diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2015-11-04 09:41:05 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2015-11-04 09:41:05 -0800 |
commit | b0f85fa11aefc4f3e03306b4cd47f113bd57dcba (patch) | |
tree | 1333d36d99fde3f97210795941fc246f0ad08a75 /drivers/net/ethernet/intel/i40e/i40e_main.c | |
parent | ccc9d4a6d640cbde05d519edeb727881646cf71b (diff) | |
parent | f32bfb9a8ca083f8d148ea90ae5ba66f4831836e (diff) | |
download | linux-b0f85fa11aefc4f3e03306b4cd47f113bd57dcba.tar.bz2 |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
Pull networking updates from David Miller:
Changes of note:
1) Allow to schedule ICMP packets in IPVS, from Alex Gartrell.
2) Provide FIB table ID in ipv4 route dumps just as ipv6 does, from
David Ahern.
3) Allow the user to ask for the statistics to be filtered out of
ipv4/ipv6 address netlink dumps. From Sowmini Varadhan.
4) More work to pass the network namespace context around deep into
various packet path APIs, starting with the netfilter hooks. From
Eric W Biederman.
5) Add layer 2 TX/RX checksum offloading to qeth driver, from Thomas
Richter.
6) Use usec resolution for SYN/ACK RTTs in TCP, from Yuchung Cheng.
7) Support Very High Throughput in wireless MESH code, from Bob
Copeland.
8) Allow setting the ageing_time in switchdev/rocker. From Scott
Feldman.
9) Properly autoload L2TP type modules, from Stephen Hemminger.
10) Fix and enable offload features by default in 8139cp driver, from
David Woodhouse.
11) Support both ipv4 and ipv6 sockets in a single vxlan device, from
Jiri Benc.
12) Fix CWND limiting of thin streams in TCP, from Bendik Rønning
Opstad.
13) Fix IPSEC flowcache overflows on large systems, from Steffen
Klassert.
14) Convert bridging to track VLANs using rhashtable entries rather than
a bitmap. From Nikolay Aleksandrov.
15) Make TCP listener handling completely lockless, this is a major
accomplishment. Incoming request sockets now live in the
established hash table just like any other socket too.
From Eric Dumazet.
15) Provide more bridging attributes to netlink, from Nikolay
Aleksandrov.
16) Use hash based algorithm for ipv4 multipath routing, this was very
long overdue. From Peter Nørlund.
17) Several y2038 cures, mostly avoiding timespec. From Arnd Bergmann.
18) Allow non-root execution of EBPF programs, from Alexei Starovoitov.
19) Support SO_INCOMING_CPU as setsockopt, from Eric Dumazet. This
influences the port binding selection logic used by SO_REUSEPORT.
20) Add ipv6 support to VRF, from David Ahern.
21) Add support for Mellanox Spectrum switch ASIC, from Jiri Pirko.
22) Add rtl8xxxu Realtek wireless driver, from Jes Sorensen.
23) Implement RACK loss recovery in TCP, from Yuchung Cheng.
24) Support multipath routes in MPLS, from Roopa Prabhu.
25) Fix POLLOUT notification for listening sockets in AF_UNIX, from Eric
Dumazet.
26) Add new QED Qlogic river, from Yuval Mintz, Manish Chopra, and
Sudarsana Kalluru.
27) Don't fetch timestamps on AF_UNIX sockets, from Hannes Frederic
Sowa.
28) Support ipv6 geneve tunnels, from John W Linville.
29) Add flood control support to switchdev layer, from Ido Schimmel.
30) Fix CHECKSUM_PARTIAL handling of potentially fragmented frames, from
Hannes Frederic Sowa.
31) Support persistent maps and progs in bpf, from Daniel Borkmann.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1790 commits)
sh_eth: use DMA barriers
switchdev: respect SKIP_EOPNOTSUPP flag in case there is no recursion
net: sched: kill dead code in sch_choke.c
irda: Delete an unnecessary check before the function call "irlmp_unregister_service"
net: dsa: mv88e6xxx: include DSA ports in VLANs
net: dsa: mv88e6xxx: disable SA learning for DSA and CPU ports
net/core: fix for_each_netdev_feature
vlan: Invoke driver vlan hooks only if device is present
arcnet/com20020: add LEDS_CLASS dependency
bpf, verifier: annotate verbose printer with __printf
dp83640: Only wait for timestamps for packets with timestamping enabled.
ptp: Change ptp_class to a proper bitmask
dp83640: Prune rx timestamp list before reading from it
dp83640: Delay scheduled work.
dp83640: Include hash in timestamp/packet matching
ipv6: fix tunnel error handling
net/mlx5e: Fix LSO vlan insertion
net/mlx5e: Re-eanble client vlan TX acceleration
net/mlx5e: Return error in case mlx5e_set_features() fails
net/mlx5e: Don't allow more than max supported channels
...
Diffstat (limited to 'drivers/net/ethernet/intel/i40e/i40e_main.c')
-rw-r--r-- | drivers/net/ethernet/intel/i40e/i40e_main.c | 1159 |
1 files changed, 763 insertions, 396 deletions
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 3dd26cdd0bf2..b825f978d441 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -39,7 +39,7 @@ static const char i40e_driver_string[] = #define DRV_VERSION_MAJOR 1 #define DRV_VERSION_MINOR 3 -#define DRV_VERSION_BUILD 9 +#define DRV_VERSION_BUILD 46 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) DRV_KERN @@ -75,10 +75,13 @@ static const struct pci_device_id i40e_pci_tbl[] = { {PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_B), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_C), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T4), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2_A), 0}, /* required last entry */ {0, } }; @@ -213,10 +216,10 @@ static int i40e_get_lump(struct i40e_pf *pf, struct i40e_lump_tracking *pile, ret = i; pile->search_hint = i + j; break; - } else { - /* not enough, so skip over it and continue looking */ - i += j; } + + /* not enough, so skip over it and continue looking */ + i += j; } return ret; @@ -299,25 +302,69 @@ static void i40e_tx_timeout(struct net_device *netdev) struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_vsi *vsi = np->vsi; struct i40e_pf *pf = vsi->back; + struct i40e_ring *tx_ring = NULL; + unsigned int i, hung_queue = 0; + u32 head, val; pf->tx_timeout_count++; + /* find the stopped queue the same way the stack does */ + for (i = 0; i < netdev->num_tx_queues; i++) { + struct netdev_queue *q; + unsigned long trans_start; + + q = netdev_get_tx_queue(netdev, i); + trans_start = q->trans_start ? : netdev->trans_start; + if (netif_xmit_stopped(q) && + time_after(jiffies, + (trans_start + netdev->watchdog_timeo))) { + hung_queue = i; + break; + } + } + + if (i == netdev->num_tx_queues) { + netdev_info(netdev, "tx_timeout: no netdev hung queue found\n"); + } else { + /* now that we have an index, find the tx_ring struct */ + for (i = 0; i < vsi->num_queue_pairs; i++) { + if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) { + if (hung_queue == + vsi->tx_rings[i]->queue_index) { + tx_ring = vsi->tx_rings[i]; + break; + } + } + } + } + if (time_after(jiffies, (pf->tx_timeout_last_recovery + HZ*20))) - pf->tx_timeout_recovery_level = 1; + pf->tx_timeout_recovery_level = 1; /* reset after some time */ + else if (time_before(jiffies, + (pf->tx_timeout_last_recovery + netdev->watchdog_timeo))) + return; /* don't do any new action before the next timeout */ + + if (tx_ring) { + head = i40e_get_head(tx_ring); + /* Read interrupt register */ + if (pf->flags & I40E_FLAG_MSIX_ENABLED) + val = rd32(&pf->hw, + I40E_PFINT_DYN_CTLN(tx_ring->q_vector->v_idx + + tx_ring->vsi->base_vector - 1)); + else + val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0); + + netdev_info(netdev, "tx_timeout: VSI_seid: %d, Q %d, NTC: 0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x, INT: 0x%x\n", + vsi->seid, hung_queue, tx_ring->next_to_clean, + head, tx_ring->next_to_use, + readl(tx_ring->tail), val); + } + pf->tx_timeout_last_recovery = jiffies; - netdev_info(netdev, "tx_timeout recovery level %d\n", - pf->tx_timeout_recovery_level); + netdev_info(netdev, "tx_timeout recovery level %d, hung_queue %d\n", + pf->tx_timeout_recovery_level, hung_queue); switch (pf->tx_timeout_recovery_level) { - case 0: - /* disable and re-enable queues for the VSI */ - if (in_interrupt()) { - set_bit(__I40E_REINIT_REQUESTED, &pf->state); - set_bit(__I40E_REINIT_REQUESTED, &vsi->state); - } else { - i40e_vsi_reinit_locked(vsi); - } - break; case 1: set_bit(__I40E_PF_RESET_REQUESTED, &pf->state); break; @@ -329,10 +376,9 @@ static void i40e_tx_timeout(struct net_device *netdev) break; default: netdev_err(netdev, "tx_timeout recovery unsuccessful\n"); - set_bit(__I40E_DOWN_REQUESTED, &pf->state); - set_bit(__I40E_DOWN_REQUESTED, &vsi->state); break; } + i40e_service_event_schedule(pf); pf->tx_timeout_recovery_level++; } @@ -431,6 +477,7 @@ static struct rtnl_link_stats64 *i40e_get_netdev_stats_struct( stats->tx_errors = vsi_stats->tx_errors; stats->tx_dropped = vsi_stats->tx_dropped; stats->rx_errors = vsi_stats->rx_errors; + stats->rx_dropped = vsi_stats->rx_dropped; stats->rx_crc_errors = vsi_stats->rx_crc_errors; stats->rx_length_errors = vsi_stats->rx_length_errors; @@ -456,11 +503,11 @@ void i40e_vsi_reset_stats(struct i40e_vsi *vsi) memset(&vsi->eth_stats_offsets, 0, sizeof(vsi->eth_stats_offsets)); if (vsi->rx_rings && vsi->rx_rings[0]) { for (i = 0; i < vsi->num_queue_pairs; i++) { - memset(&vsi->rx_rings[i]->stats, 0 , + memset(&vsi->rx_rings[i]->stats, 0, sizeof(vsi->rx_rings[i]->stats)); - memset(&vsi->rx_rings[i]->rx_stats, 0 , + memset(&vsi->rx_rings[i]->rx_stats, 0, sizeof(vsi->rx_rings[i]->rx_stats)); - memset(&vsi->tx_rings[i]->stats, 0 , + memset(&vsi->tx_rings[i]->stats, 0, sizeof(vsi->tx_rings[i]->stats)); memset(&vsi->tx_rings[i]->tx_stats, 0, sizeof(vsi->tx_rings[i]->tx_stats)); @@ -754,7 +801,6 @@ static void i40e_update_link_xoff_rx(struct i40e_pf *pf) struct i40e_hw_port_stats *nsd = &pf->stats; struct i40e_hw *hw = &pf->hw; u64 xoff = 0; - u16 i, v; if ((hw->fc.current_mode != I40E_FC_FULL) && (hw->fc.current_mode != I40E_FC_RX_PAUSE)) @@ -769,18 +815,6 @@ static void i40e_update_link_xoff_rx(struct i40e_pf *pf) if (!