From 2950219d87b040959f23484dd4ff5856bf1bc172 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Mon, 26 Oct 2015 11:02:29 +0200 Subject: [PATCH] qede: Add basic network device support This patch includes the basic Rx/Tx support for the driver [although carrier will still never be turned on]. Following this patch the driver registers a network device, initializes it and prepares it for traffic. Signed-off-by: Sudarsana Kalluru Signed-off-by: Yuval Mintz Signed-off-by: Ariel Elior Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qede/qede.h | 128 ++ drivers/net/ethernet/qlogic/qede/qede_main.c | 1807 ++++++++++++++++++++++++++ 2 files changed, 1935 insertions(+) diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 7e2bcfa..424ef4a 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -51,6 +51,7 @@ struct qede_dev { #define QEDE_MAX_TSS_CNT(edev) ((edev)->dev_info.num_queues * \ (edev)->dev_info.num_tc) + struct qede_fastpath *fp_array; u16 num_rss; u8 num_tc; #define QEDE_RSS_CNT(edev) ((edev)->num_rss) @@ -58,6 +59,9 @@ struct qede_dev { (edev)->num_tc) #define QEDE_TSS_IDX(edev, txqidx) ((txqidx) % (edev)->num_rss) #define QEDE_TC_IDX(edev, txqidx) ((txqidx) / (edev)->num_rss) +#define QEDE_TX_QUEUE(edev, txqidx) \ + (&(edev)->fp_array[QEDE_TSS_IDX((edev), (txqidx))].txqs[QEDE_TC_IDX( \ + (edev), (txqidx))]) struct qed_int_info int_info; unsigned char primary_mac[ETH_ALEN]; @@ -65,9 +69,133 @@ struct qede_dev { /* Smaller private varaiant of the RTNL lock */ struct mutex qede_lock; u32 state; /* Protected by qede_lock */ + u16 rx_buf_size; + /* L2 header size + 2*VLANs (8 bytes) + LLC SNAP (8 bytes) */ +#define ETH_OVERHEAD (ETH_HLEN + 8 + 8) + /* Max supported alignment is 256 (8 shift) + * minimal alignment shift 6 is optimal for 57xxx HW performance + */ +#define QEDE_RX_ALIGN_SHIFT max(6, min(8, L1_CACHE_SHIFT)) + /* We assume skb_build() uses sizeof(struct skb_shared_info) bytes + * at the end of skb->data, to avoid wasting a full cache line. + * This reduces memory use (skb->truesize). + */ +#define QEDE_FW_RX_ALIGN_END \ + max_t(u64, 1UL << QEDE_RX_ALIGN_SHIFT, \ + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) + + struct qed_update_vport_rss_params rss_params; + u16 q_num_rx_buffers; /* Must be a power of two */ + u16 q_num_tx_buffers; /* Must be a power of two */ +}; + +enum QEDE_STATE { + QEDE_STATE_CLOSED, + QEDE_STATE_OPEN, +}; + +#define HILO_U64(hi, lo) ((((u64)(hi)) << 32) + (lo)) + +#define MAX_NUM_TC 8 +#define MAX_NUM_PRI 8 + +/* The driver supports the new build_skb() API: + * RX ring buffer contains pointer to kmalloc() data only, + * skb are built only after the frame was DMA-ed. + */ +struct sw_rx_data { + u8 *data; + + DEFINE_DMA_UNMAP_ADDR(mapping); +}; + +struct qede_rx_queue { + __le16 *hw_cons_ptr; + struct sw_rx_data *sw_rx_ring; + u16 sw_rx_cons; + u16 sw_rx_prod; + struct qed_chain rx_bd_ring; + struct qed_chain rx_comp_ring; + void __iomem *hw_rxq_prod_addr; + + int rx_buf_size; + + u16 num_rx_buffers; + u16 rxq_id; + + u64 rx_hw_errors; + u64 rx_alloc_errors; +}; + +union db_prod { + struct eth_db_data data; + u32 raw; +}; + +struct sw_tx_bd { + struct sk_buff *skb; + u8 flags; +/* Set on the first BD descriptor when there is a split BD */ +#define QEDE_TSO_SPLIT_BD BIT(0) +}; + +struct qede_tx_queue { + int index; /* Queue index */ + __le16 *hw_cons_ptr; + struct sw_tx_bd *sw_tx_ring; + u16 sw_tx_cons; + u16 sw_tx_prod; + struct qed_chain tx_pbl; + void __iomem *doorbell_addr; + union db_prod tx_db; + + u16 num_tx_buffers; +}; + +#define BD_UNMAP_ADDR(bd) HILO_U64(le32_to_cpu((bd)->addr.hi), \ + le32_to_cpu((bd)->addr.lo)) +#define BD_SET_UNMAP_ADDR_LEN(bd, maddr, len) \ + do { \ + (bd)->addr.hi = cpu_to_le32(upper_32_bits(maddr)); \ + (bd)->addr.lo = cpu_to_le32(lower_32_bits(maddr)); \ + (bd)->nbytes = cpu_to_le16(len); \ + } while (0) +#define BD_UNMAP_LEN(bd) (le16_to_cpu((bd)->nbytes)) + +struct qede_fastpath { + struct qede_dev *edev; + u8 rss_id; + struct napi_struct napi; + struct qed_sb_info *sb_info; + struct qede_rx_queue *rxq; + struct qede_tx_queue *txqs; + +#define VEC_NAME_SIZE (sizeof(((struct net_device *)0)->name) + 8) + char name[VEC_NAME_SIZE]; }; /* Debug print definitions */ #define DP_NAME(edev) ((edev)->ndev->name) +#define XMIT_PLAIN 0 +#define XMIT_L4_CSUM BIT(0) +#define XMIT_LSO BIT(1) +#define XMIT_ENC BIT(2) + +#define QEDE_CSUM_ERROR BIT(0) +#define QEDE_CSUM_UNNECESSARY BIT(1) +#define RX_RING_SIZE_POW 13 +#define RX_RING_SIZE BIT(RX_RING_SIZE_POW) +#define NUM_RX_BDS_MAX (RX_RING_SIZE - 1) +#define NUM_RX_BDS_MIN 128 +#define NUM_RX_BDS_DEF NUM_RX_BDS_MAX + +#define TX_RING_SIZE_POW 13 +#define TX_RING_SIZE BIT(TX_RING_SIZE_POW) +#define NUM_TX_BDS_MAX (TX_RING_SIZE - 1) +#define NUM_TX_BDS_MIN 128 +#define NUM_TX_BDS_DEF NUM_TX_BDS_MAX + +#define for_each_rss(i) for (i = 0; i < edev->num_rss; i++) + #endif /* _QEDE_H_ */ diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 02ed6db..daba118 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -85,6 +85,8 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id); #define TX_TIMEOUT (5 * HZ) static void qede_remove(struct pci_dev *pdev); +static int qede_alloc_rx_buffer(struct qede_dev *edev, + struct qede_rx_queue *rxq); static struct pci_driver qede_pci_driver = { .name = "qede", @@ -93,6 +95,41 @@ static struct pci_driver qede_pci_driver = { .remove = qede_remove, }; +static int qede_netdev_event(struct notifier_block *this, unsigned long event, + void *ptr) +{ + struct net_device *ndev = netdev_notifier_info_to_dev(ptr); + struct ethtool_drvinfo drvinfo; + struct qede_dev *edev; + + /* Currently only support name change */ + if (event != NETDEV_CHANGENAME) + goto done; + + /* Check whether this is a qede device */ + if (!ndev || !ndev->ethtool_ops || !ndev->ethtool_ops->get_drvinfo) + goto done; + + memset(&drvinfo, 0, sizeof(drvinfo)); + ndev->ethtool_ops->get_drvinfo(ndev, &drvinfo); + if (strcmp(drvinfo.driver, "qede")) + goto done; + edev = netdev_priv(ndev); + + /* Notify qed of the name change */ + if (!edev->ops || !edev->ops->common) + goto done; + edev->ops->common->set_id(edev->cdev, edev->ndev->name, + "qede"); + +done: + return NOTIFY_DONE; +} + +static struct notifier_block qede_netdev_notifier = { + .notifier_call = qede_netdev_event, +}; + static int __init qede_init(void) { @@ -115,9 +152,20 @@ int __init qede_init(void) return -EINVAL; } + /* Must register notifier before pci ops, since we might miss + * interface rename after pci probe and netdev registeration. + */ + ret = register_netdevice_notifier(&qede_netdev_notifier); + if (ret) { + pr_notice("Failed to register netdevice_notifier\n"); + qed_put_eth_ops(); + return -EINVAL; + } + ret = pci_register_driver(&qede_pci_driver); if (ret) { pr_notice("Failed to register driver\n"); + unregister_netdevice_notifier(&qede_netdev_notifier); qed_put_eth_ops(); return -EINVAL; } @@ -129,6 +177,7 @@ static void __exit qede_cleanup(void) { pr_notice("qede_cleanup called\n"); + unregister_netdevice_notifier(&qede_netdev_notifier); pci_unregister_driver(&qede_pci_driver); qed_put_eth_ops(); } @@ -137,6 +186,858 @@ module_init(qede_init); module_exit(qede_cleanup); /* ------------------------------------------------------------------------- + * START OF FAST-PATH + * ------------------------------------------------------------------------- + */ + +/* Unmap the data and free skb */ +static int qede_free_tx_pkt(struct qede_dev *edev, + struct qede_tx_queue *txq, + int *len) +{ + u16 idx = txq->sw_tx_cons & NUM_TX_BDS_MAX; + struct sk_buff *skb = txq->sw_tx_ring[idx].skb; + struct eth_tx_1st_bd *first_bd; + struct eth_tx_bd *tx_data_bd; + int bds_consumed = 0; + int nbds; + bool data_split = txq->sw_tx_ring[idx].flags & QEDE_TSO_SPLIT_BD; + int i, split_bd_len = 0; + + if (unlikely(!skb)) { + DP_ERR(edev, + "skb is null for txq idx=%d txq->sw_tx_cons=%d txq->sw_tx_prod=%d\n", + idx, txq->sw_tx_cons, txq->sw_tx_prod); + return -1; + } + + *len = skb->len; + + first_bd = (struct eth_tx_1st_bd *)qed_chain_consume(&txq->tx_pbl); + + bds_consumed++; + + nbds = first_bd->data.nbds; + + if (data_split) { + struct eth_tx_bd *split = (struct eth_tx_bd *) + qed_chain_consume(&txq->tx_pbl); + split_bd_len = BD_UNMAP_LEN(split); + bds_consumed++; + } + dma_unmap_page(&edev->pdev->dev, BD_UNMAP_ADDR(first_bd), + BD_UNMAP_LEN(first_bd) + split_bd_len, DMA_TO_DEVICE); + + /* Unmap the data of the skb frags */ + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++, bds_consumed++) { + tx_data_bd = (struct eth_tx_bd *) + qed_chain_consume(&txq->tx_pbl); + dma_unmap_page(&edev->pdev->dev, BD_UNMAP_ADDR(tx_data_bd), + BD_UNMAP_LEN(tx_data_bd), DMA_TO_DEVICE); + } + + while (bds_consumed++ < nbds) + qed_chain_consume(&txq->tx_pbl); + + /* Free skb */ + dev_kfree_skb_any(skb); + txq->sw_tx_ring[idx].skb = NULL; + txq->sw_tx_ring[idx].flags = 0; + + return 0; +} + +/* Unmap the data and free skb when mapping failed during start_xmit */ +static void qede_free_failed_tx_pkt(struct qede_dev *edev, + struct qede_tx_queue *txq, + struct eth_tx_1st_bd *first_bd, + int nbd, + bool data_split) +{ + u16 idx = txq->sw_tx_prod & NUM_TX_BDS_MAX; + struct sk_buff *skb = txq->sw_tx_ring[idx].skb; + struct eth_tx_bd *tx_data_bd; + int i, split_bd_len = 0; + + /* Return prod to its position before this skb was handled */ + qed_chain_set_prod(&txq->tx_pbl, + le16_to_cpu(txq->tx_db.data.bd_prod), + first_bd); + + first_bd = (struct eth_tx_1st_bd *)qed_chain_produce(&txq->tx_pbl); + + if (data_split) { + struct eth_tx_bd *split = (struct eth_tx_bd *) + qed_chain_produce(&txq->tx_pbl); + split_bd_len = BD_UNMAP_LEN(split); + nbd--; + } + + dma_unmap_page(&edev->pdev->dev, BD_UNMAP_ADDR(first_bd), + BD_UNMAP_LEN(first_bd) + split_bd_len, DMA_TO_DEVICE); + + /* Unmap the data of the skb frags */ + for (i = 0; i < nbd; i++) { + tx_data_bd = (struct eth_tx_bd *) + qed_chain_produce(&txq->tx_pbl); + if (tx_data_bd->nbytes) + dma_unmap_page(&edev->pdev->dev, + BD_UNMAP_ADDR(tx_data_bd), + BD_UNMAP_LEN(tx_data_bd), DMA_TO_DEVICE); + } + + /* Return again prod to its position before this skb was handled */ + qed_chain_set_prod(&txq->tx_pbl, + le16_to_cpu(txq->tx_db.data.bd_prod), + first_bd); + + /* Free skb */ + dev_kfree_skb_any(skb); + txq->sw_tx_ring[idx].skb = NULL; + txq->sw_tx_ring[idx].flags = 0; +} + +static u32 qede_xmit_type(struct qede_dev *edev, + struct sk_buff *skb, + int *ipv6_ext) +{ + u32 rc = XMIT_L4_CSUM; + __be16 l3_proto; + + if (skb->ip_summed != CHECKSUM_PARTIAL) + return XMIT_PLAIN; + + l3_proto = vlan_get_protocol(skb); + if (l3_proto == htons(ETH_P_IPV6) && + (ipv6_hdr(skb)->nexthdr == NEXTHDR_IPV6)) + *ipv6_ext = 1; + + if (skb_is_gso(skb)) + rc |= XMIT_LSO; + + return rc; +} + +static void qede_set_params_for_ipv6_ext(struct sk_buff *skb, + struct eth_tx_2nd_bd *second_bd, + struct eth_tx_3rd_bd *third_bd) +{ + u8 l4_proto; + u16 bd2_bits = 0, bd2_bits2 = 0; + + bd2_bits2 |= (1 << ETH_TX_DATA_2ND_BD_IPV6_EXT_SHIFT); + + bd2_bits |= ((((u8 *)skb_transport_header(skb) - skb->data) >> 1) & + ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_MASK) + << ETH_TX_DATA_2ND_BD_L4_HDR_START_OFFSET_W_SHIFT; + + bd2_bits2 |= (ETH_L4_PSEUDO_CSUM_CORRECT_LENGTH << + ETH_TX_DATA_2ND_BD_L4_PSEUDO_CSUM_MODE_SHIFT); + + if (vlan_get_protocol(skb) == htons(ETH_P_IPV6)) + l4_proto = ipv6_hdr(skb)->nexthdr; + else + l4_proto = ip_hdr(skb)->protocol; + + if (l4_proto == IPPROTO_UDP) + bd2_bits2 |= 1 << ETH_TX_DATA_2ND_BD_L4_UDP_SHIFT; + + if (third_bd) { + third_bd->data.bitfields |= + ((tcp_hdrlen(skb) / 4) & + ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_MASK) << + ETH_TX_DATA_3RD_BD_TCP_HDR_LEN_DW_SHIFT; + } + + second_bd->data.bitfields = cpu_to_le16(bd2_bits); + second_bd->data.bitfields2 = cpu_to_le16(bd2_bits2); +} + +static int map_frag_to_bd(struct qede_dev *edev, + skb_frag_t *frag, + struct eth_tx_bd *bd) +{ + dma_addr_t mapping; + + /* Map skb non-linear frag data for DMA */ + mapping = skb_frag_dma_map(&edev->pdev->dev, frag, 0, + skb_frag_size(frag), + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) { + DP_NOTICE(edev, "Unable to map frag - dropping packet\n"); + return -ENOMEM; + } + + /* Setup the data pointer of the frag data */ + BD_SET_UNMAP_ADDR_LEN(bd, mapping, skb_frag_size(frag)); + + return 0; +} + +/* Main transmit function */ +static +netdev_tx_t qede_start_xmit(struct sk_buff *skb, + struct net_device *ndev) +{ + struct qede_dev *edev = netdev_priv(ndev); + struct netdev_queue *netdev_txq; + struct qede_tx_queue *txq; + struct eth_tx_1st_bd *first_bd; + struct eth_tx_2nd_bd *second_bd = NULL; + struct eth_tx_3rd_bd *third_bd = NULL; + struct eth_tx_bd *tx_data_bd = NULL; + u16 txq_index; + u8 nbd = 0; + dma_addr_t mapping; + int rc, frag_idx = 0, ipv6_ext = 0; + u8 xmit_type; + u16 idx; + u16 hlen; + bool data_split; + + /* Get tx-queue context and netdev index */ + txq_index = skb_get_queue_mapping(skb); + WARN_ON(txq_index >= QEDE_TSS_CNT(edev)); + txq = QEDE_TX_QUEUE(edev, txq_index); + netdev_txq = netdev_get_tx_queue(ndev, txq_index); + + /* Current code doesn't support SKB linearization, since the max number + * of skb frags can be passed in the FW HSI. + */ + BUILD_BUG_ON(MAX_SKB_FRAGS > ETH_TX_MAX_BDS_PER_NON_LSO_PACKET); + + WARN_ON(qed_chain_get_elem_left(&txq->tx_pbl) < + (MAX_SKB_FRAGS + 1)); + + xmit_type = qede_xmit_type(edev, skb, &ipv6_ext); + + /* Fill the entry in