diff options
Diffstat (limited to 'drivers/net/ethernet/emulex/benet/be_main.c')
| -rw-r--r-- | drivers/net/ethernet/emulex/benet/be_main.c | 2899 |
1 files changed, 1819 insertions, 1080 deletions
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 5c995700e53..1e187fb760f 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2011 Emulex + * Copyright (C) 2005 - 2014 Emulex * All rights reserved. * * This program is free software; you can redistribute it and/or @@ -21,11 +21,14 @@ #include "be_cmds.h" #include <asm/div64.h> #include <linux/aer.h> +#include <linux/if_bridge.h> +#include <net/busy_poll.h> +#include <net/vxlan.h> MODULE_VERSION(DRV_VER); MODULE_DEVICE_TABLE(pci, be_dev_ids); MODULE_DESCRIPTION(DRV_DESC " " DRV_VER); -MODULE_AUTHOR("ServerEngines Corporation"); +MODULE_AUTHOR("Emulex Corporation"); MODULE_LICENSE("GPL"); static unsigned int num_vfs; @@ -119,12 +122,6 @@ static const char * const ue_status_hi_desc[] = { "Unknown" }; -/* Is BE in a multi-channel mode */ -static inline bool be_is_mc(struct be_adapter *adapter) { - return (adapter->function_mode & FLEX10_MODE || - adapter->function_mode & VNIC_MODE || - adapter->function_mode & UMC_ENABLED); -} static void be_queue_free(struct be_adapter *adapter, struct be_queue_info *q) { @@ -137,7 +134,7 @@ static void be_queue_free(struct be_adapter *adapter, struct be_queue_info *q) } static int be_queue_alloc(struct be_adapter *adapter, struct be_queue_info *q, - u16 len, u16 entry_size) + u16 len, u16 entry_size) { struct be_dma_mem *mem = &q->dma_mem; @@ -145,23 +142,19 @@ static int be_queue_alloc(struct be_adapter *adapter, struct be_queue_info *q, q->len = len; q->entry_size = entry_size; mem->size = len * entry_size; - mem->va = dma_alloc_coherent(&adapter->pdev->dev, mem->size, &mem->dma, - GFP_KERNEL); + mem->va = dma_zalloc_coherent(&adapter->pdev->dev, mem->size, &mem->dma, + GFP_KERNEL); if (!mem->va) return -ENOMEM; - memset(mem->va, 0, mem->size); return 0; } -static void be_intr_set(struct be_adapter *adapter, bool enable) +static void be_reg_intr_set(struct be_adapter *adapter, bool enable) { u32 reg, enabled; - if (adapter->eeh_error) - return; - pci_read_config_dword(adapter->pdev, PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET, - ®); + ®); enabled = reg & MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK; if (!enabled && enable) @@ -172,7 +165,23 @@ static void be_intr_set(struct be_adapter *adapter, bool enable) return; pci_write_config_dword(adapter->pdev, - PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET, reg); + PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET, reg); +} + +static void be_intr_set(struct be_adapter *adapter, bool enable) +{ + int status = 0; + + /* On lancer interrupts can't be controlled via this register */ + if (lancer_chip(adapter)) + return; + + if (adapter->eeh_error) + return; + + status = be_cmd_intr_set(adapter, enable); + if (status) + be_reg_intr_set(adapter, enable); } static void be_rxq_notify(struct be_adapter *adapter, u16 qid, u16 posted) @@ -185,23 +194,23 @@ static void be_rxq_notify(struct be_adapter *adapter, u16 qid, u16 posted) iowrite32(val, adapter->db + DB_RQ_OFFSET); } -static void be_txq_notify(struct be_adapter *adapter, u16 qid, u16 posted) +static void be_txq_notify(struct be_adapter *adapter, struct be_tx_obj *txo, + u16 posted) { u32 val = 0; - val |= qid & DB_TXULP_RING_ID_MASK; + val |= txo->q.id & DB_TXULP_RING_ID_MASK; val |= (posted & DB_TXULP_NUM_POSTED_MASK) << DB_TXULP_NUM_POSTED_SHIFT; wmb(); - iowrite32(val, adapter->db + DB_TXULP1_OFFSET); + iowrite32(val, adapter->db + txo->db_offset); } static void be_eq_notify(struct be_adapter *adapter, u16 qid, - bool arm, bool clear_int, u16 num_popped) + bool arm, bool clear_int, u16 num_popped) { u32 val = 0; val |= qid & DB_EQ_RING_ID_MASK; - val |= ((qid & DB_EQ_RING_ID_EXT_MASK) << - DB_EQ_RING_ID_EXT_MASK_SHIFT); + val |= ((qid & DB_EQ_RING_ID_EXT_MASK) << DB_EQ_RING_ID_EXT_MASK_SHIFT); if (adapter->eeh_error) return; @@ -234,54 +243,61 @@ void be_cq_notify(struct be_adapter *adapter, u16 qid, bool arm, u16 num_popped) static int be_mac_addr_set(struct net_device *netdev, void *p) { struct be_adapter *adapter = netdev_priv(netdev); + struct device *dev = &adapter->pdev->dev; struct sockaddr *addr = p; - int status = 0; - u8 current_mac[ETH_ALEN]; - u32 pmac_id = adapter->pmac_id[0]; - bool active_mac = true; + int status; + u8 mac[ETH_ALEN]; + u32 old_pmac_id = adapter->pmac_id[0], curr_pmac_id = 0; if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; - /* For BE VF, MAC address is already activated by PF. - * Hence only operation left is updating netdev->devaddr. - * Update it if user is passing the same MAC which was used - * during configuring VF MAC from PF(Hypervisor). + /* Proceed further only if, User provided MAC is different + * from active MAC */ - if (!lancer_chip(adapter) && !be_physfn(adapter)) { - status = be_cmd_mac_addr_query(adapter, current_mac, - false, adapter->if_handle, 0); - if (!status && !memcmp(current_mac, addr->sa_data, ETH_ALEN)) - goto done; - else - goto err; - } - - if (!memcmp(addr->sa_data, netdev->dev_addr, ETH_ALEN)) - goto done; + if (ether_addr_equal(addr->sa_data, netdev->dev_addr)) + return 0; - /* For Lancer check if any MAC is active. - * If active, get its mac id. + /* The PMAC_ADD cmd may fail if the VF doesn't have FILTMGMT + * privilege or if PF did not provision the new MAC address. + * On BE3, this cmd will always fail if the VF doesn't have the + * FILTMGMT privilege. This failure is OK, only if the PF programmed + * the MAC for the VF. */ - if (lancer_chip(adapter) && !be_physfn(adapter)) - be_cmd_get_mac_from_list(adapter, current_mac, &active_mac, - &pmac_id, 0); - status = be_cmd_pmac_add(adapter, (u8 *)addr->sa_data, - adapter->if_handle, - &adapter->pmac_id[0], 0); + adapter->if_handle, &adapter->pmac_id[0], 0); + if (!status) { + curr_pmac_id = adapter->pmac_id[0]; + + /* Delete the old programmed MAC. This call may fail if the + * old MAC was already deleted by the PF driver. + */ + if (adapter->pmac_id[0] != old_pmac_id) + be_cmd_pmac_del(adapter, adapter->if_handle, + old_pmac_id, 0); + } + /* Decide if the new MAC is successfully activated only after + * querying the FW + */ + status = be_cmd_get_active_mac(adapter, curr_pmac_id, mac, + adapter->if_handle, true, 0); if (status) goto err; - if (active_mac) - be_cmd_pmac_del(adapter, adapter->if_handle, - pmac_id, 0); -done: + /* The MAC change did not happen, either due to lack of privilege + * or PF didn't pre-provision. + */ + if (!ether_addr_equal(addr->sa_data, mac)) { + status = -EPERM; + goto err; + } + memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); + dev_info(dev, "MAC address changed to %pM\n", mac); return 0; err: - dev_err(&adapter->pdev->dev, "MAC %pM set Failed\n", addr->sa_data); + dev_warn(dev, "MAC address change to %pM failed\n", addr->sa_data); return status; } @@ -292,10 +308,14 @@ static void *hw_stats_from_cmd(struct be_adapter *adapter) struct be_cmd_resp_get_stats_v0 *cmd = adapter->stats_cmd.va; return &cmd->hw_stats; - } else { + } else if (BE3_chip(adapter)) { struct be_cmd_resp_get_stats_v1 *cmd = adapter->stats_cmd.va; return &cmd->hw_stats; + } else { + struct be_cmd_resp_get_stats_v2 *cmd = adapter->stats_cmd.va; + + return &cmd->hw_stats; } } @@ -306,10 +326,14 @@ static void *be_erx_stats_from_cmd(struct be_adapter *adapter) struct be_hw_stats_v0 *hw_stats = hw_stats_from_cmd(adapter); return &hw_stats->erx; - } else { + } else if (BE3_chip(adapter)) { struct be_hw_stats_v1 *hw_stats = hw_stats_from_cmd(adapter); return &hw_stats->erx; + } else { + struct be_hw_stats_v2 *hw_stats = hw_stats_from_cmd(adapter); + + return &hw_stats->erx; } } @@ -340,9 +364,9 @@ static void populate_be_v0_stats(struct be_adapter *adapter) drvs->rx_input_fifo_overflow_drop = port_stats->rx_input_fifo_overflow; drvs->rx_dropped_header_too_small = port_stats->rx_dropped_header_too_small; - drvs->rx_address_mismatch_drops = - port_stats->rx_address_mismatch_drops + - port_stats->rx_vlan_mismatch_drops; + drvs->rx_address_filtered = + port_stats->rx_address_filtered + + port_stats->rx_vlan_filtered; drvs->rx_alignment_symbol_errors = port_stats->rx_alignment_symbol_errors; @@ -391,12 +415,59 @@ static void populate_be_v1_stats(struct be_adapter *adapter) port_stats->rx_dropped_header_too_small; drvs->rx_input_fifo_overflow_drop = port_stats->rx_input_fifo_overflow_drop; - drvs->rx_address_mismatch_drops = port_stats->rx_address_mismatch_drops; + drvs->rx_address_filtered = port_stats->rx_address_filtered; + drvs->rx_alignment_symbol_errors = + port_stats->rx_alignment_symbol_errors; + drvs->rxpp_fifo_overflow_drop = port_stats->rxpp_fifo_overflow_drop; + drvs->tx_pauseframes = port_stats->tx_pauseframes; + drvs->tx_controlframes = port_stats->tx_controlframes; + drvs->tx_priority_pauseframes = port_stats->tx_priority_pauseframes; + drvs->jabber_events = port_stats->jabber_events; + drvs->rx_drops_no_pbuf = rxf_stats->rx_drops_no_pbuf; + drvs->rx_drops_no_erx_descr = rxf_stats->rx_drops_no_erx_descr; + drvs->forwarded_packets = rxf_stats->forwarded_packets; + drvs->rx_drops_mtu = rxf_stats->rx_drops_mtu; + drvs->rx_drops_no_tpre_descr = rxf_stats->rx_drops_no_tpre_descr; + drvs->rx_drops_too_many_frags = rxf_stats->rx_drops_too_many_frags; + adapter->drv_stats.eth_red_drops = pmem_sts->eth_red_drops; +} + +static void populate_be_v2_stats(struct be_adapter *adapter) +{ + struct be_hw_stats_v2 *hw_stats = hw_stats_from_cmd(adapter); + struct be_pmem_stats *pmem_sts = &hw_stats->pmem; + struct be_rxf_stats_v2 *rxf_stats = &hw_stats->rxf; + struct be_port_rxf_stats_v2 *port_stats = + &rxf_stats->port[adapter->port_num]; + struct be_drv_stats *drvs = &adapter->drv_stats; + + be_dws_le_to_cpu(hw_stats, sizeof(*hw_stats)); + drvs->pmem_fifo_overflow_drop = port_stats->pmem_fifo_overflow_drop; + drvs->rx_priority_pause_frames = port_stats->rx_priority_pause_frames; + drvs->rx_pause_frames = port_stats->rx_pause_frames; + drvs->rx_crc_errors = port_stats->rx_crc_errors; + drvs->rx_control_frames = port_stats->rx_control_frames; + drvs->rx_in_range_errors = port_stats->rx_in_range_errors; + drvs->rx_frame_too_long = port_stats->rx_frame_too_long; + drvs->rx_dropped_runt = port_stats->rx_dropped_runt; + drvs->rx_ip_checksum_errs = port_stats->rx_ip_checksum_errs; + drvs->rx_tcp_checksum_errs = port_stats->rx_tcp_checksum_errs; + drvs->rx_udp_checksum_errs = port_stats->rx_udp_checksum_errs; + drvs->rx_dropped_tcp_length = port_stats->rx_dropped_tcp_length; + drvs->rx_dropped_too_small = port_stats->rx_dropped_too_small; + drvs->rx_dropped_too_short = port_stats->rx_dropped_too_short; + drvs->rx_out_range_errors = port_stats->rx_out_range_errors; + drvs->rx_dropped_header_too_small = + port_stats->rx_dropped_header_too_small; + drvs->rx_input_fifo_overflow_drop = + port_stats->rx_input_fifo_overflow_drop; + drvs->rx_address_filtered = port_stats->rx_address_filtered; drvs->rx_alignment_symbol_errors = port_stats->rx_alignment_symbol_errors; drvs->rxpp_fifo_overflow_drop = port_stats->rxpp_fifo_overflow_drop; drvs->tx_pauseframes = port_stats->tx_pauseframes; drvs->tx_controlframes = port_stats->tx_controlframes; + drvs->tx_priority_pauseframes = port_stats->tx_priority_pauseframes; drvs->jabber_events = port_stats->jabber_events; drvs->rx_drops_no_pbuf = rxf_stats->rx_drops_no_pbuf; drvs->rx_drops_no_erx_descr = rxf_stats->rx_drops_no_erx_descr; @@ -405,14 +476,21 @@ static void populate_be_v1_stats(struct be_adapter *adapter) drvs->rx_drops_no_tpre_descr = rxf_stats->rx_drops_no_tpre_descr; drvs->rx_drops_too_many_frags = rxf_stats->rx_drops_too_many_frags; adapter->drv_stats.eth_red_drops = pmem_sts->eth_red_drops; + if (be_roce_supported(adapter)) { + drvs->rx_roce_bytes_lsd = port_stats->roce_bytes_received_lsd; + drvs->rx_roce_bytes_msd = port_stats->roce_bytes_received_msd; + drvs->rx_roce_frames = port_stats->roce_frames_received; + drvs->roce_drops_crc = port_stats->roce_drops_crc; + drvs->roce_drops_payload_len = + port_stats->roce_drops_payload_len; + } } static void populate_lancer_stats(struct be_adapter *adapter) { struct be_drv_stats *drvs = &adapter->drv_stats; - struct lancer_pport_stats *pport_stats = - pport_stats_from_cmd(adapter); + struct lancer_pport_stats *pport_stats = pport_stats_from_cmd(adapter); be_dws_le_to_cpu(pport_stats, sizeof(*pport_stats)); drvs->rx_pause_frames = pport_stats->rx_pause_frames_lo; @@ -432,9 +510,9 @@ static void populate_lancer_stats(struct be_adapter *adapter) drvs->rx_dropped_header_too_small = pport_stats->rx_dropped_header_too_small; drvs->rx_input_fifo_overflow_drop = pport_stats->rx_fifo_overflow; - drvs->rx_address_mismatch_drops = - pport_stats->rx_address_mismatch_drops + - pport_stats->rx_vlan_mismatch_drops; + drvs->rx_address_filtered = + pport_stats->rx_address_filtered + + pport_stats->rx_vlan_filtered; drvs->rx_alignment_symbol_errors = pport_stats->rx_symbol_errors_lo; drvs->rxpp_fifo_overflow_drop = pport_stats->rx_fifo_overflow; drvs->tx_pauseframes = pport_stats->tx_pause_frames_lo; @@ -458,35 +536,47 @@ static void accumulate_16bit_val(u32 *acc, u16 val) ACCESS_ONCE(*acc) = newacc; } +static void populate_erx_stats(struct be_adapter *adapter, + struct be_rx_obj *rxo, u32 erx_stat) +{ + if (!BEx_chip(adapter)) + rx_stats(rxo)->rx_drops_no_frags = erx_stat; + else + /* below erx HW counter can actually wrap around after + * 65535. Driver accumulates a 32-bit value + */ + accumulate_16bit_val(&rx_stats(rxo)->rx_drops_no_frags, + (u16)erx_stat); +} + void be_parse_stats(struct be_adapter *adapter) { - struct be_erx_stats_v1 *erx = be_erx_stats_from_cmd(adapter); + struct be_erx_stats_v2 *erx = be_erx_stats_from_cmd(adapter); struct be_rx_obj *rxo; int i; + u32 erx_stat; if (lancer_chip(adapter)) { populate_lancer_stats(adapter); } else { if (BE2_chip(adapter)) populate_be_v0_stats(adapter); - else - /* for BE3 and Skyhawk */ + else if (BE3_chip(adapter)) + /* for BE3 */ populate_be_v1_stats(adapter); + else + populate_be_v2_stats(adapter); - /* as erx_v1 is longer than v0, ok to use v1 for v0 access */ + /* erx_v2 is longer than v0, v1. use v2 for v0, v1 access */ for_all_rx_queues(adapter, rxo, i) { - /* below erx HW counter can actually wrap around after - * 65535. Driver accumulates a 32-bit value - */ - accumulate_16bit_val(&rx_stats(rxo)->rx_drops_no_frags, - (u16)erx->rx_drops_no_fragments \ - [rxo->q.id]); + erx_stat = erx->rx_drops_no_fragments[rxo->q.id]; + populate_erx_stats(adapter, rxo, erx_stat); } } } static struct rtnl_link_stats64 *be_get_stats64(struct net_device *netdev, - struct rtnl_link_stats64 *stats) + struct rtnl_link_stats64 *stats) { struct be_adapter *adapter = netdev_priv(netdev); struct be_drv_stats *drvs = &adapter->drv_stats; @@ -499,10 +589,10 @@ static struct rtnl_link_stats64 *be_get_stats64(struct net_device *netdev, for_all_rx_queues(adapter, rxo, i) { const struct be_rx_stats *rx_stats = rx_stats(rxo); do { - start = u64_stats_fetch_begin_bh(&rx_stats->sync); + start = u64_stats_fetch_begin_irq(&rx_stats->sync); pkts = rx_stats(rxo)->rx_pkts; bytes = rx_stats(rxo)->rx_bytes; - } while (u64_stats_fetch_retry_bh(&rx_stats->sync, start)); + } while (u64_stats_fetch_retry_irq(&rx_stats->sync, start)); stats->rx_packets += pkts; stats->rx_bytes += bytes; stats->multicast += rx_stats(rxo)->rx_mcast_pkts; @@ -513,10 +603,10 @@ static struct rtnl_link_stats64 *be_get_stats64(struct net_device *netdev, for_all_tx_queues(adapter, txo, i) { const struct be_tx_stats *tx_stats = tx_stats(txo); do { - start = u64_stats_fetch_begin_bh(&tx_stats->sync); + start = u64_stats_fetch_begin_irq(&tx_stats->sync); pkts = tx_stats(txo)->tx_pkts; bytes = tx_stats(txo)->tx_bytes; - } while (u64_stats_fetch_retry_bh(&tx_stats->sync, start)); + } while (u64_stats_fetch_retry_irq(&tx_stats->sync, start)); stats->tx_packets += pkts; stats->tx_bytes += bytes; } @@ -560,14 +650,15 @@ void be_link_status_update(struct be_adapter *adapter, u8 link_status) adapter->flags |= BE_FLAGS_LINK_STATUS_INIT; } - if ((link_status & LINK_STATUS_MASK) == LINK_UP) + if (link_status) netif_carrier_on(netdev); else netif_carrier_off(netdev); } static void be_tx_stats_update(struct be_tx_obj *txo, - u32 wrb_cnt, u32 copied, u32 gso_segs, bool stopped) + u32 wrb_cnt, u32 copied, u32 gso_segs, + bool stopped) { struct be_tx_stats *stats = tx_stats(txo); @@ -583,7 +674,7 @@ static void be_tx_stats_update(struct