diff options
Diffstat (limited to 'drivers/net/ethernet/intel/i40evf')
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h | 23 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40e_common.c | 1 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40e_devids.h | 1 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 226 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40e_txrx.h | 18 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h | 1 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40evf.h | 19 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c | 79 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40evf_main.c | 437 | ||||
| -rw-r--r-- | drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c | 86 |
10 files changed, 612 insertions, 279 deletions
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h index fcb9ef34cc7a..f5b2b369dc7c 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h @@ -227,6 +227,7 @@ enum i40e_admin_queue_opc { i40e_aqc_opc_nvm_update = 0x0703, i40e_aqc_opc_nvm_config_read = 0x0704, i40e_aqc_opc_nvm_config_write = 0x0705, + i40e_aqc_opc_oem_post_update = 0x0720, /* virtualization commands */ i40e_aqc_opc_send_msg_to_pf = 0x0801, @@ -1888,6 +1889,26 @@ struct i40e_aqc_nvm_config_data_immediate_field { I40E_CHECK_STRUCT_LEN(0xc, i40e_aqc_nvm_config_data_immediate_field); +/* OEM Post Update (indirect 0x0720) + * no command data struct used + */ + struct i40e_aqc_nvm_oem_post_update { +#define I40E_AQ_NVM_OEM_POST_UPDATE_EXTERNAL_DATA 0x01 + u8 sel_data; + u8 reserved[7]; +}; + +I40E_CHECK_STRUCT_LEN(0x8, i40e_aqc_nvm_oem_post_update); + +struct i40e_aqc_nvm_oem_post_update_buffer { + u8 str_len; + u8 dev_addr; + __le16 eeprom_addr; + u8 data[36]; +}; + +I40E_CHECK_STRUCT_LEN(0x28, i40e_aqc_nvm_oem_post_update_buffer); + /* Send to PF command (indirect 0x0801) id is only used by PF * Send to VF command (indirect 0x0802) id is only used by PF * Send to Peer PF command (indirect 0x0803) @@ -2311,4 +2332,4 @@ struct i40e_aqc_debug_modify_internals { I40E_CHECK_CMD_LENGTH(i40e_aqc_debug_modify_internals); -#endif +#endif /* _I40E_ADMINQ_CMD_H_ */ diff --git a/drivers/net/ethernet/intel/i40evf/i40e_common.c b/drivers/net/ethernet/intel/i40evf/i40e_common.c index 72b1942a94aa..938783e0baac 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_common.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_common.c @@ -44,7 +44,6 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw) switch (hw->device_id) { case I40E_DEV_ID_SFP_XL710: case I40E_DEV_ID_QEMU: - case I40E_DEV_ID_KX_A: case I40E_DEV_ID_KX_B: case I40E_DEV_ID_KX_C: case I40E_DEV_ID_QSFP_A: diff --git a/drivers/net/ethernet/intel/i40evf/i40e_devids.h b/drivers/net/ethernet/intel/i40evf/i40e_devids.h index e6a39c9862e8..ca8b58c3d1f5 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_devids.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_devids.h @@ -30,7 +30,6 @@ /* Device IDs */ #define I40E_DEV_ID_SFP_XL710 0x1572 #define I40E_DEV_ID_QEMU 0x1574 -#define I40E_DEV_ID_KX_A 0x157F #define I40E_DEV_ID_KX_B 0x1580 #define I40E_DEV_ID_KX_C 0x1581 #define I40E_DEV_ID_QSFP_A 0x1583 diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c index 47e9a90d6b10..7a00657dacda 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c @@ -51,11 +51,7 @@ static void i40e_unmap_and_free_tx_resource(struct i40e_ring *ring, struct i40e_tx_buffer *tx_buffer) { if (tx_buffer->skb) { - if (tx_buffer->tx_flags & I40E_TX_FLAGS_FD_SB) - kfree(tx_buffer->raw_buf); - else - dev_kfree_skb_any(tx_buffer->skb); - + dev_kfree_skb_any(tx_buffer->skb); if (dma_unmap_len(tx_buffer, len)) dma_unmap_single(ring->dev, dma_unmap_addr(tx_buffer, dma), @@ -67,6 +63,10 @@ static void i40e_unmap_and_free_tx_resource(struct i40e_ring *ring, dma_unmap_len(tx_buffer, len), DMA_TO_DEVICE); } + + if (tx_buffer->tx_flags & I40E_TX_FLAGS_FD_SB) + kfree(tx_buffer->raw_buf); + tx_buffer->next_to_watch = NULL; tx_buffer->skb = NULL; dma_unmap_len_set(tx_buffer, len, 0); @@ -127,17 +127,24 @@ void i40evf_free_tx_resources(struct i40e_ring *tx_ring) } /** - * i40e_get_head - Retrieve head from head writeback - * @tx_ring: tx ring to fetch head of + * i40evf_get_tx_pending - how many Tx descriptors not processed + * @tx_ring: the ring of descriptors * - * Returns value of Tx ring head based on value stored - * in head write-back location + * Since there is no access to the ring head register + * in XL710, we need to use our local copies **/ -static inline u32 i40e_get_head(struct i40e_ring *tx_ring) +u32 i40evf_get_tx_pending(struct i40e_ring *ring) { - void *head = (struct i40e_tx_desc *)tx_ring->desc + tx_ring->count; + u32 head, tail; + + head = i40e_get_head(ring); + tail = readl(ring->tail); - return le32_to_cpu(*(volatile __le32 *)head); + if (head != tail) + return (head < tail) ? + tail - head : (tail + ring->count - head); + + return 0; } #define WB_STRIDE 0x3 @@ -245,16 +252,6 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget) tx_ring->q_vector->tx.total_bytes += total_bytes; tx_ring->q_vector->tx.total_packets += total_packets; - /* check to see if there are any non-cache aligned descriptors - * waiting to be written back, and kick the hardware to force - * them to be written back in case of napi polling - */ - if (budget && - !((i & WB_STRIDE) == WB_STRIDE) && - !test_bit(__I40E_DOWN, &tx_ring->vsi->state) && - (I40E_DESC_UNUSED(tx_ring) != tx_ring->count)) - tx_ring->arm_wb = true; - netdev_tx_completed_queue(netdev_get_tx_queue(tx_ring->netdev, tx_ring->queue_index), total_packets, total_bytes); @@ -414,7 +411,7 @@ static bool i40e_set_new_dynamic_itr(struct i40e_ring_container *rc) return false; } -/* +/** * i40evf_setup_tx_descriptors - Allocate the Tx descriptors * @tx_ring: the tx ring to set up * @@ -889,31 +886,12 @@ checksum_fail: } /** - * i40e_rx_hash - returns the hash value from the Rx descriptor - * @ring: descriptor ring - * @rx_desc: specific descriptor - **/ -static inline u32 i40e_rx_hash(struct i40e_ring *ring, - union i40e_rx_desc *rx_desc) -{ - const __le64 rss_mask = - cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH << - I40E_RX_DESC_STATUS_FLTSTAT_SHIFT); - - if ((ring->netdev->features & NETIF_F_RXHASH) && - (rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask) - return le32_to_cpu(rx_desc->wb.qword0.hi_dword.rss); - else - return 0; -} - -/** - * i40e_ptype_to_hash - get a hash type + * i40e_ptype_to_htype - get a hash type * @ptype: the ptype value from the descriptor * * Returns a hash type to be used by skb_set_hash **/ -static inline enum pkt_hash_types i40e_ptype_to_hash(u8 ptype) +static inline enum pkt_hash_types i40e_ptype_to_htype(u8 