(nsd->link_xoff_rx - xoff)) return; - /* Clear the __I40E_HANG_CHECK_ARMED bit for all Tx rings */ - for (v = 0; v < pf->num_alloc_vsi; v++) { - struct i40e_vsi *vsi = pf->vsi[v]; - - if (!vsi || !vsi->tx_rings[0]) - continue; - - for (i = 0; i < vsi->num_queue_pairs; i++) { - struct i40e_ring *ring = vsi->tx_rings[i]; - clear_bit(__I40E_HANG_CHECK_ARMED, &ring->state); - } - } } /** @@ -796,7 +830,7 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf) bool xoff[I40E_MAX_TRAFFIC_CLASS] = {false}; struct i40e_dcbx_config *dcb_cfg; struct i40e_hw *hw = &pf->hw; - u16 i, v; + u16 i; u8 tc; dcb_cfg = &hw->local_dcbx_config; @@ -809,6 +843,7 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf) for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) { u64 prio_xoff = nsd->priority_xoff_rx[i]; + i40e_stat_update32(hw, I40E_GLPRT_PXOFFRXC(hw->port, i), pf->stat_offsets_loaded, &osd->priority_xoff_rx[i], @@ -821,23 +856,6 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf) tc = dcb_cfg->etscfg.prioritytable[i]; xoff[tc] = true; } - - /* Clear the __I40E_HANG_CHECK_ARMED bit for Tx rings */ - for (v = 0; v < pf->num_alloc_vsi; v++) { - struct i40e_vsi *vsi = pf->vsi[v]; - - if (!vsi || !vsi->tx_rings[0]) - continue; - - for (i = 0; i < vsi->num_queue_pairs; i++) { - struct i40e_ring *ring = vsi->tx_rings[i]; - - tc = ring->dcb_tc; - if (xoff[tc]) - clear_bit(__I40E_HANG_CHECK_ARMED, - &ring->state); - } - } } /** @@ -862,6 +880,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) u32 rx_page, rx_buf; u64 bytes, packets; unsigned int start; + u64 tx_linearize; u64 rx_p, rx_b; u64 tx_p, tx_b; u16 q; @@ -880,7 +899,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) */ rx_b = rx_p = 0; tx_b = tx_p = 0; - tx_restart = tx_busy = 0; + tx_restart = tx_busy = tx_linearize = 0; rx_page = 0; rx_buf = 0; rcu_read_lock(); @@ -897,6 +916,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) tx_p += packets; tx_restart += p->tx_stats.restart_queue; tx_busy += p->tx_stats.tx_busy; + tx_linearize += p->tx_stats.tx_linearize; /* Rx queue is part of the same block as Tx queue */ p = &p[1]; @@ -913,6 +933,7 @@ static void i40e_update_vsi_stats(struct i40e_vsi *vsi) rcu_read_unlock(); vsi->tx_restart = tx_restart; vsi->tx_busy = tx_busy; + vsi->tx_linearize = tx_linearize; vsi->rx_page_failed = rx_page; vsi->rx_buf_failed = rx_buf; @@ -1256,7 +1277,7 @@ bool i40e_is_vsi_in_vlan(struct i40e_vsi *vsi) * so we have to go through all the list in order to make sure */ list_for_each_entry(f, &vsi->mac_filter_list, list) { - if (f->vlan >= 0) + if (f->vlan >= 0 || vsi->info.pvid) return true; } @@ -1334,6 +1355,9 @@ static int i40e_rm_default_mac_filter(struct i40e_vsi *vsi, u8 *macaddr) * @is_netdev: make sure its a netdev filter, else doesn't matter * * Returns ptr to the filter object or NULL when no memory available. + * + * NOTE: This function is expected to be called with mac_filter_list_lock + * being held. **/ struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi, u8 *macaddr, s16 vlan, @@ -1392,6 +1416,9 @@ add_filter_out: * @vlan: the vlan * @is_vf: make sure it's a VF filter, else doesn't matter * @is_netdev: make sure it's a netdev filter, else doesn't matter + * + * NOTE: This function is expected to be called with mac_filter_list_lock + * being held. **/ void i40e_del_filter(struct i40e_vsi *vsi, u8 *macaddr, s16 vlan, @@ -1419,6 +1446,7 @@ void i40e_del_filter(struct i40e_vsi *vsi, } else { /* make sure we don't remove a filter in use by VF or netdev */ int min_f = 0; + min_f += (f->is_vf ? 1 : 0); min_f += (f->is_netdev ? 1 : 0); @@ -1477,6 +1505,7 @@ static int i40e_set_mac(struct net_device *netdev, void *p) if (vsi->type == I40E_VSI_MAIN) { i40e_status ret; + ret = i40e_aq_mac_address_write(&vsi->back->hw, I40E_AQC_WRITE_TYPE_LAA_WOL, addr->sa_data, NULL); @@ -1496,8 +1525,10 @@ static int i40e_set_mac(struct net_device *netdev, void *p) element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); } else { + spin_lock_bh(&vsi->mac_filter_list_lock); i40e_del_filter(vsi, netdev->dev_addr, I40E_VLAN_ANY, false, false); + spin_unlock_bh(&vsi->mac_filter_list_lock); } if (ether_addr_equal(addr->sa_data, hw->mac.addr)) { @@ -1508,13 +1539,15 @@ static int i40e_set_mac(struct net_device *netdev, void *p) element.flags = cpu_to_le16(I40E_AQC_MACVLAN_ADD_PERFECT_MATCH); i40e_aq_add_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); } else { + spin_lock_bh(&vsi->mac_filter_list_lock); f = i40e_add_filter(vsi, addr->sa_data, I40E_VLAN_ANY, false, false); if (f) f->is_laa = true; + spin_unlock_bh(&vsi->mac_filter_list_lock); } - i40e_sync_vsi_filters(vsi); + i40e_sync_vsi_filters(vsi, false); ether_addr_copy(netdev->dev_addr, addr->sa_data); return 0; @@ -1684,6 +1717,8 @@ static void i40e_set_rx_mode(struct net_device *netdev) struct netdev_hw_addr *mca; struct netdev_hw_addr *ha; + spin_lock_bh(&vsi->mac_filter_list_lock); + /* add addr if not already in the filter list */ netdev_for_each_uc_addr(uca, netdev) { if (!i40e_find_mac(vsi, uca->addr, false, true)) { @@ -1709,37 +1744,29 @@ static void i40e_set_rx_mode(struct net_device *netdev) /* remove filter if not in netdev list */ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { - bool found = false; if (!f->is_netdev) continue; - if (is_multicast_ether_addr(f->macaddr)) { - netdev_for_each_mc_addr(mca, netdev) { - if (ether_addr_equal(mca->addr, f->macaddr)) { - found = true; - break; - } - } - } else { - netdev_for_each_uc_addr(uca, netdev) { - if (ether_addr_equal(uca->addr, f->macaddr)) { - found = true; - break; - } - } + netdev_for_each_mc_addr(mca, netdev) + if (ether_addr_equal(mca->addr, f->macaddr)) + goto bottom_of_search_loop; - for_each_dev_addr(netdev, ha) { - if (ether_addr_equal(ha->addr, f->macaddr)) { - found = true; - break; - } - } - } - if (!found) - i40e_del_filter( - vsi, f->macaddr, I40E_VLAN_ANY, false, true); + netdev_for_each_uc_addr(uca, netdev) + if (ether_addr_equal(uca->addr, f->macaddr)) + goto bottom_of_search_loop; + + for_each_dev_addr(netdev, ha) + if (ether_addr_equal(ha->addr, f->macaddr)) + goto bottom_of_search_loop; + + /* f->macaddr wasn't found in uc, mc, or ha list so delete it */ + i40e_del_filter(vsi, f->macaddr, I40E_VLAN_ANY, false, true); + +bottom_of_search_loop: + continue; } + spin_unlock_bh(&vsi->mac_filter_list_lock); /* check for other flag changes */ if (vsi->current_netdev_flags != vsi->netdev->flags) { @@ -1749,20 +1776,96 @@ static void i40e_set_rx_mode(struct net_device *netdev) } /** + * i40e_mac_filter_entry_clone - Clones a MAC filter entry + * @src: source MAC filter entry to be clones + * + * Returns the pointer to newly cloned MAC filter entry or NULL + * in case of error + **/ +static struct i40e_mac_filter *i40e_mac_filter_entry_clone( + struct i40e_mac_filter *src) +{ + struct i40e_mac_filter *f; + + f = kzalloc(sizeof(*f), GFP_ATOMIC); + if (!f) + return NULL; + *f = *src; + + INIT_LIST_HEAD(&f->list); + + return f; +} + +/** + * i40e_undo_del_filter_entries - Undo the changes made to MAC filter entries + * @vsi: pointer to vsi struct + * @from: Pointer to list which contains MAC filter entries - changes to + * those entries needs to be undone. + * + * MAC filter entries from list were slated to be removed from device. + **/ +static void i40e_undo_del_filter_entries(struct i40e_vsi *vsi, + struct list_head *from) +{ + struct i40e_mac_filter *f, *ftmp; + + list_for_each_entry_safe(f, ftmp, from, list) { + f->changed = true; + /* Move the element back into MAC filter list*/ + list_move_tail(&f->list, &vsi->mac_filter_list); + } +} + +/** + * i40e_undo_add_filter_entries - Undo the changes made to MAC filter entries + * @vsi: pointer to vsi struct + * + * MAC filter entries from list were slated to be added from device. + **/ +static void i40e_undo_add_filter_entries(struct i40e_vsi *vsi) +{ + struct i40e_mac_filter *f, *ftmp; + + list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { + if (!f->changed && f->counter) + f->changed = true; + } +} + +/** + * i40e_cleanup_add_list - Deletes the element from add list and release + * memory + * @add_list: Pointer to list which contains MAC filter entries + **/ +static void i40e_cleanup_add_list(struct list_head *add_list) +{ + struct i40e_mac_filter *f, *ftmp; + + list_for_each_entry_safe(f, ftmp, add_list, list) { + list_del(&f->list); + kfree(f); + } +} + +/** * i40e_sync_vsi_filters - Update the VSI filter list to the HW * @vsi: ptr to the VSI + * @grab_rtnl: whether RTNL needs to be grabbed * * Push any outstanding VSI filter changes through the AdminQ. * * Returns 0 or error value **/ -int i40e_sync_vsi_filters(struct i40e_vsi *vsi) +int i40e_sync_vsi_filters(struct i40e_vsi *vsi, bool grab_rtnl) { - struct i40e_mac_filter *f, *ftmp; + struct list_head tmp_del_list, tmp_add_list; + struct i40e_mac_filter *f, *ftmp, *fclone; bool promisc_forced_on = false; bool add_happened = false; int filter_list_len = 0; u32 changed_flags = 0; + bool err_cond = false; i40e_status ret = 0; struct i40e_pf *pf; int num_add = 0; @@ -1783,17 +1886,13 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) vsi->current_netdev_flags = vsi->netdev->flags; } + INIT_LIST_HEAD(&tmp_del_list); + INIT_LIST_HEAD(&tmp_add_list); + if (vsi->flags & I40E_VSI_FLAG_FILTER_CHANGED) { vsi->flags &= ~I40E_VSI_FLAG_FILTER_CHANGED; - filter_list_len = pf->hw.aq.asq_buf_size / - sizeof(struct i40e_aqc_remove_macvlan_element_data); - del_list = kcalloc(filter_list_len, - sizeof(struct i40e_aqc_remove_macvlan_element_data), - GFP_KERNEL); - if (!del_list) - return -ENOMEM; - + spin_lock_bh(&vsi->mac_filter_list_lock); list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { if (!f->changed) continue; @@ -1801,6 +1900,58 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (f->counter != 0) continue; f->changed = false; + + /* Move the element into temporary del_list */ + list_move_tail(&f->list, &tmp_del_list); + } + + list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { + if (!f->changed) + continue; + + if (f->counter == 0) + continue; + f->changed = false; + + /* Clone MAC filter entry and add into temporary list */ + fclone = i40e_mac_filter_entry_clone(f); + if (!fclone) { + err_cond = true; + break; + } + list_add_tail(&fclone->list, &tmp_add_list); + } + + /* if failed to clone MAC filter entry - undo */ + if (err_cond) { + i40e_undo_del_filter_entries(vsi, &tmp_del_list); + i40e_undo_add_filter_entries(vsi); + } + spin_unlock_bh(&vsi->mac_filter_list_lock); + + if (err_cond) + i40e_cleanup_add_list(&tmp_add_list); + } + + /* Now process 'del_list' outside the lock */ + if (!list_empty(&tmp_del_list)) { + filter_list_len = pf->hw.aq.asq_buf_size / + sizeof(struct i40e_aqc_remove_macvlan_element_data); + del_list = kcalloc(filter_list_len, + sizeof(struct i40e_aqc_remove_macvlan_element_data), + GFP_KERNEL); + if (!del_list) { + i40e_cleanup_add_list(&tmp_add_list); + + /* Undo VSI's MAC filter entry element updates */ + spin_lock_bh(&vsi->mac_filter_list_lock); + i40e_undo_del_filter_entries(vsi, &tmp_del_list); + i40e_undo_add_filter_entries(vsi); + spin_unlock_bh(&vsi->mac_filter_list_lock); + return -ENOMEM; + } + + list_for_each_entry_safe(f, ftmp, &tmp_del_list, list) { cmd_flags = 0; /* add to delete list */ @@ -1813,10 +1964,6 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) del_list[num_del].flags = cmd_flags; num_del++; - /* unlink from filter list */ - list_del(&f->list); - kfree(f); - /* flush a full buffer */ if (num_del == filter_list_len) { ret = i40e_aq_remove_macvlan(&pf->hw, @@ -1827,12 +1974,18 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) memset(del_list, 0, sizeof(*del_list)); if (ret && aq_err != I40E_AQ_RC_ENOENT) - dev_info(&pf->pdev->dev, - "ignoring delete macvlan error, err %s, aq_err %s while flushing a full buffer\n", - i40e_stat_str(&pf->hw, ret), - i40e_aq_str(&pf->hw, aq_err)); + dev_err(&pf->pdev->dev, + "ignoring delete macvlan error, err %s, aq_err %s while flushing a full buffer\n", + i40e_stat_str(&pf->hw, ret), + i40e_aq_str(&pf->hw, aq_err)); } + /* Release memory for MAC filter entries which were + * synced up with HW. + */ + list_del(&f->list); + kfree(f); } + if (num_del) { ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid, del_list, num_del, NULL); @@ -1848,6 +2001,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) kfree(del_list); del_list = NULL; + } + + if (!list_empty(&tmp_add_list)) { /* do all the adds now */ filter_list_len = pf->hw.aq.asq_buf_size / @@ -1855,16 +2011,19 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) add_list = kcalloc(filter_list_len, sizeof(struct i40e_aqc_add_macvlan_element_data), GFP_KERNEL); - if (!add_list) + if (!add_list) { + /* Purge element from temporary lists */ + i40e_cleanup_add_list(&tmp_add_list); + + /* Undo add filter entries from VSI MAC filter list */ + spin_lock_bh(&vsi->mac_filter_list_lock); + i40e_undo_add_filter_entries(vsi); + spin_unlock_bh(&vsi->mac_filter_list_lock); return -ENOMEM; + } - list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { - if (!f->changed) - continue; + list_for_each_entry_safe(f, ftmp, &tmp_add_list, list) { - if (f->counter == 0) - continue; - f->changed = false; add_happened = true; cmd_flags = 0; @@ -1891,7 +2050,13 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) break; memset(add_list, 0, sizeof(*add_list)); } + /* Entries from tmp_add_list were cloned from MAC + * filter list, hence clean those cloned entries + */ + list_del(&f->list); + kfree(f); } + if (num_add) { ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid, add_list, num_add, NULL); @@ -1920,6 +2085,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* check for changes in promiscuous modes */ if (changed_flags & IFF_ALLMULTI) { bool cur_multipromisc; + cur_multipromisc = !!(vsi->current_netdev_flags & IFF_ALLMULTI); ret = i40e_aq_set_vsi_multicast_promiscuous(&vsi->back->hw, vsi->seid, @@ -1934,6 +2100,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } if ((changed_flags & IFF_PROMISC) || promisc_forced_on) { bool cur_promisc; + cur_promisc = (!!(vsi->current_netdev_flags & IFF_PROMISC) || test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state)); @@ -1945,7 +2112,11 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) */ if (pf->cur_promisc != cur_promisc) { pf->cur_promisc = cur_promisc; - i40e_do_reset_safe(pf, + if (grab_rtnl) + i40e_do_reset_safe(pf, + BIT(__I40E_PF_RESET_REQUESTED)); + else + i40e_do_reset(pf, BIT(__I40E_PF_RESET_REQUESTED)); } } else { @@ -1996,7 +2167,7 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf) for (v = 0; v < pf->num_alloc_vsi; v++) { if (pf->vsi[v] && (pf->vsi[v]->flags & I40E_VSI_FLAG_FILTER_CHANGED)) - i40e_sync_vsi_filters(pf->vsi[v]); + i40e_sync_vsi_filters(pf->vsi[v], true); } } @@ -2137,6 +2308,9 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) is_vf = (vsi->type == I40E_VSI_SRIOV); is_netdev = !!(vsi->netdev); + /* Locked once because all functions invoked below iterates list*/ + spin_lock_bh(&vsi->mac_filter_list_lock); + if (is_netdev) { add_f = i40e_add_filter(vsi, vsi->netdev->dev_addr, vid, is_vf, is_netdev); @@ -2144,6 +2318,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) dev_info(&vsi->back->pdev->dev, "Could not add vlan filter %d for %pM\n", vid, vsi->netdev->dev_addr); + spin_unlock_bh(&vsi->mac_filter_list_lock); return -ENOMEM; } } @@ -2154,6 +2329,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) dev_info(&vsi->back->pdev->dev, "Could not add vlan filter %d for %pM\n", vid, f->macaddr); + spin_unlock_bh(&vsi->mac_filter_list_lock); return -ENOMEM; } } @@ -2175,6 +2351,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) dev_info(&vsi->back->pdev->dev, "Could not add filter 0 for %pM\n", vsi->netdev->dev_addr); + spin_unlock_bh(&vsi->mac_filter_list_lock); return -ENOMEM; } } @@ -2183,27 +2360,33 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) /* Do not assume that I40E_VLAN_ANY should be reset to VLAN 0 */ if (vid > 0 && !vsi->info.pvid) { list_for_each_entry(f, &vsi->mac_filter_list, list) { - if (i40e_find_filter(vsi, f->macaddr, I40E_VLAN_ANY, - is_vf, is_netdev)) { - i40e_del_filter(vsi, f->macaddr, I40E_VLAN_ANY, - is_vf, is_netdev); - add_f = i40e_add_filter(vsi, f->macaddr, - 0, is_vf, is_netdev); - if (!add_f) { - dev_info(&vsi->back->pdev->dev, - "Could not add filter 0 for %pM\n", - f->macaddr); - return -ENOMEM; - } + if (!i40e_find_filter(vsi, f->macaddr, I40E_VLAN_ANY, + is_vf, is_netdev)) + continue; + i40e_del_filter(vsi, f->macaddr, I40E_VLAN_ANY, + is_vf, is_netdev); + add_f = i40e_add_filter(vsi, f->macaddr, + 0, is_vf, is_netdev); + if (!add_f) { + dev_info(&vsi->back->pdev->dev, + "Could not add filter 0 for %pM\n", + f->macaddr); + spin_unlock_bh(&vsi->mac_filter_list_lock); + return -ENOMEM; } } } + /* Make sure to release before sync_vsi_filter because that + * function will lock/unlock as necessary + */ + spin_unlock_bh(&vsi->mac_filter_list_lock); + if (test_bit(__I40E_DOWN, &vsi->back->state) || test_bit(__I40E_RESET_RECOVERY_PENDING, &vsi->back->state)) return 0; - return i40e_sync_vsi_filters(vsi); + return i40e_sync_vsi_filters(vsi, false); } /** @@ -2223,6 +2406,9 @@ int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid) is_vf = (vsi->type == I40E_VSI_SRIOV); is_netdev = !!