be_tx_obj *txo, /* Determine number of WRB entries needed to xmit data in an skb */ static u32 wrb_cnt_for_skb(struct be_adapter *adapter, struct sk_buff *skb, - bool *dummy) + bool *dummy) { int cnt = (skb->len > skb->data_len); @@ -611,7 +702,7 @@ static inline void wrb_fill(struct be_eth_wrb *wrb, u64 addr, int len) } static inline u16 be_get_tx_vlan_tag(struct be_adapter *adapter, - struct sk_buff *skb) + struct sk_buff *skb) { u8 vlan_prio; u16 vlan_tag; @@ -626,15 +717,24 @@ static inline u16 be_get_tx_vlan_tag(struct be_adapter *adapter, return vlan_tag; } -static int be_vlan_tag_chk(struct be_adapter *adapter, struct sk_buff *skb) +/* Used only for IP tunnel packets */ +static u16 skb_inner_ip_proto(struct sk_buff *skb) +{ + return (inner_ip_hdr(skb)->version == 4) ? + inner_ip_hdr(skb)->protocol : inner_ipv6_hdr(skb)->nexthdr; +} + +static u16 skb_ip_proto(struct sk_buff *skb) { - return vlan_tx_tag_present(skb) || adapter->pvid; + return (ip_hdr(skb)->version == 4) ? + ip_hdr(skb)->protocol : ipv6_hdr(skb)->nexthdr; } static void wrb_fill_hdr(struct be_adapter *adapter, struct be_eth_hdr_wrb *hdr, - struct sk_buff *skb, u32 wrb_cnt, u32 len) + struct sk_buff *skb, u32 wrb_cnt, u32 len, + bool skip_hw_vlan) { - u16 vlan_tag; + u16 vlan_tag, proto; memset(hdr, 0, sizeof(*hdr)); @@ -647,9 +747,15 @@ static void wrb_fill_hdr(struct be_adapter *adapter, struct be_eth_hdr_wrb *hdr, if (skb_is_gso_v6(skb) && !lancer_chip(adapter)) AMAP_SET_BITS(struct amap_eth_hdr_wrb, lso6, hdr, 1); } else if (skb->ip_summed == CHECKSUM_PARTIAL) { - if (is_tcp_pkt(skb)) + if (skb->encapsulation) { + AMAP_SET_BITS(struct amap_eth_hdr_wrb, ipcs, hdr, 1); + proto = skb_inner_ip_proto(skb); + } else { + proto = skb_ip_proto(skb); + } + if (proto == IPPROTO_TCP) AMAP_SET_BITS(struct amap_eth_hdr_wrb, tcpcs, hdr, 1); - else if (is_udp_pkt(skb)) + else if (proto == IPPROTO_UDP) AMAP_SET_BITS(struct amap_eth_hdr_wrb, udpcs, hdr, 1); } @@ -659,14 +765,15 @@ static void wrb_fill_hdr(struct be_adapter *adapter, struct be_eth_hdr_wrb *hdr, AMAP_SET_BITS(struct amap_eth_hdr_wrb, vlan_tag, hdr, vlan_tag); } + /* To skip HW VLAN tagging: evt = 1, compl = 0 */ + AMAP_SET_BITS(struct amap_eth_hdr_wrb, complete, hdr, !skip_hw_vlan); AMAP_SET_BITS(struct amap_eth_hdr_wrb, event, hdr, 1); - AMAP_SET_BITS(struct amap_eth_hdr_wrb, complete, hdr, 1); AMAP_SET_BITS(struct amap_eth_hdr_wrb, num_wrb, hdr, wrb_cnt); AMAP_SET_BITS(struct amap_eth_hdr_wrb, len, hdr, len); } static void unmap_tx_frag(struct device *dev, struct be_eth_wrb *wrb, - bool unmap_single) + bool unmap_single) { dma_addr_t dma; @@ -683,7 +790,8 @@ static void unmap_tx_frag(struct device *dev, struct be_eth_wrb *wrb, } static int make_tx_wrbs(struct be_adapter *adapter, struct be_queue_info *txq, - struct sk_buff *skb, u32 wrb_cnt, bool dummy_wrb) + struct sk_buff *skb, u32 wrb_cnt, bool dummy_wrb, + bool skip_hw_vlan) { dma_addr_t busaddr; int i, copied = 0; @@ -712,8 +820,7 @@ static int make_tx_wrbs(struct be_adapter *adapter, struct be_queue_info *txq, } for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { - const struct skb_frag_struct *frag = - &skb_shinfo(skb)->frags[i]; + const struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i]; busaddr = skb_frag_dma_map(dev, frag, 0, skb_frag_size(frag), DMA_TO_DEVICE); if (dma_mapping_error(dev, busaddr)) @@ -732,7 +839,7 @@ static int make_tx_wrbs(struct be_adapter *adapter, struct be_queue_info *txq, queue_head_inc(txq); } - wrb_fill_hdr(adapter, hdr, first_skb, wrb_cnt, copied); + wrb_fill_hdr(adapter, hdr, first_skb, wrb_cnt, copied, skip_hw_vlan); be_dws_cpu_to_le(hdr, sizeof(*hdr)); return copied; @@ -749,7 +856,8 @@ dma_err: } static struct sk_buff *be_insert_vlan_in_pkt(struct be_adapter *adapter, - struct sk_buff *skb) + struct sk_buff *skb, + bool *skip_hw_vlan) { u16 vlan_tag = 0; @@ -757,52 +865,182 @@ static struct sk_buff *be_insert_vlan_in_pkt(struct be_adapter *adapter, if (unlikely(!skb)) return skb; - if (vlan_tx_tag_present(skb)) { + if (vlan_tx_tag_present(skb)) vlan_tag = be_get_tx_vlan_tag(adapter, skb); - __vlan_put_tag(skb, vlan_tag); + + if (qnq_async_evt_rcvd(adapter) && adapter->pvid) { + if (!vlan_tag) + vlan_tag = adapter->pvid; + /* f/w workaround to set skip_hw_vlan = 1, informs the F/W to + * skip VLAN insertion + */ + if (skip_hw_vlan) + *skip_hw_vlan = true; + } + + if (vlan_tag) { + skb = __vlan_put_tag(skb, htons(ETH_P_8021Q), vlan_tag); + if (unlikely(!skb)) + return skb; skb->vlan_tci = 0; } + /* Insert the outer VLAN, if any */ + if (adapter->qnq_vid) { + vlan_tag = adapter->qnq_vid; + skb = __vlan_put_tag(skb, htons(ETH_P_8021Q), vlan_tag); + if (unlikely(!skb)) + return skb; + if (skip_hw_vlan) + *skip_hw_vlan = true; + } + return skb; } -static netdev_tx_t be_xmit(struct sk_buff *skb, - struct net_device *netdev) +static bool be_ipv6_exthdr_check(struct sk_buff *skb) { - struct be_adapter *adapter = netdev_priv(netdev); - struct be_tx_obj *txo = &adapter->tx_obj[skb_get_queue_mapping(skb)]; - struct be_queue_info *txq = &txo->q; - struct iphdr *ip = NULL; - u32 wrb_cnt = 0, copied = 0; - u32 start = txq->head, eth_hdr_len; - bool dummy_wrb, stopped = false; + struct ethhdr *eh = (struct ethhdr *)skb->data; + u16 offset = ETH_HLEN; - eth_hdr_len = ntohs(skb->protocol) == ETH_P_8021Q ? - VLAN_ETH_HLEN : ETH_HLEN; + if (eh->h_proto == htons(ETH_P_IPV6)) { + struct ipv6hdr *ip6h = (struct ipv6hdr *)(skb->data + offset); - /* HW has a bug which considers padding bytes as legal - * and modifies the IPv4 hdr's 'tot_len' field + offset += sizeof(struct ipv6hdr); + if (ip6h->nexthdr != NEXTHDR_TCP && + ip6h->nexthdr != NEXTHDR_UDP) { + struct ipv6_opt_hdr *ehdr = + (struct ipv6_opt_hdr *) (skb->data + offset); + + /* offending pkt: 2nd byte following IPv6 hdr is 0xff */ + if (ehdr->hdrlen == 0xff) + return true; + } + } + return false; +} + +static int be_vlan_tag_tx_chk(struct be_adapter *adapter, struct sk_buff *skb) +{ + return vlan_tx_tag_present(skb) || adapter->pvid || adapter->qnq_vid; +} + +static int be_ipv6_tx_stall_chk(struct be_adapter *adapter, struct sk_buff *skb) +{ + return BE3_chip(adapter) && be_ipv6_exthdr_check(skb); +} + +static struct sk_buff *be_lancer_xmit_workarounds(struct be_adapter *adapter, + struct sk_buff *skb, + bool *skip_hw_vlan) +{ + struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data; + unsigned int eth_hdr_len; + struct iphdr *ip; + + /* For padded packets, BE HW modifies tot_len field in IP header + * incorrecly when VLAN tag is inserted by HW. + * For padded packets, Lancer computes incorrect checksum. */ - if (skb->len <= 60 && be_vlan_tag_chk(adapter, skb) && - is_ipv4_pkt(skb)) { + eth_hdr_len = ntohs(skb->protocol) == ETH_P_8021Q ? + VLAN_ETH_HLEN : ETH_HLEN; + if (skb->len <= 60 && + (lancer_chip(adapter) || vlan_tx_tag_present(skb)) && + is_ipv4_pkt(skb)) { ip = (struct iphdr *)ip_hdr(skb); pskb_trim(skb, eth_hdr_len + ntohs(ip->tot_len)); } + /* If vlan tag is already inlined in the packet, skip HW VLAN + * tagging in pvid-tagging mode + */ + if (be_pvid_tagging_enabled(adapter) && + veh->h_vlan_proto == htons(ETH_P_8021Q)) + *skip_hw_vlan = true; + /* HW has a bug wherein it will calculate CSUM for VLAN * pkts even though it is disabled. * Manually insert VLAN in pkt. */ if (skb->ip_summed != CHECKSUM_PARTIAL && - be_vlan_tag_chk(adapter, skb)) { - skb = be_insert_vlan_in_pkt(adapter, skb); + vlan_tx_tag_present(skb)) { + skb = be_insert_vlan_in_pkt(adapter, skb, skip_hw_vlan); if (unlikely(!skb)) - goto tx_drop; + goto err; + } + + /* HW may lockup when VLAN HW tagging is requested on + * certain ipv6 packets. Drop such pkts if the HW workaround to + * skip HW tagging is not enabled by FW. + */ + if (unlikely(be_ipv6_tx_stall_chk(adapter, skb) && + (adapter->pvid || adapter->qnq_vid) && + !qnq_async_evt_rcvd(adapter))) + goto tx_drop; + + /* Manual VLAN tag insertion to prevent: + * ASIC lockup when the ASIC inserts VLAN tag into + * certain ipv6 packets. Insert VLAN tags in driver, + * and set event, completion, vlan bits accordingly + * in the Tx WRB. + */ + if (be_ipv6_tx_stall_chk(adapter, skb) && + be_vlan_tag_tx_chk(adapter, skb)) { + skb = be_insert_vlan_in_pkt(adapter, skb, skip_hw_vlan); + if (unlikely(!skb)) + goto err; + } + + return skb; +tx_drop: + dev_kfree_skb_any(skb); +err: + return NULL; +} + +static struct sk_buff *be_xmit_workarounds(struct be_adapter *adapter, + struct sk_buff *skb, + bool *skip_hw_vlan) +{ + /* Lancer, SH-R ASICs have a bug wherein Packets that are 32 bytes or + * less may cause a transmit stall on that port. So the work-around is + * to pad short packets (<= 32 bytes) to a 36-byte length. + */ + if (unlikely(!BEx_chip(adapter) && skb->len <= 32)) { + if (skb_padto(skb, 36)) + return NULL; + skb->len = 36; + } + + if (BEx_chip(adapter) || lancer_chip(adapter)) { + skb = be_lancer_xmit_workarounds(adapter, skb, skip_hw_vlan); + if (!skb) + return NULL; + } + + return skb; +} + +static netdev_tx_t be_xmit(struct sk_buff *skb, struct net_device *netdev) +{ + struct be_adapter *adapter = netdev_priv(netdev); + struct be_tx_obj *txo = &adapter->tx_obj[skb_get_queue_mapping(skb)]; + struct be_queue_info *txq = &txo->q; + bool dummy_wrb, stopped = false; + u32 wrb_cnt = 0, copied = 0; + bool skip_hw_vlan = false; + u32 start = txq->head; + + skb = be_xmit_workarounds(adapter, skb, &skip_hw_vlan); + if (!skb) { + tx_stats(txo)->tx_drv_drops++; + return NETDEV_TX_OK; } wrb_cnt = wrb_cnt_for_skb(adapter, skb, &dummy_wrb); - copied = make_tx_wrbs(adapter, txq, skb, wrb_cnt, dummy_wrb); + copied = make_tx_wrbs(adapter, txq, skb, wrb_cnt, dummy_wrb, + skip_hw_vlan); if (copied) { int gso_segs = skb_shinfo(skb)->gso_segs; @@ -821,14 +1059,14 @@ static netdev_tx_t be_xmit(struct sk_buff *skb, stopped = true; } - be_txq_notify(adapter, txq->id, wrb_cnt); + be_txq_notify(adapter, txo, wrb_cnt); be_tx_stats_update(txo, wrb_cnt, copied, gso_segs, stopped); } else { txq->head = start; + tx_stats(txo)->tx_drv_drops++; dev_kfree_skb_any(skb); } -tx_drop: return NETDEV_TX_OK; } @@ -836,16 +1074,15 @@ static int be_change_mtu(struct net_device *netdev, int new_mtu) { struct be_adapter *adapter = netdev_priv(netdev); if (new_mtu < BE_MIN_MTU || - new_mtu > (BE_MAX_JUMBO_FRAME_SIZE - - (ETH_HLEN + ETH_FCS_LEN))) { + new_mtu > (BE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN))) { dev_info(&adapter->pdev->dev, - "MTU must be between %d and %d bytes\n", - BE_MIN_MTU, - (BE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN))); + "MTU must be between %d and %d bytes\n", + BE_MIN_MTU, + (BE_MAX_JUMBO_FRAME_SIZE - (ETH_HLEN + ETH_FCS_LEN))); return -EINVAL; } dev_info(&adapter->pdev->dev, "MTU changed from %d to %d bytes\n", - netdev->mtu, new_mtu); + netdev->mtu, new_mtu); netdev->mtu = new_mtu; return 0; } @@ -857,91 +1094,108 @@ static int be_change_mtu(struct net_device *netdev, int new_mtu) static int be_vid_config(struct be_adapter *adapter) { u16 vids[BE_NUM_VLANS_SUPPORTED]; - u16 num = 0, i; + u16 num = 0, i = 0; int status = 0; /* No need to further configure vids if in promiscuous mode */ if (adapter->promiscuous) return 0; - if (adapter->vlans_added > adapter->max_vlans) + if (adapter->vlans_added > be_max_vlans(adapter)) goto set_vlan_promisc; /* Construct VLAN Table to give to HW */ - for (i = 0; i < VLAN_N_VID; i++) - if (adapter->vlan_tag[i]) - vids[num++] = cpu_to_le16(i); + for_each_set_bit(i, adapter->vids, VLAN_N_VID) + vids[num++] = cpu_to_le16(i); - status = be_cmd_vlan_config(adapter, adapter->if_handle, - vids, num, 1, 0); - - /* Set to VLAN promisc mode as setting VLAN filter failed */ + status = be_cmd_vlan_config(adapter, adapter->if_handle, vids, num); if (status) { - dev_info(&adapter->pdev->dev, "Exhausted VLAN HW filters.\n"); - dev_info(&adapter->pdev->dev, "Disabling HW VLAN filtering.\n"); - goto set_vlan_promisc; + /* Set to VLAN promisc mode as setting VLAN filter failed */ + if (addl_status(status) == + MCC_ADDL_STATUS_INSUFFICIENT_RESOURCES) + goto set_vlan_promisc; + dev_err(&adapter->pdev->dev, + "Setting HW VLAN filtering failed.\n"); + } else { + if (adapter->flags & BE_FLAGS_VLAN_PROMISC) { + /* hw VLAN filtering re-enabled. */ + status = be_cmd_rx_filter(adapter, + BE_FLAGS_VLAN_PROMISC, OFF); + if (!status) { + dev_info(&adapter->pdev->dev, + "Disabling VLAN Promiscuous mode.\n"); + adapter->flags &= ~BE_FLAGS_VLAN_PROMISC; + } + } } return status; set_vlan_promisc: - status = be_cmd_vlan_config(adapter, adapter->if_handle, - NULL, 0, 1, 1); + if (adapter->flags & BE_FLAGS_VLAN_PROMISC) + return 0; + + status = be_cmd_rx_filter(adapter, BE_FLAGS_VLAN_PROMISC, ON); + if (!status) { + dev_info(&adapter->pdev->dev, "Enable VLAN Promiscuous mode\n"); + adapter->flags |= BE_FLAGS_VLAN_PROMISC; + } else + dev_err(&adapter->pdev->dev, + "Failed to enable VLAN Promiscuous mode.\n"); return status; } -static int be_vlan_add_vid(struct net_device *netdev, u16 vid) +static int be_vlan_add_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct be_adapter *adapter = netdev_priv(netdev); int status = 0; - if (!lancer_chip(adapter) && !be_physfn(adapter)) { - status = -EINVAL; - goto ret; - } - /* Packets with VID 0 are always received by Lancer by default */ if (lancer_chip(adapter) && vid == 0) - goto ret; + return status; - adapter->vlan_tag[vid] = 1; - if (adapter->vlans_added <= (adapter->max_vlans + 1)) - status = be_vid_config(adapter); + if (test_bit(vid, adapter->vids)) + return status; + + set_bit(vid, adapter->vids); + adapter->vlans_added++; + + status = be_vid_config(adapter); + if (status) { + adapter->vlans_added--; + clear_bit(vid, adapter->vids); + } - if (!status) - adapter->vlans_added++; - else - adapter->vlan_tag[vid] = 0; -ret: return status; } -static int be_vlan_rem_vid(struct net_device *netdev, u16 vid) +static int be_vlan_rem_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct be_adapter *adapter = netdev_priv(netdev); int status = 0; - if (!lancer_chip(adapter) && !be_physfn(adapter)) { - status = -EINVAL; - goto ret; - } - /* Packets with VID 0 are always received by Lancer by default */ if (lancer_chip(adapter) && vid == 0) goto ret; - adapter->vlan_tag[vid] = 0; - if (adapter->vlans_added <= adapter->max_vlans) - status = be_vid_config(adapter); - + clear_bit(vid, adapter->vids); + status = be_vid_config(adapter); if (!status) adapter->vlans_added--; else - adapter->vlan_tag[vid] = 1; + set_bit(vid, adapter->vids); ret: return status; } +static void be_clear_promisc(struct be_adapter *adapter) +{ + adapter->promiscuous = false; + adapter->flags &= ~(BE_FLAGS_VLAN_PROMISC | BE_FLAGS_MCAST_PROMISC); + + be_cmd_rx_filter(adapter, IFF_PROMISC, OFF); +} + static void be_set_rx_mode(struct net_device *netdev) { struct be_adapter *adapter = netdev_priv(netdev); @@ -955,19 +1209,15 @@ static void be_set_rx_mode(struct net_device *netdev) /* BE was previously in promiscuous mode; disable it */ if (adapter->promiscuous) { - adapter->promiscuous = false; - be_cmd_rx_filter(adapter, IFF_PROMISC, OFF); - + be_clear_promisc(adapter); if (adapter->vlans_added) be_vid_config(adapter); } /* Enable multicast promisc if num configured exceeds what we support */ if (netdev->flags & IFF_ALLMULTI || - netdev_mc_count(netdev) > adapter->max_mcast_mac) { - be_cmd_rx_filter(adapter, IFF_ALLMULTI, ON); - goto done; - } + netdev_mc_count(netdev) > be_max_mc(adapter)) + goto set_mcast_promisc; if (netdev_uc_count(netdev) != adapter->uc_macs) { struct netdev_hw_addr *ha; @@ -978,7 +1228,7 @@ static void be_set_rx_mode(struct net_device *netdev) adapter->pmac_id[i], 0); } - if (netdev_uc_count(netdev) > adapter->max_pmac_cnt) { + if (netdev_uc_count(netdev) > be_max_uc(adapter)) { be_cmd_rx_filter(adapter, IFF_PROMISC, ON); adapter->promiscuous = true; goto done; @@ -993,13 +1243,22 @@ static void be_set_rx_mode(struct net_device *netdev) } status = be_cmd_rx_filter(adapter, IFF_MULTICAST, ON); - - /* Set to MCAST promisc mode if setting MULTICAST address fails */ - if (status) { - dev_info(&adapter->pdev->dev, "Exhausted multicast HW filters.\n"); - dev_info(&adapter->pdev->dev, "Disabling HW multicast filtering.