static int ice_ena_vsi(struct ice_vsi *vsi, bool locked) int err = 0; if (!test_bit(__ICE_NEEDS_RESTART, vsi->state)) - return err; + return 0; clear_bit(__ICE_NEEDS_RESTART, vsi->state); if (vsi->netdev && vsi->type == ICE_VSI_PF) { - struct net_device *netd = vsi->netdev; - if (netif_running(vsi->netdev)) { - if (locked) { - err = netd->netdev_ops->ndo_open(netd); - } else { + if (!locked) rtnl_lock(); - err = netd->netdev_ops->ndo_open(netd); + + err = ice_open(vsi->netdev); + + if (!locked) rtnl_unlock(); - } - } else { - err = ice_vsi_open(vsi); } } @@ -3699,9 +4124,6 @@ static int ice_ena_vsi(struct ice_vsi *vsi, bool locked) */ #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; @@ -3712,91 +4134,107 @@ static int ice_pf_ena_all_vsi(struct ice_pf *pf, bool locked) return 0; } +#endif /* CONFIG_DCB */ /** - * ice_vsi_rebuild_all - rebuild all VSIs in PF - * @pf: the PF + * ice_vsi_rebuild_by_type - Rebuild VSI of a given type + * @pf: pointer to the PF instance + * @type: VSI type to rebuild + * + * Iterates through the pf->vsi array and rebuilds VSIs of the requested type */ -static int ice_vsi_rebuild_all(struct ice_pf *pf) +static int ice_vsi_rebuild_by_type(struct ice_pf *pf, enum ice_vsi_type type) { - int i; + enum ice_status status; + int i, err; - /* loop through pf->vsi array and reinit the VSI if found */ ice_for_each_vsi(pf, i) { - int err; + struct ice_vsi *vsi = pf->vsi[i]; - if (!pf->vsi[i]) + if (!vsi || vsi->type != type) continue; - err = ice_vsi_rebuild(pf->vsi[i]); + /* rebuild the VSI */ + err = ice_vsi_rebuild(vsi); if (err) { dev_err(&pf->pdev->dev, - "VSI at index %d rebuild failed\n", - pf->vsi[i]->idx); + "rebuild VSI failed, err %d, VSI index %d, type %d\n", + err, vsi->idx, type); return err; } - dev_info(&pf->pdev->dev, - "VSI at index %d rebuilt. vsi_num = 0x%x\n", - pf->vsi[i]->idx, pf->vsi[i]->vsi_num); + /* replay filters for the VSI */ + status = ice_replay_vsi(&pf->hw, vsi->idx); + if (status) { + dev_err(&pf->pdev->dev, + "replay VSI failed, status %d, VSI index %d, type %d\n", + status, vsi->idx, type); + return -EIO; + } + + /* Re-map HW VSI number, using VSI handle that has been + * previously validated in ice_replay_vsi() call above + */ + vsi->vsi_num = ice_get_hw_vsi_num(&pf->hw, vsi->idx); + + /* enable the VSI */ + err = ice_ena_vsi(vsi, false); + if (err) { + dev_err(&pf->pdev->dev, + "enable VSI failed, err %d, VSI index %d, type %d\n", + err, vsi->idx, type); + return err; + } + + dev_info(&pf->pdev->dev, "VSI rebuilt. VSI index %d, type %d\n", + vsi->idx, type); } return 0; } /** - * ice_vsi_replay_all - replay all VSIs configuration in the PF - * @pf: the PF + * ice_update_pf_netdev_link - Update PF netdev link status + * @pf: pointer to the PF instance */ -static int ice_vsi_replay_all(struct ice_pf *pf) +static void ice_update_pf_netdev_link(struct ice_pf *pf) { - struct ice_hw *hw = &pf->hw; - enum ice_status ret; + bool link_up; int i; - /* loop through pf->vsi array and replay the VSI if found */ ice_for_each_vsi(pf, i) { - if (!pf->vsi[i]) - continue; + struct ice_vsi *vsi = pf->vsi[i]; - ret = ice_replay_vsi(hw, pf->vsi[i]->idx); - if (ret) { - dev_err(&pf->pdev->dev, - "VSI at index %d replay failed %d\n", - pf->vsi[i]->idx, ret); - return -EIO; - } - - /* Re-map HW VSI number, using VSI handle that has been - * previously validated in ice_replay_vsi() call above - */ - pf->vsi[i]->vsi_num = ice_get_hw_vsi_num(hw, pf->vsi[i]->idx); + if (!vsi || vsi->type != ICE_VSI_PF) + return; - dev_info(&pf->pdev->dev, - "VSI at index %d filter replayed successfully - vsi_num %i\n", - pf->vsi[i]->idx, pf->vsi[i]->vsi_num); + ice_get_link_status(pf->vsi[i]->port_info, &link_up); + if (link_up) { + netif_carrier_on(pf->vsi[i]->netdev); + netif_tx_wake_all_queues(pf->vsi[i]->netdev); + } else { + netif_carrier_off(pf->vsi[i]->netdev); + netif_tx_stop_all_queues(pf->vsi[i]->netdev); + } } - - /* Clean up replay filter after successful re-configuration */ - ice_replay_post(hw); - return 0; } /** * ice_rebuild - rebuild after reset * @pf: PF to rebuild + * @reset_type: type of reset */ -static void ice_rebuild(struct ice_pf *pf) +static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) { struct device *dev = &pf->pdev->dev; struct ice_hw *hw = &pf->hw; enum ice_status ret; - int err, i; + int err; if (test_bit(__ICE_DOWN, pf->state)) goto clear_recovery; - dev_dbg(dev, "rebuilding PF\n"); + dev_dbg(dev, "rebuilding PF after reset_type=%d\n", reset_type); ret = ice_init_all_ctrlq(hw); if (ret) { @@ -3804,6 +4242,16 @@ static void ice_rebuild(struct ice_pf *pf) goto err_init_ctrlq; } + /* if DDP was previously loaded successfully */ + if (!ice_is_safe_mode(pf)) { + /* reload the SW DB of filter tables */ + if (reset_type == ICE_RESET_PFR) + ice_fill_blk_tbls(hw); + else + /* Reload DDP Package after CORER/GLOBR reset */ + ice_load_pkg(NULL, pf); + } + ret = ice_clear_pf_cfg(hw); if (ret) { dev_err(dev, "clear PF configuration failed %d\n", ret); @@ -3822,65 +4270,53 @@ static void ice_rebuild(struct ice_pf *pf) if (err) goto err_sched_init_port; - ice_dcb_rebuild(pf); + err = ice_update_link_info(hw->port_info); + if (err) + dev_err(&pf->pdev->dev, "Get link status error %d\n", err); - err = ice_vsi_rebuild_all(pf); + /* start misc vector */ + err = ice_req_irq_msix_misc(pf); if (err) { - dev_err(dev, "ice_vsi_rebuild_all failed\n"); - goto err_vsi_rebuild; + dev_err(dev, "misc vector setup failed: %d\n", err); + goto err_sched_init_port; } - err = ice_update_link_info(hw->port_info); - if (err) - dev_err(&pf->pdev->dev, "Get link status error %d\n", err); + if (test_bit(ICE_FLAG_DCB_ENA, pf->flags)) + ice_dcb_rebuild(pf); - /* Replay all VSIs Configuration, including filters after reset */ - if (ice_vsi_replay_all(pf)) { - dev_err(&pf->pdev->dev, - "error replaying VSI configurations with switch filter rules\n"); + /* rebuild PF VSI */ + err = ice_vsi_rebuild_by_type(pf, ICE_VSI_PF); + if (err) { + dev_err(dev, "PF VSI rebuild failed: %d\n", err); goto err_vsi_rebuild; } - /* start misc vector */ - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) { - err = ice_req_irq_msix_misc(pf); + if (test_bit(ICE_FLAG_SRIOV_ENA, pf->flags)) { + err = ice_vsi_rebuild_by_type(pf, ICE_VSI_VF); if (err) { - dev_err(dev, "misc vector setup failed: %d\n", err); + dev_err(dev, "VF VSI rebuild failed: %d\n", err); goto err_vsi_rebuild; } } - /* restart the VSIs that were rebuilt and running before the reset */ - 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() - * since its already taken care in ice_vsi_open() - */ + ice_update_pf_netdev_link(pf); + + /* tell the firmware we are up */ + ret = ice_send_version(pf); + if (ret) { + dev_err(dev, + "Rebuild failed due to error sending driver version: %d\n", + ret); goto err_vsi_rebuild; } - ice_for_each_vsi(pf, i) { - bool link_up; - - if (!pf->vsi[i] || pf->vsi[i]->type != ICE_VSI_PF) - continue; - ice_get_link_status(pf->vsi[i]->port_info, &link_up); - if (link_up) { - netif_carrier_on(pf->vsi[i]->netdev); - netif_tx_wake_all_queues(pf->vsi[i]->netdev); - } else { - netif_carrier_off(pf->vsi[i]->netdev); - netif_tx_stop_all_queues(pf->vsi[i]->netdev); - } - } + ice_replay_post(hw); /* if we get here, reset flow is successful */ clear_bit(__ICE_RESET_FAILED, pf->state); return; err_vsi_rebuild: - ice_vsi_release_all(pf); err_sched_init_port: ice_sched_cleanup_all(hw); err_init_ctrlq: @@ -4244,9 +4680,7 @@ static void ice_tx_timeout(struct net_device *netdev) head = (rd32(hw, QTX_COMM_HEAD(vsi->txq_map[hung_queue])) & QTX_COMM_HEAD_HEAD_M) >> QTX_COMM_HEAD_HEAD_S; /* Read interrupt register */ - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - val = rd32(hw, - GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); + val = rd32(hw, 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, @@ -4295,6 +4729,7 @@ int ice_open(struct net_device *netdev) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; + struct ice_port_info *pi; int err; if (test_bit(__ICE_NEEDS_RESTART, vsi->back->state)) { @@ -4304,13 +4739,33 @@ int ice_open(struct net_device *netdev) netif_carrier_off(netdev); - err = ice_force_phys_link_state(vsi, true); + pi = vsi->port_info; + err = ice_update_link_info(pi); if (err) { - netdev_err(netdev, - "Failed to set physical link up, error %d\n", err); + netdev_err(netdev, "Failed to get link info, error %d\n", + err); return err; } + /* Set PHY if there is media, otherwise, turn off PHY */ + if (pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) { + err = ice_force_phys_link_state(vsi, true); + if (err) { + netdev_err(netdev, + "Failed to set physical link up, error %d\n", + err); + return err; + } + } else { + err = ice_aq_set_link_restart_an(pi, false, NULL); + if (err) { + netdev_err(netdev, "Failed to set PHY state, VSI %d error %d\n", + vsi->vsi_num, err); + return err; + } + set_bit(ICE_FLAG_NO_MEDIA, vsi->back->flags); + } + err = ice_vsi_open(vsi); if (err) netdev_err(netdev, "Failed to open VSI 0x%04X on switch 0x%04X\n", @@ -4388,6 +4843,17 @@ out_rm_features: return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); } +static const struct net_device_ops ice_netdev_safe_mode_ops = { + .ndo_open = ice_open, + .ndo_stop = ice_stop, + .ndo_start_xmit = ice_start_xmit, + .ndo_set_mac_address = ice_set_mac_address, + .ndo_validate_addr = eth_validate_addr, + .ndo_change_mtu = ice_change_mtu, + .ndo_get_stats64 = ice_get_stats64, + .ndo_tx_timeout = ice_tx_timeout, +}; + static const struct net_device_ops ice_netdev_ops = { .ndo_open = ice_open, .ndo_stop = ice_stop, diff --git a/drivers/net/ethernet/intel/ice/ice_sched.c b/drivers/net/ethernet/intel/ice/ice_sched.c index 2a232504379d..fc624b73d05d 100644 --- a/drivers/net/ethernet/intel/ice/ice_sched.c +++ b/drivers/net/ethernet/intel/ice/ice_sched.c @@ -260,33 +260,17 @@ 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 + * @pi: port information structure * @parent: pointer the base node of the subtree * @layer: layer number * * This function retrieves the first node of the given layer from the subtree */ static struct ice_sched_node * -ice_sched_get_first_node(struct ice_hw *hw, struct ice_sched_node *parent, - u8 layer) +ice_sched_get_first_node(struct ice_port_info *pi, + struct ice_sched_node *parent, u8 layer) { - u8 i; - - if (layer < hw->sw_entry_point_layer) - return NULL; - for (i = 0; i < parent->num_children; i++) { - struct ice_sched_node *node = parent->children[i]; - - if (node) { - if (node->tx_sched_layer == layer) - return node; - /* this recursion is intentional, and wouldn't - * go more than 9 calls - */ - return ice_sched_get_first_node(hw, node, layer); - } - } - return NULL; + return pi->sib_head[parent->tc_num][layer]; } /** @@ -300,7 +284,7 @@ struct ice_sched_node *ice_sched_get_tc_node(struct ice_port_info *pi, u8 tc) { u8 i; - if (!pi) + if (!pi || !pi->root) return NULL; for (i = 0; i < pi->root->num_children; i++) if (pi->root->children[i]->tc_num == tc) @@ -342,7 +326,7 @@ void ice_free_sched_node(struct ice_port_info *pi, struct ice_sched_node *node) parent = node->parent; /* root has no parent */ if (parent) { - struct ice_sched_node *p, *tc_node; + struct ice_sched_node *p; /* update the parent */ for (i = 0; i < parent->num_children; i++) @@ -354,16 +338,7 @@ void ice_free_sched_node(struct ice_port_info *pi, struct ice_sched_node *node) break; } - /* search for previous sibling that points to this node and - * remove the reference - */ - tc_node = ice_sched_get_tc_node(pi, node->tc_num); - if (!tc_node) { - ice_debug(hw, ICE_DBG_SCHED, - "Invalid TC number %d\n", node->tc_num); - goto err_exit; - } - p = ice_sched_get_first_node(hw, tc_node, node->tx_sched_layer); + p = ice_sched_get_first_node(pi, node, node->tx_sched_layer); while (p) { if (p->sibling == node) { p->sibling = node->sibling; @@ -371,8 +346,13 @@ void ice_free_sched_node(struct ice_port_info *pi, struct ice_sched_node *node) } p = p->sibling; } + + /* update the sibling head if head is getting removed */ + if (pi->sib_head[node->tc_num][node->tx_sched_layer] == node) + pi->sib_head[node->tc_num][node->tx_sched_layer] = + node->sibling; } -err_exit: + /* leaf nodes have no children */ if (node->children) devm_kfree(ice_hw_to_dev(hw), node->children); @@ -743,13 +723,17 @@ ice_sched_add_elems(struct ice_port_info *pi, struct ice_sched_node *tc_node, /* add it to previous node sibling pointer */ /* Note: siblings are not linked across branches */ - prev = ice_sched_get_first_node(hw, tc_node, layer); + prev = ice_sched_get_first_node(pi, tc_node, layer); if (prev && prev != new_node) { while (prev->sibling) prev = prev->sibling; prev->sibling = new_node; } + /* initialize the sibling head */ + if (!pi->sib_head[tc_node->tc_num][layer]) + pi->sib_head[tc_node->tc_num][layer] = new_node; + if (i == 0) *first_node_teid = teid; } @@ -1160,7 +1144,7 @@ ice_sched_get_free_qparent(struct ice_port_info *pi, u16 vsi_handle, u8 tc, goto lan_q_exit; /* get the first queue group node from VSI sub-tree */ - qgrp_node = ice_sched_get_first_node(pi->hw, vsi_node, qgrp_layer); + qgrp_node = ice_sched_get_first_node(pi, vsi_node, qgrp_layer); while (qgrp_node) { /* make sure the qgroup node is part of the VSI subtree */ if (ice_sched_find_node_in_subtree(pi->hw, vsi_node, qgrp_node)) @@ -1191,7 +1175,7 @@ ice_sched_get_vsi_node(struct ice_hw *hw, struct ice_sched_node *tc_node, u8 vsi_layer; vsi_layer = ice_sched_get_vsi_layer(hw); - node = ice_sched_get_first_node(hw, tc_node, vsi_layer); + node = ice_sched_get_first_node(hw->port_info, tc_node, vsi_layer); /* Check whether it already exists */ while (node) { @@ -1316,7 +1300,8 @@ ice_sched_calc_vsi_support_nodes(struct ice_hw *hw, /* If intermediate nodes are reached max children * then add a new one. */ - node = ice_sched_get_first_node(hw, tc_node, (u8)i); + node = ice_sched_get_first_node(hw->port_info, tc_node, + (u8)i); /* scan all the siblings */ while (node) { if (node->num_children < hw->max_children[i]) diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c index 8271fd651725..1acdd43a2edd 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.c +++ b/drivers/net/ethernet/intel/ice/ice_switch.c @@ -1623,12 +1623,13 @@ ice_remove_rule_internal(struct ice_hw *hw, u8 recp_id, status = ice_aq_sw_rules(hw, s_rule, ICE_SW_RULE_RX_TX_NO_HDR_SIZE, 1, ice_aqc_opc_remove_sw_rules, NULL); - if (status) - goto exit; /* Remove a book keeping from the list */ devm_kfree(ice_hw_to_dev(hw), s_rule); + if (status) + goto exit; + list_del(&list_elem->list_entry); devm_kfree(ice_hw_to_dev(hw), list_elem); } @@ -2137,6 +2138,38 @@ out: } /** + * ice_find_ucast_rule_entry - Search for a unicast MAC filter rule entry + * @hw: pointer to the hardware structure + * @recp_id: lookup type for which the specified rule needs to be searched + * @f_info: rule information + * + * Helper function to search for a unicast rule entry - this is to be used + * to remove unicast MAC filter that is not shared with other VSIs on the + * PF switch. + * + * Returns pointer to entry storing the rule if found + */ +static struct ice_fltr_mgmt_list_entry * +ice_find_ucast_rule_entry(struct ice_hw *hw, u8 recp_id, + struct ice_fltr_info *f_info) +{ + struct ice_switch_info *sw = hw->switch_info; + struct ice_fltr_mgmt_list_entry *list_itr; + struct list_head *list_head; + + list_head = &sw->recp_list[recp_id].filt_rules; + list_for_each_entry(list_itr, list_head, list_entry) { + if (!memcmp(&f_info->l_data, &list_itr->fltr_info.l_data, + sizeof(f_info->l_data)) && + f_info->fwd_id.hw_vsi_id == + list_itr->fltr_info.fwd_id.hw_vsi_id && + f_info->flag == list_itr->fltr_info.flag) + return list_itr; + } + return NULL; +} + +/** * ice_remove_mac - remove a MAC address based filter rule * @hw: pointer to the hardware structure * @m_list: list of MAC addresses and forwarding information @@ -2153,15 +2186,39 @@ enum ice_status ice_remove_mac(struct ice_hw *hw, struct list_head *m_list) { struct ice_fltr_list_entry *list_itr, *tmp; + struct mutex *rule_lock; /* Lock to protect filter rule list */ if (!m_list) return ICE_ERR_PARAM; + rule_lock = &hw->switch_info->recp_list[ICE_SW_LKUP_MAC].filt_rule_lock; list_for_each_entry_safe(list_itr, tmp, m_list, list_entry) { enum ice_sw_lkup_type l_type = list_itr->fltr_info.lkup_type; + u8 *add = &list_itr->fltr_info.l_data.mac.mac_addr[0]; + u16 vsi_handle; if (l_type != ICE_SW_LKUP_MAC) return ICE_ERR_PARAM; + + vsi_handle = list_itr->fltr_info.vsi_handle; + if (!ice_is_vsi_valid(hw, vsi_handle)) + return ICE_ERR_PARAM; + + list_itr->fltr_info.fwd_id.hw_vsi_id = + ice_get_hw_vsi_num(hw, vsi_handle); + if (is_unicast_ether_addr(add) && !hw->ucast_shared) { + /* Don't remove the unicast address that belongs to + * another VSI on the switch, since it is not being + * shared... + */ + mutex_lock(rule_lock); + if (!ice_find_ucast_rule_entry(hw, ICE_SW_LKUP_MAC, + &list_itr->fltr_info)) { + mutex_unlock(rule_lock); + return