(netdev); + /* Locked once because all functions invoked below iterates list */ + spin_lock_bh(&vsi->mac_filter_list_lock); + if (is_netdev) i40e_del_filter(vsi, netdev->dev_addr, vid, is_vf, is_netdev); @@ -2253,6 +2439,7 @@ int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid) dev_info(&vsi->back->pdev->dev, "Could not add filter %d for %pM\n", I40E_VLAN_ANY, netdev->dev_addr); + spin_unlock_bh(&vsi->mac_filter_list_lock); return -ENOMEM; } } @@ -2261,21 +2448,27 @@ int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid) list_for_each_entry(f, &vsi->mac_filter_list, list) { i40e_del_filter(vsi, f->macaddr, 0, is_vf, is_netdev); add_f = i40e_add_filter(vsi, f->macaddr, I40E_VLAN_ANY, - is_vf, is_netdev); + is_vf, is_netdev); if (!add_f) { dev_info(&vsi->back->pdev->dev, "Could not add filter %d for %pM\n", I40E_VLAN_ANY, f->macaddr); + spin_unlock_bh(&vsi->mac_filter_list_lock); return -ENOMEM; } } } + /* Make sure to release before sync_vsi_filter because that + * function with lock/unlock as necessary + */ + spin_unlock_bh(&vsi->mac_filter_list_lock); + if (test_bit(__I40E_DOWN, &vsi->back->state) || test_bit(__I40E_RESET_RECOVERY_PENDING, &vsi->back->state)) return 0; - return i40e_sync_vsi_filters(vsi); + return i40e_sync_vsi_filters(vsi, false); } /** @@ -2609,8 +2802,6 @@ static int i40e_configure_tx_ring(struct i40e_ring *ring) wr32(hw, I40E_QTX_CTL(pf_q), qtx_ctl); i40e_flush(hw); - clear_bit(__I40E_HANG_CHECK_ARMED, &ring->state); - /* cache tail off for easier writes later */ ring->tail = hw->hw_addr + I40E_QTX_TAIL(pf_q); @@ -2882,11 +3073,9 @@ static int i40e_vsi_configure(struct i40e_vsi *vsi) static void i40e_vsi_configure_msix(struct i40e_vsi *vsi) { struct i40e_pf *pf = vsi->back; - struct i40e_q_vector *q_vector; struct i40e_hw *hw = &pf->hw; u16 vector; int i, q; - u32 val; u32 qp; /* The interrupt indexing is offset by 1 in the PFINT_ITRn @@ -2896,7 +3085,9 @@ static void i40e_vsi_configure_msix(struct i40e_vsi *vsi) qp = vsi->base_queue; vector = vsi->base_vector; for (i = 0; i < vsi->num_q_vectors; i++, vector++) { - q_vector = vsi->q_vectors[i]; + struct i40e_q_vector *q_vector = vsi->q_vectors[i]; + + q_vector->itr_countdown = ITR_COUNTDOWN_START; q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting); q_vector->rx.latency_range = I40E_LOW_LATENCY; wr32(hw, I40E_PFINT_ITRN(I40E_RX_ITR, vector - 1), @@ -2905,10 +3096,14 @@ static void i40e_vsi_configure_msix(struct i40e_vsi *vsi) q_vector->tx.latency_range = I40E_LOW_LATENCY; wr32(hw, I40E_PFINT_ITRN(I40E_TX_ITR, vector - 1), q_vector->tx.itr); + wr32(hw, I40E_PFINT_RATEN(vector - 1), + INTRL_USEC_TO_REG(vsi->int_rate_limit)); /* Linked list for the queuepairs assigned to this vector */ wr32(hw, I40E_PFINT_LNKLSTN(vector - 1), qp); for (q = 0; q < q_vector->num_ringpairs; q++) { + u32 val; + val = I40E_QINT_RQCTL_CAUSE_ENA_MASK | (I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) | (vector << I40E_QINT_RQCTL_MSIX_INDX_SHIFT) | @@ -2988,6 +3183,7 @@ static void i40e_configure_msi_and_legacy(struct i40e_vsi *vsi) u32 val; /* set the ITR configuration */ + q_vector->itr_countdown = ITR_COUNTDOWN_START; q_vector->rx.itr = ITR_TO_REG(vsi->rx_itr_setting); q_vector->rx.latency_range = I40E_LOW_LATENCY; wr32(hw, I40E_PFINT_ITR0(I40E_RX_ITR), q_vector->rx.itr); @@ -3046,24 +3242,6 @@ void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf) } /** - * i40e_irq_dynamic_enable - Enable default interrupt generation settings - * @vsi: pointer to a vsi - * @vector: enable a particular Hw Interrupt vector - **/ -void i40e_irq_dynamic_enable(struct i40e_vsi *vsi, int vector) -{ - struct i40e_pf *pf = vsi->back; - struct i40e_hw *hw = &pf->hw; - u32 val; - - val = I40E_PFINT_DYN_CTLN_INTENA_MASK | - I40E_PFINT_DYN_CTLN_CLEARPBA_MASK | - (I40E_ITR_NONE << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT); - wr32(hw, I40E_PFINT_DYN_CTLN(vector - 1), val); - /* skip the flush */ -} - -/** * i40e_irq_dynamic_disable - Disable default interrupt generation settings * @vsi: pointer to a vsi * @vector: disable a particular Hw Interrupt vector @@ -3091,7 +3269,7 @@ static irqreturn_t i40e_msix_clean_rings(int irq, void *data) if (!q_vector->tx.ring && !q_vector->rx.ring) return IRQ_HANDLED; - napi_schedule(&q_vector->napi); + napi_schedule_irqoff(&q_vector->napi); return IRQ_HANDLED; } @@ -3136,8 +3314,7 @@ static int i40e_vsi_request_irq_msix(struct i40e_vsi *vsi, char *basename) q_vector); if (err) { dev_info(&pf->pdev->dev, - "%s: request_irq failed, error: %d\n", - __func__, err); + "MSIX request_irq failed, error: %d\n", err); goto free_queue_irqs; } /* assign the mask for this irq */ @@ -3202,8 +3379,7 @@ static int i40e_vsi_enable_irq(struct i40e_vsi *vsi) int i; if (pf->flags & I40E_FLAG_MSIX_ENABLED) { - for (i = vsi->base_vector; - i < (vsi->num_q_vectors + vsi->base_vector); i++) + for (i = 0; i < vsi->num_q_vectors; i++) i40e_irq_dynamic_enable(vsi, i); } else { i40e_irq_dynamic_enable_icr0(pf); @@ -3262,9 +3438,12 @@ static irqreturn_t i40e_intr(int irq, void *data) /* only q0 is used in MSI/Legacy mode, and none are used in MSIX */ if (icr0 & I40E_PFINT_ICR0_QUEUE_0_MASK) { + struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi]; + struct i40e_q_vector *q_vector = vsi->q_vectors[0]; /* temporarily disable queue cause for NAPI processing */ u32 qval = rd32(hw, I40E_QINT_RQCTL(0)); + qval &= ~I40E_QINT_RQCTL_CAUSE_ENA_MASK; wr32(hw, I40E_QINT_RQCTL(0), qval); @@ -3273,7 +3452,7 @@ static irqreturn_t i40e_intr(int irq, void *data) wr32(hw, I40E_QINT_TQCTL(0), qval); if (!test_bit(__I40E_DOWN, &pf->state)) - napi_schedule(&pf->vsi[pf->lan_vsi]->q_vectors[0]->napi); + napi_schedule_irqoff(&q_vector->napi); } if (icr0 & I40E_PFINT_ICR0_ADMINQ_MASK) { @@ -3434,10 +3613,9 @@ static bool i40e_clean_fdir_tx_irq(struct i40e_ring *tx_ring, int budget) i += tx_ring->count; tx_ring->next_to_clean = i; - if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) { - i40e_irq_dynamic_enable(vsi, - tx_ring->q_vector->v_idx + vsi->base_vector); - } + if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) + i40e_irq_dynamic_enable(vsi, tx_ring->q_vector->v_idx); + return budget > 0; } @@ -3575,14 +3753,12 @@ static void i40e_netpoll(struct net_device *netdev) if (test_bit(__I40E_DOWN, &vsi->state)) return; - pf->flags |= I40E_FLAG_IN_NETPOLL; if (pf->flags & I40E_FLAG_MSIX_ENABLED) { for (i = 0; i < vsi->num_q_vectors; i++) i40e_msix_clean_rings(0, vsi->q_vectors[i]); } else { i40e_intr(pf->pdev->irq, netdev); } - pf->flags &= ~I40E_FLAG_IN_NETPOLL; } #endif @@ -3663,9 +3839,8 @@ static int i40e_vsi_control_tx(struct i40e_vsi *vsi, bool enable) ret = i40e_pf_txq_wait(pf, pf_q, enable); if (ret) { dev_info(&pf->pdev->dev, - "%s: VSI seid %d Tx ring %d %sable timeout\n", - __func__, vsi->seid, pf_q, - (enable ? "en" : "dis")); + "VSI seid %d Tx ring %d %sable timeout\n", + vsi->seid, pf_q, (enable ? "en" : "dis")); break; } } @@ -3741,9 +3916,8 @@ static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable) ret = i40e_pf_rxq_wait(pf, pf_q, enable); if (ret) { dev_info(&pf->pdev->dev, - "%s: VSI seid %d Rx ring %d %sable timeout\n", - __func__, vsi->seid, pf_q, - (enable ? "en" : "dis")); + "VSI seid %d Rx ring %d %sable timeout\n", + vsi->seid, pf_q, (enable ? "en" : "dis")); break; } } @@ -4038,17 +4212,15 @@ static void i40e_quiesce_vsi(struct i40e_vsi *vsi) if ((test_bit(__I40E_PORT_TX_SUSPENDED, &vsi->back->state)) && vsi->type == I40E_VSI_FCOE) { dev_dbg(&vsi->back->pdev->dev, - "%s: VSI seid %d skipping FCoE VSI disable\n", - __func__, vsi->seid); + "VSI seid %d skipping FCoE VSI disable\n", vsi->seid); return; } set_bit(__I40E_NEEDS_RESTART, &vsi->state); - if (vsi->netdev && netif_running(vsi->netdev)) { + if (vsi->netdev && netif_running(vsi->netdev)) vsi->netdev->netdev_ops->ndo_stop(vsi->netdev); - } else { + else i40e_vsi_close(vsi); - } } /** @@ -4113,8 +4285,8 @@ static int i40e_vsi_wait_txq_disabled(struct i40e_vsi *vsi) ret = i40e_pf_txq_wait(pf, pf_q, false); if (ret) { dev_info(&pf->pdev->dev, - "%s: VSI seid %d Tx ring %d disable timeout\n", - __func__, vsi->seid, pf_q); + "VSI seid %d Tx ring %d disable timeout\n", + vsi->seid, pf_q); return ret; } } @@ -4146,6 +4318,108 @@ static int i40e_pf_wait_txq_disabled(struct i40e_pf *pf) } #endif + +/** + * i40e_detect_recover_hung_queue - Function to detect and recover hung_queue + * @q_idx: TX queue number + * @vsi: Pointer to VSI struct + * + * This function checks specified queue for given VSI. Detects hung condition. + * Sets hung bit since it is two step process. Before next run of service task + * if napi_poll runs, it reset 'hung' bit for respective q_vector. If not, + * hung condition remain unchanged and during subsequent run, this function + * issues SW interrupt to recover from hung condition. + **/ +static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi) +{ + struct i40e_ring *tx_ring = NULL; + struct i40e_pf *pf; + u32 head, val, tx_pending; + int i; + + pf = vsi->back; + + /* now that we have an index, find the tx_ring struct */ + for (i = 0; i < vsi->num_queue_pairs; i++) { + if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) { + if (q_idx == vsi->tx_rings[i]->queue_index) { + tx_ring = vsi->tx_rings[i]; + break; + } + } + } + + if (!tx_ring) + return; + + /* Read interrupt register */ + if (pf->flags & I40E_FLAG_MSIX_ENABLED) + val = rd32(&pf->hw, + I40E_PFINT_DYN_CTLN(tx_ring->q_vector->v_idx + + tx_ring->vsi->base_vector - 1)); + else + val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0); + + head = i40e_get_head(tx_ring); + + tx_pending = i40e_get_tx_pending(tx_ring); + + /* Interrupts are disabled and TX pending is non-zero, + * trigger the SW interrupt (don't wait). Worst case + * there will be one extra interrupt which may result + * into not cleaning any queues because queues are cleaned. + */ + if (tx_pending && (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK))) + i40e_force_wb(vsi, tx_ring->q_vector); +} + +/** + * i40e_detect_recover_hung - Function to detect and recover hung_queues + * @pf: pointer to PF struct + * + * LAN VSI has netdev and netdev has TX queues. This function is to check + * each of those TX queues if they are hung, trigger recovery by issuing + * SW interrupt. + **/ +static void i40e_detect_recover_hung(struct i40e_pf *pf) +{ + struct net_device *netdev; + struct i40e_vsi *vsi; + int i; + + /* Only for LAN VSI */ + vsi = pf->vsi[pf->lan_vsi]; + + if (!vsi) + return; + + /* Make sure, VSI state is not DOWN/RECOVERY_PENDING */ + if (test_bit(__I40E_DOWN, &vsi->back->state) || + test_bit(__I40E_RESET_RECOVERY_PENDING, &vsi->back->state)) + return; + + /* Make sure type is MAIN VSI */ + if (vsi->type != I40E_VSI_MAIN) + return; + + netdev = vsi->netdev; + if (!netdev) + return; + + /* Bail out if netif_carrier is not OK */ + if (!netif_carrier_ok(netdev)) + return; + + /* Go thru' TX queues for netdev */ + for (i = 0; i < netdev->num_tx_queues; i++) { + struct netdev_queue *q; + + q = netdev_get_tx_queue(netdev, i); + if (q) + i40e_detect_recover_hung_queue(i, vsi); + } +} + /** * i40e_get_iscsi_tc_map - Return TC map for iSCSI APP * @pf: pointer to PF @@ -4745,11 +5019,14 @@ out: * i40e_print_link_message - print link up or down * @vsi: the VSI for which link needs a message */ -static void i40e_print_link_message(struct i40e_vsi *vsi, bool isup) +void i40e_print_link_message(struct i40e_vsi *vsi, bool isup) { - char speed[SPEED_SIZE] = "Unknown"; - char fc[FC_SIZE] = "RX/TX"; + char *speed = "Unknown"; + char *fc = "Unknown"; + if (vsi->current_isup == isup) + return; + vsi->current_isup = isup; if (!isup) { netdev_info(vsi->netdev, "NIC Link is Down\n"); return; @@ -4766,19 +5043,19 @@ static void i40e_print_link_message(struct i40e_vsi *vsi, bool isup) switch (vsi->back->hw.phy.link_info.link_speed) { case I40E_LINK_SPEED_40GB: - strlcpy(speed, "40 Gbps", SPEED_SIZE); + speed = "40 G"; break; case I40E_LINK_SPEED_20GB: - strncpy(speed, "20 Gbps", SPEED_SIZE); + speed = "20 G"; break; case I40E_LINK_SPEED_10GB: - strlcpy(speed, "10 Gbps", SPEED_SIZE); + speed = "10 G"; break; case I40E_LINK_SPEED_1GB: - strlcpy(speed, "1000 Mbps", SPEED_SIZE); + speed = "1000 M"; break; case I40E_LINK_SPEED_100MB: - strncpy(speed, "100 Mbps", SPEED_SIZE); + speed = "100 M"; break; default: break; @@ -4786,20 +5063,20 @@ static void i40e_print_link_message(struct i40e_vsi *vsi, bool isup) switch (vsi->back->hw.fc.current_mode) { case I40E_FC_FULL: - strlcpy(fc, "RX/TX", FC_SIZE); + fc = "RX/TX"; break; case I40E_FC_TX_PAUSE: - strlcpy(fc, "TX", FC_SIZE); + fc = "TX"; break; case I40E_FC_RX_PAUSE: - strlcpy(fc, "RX", FC_SIZE); + fc = "RX"; break; default: - strlcpy(fc, "None", FC_SIZE); + fc = "None"; break; } - netdev_info(vsi->netdev, "NIC Link is Up %s Full Duplex, Flow Control: %s\n", + netdev_info(vsi->netdev, "NIC Link is Up %sbps Full Duplex, Flow Control: %s\n", speed, fc); } @@ -5218,15 +5495,13 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags) "VSI reinit requested\n"); for (v = 0; v < pf->num_alloc_vsi; v++) { struct i40e_vsi *vsi = pf->vsi[v]; + if (vsi != NULL && test_bit(__I40E_REINIT_REQUESTED, &vsi->state)) { i40e_vsi_reinit_locked(pf->vsi[v]); clear_bit(__I40E_REINIT_REQUESTED, &vsi->state); } } - - /* no further action needed, so return now */ - return; } else if (reset_flags & BIT_ULL(__I40E_DOWN_REQUESTED)) { int v; @@ -5234,6 +5509,7 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags) dev_info(&pf->pdev->dev, "VSI down requested\n"); for (v = 0; v < pf->num_alloc_vsi; v++) { struct i40e_vsi *vsi = pf->vsi[v]; + if (vsi != NULL && test_bit(__I40E_DOWN_REQUESTED, &vsi->state)) { set_bit(__I40E_DOWN, &vsi->state); @@ -5241,13 +5517,9 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags) clear_bit(__I40E_DOWN_REQUESTED, &vsi->state); } } - - /* no further action needed, so return now */ - return; } else { dev_info(&pf->pdev->dev, "bad reset request 0x%08x\n", reset_flags); - return; } } @@ -5303,8 +5575,7 @@ bool i40e_dcb_need_reconfig(struct i40e_pf *pf, dev_dbg(&pf->pdev->dev, "APP Table change detected.\n"); } - dev_dbg(&pf->pdev->dev, "%s: need_reconfig=%d\n", __func__, - need_reconfig); + dev_dbg(&pf->pdev->dev, "dcb need_reconfig=%d\n", need_reconfig); return need_reconfig; } @@ -5331,16 +5602,14 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf, /* Ignore if event is not for Nearest Bridge */ type = ((mib->type >> I40E_AQ_LLDP_BRIDGE_TYPE_SHIFT) & I40E_AQ_LLDP_BRIDGE_TYPE_MASK); - dev_dbg(&pf->pdev->dev, - "%s: LLDP event mib bridge type 0x%x\n", __func__, type); + dev_dbg(&pf->pdev->dev, "LLDP event mib bridge type 0x%x\n", type); if (type != I40E_AQ_LLDP_BRIDGE_TYPE_NEAREST_BRIDGE) return ret; /* Check MIB Type and return if event for Remote MIB update */ type = mib->type & I40E_AQ_LLDP_MIB_TYPE_MASK; dev_dbg(&pf->pdev->dev, - "%s: LLDP event mib type %s\n", __func__, - type ? "remote" : "local"); + "LLDP event mib type %s\n", type ? "remote" : "local"); if (type == I40E_AQ_LLDP_MIB_REMOTE) { /* Update the remote cached instance and return */ ret = i40e_aq_get_dcb_config(hw, I40E_AQ_LLDP_MIB_REMOTE, @@ -5525,7 +5794,9 @@ u32 i40e_get_global_fd_count(struct i40e_pf *pf) **/ void i40e_fdir_check_and_reenable(struct i40e_pf *pf) { + struct i40e_fdir_filter *filter; u32 fcnt_prog, fcnt_avail; + struct hlist_node *node; if (test_bit(__I40E_FD_FLUSH_REQUESTED, &pf->state)) return; @@ -5554,6 +5825,18 @@ void i40e_fdir_check_and_reenable(struct i40e_pf *pf) dev_info(&pf->pdev->dev, "ATR is being enabled since we have space in the table now\n"); } } + + /* if hw had a problem adding a filter, delete it */ + if (pf->fd_inv > 0) { + hlist_for_each_entry_safe(filter, node, + &pf->fdir_filter_list, fdir_node) { + if (filter->fd_id == pf->fd_inv) { + hlist_del(&filter->fdir_node); + kfree(filter); + pf->fdir_pf_active_filters--; + } + } + } } #define I40E_MIN_FD_FLUSH_INTERVAL 10 @@ -5573,49 +5856,51 @@ static void i40e_fdir_flush_and_replay(struct i40e_pf *pf) if (!(pf->flags & (I40E_FLAG_FD_SB_ENABLED | I40E_FLAG_FD_ATR_ENABLED))) return; - if (time_after(jiffies, pf->fd_flush_timestamp + - (I40E_MIN_FD_FLUSH_INTERVAL * HZ))) { - /* If the flush is happening too quick and we have mostly - * SB rules we should not re-enable ATR for some time. - */ - min_flush_time = pf->fd_flush_timestamp - + (I40E_MIN_FD_FLUSH_SB_ATR_UNSTABLE * HZ); - fd_room = pf->fdir_pf_filter_count - pf->fdir_pf_active_filters; + if (!time_after(jiffies, pf->fd_flush_timestamp + + (I40E_MIN_FD_FLUSH_INTERVAL * HZ))) + return; - if (!(time_after(jiffies, min_flush_time)) && - (fd_room < I40E_FDIR_BUFFER_HEAD_ROOM_FOR_ATR)) { - if (I40E_DEBUG_FD & pf->hw.debug_mask) - dev_info(&pf->pdev->dev, "ATR disabled, not enough FD filter space.\n"); - disable_atr = true; - } + /* If the flush is happening too quick and we have mostly SB rules we + * should not re-enable ATR for some time. + */ + min_flush_time = pf->fd_flush_timestamp + + (I40E_MIN_FD_FLUSH_SB_ATR_UNSTABLE * HZ); + fd_room = pf->fdir_pf_filter_count - pf->fdir_pf_active_filters; - pf->fd_flush_timestamp = jiffies; - pf->flags &= ~I40E_FLAG_FD_ATR_ENABLED; - /* flush all filters */ - wr32(&pf->hw, I40E_PFQF_CTL_1, - I40E_PFQF_CTL_1_CLEARFDTABLE_MASK); - i40e_flush(&pf->hw); - pf->fd_flush_cnt++; - pf->fd_add_err = 0; - do { - /* Check FD flush status every 5-6msec */ - usleep_range(5000, 6000); - reg = rd32(&pf->hw, I40E_PFQF_CTL_1); - if (!(reg & I40E_PFQF_CTL_1_CLEARFDTABLE_MASK)) - break; - } while (flush_wait_retry--); - if (reg & I40E_PFQF_CTL_1_CLEARFDTABLE_MASK) { - dev_warn(&pf->pdev->dev, "FD table did not flush, needs more time\n"); - } else { - /* replay sideband filters */ - i40e_fdir_filter_restore(pf->vsi[pf->lan_vsi]); - if (!disable_atr) - pf->flags |= I40E_FLAG_FD_ATR_ENABLED; - clear_bit(__I40E_FD_FLUSH_REQUESTED, &pf->state); - if (I40E_DEBUG_FD & pf->hw.debug_mask) - dev_info(&pf->pdev->dev, "FD Filter table flushed and FD-SB replayed.