\n"); - be_cmd_rx_filter(adapter, IFF_ALLMULTI, ON); + if (!status) { + if (adapter->flags & BE_FLAGS_MCAST_PROMISC) + adapter->flags &= ~BE_FLAGS_MCAST_PROMISC; + goto done; } + +set_mcast_promisc: + if (adapter->flags & BE_FLAGS_MCAST_PROMISC) + return; + + /* Set to MCAST promisc mode if setting MULTICAST address fails + * or if num configured exceeds what we support + */ + status = be_cmd_rx_filter(adapter, IFF_ALLMULTI, ON); + if (!status) + adapter->flags |= BE_FLAGS_MCAST_PROMISC; done: return; } @@ -1009,9 +1268,6 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) struct be_adapter *adapter = netdev_priv(netdev); struct be_vf_cfg *vf_cfg = &adapter->vf_cfg[vf]; int status; - bool active_mac = false; - u32 pmac_id; - u8 old_mac[ETH_ALEN]; if (!sriov_enabled(adapter)) return -EPERM; @@ -1019,25 +1275,20 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) if (!is_valid_ether_addr(mac) || vf >= adapter->num_vfs) return -EINVAL; - if (lancer_chip(adapter)) { - status = be_cmd_get_mac_from_list(adapter, old_mac, &active_mac, - &pmac_id, vf + 1); - if (!status && active_mac) - be_cmd_pmac_del(adapter, vf_cfg->if_handle, - pmac_id, vf + 1); - - status = be_cmd_set_mac_list(adapter, mac, 1, vf + 1); - } else { - status = be_cmd_pmac_del(adapter, vf_cfg->if_handle, - vf_cfg->pmac_id, vf + 1); + if (BEx_chip(adapter)) { + be_cmd_pmac_del(adapter, vf_cfg->if_handle, vf_cfg->pmac_id, + vf + 1); status = be_cmd_pmac_add(adapter, mac, vf_cfg->if_handle, &vf_cfg->pmac_id, vf + 1); + } else { + status = be_cmd_set_mac(adapter, mac, vf_cfg->if_handle, + vf + 1); } if (status) dev_err(&adapter->pdev->dev, "MAC %pM set on VF %d Failed\n", - mac, vf); + mac, vf); else memcpy(vf_cfg->mac_addr, mac, ETH_ALEN); @@ -1045,7 +1296,7 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) } static int be_get_vf_config(struct net_device *netdev, int vf, - struct ifla_vf_info *vi) + struct ifla_vf_info *vi) { struct be_adapter *adapter = netdev_priv(netdev); struct be_vf_cfg *vf_cfg = &adapter->vf_cfg[vf]; @@ -1057,54 +1308,55 @@ static int be_get_vf_config(struct net_device *netdev, int vf, return -EINVAL; vi->vf = vf; - vi->tx_rate = vf_cfg->tx_rate; - vi->vlan = vf_cfg->vlan_tag; - vi->qos = 0; + vi->max_tx_rate = vf_cfg->tx_rate; + vi->min_tx_rate = 0; + vi->vlan = vf_cfg->vlan_tag & VLAN_VID_MASK; + vi->qos = vf_cfg->vlan_tag >> VLAN_PRIO_SHIFT; memcpy(&vi->mac, vf_cfg->mac_addr, ETH_ALEN); + vi->linkstate = adapter->vf_cfg[vf].plink_tracking; return 0; } -static int be_set_vf_vlan(struct net_device *netdev, - int vf, u16 vlan, u8 qos) +static int be_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos) { struct be_adapter *adapter = netdev_priv(netdev); + struct be_vf_cfg *vf_cfg = &adapter->vf_cfg[vf]; int status = 0; if (!sriov_enabled(adapter)) return -EPERM; - if (vf >= adapter->num_vfs || vlan > 4095) + if (vf >= adapter->num_vfs || vlan > 4095 || qos > 7) return -EINVAL; - if (vlan) { - if (adapter->vf_cfg[vf].vlan_tag != vlan) { - /* If this is new value, program it. Else skip. */ - adapter->vf_cfg[vf].vlan_tag = vlan; - - status = be_cmd_set_hsw_config(adapter, vlan, - vf + 1, adapter->vf_cfg[vf].if_handle); - } + if (vlan || qos) { + vlan |= qos << VLAN_PRIO_SHIFT; + if (vf_cfg->vlan_tag != vlan) + status = be_cmd_set_hsw_config(adapter, vlan, vf + 1, + vf_cfg->if_handle, 0); } else { /* Reset Transparent Vlan Tagging. */ - adapter->vf_cfg[vf].vlan_tag = 0; - vlan = adapter->vf_cfg[vf].def_vid; - status = be_cmd_set_hsw_config(adapter, vlan, vf + 1, - adapter->vf_cfg[vf].if_handle); + status = be_cmd_set_hsw_config(adapter, BE_RESET_VLAN_TAG_ID, + vf + 1, vf_cfg->if_handle, 0); } - - if (status) + if (!status) + vf_cfg->vlan_tag = vlan; + else dev_info(&adapter->pdev->dev, - "VLAN %d config on VF %d failed\n", vlan, vf); + "VLAN %d config on VF %d failed\n", vlan, vf); return status; } -static int be_set_vf_tx_rate(struct net_device *netdev, - int vf, int rate) +static int be_set_vf_tx_rate(struct net_device *netdev, int vf, + int min_tx_rate, int max_tx_rate) { struct be_adapter *adapter = netdev_priv(netdev); - int status = 0; + struct device *dev = &adapter->pdev->dev; + int percent_rate, status = 0; + u16 link_speed = 0; + u8 link_status; if (!sriov_enabled(adapter)) return -EPERM; @@ -1112,100 +1364,148 @@ static int be_set_vf_tx_rate(struct net_device *netdev, if (vf >= adapter->num_vfs) return -EINVAL; - if (rate < 100 || rate > 10000) { - dev_err(&adapter->pdev->dev, - "tx rate must be between 100 and 10000 Mbps\n"); + if (min_tx_rate) return -EINVAL; + + if (!max_tx_rate) + goto config_qos; + + status = be_cmd_link_status_query(adapter, &link_speed, + &link_status, 0); + if (status) + goto err; + + if (!link_status) { + dev_err(dev, "TX-rate setting not allowed when link is down\n"); + status = -EPERM; + goto err; } - if (lancer_chip(adapter)) - status = be_cmd_set_profile_config(adapter, rate / 10, vf + 1); - else - status = be_cmd_set_qos(adapter, rate / 10, vf + 1); + if (max_tx_rate < 100 || max_tx_rate > link_speed) { + dev_err(dev, "TX-rate must be between 100 and %d Mbps\n", + link_speed); + status = -EINVAL; + goto err; + } + /* On Skyhawk the QOS setting must be done only as a % value */ + percent_rate = link_speed / 100; + if (skyhawk_chip(adapter) && (max_tx_rate % percent_rate)) { + dev_err(dev, "TX-rate must be a multiple of %d Mbps\n", + percent_rate); + status = -EINVAL; + goto err; + } + +config_qos: + status = be_cmd_config_qos(adapter, max_tx_rate, link_speed, vf + 1); if (status) - dev_err(&adapter->pdev->dev, - "tx rate %d on VF %d failed\n", rate, vf); - else - adapter->vf_cfg[vf].tx_rate = rate; + goto err; + + adapter->vf_cfg[vf].tx_rate = max_tx_rate; + return 0; + +err: + dev_err(dev, "TX-rate setting of %dMbps on VF%d failed\n", + max_tx_rate, vf); return status; } - -static int be_find_vfs(struct be_adapter *adapter, int vf_state) +static int be_set_vf_link_state(struct net_device *netdev, int vf, + int link_state) { - struct pci_dev *dev, *pdev = adapter->pdev; - int vfs = 0, assigned_vfs = 0, pos; - u16 offset, stride; + struct be_adapter *adapter = netdev_priv(netdev); + int status; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_SRIOV); - if (!pos) - return 0; - pci_read_config_word(pdev, pos + PCI_SRIOV_VF_OFFSET, &offset); - pci_read_config_word(pdev, pos + PCI_SRIOV_VF_STRIDE, &stride); - - dev = pci_get_device(pdev->vendor, PCI_ANY_ID, NULL); - while (dev) { - if (dev->is_virtfn && pci_physfn(dev) == pdev) { - vfs++; - if (dev->dev_flags & PCI_DEV_FLAGS_ASSIGNED) - assigned_vfs++; - } - dev = pci_get_device(pdev->vendor, PCI_ANY_ID, dev); - } - return (vf_state == ASSIGNED) ? assigned_vfs : vfs; + if (!sriov_enabled(adapter)) + return -EPERM; + + if (vf >= adapter->num_vfs) + return -EINVAL; + + status = be_cmd_set_logical_link_config(adapter, link_state, vf+1); + if (!status) + adapter->vf_cfg[vf].plink_tracking = link_state; + + return status; } -static void be_eqd_update(struct be_adapter *adapter, struct be_eq_obj *eqo) +static void be_aic_update(struct be_aic_obj *aic, u64 rx_pkts, u64 tx_pkts, + ulong now) { - struct be_rx_stats *stats = rx_stats(&adapter->rx_obj[eqo->idx]); - ulong now = jiffies; - ulong delta = now - stats->rx_jiffies; - u64 pkts; - unsigned int start, eqd; + aic->rx_pkts_prev = rx_pkts; + aic->tx_reqs_prev = tx_pkts; + aic->jiffies = now; +} - if (!eqo->enable_aic) { - eqd = eqo->eqd; - goto modify_eqd; - } +static void be_eqd_update(struct be_adapter *adapter) +{ + struct be_set_eqd set_eqd[MAX_EVT_QS]; + int eqd, i, num = 0, start; + struct be_aic_obj *aic; + struct be_eq_obj *eqo; + struct be_rx_obj *rxo; + struct be_tx_obj *txo; + u64 rx_pkts, tx_pkts; + ulong now; + u32 pps, delta; - if (eqo->idx >= adapter->num_rx_qs) - return; + for_all_evt_queues(adapter, eqo, i) { + aic = &adapter->aic_obj[eqo->idx]; + if (!aic->enable) { + if (aic->jiffies) + aic->jiffies = 0; + eqd = aic->et_eqd; + goto modify_eqd; + } - stats = rx_stats(&adapter->rx_obj[eqo->idx]); + rxo = &adapter->rx_obj[eqo->idx]; + do { + start = u64_stats_fetch_begin_irq(&rxo->stats.sync); + rx_pkts = rxo->stats.rx_pkts; + } while (u64_stats_fetch_retry_irq(&rxo->stats.sync, start)); - /* Wrapped around */ - if (time_before(now, stats->rx_jiffies)) { - stats->rx_jiffies = now; - return; - } + txo = &adapter->tx_obj[eqo->idx]; + do { + start = u64_stats_fetch_begin_irq(&txo->stats.sync); + tx_pkts = txo->stats.tx_reqs; + } while (u64_stats_fetch_retry_irq(&txo->stats.sync, start)); - /* Update once a second */ - if (delta < HZ) - return; - do { - start = u64_stats_fetch_begin_bh(&stats->sync); - pkts = stats->rx_pkts; - } while (u64_stats_fetch_retry_bh(&stats->sync, start)); - - stats->rx_pps = (unsigned long)(pkts - stats->rx_pkts_prev) / (delta / HZ); - stats->rx_pkts_prev = pkts; - stats->rx_jiffies = now; - eqd = (stats->rx_pps / 110000) << 3; - eqd = min(eqd, eqo->max_eqd); - eqd = max(eqd, eqo->min_eqd); - if (eqd < 10) - eqd = 0; + /* Skip, if wrapped around or first calculation */ + now = jiffies; + if (!aic->jiffies || time_before(now, aic->jiffies) || + rx_pkts < aic->rx_pkts_prev || + tx_pkts < aic->tx_reqs_prev) { + be_aic_update(aic, rx_pkts, tx_pkts, now); + continue; + } + + delta = jiffies_to_msecs(now - aic->jiffies); + pps = (((u32)(rx_pkts - aic->rx_pkts_prev) * 1000) / delta) + + (((u32)(tx_pkts - aic->tx_reqs_prev) * 1000) / delta); + eqd = (pps / 15000) << 2; + if (eqd < 8) + eqd = 0; + eqd = min_t(u32, eqd, aic->max_eqd); + eqd = max_t(u32, eqd, aic->min_eqd); + + be_aic_update(aic, rx_pkts, tx_pkts, now); modify_eqd: - if (eqd != eqo->cur_eqd) { - be_cmd_modify_eqd(adapter, eqo->q.id, eqd); - eqo->cur_eqd = eqd; + if (eqd != aic->prev_eqd) { + set_eqd[num].delay_multiplier = (eqd * 65)/100; + set_eqd[num].eq_id = eqo->q.id; + aic->prev_eqd = eqd; + num++; + } } + + if (num) + be_cmd_modify_eqd(adapter, set_eqd, num); } static void be_rx_stats_update(struct be_rx_obj *rxo, - struct be_rx_compl_info *rxcp) + struct be_rx_compl_info *rxcp) { struct be_rx_stats *stats = rx_stats(rxo); @@ -1223,28 +1523,34 @@ static void be_rx_stats_update(struct be_rx_obj *rxo, static inline bool csum_passed(struct be_rx_compl_info *rxcp) { /* L4 checksum is not reliable for non TCP/UDP packets. - * Also ignore ipcksm for ipv6 pkts */ + * Also ignore ipcksm for ipv6 pkts + */ return (rxcp->tcpf || rxcp->udpf) && rxcp->l4_csum && - (rxcp->ip_csum || rxcp->ipv6); + (rxcp->ip_csum || rxcp->ipv6) && !rxcp->err; } -static struct be_rx_page_info *get_rx_page_info(struct be_rx_obj *rxo, - u16 frag_idx) +static struct be_rx_page_info *get_rx_page_info(struct be_rx_obj *rxo) { struct be_adapter *adapter = rxo->adapter; struct be_rx_page_info *rx_page_info; struct be_queue_info *rxq = &rxo->q; + u16 frag_idx = rxq->tail; rx_page_info = &rxo->page_info_tbl[frag_idx]; BUG_ON(!rx_page_info->page); - if (rx_page_info->last_page_user) { + if (rx_page_info->last_frag) { dma_unmap_page(&adapter->pdev->dev, dma_unmap_addr(rx_page_info, bus), adapter->big_page_size, DMA_FROM_DEVICE); - rx_page_info->last_page_user = false; + rx_page_info->last_frag = false; + } else { + dma_sync_single_for_cpu(&adapter->pdev->dev, + dma_unmap_addr(rx_page_info, bus), + rx_frag_size, DMA_FROM_DEVICE); } + queue_tail_inc(rxq); atomic_dec(&rxq->used); return rx_page_info; } @@ -1253,15 +1559,13 @@ static struct be_rx_page_info *get_rx_page_info(struct be_rx_obj *rxo, static void be_rx_compl_discard(struct be_rx_obj *rxo, struct be_rx_compl_info *rxcp) { - struct be_queue_info *rxq = &rxo->q; struct be_rx_page_info *page_info; u16 i, num_rcvd = rxcp->num_rcvd; for (i = 0; i < num_rcvd; i++) { - page_info = get_rx_page_info(rxo, rxcp->rxq_idx); + page_info = get_rx_page_info(rxo); put_page(page_info->page); memset(page_info, 0, sizeof(*page_info)); - index_inc(&rxcp->rxq_idx, rxq->len); } } @@ -1272,13 +1576,12 @@ static void be_rx_compl_discard(struct be_rx_obj *rxo, static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb, struct be_rx_compl_info *rxcp) { - struct be_queue_info *rxq = &rxo->q; struct be_rx_page_info *page_info; u16 i, j; u16 hdr_len, curr_frag_len, remaining; u8 *start; - page_info = get_rx_page_info(rxo, rxcp->rxq_idx); + page_info = get_rx_page_info(rxo); start = page_address(page_info->page) + page_info->page_offset; prefetch(start); @@ -1299,7 +1602,8 @@ static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb, skb_frag_set_page(skb, 0, page_info->page); skb_shinfo(skb)->frags[0].page_offset = page_info->page_offset + hdr_len; - skb_frag_size_set(&skb_shinfo(skb)->frags[0], curr_frag_len - hdr_len); + skb_frag_size_set(&skb_shinfo(skb)->frags[0], + curr_frag_len - hdr_len); skb->data_len = curr_frag_len - hdr_len; skb->truesize += rx_frag_size; skb->tail += hdr_len; @@ -1312,10 +1616,9 @@ static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb, } /* More frags present for this completion */ - index_inc(&rxcp->rxq_idx, rxq->len); remaining = rxcp->pkt_size - curr_frag_len; for (i = 1, j = 0; i < rxcp->num_rcvd; i++) { - page_info = get_rx_page_info(rxo, rxcp->rxq_idx); + page_info = get_rx_page_info(rxo); curr_frag_len = min(remaining, rx_frag_size); /* Coalesce all frags from the same physical page in one slot */ @@ -1336,14 +1639,13 @@ static void skb_fill_rx_data(struct be_rx_obj *rxo, struct sk_buff *skb, skb->data_len += curr_frag_len; skb->truesize += rx_frag_size; remaining -= curr_frag_len; - index_inc(&rxcp->rxq_idx, rxq->len); page_info->page = NULL; } BUG_ON(j > MAX_SKB_FRAGS); } /* Process the RX completion indicated by rxcp when GRO is disabled */ -static void be_rx_compl_process(struct be_rx_obj *rxo, +static void be_rx_compl_process(struct be_rx_obj *rxo, struct napi_struct *napi, struct be_rx_compl_info *rxcp) { struct be_adapter *adapter = rxo->adapter; @@ -1367,23 +1669,25 @@ static void be_rx_compl_process(struct be_rx_obj *rxo, skb->protocol = eth_type_trans(skb, netdev); skb_record_rx_queue(skb, rxo - &adapter->rx_obj[0]); if (netdev->features & NETIF_F_RXHASH) - skb->rxhash = rxcp->rss_hash; + skb_set_hash(skb, rxcp->rss_hash, PKT_HASH_TYPE_L3); + skb->encapsulation = rxcp->tunneled; + skb_mark_napi_id(skb, napi); if (rxcp->vlanf) - __vlan_hwaccel_put_tag(skb, rxcp->vlan_tag); + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), rxcp->vlan_tag); netif_receive_skb(skb); } /* Process the RX completion indicated by rxcp when GRO is enabled */ -void be_rx_compl_process_gro(struct be_rx_obj *rxo, struct napi_struct *napi, - struct be_rx_compl_info *rxcp) +static void be_rx_compl_process_gro(struct be_rx_obj *rxo, + struct napi_struct *napi, + struct be_rx_compl_info *rxcp) { struct be_adapter *adapter = rxo->adapter; struct be_rx_page_info *page_info; struct sk_buff *skb = NULL; - struct be_queue_info *rxq = &rxo->q; u16 remaining, curr_frag_len; u16 i, j; @@ -1395,7 +1699,7 @@ void be_rx_compl_process_gro(struct be_rx_obj *rxo, struct napi_struct *napi, remaining = rxcp->pkt_size; for (i = 0, j = -1; i < rxcp->num_rcvd; i++) { - page_info = get_rx_page_info(rxo, rxcp->rxq_idx); + page_info = get_rx_page_info(rxo); curr_frag_len = min(remaining, rx_frag_size); @@ -1413,7 +1717,6 @@ void be_rx_compl_process_gro(struct be_rx_obj *rxo, struct napi_struct *napi, skb_frag_size_add(&skb_shinfo(skb)->frags[j], curr_frag_len); skb->truesize += rx_frag_size; remaining -= curr_frag_len; - index_inc(&rxcp->rxq_idx, rxq->len); memset(page_info, 0, sizeof(*page_info)); } BUG_ON(j > MAX_SKB_FRAGS); @@ -1424,10 +1727,13 @@ void be_rx_compl_process_gro(struct be_rx_obj *rxo, struct napi_struct *napi, skb->ip_summed = CHECKSUM_UNNECESSARY; skb_record_rx_queue(skb, rxo - &adapter->rx_obj[0]); if (adapter->netdev->features & NETIF_F_RXHASH) - skb->rxhash = rxcp->rss_hash; + skb_set_hash(skb, rxcp->rss_hash, PKT_HASH_TYPE_L3); + + skb->encapsulation = rxcp->tunneled; + skb_mark_napi_id(skb, napi); if (rxcp->vlanf) - __vlan_hwaccel_put_tag(skb, rxcp->vlan_tag); + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), rxcp->vlan_tag); napi_gro_frags(napi); } @@ -1447,8 +1753,6 @@ static void be_parse_rx_compl_v1(struct be_eth_rx_compl *compl, AMAP_GET_BITS(struct amap_eth_rx_compl_v1, l4_cksm, compl); rxcp->ipv6 = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, ip_version, compl); - rxcp->rxq_idx = - AMAP_GET_BITS(struct amap_eth_rx_compl_v1, fragndx, compl); rxcp->num_rcvd = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, numfrags, compl); rxcp->pkt_type = @@ -1456,12 +1760,14 @@ static void be_parse_rx_compl_v1(struct be_eth_rx_compl *compl, rxcp->rss_hash = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, rsshash, compl); if (rxcp->vlanf) { - rxcp->vtm = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, vtm, + rxcp->qnq = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, qnq, compl); - rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, vlan_tag, - compl); + rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, + vlan_tag, compl); } rxcp->port = AMAP_GET_BITS(struct amap_eth_rx_compl_v1, port, compl); + rxcp->tunneled = + AMAP_GET_BITS(struct amap_eth_rx_compl_v1, tunneled, compl); } static void be_parse_rx_compl_v0(struct be_eth_rx_compl *compl, @@ -1479,8 +1785,6 @@ static void be_parse_rx_compl_v0(struct be_eth_rx_compl *compl, AMAP_GET_BITS(struct amap_eth_rx_compl_v0, l4_cksm, compl); rxcp->ipv6 = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, ip_version, compl); - rxcp->rxq_idx = - AMAP_GET_BITS(struct amap_eth_rx_compl_v0, fragndx, compl); rxcp->num_rcvd = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, numfrags, compl); rxcp->pkt_type = @@ -1488,12 +1792,14 @@ static void be_parse_rx_compl_v0(struct be_eth_rx_compl *compl, rxcp->rss_hash = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, rsshash, compl); if (rxcp->vlanf) { - rxcp->vtm = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, vtm, + rxcp->qnq = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, qnq, compl); - rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, vlan_tag, - compl); + rxcp->vlan_tag = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, + vlan_tag, compl); } rxcp->port = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, port, compl); + rxcp->ip_frag = AMAP_GET_BITS(struct amap_eth_rx_compl_v0, + ip_frag, compl); } static struct be_rx_compl_info *be_rx_compl_get(struct be_rx_obj *rxo) @@ -1515,17 +1821,22 @@ static struct be_rx_compl_info *be_rx_compl_get(struct be_rx_obj *rxo) else be_parse_rx_compl_v0(compl, rxcp); + if (rxcp->ip_frag) + rxcp->l4_csum = 0; + if (rxcp->vlanf) { - /* vlanf could be wrongly set in some cards. - * ignore if vtm is not set */ - if ((adapter->function_mode & FLEX10_MODE) && !rxcp->vtm) + /* In QNQ modes, if qnq bit is not set, then the packet was + * tagged only with the transparent outer vlan-tag and must + * not be treated as a vlan packet by host + */ + if (be_is_qnq_mode(adapter) && !rxcp->qnq) rxcp->vlanf = 0; if (!lancer_chip(adapter)) rxcp->vlan_tag = swab16(rxcp->vlan_tag); if (adapter->pvid == (rxcp->vlan_tag & VLAN_VID_MASK) && - !adapter->vlan_tag[rxcp->vlan_tag]) + !test_bit(rxcp->vlan_tag, adapter->vids)) rxcp->vlanf = 0; } @@ -1555,6 +1866,7 @@ static void be_post_rx_frags(struct be_rx_obj *rxo, gfp_t gfp) struct be_rx_page_info *page_info = NULL, *prev_page_info = NULL; struct be_queue_info *rxq = &rxo->q; struct page *pagep = NULL; + struct device *dev = &adapter->pdev->dev; struct be_eth_rx_d *rxd; u64 page_dmaaddr = 0, frag_dmaaddr; u32 posted, page_offset = 0; @@ -1567,20 +1879,25 @@ static void be_post_rx_frags(struct be_rx_obj *rxo, gfp_t gfp) rx_stats(rxo)->rx_post_fail++; break; } - page_dmaaddr = dma_map_page(&adapter->pdev->dev, pagep, - 0, adapter->big_page_size, + page_dmaaddr = dma_map_page(dev, pagep, 0, + adapter->big_page_size, DMA_FROM_DEVICE); - page_info->page_offset = 0; + if (dma_mapping_error(dev, page_dmaaddr)) { + put_page(pagep); + pagep = NULL; + rx_stats(rxo)->rx_post_fail++; + break; + } + page_offset = 0; } else { get_page(pagep); - page_info->page_offset = page_offset + rx_frag_size; + page_offset += rx_frag_size; } - page_offset = page_info->page_offset; + page_info->page_offset = page_offset; page_info->page = pagep; - dma_unmap_addr_set(page_info, bus, page_dmaaddr); - frag_dmaaddr = page_dmaaddr + page_info->page_offset; rxd = queue_head_node(rxq); + frag_dmaaddr = page_dmaaddr + page_info->page_offset; rxd->fragpa_lo = cpu_to_le32(frag_dmaaddr & 0xFFFFFFFF); rxd->fragpa_hi = cpu_to_le32(upper_32_bits(frag_dmaaddr)); @@ -1588,18 +1905,29 @@ static void be_post_rx_frags(struct be_rx_obj *rxo, gfp_t gfp) if ((page_offset + rx_frag_size + rx_frag_size) > adapter->big_page_size) { pagep = NULL; - page_info->last_page_user = true; + page_info->last_frag = true; + dma_unmap_addr_set(page_info, bus, page_dmaaddr); + } else { + dma_unmap_addr_set(page_info, bus, frag_dmaaddr); } prev_page_info = page_info; queue_head_inc(rxq); page_info = &rxo->page_info_tbl[rxq->head]; } - if (pagep) - prev_page_info->last_page_user = true; + + /* Mark the last frag of a page when we break out of the above loop + * with no more slots available in the RXQ + */ + if (pagep) { + prev_page_info->last_frag = true; + dma_unmap_addr_set(prev_page_info, bus, page_dmaaddr); + } if (posted) { atomic_add(posted, &rxq->used); + if (rxo->rx_post_starved) + rxo->rx_post_starved = false; be_rxq_notify(adapter, rxq->id, posted); } else if (atomic_read(&rxq->used) == 0) { /* Let be_worker replenish when memory is available */ @@ -1624,7 +1952,7 @@ static struct be_eth_tx_compl *be_tx_compl_get(struct be_queue_info *tx_cq) } static u16 be_tx_compl_process(struct be_adapter *adapter, - struct be_tx_obj *txo, u16 last_index) + struct be_tx_obj *txo, u16 last_index) { struct be_queue_info *txq = &txo->q; struct be_eth_wrb *wrb; @@ -1651,7 +1979,7 @@ static u16 be_tx_compl_process(struct be_adapter *adapter, queue_tail_inc(txq); } while (cur_index != last_index); - kfree_skb(sent_skb); + dev_kfree_skb_any(sent_skb); return num_wrbs; } @@ -1691,7 +2019,6 @@ static void be_rx_cq_clean(struct be_rx_obj *rxo) struct be_rx_compl_info *rxcp; struct be_adapter *adapter = rxo->adapter; int flush_wait = 0; - u16 tail; /* Consume pending rx completions. * Wait for the flush completion (identified by zero num_rcvd) @@ -1714,7 +2041,7 @@ static void be_rx_cq_clean(struct be_rx_obj *rxo) mdelay(1); } else { be_rx_compl_discard(rxo, rxcp); - be_cq_notify(adapter, rx_cq->id, true, 1); + be_cq_notify(adapter, rx_cq->id, false, 1); if (rxcp->num_rcvd == 0) break; } @@ -1724,9 +2051,8 @@ static void be_rx_cq_clean(struct be_rx_obj *rxo) be_cq_notify(adapter, rx_cq->id, false, 0); /* Then free posted rx buffers that were not used */ - tail = (rxq->head + rxq->len - atomic_read(&rxq->used)) % rxq->len; - for (; atomic_read(&rxq->used) > 0; index_inc(&tail, rxq->len)) { - page_info = get_rx_page_info(rxo, tail); + while (atomic_read(&rxq->used) > 0) { + page_info = get_rx_page_info(rxo); put_page(page_info->page); memset(page_info, 0, sizeof(*page_info)); } @@ -1744,11 +2070,13 @@ static void be_tx_compl_clean(struct be_adapter *adapter) bool dummy_wrb; int i, pending_txqs; - /* Wait for a max of 200ms for all the tx-completions to arrive. */ + /* Stop polling for compls when HW has been silent for 10ms */ do { pending_txqs = adapter->num_tx_qs; for_all_tx_queues(adapter, txo, i) { + cmpl = 0; + num_wrbs = 0; txq = &txo->q; while ((txcp = be_tx_compl_get(&txo->cq))) { end_idx = @@ -1761,14 +2089,13 @@ static void be_tx_compl_clean(struct be_adapter *adapter) if (cmpl) { be_cq_notify(adapter, txo->cq.id, false, cmpl); atomic_sub(num_wrbs, &txq->used); - cmpl = 0; - num_wrbs = 0; + timeo = 0; } if (atomic_read(&txq->used) == 0) pending_txqs--; } - if (pending_txqs == 0 || ++timeo > 200) + if (pending_txqs == 0 || ++timeo > 10 || be_hw_error(adapter)) break; mdelay(1); @@ -1802,6 +2129,8 @@ static void be_evt_queues_destroy(struct be_adapter *adapter) if (eqo->q.created) { be_eq_clean(eqo); be_cmd_q_destroy(adapter, &eqo->q, QTYPE_EQ); + napi_hash_del(&eqo->napi); + netif_napi_del(&eqo->napi); } be_queue_free(adapter, &eqo->q); } @@ -1811,24 +2140,30 @@ static int be_evt_queues_create(struct be_adapter *adapter) { struct be_queue_info *eq; struct be_eq_obj *eqo; + struct be_aic_obj *aic; int i, rc; - adapter->num_evt_qs = num_irqs(adapter); + adapter->num_evt_qs = min_t(u16, num_irqs(adapter), + adapter->cfg_num_qs); for_all_evt_queues(adapter, eqo, i) { + netif_napi_add(adapter->netdev, &eqo->napi, be_poll, + BE_NAPI_WEIGHT); + napi_hash_add(&eqo->napi); + aic = &adapter->aic_obj[i]; eqo->adapter = adapter; eqo->tx_budget = BE_TX_BUDGET; eqo->idx = i; - eqo->max_eqd = BE_MAX_EQD; - eqo->enable_aic = true; + aic->max_eqd = BE_MAX_EQD; + aic->enable = true; eq = &eqo->q; rc = be_queue_alloc(adapter, eq, EVNT_Q_LEN, - sizeof(struct be_eq_entry)); + sizeof(struct be_eq_entry)); if (rc) return rc; - rc = be_cmd_eq_create(adapter, eq, eqo->cur_eqd); + rc = be_cmd_eq_create(adapter, eqo); if (rc) return rc; } @@ -1857,7 +2192,7 @@ static int be_mcc_queues_create(struct be_adapter *adapter) cq = &adapter->mcc_obj.cq; if (be_queue_alloc(adapter, cq, MCC_CQ_LEN, - sizeof(struct be_mcc_compl))) + sizeof(struct be_mcc_compl))) goto err; /* Use the default EQ for MCC completions */ @@ -1902,31 +2237,13 @@ static void be_tx_queues_destroy(struct be_adapter *adapter) } } -static int be_num_txqs_want(struct be_adapter *adapter) -{ - if ((!lancer_chip(adapter) && sriov_want(adapter)) || - be_is_mc(adapter) || - (!lancer_chip(adapter) && !be_physfn(adapter)) || - BE2_chip(adapter)) - return 1; - else - return adapter->max_tx_queues; -} - -static int be_tx_cqs_create(struct be_adapter *adapter) +static int be_tx_qs_create(struct be_adapter *adapter) { struct be_queue_info *cq, *eq; - int status; struct be_tx_obj *txo; - u8 i; + int status, i; - adapter->num_tx_qs = be_num_txqs_want(adapter); - if (adapter->num_tx_qs != MAX_TX_QS) { - rtnl_lock(); - netif_set_real_num_tx_queues(adapter->netdev, - adapter->num_tx_qs); - rtnl_unlock(); - } + adapter->num_tx_qs = min(adapter->num_evt_qs, be_max_txqs(adapter)); for_all_tx_queues(adapter, txo, i) { cq = &txo->cq; @@ -1935,6 +2252,9 @@ static int be_tx_cqs_create(struct be_adapter *adapter) if (status) return status; + u64_stats_init(&txo->stats.sync); + u64_stats_init(&txo->stats.sync_compl); + /* If num_evt_qs is less than num_tx_qs, then more than * one txq share an eq */ @@ -1942,22 +2262,13 @@ static int be_tx_cqs_create(struct be_adapter *adapter) status = be_cmd_cq_create(adapter, cq, eq, false, 3); if (status) return status; - } - return 0; -} - -static int be_tx_qs_create(struct be_adapter *adapter) -{ - struct be_tx_obj *txo; - int i, status; - for_all_tx_queues(adapter, txo, i) { status = be_queue_alloc(adapter, &txo->q, TX_Q_LEN, sizeof(struct be_eth_wrb)); if (status) return status; - status = be_cmd_txq_create(adapter, &txo->q, &txo->cq); + status = be_cmd_txq_create(adapter, txo); if (status) return status; } @@ -1987,27 +2298,25 @@ static int be_rx_cqs_create(struct be_adapter *adapter) struct be_rx_obj *rxo; int rc, i; - /* We'll create as many RSS rings as there are irqs. - * But when there's only one irq there's no use creating RSS rings + /* We can create as many RSS rings as there are EQs. */ + adapter->num_rx_qs = adapter->num_evt_qs; + + /* We'll use RSS only if atleast 2 RSS rings are supported. + * When RSS is used, we'll need a default RXQ for non-IP traffic. */ - adapter->num_rx_qs = (num_irqs(adapter) > 1) ? - num_irqs(adapter) + 1 : 1; - if (adapter->num_rx_qs != MAX_RX_QS) { - rtnl_lock(); - netif_set_real_num_rx_queues(adapter->netdev, - adapter->num_rx_qs); - rtnl_unlock(); - } + if (adapter->num_rx_qs > 1) + adapter->num_rx_qs++; adapter->big_page_size = (1 << get_order(rx_frag_size)) * PAGE_SIZE; for_all_rx_queues(adapter, rxo, i) { rxo->adapter = adapter; cq = &rxo->cq; rc = be_queue_alloc(adapter, cq, RX_CQ_LEN, - sizeof(struct be_eth_rx_compl)); + sizeof(struct be_eth_rx_compl)); if (rc) return rc; + u64_stats_init(&rxo->stats.sync); eq = &adapter->eq_obj[i % adapter->num_evt_qs].q; rc = be_cmd_cq_create(adapter, cq, eq, false, 3); if (rc) @@ -2063,11 +2372,11 @@ static irqreturn_t be_msix(int irq, void *dev) static inline bool do_gro(struct be_rx_compl_info *rxcp) { - return (rxcp->tcpf && !rxcp->err) ? true : false; + return (rxcp->tcpf && !rxcp->err && rxcp->l4_csum) ? true : false; } static int be_process_rx(struct be_rx_obj *rxo, struct napi_struct *napi, - int budget) + int budget, int polling) { struct be_adapter *adapter = rxo->adapter; struct be_queue_info *rx_cq = &rxo->cq; @@ -2093,15 +2402,17 @@ static int be_process_rx(struct be_rx_obj *rxo, struct napi_struct *napi, * promiscuous mode on some skews */ if (unlikely(rxcp->port != adapter->port_num && - !lancer_chip(adapter))) { + !lancer_chip(adapter))) { be_rx_compl_discard(rxo, rxcp); goto loop_continue; } - if (do_gro(rxcp)) + /* Don't do gro when we're busy_polling */ + if (do_gro(rxcp) && polling != BUSY_POLLING) be_rx_compl_process_gro(rxo, napi, rxcp); else - be_rx_compl_process(rxo, rxcp); + be_rx_compl_process(rxo, napi, rxcp); + loop_continue: be_rx_stats_update(rxo, rxcp); } @@ -2109,7 +2420,11 @@ loop_continue: if (work_done) { be_cq_notify(adapter, rx_cq->id, true, work_done); - if (atomic_read(&rxo->q.used) < RX_FRAGS_REFILL_WM) + /* When an rx-obj gets into post_starved state, just + * let be_worker do the posting. + */ + if (atomic_read(&rxo->q.used) < RX_FRAGS_REFILL_WM && + !rxo->rx_post_starved) be_post_rx_frags(rxo, GFP_ATOMIC); } @@ -2127,8 +2442,9 @@ static bool be_process_tx(struct be_adapter *adapter, struct be_tx_obj *txo, if (!txcp) break; num_wrbs += be_tx_compl_process(adapter, txo, - AMAP_GET_BITS(struct amap_eth_tx_compl, - wrb_index, txcp)); + AMAP_GET_BITS(struct + amap_eth_tx_compl, + wrb_index, txcp)); } if (work_done) { @@ -2138,7 +2454,7 @@ static bool be_process_tx(struct be_adapter *adapter, struct be_tx_obj *txo, /* As Tx wrbs have been freed up, wake up netdev queue * if it was stopped due to lack of tx wrbs. */ if (__netif_subqueue_stopped(adapter->netdev, idx) && - atomic_read(&txo->q.used) < txo->q.len / 2) { + atomic_read(&txo->q.used) < txo->q.len / 2) { netif_wake_subqueue(adapter->netdev, idx); } @@ -2154,6 +2470,7 @@ int be_poll(struct napi_struct *napi, int budget) struct be_eq_obj *eqo = container_of(napi, struct be_eq_obj, napi); struct be_adapter *adapter = eqo->adapter; int max_work = 0, work, i, num_evts; + struct be_rx_obj *rxo; bool tx_done; num_evts = events_get(eqo); @@ -2166,13 +2483,18 @@ int be_poll(struct napi_struct *napi, int budget) max_work = budget; } - /* This loop will iterate twice for EQ0 in which - * completions of the last RXQ (default one) are also processed - * For other EQs the loop iterates only once - */ - for (i = eqo->idx; i < adapter->num_rx_qs; i += adapter->num_evt_qs) { - work = be_process_rx(&adapter->rx_obj[i], napi, budget); - max_work = max(work, max_work); + if (be_lock_napi(eqo)) { + /* This loop will iterate twice for EQ0 in which + * completions of the last RXQ (default one) are also processed + * For other EQs the loop iterates only once + */ + for_all_rx_queues_on_eq(adapter, eqo, rxo, i) { + work = be_process_rx(rxo, napi, budget, NAPI_POLLING); + max_work = max(work, max_work); + } + be_unlock_napi(eqo); + } else { + max_work = budget; } if (is_mcc_eqo(eqo)) @@ -2188,11 +2510,36 @@ int be_poll(struct napi_struct *napi, int budget) return max_work; } +#ifdef CONFIG_NET_RX_BUSY_POLL +static int be_busy_poll(struct napi_struct *napi) +{ + struct be_eq_obj *eqo = container_of(napi, struct be_eq_obj, napi); + struct be_adapter *adapter = eqo->adapter; + struct be_rx_obj *rxo; + int i, work = 0; + + if (!be_lock_busy_poll(eqo)) + return LL_FLUSH_BUSY; + + for_all_rx_queues_on_eq(adapter, eqo, rxo, i) { + work = be_process_rx(rxo, napi, 4, BUSY_POLLING); + if (work) + break; + } + + be_unlock_busy_poll(eqo); + return work; +} +#endif + void be_detect_error(struct be_adapter *adapter) { u32 ue_lo = 0, ue_hi = 0, ue_lo_mask = 0, ue_hi_mask = 0; u32 sliport_status = 0, sliport_err1 = 0, sliport_err2 = 0; u32 i; + bool error_detected = false; + struct device *dev = &adapter->pdev->dev; + struct net_device *netdev = adapter->netdev; if (be_hw_error(adapter)) return; @@ -2201,59 +2548,64 @@ void be_detect_error(struct be_adapter *adapter) sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET); if (sliport_status & SLIPORT_STATUS_ERR_MASK) { sliport_err1 = ioread32(adapter->db + - SLIPORT_ERROR1_OFFSET); + SLIPORT_ERROR1_OFFSET); sliport_err2 = ioread32(adapter->db + - SLIPORT_ERROR2_OFFSET); + SLIPORT_ERROR2_OFFSET); + adapter->hw_error = true; + /* Do not log error messages if its a FW reset */ + if (sliport_err1 == SLIPORT_ERROR_FW_RESET1 && + sliport_err2 == SLIPORT_ERROR_FW_RESET2) { + dev_info(dev, "Firmware update in progress\n"); + } else { + error_detected = true; + dev_err(dev, "Error detected in the card\n"); + dev_err(dev, "ERR: sliport status 0x%x\n", + sliport_status); + dev_err(dev, "ERR: sliport error1 0x%x\n", + sliport_err1); + dev_err(dev, "ERR: sliport error2 0x%x\n", + sliport_err2); + } } } else { pci_read_config_dword(adapter->pdev, - PCICFG_UE_STATUS_LOW, &ue_lo); + PCICFG_UE_STATUS_LOW, &ue_lo); pci_read_config_dword(adapter->pdev, - PCICFG_UE_STATUS_HIGH, &ue_hi); + PCICFG_UE_STATUS_HIGH, &ue_hi); pci_read_config_dword(adapter->pdev, - PCICFG_UE_STATUS_LOW_MASK, &ue_lo_mask); + PCICFG_UE_STATUS_LOW_MASK, &ue_lo_mask); pci_read_config_dword(adapter->pdev, - PCICFG_UE_STATUS_HI_MASK, &ue_hi_mask); + PCICFG_UE_STATUS_HI_MASK, &ue_hi_mask); ue_lo = (ue_lo & ~ue_lo_mask); ue_hi = (ue_hi & ~ue_hi_mask); - } - /* On certain platforms BE hardware can indicate spurious UEs. - * Allow the h/w to stop working completely in case of a real UE. - * Hence not setting the hw_error for UE detection. - */ - if (sliport_status & SLIPORT_STATUS_ERR_MASK) { - adapter->hw_error = true; - dev_err(&adapter->pdev->dev, - "Error detected in the card\n"); - } - - if (sliport_status & SLIPORT_STATUS_ERR_MASK) { - dev_err(&adapter->pdev->dev, - "ERR: sliport status 0x%x\n", sliport_status); - dev_err(&adapter->pdev->dev, - "ERR: sliport error1 0x%x\n", sliport_err1); - dev_err(&adapter->pdev->dev, - "ERR: sliport error2 0x%x\n", sliport_err2); - } - - if (ue_lo) { - for (i = 0; ue_lo; ue_lo >>= 1, i++) { - if (ue_lo & 1) - dev_err(&adapter->pdev->dev, - "UE: %s bit set\n", ue_status_low_desc[i]); - } - } + /* On certain platforms BE hardware can indicate spurious UEs. + * Allow HW to stop working completely in case of a real UE. + * Hence not setting the hw_error for UE detection. + */ - if (ue_hi) { - for (i = 0; ue_hi; ue_hi >>= 1, i++) { - if (ue_hi & 1) - dev_err(&adapter->pdev->dev, - "UE: %s bit set\n", ue_status_hi_desc[i]); + if (ue_lo || ue_hi) { + error_detected = true; + dev_err(dev, + "Unrecoverable Error detected in the adapter"); + dev_err(dev, "Please reboot server to recover"); + if (skyhawk_chip(adapter)) + adapter->hw_error = true; + for (i = 0; ue_lo; ue_lo >>= 1, i++) { + if (ue_lo & 1) + dev_err(dev, "UE: %s bit set\n", + ue_status_low_desc[i]); + } + for (i = 0; ue_hi; ue_hi >>= 1, i++) { + if (ue_hi & 1) + dev_err(dev, "UE: %s bit set\n", + ue_status_hi_desc[i]); + } } } - + if (error_detected) + netif_carrier_off(netdev); } static void be_msix_disable(struct be_adapter *adapter) @@ -2261,74 +2613,58 @@ static void be_msix_disable(struct be_adapter *adapter) if (msix_enabled(adapter)) { pci_disable_msix(adapter->pdev); adapter->num_msix_vec = 0; + adapter->num_msix_roce_vec = 0; } } -static uint be_num_rss_want(struct be_adapter *adapter) -{ - u32 num = 0; - - if ((adapter->function_caps & BE_FUNCTION_CAPS_RSS) && - (lancer_chip(adapter) || - (!sriov_want(adapter) && be_physfn(adapter)))) { - num = adapter->max_rss_queues; - num = min_t(u32, num, (u32)netif_get_num_default_rss_queues()); - } - return num; -} - -static void be_msix_enable(struct be_adapter *adapter) +static int be_msix_enable(struct be_adapter *adapter) { -#define BE_MIN_MSIX_VECTORS 1 - int i, status, num_vec, num_roce_vec = 0; + int i, num_vec; struct device *dev = &adapter->pdev->dev; - /* If RSS queues are not used, need a vec for default RX Q */ - num_vec = min(be_num_rss_want(adapter), num_online_cpus()); - if (be_roce_supported(adapter)) { - num_roce_vec = min_t(u32, MAX_ROCE_MSIX_VECTORS, - (num_online_cpus() + 1)); - num_roce_vec = min(num_roce_vec, MAX_ROCE_EQS); - num_vec += num_roce_vec; - num_vec = min(num_vec, MAX_MSIX_VECTORS); - } - num_vec = max(num_vec, BE_MIN_MSIX_VECTORS); + /* If RoCE is supported, program the max number of NIC vectors that + * may be configured via set-channels, along with vectors needed for + * RoCe. Else, just program the number we'll use initially. + */ + if (be_roce_supported(adapter)) + num_vec = min_t(int, 2 * be_max_eqs(adapter), + 2 * num_online_cpus()); + else + num_vec = adapter->cfg_num_qs; for (i = 0; i < num_vec; i++) adapter->msix_entries[i].entry = i; - status = pci_enable_msix(adapter->pdev, adapter->msix_entries, num_vec); - if (status == 0) { - goto done; - } else if (status >= BE_MIN_MSIX_VECTORS) { - num_vec = status; - if (pci_enable_msix(adapter->pdev, adapter->msix_entries, - num_vec) == 0) - goto done; + num_vec = pci_enable_msix_range(adapter->pdev, adapter->msix_entries, + MIN_MSIX_VECTORS, num_vec); + if (num_vec < 0) + goto fail; + + if (be_roce_supported(adapter) && num_vec > MIN_MSIX_VECTORS) { + adapter->num_msix_roce_vec = num_vec / 2; + dev_info(dev, "enabled %d MSI-x vector(s) for RoCE\n", + adapter->num_msix_roce_vec); } + adapter->num_msix_vec = num_vec - adapter->num_msix_roce_vec; + + dev_info(dev, "enabled %d MSI-x vector(s) for NIC\n", + adapter->num_msix_vec); + return 0; + +fail: dev_warn(dev, "MSIx enable failed\n"); - return; -done: - if (be_roce_supported(adapter)) { - if (num_vec > num_roce_vec) { - adapter->num_msix_vec = num_vec - num_roce_vec; - adapter->num_msix_roce_vec = - num_vec - adapter->num_msix_vec; - } else { - adapter->num_msix_vec = num_vec; - adapter->num_msix_roce_vec = 0; - } - } else - adapter->num_msix_vec = num_vec; - dev_info(dev, "enabled %d MSI-x vector(s)\n", adapter->num_msix_vec); - return; + + /* INTx is not supported in VFs, so fail probe if enable_msix fails */ + if (!be_physfn(adapter)) + return num_vec; + return 0; } static inline int be_msix_vec_get(struct be_adapter *adapter, - struct be_eq_obj *eqo) + struct be_eq_obj *eqo) { - return adapter->msix_entries[eqo->idx].vector; + return adapter->msix_entries[eqo->msix_idx].vector; } static int be_msix_register(struct be_adapter *adapter) @@ -2350,7 +2686,7 @@ err_msix: for (i--, eqo = &adapter->eq_obj[i]; i >= 0; i--, eqo--) free_irq(be_msix_vec_get(adapter, eqo), eqo); dev_warn(&adapter->pdev->dev, "MSIX Request IRQ failed - err %d\n", - status); + status); be_msix_disable(adapter); return status; } @@ -2416,11 +2752,6 @@ static void be_rx_qs_destroy(struct be_adapter *adapter) q = &rxo->q; if (q->created) { be_cmd_rxq_destroy(adapter, q); - /* After the rxq is invalidated, wait for a grace time - * of 1ms for all dma to end and the flush compl to - * arrive - */ - mdelay(1); be_rx_cq_clean(rxo); } be_queue_free(adapter, q); @@ -2433,23 +2764,37 @@ static int be_close(struct net_device *netdev) struct be_eq_obj *eqo; int i; - be_roce_dev_close(adapter); + /* This protection is needed as be_close() may be called even when the + * adapter is in cleared state (after eeh perm failure) + */ + if (!(adapter->flags & BE_FLAGS_SETUP_DONE)) + return 0; - if (!lancer_chip(adapter)) - be_intr_set(adapter, false); + be_roce_dev_close(adapter); - for_all_evt_queues(adapter, eqo, i) - napi_disable(&eqo->napi); + if (adapter->flags & BE_FLAGS_NAPI_ENABLED) { + for_all_evt_queues(adapter, eqo, i) { + napi_disable(&eqo->napi); + be_disable_busy_poll(eqo); + } + adapter->flags &= ~BE_FLAGS_NAPI_ENABLED; + } be_async_mcc_disable(adapter); /* Wait for all pending tx completions to arrive so that * all tx skbs are freed. */ + netif_tx_disable(netdev); be_tx_compl_clean(adapter); be_rx_qs_destroy(adapter); + for (i = 1; i < (adapter->uc_macs + 1); i++) + be_cmd_pmac_del(adapter, adapter->if_handle, + adapter->pmac_id[i], 0); + adapter->uc_macs = 0; + for_all_evt_queues(adapter, eqo, i) { if (msix_enabled(adapter)) synchronize_irq(be_msix_vec_get(adapter, eqo)); @@ -2467,7 +2812,8 @@ static int be_rx_qs_create(struct be_adapter *adapter) { struct be_rx_obj *rxo; int rc, i, j; - u8 rsstable[128]; + u8 rss_hkey[RSS_HASH_KEY_LEN]; + struct rss_info *rss = &adapter->rss_info; for_all_rx_queues(adapter, rxo, i) { rc = be_queue_alloc(adapter, &rxo->q, RX_Q_LEN, @@ -2492,18 +2838,36 @@ static int be_rx_qs_create(struct be_adapter *adapter) } if (be_multi_rxq(adapter)) { - for (j = 0; j < 128; j += adapter->num_rx_qs - 1) { + for (j = 0; j < RSS_INDIR_TABLE_LEN; + j += adapter->num_rx_qs - 1) { for_all_rss_queues(adapter, rxo, i) { - if ((j + i) >= 128) + if ((j + i) >= RSS_INDIR_TABLE_LEN) break; - rsstable[j + i] = rxo->rss_id; + rss->rsstable[j + i] = rxo->rss_id; + rss->rss_queue[j + i] = i; } } - rc = be_cmd_rss_config(adapter, rsstable, 128); - if (rc) - return rc; + rss->rss_flags = RSS_ENABLE_TCP_IPV4 | RSS_ENABLE_IPV4 | + RSS_ENABLE_TCP_IPV6 | RSS_ENABLE_IPV6; + + if (!BEx_chip(adapter)) + rss->rss_flags |= RSS_ENABLE_UDP_IPV4 | + RSS_ENABLE_UDP_IPV6; + } else { + /* Disable RSS, if only default RX Q is created */ + rss->rss_flags = RSS_ENABLE_NONE; + } + + get_random_bytes(rss_hkey, RSS_HASH_KEY_LEN); + rc = be_cmd_rss_config(adapter, rss->rsstable, rss->rss_flags, + 128, rss_hkey); + if (rc) { + rss->rss_flags = RSS_ENABLE_NONE; + return rc; } + memcpy(rss->rss_hkey, rss_hkey, RSS_HASH_KEY_LEN); + /* First time posting */ for_all_rx_queues(adapter, rxo, i) be_post_rx_frags(rxo, GFP_KERNEL); @@ -2523,10 +2887,9 @@ static int be_open(struct net_device *netdev) if (status) goto err; - be_irq_register(adapter); - - if (!lancer_chip(adapter)) - be_intr_set(adapter, true); + status = be_irq_register(adapter); + if (status) + goto err; for_all_rx_queues(adapter, rxo, i) be_cq_notify(adapter, rxo->cq.id, true, 0); @@ -2538,14 +2901,23 @@ static int be_open(struct net_device *netdev) for_all_evt_queues(adapter, eqo, i) { napi_enable(&eqo->napi); - be_eq_notify(adapter, eqo->q.id, true, false, 0); + be_enable_busy_poll(eqo); + be_eq_notify(adapter, eqo->q.id, true, true, 0); } + adapter->flags |= BE_FLAGS_NAPI_ENABLED; status = be_cmd_link_status_query(adapter, NULL, &link_status, 0); if (!status) be_link_status_update(adapter, link_status); + netif_tx_start_all_queues(netdev); be_roce_dev_open(adapter); + +#ifdef CONFIG_BE2NET_VXLAN + if (skyhawk_chip(adapter)) + vxlan_get_rx_port(netdev); +#endif + return 0; err: be_close(adapter->netdev); @@ -2561,15 +2933,15 @@ static int be_setup_wol(struct be_adapter *adapter, bool enable) memset(mac, 0, ETH_ALEN); cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config); - cmd.va = dma_alloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma, - GFP_KERNEL); + cmd.va = dma_zalloc_coherent(&adapter->pdev->dev, cmd.size, &cmd.dma, + GFP_KERNEL); if (cmd.va == NULL) return -1; - memset(cmd.va, 0, cmd.size); if (enable) { status = pci_write_config_dword(adapter->pdev, - PCICFG_PM_CONTROL_OFFSET, PCICFG_PM_CONTROL_MASK); + PCICFG_PM_CONTROL_OFFSET, + PCICFG_PM_CONTROL_MASK); if (status) { dev_err(&adapter->pdev->dev, "Could not enable Wake-on-lan\n"); @@ -2578,7 +2950,8 @@ static int be_setup_wol(struct be_adapter *adapter, bool enable) return status; } status = be_cmd_enable_magic_wol(adapter, - adapter->netdev->dev_addr, &cmd); + adapter->netdev->dev_addr, + &cmd); pci_enable_wake(adapter->pdev, PCI_D3hot, 1); pci_enable_wake(adapter->pdev, PCI_D3cold, 1); } else { @@ -2597,7 +2970,7 @@ static int be_setup_wol(struct be_adapter *adapter, bool enable) * These addresses are programmed in the ASIC by the PF and the VF driver * queries for the MAC address during its probe. */ -static inline int be_vf_eth_addr_config(struct be_adapter *adapter) +static int be_vf_eth_addr_config(struct be_adapter *adapter) { u32 vf; int status = 0; @@ -2607,17 +2980,18 @@ static inline int be_vf_eth_addr_config(struct be_adapter *adapter) be_vf_eth_addr_generate(adapter, mac); for_all_vfs(adapter, vf_cfg, vf) { - if (lancer_chip(adapter)) { - status = be_cmd_set_mac_list(adapter, mac, 1, vf + 1); - } else { + if (BEx_chip(adapter)) status = be_cmd_pmac_add(adapter, mac, vf_cfg->if_handle, &vf_cfg->pmac_id, vf + 1); - } + else + status = be_cmd_set_mac(adapter, mac, vf_cfg->if_handle, + vf + 1); if (status) dev_err(&adapter->pdev->dev, - "Mac address assignment failed for VF %d\n", vf); + "Mac address assignment failed for VF %d\n", + vf); else memcpy(vf_cfg->mac_addr, mac, ETH_ALEN); @@ -2626,76 +3000,149 @@ static inline int be_vf_eth_addr_config(struct be_adapter *adapter) return status; } +static int be_vfs_mac_query(struct be_adapter *adapter) +{ + int status, vf; + u8 mac[ETH_ALEN]; + struct be_vf_cfg *vf_cfg; + + for_all_vfs(adapter, vf_cfg, vf) { + status = be_cmd_get_active_mac(adapter, vf_cfg->pmac_id, + mac, vf_cfg->if_handle, + false, vf+1); + if (status) + return status; + memcpy(vf_cfg->mac_addr, mac, ETH_ALEN); + } + return 0; +} + static void be_vf_clear(struct be_adapter *adapter) { struct be_vf_cfg *vf_cfg; u32 vf; - if (be_find_vfs(adapter, ASSIGNED)) { - dev_warn(&adapter->pdev->dev, "VFs are assigned to VMs\n"); + if (pci_vfs_assigned(adapter->pdev)) { + dev_warn(&adapter->pdev->dev, + "VFs are assigned to VMs: not disabling VFs\n"); goto done; } + pci_disable_sriov(adapter->pdev); + for_all_vfs(adapter, vf_cfg, vf) { - if (lancer_chip(adapter)) - be_cmd_set_mac_list(adapter, NULL, 0, vf + 1); - else + if (BEx_chip(adapter)) be_cmd_pmac_del(adapter, vf_cfg->if_handle, vf_cfg->pmac_id, vf + 1); + else + be_cmd_set_mac(adapter, NULL, vf_cfg->if_handle, + vf + 1); be_cmd_if_destroy(adapter, vf_cfg->if_handle, vf + 1); } - pci_disable_sriov(adapter->pdev); done: kfree(adapter->vf_cfg); adapter->num_vfs = 0; } -static int be_clear(struct be_adapter *adapter) +static void be_clear_queues(struct be_adapter *adapter) { - int i = 1; + be_mcc_queues_destroy(adapter); + be_rx_cqs_destroy(adapter); + be_tx_queues_destroy(adapter); + be_evt_queues_destroy(adapter); +} +static void be_cancel_worker(struct be_adapter *adapter) +{ if (adapter->flags & BE_FLAGS_WORKER_SCHEDULED) { cancel_delayed_work_sync(&adapter->work); adapter->flags &= ~BE_FLAGS_WORKER_SCHEDULED; } +} + +static void be_mac_clear(struct be_adapter *adapter) +{ + int i; + + if (adapter->pmac_id) { + for (i = 0; i < (adapter->uc_macs + 1); i++) + be_cmd_pmac_del(adapter, adapter->if_handle, + adapter->pmac_id[i], 0); + adapter->uc_macs = 0; + + kfree(adapter->pmac_id); + adapter->pmac_id = NULL; + } +} + +#ifdef CONFIG_BE2NET_VXLAN +static void be_disable_vxlan_offloads(struct be_adapter *adapter) +{ + if (adapter->flags & BE_FLAGS_VXLAN_OFFLOADS) + be_cmd_manage_iface(adapter, adapter->if_handle, + OP_CONVERT_TUNNEL_TO_NORMAL); + + if (adapter->vxlan_port) + be_cmd_set_vxlan_port(adapter, 0); + + adapter->flags &= ~BE_FLAGS_VXLAN_OFFLOADS; + adapter->vxlan_port = 0; +} +#endif + +static int be_clear(struct be_adapter *adapter) +{ + be_cancel_worker(adapter); if (sriov_enabled(adapter)) be_vf_clear(adapter); - for (; adapter->uc_macs > 0; adapter->uc_macs--, i++) - be_cmd_pmac_del(adapter, adapter->if_handle, - adapter->pmac_id[i], 0); +#ifdef CONFIG_BE2NET_VXLAN + be_disable_vxlan_offloads(adapter); +#endif + /* delete the primary mac along with the uc-mac list */ + be_mac_clear(adapter); be_cmd_if_destroy(adapter, adapter->if_handle, 0); - be_mcc_queues_destroy(adapter); - be_rx_cqs_destroy(adapter); - be_tx_queues_destroy(adapter); - be_evt_queues_destroy(adapter); - - kfree(adapter->pmac_id); - adapter->pmac_id = NULL; + be_clear_queues(adapter); be_msix_disable(adapter); + adapter->flags &= ~BE_FLAGS_SETUP_DONE; return 0; } -static void be_get_vf_if_cap_flags(struct be_adapter *adapter, - u32 *cap_flags, u8 domain) +static int be_vfs_if_create(struct be_adapter *adapter) { - bool profile_present = false; - int status; + struct be_resources res = {0}; + struct be_vf_cfg *vf_cfg; + u32 cap_flags, en_flags, vf; + int status = 0; - if (lancer_chip(adapter)) { - status = be_cmd_get_profile_config(adapter, cap_flags, domain); - if (!status) - profile_present = true; - } + cap_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST | + BE_IF_FLAGS_MULTICAST; - if (!profile_present) - *cap_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST | - BE_IF_FLAGS_MULTICAST; + for_all_vfs(adapter, vf_cfg, vf) { + if (!BE3_chip(adapter)) { + status = be_cmd_get_profile_config(adapter, &res, + vf + 1); + if (!status) + cap_flags = res.if_cap_flags; + } + + /* If a FW profile exists, then cap_flags are updated */ + en_flags = cap_flags & (BE_IF_FLAGS_UNTAGGED | + BE_IF_FLAGS_BROADCAST | + BE_IF_FLAGS_MULTICAST); + status = + be_cmd_if_create(adapter, cap_flags, en_flags, + &vf_cfg->if_handle, vf + 1); + if (status) + goto err; + } +err: + return status; } static int be_vf_setup_init(struct be_adapter *adapter) @@ -2717,77 +3164,178 @@ static int be_vf_setup_init(struct be_adapter *adapter) static int be_vf_setup(struct be_adapter *adapter) { - struct be_vf_cfg *vf_cfg; struct device *dev = &adapter->pdev->dev; - u32 cap_flags, en_flags, vf; - u16 def_vlan, lnk_speed; - int status, enabled_vfs; - - enabled_vfs = be_find_vfs(adapter, ENABLED); - if (enabled_vfs) { - dev_warn(dev, "%d VFs are already enabled\n", enabled_vfs); - dev_warn(dev, "Ignoring num_vfs=%d setting\n", num_vfs); - return 0; - } - - if (num_vfs > adapter->dev_num_vfs) { - dev_warn(dev, "Device supports %d VFs and not %d\n", - adapter->dev_num_vfs, num_vfs); - num_vfs = adapter->dev_num_vfs; - } - - status = pci_enable_sriov(adapter->pdev, num_vfs); - if (!status) { - adapter->num_vfs = num_vfs; + struct be_vf_cfg *vf_cfg; + int status, old_vfs, vf; + u32 privileges; + + old_vfs = pci_num_vf(adapter->pdev); + if (old_vfs) { + dev_info(dev, "%d VFs are already enabled\n", old_vfs); + if (old_vfs != num_vfs) + dev_warn(dev, "Ignoring num_vfs=%d setting\n", num_vfs); + adapter->num_vfs = old_vfs; } else { - /* Platform doesn't support SRIOV though device supports it */ - dev_warn(dev, "SRIOV enable failed\n"); - return 0; + if (num_vfs > be_max_vfs(adapter)) + dev_info(dev, "Device supports %d VFs and not %d\n", + be_max_vfs(adapter), num_vfs); + adapter->num_vfs = min_t(u16, num_vfs, be_max_vfs(adapter)); + if (!adapter->num_vfs) + return 0; } status = be_vf_setup_init(adapter); if (status) goto err; - for_all_vfs(adapter, vf_cfg, vf) { - be_get_vf_if_cap_flags(adapter, &cap_flags, vf + 1); - - en_flags = cap_flags & (BE_IF_FLAGS_UNTAGGED | - BE_IF_FLAGS_BROADCAST | - BE_IF_FLAGS_MULTICAST); - - status = be_cmd_if_create(adapter, cap_flags, en_flags, - &vf_cfg->if_handle, vf + 1); + if (old_vfs) { + for_all_vfs(adapter, vf_cfg, vf) { + status = be_cmd_get_if_id(adapter, vf_cfg, vf); + if (status) + goto err; + } + } else { + status = be_vfs_if_create(adapter); if (status) goto err; } - if (!enabled_vfs) { + if (old_vfs) { + status = be_vfs_mac_query(adapter); + if (status) + goto err; + } else { status = be_vf_eth_addr_config(adapter); if (status) goto err; } for_all_vfs(adapter, vf_cfg, vf) { - lnk_speed = 1000; - status = be_cmd_set_qos(adapter, lnk_speed, vf + 1); - if (status) - goto err; - vf_cfg->tx_rate = lnk_speed * 10; + /* Allow VFs to programs MAC/VLAN filters */ + status = be_cmd_get_fn_privileges(adapter, &privileges, vf + 1); + if (!status && !