diff options
Diffstat (limited to 'drivers/net/ethernet/intel/i40e')
22 files changed, 1472 insertions, 468 deletions
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index 4d939af0a626..9a60d6b207f7 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -37,6 +37,7 @@ #include <net/tc_act/tc_mirred.h> #include <net/udp_tunnel.h> #include <net/xdp_sock.h> +#include <linux/bitfield.h> #include "i40e_type.h" #include "i40e_prototype.h" #include <linux/net/intel/i40e_client.h> @@ -144,6 +145,7 @@ enum i40e_state_t { __I40E_VIRTCHNL_OP_PENDING, __I40E_RECOVERY_MODE, __I40E_VF_RESETS_DISABLED, /* disable resets during i40e_remove */ + __I40E_IN_REMOVE, __I40E_VFS_RELEASING, /* This must be last as it determines the size of the BITMAP */ __I40E_STATE_SIZE__, @@ -174,7 +176,6 @@ enum i40e_interrupt_policy { struct i40e_lump_tracking { u16 num_entries; - u16 search_hint; u16 list[0]; #define I40E_PILE_VALID_BIT 0x8000 #define I40E_IWARP_IRQ_PILE_ID (I40E_PILE_VALID_BIT - 2) @@ -398,6 +399,20 @@ struct i40e_ddp_old_profile_list { I40E_FLEX_54_MASK | I40E_FLEX_55_MASK | \ I40E_FLEX_56_MASK | I40E_FLEX_57_MASK) +#define I40E_QINT_TQCTL_VAL(qp, vector, nextq_type) \ + (I40E_QINT_TQCTL_CAUSE_ENA_MASK | \ + (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) | \ + ((vector) << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) | \ + ((qp) << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT) | \ + (I40E_QUEUE_TYPE_##nextq_type << I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT)) + +#define I40E_QINT_RQCTL_VAL(qp, vector, nextq_type) \ + (I40E_QINT_RQCTL_CAUSE_ENA_MASK | \ + (I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) | \ + ((vector) << I40E_QINT_RQCTL_MSIX_INDX_SHIFT) | \ + ((qp) << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT) | \ + (I40E_QUEUE_TYPE_##nextq_type << I40E_QINT_RQCTL_NEXTQ_TYPE_SHIFT)) + struct i40e_flex_pit { struct list_head list; u16 src_offset; @@ -565,6 +580,7 @@ struct i40e_pf { #define I40E_FLAG_DISABLE_FW_LLDP BIT(24) #define I40E_FLAG_RS_FEC BIT(25) #define I40E_FLAG_BASE_R_FEC BIT(26) +#define I40E_FLAG_VF_VLAN_PRUNING BIT(27) /* TOTAL_PORT_SHUTDOWN * Allows to physically disable the link on the NIC's port. * If enabled, (after link down request from the OS) @@ -848,12 +864,17 @@ struct i40e_vsi { struct rtnl_link_stats64 net_stats_offsets; struct i40e_eth_stats eth_stats; struct i40e_eth_stats eth_stats_offsets; - u32 tx_restart; - u32 tx_busy; + u64 tx_restart; + u64 tx_busy; u64 tx_linearize; u64 tx_force_wb; - u32 rx_buf_failed; - u32 rx_page_failed; + u64 tx_stopped; + u64 rx_buf_failed; + u64 rx_page_failed; + u64 rx_page_reuse; + u64 rx_page_alloc; + u64 rx_page_waive; + u64 rx_page_busy; /* These are containers of ring pointers, allocated at run-time */ struct i40e_ring **rx_rings; @@ -1087,6 +1108,21 @@ static inline void i40e_write_fd_input_set(struct i40e_pf *pf, (u32)(val & 0xFFFFFFFFULL)); } +/** + * i40e_get_pf_count - get PCI PF count. + * @hw: pointer to a hw. + * + * Reports the function number of the highest PCI physical + * function plus 1 as it is loaded from the NVM. + * + * Return: PCI PF count. + **/ +static inline u32 i40e_get_pf_count(struct i40e_hw *hw) +{ + return FIELD_GET(I40E_GLGEN_PCIFCNCNT_PCIPFCNT_MASK, + rd32(hw, I40E_GLGEN_PCIFCNCNT)); +} + /* needed by i40e_ethtool.c */ int i40e_up(struct i40e_vsi *vsi); void i40e_down(struct i40e_vsi *vsi); @@ -1270,4 +1306,18 @@ int i40e_add_del_cloud_filter(struct i40e_vsi *vsi, int i40e_add_del_cloud_filter_big_buf(struct i40e_vsi *vsi, struct i40e_cloud_filter *filter, bool add); + +/** + * i40e_is_tc_mqprio_enabled - check if TC MQPRIO is enabled on PF + * @pf: pointer to a pf. + * + * Check and return value of flag I40E_FLAG_TC_MQPRIO. + * + * Return: I40E_FLAG_TC_MQPRIO set state. + **/ +static inline u32 i40e_is_tc_mqprio_enabled(struct i40e_pf *pf) +{ + return pf->flags & I40E_FLAG_TC_MQPRIO; +} + #endif /* _I40E_H_ */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c index 593912b17609..42439f725aa4 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c @@ -769,21 +769,24 @@ static bool i40e_asq_done(struct i40e_hw *hw) } /** - * i40e_asq_send_command - send command to Admin Queue + * i40e_asq_send_command_atomic_exec - send command to Admin Queue * @hw: pointer to the hw struct * @desc: prefilled descriptor describing the command (non DMA mem) * @buff: buffer to use for indirect commands * @buff_size: size of buffer for indirect commands * @cmd_details: pointer to command details structure + * @is_atomic_context: is the function called in an atomic context? * * This is the main send command driver routine for the Admin Queue send * queue. It runs the queue, cleans the queue, etc **/ -i40e_status i40e_asq_send_command(struct i40e_hw *hw, - struct i40e_aq_desc *desc, - void *buff, /* can be NULL */ - u16 buff_size, - struct i40e_asq_cmd_details *cmd_details) +static i40e_status +i40e_asq_send_command_atomic_exec(struct i40e_hw *hw, + struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ + u16 buff_size, + struct i40e_asq_cmd_details *cmd_details, + bool is_atomic_context) { i40e_status status = 0; struct i40e_dma_mem *dma_buff = NULL; @@ -793,8 +796,6 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw, u16 retval = 0; u32 val = 0; - mutex_lock(&hw->aq.asq_mutex); - if (hw->aq.asq.count == 0) { i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, "AQTX: Admin queue not initialized.\n"); @@ -910,7 +911,12 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw, */ if (i40e_asq_done(hw)) break; - udelay(50); + + if (is_atomic_context) + udelay(50); + else + usleep_range(40, 60); + total_delay += 50; } while (total_delay < hw->aq.asq_cmd_timeout); } @@ -963,10 +969,95 @@ i40e_status i40e_asq_send_command(struct i40e_hw *hw, } asq_send_command_error: + return status; +} + +/** + * i40e_asq_send_command_atomic - send command to Admin Queue + * @hw: pointer to the hw struct + * @desc: prefilled descriptor describing the command (non DMA mem) + * @buff: buffer to use for indirect commands + * @buff_size: size of buffer for indirect commands + * @cmd_details: pointer to command details structure + * @is_atomic_context: is the function called in an atomic context? + * + * Acquires the lock and calls the main send command execution + * routine. + **/ +i40e_status +i40e_asq_send_command_atomic(struct i40e_hw *hw, + struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ + u16 buff_size, + struct i40e_asq_cmd_details *cmd_details, + bool is_atomic_context) +{ + i40e_status status; + + mutex_lock(&hw->aq.asq_mutex); + status = i40e_asq_send_command_atomic_exec(hw, desc, buff, buff_size, + cmd_details, + is_atomic_context); + + mutex_unlock(&hw->aq.asq_mutex); + return status; +} + +i40e_status +i40e_asq_send_command(struct i40e_hw *hw, struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ u16 buff_size, + struct i40e_asq_cmd_details *cmd_details) +{ + return i40e_asq_send_command_atomic(hw, desc, buff, buff_size, + cmd_details, false); +} + +/** + * i40e_asq_send_command_atomic_v2 - send command to Admin Queue + * @hw: pointer to the hw struct + * @desc: prefilled descriptor describing the command (non DMA mem) + * @buff: buffer to use for indirect commands + * @buff_size: size of buffer for indirect commands + * @cmd_details: pointer to command details structure + * @is_atomic_context: is the function called in an atomic context? + * @aq_status: pointer to Admin Queue status return value + * + * Acquires the lock and calls the main send command execution + * routine. Returns the last Admin Queue status in aq_status + * to avoid race conditions in access to hw->aq.asq_last_status. + **/ +i40e_status +i40e_asq_send_command_atomic_v2(struct i40e_hw *hw, + struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ + u16 buff_size, + struct i40e_asq_cmd_details *cmd_details, + bool is_atomic_context, + enum i40e_admin_queue_err *aq_status) +{ + i40e_status status; + + mutex_lock(&hw->aq.asq_mutex); + status = i40e_asq_send_command_atomic_exec(hw, desc, buff, + buff_size, + cmd_details, + is_atomic_context); + if (aq_status) + *aq_status = hw->aq.asq_last_status; mutex_unlock(&hw->aq.asq_mutex); return status; } +i40e_status +i40e_asq_send_command_v2(struct i40e_hw *hw, struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ u16 buff_size, + struct i40e_asq_cmd_details *cmd_details, + enum i40e_admin_queue_err *aq_status) +{ + return i40e_asq_send_command_atomic_v2(hw, desc, buff, buff_size, + cmd_details, true, aq_status); +} + /** * i40e_fill_default_direct_cmd_desc - AQ descriptor helper function * @desc: pointer to the temp descriptor (non DMA mem) diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h index 140b677f114d..60f9e0a6aaca 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h @@ -11,8 +11,8 @@ */ #define I40E_FW_API_VERSION_MAJOR 0x0001 -#define I40E_FW_API_VERSION_MINOR_X722 0x0009 -#define I40E_FW_API_VERSION_MINOR_X710 0x0009 +#define I40E_FW_API_VERSION_MINOR_X722 0x000C +#define I40E_FW_API_VERSION_MINOR_X710 0x000F #define I40E_FW_MINOR_VERSION(_h) ((_h)->mac.type == I40E_MAC_XL710 ? \ I40E_FW_API_VERSION_MINOR_X710 : \ diff --git a/drivers/net/ethernet/intel/i40e/i40e_client.c b/drivers/net/ethernet/intel/i40e/i40e_client.c index ea2bb0140a6e..10d7a982a5b9 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_client.c +++ b/drivers/net/ethernet/intel/i40e/i40e_client.c @@ -177,6 +177,10 @@ void i40e_notify_client_of_netdev_close(struct i40e_vsi *vsi, bool reset) "Cannot locate client instance close routine\n"); return; } + if (!test_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state)) { + dev_dbg(&pf->pdev->dev, "Client is not open, abort close\n"); + return; + } cdev->client->ops->close(&cdev->lan_info, cdev->client, reset); clear_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state); i40e_client_release_qvlist(&cdev->lan_info); @@ -429,7 +433,6 @@ void i40e_client_subtask(struct i40e_pf *pf) /* Remove failed client instance */ clear_bit(__I40E_CLIENT_INSTANCE_OPENED, &cdev->state); - i40e_client_del_instance(pf); return; } } diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index b4d3fed0d2f2..4f01e2a6b6bb 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -27,6 +27,7 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw) case I40E_DEV_ID_QSFP_A: case I40E_DEV_ID_QSFP_B: case I40E_DEV_ID_QSFP_C: + case I40E_DEV_ID_1G_BASE_T_BC: case I40E_DEV_ID_5G_BASE_T_BC: case I40E_DEV_ID_10G_BASE_T: case I40E_DEV_ID_10G_BASE_T4: @@ -47,6 +48,7 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw) case I40E_DEV_ID_1G_BASE_T_X722: case I40E_DEV_ID_10G_BASE_T_X722: case I40E_DEV_ID_SFP_I_X722: + case I40E_DEV_ID_SFP_X722_A: hw->mac.type = I40E_MAC_X722; break; default: @@ -154,8 +156,8 @@ const char *i40e_stat_str(struct i40e_hw *hw, i40e_status stat_err) return "I40E_ERR_INVALID_MAC_ADDR"; case I40E_ERR_DEVICE_NOT_SUPPORTED: return "I40E_ERR_DEVICE_NOT_SUPPORTED"; - case I40E_ERR_MASTER_REQUESTS_PENDING: - return "I40E_ERR_MASTER_REQUESTS_PENDING"; + case I40E_ERR_PRIMARY_REQUESTS_PENDING: + return "I40E_ERR_PRIMARY_REQUESTS_PENDING"; case I40E_ERR_INVALID_LINK_SETTINGS: return "I40E_ERR_INVALID_LINK_SETTINGS"; case I40E_ERR_AUTONEG_NOT_COMPLETE: @@ -1899,8 +1901,9 @@ i40e_status i40e_aq_add_vsi(struct i40e_hw *hw, desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD)); - status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info, - sizeof(vsi_ctx->info), cmd_details); + status = i40e_asq_send_command_atomic(hw, &desc, &vsi_ctx->info, + sizeof(vsi_ctx->info), + cmd_details, true); if (status) goto aq_add_vsi_exit; @@ -2073,7 +2076,8 @@ enum i40e_status_code i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw, cmd->seid = cpu_to_le16(seid); cmd->vlan_tag = cpu_to_le16(vid | I40E_AQC_SET_VSI_VLAN_VALID); - status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + status = i40e_asq_send_command_atomic(hw, &desc, NULL, 0, + cmd_details, true); return status; } @@ -2114,7 +2118,8 @@ enum i40e_status_code i40e_aq_set_vsi_uc_promisc_on_vlan(struct i40e_hw *hw, cmd->seid = cpu_to_le16(seid); cmd->vlan_tag = cpu_to_le16(vid | I40E_AQC_SET_VSI_VLAN_VALID); - status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + status = i40e_asq_send_command_atomic(hw, &desc, NULL, 0, + cmd_details, true); return status; } @@ -2285,8 +2290,9 @@ i40e_status i40e_aq_update_vsi_params(struct i40e_hw *hw, desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD)); - status = i40e_asq_send_command(hw, &desc, &vsi_ctx->info, - sizeof(vsi_ctx->info), cmd_details); + status = i40e_asq_send_command_atomic(hw, &desc, &vsi_ctx->info, + sizeof(vsi_ctx->info), + cmd_details, true); vsi_ctx->vsis_allocated = le16_to_cpu(resp->vsi_used); vsi_ctx->vsis_unallocated = le16_to_cpu(resp->vsi_free); @@ -2630,33 +2636,28 @@ get_veb_exit: } /** - * i40e_aq_add_macvlan - * @hw: pointer to the hw struct - * @seid: VSI for the mac address + * i40e_prepare_add_macvlan * @mv_list: list of macvlans to be added + * @desc: pointer to AQ descriptor structure * @count: length of the list - * @cmd_details: pointer to command details structure or NULL + * @seid: VSI for the mac address * - * Add MAC/VLAN addresses to the HW filtering + * Internal helper function that prepares the add macvlan request + * and returns the buffer size. **/ -i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid, - struct i40e_aqc_add_macvlan_element_data *mv_list, - u16 count, struct i40e_asq_cmd_details *cmd_details) +static u16 +i40e_prepare_add_macvlan(struct i40e_aqc_add_macvlan_element_data *mv_list, + struct i40e_aq_desc *desc, u16 count, u16 seid) { - struct i40e_aq_desc desc; struct i40e_aqc_macvlan *cmd = - (struct i40e_aqc_macvlan *)&desc.params.raw; - i40e_status status; + (struct i40e_aqc_macvlan *)&desc->params.raw; u16 buf_size; int i; - if (count == 0 || !mv_list || !hw) - return I40E_ERR_PARAM; - buf_size = count * sizeof(*mv_list); /* prep the rest of the request */ - i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_macvlan); + i40e_fill_default_direct_cmd_desc(desc, i40e_aqc_opc_add_macvlan); cmd->num_addresses = cpu_to_le16(count); cmd->seid[0] = cpu_to_le16(I40E_AQC_MACVLAN_CMD_SEID_VALID | seid); cmd->seid[1] = 0; @@ -2667,14 +2668,71 @@ i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid, mv_list[i].flags |= cpu_to_le16(I40E_AQC_MACVLAN_ADD_USE_SHARED_MAC); - desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD)); + desc->flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD)); if (buf_size > I40E_AQ_LARGE_BUF) - desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB); + desc->flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB); - status = i40e_asq_send_command(hw, &desc, mv_list, buf_size, - cmd_details); + return buf_size; +} - return status; +/** + * i40e_aq_add_macvlan + * @hw: pointer to the hw struct + * @seid: VSI for the mac address + * @mv_list: list of macvlans to be added + * @count: length of the list + * @cmd_details: pointer to command details structure or NULL + * + * Add MAC/VLAN addresses to the HW filtering + **/ +i40e_status +i40e_aq_add_macvlan(struct i40e_hw *hw, u16 seid, + struct i40e_aqc_add_macvlan_element_data *mv_list, + u16 count, struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + u16 buf_size; + + if (count == 0 || !mv_list || !hw) + return I40E_ERR_PARAM; + + buf_size = i40e_prepare_add_macvlan(mv_list, &desc, count, seid); + + return i40e_asq_send_command_atomic(hw, &desc, mv_list, buf_size, + cmd_details, true); +} + +/** + * i40e_aq_add_macvlan_v2 + * @hw: pointer to the hw struct + * @seid: VSI for the mac address + * @mv_list: list of macvlans to be added + * @count: length of the list + * @cmd_details: pointer to command details structure or NULL + * @aq_status: pointer to Admin Queue status return value + * + * Add MAC/VLAN addresses to the HW filtering. + * The _v2 version returns the last Admin Queue status in aq_status + * to avoid race conditions in access to hw->aq.asq_last_status. + * It also calls _v2 versions of asq_send_command functions to + * get the aq_status on the stack. + **/ +i40e_status +i40e_aq_add_macvlan_v2(struct