ICE_ERR_DOES_NOT_EXIST; + } + mutex_unlock(rule_lock); + } list_itr->status = ice_remove_rule_internal(hw, ICE_SW_LKUP_MAC, list_itr); diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 3c83230434b6..33dd103035dc 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -95,17 +95,16 @@ void ice_free_tx_ring(struct ice_ring *tx_ring) /** * ice_clean_tx_irq - Reclaim resources after transmit completes - * @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll * * 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_ring *tx_ring, int napi_budget) { unsigned int total_bytes = 0, total_pkts = 0; - unsigned int budget = vsi->work_lmt; + unsigned int budget = ICE_DFLT_IRQ_WORK; + struct ice_vsi *vsi = tx_ring->vsi; s16 i = tx_ring->next_to_clean; struct ice_tx_desc *tx_desc; struct ice_tx_buf *tx_buf; @@ -114,6 +113,8 @@ ice_clean_tx_irq(struct ice_vsi *vsi, struct ice_ring *tx_ring, int napi_budget) tx_desc = ICE_TX_DESC(tx_ring, i); i -= tx_ring->count; + prefetch(&vsi->state); + do { struct ice_tx_desc *eop_desc = tx_buf->next_to_watch; @@ -206,7 +207,7 @@ ice_clean_tx_irq(struct ice_vsi *vsi, struct ice_ring *tx_ring, int napi_budget) smp_mb(); if (__netif_subqueue_stopped(tx_ring->netdev, tx_ring->q_index) && - !test_bit(__ICE_DOWN, vsi->state)) { + !test_bit(__ICE_DOWN, vsi->state)) { netif_wake_subqueue(tx_ring->netdev, tx_ring->q_index); ++tx_ring->tx_stats.restart_q; @@ -377,18 +378,28 @@ err: */ static void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val) { + u16 prev_ntu = rx_ring->next_to_use; + rx_ring->next_to_use = val; /* update next to alloc since we have filled the ring */ rx_ring->next_to_alloc = val; - /* Force memory writes to complete before letting h/w - * know there are new descriptors to fetch. (Only - * applicable for weak-ordered memory model archs, - * such as IA-64). + /* QRX_TAIL will be updated with any tail value, but hardware ignores + * the lower 3 bits. This makes it so we only bump tail on meaningful + * boundaries. Also, this allows us to bump tail on intervals of 8 up to + * the budget depending on the current traffic load. */ - wmb(); - writel(val, rx_ring->tail); + val &= ~0x7; + if (prev_ntu != val) { + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). + */ + wmb(); + writel(val, rx_ring->tail); + } } /** @@ -445,7 +456,13 @@ ice_alloc_mapped_page(struct ice_ring *rx_ring, struct ice_rx_buf *bi) * @rx_ring: ring to place buffers on * @cleaned_count: number of buffers to replace * - * Returns false if all allocations were successful, true if any fail + * Returns false if all allocations were successful, true if any fail. Returning + * true signals to the caller that we didn't replace cleaned_count buffers and + * there is more work to do. + * + * First, try to clean "cleaned_count" Rx buffers. Then refill the cleaned Rx + * buffers. Then bump tail at most one time. Grouping like this lets us avoid + * multiple tail writes per call. */ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count) { @@ -462,8 +479,9 @@ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count) bi = &rx_ring->rx_buf[ntu]; do { + /* if we fail here, we have work remaining */ if (!ice_alloc_mapped_page(rx_ring, bi)) - goto no_bufs; + break; /* sync the buffer for use by the device */ dma_sync_single_range_for_device(rx_ring->dev, bi->dma, @@ -494,16 +512,7 @@ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count) if (rx_ring->next_to_use != ntu) ice_release_rx_desc(rx_ring, ntu); - return false; - -no_bufs: - if (rx_ring->next_to_use != ntu) - ice_release_rx_desc(rx_ring, ntu); - - /* make sure to come back via polling to try again after - * allocation failure - */ - return true; + return !!cleaned_count; } /** @@ -599,6 +608,8 @@ ice_add_rx_frag(struct ice_rx_buf *rx_buf, struct sk_buff *skb, unsigned int truesize = ICE_RXBUF_2048; #endif + if (!size) + return; skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page, rx_buf->page_offset, size, truesize); @@ -654,6 +665,8 @@ ice_get_rx_buf(struct ice_ring *rx_ring, struct sk_buff **skb, prefetchw(rx_buf->page); *skb = rx_buf->skb; + if (!size) + return rx_buf; /* 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, @@ -737,8 +750,11 @@ ice_construct_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, */ static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) { - /* hand second half of page back to the ring */ + if (!rx_buf) + return; + if (ice_can_reuse_rx_page(rx_buf)) { + /* hand second half of page back to the ring */ ice_reuse_rx_page(rx_ring, rx_buf); rx_ring->rx_stats.page_reuse_count++; } else { @@ -864,7 +880,7 @@ ice_rx_hash(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, /** * ice_rx_csum - Indicate in skb if checksum is good - * @vsi: the VSI we care about + * @ring: the ring we care about * @skb: skb currently being received and modified * @rx_desc: the receive descriptor * @ptype: the packet type decoded by hardware @@ -872,7 +888,7 @@ 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, +ice_rx_csum(struct ice_ring *ring, struct sk_buff *skb, union ice_32b_rx_flex_desc *rx_desc, u8 ptype) { struct ice_rx_ptype_decoded decoded; @@ -889,7 +905,7 @@ ice_rx_csum(struct ice_vsi *vsi, struct sk_buff *skb, skb_checksum_none_assert(skb); /* check if Rx checksum is enabled */ - if (!(vsi->netdev->features & NETIF_F_RXCSUM)) + if (!(ring->netdev->features & NETIF_F_RXCSUM)) return; /* check if HW has decoded the packet and checksum */ @@ -929,7 +945,7 @@ ice_rx_csum(struct ice_vsi *vsi, struct sk_buff *skb, return; checksum_fail: - vsi->back->hw_csum_rx_error++; + ring->vsi->back->hw_csum_rx_error++; } /** @@ -953,7 +969,7 @@ ice_process_skb_fields(struct ice_ring *rx_ring, /* modifies the skb - consumes the enet header */ skb->protocol = eth_type_trans(skb, rx_ring->netdev); - ice_rx_csum(rx_ring->vsi, skb, rx_desc, ptype); + ice_rx_csum(rx_ring, skb, rx_desc, ptype); } /** @@ -990,7 +1006,7 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) { unsigned int total_rx_bytes = 0, total_rx_pkts = 0; u16 cleaned_count = ICE_DESC_UNUSED(rx_ring); - bool failure = false; + bool failure; /* start the loop to process Rx packets bounded by 'budget' */ while (likely(total_rx_pkts < (unsigned int)budget)) { @@ -1002,13 +1018,6 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) u16 vlan_tag = 0; u8 rx_ptype; - /* return some buffers to hardware, one at a time is too slow */ - if (cleaned_count >= ICE_RX_BUF_WRITE) { - failure = failure || - ice_alloc_rx_bufs(rx_ring, cleaned_count); - cleaned_count = 0; - } - /* get the Rx desc from Rx ring based on 'next_to_clean' */ rx_desc = ICE_RX_DESC(rx_ring, rx_ring->next_to_clean); @@ -1030,8 +1039,9 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) size = le16_to_cpu(rx_desc->wb.pkt_len) & ICE_RX_FLX_DESC_PKT_LEN_M; + /* retrieve a buffer from the ring */ rx_buf = ice_get_rx_buf(rx_ring, &skb, size); - /* allocate (if needed) and populate skb */ + if (skb) ice_add_rx_frag(rx_buf, skb, size); else @@ -1040,7 +1050,8 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) /* exit if we failed to retrieve a buffer */ if (!skb) { rx_ring->rx_stats.alloc_buf_failed++; - rx_buf->pagecnt_bias++; + if (rx_buf) + rx_buf->pagecnt_bias++; break; } @@ -1057,9 +1068,6 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) continue; } - rx_ptype = le16_to_cpu(rx_desc->wb.ptype_flex_flags0) & - ICE_RX_FLEX_DESC_PTYPE_M; - stat_err_bits = BIT(ICE_RX_FLEX_DESC_STATUS0_L2TAG1P_S); if (ice_test_staterr(rx_desc, stat_err_bits)) vlan_tag = le16_to_cpu(rx_desc->wb.l2tag1); @@ -1076,6 +1084,9 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) total_rx_bytes += skb->len; /* populate checksum, VLAN, and protocol */ + rx_ptype = le16_to_cpu(rx_desc->wb.ptype_flex_flags0) & + ICE_RX_FLEX_DESC_PTYPE_M; + ice_process_skb_fields(rx_ring, rx_desc, skb, rx_ptype); /* send completed skb up the stack */ @@ -1085,6 +1096,9 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) total_rx_pkts++; } + /* return up to cleaned_count buffers to hardware */ + failure = ice_alloc_rx_bufs(rx_ring, cleaned_count); + /* update queue and vector specific stats */ u64_stats_update_begin(&rx_ring->syncp); rx_ring->stats.pkts += total_rx_pkts; @@ -1212,6 +1226,8 @@ ice_update_itr(struct ice_q_vector *q_vector, struct ice_ring_container *rc) if (time_after(next_update, rc->next_update)) goto clear_counts; + prefetch(q_vector->vsi->port_info); + packets = rc->total_pkts; bytes = rc->total_bytes; @@ -1341,16 +1357,32 @@ static u32 ice_buildreg_itr(u16 itr_idx, u16 itr) /** * ice_update_ena_itr - Update ITR and re-enable MSIX interrupt - * @vsi: the VSI associated with the q_vector * @q_vector: q_vector for which ITR is being updated and interrupt enabled */ -static void -ice_update_ena_itr(struct ice_vsi *vsi, struct ice_q_vector *q_vector) +static void ice_update_ena_itr(struct ice_q_vector *q_vector) { struct ice_ring_container *tx = &q_vector->tx; struct ice_ring_container *rx = &q_vector->rx; + struct ice_vsi *vsi = q_vector->vsi; u32 itr_val; + /* when exiting WB_ON_ITR lets set a low ITR value and trigger + * interrupts to expire right away in case we have more work ready to go + * already + */ + if (q_vector->itr_countdown == ICE_IN_WB_ON_ITR_MODE) { + itr_val = ice_buildreg_itr(rx->itr_idx, ICE_WB_ON_ITR_USECS); + wr32(&vsi->back->hw, GLINT_DYN_CTL(q_vector->reg_idx), itr_val); + /* set target back to last user set value */ + rx->target_itr = rx->itr_setting; + /* set current to what we just wrote and dynamic if needed */ + rx->current_itr = ICE_WB_ON_ITR_USECS | + (rx->itr_setting & ICE_ITR_DYNAMIC); + /* allow normal interrupt flow to start */ + q_vector->itr_countdown = 0; + return; + } + /* This will do nothing if dynamic updates are not enabled */ ice_update_itr(q_vector, tx); ice_update_itr(q_vector, rx); @@ -1389,13 +1421,48 @@ ice_update_ena_itr(struct ice_vsi *vsi, struct ice_q_vector *q_vector) q_vector->itr_countdown--; } - if (!test_bit(__ICE_DOWN, vsi->state)) - wr32(&vsi->back->hw, + if (!test_bit(__ICE_DOWN, q_vector->vsi->state)) + wr32(&q_vector->vsi->back->hw, GLINT_DYN_CTL(q_vector->reg_idx), itr_val); } /** + * ice_set_wb_on_itr - set WB_ON_ITR for this q_vector + * @q_vector: q_vector to set WB_ON_ITR on + * + * We need to tell hardware to write-back completed descriptors even when + * interrupts are disabled. Descriptors will be written back on cache line + * boundaries without WB_ON_ITR enabled, but if we don't enable WB_ON_ITR + * descriptors may not be written back if they don't fill a cache line until the + * next interrupt. + * + * This sets the write-back frequency to 2 microseconds as that is the minimum + * value that's not 0 due to ITR granularity. Also, set the INTENA_MSK bit to + * make sure hardware knows we aren't meddling with the INTENA_M bit. + */ +static void ice_set_wb_on_itr(struct ice_q_vector *q_vector) +{ + struct ice_vsi *vsi = q_vector->vsi; + + /* already in WB_ON_ITR mode no need to change it */ + if (q_vector->itr_countdown == ICE_IN_WB_ON_ITR_MODE) + return; + + if (q_vector->num_ring_rx) + wr32(&vsi->back->hw, GLINT_DYN_CTL(q_vector->reg_idx), + ICE_GLINT_DYN_CTL_WB_ON_ITR(ICE_WB_ON_ITR_USECS, + ICE_RX_ITR)); + + if (q_vector->num_ring_tx) + wr32(&vsi->back->hw, GLINT_DYN_CTL(q_vector->reg_idx), + ICE_GLINT_DYN_CTL_WB_ON_ITR(ICE_WB_ON_ITR_USECS, + ICE_TX_ITR)); + + q_vector->itr_countdown = ICE_IN_WB_ON_ITR_MODE; +} + +/** * ice_napi_poll - NAPI polling Rx/Tx cleanup routine * @napi: napi struct with our devices info in it * @budget: amount of work driver is allowed to do this pass, in packets @@ -1408,29 +1475,32 @@ int ice_napi_poll(struct napi_struct *napi, int budget) { struct ice_q_vector *q_vector = container_of(napi, struct ice_q_vector, napi); - struct ice_vsi *vsi = q_vector->vsi; - struct ice_pf *pf = vsi->back; bool clean_complete = true; - int budget_per_ring = 0; struct ice_ring *ring; + int budget_per_ring; int work_done = 0; /* Since the actual Tx work is minimal, we can give the Tx a larger * budget and be more aggressive about cleaning up the Tx descriptors. */ ice_for_each_ring(ring, q_vector->tx) - if (!ice_clean_tx_irq(vsi, ring, budget)) + if (!ice_clean_tx_irq(ring, budget)) clean_complete = false; /* Handle case where we are called by netpoll with a budget of 0 */ - if (budget <= 0) + if (unlikely(budget <= 0)) return budget; - /* We attempt to distribute budget to each Rx queue fairly, but don't - * allow the budget to go below 1 because that would exit polling early. - */ - if (q_vector->num_ring_rx) + /* normally we have 1 Rx ring per q_vector */ + if (unlikely(q_vector->num_ring_rx > 1)) + /* We attempt to distribute budget to each Rx queue fairly, but + * don't allow the budget to go below 1 because that would exit + * polling early. + */ budget_per_ring = max(budget / q_vector->num_ring_rx, 1); + else + /* Max of 1 Rx ring in this q_vector so give it the budget */ + budget_per_ring = budget; ice_for_each_ring(ring, q_vector->rx) { int cleaned; @@ -1450,8 +1520,9 @@ int ice_napi_poll(struct napi_struct *napi, int budget) * poll us due to busy-polling */ if (likely(napi_complete_done(napi, work_done))) - if (test_bit(ICE_FLAG_MSIX_ENA, pf->flags)) - ice_update_ena_itr(vsi, q_vector); + ice_update_ena_itr(q_vector); + else + ice_set_wb_on_itr(q_vector); return min_t(int, work_done, budget - 1); } @@ -1521,7 +1592,7 @@ ice_tx_map(struct ice_ring *tx_ring, struct ice_tx_buf *first, { u64 td_offset, td_tag, td_cmd; u16 i = tx_ring->next_to_use; - struct skb_frag_struct *frag; + skb_frag_t *frag; unsigned int data_len, size; struct ice_tx_desc *tx_desc; struct ice_tx_buf *tx_buf; @@ -1923,7 +1994,7 @@ static unsigned int ice_txd_use_count(unsigned int size) */ static unsigned int ice_xmit_desc_count(struct sk_buff *skb) { - const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0]; + const skb_frag_t *frag = &skb_shinfo(skb)->frags[0]; unsigned int nr_frags = skb_shinfo(skb)->nr_frags; unsigned int count = 0, size = skb_headlen(skb); @@ -1954,7 +2025,7 @@ static unsigned int ice_xmit_desc_count(struct sk_buff *skb) */ static bool __ice_chk_linearize(struct sk_buff *skb) { - const struct skb_frag_struct *frag, *stale; + const skb_frag_t *frag, *stale; int nr_frags, sum; /* no need to check if number of frags is less than 7 */ @@ -2036,6 +2107,7 @@ static netdev_tx_t ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring) { struct ice_tx_offload_params offload = { 0 }; + struct ice_vsi *vsi = tx_ring->vsi; struct ice_tx_buf *first; unsigned int count; int tso, csum; @@ -2083,7 +2155,15 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring) if (csum < 0) goto out_drop; - if (tso || offload.cd_tunnel_params) { + /* allow CONTROL frames egress from main VSI if FW LLDP disabled */ + if (unlikely(skb->priority == TC_PRIO_CONTROL && + vsi->type == ICE_VSI_PF && + vsi->port_info->is_sw_lldp)) + offload.cd_qw1 |= (u64)(ICE_TX_DESC_DTYPE_CTX | + ICE_TX_CTX_DESC_SWTCH_UPLINK << + ICE_TXD_CTX_QW1_CMD_S); + + if (offload.cd_qw1 & ICE_TX_DESC_DTYPE_CTX) { struct ice_tx_ctx_desc *cdesc; int i = tx_ring->next_to_use; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index ec76aba347b9..94a9280193e2 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -144,6 +144,19 @@ enum ice_rx_dtype { #define ICE_DFLT_INTRL 0 #define ICE_MAX_INTRL 236 +#define ICE_WB_ON_ITR_USECS 2 +#define ICE_IN_WB_ON_ITR_MODE 255 +/* Sets WB_ON_ITR and assumes INTENA bit is already cleared, which allows + * setting the MSK_M bit to tell hardware to ignore the INTENA_M bit. Also, + * set the write-back latency to the usecs passed in. + */ +#define ICE_GLINT_DYN_CTL_WB_ON_ITR(usecs, itr_idx) \ + ((((usecs) << (GLINT_DYN_CTL_INTERVAL_S - ICE_ITR_GRAN_S)) & \ + GLINT_DYN_CTL_INTERVAL_M) | \ + (((itr_idx) << GLINT_DYN_CTL_ITR_INDX_S) & \ + GLINT_DYN_CTL_ITR_INDX_M) | GLINT_DYN_CTL_INTENA_MSK_M | \ + GLINT_DYN_CTL_WB_ON_ITR_M) + /* Legacy or Advanced Mode Queue */ #define ICE_TX_ADVANCED 0 #define ICE_TX_LEGACY 1 diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 24bbef8bbe69..6667d17a4206 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -4,18 +4,19 @@ #ifndef _ICE_TYPE_H_ #define _ICE_TYPE_H_ +#define ICE_BYTES_PER_WORD 2 +#define ICE_BYTES_PER_DWORD 4 + #include "ice_status.h" #include "ice_hw_autogen.h" #include "ice_osdep.h" #include "ice_controlq.h" #include "ice_lan_tx_rx.h" +#include "ice_flex_type.h" -#define ICE_BYTES_PER_WORD 2 -#define ICE_BYTES_PER_DWORD 4 - -static inline bool ice_is_tc_ena(u8 bitmap, u8 tc) +static inline bool ice_is_tc_ena(unsigned long bitmap, u8 tc) { - return test_bit(tc, (unsigned long *)&bitmap); + return test_bit(tc, &bitmap); } /* Driver always calls main vsi_handle first */ @@ -31,6 +32,7 @@ static inline bool ice_is_tc_ena(u8 bitmap, u8 tc) #define ICE_DBG_LAN BIT_ULL(8) #define ICE_DBG_SW BIT_ULL(13) #define ICE_DBG_SCHED BIT_ULL(14) +#define