the SW ring and the BDs in the FW ring */ + idx = txq->sw_tx_prod & NUM_TX_BDS_MAX; + txq->sw_tx_ring[idx].skb = skb; + first_bd = (struct eth_tx_1st_bd *) + qed_chain_produce(&txq->tx_pbl); + memset(first_bd, 0, sizeof(*first_bd)); + first_bd->data.bd_flags.bitfields = + 1 << ETH_TX_1ST_BD_FLAGS_START_BD_SHIFT; + + /* Map skb linear data for DMA and set in the first BD */ + mapping = dma_map_single(&edev->pdev->dev, skb->data, + skb_headlen(skb), DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) { + DP_NOTICE(edev, "SKB mapping failed\n"); + qede_free_failed_tx_pkt(edev, txq, first_bd, 0, false); + return NETDEV_TX_OK; + } + nbd++; + BD_SET_UNMAP_ADDR_LEN(first_bd, mapping, skb_headlen(skb)); + + /* In case there is IPv6 with extension headers or LSO we need 2nd and + * 3rd BDs. + */ + if (unlikely((xmit_type & XMIT_LSO) | ipv6_ext)) { + second_bd = (struct eth_tx_2nd_bd *) + qed_chain_produce(&txq->tx_pbl); + memset(second_bd, 0, sizeof(*second_bd)); + + nbd++; + third_bd = (struct eth_tx_3rd_bd *) + qed_chain_produce(&txq->tx_pbl); + memset(third_bd, 0, sizeof(*third_bd)); + + nbd++; + /* We need to fill in additional data in second_bd... */ + tx_data_bd = (struct eth_tx_bd *)second_bd; + } + + if (skb_vlan_tag_present(skb)) { + first_bd->data.vlan = cpu_to_le16(skb_vlan_tag_get(skb)); + first_bd->data.bd_flags.bitfields |= + 1 << ETH_TX_1ST_BD_FLAGS_VLAN_INSERTION_SHIFT; + } + + /* Fill the parsing flags & params according to the requested offload */ + if (xmit_type & XMIT_L4_CSUM) { + /* We don't re-calculate IP checksum as it is already done by + * the upper stack + */ + first_bd->data.bd_flags.bitfields |= + 1 << ETH_TX_1ST_BD_FLAGS_L4_CSUM_SHIFT; + + /* If the packet is IPv6 with extension header, indicate that + * to FW and pass few params, since the device cracker doesn't + * support parsing IPv6 with extension header/s. + */ + if (unlikely(ipv6_ext)) + qede_set_params_for_ipv6_ext(skb, second_bd, third_bd); + } + + if (xmit_type & XMIT_LSO) { + first_bd->data.bd_flags.bitfields |= + (1 << ETH_TX_1ST_BD_FLAGS_LSO_SHIFT); + third_bd->data.lso_mss = + cpu_to_le16(skb_shinfo(skb)->gso_size); + + first_bd->data.bd_flags.bitfields |= + 1 << ETH_TX_1ST_BD_FLAGS_IP_CSUM_SHIFT; + hlen = skb_transport_header(skb) + + tcp_hdrlen(skb) - skb->data; + + /* @@@TBD - if will not be removed need to check */ + third_bd->data.bitfields |= + (1 << ETH_TX_DATA_3RD_BD_HDR_NBD_SHIFT); + + /* Make life easier for FW guys who can't deal with header and + * data on same BD. If we need to split, use the second bd... + */ + if (unlikely(skb_headlen(skb) > hlen)) { + DP_VERBOSE(edev, NETIF_MSG_TX_QUEUED, + "TSO split header size is %d (%x:%x)\n", + first_bd->nbytes, first_bd->addr.hi, + first_bd->addr.lo); + + mapping = HILO_U64(le32_to_cpu(first_bd->addr.hi), + le32_to_cpu(first_bd->addr.lo)) + + hlen; + + BD_SET_UNMAP_ADDR_LEN(tx_data_bd, mapping, + le16_to_cpu(first_bd->nbytes) - + hlen); + + /* this marks the BD as one that has no + * individual mapping + */ + txq->sw_tx_ring[idx].flags |= QEDE_TSO_SPLIT_BD; + + first_bd->nbytes = cpu_to_le16(hlen); + + tx_data_bd = (struct eth_tx_bd *)third_bd; + data_split = true; + } + } + + /* Handle fragmented skb */ + /* special handle for frags inside 2nd and 3rd bds.. */ + while (tx_data_bd && frag_idx < skb_shinfo(skb)->nr_frags) { + rc = map_frag_to_bd(edev, + &skb_shinfo(skb)->frags[frag_idx], + tx_data_bd); + if (rc) { + qede_free_failed_tx_pkt(edev, txq, first_bd, nbd, + data_split); + return NETDEV_TX_OK; + } + + if (tx_data_bd == (struct eth_tx_bd *)second_bd) + tx_data_bd = (struct eth_tx_bd *)third_bd; + else + tx_data_bd = NULL; + + frag_idx++; + } + + /* map last frags into 4th, 5th .... */ + for (; frag_idx < skb_shinfo(skb)->nr_frags; frag_idx++, nbd++) { + tx_data_bd = (struct eth_tx_bd *) + qed_chain_produce(&txq->tx_pbl); + + memset(tx_data_bd, 0, sizeof(*tx_data_bd)); + + rc = map_frag_to_bd(edev, + &skb_shinfo(skb)->frags[frag_idx], + tx_data_bd); + if (rc) { + qede_free_failed_tx_pkt(edev, txq, first_bd, nbd, + data_split); + return NETDEV_TX_OK; + } + } + + /* update the first BD with the actual num BDs */ + first_bd->data.nbds = nbd; + + netdev_tx_sent_queue(netdev_txq, skb->len); + + skb_tx_timestamp(skb); + + /* Advance packet producer only before sending the packet since mapping + * of pages may fail. + */ + txq->sw_tx_prod++; + + /* 'next page' entries are counted in the producer value */ + txq->tx_db.data.bd_prod = + cpu_to_le16(qed_chain_get_prod_idx(&txq->tx_pbl)); + + /* wmb makes sure that the BDs data is updated before updating the + * producer, otherwise FW may read old data from the BDs. + */ + wmb(); + barrier(); + writel(txq->tx_db.raw, txq->doorbell_addr); + + /* mmiowb is needed to synchronize doorbell writes from more than one + * processor. It guarantees that the write arrives to the device before + * the queue lock is released and another start_xmit is called (possibly + * on another CPU). Without this barrier, the next doorbell can bypass + * this doorbell. This is applicable to IA64/Altix systems. + */ + mmiowb(); + + if (unlikely(qed_chain_get_elem_left(&txq->tx_pbl) + < (MAX_SKB_FRAGS + 1))) { + netif_tx_stop_queue(netdev_txq); + DP_VERBOSE(edev, NETIF_MSG_TX_QUEUED, + "Stop queue was called\n"); + /* paired memory barrier is in qede_tx_int(), we have to keep + * ordering of set_bit() in netif_tx_stop_queue() and read of + * fp->bd_tx_cons + */ + smp_mb(); + + if (qed_chain_get_elem_left(&txq->tx_pbl) + >= (MAX_SKB_FRAGS + 1) && + (edev->state == QEDE_STATE_OPEN)) { + netif_tx_wake_queue(netdev_txq); + DP_VERBOSE(edev, NETIF_MSG_TX_QUEUED, + "Wake queue was called\n"); + } + } + + return NETDEV_TX_OK; +} + +static int qede_txq_has_work(struct qede_tx_queue *txq) +{ + u16 hw_bd_cons; + + /* Tell compiler that consumer and producer can change */ + barrier(); + hw_bd_cons = le16_to_cpu(*txq->hw_cons_ptr); + if (qed_chain_get_cons_idx(&txq->tx_pbl) == hw_bd_cons + 1) + return 0; + + return hw_bd_cons != qed_chain_get_cons_idx(&txq->tx_pbl); +} + +static int qede_tx_int(struct qede_dev *edev, + struct qede_tx_queue *txq) +{ + struct netdev_queue *netdev_txq; + u16 hw_bd_cons; + unsigned int pkts_compl = 0, bytes_compl = 0; + int rc; + + netdev_txq = netdev_get_tx_queue(edev->ndev, txq->index); + + hw_bd_cons = le16_to_cpu(*txq->hw_cons_ptr); + barrier(); + + while (hw_bd_cons != qed_chain_get_cons_idx(&txq->tx_pbl)) { + int len = 0; + + rc = qede_free_tx_pkt(edev, txq, &len); + if (rc) { + DP_NOTICE(edev, "hw_bd_cons = %d, chain_cons=%d\n", + hw_bd_cons, + qed_chain_get_cons_idx(&txq->tx_pbl)); + break; + } + + bytes_compl += len; + pkts_compl++; + txq->sw_tx_cons++; + } + + netdev_tx_completed_queue(netdev_txq, pkts_compl, bytes_compl); + + /* Need