i40e_hw *hw, u16 seid, + struct i40e_aqc_add_macvlan_element_data *mv_list, + u16 count, struct i40e_asq_cmd_details *cmd_details, + enum i40e_admin_queue_err *aq_status) +{ + struct i40e_aq_desc desc; + u16 buf_size; + + if (count == 0 || !mv_list || !hw) + return I40E_ERR_PARAM; + + buf_size = i40e_prepare_add_macvlan(mv_list, &desc, count, seid); + + return i40e_asq_send_command_atomic_v2(hw, &desc, mv_list, buf_size, + cmd_details, true, aq_status); } /** @@ -2713,13 +2771,59 @@ i40e_status i40e_aq_remove_macvlan(struct i40e_hw *hw, u16 seid, if (buf_size > I40E_AQ_LARGE_BUF) desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB); - status = i40e_asq_send_command(hw, &desc, mv_list, buf_size, - cmd_details); + status = i40e_asq_send_command_atomic(hw, &desc, mv_list, buf_size, + cmd_details, true); return status; } /** + * i40e_aq_remove_macvlan_v2 + * @hw: pointer to the hw struct + * @seid: VSI for the mac address + * @mv_list: list of macvlans to be removed + * @count: length of the list + * @cmd_details: pointer to command details structure or NULL + * @aq_status: pointer to Admin Queue status return value + * + * Remove MAC/VLAN addresses from the HW filtering. + * The _v2 version returns the last Admin Queue status in aq_status + * to avoid race conditions in access to hw->aq.asq_last_status. + * It also calls _v2 versions of asq_send_command functions to + * get the aq_status on the stack. + **/ +i40e_status +i40e_aq_remove_macvlan_v2(struct i40e_hw *hw, u16 seid, + struct i40e_aqc_remove_macvlan_element_data *mv_list, + u16 count, struct i40e_asq_cmd_details *cmd_details, + enum i40e_admin_queue_err *aq_status) +{ + struct i40e_aqc_macvlan *cmd; + struct i40e_aq_desc desc; + u16 buf_size; + + if (count == 0 || !mv_list || !hw) + return I40E_ERR_PARAM; + + buf_size = count * sizeof(*mv_list); + + /* prep the rest of the request */ + i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_remove_macvlan); + cmd = (struct i40e_aqc_macvlan *)&desc.params.raw; + cmd->num_addresses = cpu_to_le16(count); + cmd->seid[0] = cpu_to_le16(I40E_AQC_MACVLAN_CMD_SEID_VALID | seid); + cmd->seid[1] = 0; + cmd->seid[2] = 0; + + desc.flags |= cpu_to_le16((u16)(I40E_AQ_FLAG_BUF | I40E_AQ_FLAG_RD)); + if (buf_size > I40E_AQ_LARGE_BUF) + desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_LB); + + return i40e_asq_send_command_atomic_v2(hw, &desc, mv_list, buf_size, + cmd_details, true, aq_status); +} + +/** * i40e_mirrorrule_op - Internal helper function to add/delete mirror rule * @hw: pointer to the hw struct * @opcode: AQ opcode for add or delete mirror rule @@ -3866,7 +3970,8 @@ i40e_status i40e_aq_delete_element(struct i40e_hw *hw, u16 seid, cmd->seid = cpu_to_le16(seid); - status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + status = i40e_asq_send_command_atomic(hw, &desc, NULL, 0, + cmd_details, true); return status; } @@ -4136,7 +4241,6 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw, struct i40e_filter_control_settings *settings) { u32 fcoe_cntx_size, fcoe_filt_size; - u32 pe_cntx_size, pe_filt_size; u32 fcoe_fmax; u32 val; @@ -4180,8 +4284,6 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw, case I40E_HASH_FILTER_SIZE_256K: case I40E_HASH_FILTER_SIZE_512K: case I40E_HASH_FILTER_SIZE_1M: - pe_filt_size = I40E_HASH_FILTER_BASE_SIZE; - pe_filt_size <<= (u32)settings->pe_filt_num; break; default: return I40E_ERR_PARAM; @@ -4198,8 +4300,6 @@ static i40e_status i40e_validate_filter_settings(struct i40e_hw *hw, case I40E_DMA_CNTX_SIZE_64K: case I40E_DMA_CNTX_SIZE_128K: case I40E_DMA_CNTX_SIZE_256K: - pe_cntx_size = I40E_DMA_CNTX_BASE_SIZE; - pe_cntx_size <<= (u32)settings->pe_cntx_num; break; default: return I40E_ERR_PARAM; @@ -4875,6 +4975,7 @@ i40e_status i40e_write_phy_register(struct i40e_hw *hw, status = i40e_write_phy_register_clause22(hw, reg, phy_addr, value); break; + case I40E_DEV_ID_1G_BASE_T_BC: case I40E_DEV_ID_5G_BASE_T_BC: case I40E_DEV_ID_10G_BASE_T: case I40E_DEV_ID_10G_BASE_T4: @@ -4913,6 +5014,7 @@ i40e_status i40e_read_phy_register(struct i40e_hw *hw, status = i40e_read_phy_register_clause22(hw, reg, phy_addr, value); break; + case I40E_DEV_ID_1G_BASE_T_BC: case I40E_DEV_ID_5G_BASE_T_BC: case I40E_DEV_ID_10G_BASE_T: case I40E_DEV_ID_10G_BASE_T4: diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c index 2c1b1da1220e..c9dcd6d92c83 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c +++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c @@ -240,7 +240,7 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid) (unsigned long int)vsi->net_stats_offsets.rx_compressed, (unsigned long int)vsi->net_stats_offsets.tx_compressed); dev_info(&pf->pdev->dev, - " tx_restart = %d, tx_busy = %d, rx_buf_failed = %d, rx_page_failed = %d\n", + " tx_restart = %llu, tx_busy = %llu, rx_buf_failed = %llu, rx_page_failed = %llu\n", vsi->tx_restart, vsi->tx_busy, vsi->rx_buf_failed, vsi->rx_page_failed); rcu_read_lock(); @@ -275,9 +275,8 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid) rx_ring->rx_stats.alloc_page_failed, rx_ring->rx_stats.alloc_buff_failed); dev_info(&pf->pdev->dev, - " rx_rings[%i]: rx_stats: realloc_count = %lld, page_reuse_count = %lld\n", + " rx_rings[%i]: rx_stats: realloc_count = 0, page_reuse_count = %lld\n", i, - rx_ring->rx_stats.realloc_count, rx_ring->rx_stats.page_reuse_count); dev_info(&pf->pdev->dev, " rx_rings[%i]: size = %i\n", @@ -310,10 +309,11 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid) tx_ring->stats.bytes, tx_ring->tx_stats.restart_queue); dev_info(&pf->pdev->dev, - " tx_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld\n", + " tx_rings[%i]: tx_stats: tx_busy = %lld, tx_done_old = %lld, tx_stopped = %lld\n", i, tx_ring->tx_stats.tx_busy, - tx_ring->tx_stats.tx_done_old); + tx_ring->tx_stats.tx_done_old, + tx_ring->tx_stats.tx_stopped); dev_info(&pf->pdev->dev, " tx_rings[%i]: size = %i\n", i, tx_ring->size); @@ -742,10 +742,8 @@ static void i40e_dbg_dump_vf(struct i40e_pf *pf, int vf_id) vsi = pf->vsi[vf->lan_vsi_idx]; dev_info(&pf->pdev->dev, "vf %2d: VSI id=%d, seid=%d, qps=%d\n", vf_id, vf->lan_vsi_id, vsi->seid, vf->num_queue_pairs); - dev_info(&pf->pdev->dev, " num MDD=%lld, invalid msg=%lld, valid msg=%lld\n", - vf->num_mdd_events, - vf->num_invalid_msgs, - vf->num_valid_msgs); + dev_info(&pf->pdev->dev, " num MDD=%lld\n", + vf->num_mdd_events); } else { dev_info(&pf->pdev->dev, "invalid VF id %d\n", vf_id); } diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h index 1bcb0ec0f0c0..d9c51a238dcc 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_devids.h +++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h @@ -24,8 +24,10 @@ #define I40E_DEV_ID_10G_B 0x104F #define I40E_DEV_ID_10G_SFP 0x104E #define I40E_DEV_ID_5G_BASE_T_BC 0x101F +#define I40E_DEV_ID_1G_BASE_T_BC 0x0DD2 #define I40E_IS_X710TL_DEVICE(d) \ - (((d) == I40E_DEV_ID_5G_BASE_T_BC) || \ + (((d) == I40E_DEV_ID_1G_BASE_T_BC) || \ + ((d) == I40E_DEV_ID_5G_BASE_T_BC) || \ ((d) == I40E_DEV_ID_10G_BASE_T_BC)) #define I40E_DEV_ID_KX_X722 0x37CE #define I40E_DEV_ID_QSFP_X722 0x37CF @@ -33,6 +35,7 @@ #define I40E_DEV_ID_1G_BASE_T_X722 0x37D1 #define I40E_DEV_ID_10G_BASE_T_X722 0x37D2 #define I40E_DEV_ID_SFP_I_X722 0x37D3 +#define I40E_DEV_ID_SFP_X722_A 0x0DDA #endif /* _I40E_DEVIDS_H_ */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 513ba6974355..4a6a6e48c615 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -236,8 +236,6 @@ static void __i40e_add_stat_strings(u8 **p, const struct i40e_stats stats[], I40E_STAT(struct i40e_cp_veb_tc_stats, _name, _stat) #define I40E_PFC_STAT(_name, _stat) \ I40E_STAT(struct i40e_pfc_stats, _name, _stat) -#define I40E_QUEUE_STAT(_name, _stat) \ - I40E_STAT(struct i40e_ring, _name, _stat) static const struct i40e_stats i40e_gstrings_net_stats[] = { I40E_NETDEV_STAT(rx_packets), @@ -293,8 +291,14 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = { I40E_VSI_STAT("tx_linearize", tx_linearize), I40E_VSI_STAT("tx_force_wb", tx_force_wb), I40E_VSI_STAT("tx_busy", tx_busy), + I40E_VSI_STAT("tx_stopped", tx_stopped), I40E_VSI_STAT("rx_alloc_fail", rx_buf_failed), I40E_VSI_STAT("rx_pg_alloc_fail", rx_page_failed), + I40E_VSI_STAT("rx_cache_reuse", rx_page_reuse), + I40E_VSI_STAT("rx_cache_alloc", rx_page_alloc), + I40E_VSI_STAT("rx_cache_waive", rx_page_waive), + I40E_VSI_STAT("rx_cache_busy", rx_page_busy), + I40E_VSI_STAT("tx_restart", tx_restart), }; /* These PF_STATs might look like duplicates of some NETDEV_STATs, @@ -451,6 +455,8 @@ static const struct i40e_priv_flags i40e_gstrings_priv_flags[] = { I40E_PRIV_FLAG("disable-fw-lldp", I40E_FLAG_DISABLE_FW_LLDP, 0), I40E_PRIV_FLAG("rs-fec", I40E_FLAG_RS_FEC, 0), I40E_PRIV_FLAG("base-r-fec", I40E_FLAG_BASE_R_FEC, 0), + I40E_PRIV_FLAG("vf-vlan-pruning", + I40E_FLAG_VF_VLAN_PRUNING, 0), }; #define I40E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40e_gstrings_priv_flags) @@ -1135,6 +1141,71 @@ static int i40e_get_link_ksettings(struct net_device *netdev, return 0; } +#define I40E_LBIT_SIZE 8 +/** + * i40e_speed_to_link_speed - Translate decimal speed to i40e_aq_link_speed + * @speed: speed in decimal + * @ks: ethtool ksettings + * + * Return i40e_aq_link_speed based on speed + **/ +static enum i40e_aq_link_speed +i40e_speed_to_link_speed(__u32 speed, const struct ethtool_link_ksettings *ks) +{ + enum i40e_aq_link_speed link_speed = I40E_LINK_SPEED_UNKNOWN; + bool speed_changed = false; + int i, j; + + static const struct { + __u32 speed; + enum i40e_aq_link_speed link_speed; + __u8 bit[I40E_LBIT_SIZE]; + } i40e_speed_lut[] = { +#define I40E_LBIT(mode) ETHTOOL_LINK_MODE_ ## mode ##_Full_BIT + {SPEED_100, I40E_LINK_SPEED_100MB, {I40E_LBIT(100baseT)} }, + {SPEED_1000, I40E_LINK_SPEED_1GB, + {I40E_LBIT(1000baseT), I40E_LBIT(1000baseX), + I40E_LBIT(1000baseKX)} }, + {SPEED_10000, I40E_LINK_SPEED_10GB, + {I40E_LBIT(10000baseT), I40E_LBIT(10000baseKR), + I40E_LBIT(10000baseLR), I40E_LBIT(10000baseCR), + I40E_LBIT(10000baseSR), I40E_LBIT(10000baseKX4)} }, + + {SPEED_25000, I40E_LINK_SPEED_25GB, + {I40E_LBIT(25000baseCR), I40E_LBIT(25000baseKR), + I40E_LBIT(25000baseSR)} }, + {SPEED_40000, I40E_LINK_SPEED_40GB, + {I40E_LBIT(40000baseKR4), I40E_LBIT(40000baseCR4), + I40E_LBIT(40000baseSR4), I40E_LBIT(40000baseLR4)} }, + {SPEED_20000, I40E_LINK_SPEED_20GB, + {I40E_LBIT(20000baseKR2)} }, + {SPEED_2500, I40E_LINK_SPEED_2_5GB, {I40E_LBIT(2500baseT)} }, + {SPEED_5000, I40E_LINK_SPEED_5GB, {I40E_LBIT(2500baseT)} } +#undef I40E_LBIT +}; + + for (i = 0; i < ARRAY_SIZE(i40e_speed_lut); i++) { + if (i40e_speed_lut[i].speed == speed) { + for (j = 0; j < I40E_LBIT_SIZE; j++) { + if (test_bit(i40e_speed_lut[i].bit[j], + ks->link_modes.supported)) { + speed_changed = true; + break; + } + if (!i40e_speed_lut[i].bit[j]) + break; + } + if (speed_changed) { + link_speed = i40e_speed_lut[i].link_speed; + break; + } + } + } + return link_speed; +} + +#undef I40E_LBIT_SIZE + /** * i40e_set_link_ksettings - Set Speed and Duplex * @netdev: network interface device structure @@ -1151,12 +1222,14 @@ static int i40e_set_link_ksettings(struct net_device *netdev, struct ethtool_link_ksettings copy_ks; struct i40e_aq_set_phy_config config; struct i40e_pf *pf = np->vsi->back; + enum i40e_aq_link_speed link_speed; struct i40e_vsi *vsi = np->vsi; struct i40e_hw *hw = &pf->hw; bool autoneg_changed = false; i40e_status status = 0; int timeout = 50; int err = 0; + __u32 speed; u8 autoneg; /* Changing port settings is not supported if this isn't the @@ -1189,6 +1262,7 @@ static int i40e_set_link_ksettings(struct net_device *netdev, /* save autoneg out of ksettings */ autoneg = copy_ks.base.autoneg; + speed = copy_ks.base.speed; /* get our own copy of the bits to check against */ memset(&safe_ks, 0, sizeof(struct ethtool_link_ksettings)); @@ -1207,6 +1281,7 @@ static int i40e_set_link_ksettings(struct net_device *netdev, /* set autoneg back to what it currently is */ copy_ks.base.autoneg = safe_ks.base.autoneg; + copy_ks.base.speed = safe_ks.base.speed; /* If copy_ks.base and safe_ks.base are not the same now, then they are * trying to set something that we do not support. @@ -1323,6 +1398,27 @@ static int i40e_set_link_ksettings(struct net_device *netdev, 40000baseLR4_Full)) config.link_speed |= I40E_LINK_SPEED_40GB; + /* Autonegotiation must be disabled to change speed */ + if ((speed != SPEED_UNKNOWN && safe_ks.base.speed != speed) && + (autoneg == AUTONEG_DISABLE || + (safe_ks.base.autoneg == AUTONEG_DISABLE && !autoneg_changed))) { + link_speed = i40e_speed_to_link_speed(speed, ks); + if (link_speed == I40E_LINK_SPEED_UNKNOWN) { + netdev_info(netdev, "Given speed is not supported\n"); + err = -EOPNOTSUPP; + goto done; + } else { + config.link_speed = link_speed; + } + } else { + if (safe_ks.base.speed != speed) { + netdev_info(netdev, + "Unable to set speed, disable autoneg\n"); + err = -EOPNOTSUPP; + goto done; + } + } + /* If speed didn't get set, set it to what it currently is. * This is needed because if advertise is 0 (as it is when autoneg * is disabled) then speed won't get set. @@ -1905,10 +2001,10 @@ static void i40e_get_drvinfo(struct net_device *netdev, struct i40e_vsi *vsi = np->vsi; struct i40e_pf *pf = vsi->back; - strlcpy(drvinfo->driver, i40e_driver_name, sizeof(drvinfo->driver)); - strlcpy(drvinfo->fw_version, i40e_nvm_version_str(&pf->hw), + strscpy(drvinfo->driver, i40e_driver_name, sizeof(drvinfo->driver)); + strscpy(drvinfo->fw_version, i40e_nvm_version_str(&pf->hw), sizeof(drvinfo->fw_version)); - strlcpy(drvinfo->bus_info, pci_name(pf->pdev), + strscpy(drvinfo->bus_info, pci_name(pf->pdev), sizeof(drvinfo->bus_info)); drvinfo->n_priv_flags = I40E_PRIV_FLAGS_STR_LEN; if (pf->hw.pf_id == 0) @@ -1916,7 +2012,9 @@ static void i40e_get_drvinfo(struct net_device *netdev, } static void i40e_get_ringparam(struct net_device *netdev, - struct ethtool_ringparam *ring) + struct ethtool_ringparam *ring, + struct kernel_ethtool_ringparam *kernel_ring, + struct netlink_ext_ack *extack) { struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_pf *pf = np->vsi->back; @@ -1944,7 +2042,9 @@ static bool i40e_active_tx_ring_index(struct i40e_vsi *vsi, u16 index) } static int i40e_set_ringparam(struct net_device *netdev, - struct ethtool_ringparam *ring) + struct ethtool_ringparam *ring, + struct kernel_ethtool_ringparam *kernel_ring, + struct netlink_ext_ack *extack) { struct i40e_ring *tx_rings = NULL, *rx_rings = NULL; struct i40e_netdev_priv *np = netdev_priv(netdev); @@ -2083,9 +2183,6 @@ static int i40e_set_ringparam(struct net_device *netdev, err = i40e_setup_rx_descriptors(&rx_rings[i]); if (err) goto rx_unwind; - err = i40e_alloc_rx_bi(&rx_rings[i]); - if (err) - goto rx_unwind; /* now allocate the Rx buffers to make sure the OS * has enough memory, any failure here means abort @@ -2576,15 +2673,16 @@ static void i40e_diag_test(struct net_device *netdev, set_bit(__I40E_TESTING, pf->state); + if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) || + test_bit(__I40E_RESET_INTR_RECEIVED, pf->state)) { + dev_warn(&pf->pdev->dev, + "Cannot start offline testing when PF is in reset state.