\n"); - } + if (!(time_after(jiffies, min_flush_time)) && + (fd_room < I40E_FDIR_BUFFER_HEAD_ROOM_FOR_ATR)) { + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "ATR disabled, not enough FD filter space.\n"); + disable_atr = true; + } + + pf->fd_flush_timestamp = jiffies; + pf->flags &= ~I40E_FLAG_FD_ATR_ENABLED; + /* flush all filters */ + wr32(&pf->hw, I40E_PFQF_CTL_1, + I40E_PFQF_CTL_1_CLEARFDTABLE_MASK); + i40e_flush(&pf->hw); + pf->fd_flush_cnt++; + pf->fd_add_err = 0; + do { + /* Check FD flush status every 5-6msec */ + usleep_range(5000, 6000); + reg = rd32(&pf->hw, I40E_PFQF_CTL_1); + if (!(reg & I40E_PFQF_CTL_1_CLEARFDTABLE_MASK)) + break; + } while (flush_wait_retry--); + if (reg & I40E_PFQF_CTL_1_CLEARFDTABLE_MASK) { + dev_warn(&pf->pdev->dev, "FD table did not flush, needs more time\n"); + } else { + /* replay sideband filters */ + i40e_fdir_filter_restore(pf->vsi[pf->lan_vsi]); + if (!disable_atr) + pf->flags |= I40E_FLAG_FD_ATR_ENABLED; + clear_bit(__I40E_FD_FLUSH_REQUESTED, &pf->state); + if (I40E_DEBUG_FD & pf->hw.debug_mask) + dev_info(&pf->pdev->dev, "FD Filter table flushed and FD-SB replayed.\n"); } + } /** @@ -5723,15 +6008,23 @@ static void i40e_veb_link_event(struct i40e_veb *veb, bool link_up) **/ static void i40e_link_event(struct i40e_pf *pf) { - bool new_link, old_link; struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi]; u8 new_link_speed, old_link_speed; + i40e_status status; + bool new_link, old_link; /* set this to force the get_link_status call to refresh state */ pf->hw.phy.get_link_info = true; old_link = (pf->hw.phy.link_info_old.link_info & I40E_AQ_LINK_UP); - new_link = i40e_get_link_status(&pf->hw); + + status = i40e_get_link_status(&pf->hw, &new_link); + if (status) { + dev_dbg(&pf->pdev->dev, "couldn't get link state, status: %d\n", + status); + return; + } + old_link_speed = pf->hw.phy.link_info_old.link_speed; new_link_speed = pf->hw.phy.link_info.link_speed; @@ -5760,68 +6053,6 @@ static void i40e_link_event(struct i40e_pf *pf) } /** - * i40e_check_hang_subtask - Check for hung queues and dropped interrupts - * @pf: board private structure - * - * Set the per-queue flags to request a check for stuck queues in the irq - * clean functions, then force interrupts to be sure the irq clean is called. - **/ -static void i40e_check_hang_subtask(struct i40e_pf *pf) -{ - int i, v; - - /* If we're down or resetting, just bail */ - if (test_bit(__I40E_DOWN, &pf->state) || - test_bit(__I40E_CONFIG_BUSY, &pf->state)) - return; - - /* for each VSI/netdev - * for each Tx queue - * set the check flag - * for each q_vector - * force an interrupt - */ - for (v = 0; v < pf->num_alloc_vsi; v++) { - struct i40e_vsi *vsi = pf->vsi[v]; - int armed = 0; - - if (!pf->vsi[v] || - test_bit(__I40E_DOWN, &vsi->state) || - (vsi->netdev && !netif_carrier_ok(vsi->netdev))) - continue; - - for (i = 0; i < vsi->num_queue_pairs; i++) { - set_check_for_tx_hang(vsi->tx_rings[i]); - if (test_bit(__I40E_HANG_CHECK_ARMED, - &vsi->tx_rings[i]->state)) - armed++; - } - - if (armed) { - if (!(pf->flags & I40E_FLAG_MSIX_ENABLED)) { - wr32(&vsi->back->hw, I40E_PFINT_DYN_CTL0, - (I40E_PFINT_DYN_CTL0_INTENA_MASK | - I40E_PFINT_DYN_CTL0_SWINT_TRIG_MASK | - I40E_PFINT_DYN_CTL0_ITR_INDX_MASK | - I40E_PFINT_DYN_CTL0_SW_ITR_INDX_ENA_MASK | - I40E_PFINT_DYN_CTL0_SW_ITR_INDX_MASK)); - } else { - u16 vec = vsi->base_vector - 1; - u32 val = (I40E_PFINT_DYN_CTLN_INTENA_MASK | - I40E_PFINT_DYN_CTLN_SWINT_TRIG_MASK | - I40E_PFINT_DYN_CTLN_ITR_INDX_MASK | - I40E_PFINT_DYN_CTLN_SW_ITR_INDX_ENA_MASK | - I40E_PFINT_DYN_CTLN_SW_ITR_INDX_MASK); - for (i = 0; i < vsi->num_q_vectors; i++, vec++) - wr32(&vsi->back->hw, - I40E_PFINT_DYN_CTLN(vec), val); - } - i40e_flush(&vsi->back->hw); - } - } -} - -/** * i40e_watchdog_subtask - periodic checks not using event driven response * @pf: board private structure **/ @@ -5840,8 +6071,8 @@ static void i40e_watchdog_subtask(struct i40e_pf *pf) return; pf->service_timer_previous = jiffies; - i40e_check_hang_subtask(pf); - i40e_link_event(pf); + if (pf->flags & I40E_FLAG_LINK_POLLING_ENABLED) + i40e_link_event(pf); /* Update the stats for active netdevs so the network stack * can look at updated numbers whenever it cares to @@ -5850,10 +6081,12 @@ static void i40e_watchdog_subtask(struct i40e_pf *pf) if (pf->vsi[i] && pf->vsi[i]->netdev) i40e_update_stats(pf->vsi[i]); - /* Update the stats for the active switching components */ - for (i = 0; i < I40E_MAX_VEB; i++) - if (pf->veb[i]) - i40e_update_veb_stats(pf->veb[i]); + if (pf->flags & I40E_FLAG_VEB_STATS_ENABLED) { + /* Update the stats for the active switching components */ + for (i = 0; i < I40E_MAX_VEB; i++) + if (pf->veb[i]) + i40e_update_veb_stats(pf->veb[i]); + } i40e_ptp_rx_hang(pf->vsi[pf->lan_vsi]); } @@ -6164,8 +6397,9 @@ static void i40e_config_bridge_mode(struct i40e_veb *veb) { struct i40e_pf *pf = veb->pf; - dev_info(&pf->pdev->dev, "enabling bridge mode: %s\n", - veb->bridge_mode == BRIDGE_MODE_VEPA ? "VEPA" : "VEB"); + if (pf->hw.debug_mask & I40E_DEBUG_LAN) + dev_info(&pf->pdev->dev, "enabling bridge mode: %s\n", + veb->bridge_mode == BRIDGE_MODE_VEPA ? "VEPA" : "VEB"); if (veb->bridge_mode & BRIDGE_MODE_VEPA) i40e_disable_pf_switch_lb(pf); else @@ -6232,6 +6466,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb) if (pf->vsi[v]->veb_idx == veb->idx) { struct i40e_vsi *vsi = pf->vsi[v]; + vsi->uplink_seid = veb->seid; ret = i40e_add_vsi(vsi); if (ret) { @@ -6296,12 +6531,6 @@ static int i40e_get_capabilities(struct i40e_pf *pf) } } while (err); - if (((pf->hw.aq.fw_maj_ver == 2) && (pf->hw.aq.fw_min_ver < 22)) || - (pf->hw.aq.fw_maj_ver < 2)) { - pf->hw.func_caps.num_msix_vectors++; - pf->hw.func_caps.num_msix_vectors_vf++; - } - if (pf->hw.debug_mask & I40E_DEBUG_USER) dev_info(&pf->pdev->dev, "pf=%d, num_vfs=%d, msix_pf=%d, msix_vf=%d, fd_g=%d, fd_b=%d, pf_max_q=%d num_vsi=%d\n", @@ -6514,9 +6743,7 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit) } #endif /* CONFIG_I40E_DCB */ #ifdef I40E_FCOE - ret = i40e_init_pf_fcoe(pf); - if (ret) - dev_info(&pf->pdev->dev, "init_pf_fcoe failed: %d\n", ret); + i40e_init_pf_fcoe(pf); #endif /* do basic switch setup */ @@ -6538,9 +6765,9 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit) /* make sure our flow control settings are restored */ ret = i40e_set_fc(&pf->hw, &set_fc_aq_fail, true); if (ret) - dev_info(&pf->pdev->dev, "set fc fail, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, ret), - i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); + dev_dbg(&pf->pdev->dev, "setting flow control: ret = %s last_status = %s\n", + i40e_stat_str(&pf->hw, ret), + i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); /* Rebuild the VSIs and VEBs that existed before reset. * They are still in our local switch element arrays, so only @@ -6610,6 +6837,15 @@ static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit) if (pf->flags & I40E_FLAG_MSIX_ENABLED) ret = i40e_setup_misc_vector(pf); + /* Add a filter to drop all Flow control frames from any VSI from being + * transmitted. By doing so we stop a malicious VF from sending out + * PAUSE or PFC frames and potentially controlling traffic for other + * PF/VF VSIs. + * The FW can still send Flow control frames if enabled. + */ + i40e_add_filter_to_drop_tx_flow_control_frames(&pf->hw, + pf->main_vsi_seid); + /* restart the VSIs that were rebuilt and running before the reset */ i40e_pf_unquiesce_all_vsi(pf); @@ -6808,6 +7044,7 @@ static void i40e_service_task(struct work_struct *work) return; } + i40e_detect_recover_hung(pf); i40e_reset_subtask(pf); i40e_handle_mdd_event(pf); i40e_vc_process_vflr_event(pf); @@ -6991,6 +7228,7 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type) vsi->idx = vsi_idx; vsi->rx_itr_setting = pf->rx_itr_default; vsi->tx_itr_setting = pf->tx_itr_default; + vsi->int_rate_limit = 0; vsi->rss_table_size = (vsi->type == I40E_VSI_MAIN) ? pf->rss_table_size : 64; vsi->netdev_registered = false; @@ -7009,6 +7247,8 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type) /* Setup default MSIX irq handler for VSI */ i40e_vsi_setup_irqhandler(vsi, i40e_msix_clean_rings); + /* Initialize VSI lock */ + spin_lock_init(&vsi->mac_filter_list_lock); pf->vsi[vsi_idx] = vsi; ret = vsi_idx; goto unlock_pf; @@ -7566,7 +7806,7 @@ static int i40e_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed) "Cannot set RSS key, err %s aq_err %s\n", i40e_stat_str(&pf->hw, ret), i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); - return ret; + goto config_rss_aq_out; } if (vsi->type == I40E_VSI_MAIN) @@ -7580,6 +7820,8 @@ static int i40e_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed) i40e_stat_str(&pf->hw, ret), i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); +config_rss_aq_out: + kfree(rss_lut); return ret; } @@ -7854,6 +8096,7 @@ static int i40e_sw_init(struct i40e_pf *pf) /* Set default capability flags */ pf->flags = I40E_FLAG_RX_CSUM_ENABLED | I40E_FLAG_MSI_ENABLED | + I40E_FLAG_LINK_POLLING_ENABLED | I40E_FLAG_MSIX_ENABLED; if (iommu_present(&pci_bus_type)) @@ -7896,12 +8139,12 @@ static int i40e_sw_init(struct i40e_pf *pf) (pf->hw.func_caps.fd_filters_best_effort > 0)) { pf->flags |= I40E_FLAG_FD_ATR_ENABLED; pf->atr_sample_rate = I40E_DEFAULT_ATR_SAMPLE_RATE; - if (!