to make the tx_bd_cons update visible to start_xmit() + * before checking for netif_tx_queue_stopped(). Without the + * memory barrier, there is a small possibility that + * start_xmit() will miss it and cause the queue to be stopped + * forever. + * On the other hand we need an rmb() here to ensure the proper + * ordering of bit testing in the following + * netif_tx_queue_stopped(txq) call. + */ + smp_mb(); + + if (unlikely(netif_tx_queue_stopped(netdev_txq))) { + /* Taking tx_lock is needed to prevent reenabling the queue + * while it's empty. This could have happen if rx_action() gets + * suspended in qede_tx_int() after the condition before + * netif_tx_wake_queue(), while tx_action (qede_start_xmit()): + * + * stops the queue->sees fresh tx_bd_cons->releases the queue-> + * sends some packets consuming the whole queue again-> + * stops the queue + */ + + __netif_tx_lock(netdev_txq, smp_processor_id()); + + if ((netif_tx_queue_stopped(netdev_txq)) && + (edev->state == QEDE_STATE_OPEN) && + (qed_chain_get_elem_left(&txq->tx_pbl) + >= (MAX_SKB_FRAGS + 1))) { + netif_tx_wake_queue(netdev_txq); + DP_VERBOSE(edev, NETIF_MSG_TX_DONE, + "Wake queue was called\n"); + } + + __netif_tx_unlock(netdev_txq); + } + + return 0; +} + +static bool qede_has_rx_work(struct qede_rx_queue *rxq) +{ + u16 hw_comp_cons, sw_comp_cons; + + /* Tell compiler that status block fields can change */ + barrier(); + + hw_comp_cons = le16_to_cpu(*rxq->hw_cons_ptr); + sw_comp_cons = qed_chain_get_cons_idx(&rxq->rx_comp_ring); + + return hw_comp_cons != sw_comp_cons; +} + +static bool qede_has_tx_work(struct qede_fastpath *fp) +{ + u8 tc; + + for (tc = 0; tc < fp->edev->num_tc; tc++) + if (qede_txq_has_work(&fp->txqs[tc])) + return true; + return false; +} + +/* This function copies the Rx buffer from the CONS position to the PROD + * position, since we failed to allocate a new Rx buffer. + */ +static void qede_reuse_rx_data(struct qede_rx_queue *rxq) +{ + struct eth_rx_bd *rx_bd_cons = qed_chain_consume(&rxq->rx_bd_ring); + struct eth_rx_bd *rx_bd_prod = qed_chain_produce(&rxq->rx_bd_ring); + struct sw_rx_data *sw_rx_data_cons = + &rxq->sw_rx_ring[rxq->sw_rx_cons & NUM_RX_BDS_MAX]; + struct sw_rx_data *sw_rx_data_prod = + &rxq->sw_rx_ring[rxq->sw_rx_prod & NUM_RX_BDS_MAX]; + + dma_unmap_addr_set(sw_rx_data_prod, mapping, + dma_unmap_addr(sw_rx_data_cons, mapping)); + + sw_rx_data_prod->data = sw_rx_data_cons->data; + memcpy(rx_bd_prod, rx_bd_cons, sizeof(struct eth_rx_bd)); + + rxq->sw_rx_cons++; + rxq->sw_rx_prod++; +} + +static inline void qede_update_rx_prod(struct qede_dev *edev, + struct qede_rx_queue *rxq) +{ + u16 bd_prod = qed_chain_get_prod_idx(&rxq->rx_bd_ring); + u16 cqe_prod = qed_chain_get_prod_idx(&rxq->rx_comp_ring); + struct eth_rx_prod_data rx_prods = {0}; + + /* Update producers */ + rx_prods.bd_prod = cpu_to_le16(bd_prod); + rx_prods.cqe_prod = cpu_to_le16(cqe_prod); + + /* Make sure that the BD and SGE data is updated before updating the + * producers since FW might read the BD/SGE right after the producer + * is updated. + */ + wmb(); + + internal_ram_wr(rxq->hw_rxq_prod_addr, sizeof(rx_prods), + (u32 *)&rx_prods); + + /* mmiowb is needed to synchronize doorbell writes from more than one + * processor. It guarantees that the write arrives to the device before + * the napi lock is released and another qede_poll is called (possibly + * on another CPU). Without this barrier, the next doorbell can bypass + * this doorbell. This is applicable to IA64/Altix systems. + */ + mmiowb(); +} + +static u32 qede_get_rxhash(struct qede_dev *edev, + u8 bitfields, + __le32 rss_hash, + enum pkt_hash_types *rxhash_type) +{ + enum rss_hash_type htype; + + htype = GET_FIELD(bitfields, ETH_FAST_PATH_RX_REG_CQE_RSS_HASH_TYPE); + + if ((edev->ndev->features & NETIF_F_RXHASH) && htype) { + *rxhash_type = ((htype == RSS_HASH_TYPE_IPV4) || + (htype == RSS_HASH_TYPE_IPV6)) ? + PKT_HASH_TYPE_L3 : PKT_HASH_TYPE_L4; + return le32_to_cpu(rss_hash); + } + *rxhash_type = PKT_HASH_TYPE_NONE; + return 0; +} + +static void qede_set_skb_csum(struct sk_buff *skb, u8 csum_flag) +{ + skb_checksum_none_assert(skb); + + if (csum_flag & QEDE_CSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_UNNECESSARY; +} + +static inline void qede_skb_receive(struct qede_dev *edev, + struct qede_fastpath *fp, + struct sk_buff *skb, + u16 vlan_tag) +{ + if (vlan_tag) + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), + vlan_tag); + + napi_gro_receive(&fp->napi, skb); +} + +static u8 qede_check_csum(u16 flag) +{ + u16 csum_flag = 0; + u8 csum = 0; + + if ((PARSING_AND_ERR_FLAGS_L4CHKSMWASCALCULATED_MASK << + PARSING_AND_ERR_FLAGS_L4CHKSMWASCALCULATED_SHIFT) & flag) { + csum_flag |= PARSING_AND_ERR_FLAGS_L4CHKSMERROR_MASK << + PARSING_AND_ERR_FLAGS_L4CHKSMERROR_SHIFT; + csum = QEDE_CSUM_UNNECESSARY; + } + + csum_flag |= PARSING_AND_ERR_FLAGS_IPHDRERROR_MASK << + PARSING_AND_ERR_FLAGS_IPHDRERROR_SHIFT; + + if (csum_flag & flag) + return QEDE_CSUM_ERROR; + + return csum; +} + +static int qede_rx_int(struct qede_fastpath *fp, int budget) +{ + struct qede_dev *edev = fp->edev; + struct qede_rx_queue *rxq = fp->rxq; + + u16 hw_comp_cons, sw_comp_cons, sw_rx_index, parse_flag; + int rx_pkt = 0; + u8 csum_flag; + + hw_comp_cons = le16_to_cpu(*rxq->hw_cons_ptr); + sw_comp_cons = qed_chain_get_cons_idx(&rxq->rx_comp_ring); + + /* Memory barrier to prevent the CPU from doing speculative reads of CQE + * / BD in the while-loop before reading hw_comp_cons. If the CQE is + * read before it is written by FW, then FW writes CQE and SB, and then + * the CPU reads the hw_comp_cons, it will use an old CQE. + */ + rmb(); + + /* Loop to complete all indicated BDs */ + while (sw_comp_cons != hw_comp_cons) { + struct eth_fast_path_rx_reg_cqe *fp_cqe; + enum pkt_hash_types rxhash_type; + enum eth_rx_cqe_type cqe_type; + struct sw_rx_data *sw_rx_data; + union eth_rx_cqe *cqe; + struct sk_buff *skb; + u16 len, pad; + u32 rx_hash; + u8 *data; + + /* Get the CQE from the completion ring */ + cqe = (union eth_rx_cqe *) + qed_chain_consume(&rxq->rx_comp_ring); + cqe_type = cqe->fast_path_regular.type; + + if (unlikely(cqe_type == ETH_RX_CQE_TYPE_SLOW_PATH)) { + edev->ops->eth_cqe_completion( + edev->cdev, fp->rss_id, + (struct eth_slow_path_rx_cqe *)cqe); + goto next_cqe; + } + + /* Get the data from the SW ring */ + sw_rx_index = rxq->sw_rx_cons & NUM_RX_BDS_MAX; + sw_rx_data = &rxq->sw_rx_ring[sw_rx_index]; + data = sw_rx_data->data; + + fp_cqe = &cqe->fast_path_regular; + len = le16_to_cpu(fp_cqe->pkt_len); + pad = fp_cqe->placement_offset; + + /* For every Rx BD consumed, we allocate a new BD so the BD ring + * is always with a fixed size. If