\n"); + goto skip_ol_tests; + } + if (i40e_active_vfs(pf) || i40e_active_vmdqs(pf)) { dev_warn(&pf->pdev->dev, "Please take active VFs and Netqueues offline and restart the adapter before running NIC diagnostics\n"); - data[I40E_ETH_TEST_REG] = 1; - data[I40E_ETH_TEST_EEPROM] = 1; - data[I40E_ETH_TEST_INTR] = 1; - data[I40E_ETH_TEST_LINK] = 1; - eth_test->flags |= ETH_TEST_FL_FAILED; - clear_bit(__I40E_TESTING, pf->state); goto skip_ol_tests; } @@ -2631,9 +2729,17 @@ static void i40e_diag_test(struct net_device *netdev, data[I40E_ETH_TEST_INTR] = 0; } -skip_ol_tests: - netif_info(pf, drv, netdev, "testing finished\n"); + return; + +skip_ol_tests: + data[I40E_ETH_TEST_REG] = 1; + data[I40E_ETH_TEST_EEPROM] = 1; + data[I40E_ETH_TEST_INTR] = 1; + data[I40E_ETH_TEST_LINK] = 1; + eth_test->flags |= ETH_TEST_FL_FAILED; + clear_bit(__I40E_TESTING, pf->state); + netif_info(pf, drv, netdev, "testing failed\n"); } static void i40e_get_wol(struct net_device *netdev, @@ -3079,10 +3185,17 @@ static int i40e_get_rss_hash_opts(struct i40e_pf *pf, struct ethtool_rxnfc *cmd) if (cmd->flow_type == TCP_V4_FLOW || cmd->flow_type == UDP_V4_FLOW) { - if (i_set & I40E_L3_SRC_MASK) - cmd->data |= RXH_IP_SRC; - if (i_set & I40E_L3_DST_MASK) - cmd->data |= RXH_IP_DST; + if (hw->mac.type == I40E_MAC_X722) { + if (i_set & I40E_X722_L3_SRC_MASK) + cmd->data |= RXH_IP_SRC; + if (i_set & I40E_X722_L3_DST_MASK) + cmd->data |= RXH_IP_DST; + } else { + if (i_set & I40E_L3_SRC_MASK) + cmd->data |= RXH_IP_SRC; + if (i_set & I40E_L3_DST_MASK) + cmd->data |= RXH_IP_DST; + } } else if (cmd->flow_type == TCP_V6_FLOW || cmd->flow_type == UDP_V6_FLOW) { if (i_set & I40E_L3_V6_SRC_MASK) @@ -3440,12 +3553,15 @@ static int i40e_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, /** * i40e_get_rss_hash_bits - Read RSS Hash bits from register + * @hw: hw structure * @nfc: pointer to user request * @i_setc: bits currently set * * Returns value of bits to be set per user request **/ -static u64 i40e_get_rss_hash_bits(struct ethtool_rxnfc *nfc, u64 i_setc) +static u64 i40e_get_rss_hash_bits(struct i40e_hw *hw, + struct ethtool_rxnfc *nfc, + u64 i_setc) { u64 i_set = i_setc; u64 src_l3 = 0, dst_l3 = 0; @@ -3464,8 +3580,13 @@ static u64 i40e_get_rss_hash_bits(struct ethtool_rxnfc *nfc, u64 i_setc) dst_l3 = I40E_L3_V6_DST_MASK; } else if (nfc->flow_type == TCP_V4_FLOW || nfc->flow_type == UDP_V4_FLOW) { - src_l3 = I40E_L3_SRC_MASK; - dst_l3 = I40E_L3_DST_MASK; + if (hw->mac.type == I40E_MAC_X722) { + src_l3 = I40E_X722_L3_SRC_MASK; + dst_l3 = I40E_X722_L3_DST_MASK; + } else { + src_l3 = I40E_L3_SRC_MASK; + dst_l3 = I40E_L3_DST_MASK; + } } else { /* Any other flow type are not supported here */ return i_set; @@ -3483,6 +3604,7 @@ static u64 i40e_get_rss_hash_bits(struct ethtool_rxnfc *nfc, u64 i_setc) return i_set; } +#define FLOW_PCTYPES_SIZE 64 /** * i40e_set_rss_hash_opt - Enable/Disable flow types for RSS hash * @pf: pointer to the physical function struct @@ -3495,9 +3617,11 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc) struct i40e_hw *hw = &pf->hw; u64 hena = (u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0)) | ((u64)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1)) << 32); - u8 flow_pctype = 0; + DECLARE_BITMAP(flow_pctypes, FLOW_PCTYPES_SIZE); u64 i_set, i_setc; + bitmap_zero(flow_pctypes, FLOW_PCTYPES_SIZE); + if (pf->flags & I40E_FLAG_MFP_ENABLED) { dev_err(&pf->pdev->dev, "Change of RSS hash input set is not supported when MFP mode is enabled\n"); @@ -3513,36 +3637,35 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc) switch (nfc->flow_type) { case TCP_V4_FLOW: - flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV4_TCP; + set_bit(I40E_FILTER_PCTYPE_NONF_IPV4_TCP, flow_pctypes); if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) - hena |= - BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK); + set_bit(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK, + flow_pctypes); break; case TCP_V6_FLOW: - flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV6_TCP; - if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) - hena |= - BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK); + set_bit(I40E_FILTER_PCTYPE_NONF_IPV6_TCP, flow_pctypes); if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) - hena |= - BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK); + set_bit(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK, + flow_pctypes); break; case UDP_V4_FLOW: - flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV4_UDP; - if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) - hena |= - BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | - BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP); - + set_bit(I40E_FILTER_PCTYPE_NONF_IPV4_UDP, flow_pctypes); + if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) { + set_bit(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP, + flow_pctypes); + set_bit(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP, + flow_pctypes); + } hena |= BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4); break; case UDP_V6_FLOW: - flow_pctype = I40E_FILTER_PCTYPE_NONF_IPV6_UDP; - if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) - hena |= - BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) | - BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP); - + set_bit(I40E_FILTER_PCTYPE_NONF_IPV6_UDP, flow_pctypes); + if (pf->hw_features & I40E_HW_MULTIPLE_TCP_UDP_RSS_PCTYPE) { + set_bit(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP, + flow_pctypes); + set_bit(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP, + flow_pctypes); + } hena |= BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6); break; case AH_ESP_V4_FLOW: @@ -3575,17 +3698,20 @@ static int i40e_set_rss_hash_opt(struct i40e_pf *pf, struct ethtool_rxnfc *nfc) return -EINVAL; } - if (flow_pctype) { - i_setc = (u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, - flow_pctype)) | - ((u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, - flow_pctype)) << 32); - i_set = i40e_get_rss_hash_bits(nfc, i_setc); - i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, flow_pctype), - (u32)i_set); - i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, flow_pctype), - (u32)(i_set >> 32)); - hena |= BIT_ULL(flow_pctype); + if (bitmap_weight(flow_pctypes, FLOW_PCTYPES_SIZE)) { + u8 flow_id; + + for_each_set_bit(flow_id, flow_pctypes, FLOW_PCTYPES_SIZE) { + i_setc = (u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, flow_id)) | + ((u64)i40e_read_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, flow_id)) << 32); + i_set = i40e_get_rss_hash_bits(&pf->hw, nfc, i_setc); + + i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(0, flow_id), + (u32)i_set); + i40e_write_rx_ctl(hw, I40E_GLQF_HASH_INSET(1, flow_id), + (u32)(i_set >> 32)); + hena |= BIT_ULL(flow_id); + } } i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (u32)hena); @@ -4376,7 +4502,7 @@ static int i40e_check_fdir_input_set(struct i40e_vsi *vsi, (struct in6_addr *)&ipv6_full_mask)) new_mask |= I40E_L3_V6_DST_MASK; else if (ipv6_addr_any((struct in6_addr *) - &usr_ip6_spec->ip6src)) + &usr_ip6_spec->ip6dst)) new_mask &= ~I40E_L3_V6_DST_MASK; else return -EOPNOTSUPP; @@ -4912,7 +5038,7 @@ static int i40e_set_channels(struct net_device *dev, /* We do not support setting channels via ethtool when TCs are * configured through mqprio */ - if (pf->flags & I40E_FLAG_TC_MQPRIO) + if (i40e_is_tc_mqprio_enabled(pf)) return -EINVAL; /* verify they are not requesting separate vectors */ @@ -5275,6 +5401,13 @@ flags_complete: return -EOPNOTSUPP; } + if ((changed_flags & I40E_FLAG_VF_VLAN_PRUNING) && + pf->num_alloc_vfs) { + dev_warn(&pf->pdev->dev, + "Changing vf-vlan-pruning flag while VF(s) are active is not supported\n"); + return -EOPNOTSUPP; + } + if ((changed_flags & new_flags & I40E_FLAG_LINK_DOWN_ON_CLOSE_ENABLED) && (new_flags & I40E_FLAG_MFP_ENABLED)) diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index e118cf9265c7..b5dcd15ced36 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -66,6 +66,7 @@ static const struct pci_device_id i40e_pci_tbl[] = { {PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_A), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_B), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_C), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_BC), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T4), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_BC), 0}, @@ -77,6 +78,7 @@ static const struct pci_device_id i40e_pci_tbl[] = { {PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_I_X722), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722_A), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2_A), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_X710_N3000), 0}, @@ -99,6 +101,24 @@ MODULE_LICENSE("GPL v2"); static struct workqueue_struct *i40e_wq; +static void netdev_hw_addr_refcnt(struct i40e_mac_filter *f, + struct net_device *netdev, int delta) +{ + struct netdev_hw_addr *ha; + + if (!f || !netdev) + return; + + netdev_for_each_mc_addr(ha, netdev) { + if (ether_addr_equal(ha->addr, f->macaddr)) { + ha->refcount += delta; + if (ha->refcount <= 0) + ha->refcount = 1; + break; + } + } +} + /** * i40e_allocate_dma_mem_d - OS specific memory alloc for shared code * @hw: pointer to the HW structure @@ -178,10 +198,6 @@ int i40e_free_virt_mem_d(struct i40e_hw *hw, struct i40e_virt_mem *mem) * @id: an owner id to stick on the items assigned * * Returns the base item index of the lump, or negative for error - * - * The search_hint trick and lack of advanced fit-finding only work - * because we're highly likely to have all the same size lump requests. - * Linear search time and any fragmentation should be minimal. **/ static int i40e_get_lump(struct i40e_pf *pf, struct i40e_lump_tracking *pile, u16 needed, u16 id) @@ -196,8 +212,21 @@ static int i40e_get_lump(struct i40e_pf *pf, struct i40e_lump_tracking *pile, return -EINVAL; } - /* start the linear search with an imperfect hint */ - i = pile->search_hint; + /* Allocate last queue in the pile for FDIR VSI queue + * so it doesn't fragment the qp_pile + */ + if (pile == pf->qp_pile && pf->vsi[id]->type == I40E_VSI_FDIR) { + if (pile->list[pile->num_entries - 1] & I40E_PILE_VALID_BIT) { + dev_err(&pf->pdev->dev, + "Cannot allocate queue %d for I40E_VSI_FDIR\n", + pile->num_entries - 1); + return -ENOMEM; + } + pile->list[pile->num_entries - 1] = id | I40E_PILE_VALID_BIT; + return pile->num_entries - 1; + } + + i = 0; while (i < pile->num_entries) { /* skip already allocated entries */ if (pile->list[i] & I40E_PILE_VALID_BIT) { @@ -216,7 +245,6 @@ static int i40e_get_lump(struct i40e_pf *pf, struct i40e_lump_tracking *pile, for (j = 0; j < needed; j++) pile->list[i+j] = id | I40E_PILE_VALID_BIT; ret = i; - pile->search_hint = i + j; break; } @@ -239,7 +267,7 @@ static int i40e_put_lump(struct i40e_lump_tracking *pile, u16 index, u16 id) { int valid_id = (id | I40E_PILE_VALID_BIT); int count = 0; - int i; + u16 i; if (!pile || index >= pile->num_entries) return -EINVAL; @@ -251,8 +279,6 @@ static int i40e_put_lump(struct i40e_lump_tracking *pile, u16 index, u16 id) count++; } - if (count && index < pile->search_hint) - pile->search_hint = index; return count; } @@ -359,7 +385,9 @@ static void i40e_tx_timeout(struct net_device *netdev, unsigned int txqueue) set_bit(__I40E_GLOBAL_RESET_REQUESTED, pf->state); break; default: - netdev_err(netdev, "tx_timeout recovery unsuccessful\n"); + netdev_err(netdev, "tx_timeout recovery unsuccessful, device is in non-recoverable state.\n"); + set_bit(__I40E_DOWN_REQUESTED, pf->state); + set_bit(__I40E_VSI_DOWN_REQUESTED, vsi->state); break; } @@ -526,6 +554,47 @@ void i40e_pf_reset_stats(struct i40e_pf *pf) } /** + * i40e_compute_pci_to_hw_id - compute index form PCI function. + * @vsi: ptr to the VSI to read from. + * @hw: ptr to the hardware info. + **/ +static u32 i40e_compute_pci_to_hw_id(struct i40e_vsi *vsi, struct i40e_hw *hw) +{ + int pf_count = i40e_get_pf_count(hw); + + if (vsi->type == I40E_VSI_SRIOV) + return (hw->port * BIT(7)) / pf_count + vsi->vf_id; + + return hw->port + BIT(7); +} + +/** + * i40e_stat_update64 - read and update a 64 bit stat from the chip. + * @hw: ptr to the hardware info. + * @hireg: the high 32 bit reg to read. + * @loreg: the low 32 bit reg to read. + * @offset_loaded: has the initial offset been loaded yet. + * @offset: ptr to current offset value. + * @stat: ptr to the stat. + * + * Since the device stats are not reset at PFReset, they will not + * be zeroed when the driver starts. We'll 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. + **/ +static void i40e_stat_update64(struct i40e_hw *hw, u32 hireg, u32 loreg, + bool offset_loaded, u64 *offset, u64 *stat) +{ + u64 new_data; + + new_data = rd64(hw, loreg); + + if (!offset_loaded || new_data < *offset) + *offset = new_data; + *stat = new_data - *offset; +} + +/** * i40e_stat_update48 - read and update a 48 bit stat from the chip * @hw: ptr to the hardware info * @hireg: the high 32 bit reg to read @@ -597,6 +666,34 @@ static void i40e_stat_update_and_clear32(struct i40e_hw *hw, u32 reg, u64 *stat) } /** + * i40e_stats_update_rx_discards - update rx_discards. + * @vsi: ptr to the VSI to be updated. + * @hw: ptr to the hardware info. + * @stat_idx: VSI's stat_counter_idx. + * @offset_loaded: ptr to the VSI's stat_offsets_loaded. + * @stat_offset: ptr to stat_offset to store first read of specific register. + * @stat: ptr to VSI's stat to be updated. + **/ +static void +i40e_stats_update_rx_discards(struct i40e_vsi *vsi, struct i40e_hw *hw, + int stat_idx, bool offset_loaded, + struct i40e_eth_stats *stat_offset, + struct i40e_eth_stats *stat) +{ + u64 rx_rdpc, rx_rxerr; + + i40e_stat_update32(hw, I40E_GLV_RDPC(stat_idx), offset_loaded, + &stat_offset->rx_discards, &rx_rdpc); + i40e_stat_update64(hw, + I40E_GL_RXERR1H(i40e_compute_pci_to_hw_id(vsi, hw)), + I40E_GL_RXERR1L(i40e_compute_pci_to_hw_id(vsi, hw)), + offset_loaded, &stat_offset->rx_discards_other, + &rx_rxerr); + + stat->rx_discards = rx_rdpc + rx_rxerr; +} + +/** * i40e_update_eth_stats - Update VSI-specific ethernet statistics counters. * @vsi: the VSI to be updated **/ @@ -655,6 +752,10 @@ void i40e_update_eth_stats(struct i40e_vsi *vsi) I40E_GLV_BPTCL(stat_idx), vsi->stat_offsets_loaded, &oes->tx_broadcast, &es->tx_broadcast); + + i40e_stats_update_rx_discards(vsi, hw, stat_idx, + vsi->stat_offsets_loaded, oes, es); + vsi->stat_offsets_loaded = true; } @@ -749,18 +850,19 @@ void i40e_update_veb_stats(struct i40e_veb *veb) **/ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) { + u64 rx_page, rx_buf, rx_reuse, rx_alloc, rx_waive, rx_busy; struct i40e_pf *pf = vsi->back; struct rtnl_link_stats64 *ons; struct rtnl_link_stats64 *ns; /* netdev stats */ struct i40e_eth_stats *oes; struct i40e_eth_stats *es; /* device's eth stats */ - u32 tx_restart, tx_busy; + u64 tx_restart, tx_busy; struct i40e_ring *p; - u32 rx_page, rx_buf; u64 bytes, packets; unsigned int start; u64 tx_linearize; u64 tx_force_wb; + u64 tx_stopped; u64 rx_p, rx_b; u64 tx_p, tx_b; u16 q; @@ -780,8 +882,13 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) rx_b = rx_p = 0; tx_b = tx_p = 0; tx_restart = tx_busy = tx_linearize = tx_force_wb = 0; + tx_stopped = 0; rx_page = 0; rx_buf = 0; + rx_reuse = 0; + rx_alloc = 0; + rx_waive = 0; + rx_busy = 0; rcu_read_lock(); for (q = 0; q < vsi->num_queue_pairs; q++) { /* locate Tx ring */ @@ -800,6 +907,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) tx_busy += p->tx_stats.tx_busy; tx_linearize += p->tx_stats.tx_linearize; tx_force_wb += p->tx_stats.tx_force_wb; + tx_stopped += p->tx_stats.tx_stopped; /* locate Rx ring */ p = READ_ONCE(vsi->rx_rings[q]); @@ -815,6 +923,10 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) rx_p += packets; rx_buf += p->rx_stats.alloc_buff_failed; rx_page += p->rx_stats.alloc_page_failed; + rx_reuse += p->rx_stats.page_reuse_count; + rx_alloc += p->rx_stats.page_alloc_count; + rx_waive += p->rx_stats.page_waive_count; + rx_busy += p->rx_stats.page_busy_count; if (i40e_enabled_xdp_vsi(vsi)) { /* locate XDP ring */ @@ -840,8 +952,13 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) vsi->tx_busy = tx_busy; vsi->tx_linearize = tx_linearize; vsi->tx_force_wb = tx_force_wb; + vsi->tx_stopped = tx_stopped; vsi->rx_page_failed = rx_page; vsi->rx_buf_failed = rx_buf; + vsi->rx_page_reuse = rx_reuse; + vsi->rx_page_alloc = rx_alloc; + vsi->rx_page_waive = rx_waive; + vsi->rx_page_busy = rx_busy; ns->rx_packets = rx_p; ns->rx_bytes = rx_b; @@ -1328,6 +1445,114 @@ static int i40e_correct_mac_vlan_filters(struct i40e_vsi *vsi, } /** + * i40e_get_vf_new_vlan - Get new vlan id on a vf + * @vsi: the vsi to configure + * @new_mac: new mac filter to be added + * @f: existing mac filter, replaced with new_mac->f if new_mac is not NULL + * @vlan_filters: the number of active VLAN filters + * @trusted: flag if the VF is trusted + * + * Get new VLAN id based on current VLAN filters, trust, PVID + * and vf-vlan-prune-disable flag. + * + * Returns the value of the new vlan filter or + * the old value if no new filter is needed. + */ +static s16 i40e_get_vf_new_vlan(struct i40e_vsi *vsi, + struct i40e_new_mac_filter *new_mac, + struct i40e_mac_filter *f, + int vlan_filters, + bool trusted) +{ + s16 pvid = le16_to_cpu(vsi->info.pvid); + struct i40e_pf *pf = vsi->back; + bool is_any; + + if (new_mac) + f = new_mac->f; + + if (pvid && f->vlan != pvid) + return pvid; + + is_any = (trusted || + !