ptype) { struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(ptype); @@ -931,6 +909,30 @@ static inline enum pkt_hash_types i40e_ptype_to_hash(u8 ptype) } /** + * i40e_rx_hash - set the hash value in the skb + * @ring: descriptor ring + * @rx_desc: specific descriptor + **/ +static inline void i40e_rx_hash(struct i40e_ring *ring, + union i40e_rx_desc *rx_desc, + struct sk_buff *skb, + u8 rx_ptype) +{ + u32 hash; + const __le64 rss_mask = + cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH << + I40E_RX_DESC_STATUS_FLTSTAT_SHIFT); + + if (ring->netdev->features & NETIF_F_RXHASH) + return; + + if ((rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask) { + hash = le32_to_cpu(rx_desc->wb.qword0.hi_dword.rss); + skb_set_hash(skb, hash, i40e_ptype_to_htype(rx_ptype)); + } +} + +/** * i40e_clean_rx_irq_ps - Reclaim resources after receive; packet split * @rx_ring: rx ring to clean * @budget: how many cleans we're allowed @@ -1071,8 +1073,8 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget) continue; } - skb_set_hash(skb, i40e_rx_hash(rx_ring, rx_desc), - i40e_ptype_to_hash(rx_ptype)); + i40e_rx_hash(rx_ring, rx_desc, skb, rx_ptype); + /* probably a little skewed due to removing CRC */ total_rx_bytes += skb->len; total_rx_packets++; @@ -1090,7 +1092,6 @@ static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget) continue; } #endif - skb_mark_napi_id(skb, &rx_ring->q_vector->napi); i40e_receive_skb(rx_ring, skb, vlan_tag); rx_desc->wb.qword1.status_error_len = 0; @@ -1189,8 +1190,7 @@ static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget) continue; } - skb_set_hash(skb, i40e_rx_hash(rx_ring, rx_desc), - i40e_ptype_to_hash(rx_ptype)); + i40e_rx_hash(rx_ring, rx_desc, skb, rx_ptype); /* probably a little skewed due to removing CRC */ total_rx_bytes += skb->len; total_rx_packets++; @@ -1263,10 +1263,12 @@ static inline void i40e_update_enable_itr(struct i40e_vsi *vsi, rx = i40e_set_new_dynamic_itr(&q_vector->rx); rxval = i40e_buildreg_itr(I40E_RX_ITR, q_vector->rx.itr); } + if (ITR_IS_DYNAMIC(vsi->tx_itr_setting)) { tx = i40e_set_new_dynamic_itr(&q_vector->tx); txval = i40e_buildreg_itr(I40E_TX_ITR, q_vector->tx.itr); } + if (rx || tx) { /* get the higher of the two ITR adjustments and * use the same value for both ITR registers @@ -1302,7 +1304,6 @@ enable_int: q_vector->itr_countdown--; else q_vector->itr_countdown = ITR_COUNTDOWN_START; - } /** @@ -1335,7 +1336,7 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget) */ i40e_for_each_ring(ring, q_vector->tx) { clean_complete &= i40e_clean_tx_irq(ring, vsi->work_limit); - arm_wb |= ring->arm_wb; + arm_wb = arm_wb || ring->arm_wb; ring->arm_wb = false; } @@ -1364,8 +1365,10 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget) /* If work not completed, return budget and polling will return */ if (!clean_complete) { tx_only: - if (arm_wb) + if (arm_wb) { + q_vector->tx.ring[0].tx_stats.tx_force_wb++; i40evf_force_wb(vsi, q_vector); + } return budget; } @@ -1437,13 +1440,12 @@ out: * @tx_ring: ptr to the ring to send * @skb: ptr to the skb we're sending * @hdr_len: ptr to the size of the packet header - * @cd_tunneling: ptr to context descriptor bits + * @cd_type_cmd_tso_mss: Quad Word 1 * * Returns 0 if no TSO can happen, 1 if tso is going, or error **/ static int i40e_tso(struct i40e_ring *tx_ring, struct sk_buff *skb, - u8 *hdr_len, u64 *cd_type_cmd_tso_mss, - u32 *cd_tunneling) + u8 *hdr_len, u64 *cd_type_cmd_tso_mss) { u32 cd_cmd, cd_tso_len, cd_mss; struct ipv6hdr *ipv6h; @@ -1555,7 +1557,6 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags, *tx_flags |= I40E_TX_FLAGS_IPV6; } - if ((tx_ring->flags & I40E_TXR_FLAGS_OUTER_UDP_CSUM) && (l4_tunnel == I40E_TXD_CTX_UDP_TUNNELING) && (*cd_tunneling & I40E_TXD_CTX_QW0_EXT_IP_MASK)) { @@ -1654,7 +1655,7 @@ static void i40e_create_tx_ctx(struct i40e_ring *tx_ring, context_desc->type_cmd_tso_mss = cpu_to_le64(cd_type_cmd_tso_mss); } - /** +/** * i40e_chk_linearize - Check if there are more than 8 fragments per packet * @skb: send buffer * @tx_flags: collected send information @@ -1770,6 +1771,9 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, u32 td_tag = 0; dma_addr_t dma; u16 gso_segs; + u16 desc_count = 0; + bool tail_bump = true; + bool do_rs = false; if (tx_flags & I40E_TX_FLAGS_HW_VLAN) { td_cmd |= I40E_TX_DESC_CMD_IL2TAG1; @@ -1810,6 +1814,8 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_desc++; i++; + desc_count++; + if (i == tx_ring->count) { tx_desc = I40E_TX_DESC(tx_ring, 0); i = 0; @@ -1829,6 +1835,8 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_desc++; i++; + desc_count++; + if (i == tx_ring->count) { tx_desc = I40E_TX_DESC(tx_ring, 0); i = 0; @@ -1843,35 +1851,6 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_bi = &tx_ring->tx_bi[i]; } - /* Place RS bit on last descriptor of any packet that spans across the - * 4th descriptor (WB_STRIDE aka 0x3) in a 64B cacheline. - */ -#define WB_STRIDE 0x3 - if (((i & WB_STRIDE) != WB_STRIDE) && - (first <= &tx_ring->tx_bi[i]) && - (first >= &tx_ring->tx_bi[i & ~WB_STRIDE])) { - tx_desc->cmd_type_offset_bsz = - build_ctob(td_cmd, td_offset, size, td_tag) | - cpu_to_le64((u64)I40E_TX_DESC_CMD_EOP << - I40E_TXD_QW1_CMD_SHIFT); - } else { - tx_desc->cmd_type_offset_bsz = - build_ctob(td_cmd, td_offset, size, td_tag) | - cpu_to_le64((u64)I40E_TXD_CMD << - I40E_TXD_QW1_CMD_SHIFT); - } - - netdev_tx_sent_queue(netdev_get_tx_queue(tx_ring->netdev, - tx_ring->queue_index), - first->bytecount); - - /* Force memory writes to complete before letting h/w - * know there are new descriptors to fetch. (Only - * applicable for weak-ordered memory model archs, - * such as IA-64). - */ - wmb(); - /* set next_to_watch value indicating a packet is present */ first->next_to_watch = tx_desc; @@ -1881,15 +1860,72 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb, tx_ring->next_to_use = i; + netdev_tx_sent_queue(netdev_get_tx_queue(tx_ring->netdev, + tx_ring->queue_index), + first->bytecount); i40evf_maybe_stop_tx(tx_ring, DESC_NEEDED); + + /* Algorithm to optimize tail and RS bit setting: + * if xmit_more is supported + * if xmit_more is true + * do not update tail and do not mark RS bit. + * if xmit_more is false and last xmit_more was false + * if every packet spanned less than 4 desc + * then set RS bit on 4th packet and update tail + * on every packet + * else + * update tail and set RS bit on every packet. + * if xmit_more is false and last_xmit_more was true + * update tail and set RS bit. + * + * Optimization: wmb to be issued only in case of tail update. + * Also optimize the Descriptor WB path for RS bit with the same + * algorithm. + * + * Note: If there are less than 4 packets + * pending and interrupts were disabled the service task will + * trigger a force WB. + */ + if (skb->xmit_more && + !netif_xmit_stopped(netdev_get_tx_queue(tx_ring->netdev, + tx_ring->queue_index))) { + tx_ring->flags |= I40E_TXR_FLAGS_LAST_XMIT_MORE_SET; + tail_bump = false; + } else if (!skb->xmit_more && + !netif_xmit_stopped(netdev_get_tx_queue(tx_ring->netdev, + tx_ring->queue_index)) && + (!