ICE_DBG_PKG BIT_ULL(16) #define ICE_DBG_RES BIT_ULL(17) #define ICE_DBG_AQ_MSG BIT_ULL(24) #define ICE_DBG_AQ_CMD BIT_ULL(27) @@ -53,6 +55,14 @@ enum ice_aq_res_access_type { ICE_RES_WRITE }; +struct ice_driver_ver { + u8 major_ver; + u8 minor_ver; + u8 build_ver; + u8 subbuild_ver; + u8 driver_string[32]; +}; + enum ice_fc_mode { ICE_FC_NONE = 0, ICE_FC_RX_PAUSE, @@ -139,6 +149,9 @@ struct ice_phy_info { /* Common HW capabilities for SW use */ struct ice_hw_common_caps { u32 valid_functions; + /* DCB capabilities */ + u32 active_tc_bitmap; + u32 maxtc; /* Tx/Rx queues */ u16 num_rxq; /* Number/Total Rx queues */ @@ -219,6 +232,8 @@ struct ice_nvm_info { u8 blank_nvm_mode; /* is NVM empty (no FW present) */ }; +#define ICE_NVM_VER_LEN 32 + /* Max number of port to queue branches w.r.t topology */ #define ICE_MAX_TRAFFIC_CLASS 8 #define ICE_TXSCHED_MAX_BRANCHES ICE_MAX_TRAFFIC_CLASS @@ -347,6 +362,8 @@ struct ice_port_info { struct ice_mac_info mac; struct ice_phy_info phy; struct mutex sched_lock; /* protect access to TXSched tree */ + struct ice_sched_node * + sib_head[ICE_MAX_TRAFFIC_CLASS][ICE_AQC_TOPO_MAX_LEVEL_NUM]; struct ice_dcbx_cfg local_dcbx_cfg; /* Oper/Local Cfg */ /* DCBX info */ struct ice_dcbx_cfg remote_dcbx_cfg; /* Peer Cfg */ @@ -454,6 +471,30 @@ struct ice_hw { u8 ucast_shared; /* true if VSIs can share unicast addr */ + /* Active package version (currently active) */ + struct ice_pkg_ver active_pkg_ver; + u8 active_pkg_name[ICE_PKG_NAME_SIZE]; + u8 active_pkg_in_nvm; + + enum ice_aq_err pkg_dwnld_status; + + /* Driver's package ver - (from the Metadata seg) */ + struct ice_pkg_ver pkg_ver; + u8 pkg_name[ICE_PKG_NAME_SIZE]; + + /* Driver's Ice package version (from the Ice seg) */ + struct ice_pkg_ver ice_pkg_ver; + u8 ice_pkg_name[ICE_PKG_NAME_SIZE]; + + /* Pointer to the ice segment */ + struct ice_seg *seg; + + /* Pointer to allocated copy of pkg memory */ + u8 *pkg_copy; + u32 pkg_size; + + /* HW block tables */ + struct ice_blk_info blk[ICE_BLK_COUNT]; }; /* Statistics collected by each port, VSI, VEB, and S-channel */ diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 5d24b539648f..b45797f39b2f 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -129,7 +129,10 @@ static void ice_vc_notify_vf_link_state(struct ice_vf *vf) pfe.event = VIRTCHNL_EVENT_LINK_CHANGE; pfe.severity = PF_EVENT_SEVERITY_INFO; - if (vf->link_forced) + /* Always report link is down if the VF queues aren't enabled */ + if (!vf->num_qs_ena) + ice_set_pfe_link(vf, &pfe, ICE_AQ_LINK_SPEED_UNKNOWN, false); + else if (vf->link_forced) ice_set_pfe_link_forced(vf, &pfe, vf->link_up); else ice_set_pfe_link(vf, &pfe, ls->link_speed, ls->link_info & @@ -252,6 +255,35 @@ static int ice_sriov_free_msix_res(struct ice_pf *pf) } /** + * ice_set_vf_state_qs_dis - Set VF queues state to disabled + * @vf: pointer to the VF structure + */ +void ice_set_vf_state_qs_dis(struct ice_vf *vf) +{ + /* Clear Rx/Tx enabled queues flag */ + bitmap_zero(vf->txq_ena, ICE_MAX_BASE_QS_PER_VF); + bitmap_zero(vf->rxq_ena, ICE_MAX_BASE_QS_PER_VF); + vf->num_qs_ena = 0; + clear_bit(ICE_VF_STATE_QS_ENA, vf->vf_states); +} + +/** + * ice_dis_vf_qs - Disable the VF queues + * @vf: pointer to the VF structure + */ +static void ice_dis_vf_qs(struct ice_vf *vf) +{ + struct ice_pf *pf = vf->pf; + struct ice_vsi *vsi; + + vsi = pf->vsi[vf->lan_vsi_idx]; + + ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, vf->vf_id); + ice_vsi_stop_rx_rings(vsi); + ice_set_vf_state_qs_dis(vf); +} + +/** * ice_free_vfs - Free all VFs * @pf: pointer to the PF structure */ @@ -267,19 +299,9 @@ void ice_free_vfs(struct ice_pf *pf) usleep_range(1000, 2000); /* Avoid wait time by stopping all VFs at the same time */ - for (i = 0; i < pf->num_alloc_vfs; i++) { - struct ice_vsi *vsi; - - if (!test_bit(ICE_VF_STATE_ENA, pf->vf[i].vf_states)) - continue; - - vsi = pf->vsi[pf->vf[i].lan_vsi_idx]; - /* stop rings without wait time */ - ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, i); - ice_vsi_stop_rx_rings(vsi); - - clear_bit(ICE_VF_STATE_ENA, pf->vf[i].vf_states); - } + for (i = 0; i < pf->num_alloc_vfs; i++) + if (test_bit(ICE_VF_STATE_QS_ENA, pf->vf[i].vf_states)) + ice_dis_vf_qs(&pf->vf[i]); /* Disable IOV before freeing resources. This lets any VF drivers * running in the host get themselves cleaned up before we yank @@ -297,13 +319,6 @@ void ice_free_vfs(struct ice_pf *pf) if (test_bit(ICE_VF_STATE_INIT, pf->vf[i].vf_states)) { /* disable VF qp mappings */ ice_dis_vf_mappings(&pf->vf[i]); - - /* Set this state so that assigned VF vectors can be - * reclaimed by PF for reuse in ice_vsi_release(). No - * need to clear this bit since pf->vf array is being - * freed anyways after this for loop - */ - set_bit(ICE_VF_STATE_CFG_INTR, pf->vf[i].vf_states); ice_free_vf_res(&pf->vf[i]); } } @@ -341,12 +356,13 @@ void ice_free_vfs(struct ice_pf *pf) * ice_trigger_vf_reset - Reset a VF on HW * @vf: pointer to the VF structure * @is_vflr: true if VFLR was issued, false if not + * @is_pfr: true if the reset was triggered due to a previous PFR * * Trigger hardware to start a reset for a particular VF. Expects the caller * to wait the proper amount of time to allow hardware to reset the VF before * it cleans up and restores VF functionality. */ -static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr) +static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr, bool is_pfr) { struct ice_pf *pf = vf->pf; u32 reg, reg_idx, bit_idx; @@ -367,10 +383,13 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr) */ clear_bit(ICE_VF_STATE_INIT, vf->vf_states); - /* Clear the VF's ARQLEN register. This is how the VF detects reset, - * since the VFGEN_RSTAT register doesn't stick at 0 after reset. + /* VF_MBX_ARQLEN is cleared by PFR, so the driver needs to clear it + * in the case of VFR. If this is done for PFR, it can mess up VF + * resets because the VF driver may already have started cleanup + * by the time we get here. */ - wr32(hw, VF_MBX_ARQLEN(vf_abs_id), 0); + if (!is_pfr) + wr32(hw, VF_MBX_ARQLEN(vf_abs_id), 0); /* In the case of a VFLR, the HW has already reset the VF and we * just need to clean up, so don't hit the VFRTRIG register. @@ -389,12 +408,15 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr) wr32(hw, PF_PCI_CIAA, VF_DEVICE_STATUS | (vf_abs_id << PF_PCI_CIAA_VF_NUM_S)); - for (i = 0; i < 100; i++) { + for (i = 0; i < ICE_PCI_CIAD_WAIT_COUNT; i++) { reg = rd32(hw, PF_PCI_CIAD); - if ((reg & VF_TRANS_PENDING_M) != 0) - dev_err(&pf->pdev->dev, - "VF %d PCI transactions stuck\n", vf->vf_id); - udelay(1); + /* no transactions pending so stop polling */ + if ((reg & VF_TRANS_PENDING_M) == 0) + break; + + dev_err(&pf->pdev->dev, + "VF %d PCI transactions stuck\n", vf->vf_id); + udelay(ICE_PCI_CIAD_WAIT_DELAY_US); } } @@ -481,19 +503,20 @@ 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 + * ice_calc_vf_first_vector_idx - Calculate MSIX vector index in the PF space * @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 + * This returns the first MSIX vector index in PF space that is used by this VF. + * This index is used when accessing PF relative registers such as + * GLINT_VECT2FUNC and GLINT_DYN_CTL. + * 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; + return pf->sriov_base_vector + vf->vf_id * pf->num_vf_msix; } /** @@ -543,7 +566,10 @@ static int ice_alloc_vsi_res(struct ice_vf *vf) status = ice_add_mac(&pf->hw, &tmp_add_list); if (status) - dev_err(&pf->pdev->dev, "could not add mac filters\n"); + dev_err(&pf->pdev->dev, + "could not add mac filters error %d\n", status); + else + vf->num_mac = 1; /* Clear this bit after VF initialization since we shouldn't reclaim * and reassign interrupts for synchronous or asynchronous VFR events. @@ -551,7 +577,6 @@ static int ice_alloc_vsi_res(struct ice_vf *vf) * expect vector assignment to be changed unless there is a request for * more vectors. */ - clear_bit(ICE_VF_STATE_CFG_INTR, vf->vf_states); ice_alloc_vsi_res_exit: ice_free_fltr_list(&pf->pdev->dev, &tmp_add_list); return status; @@ -567,20 +592,21 @@ static int ice_alloc_vf_res(struct ice_vf *vf) int tx_rx_queue_left; int status; - /* setup VF VSI and necessary resources */ - status = ice_alloc_vsi_res(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 = min_t(int, ice_get_avail_txq_count(pf), + ice_get_avail_rxq_count(pf)); 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; + /* setup VF VSI and necessary resources */ + status = ice_alloc_vsi_res(vf); + if (status) + goto ice_alloc_vf_res_exit; + if (vf->trusted) set_bit(ICE_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps); else @@ -605,27 +631,30 @@ ice_alloc_vf_res_exit: */ static void ice_ena_vf_mappings(struct ice_vf *vf) { + int abs_vf_id, abs_first, abs_last; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; int first, last, v; struct ice_hw *hw; - int abs_vf_id; u32 reg; hw = &pf->hw; vsi = pf->vsi[vf->lan_vsi_idx]; first = vf->first_vector_idx; last = (first + pf->num_vf_msix) - 1; + abs_first = first + pf->hw.func_caps.common_cap.msix_vector_first_id; + abs_last = (abs_first + pf->num_vf_msix) - 1; abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; /* VF Vector allocation */ - reg = (((first << VPINT_ALLOC_FIRST_S) & VPINT_ALLOC_FIRST_M) | - ((last << VPINT_ALLOC_LAST_S) & VPINT_ALLOC_LAST_M) | + reg = (((abs_first << VPINT_ALLOC_FIRST_S) & VPINT_ALLOC_FIRST_M) | + ((abs_last << VPINT_ALLOC_LAST_S) & VPINT_ALLOC_LAST_M) | VPINT_ALLOC_VALID_M); wr32(hw, VPINT_ALLOC(vf->vf_id), reg); - reg = (((first << VPINT_ALLOC_PCI_FIRST_S) & VPINT_ALLOC_PCI_FIRST_M) | - ((last << VPINT_ALLOC_PCI_LAST_S) & VPINT_ALLOC_PCI_LAST_M) | + reg = (((abs_first << VPINT_ALLOC_PCI_FIRST_S) + & VPINT_ALLOC_PCI_FIRST_M) | + ((abs_last << VPINT_ALLOC_PCI_LAST_S) & VPINT_ALLOC_PCI_LAST_M) | VPINT_ALLOC_PCI_VALID_M); wr32(hw, VPINT_ALLOC_PCI(vf->vf_id), reg); /* map the interrupts to its functions */ @@ -870,11 +899,11 @@ static int ice_check_avail_res(struct ice_pf *pf) * at runtime through Virtchnl, that is the reason we start by reserving * few queues. */ - num_txq = ice_determine_res(pf, pf->q_left_tx, ICE_DFLT_QS_PER_VF, - ICE_MIN_QS_PER_VF); + num_txq = ice_determine_res(pf, ice_get_avail_txq_count(pf), + ICE_DFLT_QS_PER_VF, ICE_MIN_QS_PER_VF); - num_rxq = ice_determine_res(pf, pf->q_left_rx, ICE_DFLT_QS_PER_VF, - ICE_MIN_QS_PER_VF); + num_rxq = ice_determine_res(pf, ice_get_avail_rxq_count(pf), + ICE_DFLT_QS_PER_VF, ICE_MIN_QS_PER_VF); if (!num_txq || !num_rxq) return -EIO; @@ -983,6 +1012,47 @@ ice_vf_set_vsi_promisc(struct ice_vf *vf, struct ice_vsi *vsi, u8 promisc_m, } /** + * ice_config_res_vfs - Finalize allocation of VFs resources in one go + * @pf: pointer to the PF structure + * + * This function is being called as last part of resetting all VFs, or when + * configuring VFs for the first time, where there is no resource to be freed + * Returns true if resources were properly allocated for all VFs, and false + * otherwise. + */ +static bool ice_config_res_vfs(struct ice_pf *pf) +{ + struct ice_hw *hw = &pf->hw; + int v; + + if (ice_check_avail_res(pf)) { + dev_err(&pf->pdev->dev, + "Cannot allocate VF resources, try with fewer number of VFs\n"); + return false; + } + + /* rearm global interrupts */ + if (test_and_clear_bit(__ICE_OICR_INTR_DIS, pf->state)) + ice_irq_dynamic_ena(hw, NULL, NULL); + + /* Finish resetting each VF and allocate resources */ + for (v = 0; v < pf->num_alloc_vfs; v++) { + struct ice_vf *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); + + return true; +} + +/** * 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 @@ -1010,18 +1080,17 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) /* Begin reset on all VFs at once */ for (v = 0; v < pf->num_alloc_vfs; v++) - ice_trigger_vf_reset(&pf->vf[v], is_vflr); + ice_trigger_vf_reset(&pf->vf[v], is_vflr, true); 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); - } + if (test_bit(ICE_VF_STATE_QS_ENA, vf->vf_states)) + ice_dis_vf_qs(vf); + ice_dis_vsi_txq(vsi->port_info, vsi->idx, 0, 0, NULL, NULL, + NULL, ICE_VF_RESET, vf->vf_id, NULL); } /* HW requires some time to make sure it can flush the FIFO for a VF @@ -1031,7 +1100,6 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) * finished resetting. */ for (i = 0, v = 0; i < 10 && v < pf->num_alloc_vfs; i++) { - usleep_range(10000, 20000); /* Check each VF in sequence */ while (v < pf->num_alloc_vfs) { @@ -1039,8 +1107,11 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) vf = &pf->vf[v]; reg = rd32(hw, VPGEN_VFRSTAT(vf->vf_id)); - if (!(reg & VPGEN_VFRSTAT_VFRD_M)) + if (!(reg & VPGEN_VFRSTAT_VFRD_M)) { + /* only delay if the check failed */ + usleep_range(10, 20); break; + } /* If the current VF has finished resetting, move on * to the next VF in sequence. @@ -1054,7 +1125,6 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) */ if (v < pf->num_alloc_vfs) dev_warn(&pf->pdev->dev, "VF reset check timeout\n"); - usleep_range(10000, 20000); /* free VF resources to begin resetting the VSI state */ for (v = 0; v < pf->num_alloc_vfs; v++) { @@ -1074,25 +1144,8 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) 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, - "Cannot allocate VF resources, try with fewer number of VFs\n"); + if (!ice_config_res_vfs(pf)) return false; - } - - /* Finish the reset on each VF */ - 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); return true; } @@ -1114,27 +1167,31 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) u32 reg; int i; - /* If the VFs have been disabled, this means something else is - * resetting the VF, so we shouldn't continue. + /* If the PF has been disabled, there is no need resetting VF until + * PF is active again. */ - if (test_and_set_bit(__ICE_VF_DIS, pf->state)) + if (test_bit(__ICE_VF_DIS, pf->state)) return false; - ice_trigger_vf_reset(vf, is_vflr); + /* If the VF has been disabled, this means something else is + * resetting the VF, so we shouldn't continue. Otherwise, set + * disable VF state bit for actual reset, and continue. + */ + if (test_and_set_bit(ICE_VF_STATE_DIS, vf->vf_states)) + return false; + + ice_trigger_vf_reset(vf, is_vflr, false); 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); - } else { - /* 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, vsi->idx, 0, 0, NULL, NULL, - NULL, ICE_VF_RESET, vf->vf_id, NULL); - } + if (test_bit(ICE_VF_STATE_QS_ENA, vf->vf_states)) + ice_dis_vf_qs(vf); + + /* Call Disable LAN Tx queue AQ whether or not queues are + * enabled. This is needed for successful completion of VFR. + */ + 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 @@ -1145,12 +1202,14 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) * poll the status register to make sure that the reset * completed successfully. */ - usleep_range(10000, 20000); reg = rd32(hw, VPGEN_VFRSTAT(vf->vf_id)); if (reg & VPGEN_VFRSTAT_VFRD_M) { rsd = true; break; } + + /* only sleep if the reset is not done */ + usleep_range(10, 20); } /* Display a warning if VF didn't manage to reset in time, but need to @@ -1160,8 +1219,6 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) dev_warn(&pf->pdev->dev, "VF reset check timeout on VF %d\n", vf->vf_id); - usleep_range(10000, 20000); - /* disable promiscuous modes in case they were enabled * ignore any error if disabling process failed */ @@ -1183,7 +1240,6 @@ static bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) ice_cleanup_and_realloc_vf(vf); ice_flush(hw); - clear_bit(__ICE_VF_DIS, pf->state); return true; } @@ -1257,7 +1313,7 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) /* Disable global interrupt 0 so we don't try to handle the VFLR. */ wr32(hw, GLINT_DYN_CTL(pf->oicr_idx), ICE_ITR_NONE << GLINT_DYN_CTL_ITR_INDX_S); - + set_bit(__ICE_OICR_INTR_DIS, pf->state); ice_flush(hw); ret = pci_enable_sriov(pf->pdev, num_alloc_vfs); @@ -1283,19 +1339,16 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) /* assign default capabilities */ set_bit(ICE_VIRTCHNL_VF_CAP_L2, &vfs[i].vf_caps); vfs[i].spoofchk = true; - - /* Set this state so that PF driver does VF vector assignment */ - set_bit(ICE_VF_STATE_CFG_INTR, vfs[i].vf_states); } pf->num_alloc_vfs = num_alloc_vfs; - /* VF resources get allocated during reset */ - if (!ice_reset_all_vfs(pf, true)) { + /* VF resources get allocated with initialization */ + if (!ice_config_res_vfs(pf)) { ret = -EIO; goto err_unroll_sriov; } - goto err_unroll_intr; + return ret; err_unroll_sriov: pf->vf = NULL; @@ -1307,6 +1360,7 @@ err_pci_disable_sriov: err_unroll_intr: /* rearm interrupts here */ ice_irq_dynamic_ena(hw, NULL, NULL); + clear_bit(__ICE_OICR_INTR_DIS, pf->state); return ret; } @@ -1389,6 +1443,12 @@ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs) { struct ice_pf *pf = pci_get_drvdata(pdev); + if (ice_is_safe_mode(pf)) { + dev_err(&pf->pdev->dev, + "SR-IOV cannot be configured - Device is in Safe Mode\n"); + return -EOPNOTSUPP; + } + if (num_vfs) return ice_pci_sriov_ena(pf, num_vfs); @@ -1490,10 +1550,10 @@ ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode, aq_ret = ice_aq_send_msg_to_vf(&pf->hw, vf->vf_id, v_opcode, v_retval, msg, msglen, NULL); - if (aq_ret) { + if (aq_ret && pf->hw.mailboxq.sq_last_status != ICE_AQ_RC_ENOSYS) { dev_info(&pf->pdev->dev, - "Unable to send the message to VF %d aq_err %d\n", - vf->vf_id, pf->hw.mailboxq.sq_last_status); + "Unable to send the message to VF %d ret %d aq_err %d\n", + vf->vf_id, aq_ret, pf->hw.mailboxq.sq_last_status); return -EIO; } @@ -1688,6 +1748,21 @@ static bool ice_vc_isvalid_q_id(struct ice_vf *vf, u16 vsi_id, u8 qid) } /** + * ice_vc_isvalid_ring_len + * @ring_len: length of ring + * + * check for the valid ring count, should be multiple of ICE_REQ_DESC_MULTIPLE + * or zero + */ +static bool ice_vc_isvalid_ring_len(u16 ring_len) +{ + return ring_len == 0 || + (ring_len >= ICE_MIN_NUM_DESC && + ring_len <= ICE_MAX_NUM_DESC && + !