(pf->flags & I40E_FLAG_VF_VLAN_PRUNING)); + + if ((vlan_filters && f->vlan == I40E_VLAN_ANY) || + (!is_any && !vlan_filters && f->vlan == I40E_VLAN_ANY) || + (is_any && !vlan_filters && f->vlan == 0)) { + if (is_any) + return I40E_VLAN_ANY; + else + return 0; + } + + return f->vlan; +} + +/** + * i40e_correct_vf_mac_vlan_filters - Correct non-VLAN VF filters if necessary + * @vsi: the vsi to configure + * @tmp_add_list: list of filters ready to be added + * @tmp_del_list: list of filters ready to be deleted + * @vlan_filters: the number of active VLAN filters + * @trusted: flag if the VF is trusted + * + * Correct VF VLAN filters based on current VLAN filters, trust, PVID + * and vf-vlan-prune-disable flag. + * + * In case of memory allocation failure return -ENOMEM. Otherwise, return 0. + * + * This function is only expected to be called from within + * i40e_sync_vsi_filters. + * + * NOTE: This function expects to be called while under the + * mac_filter_hash_lock + */ +static int i40e_correct_vf_mac_vlan_filters(struct i40e_vsi *vsi, + struct hlist_head *tmp_add_list, + struct hlist_head *tmp_del_list, + int vlan_filters, + bool trusted) +{ + struct i40e_mac_filter *f, *add_head; + struct i40e_new_mac_filter *new_mac; + struct hlist_node *h; + int bkt, new_vlan; + + hlist_for_each_entry(new_mac, tmp_add_list, hlist) { + new_mac->f->vlan = i40e_get_vf_new_vlan(vsi, new_mac, NULL, + vlan_filters, trusted); + } + + hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) { + new_vlan = i40e_get_vf_new_vlan(vsi, NULL, f, vlan_filters, + trusted); + if (new_vlan != f->vlan) { + add_head = i40e_add_filter(vsi, f->macaddr, new_vlan); + if (!add_head) + return -ENOMEM; + /* Create a temporary i40e_new_mac_filter */ + new_mac = kzalloc(sizeof(*new_mac), GFP_ATOMIC); + if (!new_mac) + return -ENOMEM; + new_mac->f = add_head; + new_mac->state = add_head->state; + + /* Add the new filter to the tmp list */ + hlist_add_head(&new_mac->hlist, tmp_add_list); + + /* Put the original filter into the delete list */ + f->state = I40E_FILTER_REMOVE; + hash_del(&f->hlist); + hlist_add_head(&f->hlist, tmp_del_list); + } + } + + vsi->has_vlan_filter = !!vlan_filters; + return 0; +} + +/** * i40e_rm_default_mac_filter - Remove the default MAC filter set by NVM * @vsi: the PF Main VSI - inappropriate for any other VSI * @macaddr: the MAC address @@ -1811,11 +2036,15 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi, * non-zero req_queue_pairs says that user requested a new * queue count via ethtool's set_channels, so use this * value for queues distribution across traffic classes + * We need at least one queue pair for the interface + * to be usable as we see in else statement. */ if (vsi->req_queue_pairs > 0) vsi->num_queue_pairs = vsi->req_queue_pairs; else if (pf->flags & I40E_FLAG_MSIX_ENABLED) vsi->num_queue_pairs = pf->num_lan_msix; + else + vsi->num_queue_pairs = 1; } /* Number of queues per enabled TC */ @@ -2036,6 +2265,7 @@ static void i40e_undo_add_filter_entries(struct i40e_vsi *vsi, hlist_for_each_entry_safe(new, h, from, hlist) { /* We can simply free the wrapper structure */ hlist_del(&new->hlist); + netdev_hw_addr_refcnt(new->f, vsi->netdev, -1); kfree(new); } } @@ -2118,19 +2348,19 @@ void i40e_aqc_del_filters(struct i40e_vsi *vsi, const char *vsi_name, int num_del, int *retval) { struct i40e_hw *hw = &vsi->back->hw; + enum i40e_admin_queue_err aq_status; i40e_status aq_ret; - int aq_err; - aq_ret = i40e_aq_remove_macvlan(hw, vsi->seid, list, num_del, NULL); - aq_err = hw->aq.asq_last_status; + aq_ret = i40e_aq_remove_macvlan_v2(hw, vsi->seid, list, num_del, NULL, + &aq_status); /* Explicitly ignore and do not report when firmware returns ENOENT */ - if (aq_ret && !(aq_err == I40E_AQ_RC_ENOENT)) { + if (aq_ret && !(aq_status == I40E_AQ_RC_ENOENT)) { *retval = -EIO; dev_info(&vsi->back->pdev->dev, "ignoring delete macvlan error on %s, err %s, aq_err %s\n", vsi_name, i40e_stat_str(hw, aq_ret), - i40e_aq_str(hw, aq_err)); + i40e_aq_str(hw, aq_status)); } } @@ -2153,10 +2383,10 @@ void i40e_aqc_add_filters(struct i40e_vsi *vsi, const char *vsi_name, int num_add) { struct i40e_hw *hw = &vsi->back->hw; - int aq_err, fcnt; + enum i40e_admin_queue_err aq_status; + int fcnt; - i40e_aq_add_macvlan(hw, vsi->seid, list, num_add, NULL); - aq_err = hw->aq.asq_last_status; + i40e_aq_add_macvlan_v2(hw, vsi->seid, list, num_add, NULL, &aq_status); fcnt = i40e_update_filter_state(num_add, list, add_head); if (fcnt != num_add) { @@ -2164,17 +2394,19 @@ void i40e_aqc_add_filters(struct i40e_vsi *vsi, const char *vsi_name, set_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state); dev_warn(&vsi->back->pdev->dev, "Error %s adding RX filters on %s, promiscuous mode forced on\n", - i40e_aq_str(hw, aq_err), vsi_name); + i40e_aq_str(hw, aq_status), vsi_name); } else if (vsi->type == I40E_VSI_SRIOV || vsi->type == I40E_VSI_VMDQ1 || vsi->type == I40E_VSI_VMDQ2) { dev_warn(&vsi->back->pdev->dev, "Error %s adding RX filters on %s, please set promiscuous on manually for %s\n", - i40e_aq_str(hw, aq_err), vsi_name, vsi_name); + i40e_aq_str(hw, aq_status), vsi_name, + vsi_name); } else { dev_warn(&vsi->back->pdev->dev, "Error %s adding RX filters on %s, incorrect VSI type: %i.\n", - i40e_aq_str(hw, aq_err), vsi_name, vsi->type); + i40e_aq_str(hw, aq_status), vsi_name, + vsi->type); } } } @@ -2379,10 +2611,18 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) vlan_filters++; } - retval = i40e_correct_mac_vlan_filters(vsi, - &tmp_add_list, - &tmp_del_list, - vlan_filters); + if (vsi->type != I40E_VSI_SRIOV) + retval = i40e_correct_mac_vlan_filters + (vsi, &tmp_add_list, &tmp_del_list, + vlan_filters); + else + retval = i40e_correct_vf_mac_vlan_filters + (vsi, &tmp_add_list, &tmp_del_list, + vlan_filters, pf->vf[vsi->vf_id].trusted); + + hlist_for_each_entry(new, &tmp_add_list, hlist) + netdev_hw_addr_refcnt(new->f, vsi->netdev, 1); + if (retval) goto err_no_memory_locked; @@ -2515,6 +2755,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (new->f->state == I40E_FILTER_NEW) new->f->state = new->state; hlist_del(&new->hlist); + netdev_hw_addr_refcnt(new->f, vsi->netdev, -1); kfree(new); } spin_unlock_bh(&vsi->mac_filter_hash_lock); @@ -2806,8 +3047,21 @@ int i40e_add_vlan_all_mac(struct i40e_vsi *vsi, s16 vid) int bkt; hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) { - if (f->state == I40E_FILTER_REMOVE) + /* If we're asked to add a filter that has been marked for + * removal, it is safe to simply restore it to active state. + * __i40e_del_filter will have simply deleted any filters which + * were previously marked NEW or FAILED, so if it is currently + * marked REMOVE it must have previously been ACTIVE. Since we + * haven't yet run the sync filters task, just restore this + * filter to the ACTIVE state so that the sync task leaves it + * in place. + */ + if (f->state == I40E_FILTER_REMOVE && f->vlan == vid) { + f->state = I40E_FILTER_ACTIVE; + continue; + } else if (f->state == I40E_FILTER_REMOVE) { continue; + } add_f = i40e_add_filter(vsi, f->macaddr, vid); if (!add_f) { dev_info(&vsi->back->pdev->dev, @@ -3312,12 +3566,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) if (ring->vsi->type == I40E_VSI_MAIN) xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); - kfree(ring->rx_bi); ring->xsk_pool = i40e_xsk_pool(ring); if (ring->xsk_pool) { - ret = i40e_alloc_rx_bi_zc(ring); - if (ret) - return ret; ring->rx_buf_len = xsk_pool_get_rx_frame_size(ring->xsk_pool); /* For AF_XDP ZC, we disallow packets to span on @@ -3335,9 +3585,6 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) ring->queue_index); } else { - ret = i40e_alloc_rx_bi(ring); - if (ret) - return ret; ring->rx_buf_len = vsi->rx_buf_len; if (ring->vsi->type == I40E_VSI_MAIN) { ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, @@ -3625,7 +3872,7 @@ static void i40e_vsi_configure_msix(struct i40e_vsi *vsi) wr32(hw, I40E_PFINT_RATEN(vector - 1), i40e_intrl_usec_to_reg(vsi->int_rate_limit)); - /* Linked list for the queuepairs assigned to this vector */ + /* begin of linked list for RX queue assigned to this vector */ wr32(hw, I40E_PFINT_LNKLSTN(vector - 1), qp); for (q = 0; q < q_vector->num_ringpairs; q++) { u32 nextqp = has_xdp ? qp + vsi->alloc_queue_pairs : qp; @@ -3641,6 +3888,7 @@ static void i40e_vsi_configure_msix(struct i40e_vsi *vsi) wr32(hw, I40E_QINT_RQCTL(qp), val); if (has_xdp) { + /* TX queue with next queue set to TX */ val = I40E_QINT_TQCTL_CAUSE_ENA_MASK | (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) | (vector << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) | @@ -3650,7 +3898,7 @@ static void i40e_vsi_configure_msix(struct i40e_vsi *vsi) wr32(hw, I40E_QINT_TQCTL(nextqp), val); } - + /* TX queue with next RX or end of linked list */ val = I40E_QINT_TQCTL_CAUSE_ENA_MASK | (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) | (vector << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) | @@ -3719,7 +3967,6 @@ static void i40e_configure_msi_and_legacy(struct i40e_vsi *vsi) struct i40e_q_vector *q_vector = vsi->q_vectors[0]; struct i40e_pf *pf = vsi->back; struct i40e_hw *hw = &pf->hw; - u32 val; /* set the ITR configuration */ q_vector->rx.next_update = jiffies + 1; @@ -3736,28 +3983,20 @@ static void i40e_configure_msi_and_legacy(struct i40e_vsi *vsi) /* FIRSTQ_INDX = 0, FIRSTQ_TYPE = 0 (rx) */ wr32(hw, I40E_PFINT_LNKLST0, 0); - /* Associate the queue pair to the vector and enable the queue int */ - val = I40E_QINT_RQCTL_CAUSE_ENA_MASK | - (I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) | - (nextqp << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT)| - (I40E_QUEUE_TYPE_TX << I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT); - - wr32(hw, I40E_QINT_RQCTL(0), val); + /* Associate the queue pair to the vector and enable the queue + * interrupt RX queue in linked list with next queue set to TX + */ + wr32(hw, I40E_QINT_RQCTL(0), I40E_QINT_RQCTL_VAL(nextqp, 0, TX)); if (i40e_enabled_xdp_vsi(vsi)) { - val = I40E_QINT_TQCTL_CAUSE_ENA_MASK | - (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT)| - (I40E_QUEUE_TYPE_TX - << I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT); - - wr32(hw, I40E_QINT_TQCTL(nextqp), val); + /* TX queue in linked list with next queue set to TX */ + wr32(hw, I40E_QINT_TQCTL(nextqp), + I40E_QINT_TQCTL_VAL(nextqp, 0, TX)); } - val = I40E_QINT_TQCTL_CAUSE_ENA_MASK | - (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) | - (I40E_QUEUE_END_OF_LIST << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT); - - wr32(hw, I40E_QINT_TQCTL(0), val); + /* last TX queue so the next RX queue doesn't matter */ + wr32(hw, I40E_QINT_TQCTL(0), + I40E_QINT_TQCTL_VAL(I40E_QUEUE_END_OF_LIST, 0, RX)); i40e_flush(hw); } @@ -3891,10 +4130,10 @@ static int i40e_vsi_request_irq_msix(struct i40e_vsi *vsi, char *basename) * * get_cpu_mask returns a static constant mask with * a permanent lifetime so it's ok to pass to - * irq_set_affinity_hint without making a copy. + * irq_update_affinity_hint without making a copy. */ cpu = cpumask_local_spread(q_vector->v_idx, -1); - irq_set_affinity_hint(irq_num, get_cpu_mask(cpu)); + irq_update_affinity_hint(irq_num, get_cpu_mask(cpu)); } vsi->irqs_ready = true; @@ -3905,7 +4144,7 @@ free_queue_irqs: vector--; irq_num = pf->msix_entries[base + vector].vector; irq_set_affinity_notifier(irq_num, NULL); - irq_set_affinity_hint(irq_num, NULL); + irq_update_affinity_hint(irq_num, NULL); free_irq(irq_num, &vsi->q_vectors[vector]); } return err; @@ -3988,7 +4227,6 @@ static void i40e_free_misc_vector(struct i40e_pf *pf) i40e_flush(&pf->hw); if (pf->flags & I40E_FLAG_MSIX_ENABLED && pf->msix_entries) { - synchronize_irq(pf->msix_entries[0].vector); free_irq(pf->msix_entries[0].vector, pf); clear_bit(__I40E_MISC_IRQ_REQUESTED, pf->state); } @@ -4726,8 +4964,7 @@ static void i40e_vsi_free_irq(struct i40e_vsi *vsi) /* clear the affinity notifier in the IRQ descriptor */ irq_set_affinity_notifier(irq_num, NULL); /* remove our suggested affinity mask for this IRQ */ - irq_set_affinity_hint(irq_num, NULL); - synchronize_irq(irq_num); + irq_update_affinity_hint(irq_num, NULL); free_irq(irq_num, vsi->q_vectors[i]); /* Tear down the interrupt queue link list @@ -5211,7 +5448,7 @@ static u8 i40e_pf_get_num_tc(struct i40e_pf *pf) u8 num_tc = 0; struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config; - if (pf->flags & I40E_FLAG_TC_MQPRIO) + if (i40e_is_tc_mqprio_enabled(pf)) return pf->vsi[pf->lan_vsi]->mqprio_qopt.qopt.num_tc; /* If neither MQPRIO nor DCB is enabled, then always use single TC */ @@ -5243,7 +5480,7 @@ static u8 i40e_pf_get_num_tc(struct i40e_pf *pf) **/ static u8 i40e_pf_get_tc_map(struct i40e_pf *pf) { - if (pf->flags & I40E_FLAG_TC_MQPRIO) + if (i40e_is_tc_mqprio_enabled(pf)) return i40e_mqprio_get_enabled_tc(pf); /* If neither MQPRIO nor DCB is enabled for this PF then just return @@ -5340,7 +5577,7 @@ static int i40e_vsi_configure_bw_alloc(struct i40e_vsi *vsi, u8 enabled_tc, int i; /* There is no need to reset BW when mqprio mode is on. */ - if (pf->flags & I40E_FLAG_TC_MQPRIO) + if (i40e_is_tc_mqprio_enabled(pf)) return 0; if (!vsi->mqprio_qopt.qopt.hw && !(pf->flags & I40E_FLAG_DCB_ENABLED)) { ret = i40e_set_bw_limit(vsi, vsi->seid, 0); @@ -5412,7 +5649,7 @@ static void i40e_vsi_config_netdev_tc(struct i40e_vsi *vsi, u8 enabled_tc) vsi->tc_config.tc_info[i].qoffset); } - if (pf->flags & I40E_FLAG_TC_MQPRIO) + if (i40e_is_tc_mqprio_enabled(pf)) return; /* Assign UP2TC map for the VSI */ @@ -5573,7 +5810,7 @@ static int i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc) ctxt.vf_num = 0; ctxt.uplink_seid = vsi->uplink_seid; ctxt.info = vsi->info; - if (vsi->back->flags & I40E_FLAG_TC_MQPRIO) { + if (i40e_is_tc_mqprio_enabled(pf)) { ret = i40e_vsi_setup_queue_map_mqprio(vsi, &ctxt, enabled_tc); if (ret) goto out; @@ -5658,6 +5895,26 @@ static int i40e_get_link_speed(struct i40e_vsi *vsi) } /** + * i40e_bw_bytes_to_mbits - Convert max_tx_rate from bytes to mbits + * @vsi: Pointer to vsi structure + * @max_tx_rate: max TX rate in bytes to be converted into Mbits + * + * Helper function to convert units before send to set BW limit + **/ +static u64 i40e_bw_bytes_to_mbits(struct i40e_vsi *vsi, u64 max_tx_rate) +{ + if (max_tx_rate < I40E_BW_MBPS_DIVISOR) { + dev_warn(&vsi->back->pdev->dev, + "Setting max tx rate to minimum usable value of 50Mbps.\n"); + max_tx_rate = I40E_BW_CREDIT_DIVISOR; + } else { + do_div(max_tx_rate, I40E_BW_MBPS_DIVISOR); + } + + return max_tx_rate; +} + +/** * i40e_set_bw_limit - setup BW limit for Tx traffic based on max_tx_rate * @vsi: VSI to be configured * @seid: seid of the channel/VSI @@ -5679,10 +5936,10 @@ int i40e_set_bw_limit(struct i40e_vsi *vsi, u16 seid, u64 max_tx_rate) max_tx_rate, seid); return -EINVAL; } - if (max_tx_rate && max_tx_rate < 50) { + if (max_tx_rate && max_tx_rate < I40E_BW_CREDIT_DIVISOR) { dev_warn(&pf->pdev->dev, "Setting max tx rate to minimum usable value of 50Mbps.