(tx_ring->flags & I40E_TXR_FLAGS_LAST_XMIT_MORE_SET)) && + (tx_ring->packet_stride < WB_STRIDE) && + (desc_count < WB_STRIDE)) { + tx_ring->packet_stride++; + } else { + tx_ring->packet_stride = 0; + tx_ring->flags &= ~I40E_TXR_FLAGS_LAST_XMIT_MORE_SET; + do_rs = true; + } + if (do_rs) + tx_ring->packet_stride = 0; + + tx_desc->cmd_type_offset_bsz = + build_ctob(td_cmd, td_offset, size, td_tag) | + cpu_to_le64((u64)(do_rs ? I40E_TXD_CMD : + I40E_TX_DESC_CMD_EOP) << + I40E_TXD_QW1_CMD_SHIFT); + /* notify HW of packet */ - if (!skb->xmit_more || - netif_xmit_stopped(netdev_get_tx_queue(tx_ring->netdev, - tx_ring->queue_index))) - writel(i, tx_ring->tail); - else + if (!tail_bump) prefetchw(tx_desc + 1); + if (tail_bump) { + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). + */ + wmb(); + writel(i, tx_ring->tail); + } + return; dma_error: @@ -1961,6 +1997,9 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, u8 hdr_len = 0; int tso; + /* prefetch the data, we'll need it later */ + prefetch(skb->data); + if (0 == i40evf_xmit_descriptor_count(skb, tx_ring)) return NETDEV_TX_BUSY; @@ -1980,8 +2019,7 @@ static netdev_tx_t i40e_xmit_frame_ring(struct sk_buff *skb, else if (protocol == htons(ETH_P_IPV6)) tx_flags |= I40E_TX_FLAGS_IPV6; - tso = i40e_tso(tx_ring, skb, &hdr_len, - &cd_type_cmd_tso_mss, &cd_tunneling); + tso = i40e_tso(tx_ring, skb, &hdr_len, &cd_type_cmd_tso_mss); if (tso < 0) goto out_drop; @@ -2029,7 +2067,7 @@ out_drop: netdev_tx_t i40evf_xmit_frame(struct sk_buff *skb, struct net_device *netdev) { struct i40evf_adapter *adapter = netdev_priv(netdev); - struct i40e_ring *tx_ring = adapter->tx_rings[skb->queue_mapping]; + struct i40e_ring *tx_ring = &adapter->tx_rings[skb->queue_mapping]; /* hardware can't handle really short frames, hardware padding works * beyond this point diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h index ebc1bf77f036..e29bb3e86cfd 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h @@ -201,6 +201,7 @@ struct i40e_tx_queue_stats { u64 tx_busy; u64 tx_done_old; u64 tx_linearize; + u64 tx_force_wb; }; struct i40e_rx_queue_stats { @@ -267,6 +268,8 @@ struct i40e_ring { bool ring_active; /* is ring online or not */ bool arm_wb; /* do something to arm write back */ + u8 packet_stride; +#define I40E_TXR_FLAGS_LAST_XMIT_MORE_SET BIT(2) u16 flags; #define I40E_TXR_FLAGS_WB_ON_ITR BIT(0) @@ -321,4 +324,19 @@ int i40evf_setup_rx_descriptors(struct i40e_ring *rx_ring); void i40evf_free_tx_resources(struct i40e_ring *tx_ring); void i40evf_free_rx_resources(struct i40e_ring *rx_ring); int i40evf_napi_poll(struct napi_struct *napi, int budget); +u32 i40evf_get_tx_pending(struct i40e_ring *ring); + +/** + * i40e_get_head - Retrieve head from head writeback + * @tx_ring: Tx ring to fetch head of + * + * Returns value of Tx ring head based on value stored + * in head write-back location + **/ +static inline u32 i40e_get_head(struct i40e_ring *tx_ring) +{ + void *head = (struct i40e_tx_desc *)tx_ring->desc + tx_ring->count; + + return le32_to_cpu(*(volatile __le32 *)head); +} #endif /* _I40E_TXRX_H_ */ diff --git a/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h b/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h index 9f7b279b9d9c..3b9d2037456c 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_virtchnl.h @@ -153,6 +153,7 @@ struct i40e_virtchnl_vsi_resource { #define I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR 0x00000020 #define I40E_VIRTCHNL_VF_OFFLOAD_VLAN 0x00010000 #define I40E_VIRTCHNL_VF_OFFLOAD_RX_POLLING 0x00020000 +#define I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2 0x00040000 struct i40e_virtchnl_vf_resource { u16 num_vsis; diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h index 22fc3d49c4b9..be1b72b93888 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf.h +++ b/drivers/net/ethernet/intel/i40evf/i40evf.h @@ -67,6 +67,8 @@ struct i40e_vsi { u16 rx_itr_setting; u16 tx_itr_setting; u16 qs_handle; + u8 *rss_hkey_user; /* User configured hash keys */ + u8 *rss_lut_user; /* User configured lookup table entries */ }; /* How many Rx Buffers do we bundle into one write to the hardware ? */ @@ -95,10 +97,10 @@ struct i40e_vsi { #define I40E_TX_DESC(R, i) (&(((struct i40e_tx_desc *)((R)->desc))[i])) #define I40E_TX_CTXTDESC(R, i) \ (&(((struct i40e_tx_context_desc *)((R)->desc))[i])) -#define MAX_RX_QUEUES 8 -#define MAX_TX_QUEUES MAX_RX_QUEUES +#define MAX_QUEUES 16 #define I40EVF_HKEY_ARRAY_SIZE ((I40E_VFQF_HKEY_MAX_INDEX + 1) * 4) +#define I40EVF_HLUT_ARRAY_SIZE ((I40E_VFQF_HLUT_MAX_INDEX + 1) * 4) /* MAX_MSIX_Q_VECTORS of these are allocated, * but we only use one per queue-specific vector. @@ -142,9 +144,6 @@ struct i40e_q_vector { #define OTHER_VECTOR 1 #define NONQ_VECS (OTHER_VECTOR) -#define MAX_MSIX_Q_VECTORS 4 -#define MAX_MSIX_COUNT 5 - #define MIN_MSIX_Q_VECTORS 1 #define MIN_MSIX_COUNT (MIN_MSIX_Q_VECTORS + NONQ_VECS) @@ -190,19 +189,19 @@ struct i40evf_adapter { struct work_struct reset_task; struct work_struct adminq_task; struct delayed_work init_task; - struct i40e_q_vector *q_vector[MAX_MSIX_Q_VECTORS]; + struct i40e_q_vector *q_vectors; struct list_head vlan_filter_list; char misc_vector_name[IFNAMSIZ + 9]; int num_active_queues; /* TX */ - struct i40e_ring *tx_rings[I40E_MAX_VSI_QP]; + struct i40e_ring *tx_rings; u32 tx_timeout_count; struct list_head mac_filter_list; u32 tx_desc_count; /* RX */ - struct i40e_ring *rx_rings[I40E_MAX_VSI_QP]; + struct i40e_ring *rx_rings; u64 hw_csum_rx_error; u32 rx_desc_count; int num_msix_vectors; @@ -313,4 +312,8 @@ void i40evf_request_reset(struct i40evf_adapter *adapter); void i40evf_virtchnl_completion(struct i40evf_adapter *adapter, enum i40e_virtchnl_ops v_opcode, i40e_status v_retval, u8 *msg, u16 msglen); +int i40evf_config_rss(struct i40e_vsi *vsi, const u8 *seed, u8 *lut, + u16 lut_size); +int i40evf_get_rss(struct i40e_vsi *vsi, const u8 *seed, u8 *lut, + u16 lut_size); #endif /* _I40EVF_H_ */ diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c index 4790437a50ac..a4c9feb589e7 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c @@ -121,12 +121,12 @@ static void i40evf_get_ethtool_stats(struct net_device *netdev, data[i] = *(u64 *)p; } for (j = 0; j < adapter->num_active_queues; j++) { - data[i++] = adapter->tx_rings[j]->stats.packets; - data[i++] = adapter->tx_rings[j]->stats.bytes; + data[i++] = adapter->tx_rings[j].stats.packets; + data[i++] = adapter->tx_rings[j].stats.bytes; } for (j = 0; j < adapter->num_active_queues; j++) { - data[i++] = adapter->rx_rings[j]->stats.packets; - data[i++] = adapter->rx_rings[j]->stats.bytes; + data[i++] = adapter->rx_rings[j].stats.packets; + data[i++] = adapter->rx_rings[j].stats.bytes; } } @@ -351,7 +351,7 @@ static int i40evf_set_coalesce(struct net_device *netdev, vsi->tx_itr_setting &= ~I40E_ITR_DYNAMIC; for (i = 0; i < adapter->num_msix_vectors - NONQ_VECS; i++) { - q_vector = adapter->q_vector[i]; + q_vector = &adapter->q_vectors[i]; q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting); wr32(hw, I40E_VFINT_ITRN1(0, i), q_vector->rx.itr); q_vector->tx.itr = ITR_TO_REG(vsi->tx_itr_setting); @@ -634,25 +634,34 @@ static int i40evf_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, u8 *hfunc) { struct i40evf_adapter *adapter = netdev_priv(netdev); - struct i40e_hw *hw = &adapter->hw; - u32 hlut_val; - int i, j; + struct i40e_vsi *vsi = &adapter->vsi; + u8 *seed = NULL, *lut; + int ret; + u16 i; if (hfunc) *hfunc = ETH_RSS_HASH_TOP; if (!indir) return 0; - if (indir) { - for (i = 0, j = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) { - hlut_val = rd32(hw, I40E_VFQF_HLUT(i)); - indir[j++] = hlut_val & 0xff; - indir[j++] = (hlut_val >> 8) & 0xff; - indir[j++] = (hlut_val >> 16) & 0xff; - indir[j++] = (hlut_val >> 24) & 0xff; - } - } - return 0; + seed = key; + + lut = kzalloc(I40EVF_HLUT_ARRAY_SIZE, GFP_KERNEL); + if (!lut) + return -ENOMEM; + + ret = i40evf_get_rss(vsi, seed, lut, I40EVF_HLUT_ARRAY_SIZE); + if (ret) + goto out; + + /* Each 32 bits pointed by 'indir' is stored with a lut entry */ + for (i = 0; i < I40EVF_HLUT_ARRAY_SIZE; i++) + indir[i] = (u32)lut[i]; + +out: + kfree(lut); + + return ret; } /** @@ -668,9 +677,9 @@ static int i40evf_set_rxfh(struct net_device *netdev, const u32 *indir, const u8 *key, const u8 hfunc) { struct i40evf_adapter *adapter = netdev_priv(netdev); - struct i40e_hw *hw = &adapter->hw; - u32 hlut_val; - int i, j; + struct i40e_vsi *vsi = &adapter->vsi; + u8 *seed = NULL; + u16 i; /* We do not allow change in unsupported parameters */ if (key || @@ -679,15 +688,29 @@ static int i40evf_set_rxfh(struct net_device *netdev, const u32 *indir, if (!indir) return 0; - for (i = 0, j = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) { - hlut_val = indir[j++]; - hlut_val |= indir[j++] << 8; - hlut_val |= indir[j++] << 16; - hlut_val |= indir[j++] << 24; - wr32(hw, I40E_VFQF_HLUT(i), hlut_val); + if (key) { + if (!vsi->rss_hkey_user) { + vsi->rss_hkey_user = kzalloc(I40EVF_HKEY_ARRAY_SIZE, + GFP_KERNEL); + if (!vsi->rss_hkey_user) + return -ENOMEM; + } + memcpy(vsi->rss_hkey_user, key, I40EVF_HKEY_ARRAY_SIZE); + seed = vsi->rss_hkey_user; + } + if (!vsi->rss_lut_user) { + vsi->rss_lut_user = kzalloc(I40EVF_HLUT_ARRAY_SIZE, + GFP_KERNEL); + if (!vsi->rss_lut_user) + return -ENOMEM; } - return 0; + /* Each 32 bits pointed by 'indir' is stored with a lut entry */ + for (i = 0; i < I40EVF_HLUT_ARRAY_SIZE; i++) + vsi->rss_lut_user[i] = (u8)(indir[i]); + + return i40evf_config_rss(vsi, seed, vsi->rss_lut_user, + I40EVF_HLUT_ARRAY_SIZE); } static const struct ethtool_ops i40evf_ethtool_ops = { diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index 99d2cffae0cd..94da913b151d 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -34,7 +34,15 @@ char i40evf_driver_name[] = "i40evf"; static const char i40evf_driver_string[] = "Intel(R) XL710/X710 Virtual Function Network Driver"; -#define DRV_VERSION "1.3.33" +#define DRV_KERN "-k" + +#define DRV_VERSION_MAJOR 1 +#define DRV_VERSION_MINOR 4 +#define DRV_VERSION_BUILD 4 +#define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ + __stringify(DRV_VERSION_MINOR) "." \ + __stringify(DRV_VERSION_BUILD) \ + DRV_KERN const char i40evf_driver_version[] = DRV_VERSION; static const char i40evf_copyright[] = "Copyright (c) 2013 - 2015 Intel Corporation."; @@ -259,7 +267,7 @@ static void i40evf_fire_sw_int(struct i40evf_adapter *adapter, u32 mask) { struct i40e_hw *hw = &adapter->hw; int i; - uint32_t dyn_ctl; + u32 dyn_ctl; if (mask & 1) { dyn_ctl = rd32(hw, I40E_VFINT_DYN_CTL01); @@ -307,10 +315,9 @@ static irqreturn_t i40evf_msix_aq(int irq, void *data) struct i40e_hw *hw = &adapter->hw; u32 val; - /* handle non-queue interrupts */ - rd32(hw, I40E_VFINT_ICR01); - rd32(hw, I40E_VFINT_ICR0_ENA1); - + /* handle non-queue interrupts, these reads clear the registers */ + val = rd32(hw, I40E_VFINT_ICR01); + val = rd32(hw, I40E_VFINT_ICR0_ENA1); val = rd32(hw, I40E_VFINT_DYN_CTL01) | I40E_VFINT_DYN_CTL01_CLEARPBA_MASK; @@ -348,8 +355,8 @@ static irqreturn_t i40evf_msix_clean_rings(int irq, void *data) static void i40evf_map_vector_to_rxq(struct i40evf_adapter *adapter, int v_idx, int r_idx) { - struct i40e_q_vector *q_vector = adapter->q_vector[v_idx]; - struct i40e_ring *rx_ring = adapter->rx_rings[r_idx]; + struct i40e_q_vector *q_vector = &adapter->q_vectors[v_idx]; + struct i40e_ring *rx_ring = &adapter->rx_rings[r_idx]; rx_ring->q_vector = q_vector; rx_ring->next = q_vector->rx.ring; @@ -369,8 +376,8 @@ i40evf_map_vector_to_rxq(struct i40evf_adapter *adapter, int v_idx, int r_idx) static void i40evf_map_vector_to_txq(struct i40evf_adapter *adapter, int v_idx, int t_idx) { - struct i40e_q_vector *q_vector = adapter->q_vector[v_idx]; - struct i40e_ring *tx_ring = adapter->tx_rings[t_idx]; + struct i40e_q_vector *q_vector = &adapter->q_vectors[v_idx]; + struct i40e_ring *tx_ring = &adapter->tx_rings[t_idx]; tx_ring->q_vector = q_vector; tx_ring->next = q_vector->tx.ring; @@ -465,7 +472,7 @@ static void i40evf_netpoll(struct net_device *netdev) return; for (i = 0; i < q_vectors; i++) - i40evf_msix_clean_rings(0, adapter->q_vector[i]); + i40evf_msix_clean_rings(0, &adapter->q_vectors[i]); } #endif @@ -487,7 +494,7 @@ i40evf_request_traffic_irqs(struct i40evf_adapter *adapter, char *basename) q_vectors = adapter->num_msix_vectors - NONQ_VECS; for (vector = 0; vector < q_vectors; vector++) { - struct i40e_q_vector *q_vector = adapter->q_vector[vector]; + struct i40e_q_vector *q_vector = &adapter->q_vectors[vector]; if (q_vector->tx.ring && q_vector->rx.ring) { snprintf(q_vector->name, sizeof(q_vector->name) - 1, @@ -532,7 +539,7 @@ free_queue_irqs: adapter->msix_entries[vector + NONQ_VECS].vector, NULL); free_irq(adapter->msix_entries[vector + NONQ_VECS].vector, - adapter->q_vector[vector]); + &adapter->q_vectors[vector]); } return err; } @@ -582,7 +589,7 @@ static void i40evf_free_traffic_irqs(struct i40evf_adapter *adapter) irq_set_affinity_hint(adapter->msix_entries[i+1].vector, NULL); free_irq(adapter->msix_entries[i+1].vector, - adapter->q_vector[i]); + &adapter->q_vectors[i]); } } @@ -611,7 +618,7 @@ static void i40evf_configure_tx(struct i40evf_adapter *adapter) int i; for (i = 0; i < adapter->num_active_queues; i++) - adapter->tx_rings[i]->tail = hw->hw_addr + I40E_QTX_TAIL1(i); + adapter->tx_rings[i].tail = hw->hw_addr + I40E_QTX_TAIL1(i); } /** @@ -656,8 +663,8 @@ static void i40evf_configure_rx(struct i40evf_adapter *adapter) } for (i = 0; i < adapter->num_active_queues; i++) { - adapter->rx_rings[i]->tail = hw->hw_addr + I40E_QRX_TAIL1(i); - adapter->rx_rings[i]->rx_buf_len = rx_buf_len; + adapter->rx_rings[i].tail = hw->hw_addr + I40E_QRX_TAIL1(i); + adapter->rx_rings[i].rx_buf_len = rx_buf_len; } } @@ -954,7 +961,7 @@ static void i40evf_napi_enable_all(struct i40evf_adapter *adapter) for (q_idx = 0; q_idx < q_vectors; q_idx++) { struct napi_struct *napi; - q_vector = adapter->q_vector[q_idx]; + q_vector = &adapter->q_vectors[q_idx]; napi = &q_vector->napi; napi_enable(napi); } @@ -971,7 +978,7 @@ static void i40evf_napi_disable_all(struct i40evf_adapter *adapter) int q_vectors = adapter->num_msix_vectors - NONQ_VECS; for (q_idx = 0; q_idx < q_vectors; q_idx++) { - q_vector = adapter->q_vector[q_idx]; + q_vector = &adapter->q_vectors[q_idx]; napi_disable(&q_vector->napi); } } @@ -992,7 +999,7 @@ static void i40evf_configure(struct i40evf_adapter *adapter) adapter->aq_required |= I40EVF_FLAG_AQ_CONFIGURE_QUEUES; for (i = 0; i < adapter->num_active_queues; i++) { - struct i40e_ring *ring = adapter->rx_rings[i]; + struct i40e_ring *ring = &adapter->rx_rings[i]; i40evf_alloc_rx_buffers_1buf(ring, ring->count); ring->next_to_use = ring->count - 1; @@ -1112,16 +1119,10 @@ i40evf_acquire_msix_vectors(struct i40evf_adapter *adapter, int vectors) **/ static void i40evf_free_queues(struct i40evf_adapter *adapter) { - int i; - if (!adapter->vsi_res) return; - for (i = 0; i < adapter->num_active_queues; i++) { - if (adapter->tx_rings[i]) - kfree_rcu(adapter->tx_rings[i], rcu); - adapter->tx_rings[i] = NULL; - adapter->rx_rings[i] = NULL; - } + kfree(adapter->tx_rings); + kfree(adapter->rx_rings); } /** @@ -1136,13 +1137,20 @@ static int i40evf_alloc_queues(struct i40evf_adapter *adapter) { int i; + adapter->tx_rings = kcalloc(adapter->num_active_queues, + sizeof(struct i40e_ring), GFP_KERNEL); + if (!adapter->tx_rings) + goto err_out; + adapter->rx_rings = kcalloc(adapter->num_active_queues, + sizeof(struct i40e_ring), GFP_KERNEL); + if (!adapter->rx_rings) + goto err_out; + for (i = 0; i < adapter->num_active_queues; i++) { struct i40e_ring *tx_ring; struct i40e_ring *rx_ring; - tx_ring = kzalloc(sizeof(*tx_ring) * 2, GFP_KERNEL); - if (!tx_ring) - goto err_out; + tx_ring = &adapter->tx_rings[i]; tx_ring->queue_index = i; tx_ring->netdev = adapter->netdev; @@ -1150,14 +1158,12 @@ static int i40evf_alloc_queues(struct i40evf_adapter *adapter) tx_ring->count = adapter->tx_desc_count; if (adapter->flags & I40E_FLAG_WB_ON_ITR_CAPABLE) tx_ring->flags |= I40E_TXR_FLAGS_WB_ON_ITR; - adapter->tx_rings[i] = tx_ring; - rx_ring = &tx_ring[1]; + rx_ring = &adapter->rx_rings[i]; rx_ring->queue_index = i; rx_ring->netdev = adapter->netdev; rx_ring->dev = &adapter->pdev->dev; rx_ring->count = adapter->rx_desc_count; - adapter->rx_rings[i] = rx_ring; } return 0; @@ -1207,115 +1213,273 @@ static int i40evf_set_interrupt_capability(struct i40evf_adapter *adapter) err = i40evf_acquire_msix_vectors(adapter, v_budget); out: - adapter->netdev->real_num_tx_queues = pairs; + netif_set_real_num_rx_queues(adapter->netdev, pairs); + netif_set_real_num_tx_queues(adapter->netdev, pairs); return err; } /** - * i40e_configure_rss_aq - Prepare for RSS using AQ commands + * i40e_config_rss_aq - Prepare for RSS using AQ commands * @vsi: vsi structure * @seed: RSS hash seed + * @lut: Lookup table + * @lut_size: Lookup table size + * + * Return 0 on success, negative on failure **/ -static void i40evf_configure_rss_aq(struct i40e_vsi *vsi, const u8 *seed) +static int i40evf_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed, + u8 *lut, u16 lut_size) { - struct i40e_aqc_get_set_rss_key_data rss_key; struct i40evf_adapter *adapter = vsi->back; struct i40e_hw *hw = &adapter->hw; - int ret = 0, i; - u8 *rss_lut; + int ret = 0; if (!vsi->id) - return; + return -EINVAL; if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) { /* bail because we already have a command pending */ - dev_err(&adapter->pdev->dev, "Cannot confiure RSS, command %d pending\n", + dev_err(&adapter->pdev->dev, "Cannot configure RSS, command %d pending\n", adapter->current_op); - return; + return -EBUSY; } - memset(&rss_key, 0, sizeof(rss_key)); - memcpy(&rss_key, seed, sizeof(rss_key)); + if (seed) { + struct i40e_aqc_get_set_rss_key_data *rss_key = + (struct i40e_aqc_get_set_rss_key_data *)seed; + ret = i40evf_aq_set_rss_key(hw, vsi->id, rss_key); + if (ret) { + dev_err(&adapter->pdev->dev, "Cannot set RSS key, err %s aq_err %s\n", + i40evf_stat_str(hw, ret), + i40evf_aq_str(hw, hw->aq.asq_last_status)); + return ret; + } + } - rss_lut = kzalloc(((I40E_VFQF_HLUT_MAX_INDEX + 1) * 4), GFP_KERNEL); - if (!rss_lut) - return; + if (lut) { + ret = i40evf_aq_set_rss_lut(hw, vsi->id, false, lut, lut_size); + if (ret) { + dev_err(&adapter->pdev->dev, + "Cannot set RSS lut, err %s aq_err %s\n", + i40evf_stat_str(hw, ret), + i40evf_aq_str(hw, hw->aq.asq_last_status)); + return ret; + } + } - /* Populate the LUT with max no. PF queues in round robin fashion */ - for (i = 0; i <= (I40E_VFQF_HLUT_MAX_INDEX * 4); i++) - rss_lut[i] = i % adapter->num_active_queues; + return ret; +} - ret = i40evf_aq_set_rss_key(hw, vsi->id, &rss_key); - if (ret) { - dev_err(&adapter->pdev->dev, - "Cannot set RSS key, err %s aq_err %s\n", - i40evf_stat_str(hw, ret), - i40evf_aq_str(hw, hw->aq.asq_last_status)); - return; +/** + * i40evf_config_rss_reg - Configure RSS keys and lut by writing registers + * @vsi: Pointer to vsi structure + * @seed: RSS hash seed + * @lut: Lookup table + * @lut_size: Lookup table size + * + * Returns 0 on success, negative on failure + **/ +static int i40evf_config_rss_reg(struct i40e_vsi *vsi, const u8 *seed, + const u8 *lut, u16 lut_size) +{ + struct i40evf_adapter *adapter = vsi->back; + struct i40e_hw *hw = &adapter->hw; + u16 i; + + if (seed) { + u32 *seed_dw = (u32 *)seed; + + for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++) + wr32(hw, I40E_VFQF_HKEY(i), seed_dw[i]); } - ret = i40evf_aq_set_rss_lut(hw, vsi->id, false, rss_lut, - (I40E_VFQF_HLUT_MAX_INDEX + 1) * 4); - if (ret) - dev_err(&adapter->pdev->dev, - "Cannot set RSS lut, err %s aq_err %s\n", - i40evf_stat_str(hw, ret), - i40evf_aq_str(hw, hw->aq.asq_last_status)); + if (lut) { + u32 *lut_dw = (u32 *)lut; + + if (lut_size != I40EVF_HLUT_ARRAY_SIZE) + return -EINVAL; + + for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) + wr32(hw, I40E_VFQF_HLUT(i), lut_dw[i]); + } + i40e_flush(hw); + + return 0; } /** - * i40e_configure_rss_reg - Prepare for RSS if used - * @adapter: board private structure - * @seed: RSS hash seed + * * i40evf_get_rss_aq - Get RSS keys and lut by using AQ commands + * @vsi: Pointer to vsi structure + * @seed: RSS hash seed + * @lut: Lookup table + * @lut_size: Lookup table size + * + * Return 0 on success, negative on failure **/ -static void i40evf_configure_rss_reg(struct i40evf_adapter *adapter, - const u8 *seed) +static int i40evf_get_rss_aq(struct i40e_vsi *vsi, const u8 *seed, + u8 *lut, u16 lut_size) { + struct i40evf_adapter *adapter = vsi->back; struct i40e_hw *hw = &adapter->hw; - u32 *seed_dw = (u32 *)seed; - u32 cqueue = 0; - u32 lut = 0; - int i, j; + int ret = 0; - /* Fill out hash function seed */ - for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++) - wr32(hw, I40E_VFQF_HKEY(i), seed_dw[i]); - - /* Populate the LUT with max no. PF queues in round robin fashion */ - for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) { - lut = 0; - for (j = 0; j < 4; j++) { - if (cqueue == adapter->num_active_queues) - cqueue = 0; - lut |= ((cqueue) << (8 * j)); - cqueue++; + if (seed) { + ret = i40evf_aq_get_rss_key(hw, vsi->id, + (struct i40e_aqc_get_set_rss_key_data *)seed); + if (ret) { + dev_err(&adapter->pdev->dev, + "Cannot get RSS key, err %s aq_err %s\n", + i40evf_stat_str(hw, ret), + i40evf_aq_str(hw, hw->aq.asq_last_status)); + return ret; } - wr32(hw, I40E_VFQF_HLUT(i), lut); } - i40e_flush(hw); + + if (lut) { + ret = i40evf_aq_get_rss_lut(hw, vsi->id, seed, lut, lut_size); + if (ret) { + dev_err(&adapter->pdev->dev, + "Cannot get RSS lut, err %s aq_err %s\n", + i40evf_stat_str(hw, ret), + i40evf_aq_str(hw, hw->aq.asq_last_status)); + return ret; + } + } + + return ret; } /** - * i40evf_configure_rss - Prepare for RSS + * * i40evf_get_rss_reg - Get RSS keys and lut by reading registers + * @vsi: Pointer to vsi structure + * @seed: RSS hash seed + * @lut: Lookup table + * @lut_size: Lookup table size + * + * Returns 0 on success, negative on failure + **/ +static int i40evf_get_rss_reg(struct i40e_vsi *vsi, const u8 *seed, + const u8 *lut, u16 lut_size) +{ + struct i40evf_adapter *adapter = vsi->back; + struct i40e_hw *hw = &adapter->hw; + u16 i; + + if (seed) { + u32 *seed_dw = (u32 *)seed; + + for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++) + seed_dw[i] = rd32(hw, I40E_VFQF_HKEY(i)); + } + + if (lut) { + u32 *lut_dw = (u32 *)lut; + + if (lut_size != I40EVF_HLUT_ARRAY_SIZE) + return -EINVAL; + + for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) + lut_dw[i] = rd32(hw, I40E_VFQF_HLUT(i)); + } + + return 0; +} + +/** + * i40evf_config_rss - Configure RSS keys and lut + * @vsi: Pointer to vsi structure + * @seed: RSS hash seed + * @lut: Lookup table + * @lut_size: Lookup table size + * + * Returns 0 on success, negative on failure + **/ +int i40evf_config_rss(struct i40e_vsi *vsi, const u8 *seed, + u8 *lut, u16 lut_size) +{ + struct i40evf_adapter *adapter = vsi->back; + + if (RSS_AQ(adapter)) + return i40evf_config_rss_aq(vsi, seed, lut, lut_size); + else + return i40evf_config_rss_reg(vsi, seed, lut, lut_size); +} + +/** + * i40evf_get_rss - Get RSS keys and lut + * @vsi: Pointer to vsi structure + * @seed: RSS hash seed + * @lut: Lookup table + * @lut_size: Lookup table size + * + * Returns 0 on success, negative on failure + **/ +int i40evf_get_rss(struct i40e_vsi *vsi, const u8 *seed, u8 *lut, u16 lut_size) +{ + struct i40evf_adapter *adapter = vsi->back; + + if (RSS_AQ(adapter)) + return i40evf_get_rss_aq(vsi, seed, lut, lut_size); + else + return i40evf_get_rss_reg(vsi, seed, lut, lut_size); +} + +/** + * i40evf_fill_rss_lut - Fill the lut with default values + * @lut: Lookup table to be filled with + * @rss_table_size: Lookup table size + * @rss_size: Range of queue number for hashing + **/ +static void i40evf_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size) +{ + u16 i; + + for (i = 0; i < rss_table_size; i++) + lut[i] = i % rss_size; +} + +/** + * i40evf_init_rss - Prepare for RSS * @adapter: board private structure + * + * Return 0 on success, negative on failure **/ -static void i40evf_configure_rss(struct i40evf_adapter *adapter) +static int i40evf_init_rss(struct i40evf_adapter *adapter) { + struct i40e_vsi *vsi = &adapter->vsi; struct i40e_hw *hw = &adapter->hw; u8 seed[I40EVF_HKEY_ARRAY_SIZE]; u64 hena; - - netdev_rss_key_fill((void *)seed, I40EVF_HKEY_ARRAY_SIZE); + u8 *lut; + int ret; /* Enable PCTYPES for RSS, TCP/UDP with IPv4/IPv6 */ hena = I40E_DEFAULT_RSS_HENA; wr32(hw, I40E_VFQF_HENA(0), (u32)hena); wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32)); - if (RSS_AQ(adapter)) - i40evf_configure_rss_aq(&adapter->vsi, seed); + lut = kzalloc(I40EVF_HLUT_ARRAY_SIZE, GFP_KERNEL); + if (!lut) + return -ENOMEM; + + /* Use user configured lut if there is one, otherwise use default */ + if (vsi->rss_lut_user) + memcpy(lut, vsi->rss_lut_user, I40EVF_HLUT_ARRAY_SIZE); + else + i40evf_fill_rss_lut(lut, I40EVF_HLUT_ARRAY_SIZE, + adapter->num_active_queues); + + /* Use user configured hash key if there is one, otherwise + * user default. + */ + if (vsi->rss_hkey_user) + memcpy(seed, vsi->rss_hkey_user, I40EVF_HKEY_ARRAY_SIZE); else - i40evf_configure_rss_reg(adapter, seed); + netdev_rss_key_fill((void *)seed, I40EVF_HKEY_ARRAY_SIZE); + ret = i40evf_config_rss(vsi, seed, lut, I40EVF_HLUT_ARRAY_SIZE); + kfree(lut); + + return ret; } /** @@ -1327,21 +1491,22 @@ static void i40evf_configure_rss(struct i40evf_adapter *adapter) **/ static int i40evf_alloc_q_vectors(struct i40evf_adapter *adapter) { - int q_idx, num_q_vectors; + int q_idx = 0, num_q_vectors; struct i40e_q_vector *q_vector; num_q_vectors = adapter->num_msix_vectors - NONQ_VECS; + adapter->q_vectors = kcalloc(num_q_vectors, sizeof(*q_vector), + GFP_KERNEL); + if (!adapter->q_vectors) + goto err_out; for (q_idx = 0; q_idx < num_q_vectors; q_idx++) { - q_vector = kzalloc(sizeof(*q_vector), GFP_KERNEL); - if (!q_vector) - goto err_out; + q_vector = &adapter->q_vectors[q_idx]; q_vector->adapter = adapter; q_vector->vsi = &adapter->vsi; q_vector->v_idx = q_idx; netif_napi_add(adapter->netdev, &q_vector->napi, i40evf_napi_poll, NAPI_POLL_WEIGHT); - adapter->q_vector[q_idx] = q_vector; } return 0; @@ -1349,11 +1514,10 @@ static int i40evf_alloc_q_vectors(struct i40evf_adapter *adapter) err_out: while (q_idx) { q_idx--; - q_vector = adapter->q_vector[q_idx]; + q_vector = &adapter->q_vectors[q_idx]; netif_napi_del(&q_vector->napi); - kfree(q_vector); - adapter->q_vector[q_idx] = NULL; } + kfree(adapter->q_vectors); return -ENOMEM; } @@ -1374,13 +1538,11 @@ static void i40evf_free_q_vectors(struct i40evf_adapter *adapter) napi_vectors = adapter->num_active_queues; for (q_idx = 0; q_idx < num_q_vectors; q_idx++) { - struct i40e_q_vector *q_vector = adapter->q_vector[q_idx]; - - adapter->q_vector[q_idx] = NULL; + struct i40e_q_vector *q_vector = &adapter->q_vectors[q_idx]; if (q_idx < napi_vectors) netif_napi_del(&q_vector->napi); - kfree(q_vector); } + kfree(adapter->q_vectors); } /** @@ -1439,6 +1601,22 @@ err_set_interrupt: } /** + * i40evf_clear_rss_config_user - Clear user configurations of RSS + * @vsi: Pointer to VSI structure + **/ +static void i40evf_clear_rss_config_user(struct i40e_vsi *vsi) +{ + if (!vsi) + return; + + kfree(vsi->rss_hkey_user); + vsi->rss_hkey_user = NULL; + + kfree(vsi->rss_lut_user); + vsi->rss_lut_user = NULL; +} + +/** * i40evf_watchdog_timer - Periodic call-back timer * @data: pointer to adapter disguised as unsigned long **/ @@ -1565,7 +1743,7 @@ static void i40evf_watchdog_task(struct work_struct *work) * PF, so we don't have to set current_op as we will * not get a response through the ARQ. */ - i40evf_configure_rss(adapter); + i40evf_init_rss(adapter); adapter->aq_required &= ~I40EVF_FLAG_AQ_CONFIGURE_RSS; goto watchdog_done; } @@ -1864,9 +2042,12 @@ void i40evf_free_all_tx_resources(struct i40evf_adapter *adapter) { int i; + if (!adapter->tx_rings) + return; + for (i = 0; i < adapter->num_active_queues; i++) - if (adapter->tx_rings[i]->desc) - i40evf_free_tx_resources(adapter->tx_rings[i]); + if (adapter->tx_rings[i].desc) + i40evf_free_tx_resources(&adapter->tx_rings[i]); } /** @@ -1884,8 +2065,8 @@ static int i40evf_setup_all_tx_resources(struct i40evf_adapter *adapter) int i, err = 0; for (i = 0; i < adapter->num_active_queues; i++) { - adapter->tx_rings[i]->count = adapter->tx_desc_count; - err = i40evf_setup_tx_descriptors(adapter->tx_rings[i]); + adapter->tx_rings[i].count = adapter->tx_desc_count; + err = i40evf_setup_tx_descriptors(&adapter->tx_rings[i]); if (!err) continue; dev_err(&adapter->pdev->dev, @@ -1911,8 +2092,8 @@ static int i40evf_setup_all_rx_resources(struct i40evf_adapter *adapter) int i, err = 0; for (i = 0; i < adapter->num_active_queues; i++) { - adapter->rx_rings[i]->count = adapter->rx_desc_count; - err = i40evf_setup_rx_descriptors(adapter->rx_rings[i]); + adapter->rx_rings[i].count = adapter->rx_desc_count; + err = i40evf_setup_rx_descriptors(&adapter->rx_rings[i]); if (!err) continue; dev_err(&adapter->pdev->dev, @@ -1932,9 +2113,12 @@ void i40evf_free_all_rx_resources(struct i40evf_adapter *adapter) { int i; + if (!adapter->rx_rings) + return; + for (i = 0; i < adapter->num_active_queues; i++) - if (adapter->rx_rings[i]->desc) - i40evf_free_rx_resources(adapter->rx_rings[i]); + if (adapter->rx_rings[i].desc) + i40evf_free_rx_resources(&adapter->rx_rings[i]); } /** @@ -2137,7 +2321,7 @@ int i40evf_process_config(struct i40evf_adapter *adapter) netdev->features |= NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_IP_CSUM | - NETIF_F_SCTP_CSUM | + NETIF_F_SCTP_CRC | NETIF_F_IPV6_CSUM | NETIF_F_TSO | NETIF_F_TSO6 | @@ -2263,6 +2447,14 @@ static void i40evf_init_task(struct work_struct *work) if (err == I40E_ERR_ADMIN_QUEUE_NO_WORK) { err = i40evf_send_vf_config_msg(adapter); goto err; + } else if (err == I40E_ERR_PARAM) { + /* We only get ERR_PARAM if the device is in a very bad + * state or if we've been disabled for previous bad + * behavior. Either way, we're done now. + */ + i40evf_shutdown_adminq(hw); + dev_err(&pdev->dev, "Unable to get VF config due to PF error condition, not retrying\n"); + return; } if (err) { dev_err(&pdev->dev, "Unable to get VF config (%d)\n", @@ -2313,7 +2505,7 @@ static void i40evf_init_task(struct work_struct *work) I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR) adapter->flags |= I40EVF_FLAG_WB_ON_ITR_CAPABLE; if (!RSS_AQ(adapter)) - i40evf_configure_rss(adapter); + i40evf_init_rss(adapter); err = i40evf_request_misc_irq(adapter); if (err) goto err_sw_init; @@ -2334,7 +2526,6 @@ static void i40evf_init_task(struct work_struct *work) if (netdev->features & NETIF_F_GRO) dev_info(&pdev->dev, "GRO is enabled\n"); - dev_info(&pdev->dev, "%s\n", i40evf_driver_string); adapter->state = __I40EVF_DOWN; set_bit(__I40E_DOWN, &adapter->vsi.state); i40evf_misc_irq_enable(adapter); @@ -2343,7 +2534,7 @@ static void i40evf_init_task(struct work_struct *work) adapter->aq_required |= I40EVF_FLAG_AQ_CONFIGURE_RSS; mod_timer_pending(&adapter->watchdog_timer, jiffies + 1); } else { - i40evf_configure_rss(adapter); + i40evf_init_rss(adapter); } return; restart: @@ -2438,8 +2629,7 @@ static int i40evf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_master(pdev); - netdev = alloc_etherdev_mq(sizeof(struct i40evf_adapter), - MAX_TX_QUEUES); + netdev = alloc_etherdev_mq(sizeof(struct i40evf_adapter), MAX_QUEUES); if (!netdev) { err = -ENOMEM; goto err_alloc_etherdev; @@ -2632,6 +2822,9 @@ static void i40evf_remove(struct pci_dev *pdev) flush_scheduled_work(); + /* Clear user configurations for RSS */ + i40evf_clear_rss_config_user(&adapter->vsi); + if (hw->aq.asq.count) i40evf_shutdown_adminq(hw); diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c index 32e620e1eb5c..c1c526283757 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c @@ -157,7 +157,9 @@ int i40evf_send_vf_config_msg(struct i40evf_adapter *adapter) I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ | I40E_VIRTCHNL_VF_OFFLOAD_RSS_REG | I40E_VIRTCHNL_VF_OFFLOAD_VLAN | - I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR; + I40E_VIRTCHNL_VF_OFFLOAD_WB_ON_ITR | + I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2; + adapter->current_op = I40E_VIRTCHNL_OP_GET_VF_RESOURCES; adapter->aq_required &= ~I40EVF_FLAG_AQ_GET_CONFIG; if (PF_IS_V11(adapter)) @@ -242,7 +244,7 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter) adapter->current_op = I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES; len = sizeof(struct i40e_virtchnl_vsi_queue_config_info) + (sizeof(struct i40e_virtchnl_queue_pair_info) * pairs); - vqci = kzalloc(len, GFP_ATOMIC); + vqci = kzalloc(len, GFP_KERNEL); if (!vqci) return; @@ -255,19 +257,19 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter) for (i = 0; i < pairs; i++) { vqpi->txq.vsi_id = vqci->vsi_id; vqpi->txq.queue_id = i; - vqpi->txq.ring_len = adapter->tx_rings[i]->count; - vqpi->txq.dma_ring_addr = adapter->tx_rings[i]->dma; + vqpi->txq.ring_len = adapter->tx_rings[i].count; + vqpi->txq.dma_ring_addr = adapter->tx_rings[i].dma; vqpi->txq.headwb_enabled = 1; vqpi->txq.dma_headwb_addr = vqpi->txq.dma_ring_addr + (vqpi->txq.ring_len * sizeof(struct i40e_tx_desc)); vqpi->rxq.vsi_id = vqci->vsi_id; vqpi->rxq.queue_id = i; - vqpi->rxq.ring_len = adapter->rx_rings[i]->count; - vqpi->rxq.dma_ring_addr = adapter->rx_rings[i]->dma; + vqpi->rxq.ring_len = adapter->rx_rings[i].count; + vqpi->rxq.dma_ring_addr = adapter->rx_rings[i].dma; vqpi->rxq.max_pkt_size = adapter->netdev->mtu + ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN; - vqpi->rxq.databuffer_size = adapter->rx_rings[i]->rx_buf_len; + vqpi->rxq.databuffer_size = adapter->rx_rings[i].rx_buf_len; vqpi++; } @@ -353,14 +355,14 @@ void i40evf_map_queues(struct i40evf_adapter *adapter) len = sizeof(struct i40e_virtchnl_irq_map_info) + (adapter->num_msix_vectors * sizeof(struct i40e_virtchnl_vector_map)); - vimi = kzalloc(len, GFP_ATOMIC); + vimi = kzalloc(len, GFP_KERNEL); if (!vimi) return; vimi->num_vectors = adapter->num_msix_vectors; /* Queue vectors first */ for (v_idx = 0; v_idx < q_vectors; v_idx++) { - q_vector = adapter->q_vector[v_idx]; + q_vector = adapter->q_vectors + v_idx; vimi->vecmap[v_idx].vsi_id = adapter->vsi_res->vsi_id; vimi->vecmap[v_idx].vector_id = v_idx + NONQ_VECS; vimi->vecmap[v_idx].txq_map = q_vector->ring_mask; @@ -391,6 +393,7 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter) struct i40e_virtchnl_ether_addr_list *veal; int len, i = 0, count = 0; struct i40evf_mac_filter *f; + bool more = false; if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) { /* bail because we already have a command pending */ @@ -415,10 +418,12 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter) count = (I40EVF_MAX_AQ_BUF_SIZE - sizeof(struct i40e_virtchnl_ether_addr_list)) / sizeof(struct i40e_virtchnl_ether_addr); - len = I40EVF_MAX_AQ_BUF_SIZE; + len = sizeof(struct i40e_virtchnl_ether_addr_list) + + (count * sizeof(struct i40e_virtchnl_ether_addr)); + more = true; } - veal = kzalloc(len, GFP_ATOMIC); + veal = kzalloc(len, GFP_KERNEL); if (!veal) return; @@ -431,7 +436,8 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter) f->add = false; } } - adapter->aq_required &= ~I40EVF_FLAG_AQ_ADD_MAC_FILTER; + if (!more) + adapter->aq_required &= ~I40EVF_FLAG_AQ_ADD_MAC_FILTER; i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS, (u8 *)veal, len); kfree(veal); @@ -450,6 +456,7 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter) struct i40e_virtchnl_ether_addr_list *veal; struct i40evf_mac_filter *f, *ftmp; int len, i = 0, count = 0; + bool more = false; if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) { /* bail because we already have a command pending */ @@ -474,9 +481,11 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter) count = (I40EVF_MAX_AQ_BUF_SIZE - sizeof(struct i40e_virtchnl_ether_addr_list)) / sizeof(struct i40e_virtchnl_ether_addr); - len = I40EVF_MAX_AQ_BUF_SIZE; + len = sizeof(struct i40e_virtchnl_ether_addr_list) + + (count * sizeof(struct i40e_virtchnl_ether_addr)); + more = true; } - veal = kzalloc(len, GFP_ATOMIC); + veal = kzalloc(len, GFP_KERNEL); if (!veal) return; @@ -490,7 +499,8 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter) kfree(f); } } - adapter->aq_required &= ~I40EVF_FLAG_AQ_DEL_MAC_FILTER; + if (!more) + adapter->aq_required &= ~I40EVF_FLAG_AQ_DEL_MAC_FILTER; i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS, (u8 *)veal, len); kfree(veal); @@ -509,6 +519,7 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter) struct i40e_virtchnl_vlan_filter_list *vvfl; int len, i = 0, count = 0; struct i40evf_vlan_filter *f; + bool more = false; if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) { /* bail because we already have a command pending */ @@ -534,9 +545,11 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter) count = (I40EVF_MAX_AQ_BUF_SIZE - sizeof(struct i40e_virtchnl_vlan_filter_list)) / sizeof(u16); - len = I40EVF_MAX_AQ_BUF_SIZE; + len = sizeof(struct i40e_virtchnl_vlan_filter_list) + + (count * sizeof(u16)); + more = true; } - vvfl = kzalloc(len, GFP_ATOMIC); + vvfl = kzalloc(len, GFP_KERNEL); if (!vvfl) return; @@ -549,7 +562,8 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter) f->add = false; } } - adapter->aq_required &= ~I40EVF_FLAG_AQ_ADD_VLAN_FILTER; + if (!more) + adapter->aq_required &= ~I40EVF_FLAG_AQ_ADD_VLAN_FILTER; i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_ADD_VLAN, (u8 *)vvfl, len); kfree(vvfl); } @@ -567,6 +581,7 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter) struct i40e_virtchnl_vlan_filter_list *vvfl; struct i40evf_vlan_filter *f, *ftmp; int len, i = 0, count = 0; + bool more = false; if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) { /* bail because we already have a command pending */ @@ -592,9 +607,11 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter) count = (I40EVF_MAX_AQ_BUF_SIZE - sizeof(struct i40e_virtchnl_vlan_filter_list)) / sizeof(u16); - len = I40EVF_MAX_AQ_BUF_SIZE; + len = sizeof(struct i40e_virtchnl_vlan_filter_list) + + (count * sizeof(u16)); + more = true; } - vvfl = kzalloc(len, GFP_ATOMIC); + vvfl = kzalloc(len, GFP_KERNEL); if (!vvfl) return; @@ -608,7 +625,8 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter) kfree(f); } } - adapter->aq_required &= ~I40EVF_FLAG_AQ_DEL_VLAN_FILTER; + if (!more) + adapter->aq_required &= ~I40EVF_FLAG_AQ_DEL_VLAN_FILTER; i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_DEL_VLAN, (u8 *)vvfl, len); kfree(vvfl); } @@ -724,9 +742,29 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter, return; } if (v_retval) { - dev_err(&adapter->pdev->dev, "PF returned error %d (%s) to our request %d\n", - v_retval, i40evf_stat_str(&adapter->hw, v_retval), - v_opcode); + switch (v_opcode) { + case I40E_VIRTCHNL_OP_ADD_VLAN: + dev_err(&adapter->pdev->dev, "Failed to add VLAN filter, error %s\n", + i40evf_stat_str(&adapter->hw, v_retval)); + break; + case I40E_VIRTCHNL_OP_ADD_ETHER_ADDRESS: + dev_err(&adapter->pdev->dev, "Failed to add MAC filter, error %s\n", + i40evf_stat_str(&adapter->hw, v_retval)); + break; + case I40E_VIRTCHNL_OP_DEL_VLAN: + dev_err(&adapter->pdev->dev, "Failed to delete VLAN filter, error %s\n", + i40evf_stat_str(&adapter->hw, v_retval)); + break; + case I40E_VIRTCHNL_OP_DEL_ETHER_ADDRESS: + dev_err(&adapter->pdev->dev, "Failed to delete MAC filter, error %s\n", + i40evf_stat_str(&adapter->hw, v_retval)); + break; + default: + dev_err(&adapter->pdev->dev, "PF returned error %d (%s) to our request %d\n", + v_retval, + i40evf_stat_str(&adapter->hw, v_retval), + v_opcode); + } } switch (v_opcode) { case I40E_VIRTCHNL_OP_GET_STATS: { |