allocation fails, we take the + * consumed BD and return it to the ring in the PROD position. + * The packet that was received on that BD will be dropped (and + * not passed to the upper stack). + */ + if (likely(qede_alloc_rx_buffer(edev, rxq) == 0)) { + dma_unmap_single(&edev->pdev->dev, + dma_unmap_addr(sw_rx_data, mapping), + rxq->rx_buf_size, DMA_FROM_DEVICE); + + /* If this is an error packet then drop it */ + parse_flag = + le16_to_cpu(cqe->fast_path_regular.pars_flags.flags); + csum_flag = qede_check_csum(parse_flag); + if (csum_flag == QEDE_CSUM_ERROR) { + DP_NOTICE(edev, + "CQE in CONS = %u has error, flags = %x, dropping incoming packet\n", + sw_comp_cons, parse_flag); + rxq->rx_hw_errors++; + kfree(data); + goto next_rx; + } + + skb = build_skb(data, 0); + + if (unlikely(!skb)) { + DP_NOTICE(edev, + "Build_skb failed, dropping incoming packet\n"); + kfree(data); + rxq->rx_alloc_errors++; + goto next_rx; + } + + skb_reserve(skb, pad); + + } else { + DP_NOTICE(edev, + "New buffer allocation failed, dropping incoming packet and reusing its buffer\n"); + qede_reuse_rx_data(rxq); + rxq->rx_alloc_errors++; + goto next_cqe; + } + + sw_rx_data->data = NULL; + + skb_put(skb, len); + + skb->protocol = eth_type_trans(skb, edev->ndev); + + rx_hash = qede_get_rxhash(edev, fp_cqe->bitfields, + fp_cqe->rss_hash, + &rxhash_type); + + skb_set_hash(skb, rx_hash, rxhash_type); + + qede_set_skb_csum(skb, csum_flag); + + skb_record_rx_queue(skb, fp->rss_id); + + qede_skb_receive(edev, fp, skb, le16_to_cpu(fp_cqe->vlan_tag)); + + qed_chain_consume(&rxq->rx_bd_ring); + +next_rx: + rxq->sw_rx_cons++; + rx_pkt++; + +next_cqe: /* don't consume bd rx buffer */ + qed_chain_recycle_consumed(&rxq->rx_comp_ring); + sw_comp_cons = qed_chain_get_cons_idx(&rxq->rx_comp_ring); + /* CR TPA - revisit how to handle budget in TPA perhaps + * increase on "end" + */ + if (rx_pkt == budget) + break; + } /* repeat while sw_comp_cons != hw_comp_cons... */ + + /* Update producers */ + qede_update_rx_prod(edev, rxq); + + return rx_pkt; +} + +static int qede_poll(struct napi_struct *napi, int budget) +{ + int work_done = 0; + struct qede_fastpath *fp = container_of(napi, struct qede_fastpath, + napi); + struct qede_dev *edev = fp->edev; + + while (1) { + u8 tc; + + for (tc = 0; tc < edev->num_tc; tc++) + if (qede_txq_has_work(&fp->txqs[tc])) + qede_tx_int(edev, &fp->txqs[tc]); + + if (qede_has_rx_work(fp->rxq)) { + work_done += qede_rx_int(fp, budget - work_done); + + /* must not complete if we consumed full budget */ + if (work_done >= budget) + break; + } + + /* Fall out from the NAPI loop if needed */ + if (!(qede_has_rx_work(fp->rxq) || qede_has_tx_work(fp))) { + qed_sb_update_sb_idx(fp->sb_info); + /* *_has_*_work() reads the status block, + * thus we need to ensure that status block indices + * have been actually read (qed_sb_update_sb_idx) + * prior to this check (*_has_*_work) so that + * we won't write the "newer" value of the status block + * to HW (if there was a DMA right after + * qede_has_rx_work and if there is no rmb, the memory + * reading (qed_sb_update_sb_idx) may be postponed + * to right before *_ack_sb). In this case there + * will never be another interrupt until there is + * another update of the status block, while there + * is still unhandled work. + */ + rmb(); + + if (!(qede_has_rx_work(fp->rxq) || + qede_has_tx_work(fp))) { + napi_complete(napi); + /* Update and reenable interrupts */ + qed_sb_ack(fp->sb_info, IGU_INT_ENABLE, + 1 /*update*/); + break; + } + } + } + + return work_done; +} + +static irqreturn_t qede_msix_fp_int(int irq, void *fp_cookie) +{ + struct qede_fastpath *fp = fp_cookie; + + qed_sb_ack(fp->sb_info, IGU_INT_DISABLE, 0 /*do not update*/); + + napi_schedule_irqoff(&fp->napi); + return IRQ_HANDLED; +} + +/* ------------------------------------------------------------------------- + * END OF FAST-PATH + * ------------------------------------------------------------------------- + */ + +static int qede_open(struct net_device *ndev); +static int qede_close(struct net_device *ndev); +static const struct net_device_ops qede_netdev_ops = { + .ndo_open = qede_open, + .ndo_stop = qede_close, + .ndo_start_xmit = qede_start_xmit, + .ndo_validate_addr = eth_validate_addr, +}; + +/* ------------------------------------------------------------------------- * START OF PROBE / REMOVE * ------------------------------------------------------------------------- */ @@ -165,6 +1066,8 @@ static struct qede_dev *qede_alloc_etherdev(struct qed_dev *cdev, edev->dp_module = dp_module; edev->dp_level = dp_level; edev->ops = qed_ops; + edev->q_num_rx_buffers = NUM_RX_BDS_DEF; + edev->q_num_tx_buffers = NUM_TX_BDS_DEF; DP_INFO(edev, "Allocated netdev with 64 tx queues and 64 rx queues\n"); @@ -192,6 +1095,8 @@ static void qede_init_ndev(struct qede_dev *edev) ndev->watchdog_timeo = TX_TIMEOUT; + ndev->netdev_ops = &qede_netdev_ops; + /* user-changeble features */ hw_features = NETIF_F_GRO | NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | @@ -235,6 +1140,64 @@ static void qede_config_debug(uint debug, u32 *p_dp_module, u8 *p_dp_level) } } +static void qede_free_fp_array(struct qede_dev *edev) +{ + if (edev->fp_array) { + struct qede_fastpath *fp; + int i; + + for_each_rss(i) { + fp = &edev->fp_array[i]; + + kfree(fp->sb_info); + kfree(fp->rxq); + kfree(fp->txqs); + } + kfree(edev->fp_array); + } + edev->num_rss = 0; +} + +static int qede_alloc_fp_array(struct qede_dev *edev) +{ + struct qede_fastpath *fp; + int i; + + edev->fp_array = kcalloc(QEDE_RSS_CNT(edev), + sizeof(*edev->fp_array), GFP_KERNEL); + if (!edev->fp_array) { + DP_NOTICE(edev, "fp array allocation failed\n"); + goto err; + } + + for_each_rss(i) { + fp = &edev->fp_array[i]; + + fp->sb_info = kcalloc(1, sizeof(*fp->sb_info), GFP_KERNEL); + if (!fp->sb_info) { + DP_NOTICE(edev, "sb info struct allocation failed\n"); + goto err; + } + + fp->rxq = kcalloc(1, sizeof(*fp->rxq), GFP_KERNEL); + if (!fp->rxq) { + DP_NOTICE(edev, "RXQ struct allocation failed\n"); + goto err; + } + + fp->txqs = kcalloc(edev->num_tc, sizeof(*fp->txqs), GFP_KERNEL); + if (!fp->txqs) { + DP_NOTICE(edev, "TXQ array allocation failed\n"); + goto err; + } + } + + return 0; +err: + qede_free_fp_array(edev); + return -ENOMEM; +} + static void qede_update_pf_params(struct qed_dev *cdev) { struct qed_pf_params pf_params; @@ -298,12 +1261,20 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, qede_init_ndev(edev); + rc = register_netdev(edev->ndev); + if (rc) { + DP_NOTICE(edev, "Cannot register net-device\n"); + goto err3; + } + edev->ops->common->set_id(cdev, edev->ndev->name, DRV_MODULE_VERSION); DP_INFO(edev, "Ending successfully qede probe\n"); return 0; +err3: + free_netdev(edev->ndev); err2: qed_ops->common->slowpath_stop(cdev); err1: @@ -335,6 +1306,8 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) DP_INFO(edev, "Starting qede_remove\n"); + unregister_netdev(ndev); + edev->ops->common->set_power_state(cdev, PCI_D0); pci_set_drvdata(pdev, NULL); @@ -352,3 +1325,837 @@ static void qede_remove(struct pci_dev *pdev) { __qede_remove(pdev, QEDE_REMOVE_NORMAL); } + +/* ------------------------------------------------------------------------- + * START OF LOAD / UNLOAD + * ------------------------------------------------------------------------- + */ + +static int qede_set_num_queues(struct qede_dev *edev) +{ + int rc; + u16 rss_num; + + /* Setup queues according to possible resources*/ + rss_num = netif_get_num_default_rss_queues() * + edev->dev_info.common.num_hwfns; + + rss_num = min_t(u16, QEDE_MAX_RSS_CNT(edev), rss_num); + + rc = edev->ops->common->set_fp_int(edev->cdev, rss_num); + if (rc > 0) { + /* Managed to request interrupts for our queues */ + edev->num_rss = rc; + DP_INFO(edev, "Managed %d [of %d] RSS queues\n", + QEDE_RSS_CNT(edev), rss_num); + rc = 0; + } + return rc; +} + +static void qede_free_mem_sb(struct qede_dev *edev, + struct qed_sb_info *sb_info) +{ + if (sb_info->sb_virt) + dma_free_coherent(&edev->pdev->dev, sizeof(*sb_info->sb_virt), + (void *)sb_info->sb_virt, sb_info->sb_phys); +} + +/* This function allocates fast-path status block memory */ +static int qede_alloc_mem_sb(struct qede_dev *edev, + struct qed_sb_info *sb_info, + u16 sb_id) +{ + struct status_block *sb_virt; + dma_addr_t sb_phys; + int rc; + + sb_virt = dma_alloc_coherent(&edev->pdev->dev, + sizeof(*sb_virt), + &sb_phys, GFP_KERNEL); + if (!sb_virt) { + DP_ERR(edev, "Status block allocation failed\n"); + return -ENOMEM; + } + + rc = edev->ops->common->sb_init(edev->cdev, sb_info, + sb_virt, sb_phys, sb_id, + QED_SB_TYPE_L2_QUEUE); + if (rc) { + DP_ERR(edev, "Status block initialization failed\n"); + dma_free_coherent(&edev->pdev->dev, sizeof(*sb_virt), + sb_virt, sb_phys); + return rc; + } + + return 0; +} + +static void qede_free_rx_buffers(struct qede_dev *edev, + struct qede_rx_queue *rxq) +{ + u16 i; + + for (i = rxq->sw_rx_cons; i != rxq->sw_rx_prod; i++) { + struct sw_rx_data *rx_buf; + u8 *data; + + rx_buf = &rxq->sw_rx_ring[i & NUM_RX_BDS_MAX]; + data = rx_buf->data; + + dma_unmap_single(&edev->pdev->dev, + dma_unmap_addr(rx_buf, mapping), + rxq->rx_buf_size, DMA_FROM_DEVICE); + + rx_buf->data = NULL; + kfree(data); + } +} + +static void qede_free_mem_rxq(struct qede_dev *edev, + struct qede_rx_queue *rxq) +{ + /* Free rx buffers */ + qede_free_rx_buffers(edev, rxq); + + /* Free the parallel SW ring */ + kfree(rxq->sw_rx_ring); + + /* Free the real RQ ring used by FW */ + edev->ops->common->chain_free(edev->cdev, &rxq->rx_bd_ring); + edev->ops->common->chain_free(edev->cdev, &rxq->rx_comp_ring); +} + +static int qede_alloc_rx_buffer(struct qede_dev *edev, + struct qede_rx_queue *rxq) +{ + struct sw_rx_data *sw_rx_data; + struct eth_rx_bd *rx_bd; + dma_addr_t mapping; + u16 rx_buf_size; + u8 *data; + + rx_buf_size = rxq->rx_buf_size; + + data = kmalloc(rx_buf_size, GFP_ATOMIC); + if (unlikely(!data)) { + DP_NOTICE(edev, "Failed to allocate Rx data\n"); + return -ENOMEM; + } + + mapping = dma_map_single(&edev->pdev->dev, data, + rx_buf_size, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) { + kfree(data); + DP_NOTICE(edev, "Failed to map Rx buffer\n"); + return -ENOMEM; + } + + sw_rx_data = &rxq->sw_rx_ring[rxq->sw_rx_prod & NUM_RX_BDS_MAX]; + sw_rx_data->data = data; + + dma_unmap_addr_set(sw_rx_data, mapping, mapping); + + /* Advance PROD and get BD pointer */ + rx_bd = (struct eth_rx_bd *)qed_chain_produce(&rxq->rx_bd_ring); + WARN_ON(!rx_bd); + rx_bd->addr.hi = cpu_to_le32(upper_32_bits(mapping)); + rx_bd->addr.lo = cpu_to_le32(lower_32_bits(mapping)); + + rxq->sw_rx_prod++; + + return 0; +} + +/* This function allocates all memory needed per Rx queue */ +static int qede_alloc_mem_rxq(struct qede_dev *edev, + struct qede_rx_queue *rxq) +{ + int i, rc, size, num_allocated; + + rxq->num_rx_buffers = edev->q_num_rx_buffers; + + rxq->rx_buf_size = NET_IP_ALIGN + + ETH_OVERHEAD + + edev->ndev->mtu + + QEDE_FW_RX_ALIGN_END; + + /* Allocate the parallel driver ring for Rx buffers */ + size = sizeof(*rxq->sw_rx_ring) * NUM_RX_BDS_MAX; + rxq->sw_rx_ring = kzalloc(size, GFP_KERNEL); + if (!rxq->sw_rx_ring) { + DP_ERR(edev, "Rx buffers ring allocation failed\n"); + goto err; + } + + /* Allocate FW Rx ring */ + rc = edev->ops->common->chain_alloc(edev->cdev, + QED_CHAIN_USE_TO_CONSUME_PRODUCE, + QED_CHAIN_MODE_NEXT_PTR, + NUM_RX_BDS_MAX, + sizeof(struct eth_rx_bd), + &rxq->rx_bd_ring); + + if (rc) + goto err; + + /* Allocate FW completion ring */ + rc = edev->ops->common->chain_alloc(edev->cdev, + QED_CHAIN_USE_TO_CONSUME, + QED_CHAIN_MODE_PBL, + NUM_RX_BDS_MAX, + sizeof(union eth_rx_cqe), + &rxq->rx_comp_ring); + if (rc) + goto err; + + /* Allocate buffers for the Rx ring */ + for (i = 0; i < rxq->num_rx_buffers; i++) { + rc = qede_alloc_rx_buffer(edev, rxq); + if (rc) + break; + } + num_allocated = i; + if (!num_allocated) { + DP_ERR(edev, "Rx buffers allocation failed\n"); + goto err; + } else if (num_allocated < rxq->num_rx_buffers) { + DP_NOTICE(edev, + "Allocated less buffers than desired (%d allocated)\n", + num_allocated); + } + + return 0; + +err: + qede_free_mem_rxq(edev, rxq); + return -ENOMEM; +} + +static void qede_free_mem_txq(struct qede_dev *edev, + struct qede_tx_queue *txq) +{ + /* Free the parallel SW ring */ + kfree(txq->sw_tx_ring); + + /* Free the real RQ ring used by FW */ + edev->ops->common->chain_free(edev->cdev, &txq->tx_pbl); +} + +/* This function allocates all memory needed per Tx queue */ +static int qede_alloc_mem_txq(struct qede_dev *edev, + struct qede_tx_queue *txq) +{ + int size, rc; + union eth_tx_bd_types *p_virt; + + txq->num_tx_buffers = edev->q_num_tx_buffers; + + /* Allocate the parallel driver ring for Tx buffers */ + size = sizeof(*txq->sw_tx_ring) * NUM_TX_BDS_MAX; + txq->sw_tx_ring = kzalloc(size, GFP_KERNEL); + if (!txq->sw_tx_ring) { + DP_NOTICE(edev, "Tx buffers ring allocation failed\n"); + goto err; + } + + rc = edev->ops->common->chain_alloc(edev->cdev, + QED_CHAIN_USE_TO_CONSUME_PRODUCE, + QED_CHAIN_MODE_PBL, + NUM_TX_BDS_MAX, + sizeof(*p_virt), + &txq->tx_pbl); + if (rc) + goto err; + + return 0; + +err: + qede_free_mem_txq(edev, txq); + return -ENOMEM; +} + +/* This function frees all memory of a single fp */ +static void qede_free_mem_fp(struct qede_dev *edev, + struct qede_fastpath *fp) +{ + int tc; + + qede_free_mem_sb(edev, fp->sb_info); + + qede_free_mem_rxq(edev, fp->rxq); + + for (tc = 0; tc < edev->num_tc; tc++) + qede_free_mem_txq(edev, &fp->txqs[tc]); +} + +/* This function allocates all memory needed for a single fp (i.e. an entity + * which contains status block, one rx queue and multiple per-TC tx queues. + */ +static int qede_alloc_mem_fp(struct qede_dev *edev, + struct qede_fastpath *fp) +{ + int rc, tc; + + rc = qede_alloc_mem_sb(edev, fp->sb_info, fp->rss_id); + if (rc) + goto err; + + rc = qede_alloc_mem_rxq(edev, fp->rxq); + if (rc) + goto err; + + for (tc = 0; tc < edev->num_tc; tc++) { + rc = qede_alloc_mem_txq(edev, &fp->txqs[tc]); + if (rc) + goto err; + } + + return 0; + +err: + qede_free_mem_fp(edev, fp); + return -ENOMEM; +} + +static void qede_free_mem_load(struct qede_dev *edev) +{ + int i; + + for_each_rss(i) { + struct qede_fastpath *fp = &edev->fp_array[i]; + + qede_free_mem_fp(edev, fp); + } +} + +/* This function allocates all qede memory at NIC load. */ +static int qede_alloc_mem_load(struct qede_dev *edev) +{ + int rc = 0, rss_id; + + for (rss_id = 0; rss_id < QEDE_RSS_CNT(edev); rss_id++) { + struct qede_fastpath *fp = &edev->fp_array[rss_id]; + + rc = qede_alloc_mem_fp(edev, fp); + if (rc) + break; + } + + if (rss_id != QEDE_RSS_CNT(edev)) { + /* Failed allocating memory for all the queues */ + if (!rss_id) { + DP_ERR(edev, + "Failed to allocate memory for the leading queue\n"); + rc = -ENOMEM; + } else { + DP_NOTICE(edev, + "Failed to allocate memory for all of RSS queues\n Desired: %d queues, allocated: %d queues\n", + QEDE_RSS_CNT(edev), rss_id); + } + edev->num_rss = rss_id; + } + + return 0; +} + +/* This function inits fp content and resets the SB, RXQ and TXQ structures */ +static void qede_init_fp(struct qede_dev *edev) +{ + int rss_id, txq_index, tc; + struct qede_fastpath *fp; + + for_each_rss(rss_id) { + fp = &edev->fp_array[rss_id]; + + fp->edev = edev; + fp->rss_id = rss_id; + + memset((void *)&fp->napi, 0, sizeof(fp->napi)); + + memset((void *)fp->sb_info, 0, sizeof(*fp->sb_info)); + + memset((void *)fp->rxq, 0, sizeof(*fp->rxq)); + fp->rxq->rxq_id = rss_id; + + memset((void *)fp->txqs, 0, (edev->num_tc * sizeof(*fp->txqs))); + for (tc = 0; tc < edev->num_tc; tc++) { + txq_index = tc * QEDE_RSS_CNT(edev) + rss_id; + fp->txqs[tc].index = txq_index; + } + + snprintf(fp->name, sizeof(fp->name), "%s-fp-%d", + edev->ndev->name, rss_id); + } +} + +static int qede_set_real_num_queues(struct qede_dev *edev) +{ + int rc = 0; + + rc = netif_set_real_num_tx_queues(edev->ndev, QEDE_TSS_CNT(edev)); + if (rc) { + DP_NOTICE(edev, "Failed to set real number of Tx queues\n"); + return rc; + } + rc = netif_set_real_num_rx_queues(edev->ndev, QEDE_RSS_CNT(edev)); + if (rc) { + DP_NOTICE(edev, "Failed to set real number of Rx queues\n"); + return rc; + } + + return 0; +} + +static void qede_napi_disable_remove(struct qede_dev *edev) +{ + int i; + + for_each_rss(i) { + napi_disable(&edev->fp_array[i].napi); + + netif_napi_del(&edev->fp_array[i].napi); + } +} + +static void qede_napi_add_enable(struct qede_dev *edev) +{ + int i; + + /* Add NAPI objects */ + for_each_rss(i) { + netif_napi_add(edev->ndev, &edev->fp_array[i].napi, + qede_poll, NAPI_POLL_WEIGHT); + napi_enable(&edev->fp_array[i].napi); + } +} + +static void qede_sync_free_irqs(struct qede_dev *edev) +{ + int i; + + for (i = 0; i < edev->int_info.used_cnt; i++) { + if (edev->int_info.msix_cnt) { + synchronize_irq(edev->int_info.msix[i].vector); + free_irq(edev->int_info.msix[i].vector, + &edev->fp_array[i]); + } else { + edev->ops->common->simd_handler_clean(edev->cdev, i); + } + } + + edev->int_info.used_cnt = 0; +} + +static int qede_req_msix_irqs(struct qede_dev *edev) +{ + int i, rc; + + /* Sanitize number of interrupts == number of prepared RSS queues */ + if (QEDE_RSS_CNT(edev) > edev->int_info.msix_cnt) { + DP_ERR(edev, + "Interrupt mismatch: %d RSS queues > %d MSI-x vectors\n", + QEDE_RSS_CNT(edev), edev->int_info.msix_cnt); + return -EINVAL; + } + + for (i = 0; i < QEDE_RSS_CNT(edev); i++) { + rc = request_irq(edev->int_info.msix[i].vector, + qede_msix_fp_int, 0, edev->fp_array[i].name, + &edev->fp_array[i]); + if (rc) { + DP_ERR(edev, "Request fp %d irq failed\n", i); + qede_sync_free_irqs(edev); + return rc; + } + DP_VERBOSE(edev, NETIF_MSG_INTR, + "Requested fp irq for %s [entry %d]. Cookie is at %p\n", + edev->fp_array[i].name, i, + &edev->fp_array[i]); + edev->int_info.used_cnt++; + } + + return 0; +} + +static void qede_simd_fp_handler(void *cookie) +{ + struct qede_fastpath *fp = (struct qede_fastpath *)cookie; + + napi_schedule_irqoff(&fp->napi); +} + +static int qede_setup_irqs(struct qede_dev *edev) +{ + int i, rc = 0; + + /* Learn Interrupt configuration */ + rc = edev->ops->common->get_fp_int(edev->cdev, &edev->int_info); + if (rc) + return rc; + + if (edev->int_info.msix_cnt) { + rc = qede_req_msix_irqs(edev); + if (rc) + return rc; + edev->ndev->irq = edev->int_info.msix[0].vector; + } else { + const struct qed_common_ops *ops; + + /* qed should learn receive the RSS ids and callbacks */ + ops = edev->ops->common; + for (i = 0; i < QEDE_RSS_CNT(edev); i++) + ops->simd_handler_config(edev->cdev, + &edev->fp_array[i], i, + qede_simd_fp_handler); + edev->int_info.used_cnt = QEDE_RSS_CNT(edev); + } + return 0; +} + +static int qede_drain_txq(struct qede_dev *edev, + struct qede_tx_queue *txq, + bool allow_drain) +{ + int rc, cnt = 1000; + + while (txq->sw_tx_cons != txq->sw_tx_prod) { + if (!cnt) { + if (allow_drain) { + DP_NOTICE(edev, + "Tx queue[%d] is stuck, requesting MCP to drain\n", + txq->index); + rc = edev->ops->common->drain(edev->cdev); + if (rc) + return rc; + return qede_drain_txq(edev, txq, false); + } + DP_NOTICE(edev, + "Timeout waiting for tx queue[%d]: PROD=%d, CONS=%d\n", + txq->index, txq->sw_tx_prod, + txq->sw_tx_cons); + return -ENODEV; + } + cnt--; + usleep_range(1000, 2000); + barrier(); + } + + /* FW finished processing, wait for HW to transmit all tx packets */ + usleep_range(1000, 2000); + + return 0; +} + +static int qede_stop_queues(struct qede_dev *edev) +{ + struct qed_update_vport_params vport_update_params; + struct qed_dev *cdev = edev->cdev; + int rc, tc, i; + + /* Disable the vport */ + memset(&vport_update_params, 0, sizeof(vport_update_params)); + vport_update_params.vport_id = 0; + vport_update_params.update_vport_active_flg = 1; + vport_update_params.vport_active_flg = 0; + vport_update_params.update_rss_flg = 0; + + rc = edev->ops->vport_update(cdev, &vport_update_params); + if (rc) { + DP_ERR(edev, "Failed to update vport\n"); + return rc; + } + + /* Flush Tx queues. If needed, request drain from MCP */ + for_each_rss(i) { + struct qede_fastpath *fp = &edev->fp_array[i]; + + for (tc = 0; tc < edev->num_tc; tc++) { + struct