\n"); - max_tx_rate = 50; + max_tx_rate = I40E_BW_CREDIT_DIVISOR; } /* Tx rate credits are in values of 50Mbps, 0 is disabled */ @@ -6297,7 +6554,7 @@ int i40e_create_queue_channel(struct i40e_vsi *vsi, pf->flags |= I40E_FLAG_VEB_MODE_ENABLED; if (vsi->type == I40E_VSI_MAIN) { - if (pf->flags & I40E_FLAG_TC_MQPRIO) + if (i40e_is_tc_mqprio_enabled(pf)) i40e_do_reset(pf, I40E_PF_RESET_FLAG, true); else i40e_do_reset_safe(pf, I40E_PF_RESET_FLAG); @@ -6408,6 +6665,9 @@ static int i40e_configure_queue_channels(struct i40e_vsi *vsi) vsi->tc_seid_map[i] = ch->seid; } } + + /* reset to reconfigure TX queue contexts */ + i40e_do_reset(vsi->back, I40E_PF_RESET_FLAG, true); return ret; err_free: @@ -7505,42 +7765,43 @@ static void i40e_free_macvlan_channels(struct i40e_vsi *vsi) static int i40e_fwd_ring_up(struct i40e_vsi *vsi, struct net_device *vdev, struct i40e_fwd_adapter *fwd) { + struct i40e_channel *ch = NULL, *ch_tmp, *iter; int ret = 0, num_tc = 1, i, aq_err; - struct i40e_channel *ch, *ch_tmp; struct i40e_pf *pf = vsi->back; struct i40e_hw *hw = &pf->hw; - if (list_empty(&vsi->macvlan_list)) - return -EINVAL; - /* Go through the list and find an available channel */ - list_for_each_entry_safe(ch, ch_tmp, &vsi->macvlan_list, list) { - if (!i40e_is_channel_macvlan(ch)) { - ch->fwd = fwd; + list_for_each_entry_safe(iter, ch_tmp, &vsi->macvlan_list, list) { + if (!i40e_is_channel_macvlan(iter)) { + iter->fwd = fwd; /* record configuration for macvlan interface in vdev */ for (i = 0; i < num_tc; i++) netdev_bind_sb_channel_queue(vsi->netdev, vdev, i, - ch->num_queue_pairs, - ch->base_queue); - for (i = 0; i < ch->num_queue_pairs; i++) { + iter->num_queue_pairs, + iter->base_queue); + for (i = 0; i < iter->num_queue_pairs; i++) { struct i40e_ring *tx_ring, *rx_ring; u16 pf_q; - pf_q = ch->base_queue + i; + pf_q = iter->base_queue + i; /* Get to TX ring ptr */ tx_ring = vsi->tx_rings[pf_q]; - tx_ring->ch = ch; + tx_ring->ch = iter; /* Get the RX ring ptr */ rx_ring = vsi->rx_rings[pf_q]; - rx_ring->ch = ch; + rx_ring->ch = iter; } + ch = iter; break; } } + if (!ch) + return -EINVAL; + /* Guarantee all rings are updated before we update the * MAC address filter. */ @@ -7690,7 +7951,7 @@ static void *i40e_fwd_add(struct net_device *netdev, struct net_device *vdev) netdev_info(netdev, "Macvlans are not supported when DCB is enabled\n"); return ERR_PTR(-EINVAL); } - if ((pf->flags & I40E_FLAG_TC_MQPRIO)) { + if (i40e_is_tc_mqprio_enabled(pf)) { netdev_info(netdev, "Macvlans are not supported when HW TC offload is on\n"); return ERR_PTR(-EINVAL); } @@ -7943,7 +8204,7 @@ config_tc: /* Quiesce VSI queues */ i40e_quiesce_vsi(vsi); - if (!hw && !(pf->flags & I40E_FLAG_TC_MQPRIO)) + if (!hw && !i40e_is_tc_mqprio_enabled(pf)) i40e_remove_queue_channels(vsi); /* Configure VSI for enabled TCs */ @@ -7967,11 +8228,11 @@ config_tc: "Setup channel (id:%u) utilizing num_queues %d\n", vsi->seid, vsi->tc_config.tc_info[0].qcount); - if (pf->flags & I40E_FLAG_TC_MQPRIO) { + if (i40e_is_tc_mqprio_enabled(pf)) { if (vsi->mqprio_qopt.max_rate[0]) { - u64 max_tx_rate = vsi->mqprio_qopt.max_rate[0]; + u64 max_tx_rate = i40e_bw_bytes_to_mbits(vsi, + vsi->mqprio_qopt.max_rate[0]); - do_div(max_tx_rate, I40E_BW_MBPS_DIVISOR); ret = i40e_set_bw_limit(vsi, vsi->seid, max_tx_rate); if (!ret) { u64 credits = max_tx_rate; @@ -8492,6 +8753,11 @@ static int i40e_configure_clsflower(struct i40e_vsi *vsi, return -EOPNOTSUPP; } + if (!tc) { + dev_err(&pf->pdev->dev, "Unable to add filter because of invalid destination"); + return -EINVAL; + } + if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) || test_bit(__I40E_RESET_INTR_RECEIVED, pf->state)) return -EBUSY; @@ -8717,6 +8983,27 @@ int i40e_open(struct net_device *netdev) } /** + * i40e_netif_set_realnum_tx_rx_queues - Update number of tx/rx queues + * @vsi: vsi structure + * + * This updates netdev's number of tx/rx queues + * + * Returns status of setting tx/rx queues + **/ +static int i40e_netif_set_realnum_tx_rx_queues(struct i40e_vsi *vsi) +{ + int ret; + + ret = netif_set_real_num_rx_queues(vsi->netdev, + vsi->num_queue_pairs); + if (ret) + return ret; + + return netif_set_real_num_tx_queues(vsi->netdev, + vsi->num_queue_pairs); +} + +/** * i40e_vsi_open - * @vsi: the VSI to open * @@ -8752,13 +9039,7 @@ int i40e_vsi_open(struct i40e_vsi *vsi) goto err_setup_rx; /* Notify the stack of the actual queue counts. */ - err = netif_set_real_num_tx_queues(vsi->netdev, - vsi->num_queue_pairs); - if (err) - goto err_set_queues; - - err = netif_set_real_num_rx_queues(vsi->netdev, - vsi->num_queue_pairs); + err = i40e_netif_set_realnum_tx_rx_queues(vsi); if (err) goto err_set_queues; @@ -10429,7 +10710,7 @@ static void i40e_send_version(struct i40e_pf *pf) dv.minor_version = 0xff; dv.build_version = 0xff; dv.subbuild_version = 0; - strlcpy(dv.driver_string, UTS_RELEASE, sizeof(dv.driver_string)); + strscpy(dv.driver_string, UTS_RELEASE, sizeof(dv.driver_string)); i40e_aq_send_driver_version(&pf->hw, &dv, NULL); } @@ -10507,7 +10788,7 @@ static int i40e_reset(struct i40e_pf *pf) **/ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) { - int old_recovery_mode_bit = test_bit(__I40E_RECOVERY_MODE, pf->state); + const bool is_recovery_mode_reported = i40e_check_recovery_mode(pf); struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi]; struct i40e_hw *hw = &pf->hw; i40e_status ret; @@ -10515,13 +10796,11 @@ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) int v; if (test_bit(__I40E_EMP_RESET_INTR_RECEIVED, pf->state) && - i40e_check_recovery_mode(pf)) { + is_recovery_mode_reported) i40e_set_ethtool_ops(pf->vsi[pf->lan_vsi]->netdev); - } if (test_bit(__I40E_DOWN, pf->state) && - !test_bit(__I40E_RECOVERY_MODE, pf->state) && - !old_recovery_mode_bit) + !test_bit(__I40E_RECOVERY_MODE, pf->state)) goto clear_recovery; dev_dbg(&pf->pdev->dev, "Rebuilding internal switch\n"); @@ -10535,15 +10814,9 @@ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) } i40e_get_oem_version(&pf->hw); - if (test_bit(__I40E_EMP_RESET_INTR_RECEIVED, pf->state) && - ((hw->aq.fw_maj_ver == 4 && hw->aq.fw_min_ver <= 33) || - hw->aq.fw_maj_ver < 4) && hw->mac.type == I40E_MAC_XL710) { - /* The following delay is necessary for 4.33 firmware and older - * to recover after EMP reset. 200 ms should suffice but we - * put here 300 ms to be sure that FW is ready to operate - * after reset. - */ - mdelay(300); + if (test_and_clear_bit(__I40E_EMP_RESET_INTR_RECEIVED, pf->state)) { + /* The following delay is necessary for firmware update. */ + mdelay(1000); } /* re-verify the eeprom if we just had an EMP reset */ @@ -10554,13 +10827,12 @@ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) * accordingly with regard to resources initialization * and deinitialization */ - if (test_bit(__I40E_RECOVERY_MODE, pf->state) || - old_recovery_mode_bit) { + if (test_bit(__I40E_RECOVERY_MODE, pf->state)) { if (i40e_get_capabilities(pf, i40e_aqc_opc_list_func_capabilities)) goto end_unlock; - if (test_bit(__I40E_RECOVERY_MODE, pf->state)) { + if (is_recovery_mode_reported) { /* we're staying in recovery mode so we'll reinitialize * misc vector here */ @@ -10610,7 +10882,7 @@ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) * unless I40E_FLAG_TC_MQPRIO was enabled or DCB * is not supported with new link speed */ - if (pf->flags & I40E_FLAG_TC_MQPRIO) { + if (i40e_is_tc_mqprio_enabled(pf)) { i40e_aq_set_dcb_parameters(hw, false, NULL); } else { if (I40E_IS_X710TL_DEVICE(hw->device_id) && @@ -10705,10 +10977,10 @@ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) } if (vsi->mqprio_qopt.max_rate[0]) { - u64 max_tx_rate = vsi->mqprio_qopt.max_rate[0]; + u64 max_tx_rate = i40e_bw_bytes_to_mbits(vsi, + vsi->mqprio_qopt.max_rate[0]); u64 credits = 0; - do_div(max_tx_rate, I40E_BW_MBPS_DIVISOR); ret = i40e_set_bw_limit(vsi, vsi->seid, max_tx_rate); if (ret) goto end_unlock; @@ -10814,6 +11086,9 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired) { int ret; + + if (test_bit(__I40E_IN_REMOVE, pf->state)) + return; /* Now we wait for GRST to settle out. * We don't have to delete the VEBs or VSIs from the hw switch * because the reset will make them disappear. @@ -11659,8 +11934,7 @@ static int i40e_vsi_alloc_q_vector(struct i40e_vsi *vsi, int v_idx) cpumask_copy(&q_vector->affinity_mask, cpu_possible_mask); if (vsi->netdev) - netif_napi_add(vsi->netdev, &q_vector->napi, - i40e_napi_poll, NAPI_POLL_WEIGHT); + netif_napi_add(vsi->netdev, &q_vector->napi, i40e_napi_poll); /* tie q_vector and vsi together */ vsi->q_vectors[v_idx] = q_vector; @@ -11753,7 +12027,6 @@ static int i40e_init_interrupt_scheme(struct i40e_pf *pf) return -ENOMEM; pf->irq_pile->num_entries = vectors; - pf->irq_pile->search_hint = 0; /* track first vector for misc interrupts, ignore return */ (void)i40e_get_lump(pf, pf->irq_pile, 1, I40E_PILE_VALID_BIT - 1); @@ -12174,6 +12447,8 @@ int i40e_reconfig_rss_queues(struct i40e_pf *pf, int queue_count) vsi->req_queue_pairs = queue_count; i40e_prep_for_reset(pf); + if (test_bit(__I40E_IN_REMOVE, pf->state)) + return pf->alloc_rss_size; pf->alloc_rss_size = new_rss_size; @@ -12556,7 +12831,6 @@ static int i40e_sw_init(struct i40e_pf *pf) goto sw_init_done; } pf->qp_pile->num_entries = pf->hw.func_caps.num_tx_qp; - pf->qp_pile->search_hint = 0; pf->tx_timeout_recovery_level = 1; @@ -12670,7 +12944,8 @@ static int i40e_set_features(struct net_device *netdev, else i40e_vlan_stripping_disable(vsi); - if (!(features & NETIF_F_HW_TC) && pf->num_cloud_filters) { + if (!(features & NETIF_F_HW_TC) && + (netdev->features & NETIF_F_HW_TC) && pf->num_cloud_filters) { dev_err(&pf->pdev->dev, "Offloaded tc filters active, can't turn hw_tc_offload off"); return -EINVAL; @@ -12965,7 +13240,7 @@ static netdev_features_t i40e_features_check(struct sk_buff *skb, } /* No need to validate L4LEN as TCP is the only protocol with a - * a flexible value and we support all possible values supported + * flexible value and we support all possible values supported * by TCP, which is at most 15 dwords */ @@ -13001,6 +13276,10 @@ static int i40e_xdp_setup(struct i40e_vsi *vsi, struct bpf_prog *prog, if (need_reset) i40e_prep_for_reset(pf); + /* VSI shall be deleted in a moment, just return EINVAL */ + if (test_bit(__I40E_IN_REMOVE, pf->state)) + return -EINVAL; + old_prog = xchg(&vsi->xdp_prog, prog); if (need_reset) { @@ -13010,6 +13289,14 @@ static int i40e_xdp_setup(struct i40e_vsi *vsi, struct bpf_prog *prog, i40e_reset_and_rebuild(pf, true, true); } + if (!i40e_enabled_xdp_vsi(vsi) && prog) { + if (i40e_realloc_rx_bi_zc(vsi, true)) + return -ENOMEM; + } else if (i40e_enabled_xdp_vsi(vsi) && !prog) { + if (i40e_realloc_rx_bi_zc(vsi, false)) + return -ENOMEM; + } + for (i = 0; i < vsi->num_queue_pairs; i++) WRITE_ONCE(vsi->rx_rings[i]->xdp_prog, vsi->xdp_prog); @@ -13242,6 +13529,7 @@ int i40e_queue_pair_disable(struct i40e_vsi *vsi, int queue_pair) i40e_queue_pair_disable_irq(vsi, queue_pair); err = i40e_queue_pair_toggle_rings(vsi, queue_pair, false /* off */); + i40e_clean_rx_ring(vsi->rx_rings[queue_pair]); i40e_queue_pair_toggle_napi(vsi, queue_pair, false /* off */); i40e_queue_pair_clean_rings(vsi, queue_pair); i40e_queue_pair_reset_stats(vsi, queue_pair); @@ -13375,8 +13663,7 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) np->vsi = vsi; hw_enc_features = NETIF_F_SG | - NETIF_F_IP_CSUM | - NETIF_F_IPV6_CSUM | + NETIF_F_HW_CSUM | NETIF_F_HIGHDMA | NETIF_F_SOFT_FEATURES | NETIF_F_TSO | @@ -13407,6 +13694,23 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) /* record features VLANs can make use of */ netdev->vlan_features |= hw_enc_features | NETIF_F_TSO_MANGLEID; +#define I40E_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE | \ + NETIF_F_GSO_GRE_CSUM | \ + NETIF_F_GSO_IPXIP4 | \ + NETIF_F_GSO_IPXIP6 | \ + NETIF_F_GSO_UDP_TUNNEL | \ + NETIF_F_GSO_UDP_TUNNEL_CSUM) + + netdev->gso_partial_features = I40E_GSO_PARTIAL_FEATURES; + netdev->features |= NETIF_F_GSO_PARTIAL | + I40E_GSO_PARTIAL_FEATURES; + + netdev->mpls_features |= NETIF_F_SG; + netdev->mpls_features |= NETIF_F_HW_CSUM; + netdev->mpls_features |= NETIF_F_TSO; + netdev->mpls_features |= NETIF_F_TSO6; + netdev->mpls_features |= I40E_GSO_PARTIAL_FEATURES; + /* enable macvlan offloads */ netdev->hw_features |= NETIF_F_HW_L2FW_DOFFLOAD; @@ -13422,6 +13726,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) netdev->features |= hw_features | NETIF_F_HW_VLAN_CTAG_FILTER; netdev->hw_enc_features |= NETIF_F_TSO_MANGLEID; + netdev->features &= ~NETIF_F_HW_TC; + if (vsi->type == I40E_VSI_MAIN) { SET_NETDEV_DEV(netdev, &pf->pdev->dev); ether_addr_copy(mac_addr, hw->mac.perm_addr); @@ -14151,6 +14457,9 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type, ret = i40e_config_netdev(vsi); if (ret) goto err_netdev; + ret = i40e_netif_set_realnum_tx_rx_queues(vsi); + if (ret) + goto err_netdev; ret = register_netdev(vsi->netdev); if (ret) goto err_netdev; @@ -15282,12 +15591,9 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* set up for high or low dma */ err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); if (err) { - err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32)); - if (err) { - dev_err(&pdev->dev, - "DMA configuration failed: 0x%x\n", err); - goto err_dma; - } + dev_err(&pdev->dev, + "DMA configuration failed: 0x%x\n", err); + goto err_dma; } /* set up pci connections */ @@ -15451,8 +15757,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (hw->aq.api_maj_ver == I40E_FW_API_VERSION_MAJOR && hw->aq.api_min_ver > I40E_FW_MINOR_VERSION(hw)) - dev_info(&pdev->dev, - "The driver for the device detected a newer version of the NVM image v%u.%u than expected v%u.%u. Please install the most recent version of the network driver.\n", + dev_dbg(&pdev->dev, + "The driver for the device detected a newer version of the NVM image v%u.%u than v%u.%u.\n", hw->aq.api_maj_ver, hw->aq.api_min_ver, I40E_FW_API_VERSION_MAJOR, @@ -15760,23 +16066,23 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) switch (hw->bus.speed) { case i40e_bus_speed_8000: - strlcpy(speed, "8.0", PCI_SPEED_SIZE); break; + strscpy(speed, "8.0", PCI_SPEED_SIZE); break; case i40e_bus_speed_5000: - strlcpy(speed, "5.0", PCI_SPEED_SIZE); break; + strscpy(speed, "5.0", PCI_SPEED_SIZE); break; case i40e_bus_speed_2500: - strlcpy(speed, "2.5", PCI_SPEED_SIZE); break; + strscpy(speed, "2.5", PCI_SPEED_SIZE); break; default: break; } switch (hw->bus.width) { case i40e_bus_width_pcie_x8: - strlcpy(width, "8", PCI_WIDTH_SIZE); break; + strscpy(width, "8", PCI_WIDTH_SIZE); break; case i40e_bus_width_pcie_x4: - strlcpy(width, "4", PCI_WIDTH_SIZE); break; + strscpy(width, "4", PCI_WIDTH_SIZE); break; case i40e_bus_width_pcie_x2: - strlcpy(width, "2", PCI_WIDTH_SIZE); break; + strscpy(width, "2", PCI_WIDTH_SIZE); break; case i40e_bus_width_pcie_x1: - strlcpy(width, "1", PCI_WIDTH_SIZE); break; + strscpy(width, "1", PCI_WIDTH_SIZE); break; default: break; } @@ -15888,8 +16194,13 @@ static void i40e_remove(struct pci_dev *pdev) i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), 0); i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), 0); - while (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state)) + /* Grab __I40E_RESET_RECOVERY_PENDING and set __I40E_IN_REMOVE + * flags, once they are set, i40e_rebuild should not be called as + * i40e_prep_for_reset always returns early. + */ + while (test_and_set_bit(__I40E_RESET_RECOVERY_PENDING, pf->state)) usleep_range(1000, 2000); + set_bit(__I40E_IN_REMOVE, pf->state); if (pf->flags & I40E_FLAG_SRIOV_ENABLED) { set_bit(__I40E_VF_RESETS_DISABLED, pf->state); @@ -16088,6 +16399,9 @@ static void i40e_pci_error_reset_done(struct pci_dev *pdev) { struct i40e_pf *pf = pci_get_drvdata(pdev); + if (test_bit(__I40E_IN_REMOVE, pf->state)) + return; + i40e_reset_and_rebuild(pf, false, false); } diff --git a/drivers/net/ethernet/intel/i40e/i40e_nvm.c b/drivers/net/ethernet/intel/i40e/i40e_nvm.c index fe6dca846028..3a38bf8bcde7 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_nvm.c +++ b/drivers/net/ethernet/intel/i40e/i40e_nvm.c @@ -682,10 +682,11 @@ i40e_status i40e_update_nvm_checksum(struct i40e_hw *hw) __le16 le_sum; ret_code = i40e_calc_nvm_checksum(hw, &checksum); - le_sum = cpu_to_le16(checksum); - if (!ret_code) + if (!ret_code) { + le_sum = cpu_to_le16(checksum); ret_code = i40e_write_nvm_aq(hw, 0x00, I40E_SR_SW_CHECKSUM_WORD, 1, &le_sum, true); + } return ret_code; } diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h index aaea297640e0..ebdcde6f1aeb 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h +++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h @@ -22,11 +22,30 @@ void i40e_adminq_init_ring_data(struct i40e_hw *hw); i40e_status i40e_clean_arq_element(struct i40e_hw *hw, struct i40e_arq_event_info *e, u16 *events_pending); -i40e_status i40e_asq_send_command(struct i40e_hw *hw, +i40e_status +i40e_asq_send_command(struct i40e_hw *hw, struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ u16 buff_size, + struct i40e_asq_cmd_details *cmd_details); +i40e_status +i40e_asq_send_command_v2(struct i40e_hw *hw, + struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ + u16 buff_size, + struct i40e_asq_cmd_details *cmd_details, + enum i40e_admin_queue_err *aq_status); +i40e_status +i40e_asq_send_command_atomic(struct i40e_hw *hw, struct i40e_aq_desc *desc, + void *buff, /* can be NULL */ u16 buff_size, + struct i40e_asq_cmd_details *cmd_details, + bool is_atomic_context); +i40e_status +i40e_asq_send_command_atomic_v2(struct i40e_hw *hw, struct i40e_aq_desc *desc, void *buff, /* can be NULL */ u16 buff_size, - struct i40e_asq_cmd_details *cmd_details); + struct i40e_asq_cmd_details *cmd_details, + bool is_atomic_context, + enum i40e_admin_queue_err *aq_status); /* debug function for adminq */ void i40e_debug_aq(struct i40e_hw *hw, enum i40e_debug_mask mask, @@ -146,9 +165,19 @@ i40e_status i40e_aq_get_veb_parameters(struct i40e_hw *hw, i40e_status i40e_aq_add_macvlan(struct i40e_hw *hw, u16 vsi_id, struct i40e_aqc_add_macvlan_element_data *mv_list, u16 count, struct i40e_asq_cmd_details *cmd_details); +i40e_status +i40e_aq_add_macvlan_v2(struct i40e_hw *hw, u16 seid, + struct i40e_aqc_add_macvlan_element_data *mv_list, + u16 count, struct i40e_asq_cmd_details *cmd_details, + enum i40e_admin_queue_err *aq_status); i40e_status i40e_aq_remove_macvlan(struct i40e_hw *hw, u16 vsi_id, struct i40e_aqc_remove_macvlan_element_data *mv_list, u16 count, struct i40e_asq_cmd_details *cmd_details); +i40e_status +i40e_aq_remove_macvlan_v2(struct i40e_hw *hw, u16 seid, + struct i40e_aqc_remove_macvlan_element_data *mv_list, + u16 count, struct i40e_asq_cmd_details *cmd_details, + enum i40e_admin_queue_err *aq_status); i40e_status i40e_aq_add_mirrorrule(struct i40e_hw *hw, u16 sw_seid, u16 rule_type, u16 dest_vsi, u16 count, __le16 *mr_list, struct i40e_asq_cmd_details *cmd_details, diff --git a/drivers/net/ethernet/intel/i40e/i40e_ptp.c b/drivers/net/ethernet/intel/i40e/i40e_ptp.c index 09b1d5aed1c9..ffea0c9c82f1 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ptp.