(privileges & BE_PRIV_FILTMGMT)) { + status = be_cmd_set_fn_privileges(adapter, + privileges | + BE_PRIV_FILTMGMT, + vf + 1); + if (!status) + dev_info(dev, "VF%d has FILTMGMT privilege\n", + vf); + } - status = be_cmd_get_hsw_config(adapter, &def_vlan, - vf + 1, vf_cfg->if_handle); - if (status) - goto err; - vf_cfg->def_vid = def_vlan; + /* Allow full available bandwidth */ + if (!old_vfs) + be_cmd_config_qos(adapter, 0, 0, vf + 1); + + if (!old_vfs) { + be_cmd_enable_vf(adapter, vf + 1); + be_cmd_set_logical_link_config(adapter, + IFLA_VF_LINK_STATE_AUTO, + vf+1); + } + } - be_cmd_enable_vf(adapter, vf + 1); + if (!old_vfs) { + status = pci_enable_sriov(adapter->pdev, adapter->num_vfs); + if (status) { + dev_err(dev, "SRIOV enable failed\n"); + adapter->num_vfs = 0; + goto err; + } } return 0; err: + dev_err(dev, "VF setup failed\n"); + be_vf_clear(adapter); return status; } +/* Converting function_mode bits on BE3 to SH mc_type enums */ + +static u8 be_convert_mc_type(u32 function_mode) +{ + if (function_mode & VNIC_MODE && function_mode & QNQ_MODE) + return vNIC1; + else if (function_mode & QNQ_MODE) + return FLEX10; + else if (function_mode & VNIC_MODE) + return vNIC2; + else if (function_mode & UMC_ENABLED) + return UMC; + else + return MC_NONE; +} + +/* On BE2/BE3 FW does not suggest the supported limits */ +static void BEx_get_resources(struct be_adapter *adapter, + struct be_resources *res) +{ + struct pci_dev *pdev = adapter->pdev; + bool use_sriov = false; + int max_vfs = 0; + + if (be_physfn(adapter) && BE3_chip(adapter)) { + be_cmd_get_profile_config(adapter, res, 0); + /* Some old versions of BE3 FW don't report max_vfs value */ + if (res->max_vfs == 0) { + max_vfs = pci_sriov_get_totalvfs(pdev); + res->max_vfs = max_vfs > 0 ? min(MAX_VFS, max_vfs) : 0; + } + use_sriov = res->max_vfs && sriov_want(adapter); + } + + if (be_physfn(adapter)) + res->max_uc_mac = BE_UC_PMAC_COUNT; + else + res->max_uc_mac = BE_VF_UC_PMAC_COUNT; + + adapter->mc_type = be_convert_mc_type(adapter->function_mode); + + if (be_is_mc(adapter)) { + /* Assuming that there are 4 channels per port, + * when multi-channel is enabled + */ + if (be_is_qnq_mode(adapter)) + res->max_vlans = BE_NUM_VLANS_SUPPORTED/8; + else + /* In a non-qnq multichannel mode, the pvid + * takes up one vlan entry + */ + res->max_vlans = (BE_NUM_VLANS_SUPPORTED / 4) - 1; + } else { + res->max_vlans = BE_NUM_VLANS_SUPPORTED; + } + + res->max_mcast_mac = BE_MAX_MC; + + /* 1) For BE3 1Gb ports, FW does not support multiple TXQs + * 2) Create multiple TX rings on a BE3-R multi-channel interface + * *only* if it is RSS-capable. + */ + if (BE2_chip(adapter) || use_sriov || (adapter->port_num > 1) || + !be_physfn(adapter) || (be_is_mc(adapter) && + !(adapter->function_caps & BE_FUNCTION_CAPS_RSS))) + res->max_tx_qs = 1; + else + res->max_tx_qs = BE3_MAX_TX_QS; + + if ((adapter->function_caps & BE_FUNCTION_CAPS_RSS) && + !use_sriov && be_physfn(adapter)) + res->max_rss_qs = (adapter->be3_native) ? + BE3_MAX_RSS_QS : BE2_MAX_RSS_QS; + res->max_rx_qs = res->max_rss_qs + 1; + + if (be_physfn(adapter)) + res->max_evt_qs = (res->max_vfs > 0) ? + BE3_SRIOV_MAX_EVT_QS : BE3_MAX_EVT_QS; + else + res->max_evt_qs = 1; + + res->if_cap_flags = BE_IF_CAP_FLAGS_WANT; + if (!(adapter->function_caps & BE_FUNCTION_CAPS_RSS)) + res->if_cap_flags &= ~BE_IF_FLAGS_RSS; +} + static void be_setup_init(struct be_adapter *adapter) { adapter->vlan_prio_bmap = 0xff; @@ -2801,148 +3349,191 @@ static void be_setup_init(struct be_adapter *adapter) adapter->cmd_privileges = MIN_PRIVILEGES; } -static int be_get_mac_addr(struct be_adapter *adapter, u8 *mac, u32 if_handle, - bool *active_mac, u32 *pmac_id) +static int be_get_resources(struct be_adapter *adapter) { - int status = 0; - - if (!is_zero_ether_addr(adapter->netdev->perm_addr)) { - memcpy(mac, adapter->netdev->dev_addr, ETH_ALEN); - if (!lancer_chip(adapter) && !be_physfn(adapter)) - *active_mac = true; - else - *active_mac = false; + struct device *dev = &adapter->pdev->dev; + struct be_resources res = {0}; + int status; - return status; + if (BEx_chip(adapter)) { + BEx_get_resources(adapter, &res); + adapter->res = res; } - if (lancer_chip(adapter)) { - status = be_cmd_get_mac_from_list(adapter, mac, - active_mac, pmac_id, 0); - if (*active_mac) { - status = be_cmd_mac_addr_query(adapter, mac, false, - if_handle, *pmac_id); + /* For Lancer, SH etc read per-function resource limits from FW. + * GET_FUNC_CONFIG returns per function guaranteed limits. + * GET_PROFILE_CONFIG returns PCI-E related limits PF-pool limits + */ + if (!BEx_chip(adapter)) { + status = be_cmd_get_func_config(adapter, &res); + if (status) + return status; + + /* If RoCE may be enabled stash away half the EQs for RoCE */ + if (be_roce_supported(adapter)) + res.max_evt_qs /= 2; + adapter->res = res; + + if (be_physfn(adapter)) { + status = be_cmd_get_profile_config(adapter, &res, 0); + if (status) + return status; + adapter->res.max_vfs = res.max_vfs; } - } else if (be_physfn(adapter)) { - /* For BE3, for PF get permanent MAC */ - status = be_cmd_mac_addr_query(adapter, mac, true, 0, 0); - *active_mac = false; - } else { - /* For BE3, for VF get soft MAC assigned by PF*/ - status = be_cmd_mac_addr_query(adapter, mac, false, - if_handle, 0); - *active_mac = true; + + dev_info(dev, "Max: txqs %d, rxqs %d, rss %d, eqs %d, vfs %d\n", + be_max_txqs(adapter), be_max_rxqs(adapter), + be_max_rss(adapter), be_max_eqs(adapter), + be_max_vfs(adapter)); + dev_info(dev, "Max: uc-macs %d, mc-macs %d, vlans %d\n", + be_max_uc(adapter), be_max_mc(adapter), + be_max_vlans(adapter)); } - return status; + + return 0; } -static void be_get_resources(struct be_adapter *adapter) +/* Routine to query per function resource limits */ +static int be_get_config(struct be_adapter *adapter) { + u16 profile_id; int status; - bool profile_present = false; - if (lancer_chip(adapter)) { - status = be_cmd_get_func_config(adapter); + status = be_cmd_query_fw_cfg(adapter, &adapter->port_num, + &adapter->function_mode, + &adapter->function_caps, + &adapter->asic_rev); + if (status) + return status; + if (be_physfn(adapter)) { + status = be_cmd_get_active_profile(adapter, &profile_id); if (!status) - profile_present = true; - } - - if (profile_present) { - /* Sanity fixes for Lancer */ - adapter->max_pmac_cnt = min_t(u16, adapter->max_pmac_cnt, - BE_UC_PMAC_COUNT); - adapter->max_vlans = min_t(u16, adapter->max_vlans, - BE_NUM_VLANS_SUPPORTED); - adapter->max_mcast_mac = min_t(u16, adapter->max_mcast_mac, - BE_MAX_MC); - adapter->max_tx_queues = min_t(u16, adapter->max_tx_queues, - MAX_TX_QS); - adapter->max_rss_queues = min_t(u16, adapter->max_rss_queues, - BE3_MAX_RSS_QS); - adapter->max_event_queues = min_t(u16, - adapter->max_event_queues, - BE3_MAX_RSS_QS); - - if (adapter->max_rss_queues && - adapter->max_rss_queues == adapter->max_rx_queues) - adapter->max_rss_queues -= 1; - - if (adapter->max_event_queues < adapter->max_rss_queues) - adapter->max_rss_queues = adapter->max_event_queues; + dev_info(&adapter->pdev->dev, + "Using profile 0x%x\n", profile_id); + } - } else { - if (be_physfn(adapter)) - adapter->max_pmac_cnt = BE_UC_PMAC_COUNT; - else - adapter->max_pmac_cnt = BE_VF_UC_PMAC_COUNT; + status = be_get_resources(adapter); + if (status) + return status; - if (adapter->function_mode & FLEX10_MODE) - adapter->max_vlans = BE_NUM_VLANS_SUPPORTED/8; - else - adapter->max_vlans = BE_NUM_VLANS_SUPPORTED; + adapter->pmac_id = kcalloc(be_max_uc(adapter), + sizeof(*adapter->pmac_id), GFP_KERNEL); + if (!adapter->pmac_id) + return -ENOMEM; - adapter->max_mcast_mac = BE_MAX_MC; - adapter->max_tx_queues = MAX_TX_QS; - adapter->max_rss_queues = (adapter->be3_native) ? - BE3_MAX_RSS_QS : BE2_MAX_RSS_QS; - adapter->max_event_queues = BE3_MAX_RSS_QS; + /* Sanitize cfg_num_qs based on HW and platform limits */ + adapter->cfg_num_qs = min(adapter->cfg_num_qs, be_max_qs(adapter)); - adapter->if_cap_flags = BE_IF_FLAGS_UNTAGGED | - BE_IF_FLAGS_BROADCAST | - BE_IF_FLAGS_MULTICAST | - BE_IF_FLAGS_PASS_L3L4_ERRORS | - BE_IF_FLAGS_MCAST_PROMISCUOUS | - BE_IF_FLAGS_VLAN_PROMISCUOUS | - BE_IF_FLAGS_PROMISCUOUS; + return 0; +} + +static int be_mac_setup(struct be_adapter *adapter) +{ + u8 mac[ETH_ALEN]; + int status; - if (adapter->function_caps & BE_FUNCTION_CAPS_RSS) - adapter->if_cap_flags |= BE_IF_FLAGS_RSS; + if (is_zero_ether_addr(adapter->netdev->dev_addr)) { + status = be_cmd_get_perm_mac(adapter, mac); + if (status) + return status; + + memcpy(adapter->netdev->dev_addr, mac, ETH_ALEN); + memcpy(adapter->netdev->perm_addr, mac, ETH_ALEN); + } else { + /* Maybe the HW was reset; dev_addr must be re-programmed */ + memcpy(mac, adapter->netdev->dev_addr, ETH_ALEN); } + + /* For BE3-R VFs, the PF programs the initial MAC address */ + if (!(BEx_chip(adapter) && be_virtfn(adapter))) + be_cmd_pmac_add(adapter, mac, adapter->if_handle, + &adapter->pmac_id[0], 0); + return 0; } -/* Routine to query per function resource limits */ -static int be_get_config(struct be_adapter *adapter) +static void be_schedule_worker(struct be_adapter *adapter) { - int pos, status; - u16 dev_num_vfs; + schedule_delayed_work(&adapter->work, msecs_to_jiffies(1000)); + adapter->flags |= BE_FLAGS_WORKER_SCHEDULED; +} - status = be_cmd_query_fw_cfg(adapter, &adapter->port_num, - &adapter->function_mode, - &adapter->function_caps); +static int be_setup_queues(struct be_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int status; + + status = be_evt_queues_create(adapter); if (status) goto err; - be_get_resources(adapter); + status = be_tx_qs_create(adapter); + if (status) + goto err; - /* primary mac needs 1 pmac entry */ - adapter->pmac_id = kcalloc(adapter->max_pmac_cnt + 1, - sizeof(u32), GFP_KERNEL); - if (!adapter->pmac_id) { - status = -ENOMEM; + status = be_rx_cqs_create(adapter); + if (status) goto err; - } - pos = pci_find_ext_capability(adapter->pdev, PCI_EXT_CAP_ID_SRIOV); - if (pos) { - pci_read_config_word(adapter->pdev, pos + PCI_SRIOV_TOTAL_VF, - &dev_num_vfs); - if (!lancer_chip(adapter)) - dev_num_vfs = min_t(u16, dev_num_vfs, MAX_VFS); - adapter->dev_num_vfs = dev_num_vfs; - } + status = be_mcc_queues_create(adapter); + if (status) + goto err; + + status = netif_set_real_num_rx_queues(netdev, adapter->num_rx_qs); + if (status) + goto err; + + status = netif_set_real_num_tx_queues(netdev, adapter->num_tx_qs); + if (status) + goto err; + + return 0; err: + dev_err(&adapter->pdev->dev, "queue_setup failed\n"); + return status; +} + +int be_update_queues(struct be_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int status; + + if (netif_running(netdev)) + be_close(netdev); + + be_cancel_worker(adapter); + + /* If any vectors have been shared with RoCE we cannot re-program + * the MSIx table. + */ + if (!adapter->num_msix_roce_vec) + be_msix_disable(adapter); + + be_clear_queues(adapter); + + if (!msix_enabled(adapter)) { + status = be_msix_enable(adapter); + if (status) + return status; + } + + status = be_setup_queues(adapter); + if (status) + return status; + + be_schedule_worker(adapter); + + if (netif_running(netdev)) + status = be_open(netdev); + return status; } static int be_setup(struct be_adapter *adapter) { struct device *dev = &adapter->pdev->dev; - u32 en_flags; - u32 tx_fc, rx_fc; + u32 tx_fc, rx_fc, en_flags; int status; - u8 mac[ETH_ALEN]; - bool active_mac; be_setup_init(adapter); @@ -2953,82 +3544,60 @@ static int be_setup(struct be_adapter *adapter) if (status) goto err; - be_msix_enable(adapter); - - status = be_evt_queues_create(adapter); + status = be_msix_enable(adapter); if (status) goto err; - status = be_tx_cqs_create(adapter); - if (status) - goto err; - - status = be_rx_cqs_create(adapter); - if (status) - goto err; - - status = be_mcc_queues_create(adapter); - if (status) - goto err; - - be_cmd_get_fn_privileges(adapter, &adapter->cmd_privileges, 0); - /* In UMC mode FW does not return right privileges. - * Override with correct privilege equivalent to PF. - */ - if (be_is_mc(adapter)) - adapter->cmd_privileges = MAX_PRIVILEGES; - en_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST | - BE_IF_FLAGS_MULTICAST | BE_IF_FLAGS_PASS_L3L4_ERRORS; - + BE_IF_FLAGS_MULTICAST | BE_IF_FLAGS_PASS_L3L4_ERRORS; if (adapter->function_caps & BE_FUNCTION_CAPS_RSS) en_flags |= BE_IF_FLAGS_RSS; - - en_flags = en_flags & adapter->if_cap_flags; - - status = be_cmd_if_create(adapter, adapter->if_cap_flags, en_flags, + en_flags = en_flags & be_if_cap_flags(adapter); + status = be_cmd_if_create(adapter, be_if_cap_flags(adapter), en_flags, &adapter->if_handle, 0); - if (status != 0) + if (status) goto err; - memset(mac, 0, ETH_ALEN); - active_mac = false; - status = be_get_mac_addr(adapter, mac, adapter->if_handle, - &active_mac, &adapter->pmac_id[0]); - if (status != 0) + /* Updating real_num_tx/rx_queues() requires rtnl_lock() */ + rtnl_lock(); + status = be_setup_queues(adapter); + rtnl_unlock(); + if (status) goto err; - if (!active_mac) { - status = be_cmd_pmac_add(adapter, mac, adapter->if_handle, - &adapter->pmac_id[0], 0); - if (status != 0) - goto err; - } - - if (is_zero_ether_addr(adapter->netdev->dev_addr)) { - memcpy(adapter->netdev->dev_addr, mac, ETH_ALEN); - memcpy(adapter->netdev->perm_addr, mac, ETH_ALEN); - } + be_cmd_get_fn_privileges(adapter, &adapter->cmd_privileges, 0); - status = be_tx_qs_create(adapter); + status = be_mac_setup(adapter); if (status) goto err; - be_cmd_get_fw_ver(adapter, adapter->fw_ver, NULL); + be_cmd_get_fw_ver(adapter, adapter->fw_ver, adapter->fw_on_flash); + + if (BE2_chip(adapter) && fw_major_num(adapter->fw_ver) < 4) { + dev_err(dev, "Firmware on card is old(%s), IRQs may not work.", + adapter->fw_ver); + dev_err(dev, "Please upgrade firmware to version >= 4.0\n"); + } if (adapter->vlans_added) be_vid_config(adapter); be_set_rx_mode(adapter->netdev); + be_cmd_get_acpi_wol_cap(adapter); + be_cmd_get_flow_control(adapter, &tx_fc, &rx_fc); if (rx_fc != adapter->rx_fc || tx_fc != adapter->tx_fc) be_cmd_set_flow_control(adapter, adapter->tx_fc, adapter->rx_fc); - if (be_physfn(adapter) && num_vfs) { - if (adapter->dev_num_vfs) + if (be_physfn(adapter)) + be_cmd_set_logical_link_config(adapter, + IFLA_VF_LINK_STATE_AUTO, 0); + + if (sriov_want(adapter)) { + if (be_max_vfs(adapter)) be_vf_setup(adapter); else dev_warn(dev, "device doesn't support SRIOV\n"); @@ -3038,8 +3607,8 @@ static int be_setup(struct be_adapter *adapter) if (!status && be_pause_supported(adapter)) adapter->phy.fc_autoneg = 1; - schedule_delayed_work(&adapter->work, msecs_to_jiffies(1000)); - adapter->flags |= BE_FLAGS_WORKER_SCHEDULED; + be_schedule_worker(adapter); + adapter->flags |= BE_FLAGS_SETUP_DONE; return 0; err: be_clear(adapter); @@ -3062,35 +3631,7 @@ static void be_netpoll(struct net_device *netdev) } #endif -#define FW_FILE_HDR_SIGN "ServerEngines Corp. " -char flash_cookie[2][16] = {"*** SE FLAS", "H DIRECTORY *** "}; - -static bool be_flash_redboot(struct be_adapter *adapter, - const u8 *p, u32 img_start, int image_size, - int hdr_size) -{ - u32 crc_offset; - u8 flashed_crc[4]; - int status; - - crc_offset = hdr_size + img_start + image_size - 4; - - p += crc_offset; - - status = be_cmd_get_flash_crc(adapter, flashed_crc, - (image_size - 4)); - if (status) { - dev_err(&adapter->pdev->dev, - "could not get crc from flash, not flashing redboot\n"); - return false; - } - - /*update redboot only if crc does not match*/ - if (!memcmp(flashed_crc, p, 4)) - return false; - else - return true; -} +static char flash_cookie[2][16] = {"*** SE FLAS", "H DIRECTORY *** "}; static bool phy_flashing_required(struct be_adapter *adapter) { @@ -3120,9 +3661,9 @@ static bool is_comp_in_ufi(struct be_adapter *adapter, } -struct flash_section_info *get_fsec_info(struct be_adapter *adapter, - int header_size, - const struct firmware *fw) +static struct flash_section_info *get_fsec_info(struct be_adapter *adapter, + int header_size, + const struct firmware *fw) { struct flash_section_info *fsec = NULL; const u8 *p = fw->data; @@ -3137,12 +3678,35 @@ struct flash_section_info *get_fsec_info(struct be_adapter *adapter, return NULL; } +static int be_check_flash_crc(struct be_adapter *adapter, const u8 *p, + u32 img_offset, u32 img_size, int hdr_size, + u16 img_optype, bool *crc_match) +{ + u32 crc_offset; + int status; + u8 crc[4]; + + status = be_cmd_get_flash_crc(adapter, crc, img_optype, img_size - 4); + if (status) + return status; + + crc_offset = hdr_size + img_offset + img_size - 4; + + /* Skip flashing, if crc of flashed region matches */ + if (!memcmp(crc, p + crc_offset, 4)) + *crc_match = true; + else + *crc_match = false; + + return status; +} + static int be_flash(struct be_adapter *adapter, const u8 *img, - struct be_dma_mem *flash_cmd, int optype, int img_size) + struct be_dma_mem *flash_cmd, int optype, int img_size) { - u32 total_bytes = 0, flash_op, num_bytes = 0; - int status = 0; struct be_cmd_write_flashrom *req = flash_cmd->va; + u32 total_bytes, flash_op, num_bytes; + int status; total_bytes = img_size; while (total_bytes) { @@ -3165,32 +3729,28 @@ static int be_flash(struct be_adapter *adapter, const u8 *img, memcpy(req->data_buf, img, num_bytes); img += num_bytes; status = be_cmd_write_flashrom(adapter, flash_cmd, optype, - flash_op, num_bytes); - if (status) { - if (status == ILLEGAL_IOCTL_REQ && - optype == OPTYPE_PHY_FW) - break; - dev_err(&adapter->pdev->dev, - "cmd to write to flash rom failed.\n"); + flash_op, num_bytes); + if (base_status(status) == MCC_STATUS_ILLEGAL_REQUEST && + optype == OPTYPE_PHY_FW) + break; + else if (status) return status; - } } return 0; } -/* For BE2 and BE3 */ +/* For BE2, BE3 and BE3-R */ static int be_flash_BEx(struct be_adapter *adapter, - const struct firmware *fw, - struct be_dma_mem *flash_cmd, - int num_of_images) - + const struct firmware *fw, + struct be_dma_mem *flash_cmd, int num_of_images) { - int status = 0, i, filehdr_size = 0; int img_hdrs_size = (num_of_images * sizeof(struct image_hdr)); - const u8 *p = fw->data; - const struct flash_comp *pflashcomp; - int num_comp, redboot; + struct device *dev = &adapter->pdev->dev; struct flash_section_info *fsec = NULL; + int status, i, filehdr_size, num_comp; + const struct flash_comp *pflashcomp; + bool crc_match; + const u8 *p; struct flash_comp gen3_flash_types[] = { { FLASH_iSCSI_PRIMARY_IMAGE_START_g3, OPTYPE_ISCSI_ACTIVE, @@ -3247,8 +3807,7 @@ static int be_flash_BEx(struct be_adapter *adapter, /* Get flash section info*/ fsec = get_fsec_info(adapter, filehdr_size + img_hdrs_size, fw); if (!fsec) { - dev_err(&adapter->pdev->dev, - "Invalid Cookie. UFI corrupted ?\n"); + dev_err(dev, "Invalid Cookie. FW image may be corrupted\n"); return -1; } for (i = 0; i < num_comp; i++) { @@ -3264,23 +3823,32 @@ static int be_flash_BEx(struct be_adapter *adapter, continue; if (pflashcomp[i].optype == OPTYPE_REDBOOT) { - redboot = be_flash_redboot(adapter, fw->data, - pflashcomp[i].offset, pflashcomp[i].size, - filehdr_size + img_hdrs_size); - if (!redboot) + status = be_check_flash_crc(adapter, fw->data, + pflashcomp[i].offset, + pflashcomp[i].size, + filehdr_size + + img_hdrs_size, + OPTYPE_REDBOOT, &crc_match); + if (status) { + dev_err(dev, + "Could not get CRC for 0x%x region\n", + pflashcomp[i].optype); + continue; + } + + if (crc_match) continue; } - p = fw->data; - p += filehdr_size + pflashcomp[i].offset + img_hdrs_size; + p = fw->data + filehdr_size + pflashcomp[i].offset + + img_hdrs_size; if (p + pflashcomp[i].size > fw->data + fw->size) return -1; status = be_flash(adapter, p, flash_cmd, pflashcomp[i].optype, - pflashcomp[i].size); + pflashcomp[i].size); if (status) { - dev_err(&adapter->pdev->dev, - "Flashing section type %d failed.\n", + dev_err(dev, "Flashing section type 0x%x failed\n", pflashcomp[i].img_type); return status; } @@ -3288,114 +3856,142 @@ static int be_flash_BEx(struct be_adapter *adapter, return 0; } +static u16 be_get_img_optype(struct flash_section_entry fsec_entry) +{ + u32 img_type = le32_to_cpu(fsec_entry.type); + u16 img_optype = le16_to_cpu(fsec_entry.optype); + + if (img_optype != 0xFFFF) + return img_optype; + + switch (img_type) { + case IMAGE_FIRMWARE_iSCSI: + img_optype = OPTYPE_ISCSI_ACTIVE; + break; + case IMAGE_BOOT_CODE: + img_optype = OPTYPE_REDBOOT; + break; + case IMAGE_OPTION_ROM_ISCSI: + img_optype = OPTYPE_BIOS; + break; + case IMAGE_OPTION_ROM_PXE: + img_optype = OPTYPE_PXE_BIOS; + break; + case IMAGE_OPTION_ROM_FCoE: + img_optype = OPTYPE_FCOE_BIOS; + break; + case IMAGE_FIRMWARE_BACKUP_iSCSI: + img_optype = OPTYPE_ISCSI_BACKUP; + break; + case IMAGE_NCSI: + img_optype = OPTYPE_NCSI_FW; + break; + case IMAGE_FLASHISM_JUMPVECTOR: + img_optype = OPTYPE_FLASHISM_JUMPVECTOR; + break; + case IMAGE_FIRMWARE_PHY: + img_optype = OPTYPE_SH_PHY_FW; + break; + case IMAGE_REDBOOT_DIR: + img_optype = OPTYPE_REDBOOT_DIR; + break; + case IMAGE_REDBOOT_CONFIG: + img_optype = OPTYPE_REDBOOT_CONFIG; + break; + case IMAGE_UFI_DIR: + img_optype = OPTYPE_UFI_DIR; + break; + default: + break; + } + + return img_optype; +} + static int be_flash_skyhawk(struct be_adapter *adapter, - const struct firmware *fw, - struct be_dma_mem *flash_cmd, int num_of_images) + const struct firmware *fw, + struct be_dma_mem *flash_cmd, int num_of_images) { - int status = 0, i, filehdr_size = 0; - int img_offset, img_size, img_optype, redboot; int img_hdrs_size = num_of_images * sizeof(struct image_hdr); - const u8 *p = fw->data; + struct device *dev = &adapter->pdev->dev; struct flash_section_info *fsec = NULL; + u32 img_offset, img_size, img_type; + int status, i, filehdr_size; + bool crc_match, old_fw_img; + u16 img_optype; + const u8 *p; filehdr_size = sizeof(struct flash_file_hdr_g3); fsec = get_fsec_info(adapter, filehdr_size + img_hdrs_size, fw); if (!fsec) { - dev_err(&adapter->pdev->dev, - "Invalid Cookie. UFI corrupted ?\n"); + dev_err(dev, "Invalid Cookie. FW image may be corrupted\n"); return -1; } for (i = 0; i < le32_to_cpu(fsec->fsec_hdr.num_images); i++) { img_offset = le32_to_cpu(fsec->fsec_entry[i].offset); img_size = le32_to_cpu(fsec->fsec_entry[i].pad_size); + img_type = le32_to_cpu(fsec->fsec_entry[i].type); + img_optype = be_get_img_optype(fsec->fsec_entry[i]); + old_fw_img = fsec->fsec_entry[i].optype == 0xFFFF; - switch (le32_to_cpu(fsec->fsec_entry[i].type)) { - case IMAGE_FIRMWARE_iSCSI: - img_optype = OPTYPE_ISCSI_ACTIVE; - break; - case IMAGE_BOOT_CODE: - img_optype = OPTYPE_REDBOOT; - break; - case IMAGE_OPTION_ROM_ISCSI: - img_optype = OPTYPE_BIOS; - break; - case IMAGE_OPTION_ROM_PXE: - img_optype = OPTYPE_PXE_BIOS; - break; - case IMAGE_OPTION_ROM_FCoE: - img_optype = OPTYPE_FCOE_BIOS; - break; - case IMAGE_FIRMWARE_BACKUP_iSCSI: - img_optype = OPTYPE_ISCSI_BACKUP; - break; - case IMAGE_NCSI: - img_optype = OPTYPE_NCSI_FW; - break; - default: + if (img_optype == 0xFFFF) continue; + /* Don't bother verifying CRC if an old FW image is being + * flashed + */ + if (old_fw_img) + goto flash; + + status = be_check_flash_crc(adapter, fw->data, img_offset, + img_size, filehdr_size + + img_hdrs_size, img_optype, + &crc_match); + /* The current FW image on the card does not recognize the new + * FLASH op_type. The FW download is partially complete. + * Reboot the server now to enable FW image to recognize the + * new FLASH op_type. To complete the remaining process, + * download the same FW again after the reboot. + */ + if (base_status(status) == MCC_STATUS_ILLEGAL_REQUEST || + base_status(status) == MCC_STATUS_ILLEGAL_FIELD) { + dev_err(dev, "Flash incomplete. Reset the server\n"); + dev_err(dev, "Download FW image again after reset\n"); + return -EAGAIN; + } else if (status) { + dev_err(dev, "Could not get CRC for 0x%x region\n", + img_optype); + return -EFAULT; } - if (img_optype == OPTYPE_REDBOOT) { - redboot = be_flash_redboot(adapter, fw->data, - img_offset, img_size, - filehdr_size + img_hdrs_size); - if (!redboot) - continue; - } + if (crc_match) + continue; - p = fw->data; - p += filehdr_size + img_offset + img_hdrs_size; +flash: + p = fw->data + filehdr_size + img_offset + img_hdrs_size; if (p + img_size > fw->data + fw->size) return -1; status = be_flash(adapter, p, flash_cmd, img_optype, img_size); - if (status) { - dev_err(&adapter->pdev->dev, - "Flashing section type %d failed.\n", - fsec->fsec_entry[i].type); - return status; + /* For old FW images ignore ILLEGAL_FIELD error or errors on + * UFI_DIR region + */ + if (old_fw_img && + (base_status(status) == MCC_STATUS_ILLEGAL_FIELD || + (img_optype == OPTYPE_UFI_DIR && + base_status(status) == MCC_STATUS_FAILED))) { + continue; + } else if (status) { + dev_err(dev, "Flashing section type 0x%x failed\n", + img_type); + return -EFAULT; } } return 0; } -static int lancer_wait_idle(struct be_adapter *adapter) -{ -#define SLIPORT_IDLE_TIMEOUT 30 - u32 reg_val; - int status = 0, i; - - for (i = 0; i < SLIPORT_IDLE_TIMEOUT; i++) { - reg_val = ioread32(adapter->db + PHYSDEV_CONTROL_OFFSET); - if ((reg_val & PHYSDEV_CONTROL_INP_MASK) == 0) - break; - - ssleep(1); - } - - if (i == SLIPORT_IDLE_TIMEOUT) - status = -1; - - return status; -} - -static int lancer_fw_reset(struct be_adapter *adapter) -{ - int status = 0; - - status = lancer_wait_idle(adapter); - if (status) - return status; - - iowrite32(PHYSDEV_CONTROL_FW_RESET_MASK, adapter->db + - PHYSDEV_CONTROL_OFFSET); - - return status; -} - static int lancer_fw_download(struct be_adapter *adapter, - const struct firmware *fw) + const struct firmware *fw) { #define LANCER_FW_DOWNLOAD_CHUNK (32 * 1024) #define LANCER_FW_DOWNLOAD_LOCATION "/prg" @@ -3421,11 +4017,9 @@ static int lancer_fw_download(struct be_adapter *adapter, flash_cmd.size = sizeof(struct lancer_cmd_req_write_object) + LANCER_FW_DOWNLOAD_CHUNK; flash_cmd.va = dma_alloc_coherent(&adapter->pdev->dev, flash_cmd.size, - &flash_cmd.dma, GFP_KERNEL); + &flash_cmd.dma, GFP_KERNEL); if (!flash_cmd.va) { status = -ENOMEM; - dev_err(&adapter->pdev->dev, - "Memory allocation failure while flashing\n"); goto lancer_fw_exit; } @@ -3463,7 +4057,7 @@ static int lancer_fw_download(struct be_adapter *adapter, } dma_free_coherent(&adapter->pdev->dev, flash_cmd.size, flash_cmd.va, - flash_cmd.dma); + flash_cmd.dma); if (status) { dev_err(&adapter->pdev->dev, "Firmware load error. " @@ -3473,7 +4067,10 @@ static int lancer_fw_download(struct be_adapter *adapter, } if (change_status == LANCER_FW_RESET_NEEDED) { - status = lancer_fw_reset(adapter); + dev_info(&adapter->pdev->dev, + "Resetting adapter to activate new FW\n"); + status = lancer_physdev_ctrl(adapter, + PHYSDEV_CONTROL_FW_RESET_MASK); if (status) { dev_err(&adapter->pdev->dev, "Adapter busy for FW reset.\n" @@ -3481,9 +4078,8 @@ static int lancer_fw_download(struct be_adapter *adapter, goto lancer_fw_exit; } } else if (change_status != LANCER_NO_RESET_NEEDED) { - dev_err(&adapter->pdev->dev, - "System reboot required for new FW" - " to be active\n"); + dev_err(&adapter->pdev->dev, + "System reboot required for new FW to be active\n"); } dev_info(&adapter->pdev->dev, "Firmware flashed successfully\n"); @@ -3493,18 +4089,22 @@ lancer_fw_exit: #define UFI_TYPE2 2 #define UFI_TYPE3 3 +#define UFI_TYPE3R 10 #define UFI_TYPE4 4 static int be_get_ufi_type(struct be_adapter *adapter, - struct flash_file_hdr_g2 *fhdr) + struct flash_file_hdr_g3 *fhdr) { if (fhdr == NULL) goto be_get_ufi_exit; if (skyhawk_chip(adapter) && fhdr->build[0] == '4') return UFI_TYPE4; - else if (BE3_chip(adapter) && fhdr->build[0] == '3') - return UFI_TYPE3; - else if (BE2_chip(adapter) && fhdr->build[0] == '2') + else if (BE3_chip(adapter) && fhdr->build[0] == '3') { + if (fhdr->asic_type_rev == 0x10) + return UFI_TYPE3R; + else + return UFI_TYPE3; + } else if (BE2_chip(adapter) && fhdr->build[0] == '2') return UFI_TYPE2; be_get_ufi_exit: @@ -3515,7 +4115,6 @@ be_get_ufi_exit: static int be_fw_download(struct be_adapter *adapter, const struct firmware* fw) { - struct flash_file_hdr_g2 *fhdr; struct flash_file_hdr_g3 *fhdr3; struct image_hdr *img_hdr_ptr = NULL; struct be_dma_mem flash_cmd; @@ -3527,29 +4126,41 @@ static int be_fw_download(struct be_adapter *adapter, const struct firmware* fw) &flash_cmd.dma, GFP_KERNEL); if (!flash_cmd.va) { status = -ENOMEM; - dev_err(&adapter->pdev->dev, - "Memory allocation failure while flashing\n"); goto be_fw_exit; } p = fw->data; - fhdr = (struct flash_file_hdr_g2 *)p; + fhdr3 = (struct flash_file_hdr_g3 *)p; - ufi_type = be_get_ufi_type(adapter, fhdr); + ufi_type = be_get_ufi_type(adapter, fhdr3); - fhdr3 = (struct flash_file_hdr_g3 *)fw->data; num_imgs = le32_to_cpu(fhdr3->num_imgs); for (i = 0; i < num_imgs; i++) { img_hdr_ptr = (struct image_hdr *)(fw->data + (sizeof(struct flash_file_hdr_g3) + i * sizeof(struct image_hdr))); if (le32_to_cpu(img_hdr_ptr->imageid) == 1) { - if (ufi_type == UFI_TYPE4) + switch (ufi_type) { + case UFI_TYPE4: status = be_flash_skyhawk(adapter, fw, - &flash_cmd, num_imgs); - else if (ufi_type == UFI_TYPE3) + &flash_cmd, num_imgs); + break; + case UFI_TYPE3R: status = be_flash_BEx(adapter, fw, &flash_cmd, num_imgs); + break; + case UFI_TYPE3: + /* Do not flash this ufi on BE3-R cards */ + if (adapter->asic_rev < 0x10) + status = be_flash_BEx(adapter, fw, + &flash_cmd, + num_imgs); + else { + status = -1; + dev_err(&adapter->pdev->dev, + "Can't load BE3 UFI on BE3R\n"); + } + } } } @@ -3593,11 +4204,142 @@ int be_load_fw(struct be_adapter *adapter, u8 *fw_file) else status = be_fw_download(adapter, fw); + if (!status) + be_cmd_get_fw_ver(adapter, adapter->fw_ver, + adapter->fw_on_flash); + fw_exit: release_firmware(fw); return status; } +static int be_ndo_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh) +{ + struct be_adapter *adapter = netdev_priv(dev); + struct nlattr *attr, *br_spec; + int rem; + int status = 0; + u16 mode = 0; + + if (!sriov_enabled(adapter)) + return -EOPNOTSUPP; + + br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC); + + nla_for_each_nested(attr, br_spec, rem) { + if (nla_type(attr) != IFLA_BRIDGE_MODE) + continue; + + mode = nla_get_u16(attr); + if (mode != BRIDGE_MODE_VEPA && mode != BRIDGE_MODE_VEB) + return -EINVAL; + + status = be_cmd_set_hsw_config(adapter, 0, 0, + adapter->if_handle, + mode == BRIDGE_MODE_VEPA ? + PORT_FWD_TYPE_VEPA : + PORT_FWD_TYPE_VEB); + if (status) + goto err; + + dev_info(&adapter->pdev->dev, "enabled switch mode: %s\n", + mode == BRIDGE_MODE_VEPA ? "VEPA" : "VEB"); + + return status; + } +err: + dev_err(&adapter->pdev->dev, "Failed to set switch mode %s\n", + mode == BRIDGE_MODE_VEPA ? "VEPA" : "VEB"); + + return status; +} + +static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, + struct net_device *dev, u32 filter_mask) +{ + struct be_adapter *adapter = netdev_priv(dev); + int status = 0; + u8 hsw_mode; + + if (!sriov_enabled(adapter)) + return 0; + + /* BE and Lancer chips support VEB mode only */ + if (BEx_chip(adapter) || lancer_chip(adapter)) { + hsw_mode = PORT_FWD_TYPE_VEB; + } else { + status = be_cmd_get_hsw_config(adapter, NULL, 0, + adapter->if_handle, &hsw_mode); + if (status) + return 0; + } + + return ndo_dflt_bridge_getlink(skb, pid, seq, dev, + hsw_mode == PORT_FWD_TYPE_VEPA ? + BRIDGE_MODE_VEPA : BRIDGE_MODE_VEB); +} + +#ifdef CONFIG_BE2NET_VXLAN +static void be_add_vxlan_port(struct net_device *netdev, sa_family_t sa_family, + __be16 port) +{ + struct be_adapter *adapter = netdev_priv(netdev); + struct device *dev = &adapter->pdev->dev; + int status; + + if (lancer_chip(adapter) || BEx_chip(adapter)) + return; + + if (adapter->flags & BE_FLAGS_VXLAN_OFFLOADS) { + dev_warn(dev, "Cannot add UDP port %d for VxLAN offloads\n", + be16_to_cpu(port)); + dev_info(dev, + "Only one UDP port supported for VxLAN offloads\n"); + return; + } + + status = be_cmd_manage_iface(adapter, adapter->if_handle, + OP_CONVERT_NORMAL_TO_TUNNEL); + if (status) { + dev_warn(dev, "Failed to convert normal interface to tunnel\n"); + goto err; + } + + status = be_cmd_set_vxlan_port(adapter, port); + if (status) { + dev_warn(dev, "Failed to add VxLAN port\n"); + goto err; + } + adapter->flags |= BE_FLAGS_VXLAN_OFFLOADS; + adapter->vxlan_port = port; + + dev_info(dev, "Enabled VxLAN offloads for UDP port %d\n", + be16_to_cpu(port)); + return; +err: + be_disable_vxlan_offloads(adapter); + return; +} + +static void be_del_vxlan_port(struct net_device *netdev, sa_family_t sa_family, + __be16 port) +{ + struct be_adapter *adapter = netdev_priv(netdev); + + if (lancer_chip(adapter) || BEx_chip(adapter)) + return; + + if (adapter->vxlan_port != port) + return; + + be_disable_vxlan_offloads(adapter); + + dev_info(&adapter->pdev->dev, + "Disabled VxLAN offloads for UDP port %d\n", + be16_to_cpu(port)); +} +#endif + static const struct net_device_ops be_netdev_ops = { .ndo_open = be_open, .ndo_stop = be_close, @@ -3611,27 +4353,41 @@ static const struct net_device_ops be_netdev_ops = { .ndo_vlan_rx_kill_vid = be_vlan_rem_vid, .ndo_set_vf_mac = be_set_vf_mac, .ndo_set_vf_vlan = be_set_vf_vlan, - .ndo_set_vf_tx_rate = be_set_vf_tx_rate, + .ndo_set_vf_rate = be_set_vf_tx_rate, .ndo_get_vf_config = be_get_vf_config, + .ndo_set_vf_link_state = be_set_vf_link_state, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = be_netpoll, #endif + .ndo_bridge_setlink = be_ndo_bridge_setlink, + .ndo_bridge_getlink = be_ndo_bridge_getlink, +#ifdef CONFIG_NET_RX_BUSY_POLL + .ndo_busy_poll = be_busy_poll, +#endif +#ifdef CONFIG_BE2NET_VXLAN + .ndo_add_vxlan_port = be_add_vxlan_port, + .ndo_del_vxlan_port = be_del_vxlan_port, +#endif }; static void be_netdev_init(struct net_device *netdev) { struct be_adapter *adapter = netdev_priv(netdev); - struct be_eq_obj *eqo; - int i; + if (skyhawk_chip(adapter)) { + netdev->hw_enc_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | + NETIF_F_TSO | NETIF_F_TSO6 | + NETIF_F_GSO_UDP_TUNNEL; + netdev->hw_features |= NETIF_F_GSO_UDP_TUNNEL; + } netdev->hw_features |= NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM | - NETIF_F_HW_VLAN_TX; + NETIF_F_HW_VLAN_CTAG_TX; if (be_multi_rxq(adapter)) netdev->hw_features |= NETIF_F_RXHASH; netdev->features |= netdev->hw_features | - NETIF_F_HW_VLAN_RX | NETIF_F_HW_VLAN_FILTER; + NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER; netdev->vlan_features |= NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; @@ -3644,14 +4400,13 @@ static void be_netdev_init(struct net_device *netdev) netdev->netdev_ops = &be_netdev_ops; - SET_ETHTOOL_OPS(netdev, &be_ethtool_ops); - - for_all_evt_queues(adapter, eqo, i) - netif_napi_add(netdev, &eqo->napi, be_poll, BE_NAPI_WEIGHT); + netdev->ethtool_ops = &be_ethtool_ops; } static void be_unmap_pci_bars(struct be_adapter *adapter) { + if (adapter->csr) + pci_iounmap(adapter->pdev, adapter->csr); if (adapter->db) pci_iounmap(adapter->pdev, adapter->db); } @@ -3679,11 +4434,12 @@ static int be_roce_map_pci_bars(struct be_adapter *adapter) static int be_map_pci_bars(struct be_adapter *adapter) { u8 __iomem *addr; - u32 sli_intf; - pci_read_config_dword(adapter->pdev, SLI_INTF_REG_OFFSET, &sli_intf); - adapter->if_type = (sli_intf & SLI_INTF_IF_TYPE_MASK) >> - SLI_INTF_IF_TYPE_SHIFT; + if (BEx_chip(adapter) && be_physfn(adapter)) { + adapter->csr = pci_iomap(adapter->pdev, 2, 0); + if (adapter->csr == NULL) + return -ENOMEM; + } addr = pci_iomap(adapter->pdev, db_bar(adapter), 0); if (addr == NULL) @@ -3746,18 +4502,19 @@ static int be_ctrl_init(struct be_adapter *adapter) memset(mbox_mem_align->va, 0, sizeof(struct be_mcc_mailbox)); rx_filter->size = sizeof(struct be_cmd_req_rx_filter); - rx_filter->va = dma_alloc_coherent(&adapter->pdev->dev, rx_filter->size, - &rx_filter->dma, GFP_KERNEL); + rx_filter->va = dma_zalloc_coherent(&adapter->pdev->dev, + rx_filter->size, &rx_filter->dma, + GFP_KERNEL); if (rx_filter->va == NULL) { status = -ENOMEM; goto free_mbox; } - memset(rx_filter->va, 0, rx_filter->size); + mutex_init(&adapter->mbox_lock); spin_lock_init(&adapter->mcc_lock); spin_lock_init(&adapter->mcc_cq_lock); - init_completion(&adapter->flash_compl); + init_completion(&adapter->et_cmd_compl); pci_save_state(adapter->pdev); return 0; @@ -3789,15 +4546,16 @@ static int be_stats_init(struct be_adapter *adapter) cmd->size = sizeof(struct lancer_cmd_req_pport_stats); else if (BE2_chip(adapter)) cmd->size = sizeof(struct be_cmd_req_get_stats_v0); - else - /* BE3 and Skyhawk */ + else if (BE3_chip(adapter)) cmd->size = sizeof(struct be_cmd_req_get_stats_v1); + else + /* ALL non-BE ASICs */ + cmd->size = sizeof(struct be_cmd_req_get_stats_v2); - cmd->va = dma_alloc_coherent(&adapter->pdev->dev, cmd->size, &cmd->dma, - GFP_KERNEL); + cmd->va = dma_zalloc_coherent(&adapter->pdev->dev, cmd->size, &cmd->dma, + GFP_KERNEL); if (cmd->va == NULL) return -1; - memset(cmd->va, 0, cmd->size); return 0; } @@ -3809,6 +4567,7 @@ static void be_remove(struct pci_dev *pdev) return; be_roce_dev_remove(adapter); + be_intr_set(adapter, false); cancel_delayed_work_sync(&adapter->func_recovery_work); @@ -3825,87 +4584,36 @@ static void be_remove(struct pci_dev *pdev) pci_disable_pcie_error_reporting(pdev); - pci_set_drvdata(pdev, NULL); pci_release_regions(pdev); pci_disable_device(pdev); free_netdev(adapter->netdev); } -bool be_is_wol_supported(struct be_adapter *adapter) -{ - return ((adapter->wol_cap & BE_WOL_CAP) && - !be_is_wol_excluded(adapter)) ? true : false; -} - -u32 be_get_fw_log_level(struct be_adapter *adapter) -{ - struct be_dma_mem extfat_cmd; - struct be_fat_conf_params *cfgs; - int status; - u32 level = 0; - int j; - - if (lancer_chip(adapter)) - return 0; - - memset(&extfat_cmd, 0, sizeof(struct be_dma_mem)); - extfat_cmd.size = sizeof(struct be_cmd_resp_get_ext_fat_caps); - extfat_cmd.va = pci_alloc_consistent(adapter->pdev, extfat_cmd.size, - &extfat_cmd.dma); - - if (!extfat_cmd.va) { - dev_err(&adapter->pdev->dev, "%s: Memory allocation failure\n", - __func__); - goto err; - } - - status = be_cmd_get_ext_fat_capabilites(adapter, &extfat_cmd); - if (!status) { - cfgs = (struct be_fat_conf_params *)(extfat_cmd.va + - sizeof(struct be_cmd_resp_hdr)); - for (j = 0; j < le32_to_cpu(cfgs->module[0].num_modes); j++) { - if (cfgs->module[0].trace_lvl[j].mode == MODE_UART) - level = cfgs->module[0].trace_lvl[j].dbg_lvl; - } - } - pci_free_consistent(adapter->pdev, extfat_cmd.size, extfat_cmd.va, - extfat_cmd.dma); -err: - return level; -} - static int be_get_initial_config(struct be_adapter *adapter) { - int status; - u32 level; + int status, level; status = be_cmd_get_cntl_attributes(adapter); if (status) return status; - status = be_cmd_get_acpi_wol_cap(adapter); - if (status) { - /* in case of a failure to get wol capabillities - * check the exclusion list to determine WOL capability */ - if (!be_is_wol_excluded(adapter)) - adapter->wol_cap |= BE_WOL_CAP; - } - - if (be_is_wol_supported(adapter)) - adapter->wol = true; - /* Must be a power of 2 or else MODULO will BUG_ON */ adapter->be_get_temp_freq = 64; - level = be_get_fw_log_level(adapter); - adapter->msg_enable = level <= FW_LOG_LEVEL_DEFAULT ? NETIF_MSG_HW : 0; + if (BEx_chip(adapter)) { + level = be_cmd_get_fw_log_level(adapter); + adapter->msg_enable = + level <= FW_LOG_LEVEL_DEFAULT ? NETIF_MSG_HW : 0; + } + adapter->cfg_num_qs = netif_get_num_default_rss_queues(); return 0; } static int lancer_recover_func(struct be_adapter *adapter) { + struct device *dev = &adapter->pdev->dev; int status; status = lancer_test_and_set_rdy_state(adapter); @@ -3917,8 +4625,7 @@ static int lancer_recover_func(struct be_adapter *adapter) be_clear(adapter); - adapter->hw_error = false; - adapter->fw_timeout = false; + be_clear_all_error(adapter); status = be_setup(adapter); if (status) @@ -3930,13 +4637,13 @@ static int lancer_recover_func(struct be_adapter *adapter) goto err; } - dev_err(&adapter->pdev->dev, - "Adapter SLIPORT recovery succeeded\n"); + dev_err(dev, "Adapter recovery successful\n"); return 0; err: - if (adapter->eeh_error) - dev_err(&adapter->pdev->dev, - "Adapter SLIPORT recovery failed\n"); + if (status == -EAGAIN) + dev_err(dev, "Waiting for resource provisioning\n"); + else + dev_err(dev, "Adapter recovery failed\n"); return status; } @@ -3945,28 +4652,27 @@ static void be_func_recovery_task(struct work_struct *work) { struct be_adapter *adapter = container_of(work, struct be_adapter, func_recovery_work.work); - int status; + int status = 0; be_detect_error(adapter); if (adapter->hw_error && lancer_chip(adapter)) { - if (adapter->eeh_error) - goto out; - rtnl_lock(); netif_device_detach(adapter->netdev); rtnl_unlock(); status = lancer_recover_func(adapter); - if (!status) netif_device_attach(adapter->netdev); } -out: - schedule_delayed_work(&adapter->func_recovery_work, - msecs_to_jiffies(1000)); + /* In Lancer, for all errors other than provisioning error (-EAGAIN), + * no need to attempt further recovery. + */ + if (!status || status == -EAGAIN) + schedule_delayed_work(&adapter->func_recovery_work, + msecs_to_jiffies(1000)); } static void be_worker(struct work_struct *work) @@ -3974,7 +4680,6 @@ static void be_worker(struct work_struct *work) struct be_adapter *adapter = container_of(work, struct be_adapter, work.work); struct be_rx_obj *rxo; - struct be_eq_obj *eqo; int i; /* when interrupts are not yet enabled, just reap any pending @@ -3994,39 +4699,59 @@ static void be_worker(struct work_struct *work) be_cmd_get_stats(adapter, &adapter->stats_cmd); } - if (MODULO(adapter->work_counter, adapter->be_get_temp_freq) == 0) + if (be_physfn(adapter) && + MODULO(adapter->work_counter, adapter->be_get_temp_freq) == 0) be_cmd_get_die_temperature(adapter); for_all_rx_queues(adapter, rxo, i) { - if (rxo->rx_post_starved) { - rxo->rx_post_starved = false; + /* Replenish RX-queues starved due to memory + * allocation failures. + */ + if (rxo->rx_post_starved) be_post_rx_frags(rxo, GFP_KERNEL); - } } - for_all_evt_queues(adapter, eqo, i) - be_eqd_update(adapter, eqo); + be_eqd_update(adapter); reschedule: adapter->work_counter++; schedule_delayed_work(&adapter->work, msecs_to_jiffies(1000)); } +/* If any VFs are already enabled don't FLR the PF */ static bool be_reset_required(struct be_adapter *adapter) { - return be_find_vfs(adapter, ENABLED) > 0 ? false : true; + return pci_num_vf(adapter->pdev) ? false : true; } static char *mc_name(struct be_adapter *adapter) { - if (adapter->function_mode & FLEX10_MODE) - return "FLEX10"; - else if (adapter->function_mode & VNIC_MODE) - return "vNIC"; - else if (adapter->function_mode & UMC_ENABLED) - return "UMC"; - else - return ""; + char *str = ""; /* default */ + + switch (adapter->mc_type) { + case UMC: + str = "UMC"; + break; + case FLEX10: + str = "FLEX10"; + break; + case vNIC1: + str = "vNIC-1"; + break; + case nPAR: + str = "nPAR"; + break; + case UFP: + str = "UFP"; + break; + case vNIC2: + str = "vNIC-2"; + break; + default: + str = ""; + } + + return str; } static inline char *func_name(struct be_adapter *adapter) @@ -4061,20 +4786,22 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id) adapter->netdev = netdev; SET_NETDEV_DEV(netdev, &pdev->dev); - status = dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)); + status = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); if (!status) { netdev->features |= NETIF_F_HIGHDMA; } else { - status = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); + status = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32)); if (status) { dev_err(&pdev->dev, "Could not set PCI DMA Mask\n"); goto free_netdev; } } - status = pci_enable_pcie_error_reporting(pdev); - if (status) - dev_err(&pdev->dev, "Could not use PCIe error reporting\n"); + if (be_physfn(adapter)) { + status = pci_enable_pcie_error_reporting(pdev); + if (!status) + dev_info(&pdev->dev, "PCIe error reporting enabled\n"); + } status = be_ctrl_init(adapter); if (status) @@ -4087,22 +4814,22 @@ static int be_probe(struct pci_dev *pdev, const struct pci_device_id *pdev_id) goto ctrl_clean; } - /* tell fw we're ready to fire cmds */ - status = be_cmd_fw_init(adapter); - if (status) - goto ctrl_clean; - if (be_reset_required(adapter)) { status = be_cmd_reset_function(adapter); if (status) goto ctrl_clean; + + /* Wait for interrupts to quiesce after an FLR */ + msleep(100); } - /* The INTR bit may be set in the card when probed by a kdump kernel - * after a crash. - */ - if (!lancer_chip(adapter)) - be_intr_set(adapter, false); + /* Allow interrupts for other ULPs running on NIC function */ + be_intr_set(adapter, true); + + /* tell fw we're ready to fire cmds */ + status = be_cmd_fw_init(adapter); + if (status) + goto ctrl_clean; status = be_stats_init(adapter); if (status) @@ -4145,7 +4872,6 @@ ctrl_clean: be_ctrl_cleanup(adapter); free_netdev: free_netdev(netdev); - pci_set_drvdata(pdev, NULL); rel_reg: pci_release_regions(pdev); disable_dev: @@ -4160,9 +4886,10 @@ static int be_suspend(struct pci_dev *pdev, pm_message_t state) struct be_adapter *adapter = pci_get_drvdata(pdev); struct net_device *netdev = adapter->netdev; - if (adapter->wol) + if (adapter->wol_en) be_setup_wol(adapter, true); + be_intr_set(adapter, false); cancel_delayed_work_sync(&adapter->func_recovery_work); netif_device_detach(netdev); @@ -4191,9 +4918,14 @@ static int be_resume(struct pci_dev *pdev) if (status) return status; - pci_set_power_state(pdev, 0); + pci_set_power_state(pdev, PCI_D0); pci_restore_state(pdev); + status = be_fw_wait_ready(adapter); + if (status) + return status; + + be_intr_set(adapter, true); /* tell fw we're ready to fire cmds */ status = be_cmd_fw_init(adapter); if (status) @@ -4210,7 +4942,7 @@ static int be_resume(struct pci_dev *pdev) msecs_to_jiffies(1000)); netif_device_attach(netdev); - if (adapter->wol) + if (adapter->wol_en) be_setup_wol(adapter, false); return 0; @@ -4237,27 +4969,26 @@ static void be_shutdown(struct pci_dev *pdev) } static pci_ers_result_t be_eeh_err_detected(struct pci_dev *pdev, - pci_channel_state_t state) + pci_channel_state_t state) { struct be_adapter *adapter = pci_get_drvdata(pdev); struct net_device *netdev = adapter->netdev; dev_err(&adapter->pdev->dev, "EEH error detected\n"); - adapter->eeh_error = true; - - cancel_delayed_work_sync(&adapter->func_recovery_work); + if (!adapter->eeh_error) { + adapter->eeh_error = true; - rtnl_lock(); - netif_device_detach(netdev); - rtnl_unlock(); + cancel_delayed_work_sync(&adapter->func_recovery_work); - if (netif_running(netdev)) { rtnl_lock(); - be_close(netdev); + netif_device_detach(netdev); + if (netif_running(netdev)) + be_close(netdev); rtnl_unlock(); + + be_clear(adapter); } - be_clear(adapter); if (state == pci_channel_io_perm_failure) return PCI_ERS_RESULT_DISCONNECT; @@ -4282,22 +5013,24 @@ static pci_ers_result_t be_eeh_reset(struct pci_dev *pdev) int status; dev_info(&adapter->pdev->dev, "EEH reset\n"); - be_clear_all_error(adapter); status = pci_enable_device(pdev); if (status) return PCI_ERS_RESULT_DISCONNECT; pci_set_master(pdev); - pci_set_power_state(pdev, 0); + pci_set_power_state(pdev, PCI_D0); pci_restore_state(pdev); /* Check if card is ok and fw is ready */ + dev_info(&adapter->pdev->dev, + "Waiting for FW to be ready after EEH reset\n"); status = be_fw_wait_ready(adapter); if (status) return PCI_ERS_RESULT_DISCONNECT; pci_cleanup_aer_uncorrect_error_status(pdev); + be_clear_all_error(adapter); return PCI_ERS_RESULT_RECOVERED; } @@ -4311,12 +5044,18 @@ static void be_eeh_resume(struct pci_dev *pdev) pci_save_state(pdev); - /* tell fw we're ready to fire cmds */ - status = be_cmd_fw_init(adapter); + status = be_cmd_reset_function(adapter); if (status) goto err; - status = be_cmd_reset_function(adapter); + /* On some BE3 FW versions, after a HW reset, + * interrupts will remain disabled for each function. + * So, explicitly enable interrupts + */ + be_intr_set(adapter, true); + + /* tell fw we're ready to fire cmds */ + status = be_cmd_fw_init(adapter); if (status) goto err; |