qede_tx_queue *txq = &fp->txqs[tc]; + + rc = qede_drain_txq(edev, txq, true); + if (rc) + return rc; + } + } + + /* Stop all Queues in reverse order*/ + for (i = QEDE_RSS_CNT(edev) - 1; i >= 0; i--) { + struct qed_stop_rxq_params rx_params; + + /* Stop the Tx Queue(s)*/ + for (tc = 0; tc < edev->num_tc; tc++) { + struct qed_stop_txq_params tx_params; + + tx_params.rss_id = i; + tx_params.tx_queue_id = tc * QEDE_RSS_CNT(edev) + i; + rc = edev->ops->q_tx_stop(cdev, &tx_params); + if (rc) { + DP_ERR(edev, "Failed to stop TXQ #%d\n", + tx_params.tx_queue_id); + return rc; + } + } + + /* Stop the Rx Queue*/ + memset(&rx_params, 0, sizeof(rx_params)); + rx_params.rss_id = i; + rx_params.rx_queue_id = i; + + rc = edev->ops->q_rx_stop(cdev, &rx_params); + if (rc) { + DP_ERR(edev, "Failed to stop RXQ #%d\n", i); + return rc; + } + } + + /* Stop the vport */ + rc = edev->ops->vport_stop(cdev, 0); + if (rc) + DP_ERR(edev, "Failed to stop VPORT\n"); + + return rc; +} + +static int qede_start_queues(struct qede_dev *edev) +{ + int rc, tc, i; + int vport_id = 0, drop_ttl0_flg = 1, vlan_removal_en = 1; + struct qed_dev *cdev = edev->cdev; + struct qed_update_vport_rss_params *rss_params = &edev->rss_params; + struct qed_update_vport_params vport_update_params; + struct qed_queue_start_common_params q_params; + + if (!edev->num_rss) { + DP_ERR(edev, + "Cannot update V-VPORT as active as there are no Rx queues\n"); + return -EINVAL; + } + + rc = edev->ops->vport_start(cdev, vport_id, + edev->ndev->mtu, + drop_ttl0_flg, + vlan_removal_en); + + if (rc) { + DP_ERR(edev, "Start V-PORT failed %d\n", rc); + return rc; + } + + DP_VERBOSE(edev, NETIF_MSG_IFUP, + "Start vport ramrod passed, vport_id = %d, MTU = %d, vlan_removal_en = %d\n", + vport_id, edev->ndev->mtu + 0xe, vlan_removal_en); + + for_each_rss(i) { + struct qede_fastpath *fp = &edev->fp_array[i]; + dma_addr_t phys_table = fp->rxq->rx_comp_ring.pbl.p_phys_table; + + memset(&q_params, 0, sizeof(q_params)); + q_params.rss_id = i; + q_params.queue_id = i; + q_params.vport_id = 0; + q_params.sb = fp->sb_info->igu_sb_id; + q_params.sb_idx = RX_PI; + + rc = edev->ops->q_rx_start(cdev, &q_params, + fp->rxq->rx_buf_size, + fp->rxq->rx_bd_ring.p_phys_addr, + phys_table, + fp->rxq->rx_comp_ring.page_cnt, + &fp->rxq->hw_rxq_prod_addr); + if (rc) { + DP_ERR(edev, "Start RXQ #%d failed %d\n", i, rc); + return rc; + } + + fp->rxq->hw_cons_ptr = &fp->sb_info->sb_virt->pi_array[RX_PI]; + + qede_update_rx_prod(edev, fp->rxq); + + for (tc = 0; tc < edev->num_tc; tc++) { + struct qede_tx_queue *txq = &fp->txqs[tc]; + int txq_index = tc * QEDE_RSS_CNT(edev) + i; + + memset(&q_params, 0, sizeof(q_params)); + q_params.rss_id = i; + q_params.queue_id = txq_index; + q_params.vport_id = 0; + q_params.sb = fp->sb_info->igu_sb_id; + q_params.sb_idx = TX_PI(tc); + + rc = edev->ops->q_tx_start(cdev, &q_params, + txq->tx_pbl.pbl.p_phys_table, + txq->tx_pbl.page_cnt, + &txq->doorbell_addr); + if (rc) { + DP_ERR(edev, "Start TXQ #%d failed %d\n", + txq_index, rc); + return rc; + } + + txq->hw_cons_ptr = + &fp->sb_info->sb_virt->pi_array[TX_PI(tc)]; + SET_FIELD(txq->tx_db.data.params, + ETH_DB_DATA_DEST, DB_DEST_XCM); + SET_FIELD(txq->tx_db.data.params, ETH_DB_DATA_AGG_CMD, + DB_AGG_CMD_SET); + SET_FIELD(txq->tx_db.data.params, + ETH_DB_DATA_AGG_VAL_SEL, + DQ_XCM_ETH_TX_BD_PROD_CMD); + + txq->tx_db.data.agg_flags = DQ_XCM_ETH_DQ_CF_CMD; + } + } + + /* Prepare and send the vport enable */ + memset(&vport_update_params, 0, sizeof(vport_update_params)); + vport_update_params.vport_id = vport_id; + vport_update_params.update_vport_active_flg = 1; + vport_update_params.vport_active_flg = 1; + + /* Fill struct with RSS params */ + if (QEDE_RSS_CNT(edev) > 1) { + vport_update_params.update_rss_flg = 1; + for (i = 0; i < 128; i++) + rss_params->rss_ind_table[i] = + ethtool_rxfh_indir_default(i, QEDE_RSS_CNT(edev)); + netdev_rss_key_fill(rss_params->rss_key, + sizeof(rss_params->rss_key)); + } else { + memset(rss_params, 0, sizeof(*rss_params)); + } + memcpy(&vport_update_params.rss_params, rss_params, + sizeof(*rss_params)); + + rc = edev->ops->vport_update(cdev, &vport_update_params); + if (rc) { + DP_ERR(edev, "Update V-PORT failed %d\n", rc); + return rc; + } + + return 0; +} + +enum qede_unload_mode { + QEDE_UNLOAD_NORMAL, +}; + +static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode) +{ + int rc; + + DP_INFO(edev, "Starting qede unload\n"); + + /* Close OS Tx */ + netif_tx_disable(edev->ndev); + netif_carrier_off(edev->ndev); + + rc = qede_stop_queues(edev); + if (rc) { + qede_sync_free_irqs(edev); + goto out; + } + + DP_INFO(edev, "Stopped Queues\n"); + + edev->ops->fastpath_stop(edev->cdev); + + /* Release the interrupts */ + qede_sync_free_irqs(edev); + edev->ops->common->set_fp_int(edev->cdev, 0); + + qede_napi_disable_remove(edev); + + qede_free_mem_load(edev); + qede_free_fp_array(edev); + +out: + mutex_unlock(&edev->qede_lock); + DP_INFO(edev, "Ending qede unload\n"); +} + +enum qede_load_mode { + QEDE_LOAD_NORMAL, +}; + +static int qede_load(struct qede_dev *edev, enum qede_load_mode mode) +{ + int rc; + + DP_INFO(edev, "Starting qede load\n"); + + rc = qede_set_num_queues(edev); + if (rc) + goto err0; + + rc = qede_alloc_fp_array(edev); + if (rc) + goto err0; + + qede_init_fp(edev); + + rc = qede_alloc_mem_load(edev); + if (rc) + goto err1; + DP_INFO(edev, "Allocated %d RSS queues on %d TC/s\n", + QEDE_RSS_CNT(edev), edev->num_tc); + + rc = qede_set_real_num_queues(edev); + if (rc) + goto err2; + + qede_napi_add_enable(edev); + DP_INFO(edev, "Napi added and enabled\n"); + + rc = qede_setup_irqs(edev); + if (rc) + goto err3; + DP_INFO(edev, "Setup IRQs succeeded\n"); + + rc = qede_start_queues(edev); + if (rc) + goto err4; + DP_INFO(edev, "Start VPORT, RXQ and TXQ succeeded\n"); + + /* Add primary mac and set Rx filters */ + ether_addr_copy(edev->primary_mac, edev->ndev->dev_addr); + + DP_INFO(edev, "Ending successfully qede load\n"); + + return 0; + +err4: + qede_sync_free_irqs(edev); + memset(&edev->int_info.msix_cnt, 0, sizeof(struct qed_int_info)); +err3: + qede_napi_disable_remove(edev); +err2: + qede_free_mem_load(edev); +err1: + edev->ops->common->set_fp_int(edev->cdev, 0); + qede_free_fp_array(edev); + edev->num_rss = 0; +err0: + return rc; +} + +/* called with rtnl_lock */ +static int qede_open(struct net_device *ndev) +{ + struct qede_dev *edev = netdev_priv(ndev); + + netif_carrier_off(ndev); + + edev->ops->common->set_power_state(edev->cdev, PCI_D0); + + return qede_load(edev, QEDE_LOAD_NORMAL); +} + +static int qede_close(struct net_device *ndev) +{ + struct qede_dev *edev = netdev_priv(ndev); + + qede_unload(edev, QEDE_UNLOAD_NORMAL); + + return 0; +} -- 2.7.4