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ptp.c @@ -27,7 +27,6 @@ #define I40E_PRTTSYN_CTL1_TSYNTYPE_V2 (2 << \ I40E_PRTTSYN_CTL1_TSYNTYPE_SHIFT) #define I40E_SUBDEV_ID_25G_PTP_PIN 0xB -#define to_dev(obj) container_of(obj, struct device, kobj) enum i40e_ptp_pin { SDP3_2 = 0, @@ -335,44 +334,37 @@ static void i40e_ptp_convert_to_hwtstamp(struct skb_shared_hwtstamps *hwtstamps, } /** - * i40e_ptp_adjfreq - Adjust the PHC frequency + * i40e_ptp_adjfine - Adjust the PHC frequency * @ptp: The PTP clock structure - * @ppb: Parts per billion adjustment from the base + * @scaled_ppm: Scaled parts per million adjustment from base * - * Adjust the frequency of the PHC by the indicated parts per billion from the - * base frequency. + * Adjust the frequency of the PHC by the indicated delta from the base + * frequency. + * + * Scaled parts per million is ppm with a 16 bit binary fractional field. **/ -static int i40e_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb) +static int i40e_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm) { struct i40e_pf *pf = container_of(ptp, struct i40e_pf, ptp_caps); struct i40e_hw *hw = &pf->hw; u64 adj, freq, diff; int neg_adj = 0; - if (ppb < 0) { + if (scaled_ppm < 0) { neg_adj = 1; - ppb = -ppb; + scaled_ppm = -scaled_ppm; } - freq = I40E_PTP_40GB_INCVAL; - freq *= ppb; - diff = div_u64(freq, 1000000000ULL); + smp_mb(); /* Force any pending update before accessing. */ + freq = I40E_PTP_40GB_INCVAL * READ_ONCE(pf->ptp_adj_mult); + diff = mul_u64_u64_div_u64(freq, (u64)scaled_ppm, + 1000000ULL << 16); if (neg_adj) adj = I40E_PTP_40GB_INCVAL - diff; else adj = I40E_PTP_40GB_INCVAL + diff; - /* At some link speeds, the base incval is so large that directly - * multiplying by ppb would result in arithmetic overflow even when - * using a u64. Avoid this by instead calculating the new incval - * always in terms of the 40GbE clock rate and then multiplying by the - * link speed factor afterwards. This does result in slightly lower - * precision at lower link speeds, but it is fairly minor. - */ - smp_mb(); /* Force any pending update before accessing. */ - adj *= READ_ONCE(pf->ptp_adj_mult); - wr32(hw, I40E_PRTTSYN_INC_L, adj & 0xFFFFFFFF); wr32(hw, I40E_PRTTSYN_INC_H, adj >> 32); @@ -1205,10 +1197,6 @@ static int i40e_ptp_set_timestamp_mode(struct i40e_pf *pf, INIT_WORK(&pf->ptp_extts0_work, i40e_ptp_extts0_work); - /* Reserved for future extensions. */ - if (config->flags) - return -EINVAL; - switch (config->tx_type) { case HWTSTAMP_TX_OFF: pf->ptp_tx = false; @@ -1402,11 +1390,11 @@ static long i40e_ptp_create_clock(struct i40e_pf *pf) if (!IS_ERR_OR_NULL(pf->ptp_clock)) return 0; - strlcpy(pf->ptp_caps.name, i40e_driver_name, + strscpy(pf->ptp_caps.name, i40e_driver_name, sizeof(pf->ptp_caps.name) - 1); pf->ptp_caps.owner = THIS_MODULE; pf->ptp_caps.max_adj = 999999999; - pf->ptp_caps.adjfreq = i40e_ptp_adjfreq; + pf->ptp_caps.adjfine = i40e_ptp_adjfine; pf->ptp_caps.adjtime = i40e_ptp_adjtime; pf->ptp_caps.gettimex64 = i40e_ptp_gettimex; pf->ptp_caps.settime64 = i40e_ptp_settime; diff --git a/drivers/net/ethernet/intel/i40e/i40e_register.h b/drivers/net/ethernet/intel/i40e/i40e_register.h index 8d0588a27a05..7339003aa17c 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_register.h +++ b/drivers/net/ethernet/intel/i40e/i40e_register.h @@ -211,6 +211,11 @@ #define I40E_GLGEN_MSRWD_MDIWRDATA_SHIFT 0 #define I40E_GLGEN_MSRWD_MDIRDDATA_SHIFT 16 #define I40E_GLGEN_MSRWD_MDIRDDATA_MASK I40E_MASK(0xFFFF, I40E_GLGEN_MSRWD_MDIRDDATA_SHIFT) +#define I40E_GLGEN_PCIFCNCNT 0x001C0AB4 /* Reset: PCIR */ +#define I40E_GLGEN_PCIFCNCNT_PCIPFCNT_SHIFT 0 +#define I40E_GLGEN_PCIFCNCNT_PCIPFCNT_MASK I40E_MASK(0x1F, I40E_GLGEN_PCIFCNCNT_PCIPFCNT_SHIFT) +#define I40E_GLGEN_PCIFCNCNT_PCIVFCNT_SHIFT 16 +#define I40E_GLGEN_PCIFCNCNT_PCIVFCNT_MASK I40E_MASK(0xFF, I40E_GLGEN_PCIFCNCNT_PCIVFCNT_SHIFT) #define I40E_GLGEN_RSTAT 0x000B8188 /* Reset: POR */ #define I40E_GLGEN_RSTAT_DEVSTATE_SHIFT 0 #define I40E_GLGEN_RSTAT_DEVSTATE_MASK I40E_MASK(0x3, I40E_GLGEN_RSTAT_DEVSTATE_SHIFT) @@ -413,6 +418,9 @@ #define I40E_VFINT_DYN_CTLN(_INTVF) (0x00024800 + ((_INTVF) * 4)) /* _i=0...511 */ /* Reset: VFR */ #define I40E_VFINT_DYN_CTLN_CLEARPBA_SHIFT 1 #define I40E_VFINT_DYN_CTLN_CLEARPBA_MASK I40E_MASK(0x1, I40E_VFINT_DYN_CTLN_CLEARPBA_SHIFT) +#define I40E_VFINT_ICR0_ADMINQ_SHIFT 30 +#define I40E_VFINT_ICR0_ADMINQ_MASK I40E_MASK(0x1, I40E_VFINT_ICR0_ADMINQ_SHIFT) +#define I40E_VFINT_ICR0_ENA(_VF) (0x0002C000 + ((_VF) * 4)) /* _i=0...127 */ /* Reset: CORER */ #define I40E_VPINT_AEQCTL(_VF) (0x0002B800 + ((_VF) * 4)) /* _i=0...127 */ /* Reset: CORER */ #define I40E_VPINT_AEQCTL_MSIX_INDX_SHIFT 0 #define I40E_VPINT_AEQCTL_ITR_INDX_SHIFT 11 @@ -640,6 +648,14 @@ #define I40E_VFQF_HKEY1_MAX_INDEX 12 #define I40E_VFQF_HLUT1(_i, _VF) (0x00220000 + ((_i) * 1024 + (_VF) * 4)) /* _i=0...15, _VF=0...127 */ /* Reset: CORER */ #define I40E_VFQF_HLUT1_MAX_INDEX 15 +#define I40E_GL_RXERR1H(_i) (0x00318004 + ((_i) * 8)) /* _i=0...143 */ /* Reset: CORER */ +#define I40E_GL_RXERR1H_MAX_INDEX 143 +#define I40E_GL_RXERR1H_RXERR1H_SHIFT 0 +#define I40E_GL_RXERR1H_RXERR1H_MASK I40E_MASK(0xFFFFFFFF, I40E_GL_RXERR1H_RXERR1H_SHIFT) +#define I40E_GL_RXERR1L(_i) (0x00318000 + ((_i) * 8)) /* _i=0...143 */ /* Reset: CORER */ +#define I40E_GL_RXERR1L_MAX_INDEX 143 +#define I40E_GL_RXERR1L_RXERR1L_SHIFT 0 +#define I40E_GL_RXERR1L_RXERR1L_MASK I40E_MASK(0xFFFFFFFF, I40E_GL_RXERR1L_RXERR1L_SHIFT) #define I40E_GLPRT_BPRCH(_i) (0x003005E4 + ((_i) * 8)) /* _i=0...3 */ /* Reset: CORER */ #define I40E_GLPRT_BPRCL(_i) (0x003005E0 + ((_i) * 8)) /* _i=0...3 */ /* Reset: CORER */ #define I40E_GLPRT_BPTCH(_i) (0x00300A04 + ((_i) * 8)) /* _i=0...3 */ /* Reset: CORER */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_status.h b/drivers/net/ethernet/intel/i40e/i40e_status.h index 77be0702d07c..db3714a65dc7 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_status.h +++ b/drivers/net/ethernet/intel/i40e/i40e_status.h @@ -18,7 +18,7 @@ enum i40e_status_code { I40E_ERR_ADAPTER_STOPPED = -9, I40E_ERR_INVALID_MAC_ADDR = -10, I40E_ERR_DEVICE_NOT_SUPPORTED = -11, - I40E_ERR_MASTER_REQUESTS_PENDING = -12, + I40E_ERR_PRIMARY_REQUESTS_PENDING = -12, I40E_ERR_INVALID_LINK_SETTINGS = -13, I40E_ERR_AUTONEG_NOT_COMPLETE = -14, I40E_ERR_RESET_FAILED = -15, diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index 10a83e5385c7..b97c95f89fa0 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -3,6 +3,7 @@ #include <linux/prefetch.h> #include <linux/bpf_trace.h> +#include <net/mpls.h> #include <net/xdp.h> #include "i40e.h" #include "i40e_trace.h" @@ -371,7 +372,6 @@ static void i40e_change_filter_num(bool ipv4, bool add, u16 *ipv4_filter_num, } } -#define IP_HEADER_OFFSET 14 #define I40E_UDPIP_DUMMY_PACKET_LEN 42 #define I40E_UDPIP6_DUMMY_PACKET_LEN 62 /** @@ -830,8 +830,6 @@ void i40e_free_tx_resources(struct i40e_ring *tx_ring) i40e_clean_tx_ring(tx_ring); kfree(tx_ring->tx_bi); tx_ring->tx_bi = NULL; - kfree(tx_ring->xsk_descs); - tx_ring->xsk_descs = NULL; if (tx_ring->desc) { dma_free_coherent(tx_ring->dev, tx_ring->size, @@ -1382,8 +1380,6 @@ static void i40e_reuse_rx_page(struct i40e_ring *rx_ring, new_buff->page_offset = old_buff->page_offset; new_buff->pagecnt_bias = old_buff->pagecnt_bias; - rx_ring->rx_stats.page_reuse_count++; - /* clear contents of buffer_info */ old_buff->page = NULL; } @@ -1433,13 +1429,6 @@ int i40e_setup_tx_descriptors(struct i40e_ring *tx_ring) if (!tx_ring->tx_bi) goto err; - if (ring_is_xdp(tx_ring)) { - tx_ring->xsk_descs = kcalloc(I40E_MAX_NUM_DESCRIPTORS, sizeof(*tx_ring->xsk_descs), - GFP_KERNEL); - if (!tx_ring->xsk_descs) - goto err; - } - u64_stats_init(&tx_ring->syncp); /* round up to nearest 4K */ @@ -1463,21 +1452,11 @@ int i40e_setup_tx_descriptors(struct i40e_ring *tx_ring) return 0; err: - kfree(tx_ring->xsk_descs); - tx_ring->xsk_descs = NULL; kfree(tx_ring->tx_bi); tx_ring->tx_bi = NULL; return -ENOMEM; } -int i40e_alloc_rx_bi(struct i40e_ring *rx_ring) -{ - unsigned long sz = sizeof(*rx_ring->rx_bi) * rx_ring->count; - - rx_ring->rx_bi = kzalloc(sz, GFP_KERNEL); - return rx_ring->rx_bi ? 0 : -ENOMEM; -} - static void i40e_clear_rx_bi(struct i40e_ring *rx_ring) { memset(rx_ring->rx_bi, 0, sizeof(*rx_ring->rx_bi) * rx_ring->count); @@ -1495,10 +1474,8 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring) if (!rx_ring->rx_bi) return; - if (rx_ring->skb) { - dev_kfree_skb(rx_ring->skb); - rx_ring->skb = NULL; - } + dev_kfree_skb(rx_ring->skb); + rx_ring->skb = NULL; if (rx_ring->xsk_pool) { i40e_xsk_clean_rx_ring(rx_ring); @@ -1608,6 +1585,11 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring) rx_ring->xdp_prog = rx_ring->vsi->xdp_prog; + rx_ring->rx_bi = + kcalloc(rx_ring->count, sizeof(*rx_ring->rx_bi), GFP_KERNEL); + if (!rx_ring->rx_bi) + return -ENOMEM; + return 0; } @@ -1675,6 +1657,8 @@ static bool i40e_alloc_mapped_page(struct i40e_ring *rx_ring, return false; } + rx_ring->rx_stats.page_alloc_count++; + /* map page for use */ dma = dma_map_page_attrs(rx_ring->dev, page, 0, i40e_rx_pg_size(rx_ring), @@ -1982,32 +1966,43 @@ static bool i40e_cleanup_headers(struct i40e_ring *rx_ring, struct sk_buff *skb, /** * i40e_can_reuse_rx_page - Determine if page can be reused for another Rx * @rx_buffer: buffer containing the page + * @rx_stats: rx stats structure for the rx ring * @rx_buffer_pgcnt: buffer page refcount pre xdp_do_redirect() call * * If page is reusable, we have a green light for calling i40e_reuse_rx_page, * which will assign the current buffer to the buffer that next_to_alloc is * pointing to; otherwise, the DMA mapping needs to be destroyed and - * page freed + * page freed. + * + * rx_stats will be updated to indicate whether the page was waived + * or busy if it could not be reused. */ static bool i40e_can_reuse_rx_page(struct i40e_rx_buffer *rx_buffer, + struct i40e_rx_queue_stats *rx_stats, int rx_buffer_pgcnt) { unsigned int pagecnt_bias = rx_buffer->pagecnt_bias; struct page *page = rx_buffer->page; /* Is any reuse possible? */ - if (!dev_page_is_reusable(page)) + if (!dev_page_is_reusable(page)) { + rx_stats->page_waive_count++; return false; + } #if (PAGE_SIZE < 8192) /* if we are only owner of page we can reuse it */ - if (unlikely((rx_buffer_pgcnt - pagecnt_bias) > 1)) + if (unlikely((rx_buffer_pgcnt - pagecnt_bias) > 1)) { + rx_stats->page_busy_count++; return false; + } #else #define I40E_LAST_OFFSET \ (SKB_WITH_OVERHEAD(PAGE_SIZE) - I40E_RXBUFFER_2048) - if (rx_buffer->page_offset > I40E_LAST_OFFSET) + if (rx_buffer->page_offset > I40E_LAST_OFFSET) { + rx_stats->page_busy_count++; return false; + } #endif /* If we have drained the page fragment pool we need to update @@ -2204,7 +2199,7 @@ static struct sk_buff *i40e_build_skb(struct i40e_ring *rx_ring, net_prefetch(xdp->data_meta); /* build an skb around the page buffer */ - skb = build_skb(xdp->data_hard_start, truesize); + skb = napi_build_skb(xdp->data_hard_start, truesize); if (unlikely(!skb)) return NULL; @@ -2237,7 +2232,7 @@ static void i40e_put_rx_buffer(struct i40e_ring *rx_ring, struct i40e_rx_buffer *rx_buffer, int rx_buffer_pgcnt) { - if (i40e_can_reuse_rx_page(rx_buffer, rx_buffer_pgcnt)) { + if (i40e_can_reuse_rx_page(rx_buffer, &rx_ring->rx_stats, rx_buffer_pgcnt)) { /* hand second half of page back to the ring */ i40e_reuse_rx_page(rx_ring, rx_buffer); } else { @@ -2290,16 +2285,14 @@ int i40e_xmit_xdp_tx_ring(struct xdp_buff *xdp, struct i40e_ring *xdp_ring) * i40e_run_xdp - run an XDP program * @rx_ring: Rx ring being processed * @xdp: XDP buffer containing the frame + * @xdp_prog: XDP program to run **/ -static int i40e_run_xdp(struct i40e_ring *rx_ring, struct xdp_buff *xdp) +static int i40e_run_xdp(struct i40e_ring *rx_ring, struct xdp_buff *xdp, struct bpf_prog *xdp_prog) { int err, result = I40E_XDP_PASS; struct i40e_ring *xdp_ring; - struct bpf_prog *xdp_prog; u32 act; - xdp_prog = READ_ONCE(rx_ring->xdp_prog); - if (!xdp_prog) goto xdp_out; @@ -2322,7 +2315,7 @@ static int i40e_run_xdp(struct i40e_ring *rx_ring, struct xdp_buff *xdp) result = I40E_XDP_REDIR; break; default: - bpf_warn_invalid_xdp_action(act); + bpf_warn_invalid_xdp_action(rx_ring->netdev, xdp_prog, act); fallthrough; case XDP_ABORTED: out_failure: @@ -2444,6 +2437,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget) unsigned int offset = rx_ring->rx_offset; struct sk_buff *skb = rx_ring->skb; unsigned int xdp_xmit = 0; + struct bpf_prog *xdp_prog; bool failure = false; struct xdp_buff xdp; int xdp_res = 0; @@ -2453,6 +2447,8 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget) #endif xdp_init_buff(&xdp, frame_sz, &rx_ring->xdp_rxq); + xdp_prog = READ_ONCE(rx_ring->xdp_prog); + while (likely(total_rx_packets < (unsigned int)budget)) { struct i40e_rx_buffer *rx_buffer; union i40e_rx_desc *rx_desc; @@ -2508,11 +2504,12 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget) hard_start = page_address(rx_buffer->page) + rx_buffer->page_offset - offset; xdp_prepare_buff(&xdp, hard_start, offset, size, true); + xdp_buff_clear_frags_flag(&xdp); #if (PAGE_SIZE > 4096) /* At larger PAGE_SIZE, frame_sz depend on len size */ xdp.frame_sz = i40e_rx_frame_truesize(rx_ring, size); #endif - xdp_res = i40e_run_xdp(rx_ring, &xdp); + xdp_res = i40e_run_xdp(rx_ring, &xdp, xdp_prog); } if (xdp_res) { @@ -3015,6 +3012,7 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len, { struct sk_buff *skb = first->skb; u64 cd_cmd, cd_tso_len, cd_mss; + __be16 protocol; union { struct iphdr *v4; struct ipv6hdr *v6; @@ -3026,7 +3024,7 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len, unsigned char *hdr; } l4; u32 paylen, l4_offset; - u16 gso_segs, gso_size; + u16 gso_size; int err; if (skb->ip_summed != CHECKSUM_PARTIAL) @@ -3039,15 +3037,23 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len, if (err < 0) return err; - ip.hdr = skb_network_header(skb); - l4.hdr = skb_transport_header(skb); + protocol = vlan_get_protocol(skb); + + if (eth_p_mpls(protocol)) + ip.hdr = skb_inner_network_header(skb); + else + ip.hdr = skb_network_header(skb); + l4.hdr = skb_checksum_start(skb); /* initialize outer IP header fields */ if (ip.v4->version == 4) { ip.v4->tot_len = 0; ip.v4->check = 0; + + first->tx_flags |= I40E_TX_FLAGS_TSO; } else { ip.v6->payload_len = 0; + first->tx_flags |= I40E_TX_FLAGS_TSO; } if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE | @@ -3100,10 +3106,9 @@ static int i40e_tso(struct i40e_tx_buffer *first, u8 *hdr_len, /* pull values out of skb_shinfo */ gso_size = skb_shinfo(skb)->gso_size; - gso_segs = skb_shinfo(skb)->gso_segs; /* update GSO size and bytecount with header size */ - first->gso_segs = gso_segs; + first->gso_segs = skb_shinfo(skb)->gso_segs; first->bytecount += (first->gso_segs - 1) * *hdr_len; /* find the field values */ @@ -3187,13 +3192,29 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags, unsigned char *exthdr; u32 offset, cmd = 0; __be16 frag_off; + __be16 protocol; u8 l4_proto = 0; if (skb->ip_summed != CHECKSUM_PARTIAL) return 0; - ip.hdr = skb_network_header(skb); - l4.hdr = skb_transport_header(skb); + protocol = vlan_get_protocol(skb); + + if (eth_p_mpls(protocol)) { + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_checksum_start(skb); + } else { + ip.hdr = skb_network_header(skb); + l4.hdr = skb_transport_header(skb); + } + + /* set the tx_flags to indicate the IP protocol type. this is + * required so that checksum header computation below is accurate. + */ + if (ip.v4->version == 4) + *tx_flags |= I40E_TX_FLAGS_IPV4; + else + *tx_flags |= I40E_TX_FLAGS_IPV6; /* compute outer L2 header size */ offset = ((ip.hdr - skb->data) / 2) << I40E_TX_DESC_LENGTH_MACLEN_SHIFT; @@ -3373,6 +3394,8 @@ int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size) /* Memory barrier before checking head and tail */ smp_mb(); + ++tx_ring->tx_stats.tx_stopped; + /* Check again in a case another CPU has just made room available. */ if (likely(I40E_DESC_UNUSED(tx_ring) < size)) return -EBUSY; @@ -3662,7 +3685,8 @@ u16 i40e_lan_select_queue(struct net_device *netdev, u8 prio; /* is DCB enabled at all? */ - if (vsi->tc_config.numtc == 1) + if (vsi->tc_config.numtc == 1 || + i40e_is_tc_mqprio_enabled(vsi->back)) return netdev_pick_tx(netdev, skb, sb_dev); prio = skb->priority; @@ -3688,35 +3712,55 @@ u16 i40e_lan_select_queue(struct net_device *netdev, static int i40e_xmit_xdp_ring(struct xdp_frame *xdpf, struct i40e_ring *xdp_ring) { - u16 i = xdp_ring->next_to_use; - struct i40e_tx_buffer *tx_bi; - struct i40e_tx_desc *tx_desc; + struct skb_shared_info *sinfo = xdp_get_shared_info_from_frame(xdpf); + u8 nr_frags = unlikely(xdp_frame_has_frags(xdpf)) ? sinfo->nr_frags : 0; + u16 i = 0, index = xdp_ring->next_to_use; + struct i40e_tx_buffer *tx_head = &xdp_ring->tx_bi[index]; + struct i40e_tx_buffer *tx_bi = tx_head; + struct i40e_tx_desc *tx_desc = I40E_TX_DESC(xdp_ring, index); void *data = xdpf->data; u32 size = xdpf->len; - dma_addr_t dma; - if (!unlikely(I40E_DESC_UNUSED(xdp_ring))) { + if (unlikely(I40E_DESC_UNUSED(xdp_ring) < 1 + nr_frags)) { xdp_ring->tx_stats.tx_busy++; return I40E_XDP_CONSUMED; } - dma = dma_map_single(xdp_ring->dev, data, size, DMA_TO_DEVICE); - if (dma_mapping_error(xdp_ring->dev, dma)) - return I40E_XDP_CONSUMED; - tx_bi = &xdp_ring->tx_bi[i]; - tx_bi->bytecount = size; - tx_bi->gso_segs = 1; - tx_bi->xdpf = xdpf; + tx_head->bytecount = xdp_get_frame_len(xdpf); + tx_head->gso_segs = 1; + tx_head->xdpf = xdpf; - /* record length, and DMA address */ - dma_unmap_len_set(tx_bi, len, size); - dma_unmap_addr_set(tx_bi, dma, dma); + for (;;) { + dma_addr_t dma; - tx_desc = I40E_TX_DESC(xdp_ring, i); - tx_desc->buffer_addr = cpu_to_le64(dma); - tx_desc->cmd_type_offset_bsz = build_ctob(I40E_TX_DESC_CMD_ICRC - | I40E_TXD_CMD, - 0, size, 0); + dma = dma_map_single(xdp_ring->dev, data, size, DMA_TO_DEVICE); + if (dma_mapping_error(xdp_ring->dev, dma)) + goto unmap; + + /* record length, and DMA address */ + dma_unmap_len_set(tx_bi, len, size); + dma_unmap_addr_set(tx_bi, dma, dma); + + tx_desc->buffer_addr = cpu_to_le64(dma); + tx_desc->cmd_type_offset_bsz = + build_ctob(I40E_TX_DESC_CMD_ICRC, 0, size, 0); + + if (++index == xdp_ring->count) + index = 0; + + if (i == nr_frags) + break; + + tx_bi = &xdp_ring->tx_bi[index]; + tx_desc = I40E_TX_DESC(xdp_ring, index); + + data = skb_frag_address(&sinfo->frags[i]); + size = skb_frag_size(&sinfo->frags[i]); + i++; + } + + tx_desc->cmd_type_offset_bsz |= + cpu_to_le64(I40E_TXD_CMD << I40E_TXD_QW1_CMD_SHIFT); /* Make certain all of the status bits have been updated * before next_to_watch is written. @@ -3724,14 +3768,30 @@ static int i40e_xmit_xdp_ring(struct xdp_frame *xdpf, smp_wmb(); xdp_ring->xdp_tx_active++; - i++; - if (i == xdp_ring->count) - i = 0; - tx_bi->next_to_watch = tx_desc; - xdp_ring->next_to_use = i; + tx_head->next_to_watch = tx_desc; + xdp_ring->next_to_use = index; return I40E_XDP_TX; + +unmap: + for (;;) { + tx_bi = &xdp_ring->tx_bi[index]; + if (dma_unmap_len(tx_bi, len)) + dma_unmap_page(xdp_ring->dev, + dma_unmap_addr(tx_bi, dma), + dma_unmap_len(tx_bi, len), + DMA_TO_DEVICE); + dma_unmap_len_set(tx_bi, len, 0); + if (tx_bi == tx_head) + break; + + if (!index) + index += xdp_ring->count; + index--; + } + + return I40E_XDP_CONSUMED; } /** @@ -3749,7 +3809,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, struct i40e_tx_buffer *first; u32 td_offset = 0; u32 tx_flags = 0; - __be16 protocol; u32 td_cmd = 0; u8 hdr_len = 0; int tso, count; @@ -3791,15 +3850,6 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, if (i40e_tx_prepare_vlan_flags(skb, tx_ring, &tx_flags)) goto out_drop; - /* obtain protocol of skb */ - protocol = vlan_get_protocol(skb); - - /* setup IPv4/IPv6 offloads */ - if (protocol == htons(ETH_P_IP)) - tx_flags |= I40E_TX_FLAGS_IPV4; - else if (protocol == htons(ETH_P_IPV6)) - tx_flags |= I40E_TX_FLAGS_IPV6; - tso = i40e_tso(first, &hdr_len, &cd_type_cmd_tso_mss); if (tso < 0) diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h index bfc2845c99d1..768290dc6f48 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h @@ -290,6 +290,7 @@ struct i40e_tx_queue_stats { u64 tx_done_old; u64 tx_linearize; u64 tx_force_wb; + u64 tx_stopped; int prev_pkt_ctr; }; @@ -298,7 +299,9 @@ struct i40e_rx_queue_stats { u64 alloc_page_failed; u64 alloc_buff_failed; u64 page_reuse_count; - u64 realloc_count; + u64 page_alloc_count; + u64 page_waive_count; + u64 page_busy_count; }; enum i40e_ring_state_t { @@ -390,7 +393,6 @@ struct i40e_ring { u16 rx_offset; struct xdp_rxq_info xdp_rxq; struct xsk_buff_pool *xsk_pool; - struct xdp_desc *xsk_descs; /* For storing descriptors in the AF_XDP ZC path */ } ____cacheline_internodealigned_in_smp; static inline bool ring_uses_build_skb(struct i40e_ring *ring) @@ -467,7 +469,6 @@ int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size); bool __i40e_chk_linearize(struct sk_buff *skb); int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, u32 flags); -int i40e_alloc_rx_bi(struct i40e_ring *rx_ring); /** * i40e_get_head - Retrieve head from head writeback diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h index 19da3b22160f..8c5118c8baaf 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h @@ -20,6 +20,7 @@ void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val); #define I40E_XDP_CONSUMED BIT(0) #define I40E_XDP_TX BIT(1) #define I40E_XDP_REDIR BIT(2) +#define I40E_XDP_EXIT BIT(3) /* * build_ctob - Builds the Tx descriptor (cmd, offset and type) qword diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h index 36a4ca1ffb1a..388c3d36d96a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_type.h +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h @@ -1172,6 +1172,7 @@ struct i40e_eth_stats { u64 tx_broadcast; /* bptc */ u64 tx_discards; /* tdpc */ u64 tx_errors; /* tepc */ + u64 rx_discards_other; /* rxerr1 */ }; /* Statistics collected per VEB per TC */ @@ -1403,6 +1404,10 @@ struct i40e_lldp_variables { #define I40E_PFQF_CTL_0_HASHLUTSIZE_512 0x00010000 /* INPUT SET MASK for RSS, flow director, and flexible payload */ +#define I40E_X722_L3_SRC_SHIFT 49 +#define I40E_X722_L3_SRC_MASK (0x3ULL << I40E_X722_L3_SRC_SHIFT) +#define I40E_X722_L3_DST_SHIFT 41 +#define I40E_X722_L3_DST_MASK (0x3ULL << I40E_X722_L3_DST_SHIFT) #define I40E_L3_SRC_SHIFT 47 #define I40E_L3_SRC_MASK (0x3ULL << I40E_L3_SRC_SHIFT) #define I40E_L3_V6_SRC_SHIFT 43 diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 2ea4deb8fc44..72ddcefc45b1 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -1377,6 +1377,32 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, } /** + * i40e_sync_vfr_reset + * @hw: pointer to hw struct + * @vf_id: VF identifier + * + * Before trigger hardware reset, we need to know if no other process has + * reserved the hardware for any reset operations. This check is done by + * examining the status of the RSTAT1 register used to signal the reset. + **/ +static int i40e_sync_vfr_reset(struct i40e_hw *hw, int vf_id) +{ + u32 reg; + int i; + + for (i = 0; i < I40E_VFR_WAIT_COUNT; i++) { + reg = rd32(hw, I40E_VFINT_ICR0_ENA(vf_id)) & + I40E_VFINT_ICR0_ADMINQ_MASK; + if (reg) + return 0; + + usleep_range(100, 200); + } + + return -EAGAIN; +} + +/** * i40e_trigger_vf_reset * @vf: pointer to the VF structure * @flr: VFLR was issued or not @@ -1390,9 +1416,11 @@ static void i40e_trigger_vf_reset(struct i40e_vf *vf, bool flr) struct i40e_pf *pf = vf->pf; struct i40e_hw *hw = &pf->hw; u32 reg, reg_idx, bit_idx; + bool vf_active; + u32 radq; /* warn the VF */ - clear_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states); + vf_active = test_and_clear_bit(I40E_VF_STATE_ACTIVE, &vf->vf_states); /* Disable VF's configuration API during reset. The flag is re-enabled * in i40e_alloc_vf_res(), when it's safe again to access VF's VSI. @@ -1406,7 +1434,19 @@ static void i40e_trigger_vf_reset(struct i40e_vf *vf, bool flr) * just need to clean up, so don't hit the VFRTRIG register. */ if (!flr) { - /* reset VF using VPGEN_VFRTRIG reg */ + /* Sync VFR reset before trigger next one */ + radq = rd32(hw, I40E_VFINT_ICR0_ENA(vf->vf_id)) & + I40E_VFINT_ICR0_ADMINQ_MASK; + if (vf_active && !radq) + /* waiting for finish reset by virtual driver */ + if (i40e_sync_vfr_reset(hw, vf->vf_id)) + dev_info(&pf->pdev->dev, + "Reset VF %d never finished\n", + vf->vf_id); + + /* Reset VF using VPGEN_VFRTRIG reg. It is also setting + * in progress state in rstat1 register. + */ reg = rd32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id)); reg |= I40E_VPGEN_VFRTRIG_VFSWR_MASK; wr32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id), reg); @@ -1496,10 +1536,12 @@ bool i40e_reset_vf(struct i40e_vf *vf, bool flr) if (test_bit(__I40E_VF_RESETS_DISABLED, pf->state)) return true; - /* If the VFs have been disabled, this means something else is - * resetting the VF, so we shouldn't continue. - */ - if (test_and_set_bit(__I40E_VF_DISABLE, pf->state)) + /* Bail out if VFs are disabled. */ + if (test_bit(__I40E_VF_DISABLE, pf->state)) + return true; + + /* If VF is being reset already we don't need to continue. */ + if (test_and_set_bit(I40E_VF_STATE_RESETTING, &vf->vf_states)) return true; i40e_trigger_vf_reset(vf, flr); @@ -1536,7 +1578,7 @@ bool i40e_reset_vf(struct i40e_vf *vf, bool flr) i40e_cleanup_reset_vf(vf); i40e_flush(hw); - clear_bit(__I40E_VF_DISABLE, pf->state); + clear_bit(I40E_VF_STATE_RESETTING, &vf->vf_states); return true; } @@ -1569,8 +1611,12 @@ bool i40e_reset_all_vfs(struct i40e_pf *pf, bool flr) return false; /* Begin reset on all VFs at once */ - for (v = 0; v < pf->num_alloc_vfs; v++) - i40e_trigger_vf_reset(&pf->vf[v], flr); + for (v = 0; v < pf->num_alloc_vfs; v++) { + vf = &pf->vf[v]; + /* If VF is being reset no need to trigger reset again */ + if (!test_bit(I40E_VF_STATE_RESETTING, &vf->vf_states)) + i40e_trigger_vf_reset(&pf->vf[v], flr); + } /* HW requires some time to make sure it can flush the FIFO for a VF * when it resets it. Poll the VPGEN_VFRSTAT register for each VF in @@ -1586,9 +1632,11 @@ bool i40e_reset_all_vfs(struct i40e_pf *pf, bool flr) */ while (v < pf->num_alloc_vfs) { vf = &pf->vf[v]; - reg = rd32(hw, I40E_VPGEN_VFRSTAT(vf->vf_id)); - if (!(reg & I40E_VPGEN_VFRSTAT_VFRD_MASK)) - break; + if (!test_bit(I40E_VF_STATE_RESETTING, &vf->vf_states)) { + reg = rd32(hw, I40E_VPGEN_VFRSTAT(vf->vf_id)); + if (!(reg & I40E_VPGEN_VFRSTAT_VFRD_MASK)) + break; + } /* If the current VF has finished resetting, move on * to the next VF in sequence. @@ -1616,6 +1664,10 @@ bool i40e_reset_all_vfs(struct i40e_pf *pf, bool flr) if (pf->vf[v].lan_vsi_idx == 0) continue; + /* If VF is reset in another thread just continue */ + if (test_bit(I40E_VF_STATE_RESETTING, &vf->vf_states)) + continue; + i40e_vsi_stop_rings_no_wait(pf->vsi[pf->vf[v].lan_vsi_idx]); } @@ -1627,6 +1679,10 @@ bool i40e_reset_all_vfs(struct i40e_pf *pf, bool flr) if (pf->vf[v].lan_vsi_idx == 0) continue; + /* If VF is reset in another thread just continue */ + if (test_bit(I40E_VF_STATE_RESETTING, &vf->vf_states)) + continue; + i40e_vsi_wait_queues_disabled(pf->vsi[pf->vf[v].lan_vsi_idx]); } @@ -1636,8 +1692,13 @@ bool i40e_reset_all_vfs(struct i40e_pf *pf, bool flr) mdelay(50); /* Finish the reset on each VF */ - for (v = 0; v < pf->num_alloc_vfs; v++) + for (v = 0; v < pf->num_alloc_vfs; v++) { + /* If VF is reset in another thread just continue */ + if (test_bit(I40E_VF_STATE_RESETTING, &vf->vf_states)) + continue; + i40e_cleanup_reset_vf(&pf->vf[v]); + } i40e_flush(hw); clear_bit(__I40E_VF_DISABLE, pf->state); @@ -1902,25 +1963,6 @@ static int i40e_vc_send_msg_to_vf(struct i40e_vf *vf, u32 v_opcode, hw = &pf->hw; abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; - /* single place to detect unsuccessful return values */ - if (v_retval) { - vf->num_invalid_msgs++; - dev_info(&pf->pdev->dev, "VF %d failed opcode %d, retval: %d\n", - vf->vf_id, v_opcode, v_retval); - if (vf->num_invalid_msgs > - I40E_DEFAULT_NUM_INVALID_MSGS_ALLOWED) { - dev_err(&pf->pdev->dev, - "Number of invalid messages exceeded for VF %d\n", - vf->vf_id); - dev_err(&pf->pdev->dev, "Use PF Control I/F to enable the VF\n"); - set_bit(I40E_VF_STATE_DISABLED, &vf->vf_states); - } - } else { - vf->num_valid_msgs++; - /* reset the invalid counter, if a valid message is received. */ - vf->num_invalid_msgs = 0; - } - aq_ret = i40e_aq_send_msg_to_vf(hw, abs_vf_id, v_opcode, v_retval, msg, msglen, NULL); if (aq_ret) { @@ -2018,6 +2060,25 @@ static void i40e_del_qch(struct i40e_vf *vf) } /** + * i40e_vc_get_max_frame_size + * @vf: pointer to the VF + * + * Max frame size is determined based on the current port's max frame size and + * whether a port VLAN is configured on this VF. The VF is not aware whether + * it's in a port VLAN so the PF needs to account for this in max frame size + * checks and sending the max frame size to the VF. + **/ +static u16 i40e_vc_get_max_frame_size(struct i40e_vf *vf) +{ + u16 max_frame_size = vf->pf->hw.phy.link_info.max_frame_size; + + if (vf->port_vlan_id) + max_frame_size -= VLAN_HLEN; + + return max_frame_size; +} + +/** * i40e_vc_get_vf_resources_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -2118,6 +2179,7 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) vfres->max_vectors = pf->hw.func_caps.num_msix_vectors_vf; vfres->rss_key_size = I40E_HKEY_ARRAY_SIZE; vfres->rss_lut_size = I40E_VF_HLUT_ARRAY_SIZE; + vfres->max_mtu = i40e_vc_get_max_frame_size(vf); if (vf->lan_vsi_idx) { vfres->vsi_res[0].vsi_id = vf->lan_vsi_id; @@ -2126,6 +2188,10 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg) /* VFs only use TC 0 */ vfres->vsi_res[0].qset_handle = le16_to_cpu(vsi->info.qs_handle[0]); + if (!