(ring_len % ICE_REQ_DESC_MULTIPLE)); +} + +/** * ice_vc_config_rss_key * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -1712,18 +1787,18 @@ static int ice_vc_config_rss_key(struct ice_vf *vf, u8 *msg) goto error_param; } - vsi = pf->vsi[vf->lan_vsi_idx]; - if (!vsi) { + if (vrk->key_len != ICE_VSIQF_HKEY_ARRAY_SIZE) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - if (vrk->key_len != ICE_VSIQF_HKEY_ARRAY_SIZE) { + if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { + vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1759,18 +1834,18 @@ static int ice_vc_config_rss_lut(struct ice_vf *vf, u8 *msg) goto error_param; } - vsi = pf->vsi[vf->lan_vsi_idx]; - if (!vsi) { + if (vrl->lut_entries != ICE_VSIQF_HLUT_ARRAY_SIZE) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - if (vrl->lut_entries != ICE_VSIQF_HLUT_ARRAY_SIZE) { + if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } - if (!test_bit(ICE_FLAG_RSS_ENA, vf->pf->flags)) { + vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1839,6 +1914,8 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) (struct virtchnl_queue_select *)msg; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; + unsigned long q_map; + u16 vf_q_id; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -1855,6 +1932,12 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) goto error_param; } + if (vqs->rx_queues > ICE_MAX_BASE_QS_PER_VF || + vqs->tx_queues > ICE_MAX_BASE_QS_PER_VF) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + vsi = pf->vsi[vf->lan_vsi_idx]; if (!vsi) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -1865,12 +1948,48 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) * Tx queue group list was configured and the context bits were * programmed using ice_vsi_cfg_txqs */ - if (ice_vsi_start_rx_rings(vsi)) - v_ret = VIRTCHNL_STATUS_ERR_PARAM; + q_map = vqs->rx_queues; + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + /* Skip queue if enabled */ + if (test_bit(vf_q_id, vf->rxq_ena)) + continue; + + if (ice_vsi_ctrl_rx_ring(vsi, true, vf_q_id)) { + dev_err(&vsi->back->pdev->dev, + "Failed to enable Rx ring %d on VSI %d\n", + vf_q_id, vsi->vsi_num); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + set_bit(vf_q_id, vf->rxq_ena); + vf->num_qs_ena++; + } + + vsi = pf->vsi[vf->lan_vsi_idx]; + q_map = vqs->tx_queues; + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + /* Skip queue if enabled */ + if (test_bit(vf_q_id, vf->txq_ena)) + continue; + + set_bit(vf_q_id, vf->txq_ena); + vf->num_qs_ena++; + } /* Set flag to indicate that queues are enabled */ if (v_ret == VIRTCHNL_STATUS_SUCCESS) - set_bit(ICE_VF_STATE_ENA, vf->vf_states); + set_bit(ICE_VF_STATE_QS_ENA, vf->vf_states); error_param: /* send the response to the VF */ @@ -1893,9 +2012,11 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) (struct virtchnl_queue_select *)msg; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; + unsigned long q_map; + u16 vf_q_id; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states) && - !test_bit(ICE_VF_STATE_ENA, vf->vf_states)) { + !test_bit(ICE_VF_STATE_QS_ENA, vf->vf_states)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -1910,29 +2031,81 @@ static int ice_vc_dis_qs_msg(struct ice_vf *vf, u8 *msg) goto error_param; } + if (vqs->rx_queues > ICE_MAX_BASE_QS_PER_VF || + vqs->tx_queues > ICE_MAX_BASE_QS_PER_VF) { + 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_stop_lan_tx_rings(vsi, ICE_NO_RESET, vf->vf_id)) { - dev_err(&vsi->back->pdev->dev, - "Failed to stop tx rings on VSI %d\n", - vsi->vsi_num); - v_ret = VIRTCHNL_STATUS_ERR_PARAM; + if (vqs->tx_queues) { + q_map = vqs->tx_queues; + + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + struct ice_ring *ring = vsi->tx_rings[vf_q_id]; + struct ice_txq_meta txq_meta = { 0 }; + + if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + /* Skip queue if not enabled */ + if (!test_bit(vf_q_id, vf->txq_ena)) + continue; + + ice_fill_txq_meta(vsi, ring, &txq_meta); + + if (ice_vsi_stop_tx_ring(vsi, ICE_NO_RESET, vf->vf_id, + ring, &txq_meta)) { + dev_err(&vsi->back->pdev->dev, + "Failed to stop Tx ring %d on VSI %d\n", + vf_q_id, vsi->vsi_num); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + /* Clear enabled queues flag */ + clear_bit(vf_q_id, vf->txq_ena); + vf->num_qs_ena--; + } } - 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); - v_ret = VIRTCHNL_STATUS_ERR_PARAM; + if (vqs->rx_queues) { + q_map = vqs->rx_queues; + + for_each_set_bit(vf_q_id, &q_map, ICE_MAX_BASE_QS_PER_VF) { + if (!ice_vc_isvalid_q_id(vf, vqs->vsi_id, vf_q_id)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + /* Skip queue if not enabled */ + if (!test_bit(vf_q_id, vf->rxq_ena)) + continue; + + if (ice_vsi_ctrl_rx_ring(vsi, false, vf_q_id)) { + dev_err(&vsi->back->pdev->dev, + "Failed to stop Rx ring %d on VSI %d\n", + vf_q_id, vsi->vsi_num); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + /* Clear enabled queues flag */ + clear_bit(vf_q_id, vf->rxq_ena); + vf->num_qs_ena--; + } } /* Clear enabled queues flag */ - if (v_ret == VIRTCHNL_STATUS_SUCCESS) - clear_bit(ICE_VF_STATE_ENA, vf->vf_states); + if (v_ret == VIRTCHNL_STATUS_SUCCESS && !vf->num_qs_ena) + clear_bit(ICE_VF_STATE_QS_ENA, vf->vf_states); error_param: /* send the response to the VF */ @@ -1962,12 +2135,6 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) 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; - } - /* 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 @@ -1979,6 +2146,12 @@ static int ice_vc_cfg_irq_map_msg(struct ice_vf *vf, u8 *msg) goto error_param; } + vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + for (i = 0; i < num_q_vectors_mapped; i++) { struct ice_q_vector *q_vector; @@ -2056,6 +2229,7 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) struct virtchnl_vsi_queue_config_info *qci = (struct virtchnl_vsi_queue_config_info *)msg; struct virtchnl_queue_pair_info *qpi; + u16 num_rxq = 0, num_txq = 0; struct ice_pf *pf = vf->pf; struct ice_vsi *vsi; int i; @@ -2071,13 +2245,16 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) } vsi = pf->vsi[vf->lan_vsi_idx]; - if (!vsi) + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; + } - if (qci->num_queue_pairs > ICE_MAX_BASE_QS_PER_VF) { + if (qci->num_queue_pairs > ICE_MAX_BASE_QS_PER_VF || + qci->num_queue_pairs > min_t(u16, vsi->alloc_txq, vsi->alloc_rxq)) { dev_err(&pf->pdev->dev, "VF-%d requesting more than supported number of queues: %d\n", - vf->vf_id, qci->num_queue_pairs); + vf->vf_id, min_t(u16, vsi->alloc_txq, vsi->alloc_rxq)); v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } @@ -2087,37 +2264,52 @@ static int ice_vc_cfg_qs_msg(struct ice_vf *vf, u8 *msg) if (qpi->txq.vsi_id != qci->vsi_id || qpi->rxq.vsi_id != qci->vsi_id || qpi->rxq.queue_id != qpi->txq.queue_id || + qpi->txq.headwb_enabled || + !ice_vc_isvalid_ring_len(qpi->txq.ring_len) || + !ice_vc_isvalid_ring_len(qpi->rxq.ring_len) || !ice_vc_isvalid_q_id(vf, qci->vsi_id, qpi->txq.queue_id)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; } /* copy Tx queue info from VF into VSI */ - vsi->tx_rings[i]->dma = qpi->txq.dma_ring_addr; - vsi->tx_rings[i]->count = qpi->txq.ring_len; - /* copy Rx queue info from VF into VSI */ - 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)) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; - goto error_param; + if (qpi->txq.ring_len > 0) { + num_txq++; + vsi->tx_rings[i]->dma = qpi->txq.dma_ring_addr; + vsi->tx_rings[i]->count = qpi->txq.ring_len; } - vsi->rx_buf_len = qpi->rxq.databuffer_size; - if (qpi->rxq.max_pkt_size >= (16 * 1024) || - qpi->rxq.max_pkt_size < 64) { - v_ret = VIRTCHNL_STATUS_ERR_PARAM; - goto error_param; + + /* copy Rx queue info from VF into VSI */ + if (qpi->rxq.ring_len > 0) { + num_rxq++; + vsi->rx_rings[i]->dma = qpi->rxq.dma_ring_addr; + vsi->rx_rings[i]->count = qpi->rxq.ring_len; + + if (qpi->rxq.databuffer_size != 0 && + (qpi->rxq.databuffer_size > ((16 * 1024) - 128) || + qpi->rxq.databuffer_size < 1024)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + vsi->rx_buf_len = qpi->rxq.databuffer_size; + vsi->rx_rings[i]->rx_buf_len = vsi->rx_buf_len; + if (qpi->rxq.max_pkt_size >= (16 * 1024) || + qpi->rxq.max_pkt_size < 64) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } } + vsi->max_frame = qpi->rxq.max_pkt_size; } /* VF can request to configure less than allocated queues * or default allocated queues. So update the VSI with new number */ - vsi->num_txq = qci->num_queue_pairs; - vsi->num_rxq = qci->num_queue_pairs; + vsi->num_txq = num_txq; + vsi->num_rxq = num_rxq; /* 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; + vsi->tc_cfg.tc_info[0].qcount_tx = num_txq; + vsi->tc_cfg.tc_info[0].qcount_rx = num_rxq; if (ice_vsi_cfg_lan_txqs(vsi) || ice_vsi_cfg_rxqs(vsi)) v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; @@ -2171,7 +2363,7 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) (struct virtchnl_ether_addr_list *)msg; struct ice_pf *pf = vf->pf; enum virtchnl_ops vc_op; - LIST_HEAD(mac_list); + enum ice_status status; struct ice_vsi *vsi; int mac_count = 0; int i; @@ -2245,33 +2437,32 @@ ice_vc_handle_mac_addr_msg(struct ice_vf *vf, u8 *msg, bool set) goto handle_mac_exit; } - /* 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)) { - v_ret = VIRTCHNL_STATUS_ERR_NO_MEMORY; + /* program the updated filter list */ + status = ice_vsi_cfg_mac_fltr(vsi, maddr, set); + if (status == ICE_ERR_DOES_NOT_EXIST || + status == ICE_ERR_ALREADY_EXISTS) { + dev_info(&pf->pdev->dev, + "can't %s MAC filters %pM for VF %d, error %d\n", + set ? "add" : "remove", maddr, vf->vf_id, + status); + } else if (status) { + dev_err(&pf->pdev->dev, + "can't %s MAC filters for VF %d, error %d\n", + set ? "add" : "remove", vf->vf_id, status); + v_ret = ice_err_to_virt_err(status); goto handle_mac_exit; } + mac_count++; } - /* program the updated filter list */ + /* Track number of MAC filters programmed for the VF VSI */ if (set) - v_ret = ice_err_to_virt_err(ice_add_mac(&pf->hw, &mac_list)); + vf->num_mac += mac_count; else - v_ret = ice_err_to_virt_err(ice_remove_mac(&pf->hw, &mac_list)); - - if (v_ret) { - dev_err(&pf->pdev->dev, - "can't update MAC filters for VF %d, error %d\n", - vf->vf_id, v_ret); - } else { - if (set) - vf->num_mac += mac_count; - else - vf->num_mac -= mac_count; - } + vf->num_mac -= mac_count; 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, v_ret, NULL, 0); } @@ -2315,11 +2506,11 @@ 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; + u16 req_queues = vfres->num_queue_pairs; struct ice_pf *pf = vf->pf; - int max_allowed_vf_queues; - int tx_rx_queue_left; - int cur_queues; + u16 max_allowed_vf_queues; + u16 tx_rx_queue_left; + u16 cur_queues; if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; @@ -2327,29 +2518,31 @@ static int ice_vc_request_qs_msg(struct ice_vf *vf, u8 *msg) } cur_queues = vf->num_vf_qs; - tx_rx_queue_left = min_t(int, pf->q_left_tx, pf->q_left_rx); + tx_rx_queue_left = min_t(u16, ice_get_avail_txq_count(pf), + ice_get_avail_rxq_count(pf)); max_allowed_vf_queues = tx_rx_queue_left + cur_queues; - if (req_queues <= 0) { + if (!req_queues) { dev_err(&pf->pdev->dev, - "VF %d tried to request %d queues. Ignoring.\n", - vf->vf_id, req_queues); + "VF %d tried to request 0 queues. Ignoring.\n", + vf->vf_id); } 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_BASE_QS_PER_VF); vfres->num_queue_pairs = ICE_MAX_BASE_QS_PER_VF; - } else if (req_queues - cur_queues > tx_rx_queue_left) { + } else if (req_queues > cur_queues && + 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 %d requested %u more queues, but only %u left.\n", vf->vf_id, req_queues - cur_queues, tx_rx_queue_left); - vfres->num_queue_pairs = min_t(int, max_allowed_vf_queues, + vfres->num_queue_pairs = min_t(u16, max_allowed_vf_queues, ICE_MAX_BASE_QS_PER_VF); } else { /* request is successful, then reset VF */ vf->num_req_qs = req_queues; ice_vc_dis_vf(vf); dev_info(&pf->pdev->dev, - "VF %d granted request of %d queues.\n", + "VF %d granted request of %u queues.\n", vf->vf_id, req_queues); return 0; } @@ -2589,8 +2782,9 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) } vf->num_vlan--; - /* Disable VLAN pruning when removing VLAN */ - ice_cfg_vlan_pruning(vsi, false, false); + /* Disable VLAN pruning when the last VLAN is removed */ + if (!vf->num_vlan) + ice_cfg_vlan_pruning(vsi, false, false); /* Disable Unicast/Multicast VLAN promiscuous mode */ if (vlan_promisc) { @@ -2731,20 +2925,6 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event) err = -EPERM; else err = -EINVAL; - goto error_handler; - } - - /* Perform additional checks specific to RSS and Virtchnl */ - if (v_opcode == VIRTCHNL_OP_CONFIG_RSS_KEY) { - struct virtchnl_rss_key *vrk = (struct virtchnl_rss_key *)msg; - - if (vrk->key_len != ICE_VSIQF_HKEY_ARRAY_SIZE) - err = -EINVAL; - } else if (v_opcode == VIRTCHNL_OP_CONFIG_RSS_LUT) { - struct virtchnl_rss_lut *vrl = (struct virtchnl_rss_lut *)msg; - - if (vrl->lut_entries != ICE_VSIQF_HLUT_ARRAY_SIZE) - err = -EINVAL; } error_handler: @@ -2762,6 +2942,7 @@ error_handler: break; case VIRTCHNL_OP_GET_VF_RESOURCES: err = ice_vc_get_vf_res_msg(vf, msg); + ice_vc_notify_vf_link_state(vf); break; case VIRTCHNL_OP_RESET_VF: ice_vc_reset_vf_msg(vf); diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h index c3ca522c245a..0d9880c8bba3 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h @@ -15,26 +15,38 @@ #define ICE_MAX_MACADDR_PER_VF 12 /* Malicious Driver Detection */ -#define ICE_DFLT_NUM_MDD_EVENTS_ALLOWED 3 #define ICE_DFLT_NUM_INVAL_MSGS_ALLOWED 10 +#define ICE_MDD_EVENTS_THRESHOLD 30 /* Static VF transaction/status register def */ #define VF_DEVICE_STATUS 0xAA #define VF_TRANS_PENDING_M 0x20 +/* wait defines for polling PF_PCI_CIAD register status */ +#define ICE_PCI_CIAD_WAIT_COUNT 100 +#define ICE_PCI_CIAD_WAIT_DELAY_US 1 + +/* VF resources default values and limitation */ +#define ICE_MAX_VF_COUNT 256 +#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_MAX_POLICY_INTR_PER_VF 33 +#define ICE_MIN_INTR_PER_VF (ICE_MIN_QS_PER_VF + 1) +#define ICE_DFLT_INTR_PER_VF (ICE_DFLT_QS_PER_VF + 1) + /* Specific VF states */ enum ice_vf_states { - ICE_VF_STATE_INIT = 0, - ICE_VF_STATE_ACTIVE, - ICE_VF_STATE_ENA, + ICE_VF_STATE_INIT = 0, /* PF is initializing VF */ + ICE_VF_STATE_ACTIVE, /* VF resources are allocated for use */ + ICE_VF_STATE_QS_ENA, /* VF queue(s) enabled */ ICE_VF_STATE_DIS, ICE_VF_STATE_MC_PROMISC, ICE_VF_STATE_UC_PROMISC, - /* state to indicate if PF needs to do vector assignment for VF. - * This needs to be set during first time VF initialization or later - * when VF asks for more Vectors through virtchnl OP. - */ - ICE_VF_STATE_CFG_INTR, ICE_VF_STATES_NBITS }; @@ -50,11 +62,14 @@ struct ice_vf { 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 */ + /* first vector index of this VF in the PF space */ + int first_vector_idx; 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; + DECLARE_BITMAP(txq_ena, ICE_MAX_BASE_QS_PER_VF); + DECLARE_BITMAP(rxq_ena, ICE_MAX_BASE_QS_PER_VF); u16 port_vlan_id; u8 pf_set_mac:1; /* VF MAC address set by VMM admin */ u8 trusted:1; @@ -77,6 +92,7 @@ struct ice_vf { u16 num_mac; u16 num_vlan; u16 num_vf_qs; /* num of queue configured per VF */ + u16 num_qs_ena; /* total num of Tx/Rx queue enabled */ }; #ifdef CONFIG_PCI_IOV @@ -103,12 +119,15 @@ 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); + +void ice_set_vf_state_qs_dis(struct ice_vf *vf); #else /* CONFIG_PCI_IOV */ #define ice_process_vflr_event(pf) do {} while (0) #define ice_free_vfs(pf) do {} while (0) #define ice_vc_process_vf_msg(pf, event) do {} while (0) #define ice_vc_notify_link_state(pf) do {} while (0) #define ice_vc_notify_reset(pf) do {} while (0) +#define ice_set_vf_state_qs_dis(vf) do {} while (0) static inline bool ice_reset_all_vfs(struct ice_pf __always_unused *pf, |