(pf->flags & I40E_FLAG_MFP_ENABLED)) { - pf->flags |= I40E_FLAG_FD_SB_ENABLED; - } else { + if (pf->flags & I40E_FLAG_MFP_ENABLED && + pf->hw.num_partitions > 1) dev_info(&pf->pdev->dev, "Flow Director Sideband mode Disabled in MFP mode\n"); - } + else + pf->flags |= I40E_FLAG_FD_SB_ENABLED; pf->fdir_pf_filter_count = pf->hw.func_caps.fd_filters_guaranteed; pf->hw.fdir_shared_filter_count = @@ -7915,9 +8158,7 @@ static int i40e_sw_init(struct i40e_pf *pf) } #ifdef I40E_FCOE - err = i40e_init_pf_fcoe(pf); - if (err) - dev_info(&pf->pdev->dev, "init_pf_fcoe failed: %d\n", err); + i40e_init_pf_fcoe(pf); #endif /* I40E_FCOE */ #ifdef CONFIG_PCI_IOV @@ -7941,6 +8182,9 @@ static int i40e_sw_init(struct i40e_pf *pf) pf->lan_veb = I40E_NO_VEB; pf->lan_vsi = I40E_NO_VSI; + /* By default FW has this off for performance reasons */ + pf->flags &= ~I40E_FLAG_VEB_STATS_ENABLED; + /* set up queue assignment tracking */ size = sizeof(struct i40e_lump_tracking) + (sizeof(u16) * pf->hw.func_caps.num_tx_qp); @@ -8120,9 +8364,6 @@ static void i40e_del_vxlan_port(struct net_device *netdev, pf->vxlan_ports[idx] = 0; pf->pending_vxlan_bitmap |= BIT_ULL(idx); pf->flags |= I40E_FLAG_VXLAN_FILTER_SYNC; - - dev_info(&pf->pdev->dev, "deleting vxlan port %d\n", - ntohs(port)); } else { netdev_warn(netdev, "vxlan port %d was not found, not deleting\n", ntohs(port)); @@ -8274,13 +8515,15 @@ static int i40e_ndo_bridge_setlink(struct net_device *dev, * @seq: RTNL message seq # * @dev: the netdev being configured * @filter_mask: unused + * @nlflags: netlink flags passed in * * Return the mode in which the hardware bridge is operating in * i.e VEB or VEPA. **/ static int i40e_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, - u32 filter_mask, int nlflags) + u32 __always_unused filter_mask, + int nlflags) { struct i40e_netdev_priv *np = netdev_priv(dev); struct i40e_vsi *vsi = np->vsi; @@ -8309,7 +8552,7 @@ static int i40e_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, /** * i40e_features_check - Validate encapsulated packet conforms to limits * @skb: skb buff - * @netdev: This physical port's netdev + * @dev: This physical port's netdev * @features: Offload features that the stack believes apply **/ static netdev_features_t i40e_features_check(struct sk_buff *skb, @@ -8424,17 +8667,26 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) * default a MAC-VLAN filter that accepts any tagged packet * which must be replaced by a normal filter. */ - if (!i40e_rm_default_mac_filter(vsi, mac_addr)) + if (!i40e_rm_default_mac_filter(vsi, mac_addr)) { + spin_lock_bh(&vsi->mac_filter_list_lock); i40e_add_filter(vsi, mac_addr, I40E_VLAN_ANY, false, true); + spin_unlock_bh(&vsi->mac_filter_list_lock); + } } else { /* relate the VSI_VMDQ name to the VSI_MAIN name */ snprintf(netdev->name, IFNAMSIZ, "%sv%%d", pf->vsi[pf->lan_vsi]->netdev->name); random_ether_addr(mac_addr); + + spin_lock_bh(&vsi->mac_filter_list_lock); i40e_add_filter(vsi, mac_addr, I40E_VLAN_ANY, false, false); + spin_unlock_bh(&vsi->mac_filter_list_lock); } + + spin_lock_bh(&vsi->mac_filter_list_lock); i40e_add_filter(vsi, brdcast, I40E_VLAN_ANY, false, false); + spin_unlock_bh(&vsi->mac_filter_list_lock); ether_addr_copy(netdev->dev_addr, mac_addr); ether_addr_copy(netdev->perm_addr, mac_addr); @@ -8490,12 +8742,22 @@ int i40e_is_vsi_uplink_mode_veb(struct i40e_vsi *vsi) return 1; veb = pf->veb[vsi->veb_idx]; + if (!veb) { + dev_info(&pf->pdev->dev, + "There is no veb associated with the bridge\n"); + return -ENOENT; + } + /* Uplink is a bridge in VEPA mode */ - if (veb && (veb->bridge_mode & BRIDGE_MODE_VEPA)) + if (veb->bridge_mode & BRIDGE_MODE_VEPA) { return 0; + } else { + /* Uplink is a bridge in VEB mode */ + return 1; + } - /* Uplink is a bridge in VEB mode */ - return 1; + /* VEPA is now default bridge, so return 0 */ + return 0; } /** @@ -8508,10 +8770,13 @@ int i40e_is_vsi_uplink_mode_veb(struct i40e_vsi *vsi) static int i40e_add_vsi(struct i40e_vsi *vsi) { int ret = -ENODEV; - struct i40e_mac_filter *f, *ftmp; + u8 laa_macaddr[ETH_ALEN]; + bool found_laa_mac_filter = false; struct i40e_pf *pf = vsi->back; struct i40e_hw *hw = &pf->hw; struct i40e_vsi_context ctxt; + struct i40e_mac_filter *f, *ftmp; + u8 enabled_tc = 0x1; /* TC0 enabled */ int f_count = 0; @@ -8683,32 +8948,41 @@ static int i40e_add_vsi(struct i40e_vsi *vsi) vsi->id = ctxt.vsi_number; } + spin_lock_bh(&vsi->mac_filter_list_lock); /* If macvlan filters already exist, force them to get loaded */ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { f->changed = true; f_count++; + /* Expected to have only one MAC filter entry for LAA in list */ if (f->is_laa && vsi->type == I40E_VSI_MAIN) { - struct i40e_aqc_remove_macvlan_element_data element; + ether_addr_copy(laa_macaddr, f->macaddr); + found_laa_mac_filter = true; + } + } + spin_unlock_bh(&vsi->mac_filter_list_lock); - memset(&element, 0, sizeof(element)); - ether_addr_copy(element.mac_addr, f->macaddr); - element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; - ret = i40e_aq_remove_macvlan(hw, vsi->seid, - &element, 1, NULL); - if (ret) { - /* some older FW has a different default */ - element.flags |= - I40E_AQC_MACVLAN_DEL_IGNORE_VLAN; - i40e_aq_remove_macvlan(hw, vsi->seid, - &element, 1, NULL); - } + if (found_laa_mac_filter) { + struct i40e_aqc_remove_macvlan_element_data element; - i40e_aq_mac_address_write(hw, - I40E_AQC_WRITE_TYPE_LAA_WOL, - f->macaddr, NULL); + memset(&element, 0, sizeof(element)); + ether_addr_copy(element.mac_addr, laa_macaddr); + element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; + ret = i40e_aq_remove_macvlan(hw, vsi->seid, + &element, 1, NULL); + if (ret) { + /* some older FW has a different default */ + element.flags |= + I40E_AQC_MACVLAN_DEL_IGNORE_VLAN; + i40e_aq_remove_macvlan(hw, vsi->seid, + &element, 1, NULL); } + + i40e_aq_mac_address_write(hw, + I40E_AQC_WRITE_TYPE_LAA_WOL, + laa_macaddr, NULL); } + if (f_count) { vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED; pf->flags |= I40E_FLAG_FILTER_SYNC; @@ -8771,10 +9045,13 @@ int i40e_vsi_release(struct i40e_vsi *vsi) i40e_vsi_disable_irq(vsi); } + spin_lock_bh(&vsi->mac_filter_list_lock); list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) i40e_del_filter(vsi, f->macaddr, f->vlan, f->is_vf, f->is_netdev); - i40e_sync_vsi_filters(vsi); + spin_unlock_bh(&vsi->mac_filter_list_lock); + + i40e_sync_vsi_filters(vsi, false); i40e_vsi_delete(vsi); i40e_vsi_free_q_vectors(vsi); @@ -8999,8 +9276,7 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type, if (veb) { if (vsi->seid != pf->vsi[pf->lan_vsi]->seid) { dev_info(&vsi->back->pdev->dev, - "%s: New VSI creation error, uplink seid of LAN VSI expected.\n", - __func__); + "New VSI creation error, uplink seid of LAN VSI expected.\n"); return NULL; } /* We come up by default in VEPA mode if SRIOV is not @@ -9650,6 +9926,7 @@ static int i40e_setup_pf_switch(struct i40e_pf *pf, bool reinit) } else { /* force a reset of TC and queue layout configurations */ u8 enabled_tc = pf->vsi[pf->lan_vsi]->tc_config.enabled_tc; + pf->vsi[pf->lan_vsi]->tc_config.enabled_tc = 0; pf->vsi[pf->lan_vsi]->seid = pf->main_vsi_seid; i40e_vsi_config_tc(pf->vsi[pf->lan_vsi], enabled_tc); @@ -9673,7 +9950,7 @@ static int i40e_setup_pf_switch(struct i40e_pf *pf, bool reinit) i40e_config_rss(pf); /* fill in link information and enable LSE reporting */ - i40e_aq_get_link_info(&pf->hw, true, NULL, NULL); + i40e_update_link_info(&pf->hw); i40e_link_event(pf); /* Initialize user-specific link properties */ @@ -9791,8 +10068,14 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf) } pf->queues_left = queues_left; + dev_dbg(&pf->pdev->dev, + "qs_avail=%d FD SB=%d lan_qs=%d lan_tc0=%d vf=%d*%d vmdq=%d*%d, remaining=%d\n", + pf->hw.func_caps.num_tx_qp, + !!