(vf->driver_caps & VIRTCHNL_VF_OFFLOAD_USO) && !vf->pf_set_mac) { + i40e_del_mac_filter(vsi, vf->default_lan_addr.addr); + eth_zero_addr(vf->default_lan_addr.addr); + } ether_addr_copy(vfres->vsi_res[0].default_mac_addr, vf->default_lan_addr.addr); } @@ -2261,7 +2327,7 @@ static int i40e_vc_config_queues_msg(struct i40e_vf *vf, u8 *msg) } if (vf->adq_enabled) { - for (i = 0; i < I40E_MAX_VF_VSI; i++) + for (i = 0; i < vf->num_tc; i++) num_qps_all += vf->ch[i].num_qps; if (num_qps_all != qci->num_queue_pairs) { aq_ret = I40E_ERR_PARAM; @@ -2599,6 +2665,59 @@ error_param: } /** + * i40e_check_enough_queue - find big enough queue number + * @vf: pointer to the VF info + * @needed: the number of items needed + * + * Returns the base item index of the queue, or negative for error + **/ +static int i40e_check_enough_queue(struct i40e_vf *vf, u16 needed) +{ + unsigned int i, cur_queues, more, pool_size; + struct i40e_lump_tracking *pile; + struct i40e_pf *pf = vf->pf; + struct i40e_vsi *vsi; + + vsi = pf->vsi[vf->lan_vsi_idx]; + cur_queues = vsi->alloc_queue_pairs; + + /* if current allocated queues are enough for need */ + if (cur_queues >= needed) + return vsi->base_queue; + + pile = pf->qp_pile; + if (cur_queues > 0) { + /* if the allocated queues are not zero + * just check if there are enough queues for more + * behind the allocated queues. + */ + more = needed - cur_queues; + for (i = vsi->base_queue + cur_queues; + i < pile->num_entries; i++) { + if (pile->list[i] & I40E_PILE_VALID_BIT) + break; + + if (more-- == 1) + /* there is enough */ + return vsi->base_queue; + } + } + + pool_size = 0; + for (i = 0; i < pile->num_entries; i++) { + if (pile->list[i] & I40E_PILE_VALID_BIT) { + pool_size = 0; + continue; + } + if (needed <= ++pool_size) + /* there is enough */ + return i; + } + + return -ENOMEM; +} + +/** * i40e_vc_request_queues_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -2632,6 +2751,12 @@ static int i40e_vc_request_queues_msg(struct i40e_vf *vf, u8 *msg) req_pairs - cur_pairs, pf->queues_left); vfres->num_queue_pairs = pf->queues_left + cur_pairs; + } else if (i40e_check_enough_queue(vf, req_pairs) < 0) { + dev_warn(&pf->pdev->dev, + "VF %d requested %d more queues, but there is not enough for it.\n", + vf->vf_id, + req_pairs - cur_pairs); + vfres->num_queue_pairs = cur_pairs; } else { /* successful request */ vf->num_req_queues = req_pairs; @@ -2685,12 +2810,21 @@ error_param: (u8 *)&stats, sizeof(stats)); } +#define I40E_MAX_MACVLAN_PER_HW 3072 +#define I40E_MAX_MACVLAN_PER_PF(num_ports) (I40E_MAX_MACVLAN_PER_HW / \ + (num_ports)) /* If the VF is not trusted restrict the number of MAC/VLAN it can program * MAC filters: 16 for multicast, 1 for MAC, 1 for broadcast */ #define I40E_VC_MAX_MAC_ADDR_PER_VF (16 + 1 + 1) #define I40E_VC_MAX_VLAN_PER_VF 16 +#define I40E_VC_MAX_MACVLAN_PER_TRUSTED_VF(vf_num, num_ports) \ +({ typeof(vf_num) vf_num_ = (vf_num); \ + typeof(num_ports) num_ports_ = (num_ports); \ + ((I40E_MAX_MACVLAN_PER_PF(num_ports_) - vf_num_ * \ + I40E_VC_MAX_MAC_ADDR_PER_VF) / vf_num_) + \ + I40E_VC_MAX_MAC_ADDR_PER_VF; }) /** * i40e_check_vf_permission * @vf: pointer to the VF info @@ -2713,6 +2847,7 @@ static inline int i40e_check_vf_permission(struct i40e_vf *vf, { struct i40e_pf *pf = vf->pf; struct i40e_vsi *vsi = pf->vsi[vf->lan_vsi_idx]; + struct i40e_hw *hw = &pf->hw; int mac2add_cnt = 0; int i; @@ -2752,12 +2887,26 @@ static inline int i40e_check_vf_permission(struct i40e_vf *vf, * number of addresses. Check to make sure that the additions do not * push us over the limit. */ - if (!test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) && - (i40e_count_filters(vsi) + mac2add_cnt) > + if (!test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps)) { + if ((i40e_count_filters(vsi) + mac2add_cnt) > I40E_VC_MAX_MAC_ADDR_PER_VF) { - dev_err(&pf->pdev->dev, - "Cannot add more MAC addresses, VF is not trusted, switch the VF to trusted to add more functionality\n"); - return -EPERM; + dev_err(&pf->pdev->dev, + "Cannot add more MAC addresses, VF is not trusted, switch the VF to trusted to add more functionality\n"); + return -EPERM; + } + /* If this VF is trusted, it can use more resources than untrusted. + * However to ensure that every trusted VF has appropriate number of + * resources, divide whole pool of resources per port and then across + * all VFs. + */ + } else { + if ((i40e_count_filters(vsi) + mac2add_cnt) > + I40E_VC_MAX_MACVLAN_PER_TRUSTED_VF(pf->num_alloc_vfs, + hw->num_ports)) { + dev_err(&pf->pdev->dev, + "Cannot add more MAC addresses, trusted VF exhausted it's resources\n"); + return -EPERM; + } } return 0; } @@ -2829,8 +2978,8 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg) error_param: /* send the response to the VF */ - return i40e_vc_send_resp_to_vf(vf, VIRTCHNL_OP_ADD_ETH_ADDR, - ret); + return i40e_vc_send_msg_to_vf(vf, VIRTCHNL_OP_ADD_ETH_ADDR, + ret, NULL, 0); } /** @@ -4245,6 +4394,7 @@ int i40e_ndo_set_vf_port_vlan(struct net_device *netdev, int vf_id, /* duplicate request, so just return success */ goto error_pvid; + i40e_vlan_stripping_enable(vsi); i40e_vc_reset_vf(vf, true); /* During reset the VF got a new VSI, so refresh a pointer. */ vsi = pf->vsi[vf->lan_vsi_idx]; @@ -4260,7 +4410,7 @@ int i40e_ndo_set_vf_port_vlan(struct net_device *netdev, int vf_id, * MAC addresses deleted. */ if ((!(vlan_id || qos) || - vlanprio != le16_to_cpu(vsi->info.pvid)) && + vlanprio != le16_to_cpu(vsi->info.pvid)) && vsi->info.pvid) { ret = i40e_add_vlan_all_mac(vsi, I40E_VLAN_ANY); if (ret) { @@ -4623,6 +4773,11 @@ int i40e_ndo_set_vf_trust(struct net_device *netdev, int vf_id, bool setting) goto out; vf->trusted = setting; + + /* request PF to sync mac/vlan filters for the VF */ + set_bit(__I40E_MACVLAN_SYNC_PENDING, pf->state); + pf->vsi[vf->lan_vsi_idx]->flags |= I40E_VSI_FLAG_FILTER_CHANGED; + i40e_vc_reset_vf(vf, true); dev_info(&pf->pdev->dev, "VF %u is now %strusted\n", vf_id, setting ? "" : "un"); diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h index 49575a640a84..358bbdb58795 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h @@ -10,8 +10,6 @@ #define I40E_VIRTCHNL_SUPPORTED_QTYPES 2 -#define I40E_DEFAULT_NUM_INVALID_MSGS_ALLOWED 10 - #define I40E_VLAN_PRIORITY_SHIFT 13 #define I40E_VLAN_MASK 0xFFF #define I40E_PRIORITY_MASK 0xE000 @@ -19,6 +17,7 @@ #define I40E_MAX_VF_PROMISC_FLAGS 3 #define I40E_VF_STATE_WAIT_COUNT 20 +#define I40E_VFR_WAIT_COUNT 100 /* Various queue ctrls */ enum i40e_queue_ctrl { @@ -40,6 +39,7 @@ enum i40e_vf_states { I40E_VF_STATE_MC_PROMISC, I40E_VF_STATE_UC_PROMISC, I40E_VF_STATE_PRE_ENABLE, + I40E_VF_STATE_RESETTING }; /* VF capabilities */ @@ -91,9 +91,6 @@ struct i40e_vf { u8 num_queue_pairs; /* num of qps assigned to VF vsis */ u8 num_req_queues; /* num of requested qps */ u64 num_mdd_events; /* num of mdd events detected */ - /* num of continuous malformed or invalid msgs detected */ - u64 num_invalid_msgs; - u64 num_valid_msgs; /* num of valid msgs detected */ unsigned long vf_caps; /* vf's adv. capabilities */ unsigned long vf_states; /* vf's runtime states */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c index ea06e957393e..cd7b52fb6b46 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c @@ -10,14 +10,6 @@ #include "i40e_txrx_common.h" #include "i40e_xsk.h" -int i40e_alloc_rx_bi_zc(struct i40e_ring *rx_ring) -{ - unsigned long sz = sizeof(*rx_ring->rx_bi_zc) * rx_ring->count; - - rx_ring->rx_bi_zc = kzalloc(sz, GFP_KERNEL); - return rx_ring->rx_bi_zc ? 0 : -ENOMEM; -} - void i40e_clear_rx_bi_zc(struct i40e_ring *rx_ring) { memset(rx_ring->rx_bi_zc, 0, @@ -30,6 +22,58 @@ static struct xdp_buff **i40e_rx_bi(struct i40e_ring *rx_ring, u32 idx) } /** + * i40e_realloc_rx_xdp_bi - reallocate SW ring for either XSK or normal buffer + * @rx_ring: Current rx ring + * @pool_present: is pool for XSK present + * + * Try allocating memory and return ENOMEM, if failed to allocate. + * If allocation was successful, substitute buffer with allocated one. + * Returns 0 on success, negative on failure + */ +static int i40e_realloc_rx_xdp_bi(struct i40e_ring *rx_ring, bool pool_present) +{ + size_t elem_size = pool_present ? sizeof(*rx_ring->rx_bi_zc) : + sizeof(*rx_ring->rx_bi); + void *sw_ring = kcalloc(rx_ring->count, elem_size, GFP_KERNEL); + + if (!sw_ring) + return -ENOMEM; + + if (pool_present) { + kfree(rx_ring->rx_bi); + rx_ring->rx_bi = NULL; + rx_ring->rx_bi_zc = sw_ring; + } else { + kfree(rx_ring->rx_bi_zc); + rx_ring->rx_bi_zc = NULL; + rx_ring->rx_bi = sw_ring; + } + return 0; +} + +/** + * i40e_realloc_rx_bi_zc - reallocate rx SW rings + * @vsi: Current VSI + * @zc: is zero copy set + * + * Reallocate buffer for rx_rings that might be used by XSK. + * XDP requires more memory, than rx_buf provides. + * Returns 0 on success, negative on failure + */ +int i40e_realloc_rx_bi_zc(struct i40e_vsi *vsi, bool zc) +{ + struct i40e_ring *rx_ring; + unsigned long q; + + for_each_set_bit(q, vsi->af_xdp_zc_qps, vsi->alloc_queue_pairs) { + rx_ring = vsi->rx_rings[q]; + if (i40e_realloc_rx_xdp_bi(rx_ring, zc)) + return -ENOMEM; + } + return 0; +} + +/** * i40e_xsk_pool_enable - Enable/associate an AF_XDP buffer pool to a * certain ring/qid * @vsi: Current VSI @@ -69,6 +113,10 @@ static int i40e_xsk_pool_enable(struct i40e_vsi *vsi, if (err) return err; + err = i40e_realloc_rx_xdp_bi(vsi->rx_rings[qid], true); + if (err) + return err; + err = i40e_queue_pair_enable(vsi, qid); if (err) return err; @@ -113,6 +161,9 @@ static int i40e_xsk_pool_disable(struct i40e_vsi *vsi, u16 qid) xsk_pool_dma_unmap(pool, I40E_RX_DMA_ATTR); if (if_running) { + err = i40e_realloc_rx_xdp_bi(vsi->rx_rings[qid], false); + if (err) + return err; err = i40e_queue_pair_enable(vsi, qid); if (err) return err; @@ -143,27 +194,28 @@ int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool, * i40e_run_xdp_zc - Executes an XDP program on an xdp_buff * @rx_ring: Rx ring * @xdp: xdp_buff used as input to the XDP program + * @xdp_prog: XDP program to run * * Returns any of I40E_XDP_{PASS, CONSUMED, TX, REDIR} **/ -static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) +static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp, + struct bpf_prog *xdp_prog) { int err, result = I40E_XDP_PASS; struct i40e_ring *xdp_ring; - struct bpf_prog *xdp_prog; u32 act; - /* NB! xdp_prog will always be !NULL, due to the fact that - * this path is enabled by setting an XDP program. - */ - xdp_prog = READ_ONCE(rx_ring->xdp_prog); act = bpf_prog_run_xdp(xdp_prog, xdp); if (likely(act == XDP_REDIRECT)) { err = xdp_do_redirect(rx_ring->netdev, xdp, xdp_prog); - if (err) - goto out_failure; - return I40E_XDP_REDIR; + if (!err) + return I40E_XDP_REDIR; + if (xsk_uses_need_wakeup(rx_ring->xsk_pool) && err == -ENOBUFS) + result = I40E_XDP_EXIT; + else + result = I40E_XDP_CONSUMED; + goto out_failure; } switch (act) { @@ -175,16 +227,16 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) if (result == I40E_XDP_CONSUMED) goto out_failure; break; + case XDP_DROP: + result = I40E_XDP_CONSUMED; + break; default: - bpf_warn_invalid_xdp_action(act); + bpf_warn_invalid_xdp_action(rx_ring->netdev, xdp_prog, act); fallthrough; case XDP_ABORTED: + result = I40E_XDP_CONSUMED; out_failure: trace_xdp_exception(rx_ring->netdev, xdp_prog, act); - fallthrough; /* handle aborts by dropping packet */ - case XDP_DROP: - result = I40E_XDP_CONSUMED; - break; } return result; } @@ -218,7 +270,6 @@ bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count) ntu += nb_buffs; if (ntu == rx_ring->count) { rx_desc = I40E_RX_DESC(rx_ring, 0); - xdp = i40e_rx_bi(rx_ring, 0); ntu = 0; } @@ -241,21 +292,25 @@ bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count) static struct sk_buff *i40e_construct_skb_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) { + unsigned int totalsize = xdp->data_end - xdp->data_meta; unsigned int metasize = xdp->data - xdp->data_meta; - unsigned int datasize = xdp->data_end - xdp->data; struct sk_buff *skb; + net_prefetch(xdp->data_meta); + /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, - xdp->data_end - xdp->data_hard_start, + skb = __napi_alloc_skb(&rx_ring->q_vector->napi, totalsize, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) goto out; - skb_reserve(skb, xdp->data - xdp->data_hard_start); - memcpy(__skb_put(skb, datasize), xdp->data, datasize); - if (metasize) + memcpy(__skb_put(skb, totalsize), xdp->data_meta, + ALIGN(totalsize, sizeof(long))); + + if (metasize) { skb_metadata_set(skb, metasize); + __skb_pull(skb, metasize); + } out: xsk_buff_free(xdp); @@ -268,7 +323,8 @@ static void i40e_handle_xdp_result_zc(struct i40e_ring *rx_ring, unsigned int *rx_packets, unsigned int *rx_bytes, unsigned int size, - unsigned int xdp_res) + unsigned int xdp_res, + bool *failure) { struct sk_buff *skb; @@ -278,11 +334,15 @@ static void i40e_handle_xdp_result_zc(struct i40e_ring *rx_ring, if (likely(xdp_res == I40E_XDP_REDIR) || xdp_res == I40E_XDP_TX) return; + if (xdp_res == I40E_XDP_EXIT) { + *failure = true; + return; + } + if (xdp_res == I40E_XDP_CONSUMED) { xsk_buff_free(xdp_buff); return; } - if (xdp_res == I40E_XDP_PASS) { /* NB! We are not checking for errors using * i40e_test_staterr with @@ -324,11 +384,17 @@ static void i40e_handle_xdp_result_zc(struct i40e_ring *rx_ring, int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) { unsigned int total_rx_bytes = 0, total_rx_packets = 0; - u16 cleaned_count = I40E_DESC_UNUSED(rx_ring); u16 next_to_clean = rx_ring->next_to_clean; u16 count_mask = rx_ring->count - 1; unsigned int xdp_res, xdp_xmit = 0; + struct bpf_prog *xdp_prog; bool failure = false; + u16 cleaned_count; + + /* NB! xdp_prog will always be !NULL, due to the fact that + * this path is enabled by setting an XDP program. + */ + xdp_prog = READ_ONCE(rx_ring->xdp_prog); while (likely(total_rx_packets < (unsigned int)budget)) { union i40e_rx_desc *rx_desc; @@ -366,9 +432,11 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) xsk_buff_set_size(bi, size); xsk_buff_dma_sync_for_cpu(bi, rx_ring->xsk_pool); - xdp_res = i40e_run_xdp_zc(rx_ring, bi); + xdp_res = i40e_run_xdp_zc(rx_ring, bi, xdp_prog); i40e_handle_xdp_result_zc(rx_ring, bi, rx_desc, &rx_packets, - &rx_bytes, size, xdp_res); + &rx_bytes, size, xdp_res, &failure); + if (failure) + break; total_rx_packets += rx_packets; total_rx_bytes += rx_bytes; xdp_xmit |= xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR); @@ -379,7 +447,7 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) cleaned_count = (next_to_clean - rx_ring->next_to_use - 1) & count_mask; if (cleaned_count >= I40E_RX_BUFFER_WRITE) - failure = !i40e_alloc_rx_buffers_zc(rx_ring, cleaned_count); + failure |= !i40e_alloc_rx_buffers_zc(rx_ring, cleaned_count); i40e_finalize_xdp_rx(rx_ring, xdp_xmit); i40e_update_rx_stats(rx_ring, total_rx_bytes, total_rx_packets); @@ -467,11 +535,11 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring) **/ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) { - struct xdp_desc *descs = xdp_ring->xsk_descs; + struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs; u32 nb_pkts, nb_processed = 0; unsigned int total_bytes = 0; - nb_pkts = xsk_tx_peek_release_desc_batch(xdp_ring->xsk_pool, descs, budget); + nb_pkts = xsk_tx_peek_release_desc_batch(xdp_ring->xsk_pool, budget); if (!nb_pkts) return true; @@ -591,13 +659,13 @@ int i40e_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags) return -ENETDOWN; if (!i40e_enabled_xdp_vsi(vsi)) - return -ENXIO; + return -EINVAL; if (queue_id >= vsi->num_queue_pairs) - return -ENXIO; + return -EINVAL; if (!vsi->xdp_rings[queue_id]->xsk_pool) - return -ENXIO; + return -EINVAL; ring = vsi->xdp_rings[queue_id]; diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.h b/drivers/net/ethernet/intel/i40e/i40e_xsk.h index ea88f4597a07..821df248f8be 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.h +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.h @@ -22,7 +22,6 @@ struct i40e_vsi; struct xsk_buff_pool; -struct zero_copy_allocator; int i40e_queue_pair_disable(struct i40e_vsi *vsi, int queue_pair); int i40e_queue_pair_enable(struct i40e_vsi *vsi, int queue_pair); @@ -33,7 +32,7 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget); bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi, struct i40e_ring *tx_ring); int i40e_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags); -int i40e_alloc_rx_bi_zc(struct i40e_ring *rx_ring); +int i40e_realloc_rx_bi_zc(struct i40e_vsi *vsi, bool zc); void i40e_clear_rx_bi_zc(struct i40e_ring *rx_ring); #endif /* _I40E_XSK_H_ */ |