(pf->flags & I40E_FLAG_FD_SB_ENABLED), + pf->num_lan_qps, pf->rss_size, pf->num_req_vfs, pf->num_vf_qps, + pf->num_vmdq_vsis, pf->num_vmdq_qps, queues_left); #ifdef I40E_FCOE - dev_info(&pf->pdev->dev, "fcoe queues = %d\n", pf->num_fcoe_qps); + dev_dbg(&pf->pdev->dev, "fcoe queues = %d\n", pf->num_fcoe_qps); #endif } @@ -9860,12 +10143,19 @@ static void i40e_print_features(struct i40e_pf *pf) } if (pf->flags & I40E_FLAG_DCB_CAPABLE) buf += sprintf(buf, "DCB "); +#if IS_ENABLED(CONFIG_VXLAN) + buf += sprintf(buf, "VxLAN "); +#endif if (pf->flags & I40E_FLAG_PTP) buf += sprintf(buf, "PTP "); #ifdef I40E_FCOE if (pf->flags & I40E_FLAG_FCOE_ENABLED) buf += sprintf(buf, "FCOE "); #endif + if (pf->flags & I40E_FLAG_VEB_MODE_ENABLED) + buf += sprintf(buf, "VEB "); + else + buf += sprintf(buf, "VEPA "); BUG_ON(buf > (string + INFO_STRING_LEN)); dev_info(&pf->pdev->dev, "%s\n", string); @@ -9886,14 +10176,15 @@ static void i40e_print_features(struct i40e_pf *pf) static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct i40e_aq_get_phy_abilities_resp abilities; - unsigned long ioremap_len; struct i40e_pf *pf; struct i40e_hw *hw; static u16 pfs_found; + u16 wol_nvm_bits; u16 link_status; - int err = 0; + int err; u32 len; u32 i; + u8 set_fc_aq_fail; err = pci_enable_device_mem(pdev); if (err) @@ -9939,15 +10230,15 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) hw = &pf->hw; hw->back = pf; - ioremap_len = min_t(unsigned long, pci_resource_len(pdev, 0), - I40E_MAX_CSR_SPACE); + pf->ioremap_len = min_t(int, pci_resource_len(pdev, 0), + I40E_MAX_CSR_SPACE); - hw->hw_addr = ioremap(pci_resource_start(pdev, 0), ioremap_len); + hw->hw_addr = ioremap(pci_resource_start(pdev, 0), pf->ioremap_len); if (!hw->hw_addr) { err = -EIO; dev_info(&pdev->dev, "ioremap(0x%04x, 0x%04x) failed: 0x%x\n", (unsigned int)pci_resource_start(pdev, 0), - (unsigned int)pci_resource_len(pdev, 0), err); + pf->ioremap_len, err); goto err_ioremap; } hw->vendor_id = pdev->vendor; @@ -10005,7 +10296,13 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pf->hw.fc.requested_mode = I40E_FC_NONE; err = i40e_init_adminq(hw); - dev_info(&pdev->dev, "%s\n", i40e_fw_version_str(hw)); + + /* provide nvm, fw, api versions */ + dev_info(&pdev->dev, "fw %d.%d.%05d api %d.%d nvm %s\n", + hw->aq.fw_maj_ver, hw->aq.fw_min_ver, hw->aq.fw_build, + hw->aq.api_maj_ver, hw->aq.api_min_ver, + i40e_nvm_version_str(hw)); + if (err) { dev_info(&pdev->dev, "The driver for the device stopped because the NVM image is newer than expected. You must install the most recent version of the network driver.\n"); @@ -10105,10 +10402,13 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) INIT_WORK(&pf->service_task, i40e_service_task); clear_bit(__I40E_SERVICE_SCHED, &pf->state); pf->flags |= I40E_FLAG_NEED_LINK_UPDATE; - pf->link_check_timeout = jiffies; - /* WoL defaults to disabled */ - pf->wol_en = false; + /* NVM bit on means WoL disabled for the port */ + i40e_read_nvm_word(hw, I40E_SR_NVM_WAKE_ON_LAN, &wol_nvm_bits); + if ((1 << hw->port) & wol_nvm_bits || hw->partition_id != 1) + pf->wol_en = false; + else + pf->wol_en = true; device_set_wakeup_enable(&pf->pdev->dev, pf->wol_en); /* set up the main switch operations */ @@ -10149,6 +10449,25 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) dev_info(&pdev->dev, "setup_pf_switch failed: %d\n", err); goto err_vsis; } + + /* Make sure flow control is set according to current settings */ + err = i40e_set_fc(hw, &set_fc_aq_fail, true); + if (set_fc_aq_fail & I40E_SET_FC_AQ_FAIL_GET) + dev_dbg(&pf->pdev->dev, + "Set fc with err %s aq_err %s on get_phy_cap\n", + i40e_stat_str(hw, err), + i40e_aq_str(hw, hw->aq.asq_last_status)); + if (set_fc_aq_fail & I40E_SET_FC_AQ_FAIL_SET) + dev_dbg(&pf->pdev->dev, + "Set fc with err %s aq_err %s on set_phy_config\n", + i40e_stat_str(hw, err), + i40e_aq_str(hw, hw->aq.asq_last_status)); + if (set_fc_aq_fail & I40E_SET_FC_AQ_FAIL_UPDATE) + dev_dbg(&pf->pdev->dev, + "Set fc with err %s aq_err %s on get_link_info\n", + i40e_stat_str(hw, err), + i40e_aq_str(hw, hw->aq.asq_last_status)); + /* if FDIR VSI was set up, start it now */ for (i = 0; i < pf->num_alloc_vsi; i++) { if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) { @@ -10239,37 +10558,82 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent) i40e_fcoe_vsi_setup(pf); #endif - /* Get the negotiated link width and speed from PCI config space */ - pcie_capability_read_word(pf->pdev, PCI_EXP_LNKSTA, &link_status); +#define PCI_SPEED_SIZE 8 +#define PCI_WIDTH_SIZE 8 + /* Devices on the IOSF bus do not have this information + * and will report PCI Gen 1 x 1 by default so don't bother + * checking them. + */ + if (!(pf->flags & I40E_FLAG_NO_PCI_LINK_CHECK)) { + char speed[PCI_SPEED_SIZE] = "Unknown"; + char width[PCI_WIDTH_SIZE] = "Unknown"; - i40e_set_pci_config_data(hw, link_status); + /* Get the negotiated link width and speed from PCI config + * space + */ + pcie_capability_read_word(pf->pdev, PCI_EXP_LNKSTA, + &link_status); + + i40e_set_pci_config_data(hw, link_status); + + switch (hw->bus.speed) { + case i40e_bus_speed_8000: + strncpy(speed, "8.0", PCI_SPEED_SIZE); break; + case i40e_bus_speed_5000: + strncpy(speed, "5.0", PCI_SPEED_SIZE); break; + case i40e_bus_speed_2500: + strncpy(speed, "2.5", PCI_SPEED_SIZE); break; + default: + break; + } + switch (hw->bus.width) { + case i40e_bus_width_pcie_x8: + strncpy(width, "8", PCI_WIDTH_SIZE); break; + case i40e_bus_width_pcie_x4: + strncpy(width, "4", PCI_WIDTH_SIZE); break; + case i40e_bus_width_pcie_x2: + strncpy(width, "2", PCI_WIDTH_SIZE); break; + case i40e_bus_width_pcie_x1: + strncpy(width, "1", PCI_WIDTH_SIZE); break; + default: + break; + } - dev_info(&pdev->dev, "PCI-Express: %s %s\n", - (hw->bus.speed == i40e_bus_speed_8000 ? "Speed 8.0GT/s" : - hw->bus.speed == i40e_bus_speed_5000 ? "Speed 5.0GT/s" : - hw->bus.speed == i40e_bus_speed_2500 ? "Speed 2.5GT/s" : - "Unknown"), - (hw->bus.width == i40e_bus_width_pcie_x8 ? "Width x8" : - hw->bus.width == i40e_bus_width_pcie_x4 ? "Width x4" : - hw->bus.width == i40e_bus_width_pcie_x2 ? "Width x2" : - hw->bus.width == i40e_bus_width_pcie_x1 ? "Width x1" : - "Unknown")); + dev_info(&pdev->dev, "PCI-Express: Speed %sGT/s Width x%s\n", + speed, width); - if (hw->bus.width < i40e_bus_width_pcie_x8 || - hw->bus.speed < i40e_bus_speed_8000) { - dev_warn(&pdev->dev, "PCI-Express bandwidth available for this device may be insufficient for optimal performance.\n"); - dev_warn(&pdev->dev, "Please move the device to a different PCI-e link with more lanes and/or higher transfer rate.\n"); + if (hw->bus.width < i40e_bus_width_pcie_x8 || + hw->bus.speed < i40e_bus_speed_8000) { + dev_warn(&pdev->dev, "PCI-Express bandwidth available for this device may be insufficient for optimal performance.\n"); + dev_warn(&pdev->dev, "Please move the device to a different PCI-e link with more lanes and/or higher transfer rate.\n"); + } } /* get the requested speeds from the fw */ err = i40e_aq_get_phy_capabilities(hw, false, false, &abilities, NULL); if (err) - dev_info(&pf->pdev->dev, - "get phy capabilities failed, err %s aq_err %s, advertised speed settings may not be correct\n", - i40e_stat_str(&pf->hw, err), - i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); + dev_dbg(&pf->pdev->dev, "get requested speeds ret = %s last_status = %s\n", + i40e_stat_str(&pf->hw, err), + i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); pf->hw.phy.link_info.requested_speeds = abilities.link_speed; + /* get the supported phy types from the fw */ + err = i40e_aq_get_phy_capabilities(hw, false, true, &abilities, NULL); + if (err) + dev_dbg(&pf->pdev->dev, "get supported phy types ret = %s last_status = %s\n", + i40e_stat_str(&pf->hw, err), + i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status)); + pf->hw.phy.phy_types = le32_to_cpu(abilities.phy_type); + + /* Add a filter to drop all Flow control frames from any VSI from being + * transmitted. By doing so we stop a malicious VF from sending out + * PAUSE or PFC frames and potentially controlling traffic for other + * PF/VF VSIs. + * The FW can still send Flow control frames if enabled. + */ + i40e_add_filter_to_drop_tx_flow_control_frames(&pf->hw, + pf->main_vsi_seid); + /* print a string summarizing features */ i40e_print_features(pf); @@ -10317,6 +10681,7 @@ err_dma: static void i40e_remove(struct pci_dev *pdev) { struct i40e_pf *pf = pci_get_drvdata(pdev); + struct i40e_hw *hw = &pf->hw; i40e_status ret_code; int i; @@ -10324,6 +10689,10 @@ static void i40e_remove(struct pci_dev *pdev) i40e_ptp_stop(pf); + /* Disable RSS in hw */ + wr32(hw, I40E_PFQF_HENA(0), 0); + wr32(hw, I40E_PFQF_HENA(1), 0); + /* no more scheduling of any task */ set_bit(__I40E_DOWN, &pf->state); del_timer_sync(&pf->service_timer); @@ -10440,7 +10809,7 @@ static pci_ers_result_t i40e_pci_error_slot_reset(struct pci_dev *pdev) int err; u32 reg; - dev_info(&pdev->dev, "%s\n", __func__); + dev_dbg(&pdev->dev, "%s\n", __func__); if (pci_enable_device_mem(pdev)) { dev_info(&pdev->dev, "Cannot re-enable PCI device after reset.\n"); @@ -10480,13 +10849,13 @@ static void i40e_pci_error_resume(struct pci_dev *pdev) { struct i40e_pf *pf = pci_get_drvdata(pdev); - dev_info(&pdev->dev, "%s\n", __func__); + dev_dbg(&pdev->dev, "%s\n", __func__); if (test_bit(__I40E_SUSPENDED, &pf->state)) return; rtnl_lock(); i40e_handle_reset_warning(pf); - rtnl_lock(); + rtnl_unlock(); } /** @@ -10572,9 +10941,7 @@ static int i40e_resume(struct pci_dev *pdev) err = pci_enable_device_mem(pdev); if (err) { - dev_err(&pdev->dev, - "%s: Cannot enable PCI device from suspend\n", - __func__); + dev_err(&pdev->dev, "Cannot enable PCI device from suspend\n"); return err; } pci_set_master(pdev); |